Aziridines as intermediates in diversity-oriented syntheses of alkaloids

Article abstract of DOI:10.1016/j.tetlet.2009.02.034

The efficient synthesis of small molecules that collectively comprise optimal small-molecule screening collections is an important goal. With this in mind, we have used N-alkyl aziridines in a regio- and stereochemically controlled synthesis of polycyclic

Full text of DOI:10.1016/j.tetlet.2009.02.034

Tetrahedron Letters 50 (2009) 3230–3233
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Aziridines as intermediates in diversity-oriented syntheses of alkaloids
*
Alexander M. Taylor , Stuart L. Schreiber
Howard Hughes Medical Institute, Broad Institute of Harvard and MIT, 7 Cambridge Center, Cambridge, MA 02142, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The efficient synthesis of small molecules that collectively comprise optimal small-molecule screening
collections is an important goal. With this in mind, we have used N-alkyl aziridines in a regio- and ster-
eochemically controlled synthesis of polycyclic heterocycles based on nucleophilic ring opening and sub-
sequent intramolecular cyclization.
Received 12 January 2009
Accepted 2 February 2009
Available online 10 February 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Causal disease genes are being discovered with increasing fre-
quency. The majority of the disease targets, which include tran-
scription factors, regulatory RNAs, and signaling proteins without
enzymatic activities, are believed to be difficult if not impossible
(‘undruggable’) to modulate with small molecules. The underlying
problem appears to be that the compounds comprising standard
small-molecule screening collections used in probe and drug
development lack the necessary structural elements.1 On the other
hand, small molecules synthesized using diversity-oriented syn-
thesis (DOS) have been discovered that target many of the most
challenging targets, including ones viewed as ‘undruggable’.2
The ‘build/couple/pair’ strategy in DOS addresses many of the
goals of this type of synthetic chemistry.3–5 Here, we describe
experiments aimed at using N-alkyl aziridines as building blocks
in syntheses producing several alkaloid skeletons using a nascent
build/couple/pair strategy.
groups at the nitrogen) is more facile than that to N-alkyl aziri-
dines, known sulfonamidyl, acyl, or phosphoryl activating groups
would alter the hydrogen-bonding properties of the amine they
mask and add significant steric bulk to the molecule. Fourth, the
target aziridines can be prepared on multi-gram scale without pro-
tecting groups and with control over both their relative and abso-
lute stereochemistry. A number of efficient methods of preparing
aziridines have been reported10–14, and precedent suggests that
these methods would be amenable to incorporation of a variety
of building blocks. Also, the use of protecting groups in a parallel
synthesis is undesirable; their addition and removal add two steps
onto an otherwise concise synthetic route, and the deprotection
conditions may not be compatible with all combinations of skele-
tons and building blocks.
To test the usefulness of this aziridine-based plan, compounds
2, 3, and 4 were prepared as outlined in Scheme 2. Starting from
N-alkyl aziridines were investigated for several reasons. First,
the regio- and stereochemical outcomes of nucleophilic additions
to aziridines are directed by the nature of the substituents bound
to the ring. Theoretical studies indicate that there is a strong pref-
p-methoxybenzaldehyde, dibromochalcone 1 was formed via
Horner–Wadsworth–Emmons olefination with trimethyl acet-
ophosphonate and subsequent bromination. Although 1 was iso-
lated as a single diastereomer, both the cis- and trans-isomers of
erence of nucleophiles for addition to a carbon neighboring
p orbi-
tals that can stabilize the transition state6, and such predictions
have been confirmed in many experimental studies.7 We, there-
fore, expected that incorporating an aryl group as a substituent
would promote nucleophilic addition at the adjacent carbon of
the aziridine ring (Scheme 1). Second, many ring-opening reactions
take place through an SN2 mechanism.8,9 The inversion of stereo-
chemistry during the coupling step would allow for the control
of the relative stereochemistry of the products. Third, N-alkyl azir-
idines were expected to form products with favorable hydrogen-
bonding potential and without sterically bulky activating groups.
Although precedent indicates that the addition of nucleophiles to
activated aziridines (substituted with electron-withdrawing
R
N
R'
couple
H
N
pair
Ar
CO2Me
Ar
R
CO2Me
build
regio- and stereo-
selective ring opening
different modes of
intramolecular cyclization
N
I
OMe
CO2Me
N
OMe
N
I
O
NH
N
N
OMe
CO2Me
CO2Me
MeO
* Corresponding author.
E-mail address: stuart_schreiber@harvard.edu (S.L. Schreiber).
Present address: Massachusetts Institute of Technology, 77 Massachusetts Ave.,
Rm 18-325, Cambridge, MA 02139, USA
Scheme 1. Conceptual depiction of aziridines as building blocks in diversity-
oriented syntheses using the ‘build/couple/pair’ strategy.
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.02.034

A. M. Taylor, S. L. Schreiber / Tetrahedron Letters 50 (2009) 3230–3233
3231
2 resulted from the addition of propargylamine in DMF15, an obser-
vation that has, in previous studies, been taken as evidence that the
Br
CO2Me
CHO
a, b
c
addition takes place through an
a-bromocinnamate formed
Br
in situ.16 Aziridines 3 and 4 were prepared in the same way from
related building blocks and, in all cases, the cis- and trans-partners
were separable by column chromatography. Regardless of the
mechanism of their formation, this method allowed access to both
diastereomers of a number of aziridines.
MeO
I
MeO
MeO
1
MeO
MeO
CO2Me
I
N
N
2
N
3
We next explored a number of coupling and ring-opening reac-
tions. We found that when aziridine trans-2 was treated with
trimethylsilylazide in refluxing acetonitrile17,18, tetrahydrotriazol-
opyrazine 5 was isolated in 45% yield (Scheme 3). The reaction is
presumed to proceed through a propargylamine azide intermedi-
ate that cyclizes spontaneously under thermal conditions to yield
the triazole product. The stereochemistry of the substituents on
the piperazine ring was determined to be syn by 1H NOE NMR
experiments. Interestingly, under the same conditions, aziridine
cis-2 yielded a crude reaction mixture that contained the same ma-
jor product as that for the corresponding trans-aziridine and, in
fact, the same tetrahydrotriazolopiperazine product 5 was isolated
in 24% yield following preparative TLC purification. Upon further
analysis, both mixtures contained the same minor product, which,
although not isolated, had a 1H NMR spectrum consistent with the
anti-diastereomer of 5. To test whether this presumed epimeriza-
tion occurred spontaneously under the reaction conditions, puri-
fied triazole 5 was subjected to refluxing acetonitrile for 48 h and
analyzed. 1H NMR analysis indicated that no epimerization had ta-
ken place. Scrambling of the benzylic stereocenter was observed
during Lewis acid-catalyzed nucleophilic addition to electron-rich
aziridines (described below); a similar epimerization may have oc-
curred in this case during the ring-opening step.
CO2Me
CO2Me
4
48% (3 steps)
3:1 cis:trans
50% (3 steps)
1.3:1 cis:trans
Scheme 2. Preparation of aziridines 2, 3, and 4. Reagents and conditions: (a)
(MeO)2POCH2CO2Me, NaH, THF, 0 °C to rt; (b) Br2, CH2Cl2, rt, 78% (2 steps); (c)
propargylamine, DMF, rt, 50%, 4:1 cis:trans.
N
N
N
N
a
b
NH
CO2Me
CO2Me
MeO
MeO
MeO
MeO
trans-2
5
N
N
N
N
NH
CO2Me
CO2Me
cis-2
5
Next, we turned our attention to the coupling of the aziridines
with different alcohol nucleophiles. Following a method reported
by Vessiere19, we found that exposure of the unactivated aziridine
cis-3 to BF3ÁOEt2 in methanol quickly formed amine 6 as a single
Scheme 3. Synthesis of tetrahydrotriazolopyrazine 5. Reagents and conditions: (a)
TMSN3, MeCN, reflux, 45%, >10:1 syn:anti; (b) TMSN3, MeCN, reflux, 24%, 3:1
syn:anti.
Table 1
BF3ÁOEt2-catalyzed ring-opening experiments
R
N
OR
H
BF3 OEt2
alcohol
N
Ar
R
Ar
CO2Me
CO2Me
Entry
Substrate
Alcohol
Product
Yield (%)
MeO
OMe
OMe
I
I
OMe
H
N
MeO
I
1
MeOH
82
N
CO2Me
6
CO2Me
cis-3
OMe
H
N
I
N
2
MeOH
67
CO2Me
7
CO2Me
cis-4
MeO
MeO
OMe
OMe
I
O
H
N
I
N
3
Allyl alcohol
65
CO2Me
CO2Me
8
cis-3

3232
A. M. Taylor, S. L. Schreiber / Tetrahedron Letters 50 (2009) 3230–3233
was achieved when the parent compound 8 was treated with
OMe
OMe
O
O
I
I
H
N
a
N-iodosuccinimide in methylene chloride. Importantly, many of
the products of this pathway contain reactive handles (aryliodide,
methyl ester, and primary alkyliodide) that can be modified subse-
quently to increase the building block diversity of products of this
synthetic pathway.
CO2Me
8
I
I
OMe
OMe
OMe
OMe
Aryl iodide-containing aziridines 3 and 4 were synthesized in
order to exploit reported intramolecular arylamination methods.
Using the ring-opened products 6, 7, and 11 (prepared as described
above with trimethylsilylcyanide instead of trimethylsilylazide),
copper- and palladium-catalyzed cyclization conditions were ex-
plored (Table 2). As illustrated in Table 2, entry 1, the desired
dihydroindole 13 was isolated by column chromatography after
heating aryl iodide 7 in DMF with a catalytic amount of copper(I)
iodide, cesium carbonate, and a ligand reported by Buchwald and
coworkers.20 However, the major product of the reaction was the
corresponding indole elimination product 12, while the desired
product 13 was isolated as the minor component and as a mixture
of diastereomers (although the starting material was a single dia-
stereomer). Experiments attempting a palladium-catalyzed cycli-
zation, based on a method published by both Buchwald21 and
Fukuyama22 (Table 2, entries 2 and 4), and copper-catalyzed reac-
tions performed at lower temperatures (Table 2, entries 3 and 5)
affected no reaction with amine 11, as determined by 1H NMR
and/or LC/MS analysis. Experiments with 11 at higher temperature
(Table 2, entry 6) led to the consumption of starting material and a
complex but promising mixture, as did an experiment with 6 with
a vast excess of catalyst (Table 2, entry 7).
I
O
N
N
+
CO2Me
9
CO2Me
10
Scheme 4. Intramolecular iodoamination to form substituted morpholines.
Reagents and conditions: (a) I2, THF, rt, 37% yield, 2:1 9:10.
diastereomer (Table 1, entry 1). The stereochemistry of all ring-
opened products was inferred from the corresponding cyclized
products. These conditions were applied to allyl aziridine cis-4 to
form amine 7 (Table 1, entry 2), and also facilitated the ring open-
ing of cis-3 when performed in allyl alcohol to form amine 8 (Table
1, entry 3). These conditions also induced the ring opening of trans-
aziridines effectively (data not shown).
To determine whether these coupled products are amenable to
an iodine-induced intramolecular cyclization, ring-opened allyl
amine 8 was treated with iodine in THF. A separable 2:1 mixture
of morpholine derivatives 9 and 10 was formed in 30% yield
(shown in Scheme 4). The relative stereochemistry of the major
product was determined by 1H NOE NMR studies. The same result
Table 2
Intramolecular arylamination experiments
I
R
R
H
N
R
conditions
Me
CO2
CO2Me
N
R
Entry
1
Substrate
Conditions
Result
1.2 equiv 2-acetylcyclohexanone, 30 mol % CuI, 2.0 equiv. Cs2CO3, DMF,
90 °C
I
OMe
OMe
CO2Me
H
N
CO2Me
N
CO2Me
7
N
12
13
3:1 12:13, 20%
2
3
20 mol % Pd(PPh3)4, Et3N, reflux
No reaction
OMe
OMe
OMe
I
I
H
N
CO2Me
6
1.4 equiv 2-acetylcyclohexanone, 40 mol % CuI, 2.0 equiv Cs2CO3, DMF,
50 °C
No reaction
OMe
OMe
CN
H
N
CO2Me
11
4
5
6
11
11
11
5 mol % Pd2(dba)3, 20 mol % P(o-tol)3, 4 equiv KOt-Bu, toluene, rt
2.0 equiv CuI, 5 equiv Cs2CO3, benzene, rt
20 mol % 2-acetylcyclohexanone, 5 mol % CuI, 2 equiv. Cs2CO3, DMF, rt -
>55 °C -> 90 °C
No reaction
No reaction
Starting material consumed, complex mixture
formed
7
6
4.4 equiv. 2-acetylcyclohexanone, 1.1 equiv CuI, 2.0 equiv Cs2CO3, DMF,
50 °C
Starting material consumed, complex mixture
formed

A. M. Taylor, S. L. Schreiber / Tetrahedron Letters 50 (2009) 3230–3233
OMe
3233
I
Acknowledgement
I
OMe
H
N
+ CuI
- HI
N
Cu
The NIGMS-sponsored Center of Excellence in Chemical Method-
ology and Library Design (Broad Institute CMLD; P50 GM069721)
funded this research. S.L.S. is a Howard Hughes Medical Institute
Investigator.
CO2Me
O
MeO
Figure 1. Proposed chelate to explain low reactivity of substrates in metal-
mediated coupling reactions.
Supplementary data
Related intramolecular, metal-catalyzed reactions have been re-
ported to occur in high yield at room temperature.21,22 None of the
reported substrates involved amines substituted in the a-position
Representative experimental details for the synthesis and the
characterization data for key intermediates are provided. Supple-
mentary data associated with this article can be found, in the on-
line version, at doi:10.1016/j.tetlet.2009.02.034.
with coordinating groups, however, which led to our suspicion that
the amines used in the experiments summarized in Table 2 form a
stable five-membered metal chelate that proceeded with the oxi-
dative insertion reaction only at elevated temperature (Fig. 1). If
so, changing the carbomethoxy substituent to one that cannot
form such a chelate would be expected to increase the efficiency
of this reaction and would prevent the epimerization of the neigh-
boring stereocenter.
References and notes
1. Payne, D. J.; Gwynn, M. N.; Holmes, D. J.; Pompliano, D. L. Nat. Rev. Drug Disc.
2007, 6, 29.
2. Schreiber, S. L. Nature 2009, 457, 153.
3. Schreiber, S. L. Science 2000, 287, 1964.
4. Burke, M. D.; Schreiber, S. L. Angew. Chem., Int. Ed. 2004, 43, 46.
5. Nielsen, T. E.; Schreiber, S. L. Angew. Chem., Int. Ed. 2008, 47, 48.
6. Nielsen, I. M. B. J. Phys. Chem. A 1998, 102, 3193.
7. Park, G.; Kim, S.-C.; Kang, H.-Y. Bull. Korean Chem. Soc. 2005, 26, 1339.
8. Pineschi, M.; Bertolini, F.; Crotti, P.; Macchia, F. Org. Lett. 2006, 8, 2627.
9. Hu, X. E. Tetrahedron Lett. 2002, 43, 5315.
10. Williams, A. L.; Johnston, J. N. J. Am. Chem. Soc. 2004, 126, 1612.
11. Unthank, M. G.; Hussain, N.; Aggarwal, V. K. Angew. Chem., Int. Ed. 2006, 45,
7066.
12. Evans, D. A.; Faul, M. M.; Bilodeau, M. T. J. Org. Chem. 1991, 56, 6744.
13. Li, Z.; Conser, K. R.; Jacobsen, E. N. J. Am. Chem. Soc. 1993, 115, 5326.
14. Legters, J.; Thijs, L.; Zwanenburg, B. Tetrahedron Lett. 1989, 30, 4881.
15. Davoli, P.; Spaggiari, A.; Ciamaroni, E.; Forni, A.; Torre, G.; Prati, F. Heterocyc.
2004, 63, 2495.
In summary, here we describe studies on the use of aziridines as
components of a nascent build/couple/pair synthetic strategy. Azir-
idines were prepared with aryl substituents to direct ring opening
with trimethylsilylazide and with different alcohols in the pres-
ence of BF3ÁOEt2. These products were substrates for a thermal
[3+2] cycloaddition, an iodoamination, and an intramolecular aryl-
amination, respectively. The intermediates and products synthe-
sized in these studies will be submitted for inclusion in the
Broad Institute screening collection. Although the yields of some
of these reactions are modest, this work demonstrates that aziri-
dines can be prepared with substituents that direct subsequent
additions and allow for intramolecular cyclizations to form struc-
turally diverse, rigid, and functionalized heterocycles. Incorpora-
tion of a variety of functional groups and building blocks into
this aziridine system should allow for a number of other pairing
reactions to afford heterocycles having different ring sizes and
properties.
16. Weber, F. G.; Bandlow, C. Z. Chem. 1973, 13, 467.
17. Chandrasekhar, M.; Sekar, G.; Singh, V. K. Tetrahedron Lett. 2000, 41, 10079.
18. Li, Z.; Fernandez, M.; Jacobsen, E. N. Org. Lett. 1999, 1, 1611.
19. Bodenan, J.; Chanet-Ray, J.; Vessiere, R. Synthesis 1992, 288.
20. Shafir, A.; Buchwald, S. L. J. Am. Chem. Soc. 2006, 128, 8742.
21. Peat, A. J.; Buchwald, S. L. J. Am. Chem. Soc. 1996, 118, 1028.
22. Yamada, K.; Kubo, T.; Tokuyama, H.; Fukuyama, T. Synlett 2002, 231.