Anti-selective, catalytic asymmetric vinylogous mukaiyama mannich reactions of pyrrole-based silyl dienolates with N-aryl aldimines

Article abstract of DOI:10.1021/jo1021234

Pyrrole-based silyl dienolates undergo asymmetric vinylogous Mukaiyama Mannich reactions with a series of N-aryl aldimines in the presence of the Hoveyda-Snapper amino acid-derived silver(I) catalysts. The Mannich products-α,β-unsaturated δ-amino-γ-butyro

Full text of DOI:10.1021/jo1021234

NOTE
pubs.acs.org/joc
anti-Selective, Catalytic Asymmetric Vinylogous Mukaiyama Mannich
Reactions of Pyrrole-Based Silyl Dienolates with N-Aryl Aldimines
Claudio Curti,*,† Lucia Battistini,† Beatrice Ranieri,† Giorgio Pelosi,‡ Gloria Rassu,§ Giovanni Casiraghi,*,†
and Franca Zanardi*,†
†Dipartimento Farmaceutico, Universitꢀa degli Studi di Parma, Viale G. P. Usberti 27A, I-43124 Parma, Italy
‡Dipartimento di Chimica Generale ed Inorganica, Chimica Analitica, Chimica Fisica, Universitꢀa degli Studi di Parma, Viale G. P. Usberti
17A, I-43124 Parma, Italy
§Istituto di Chimica Biomolecolare del CNR, Traversa La Crucca 3, I-07100 Li Punti, Sassari, Italy
S Supporting Information
b
ABSTRACT: Pyrrole-based silyl dienolates undergo asymmetric
vinylogous Mukaiyama Mannich reactions with a series of N-aryl
aldimines in the presence of the Hoveyda-Snapper amino acid-
derived silver(I) catalysts. The Mannich products—R,β-unsatu-
rated δ-amino-γ-butyrolactams—are typically obtained in high
yields, excellent γ-site selectivities and anti-diastereoselectivities,
and up to 80% enantioselectivity.
he vinylogous, asymmetric Mukaiyama Mannich-type reac-
tions (VMMnR) of imines with silicon dienolates provide
Scheme 1. The Synthetic Potential of the Asymmetric
VMMnR As Applied to Pyrrole Silicon Dienolates
T
very effective ways to construct δ-amino-R,β-unsaturated carbo-
nyl frameworks, which are versatile building blocks for a
myriad of important multifunctional fragments and target
compounds.1-6 There are only a few reported studies in this
emerging area of research that utilize both acyclic and hetero-
cyclic silicon dienolates reacting with preformed or in situ-
generated imine components. In a closely related series of papers,
Schneider1 and Akiyama2a independently investigated the
Brønsted acid-catalyzed, enantioselective VMMnR of acyclic silyl
dienolates with aldimines exploiting varied axially chiral BINOL
phosphoric acid catalysts, while a disclosure by Feng and Liu3
centered upon a three-component VMMnR catalyzed by chiral
scandium(III) N,N0-dioxide complexes. The utility of the hetero-
cyclic furan-based silicon dienolates in asymmetric VMMnR with
aldimines was scrutinized with copper(I) ferrocenyl complexes,4
silver(I) diphenylphosphino binaphthyl complexes,5 and iodi-
nated dinaphthyl phosphoric acids.2b A notable advance in this
technique was recently reached by Hoveyda and Snapper,6 who
utilized furan-based silyloxy dienes in the presence of designed,
amino acid-derived silver(I) catalysts. Highly valuable, chiral
nonracemic δ-aminated γ-butenolide fragments formed in this
way, with unprecedented levels of chemical efficiency and site-,
diastereo-, and enantiocontrol.
precursor candidates to be manipulated into a huge number of
structurally and stereochemically diverse aminated carbon enti-
ties as, for example, open chain R,β-substituted-γ,δ-diamino
carbonyls of type IV bearing up to four contiguous carbon
stereocenters.
With this goal in mind, we began with the model reaction
between N-Boc-2-[(tert-butyldimethyl)silyloxy]pyrrole (1a) and
preformed arylimine 2a (Table 1), which was scrutinized with
the chiral catalysts C1-C7 displayed in Figure 1.
The catalyst choice was mainly based upon two criteria: (1)
documented efficiency in related asymmetric, catalyzed vinylo-
gous Mukaiyama-type Mannich or aldol processes and (2) easy
access from readily available natural or synthetic sources. Thus,
catalysts included amino acid-derived silver(I) complexes C1-
Complementary to these seminal works, herein we describe
our efforts to expand the scope of catalytic, asymmetric VMMnR
to involve pyrrole-based silicon dienolates, a d4 donor progeny
whose utility in this domain is still substantially unexplored.7 As
illustrated in Scheme 1, the reaction between a generic pyrrole I
and imine II can deliver enantioenriched R,β-unsaturated δ-
amino-γ-butyrolactam adducts of type III that can be viewed as
Received: October 28, 2010
Published: March 07, 2011
r
2011 American Chemical Society
2248
dx.doi.org/10.1021/jo1021234 J. Org. Chem. 2011, 76, 2248–2252
|

The Journal of Organic Chemistry
NOTE
Table 1. Initial Catalyst Screening Studies of the Asymmetric
VMMnR of Pyrrole 1a with N-Arylimine 2aa
entry
catalyst
conv (%)b
drc (anti:syn)
erd (anti)
1
2
3
4
5
6
7
C1 AgOAc
20
15
30
95
71
80
30
99:1
58:42
55:45
63:37
55:45e
55:45e
51:49e
52:48e
3
C2 AgOAc
99:1
3
C3 AgOAc
99:1
3
C4
C5
60:40
65:35
55:45
80:20
Figure 1. Structures of the chiral catalyst systems of this study.
C6 SiCl4
Table 2. Examination of Silyloxypyrrole Donors 1 Bearing
3
Different Silicon Substituents and N-Protecting Groupsa
C7 TFA
3
a For each catalyst, almost the same conditions previously utilized in the
original works were adopted: entries 1-3, ref 6a; entries 4 and 5, ref 1;
entry 6, refs 8 and 9; and entry 7, ref 10. b Conversion determined by 1H
c
NMR analysis of the crude reaction product. Determined by HPLC
analysis of the crude reaction mixture. d Determined by HPLC analysis,
using a Chiralcel OD-H column. e Absolute configuration not
determined.
C3 AgOAc developed by the Hoveyda and Snapper group,6 the
3
axially chiral BINOL phosphoric acids C4 and C5 widely
entry pyrrole R3Si
PG yield (%)b drc (anti:syn) erd (anti)
exploited by Schneider et al.,1 Denmark’s bisphosphoramide/
silicon tetrachloride dual system C6 SiCl4,8,9 and the cinchona-
1
2
3
4
5
6
1a
1b
1c
1d
1e
1f
TBS
TES
Boc
Boc
30
34
60
20
99:1
99:1
99:1
99:1
63:37
65:35
68:32
51:49
3
thiourea organocatalyst C7 TFA, successfully utilized by Deng
3
et al.10
TMS Boc
As the results in Table 1 indicate, the initial move was less
rewarding and, although high diastereoselectivity in favor of the
anti-configured adduct anti-3a was attained with the three silver
TBS
TBS
allyl
Me
e
-
-
e
TMS Me
complexes C1-C3 AgOAc (entries 1-3), reactions were in-
3
a The reaction was performed with pyrrole 1/aldehyde 2a (1.0/1.0
equiv), ligand C3 (1.0 mol %), AgOAc (1.0 mol %), undistilled i-PrOH
(1.1 equiv), undistilled and stabilized (BHT, 250 ppm) THF, in air, at 25
°C for 16 h (0.21 mmol scale). b Isolated yield of the vinylogous
Mannich products. c Determined by HPLC analysis of the crude reaction
mixture. d Determined by HPLC analysis, using a Chiralcel OD-H
column. e No Mannich product obtained.
efficient (15-30% conversion after 16 h) and enantioselectivity
was low (<65:35 dr in favor of the 5R,10S-configured isomer).
With catalysts C4 and C5, reaction rates increased (entries 4 and
5) but neither the relative nor absolute stereocontrol were
substantially detected. The dual bisphosphoramide/SiCl4 system
C6 SiCl4, which proved an excellent catalyst in directing asym-
3
metric vinylogous aldol-type processes with pyrrole- and furan-
based dienoxy silanes,9 failed with the Mannich variant, and a
balanced mixture of racemic adducts anti-3a and syn-3a was
formed (entry 6). Similarly, bifunctional alkaloid-thiourea salt
complete γ-site selectivity and anti-selectivity, with moderate
enantioselectivity (68:32 er) in favor of the 5R,10S-isomer.
To optimize the reaction conditions further, the effects of the
catalyst loading, the donor:acceptor ratio, the reaction tempera-
ture, and the additive were finally investigated by using catalyst
C3, with the results grouped in Table 3. When the reaction
between 1c (1.5 equiv excess) and 2a was performed at 0 °C in
THF, using a 10.0 mol % ligand C3, in the presence of
isopropanol/water (entry 7), the enantioselectivity of the Man-
nich product anti-3a was good (90:10 er), the yield high, and
both γ-site- and diastereoselectivity excellent. In comparison, in
the absence of water (entry 6), enantioselectivity diminished to
80:20, but yield was almost maintained. Increasing the amount of
water to 3.0 equiv did not cause any additional improvement, but
more water (10.0 equiv) was detrimental for both yield and
enantioselectivity (entries 4 and 5). The dramatic role of water
C7 TFA proved to be an inefficient catalyst that afforded poor
3
yields of racemates (entry 7).
At this point, in recognizing how the nature of the pyrrole
donor component could impact both the efficiency and stereo-
control in similar vinylogous additions to aldehyde acceptors,9
we next wondered whether the low yield and enantioselectivity of
the silver-catalyzed reaction in entry 3 of Table 1 leading to anti-
3a could be improved here by altering the nitrogen and silicon
functionalities in the donor pyrrole substrates. This proved to be
the case, indeed, as shown in entry 3 of Table 2. After the
VMMnR between N-Boc-trimethylsilyl-substituted pyrrole 1c
with imine 2a had been conducted in the presence of the silver
catalyst C3 AgOAc, anti-3a was formed in 60% yield with almost
3
2249
dx.doi.org/10.1021/jo1021234 |J. Org. Chem. 2011, 76, 2248–2252

The Journal of Organic Chemistry
NOTE
Table 3. Further Optimization of the Asymmetric VMMnR of Pyrrole 1c with N-Arylimine 2a, Using Silver(I) Catalyst
C3 AgOAca
3
entry 1c:2a (equiv) C3 AgOAc (mol %) temp (°C)
additive (equiv)
i-PrOH (1.1)
yield (%)b
γ:R
drc (anti:syn) erd (anti) (5R,10S:5S,10R)
3
1
2
3
4
5
6
7
1:1
1.5:1
1.5:1
1.5:1
1.5:1
1.5:1
1.5:1
1.0
1.0
25
60
65
15
40
70
72
80
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
99:1
99:1
99:1
99:1
99:1
99:1
99:1
68:32
80:20
92:8
0
i-PrOH (1.5)
10.0
10.0
10.0
10.0
10.0
-30
i-PrOH (1.1)
0
0
0
0
i-PrOH (1.5), H2O (10.0)
i-PrOH (1.5), H2O (3.0)
i-PrOH (1.5)
68:32
78:22
80:20
90:10
i-PrOH (1.5), H2O (1.5)
a The reactions were performed with ligand C3/AgOAc (1:1), undistilled i-PrOH, undistilled and stabilized (BHT, 250 ppm) THF, in air, for 16 h at the
indicated temperature (0.21 mmol scale). b Isolated yield of the vinylogous Mannich products. c Determined by HPLC analysis of the crude reaction
mixture. d Determined by HPLC analysis, using a Chiralcel OD-H column.
and protic additives on the efficiency and stereocontrol of these
processes is noteworthy, yet precedented in similar catalytic
reactions.1a,5,6a As a plausible explanation, one can argue that
during the catalytic cycle (vide infra, Figure S2 in the Supporting
Information) protic additives act as scavengers of the evolving
silicon ion species to form inactive silanol or silyl ether bypro-
ducts. This results in catalyst turnover enhancement and, more
importantly, in depletion of the competitive racemic background
reaction catalyzed by the silicon ions themselves.
Lowering the reation temperature to -30 °C did decrease the
yield significantly (15%) with only marginal benefit of the
enantioselectivity (entry 3). Catalyst loading affected signifi-
cantly both yield and enantioselectivity (see entry 2 vs 6), and
the reactants ratio proved to be equally decisive (entry 2 vs 1). In
summary, the optimized reaction conditions were established as
rise to the expected amino lactams anti-3g and anti-3h. In these
instances, while γ-site regiocontrol, anti-diastereopreference,
and enantiocontrol proved similar to those displayed by aromatic
imines, the efficiency declined, with isolated yields ranging from
52% to 36%.
The relative configuration of the major Mannich adduct anti-
3a was assigned by single crystal X-ray analysis of the racemate
(Supporting Information), while those of the remainder adducts
anti-3b-anti-3h were proposed by analogy. The absolute
configuration of anti-3a was unambiguously determined via
chemical correlation to the known 5S,10S-configured syn-
butyrolactam (-)-(S)-N-(tert-butoxycarbonyl)-5-[(S)-hydroxy-
(phenyl)methyl]-1H-pyrrol-2(5H)-one,9 as detailed in Scheme
S1 (Supporting Information). By analogy (all the Mannich
products were dextrorotatory), the absolute configuration of
the other adducts anti-3b/anti-3h was proposed as shown in
Table 4.
On the basis of the relative and absolute configuration of our
major adducts anti-3a/anti-3h and the previous paradigmatic
studies by the Hoveyda and Snapper group with asymmetric
VMMnR of furan-based silyl dienolates,6 the simple anti-selec-
tivity and the facial 5R-selectivity observed in the silver-catalyzed
asymmetric VMMnR of pyrrole silyl dienolates with N-aryli-
mines can be rationalized by the catalytic mechanistic cycle
depicted in Figure S2 in the Supporting Information.
10.0 mol % silver catalyst C3 AgOAc (1:1 ligand:silver salt molar
3
ratio), using 1.5 equiv excess pyrrole 1c and 1.5 equiv of
isopropanol/water (1:1 molar ratio) in commercial grade, un-
distilled BHT-stabilized THF, in an open-air vessel at 0 °C.
On the basis of the optimization efforts presented above, the
scope of the acceptor substrate of the catalytic asymmetric
VMMnR catalyzed by silver(I) catalyst C3 AgOAc was carried
3
out (Table 4). In general, all the examined substrates could
furnish the desired 5,10-bis-aminated adducts in synthetically
useful yields and selectivities.
A variety of aromatic aldehyde imines with either electron-
donating or electron-withdrawing substituents in the aromatic
ring underwent the asymmetric VMMnR to afford the respective
anti-configured adducts as the exclusive products in 65-99%
yields with excellent site-selectivities (>99:1 γ:R) and diastere-
omeric ratios (dr >98:2), with acceptable levels of enantiocontrol
ranging from 71:29 to 90:10 er. The substitution pattern of the
phenyl ring in 2 did not impact stereocontrol significantly, and
heteroaromatic aldehyde imine 2f was a viable substrate too, as
was the 2-naphthaldehyde-derived imine 2e.
To conclude, we have reported the first example of a catalytic,
asymmetric VMMnR of pyrrole-based silyl dienolates with a
series of N-arylimines. The efficient amino acid-derived silver(I)
catalyst system, previously exploited by Hoveyda and Snapper6
with furan-based silyl dienolates, exhibited good performance,
giving rise to diverse anti-configured γ,δ-diamino R,β-unsatu-
rated carbonyls in good yields, virtually complete γ-site- and
diastereoselectivities, with up to 80% enantioselectivities. Ex-
ploitation of these findings in asymmetric synthesis of
relevant multifunctional target molecules which contain vicinal
diamino motifs is planned, and research in this direction is
under way.
To enlarge the scope of the reaction further, VMMnR invol-
ving two aliphatic aldimines, 2g and 2h, were scrutinized, giving
2250
dx.doi.org/10.1021/jo1021234 |J. Org. Chem. 2011, 76, 2248–2252

The Journal of Organic Chemistry
NOTE
Table 4. Asymmetric VMMnR of Pyrrole 1c with Various Aromatic, Heteroaromatic, and Aliphatic N-Arylimines 2a-h Catalyzed
by Silver(I) Complex C3 AgOAca
3
a Conditions of entry 7 in Table 3. Yields refer to isolated vinylogous Mannich products. Diastereomeric ratios were determined by HPLC analysis of the
crude reaction mixtures. Enantiomeric ratios were determined by HPLC analysis, using a Chiralcel OD-H column. b Three-component procedure, with
the aliphatic imine component formed in situ. For details, see the Supporting Information.
’ EXPERIMENTAL SECTION
MHz, CDCl3) δ 7.47-7.30 (m, 5H, Ph), 6.92 (dd, J = 6.1, 2.0 Hz, 1H,
H4), 6.77 (dd, J = 7.4, 2.1 Hz, 1H, H300), 6.70 (ddd, J = 7.5, 7.5, 1.8 Hz, 1H,
H500), 6.65 (ddd, J = 7.5, 7.5, 1.9 Hz, 1H, H400), 6.32 (dd, J = 7.2, 2.1 Hz,
1H, H600), 6.30 (dd, J = 6.0, 1.6 Hz, 1H, H3), 5.44 (bd, J = 3.3 Hz, 1H,
H10), 5.06 (ddd, J = 3.6, 1.8, 1.8 Hz, 1H, H5), 4.44 (bs, 1H, NH), 3.86 (s,
3H, CH3), 1.56 (s, 9H, t-Bu, Boc); 13C NMR (75 MHz, CDCl3) δ 169.0
(Cq, C2), 149.6 (Cq, Boc), 147.3 (Cq, C200), 146.1 (CH, C4), 139.7 (Cq,
C100), 136.9 (Cq, Ph), 129.3 (CH, C3), 129.1 (2C, CH, Ph), 128.0 (CH,
Ph), 126.6 (2C, CH Ph), 121.2 (CH, C500), 117.7 (CH, C400), 111.6 (CH,
C600), 109.8 (CH, C300), 83.7 (Cq, Boc), 67.7 (CH, C5), 57.3 (CH, C10),
53.6 (CH3, OMe), 28.3 (3C, CH3, Boc). ESI-MS m/z 417.38 [M þ Naþ]
(calcd 417.18 [M þ Naþ]). Anal. Calcd for C23H26N2O4: C, 70.03; H,
6.64; N, 7.10. Found: C, 70.13; H, 6.78; N, 6.95.
Representative Experimental Procedure for Ag-Catalyzed
VMMnR. Preparation of (R)-5-(N-tert-Butoxycarbonyl)-[(S)-
(2-methoxyphenylamino)(phenyl)methyl]-1H-pyrrol-2(5H)-
one (anti-3a). Chiral phosphine C3 (14.0 mg, 0.02 mmol, 0.10 equiv)
and AgOAc (3.5 mg, 0.02 mmol, 0.10 equiv) were dissolved in undistilled
BHT-stabilized (250 ppm) THF (4 mL) and allowed to stir for 5 min at
22 °C. A solution of imine 2a (44.0 mg, 0.21 mmol, 1.0 equiv) in THF
(1.0 mL) was added followed by addition of a mixture of i-PrOH (24.0 μL,
0.31 mmol, 1.5 equiv)/H2O (6.0 μL, 0.31 mmol, 1.5 equiv) and the
reaction vessel was capped with a septum. The mixture was allowed to cool
to 0 °C and 1c (82 mg, 0.31 mmol, 1.5 equiv) in THF (1.0 mL) was added.
After 16 h the reaction was quenched bythe addition of a saturated aqueous
solution of NaHCO3 (0.5 mL). The mixture was allowed to warm to 22 °C
with vigorous stirring for 10 min, and finally extracted with EtOAc. The
organic layers were collected, dried with MgSO4, filtered, and concentrated
in vacuo. The diastereomeric ratio of the addition products 3a was
determined to be 99:1 by analytical HPLC (CN-100, 5 μm, hexane/
anhydrous EtOH 95:5, 1.0 mL/min, 254 nm): anti-3a, Rt 11.71 min
(99.0%); syn-3a, Rt 14.33 min (1.0%). The crude residue was then purified
by silica gel flash chromatography (hexane/Et2O 70:30), to yield 62 mg
(80%) of (þ)-anti-3a as colorless crystals: TLC, Rf 0.38 (petroleum ether/
EtOAc 70:30); chiral HPLC (Chiralcel OD-H, hexane/anhydrous EtOH
90:10, 1.0 mL/min, 254 nm): (5R,10S)-3a, Rt 8.99 min (90.0%); (5S,10R,)-
3a, Rt 10.14 min (10.0%); [R]20D þ99.5 (c 0.99, EtOH). 1H NMR (300
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental procedures,
b
spectroscopic data for all new compounds, chiral HPLC traces,
crystallographic data, and copies of NMR spectra. This material is
available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: giovanni.casiraghi@unipr.it.
2251
dx.doi.org/10.1021/jo1021234 |J. Org. Chem. 2011, 76, 2248–2252

The Journal of Organic Chemistry
NOTE
’ ACKNOWLEDGMENT
Funding from Universitꢀa degli Studi di Parma and Regione
Autonoma della Sardegna is gratefully acknowledged. The
authors thank the Centro Interdipartimentale Misure “G. Cas-
nati” (Universitꢀa degli Studi di Parma) for instrumental facilities.
We are grateful to Dr. Mattia Anselmi and Dr. Carlotta Figliola
for preliminary experiments.
’ REFERENCES
(1) (a) Sickert, M.; Schneider, C. Angew. Chem., Int. Ed. 2008,
47, 3631. (b) Giera, D. S.; Sickert, M.; Schneider, C. Org. Lett. 2008,
10, 4259. (c) Giera, D. S.; Sickert, M.; Schneider, C. Synthesis
2009, 3797. (d) Sickert, M.; Abels, F.; Lang, M.; Sieler, J.; Birkemeyer,
C.; Schneider, C. Chem.—Eur. J. 2010, 16, 2806.
(2) (a) Itoh, J.; Fuchibe, K.; Akiyama, T. Angew. Chem., Int. Ed. 2006,
45, 4796. (b) Akiyama, T.; Honma, Y.; Itoh, J.; Fuchibe, K. Adv. Synth.
Catal. 2008, 350, 399.
(3) Zhang, Q.; Hui, Y.; Zhou, X.; Lin, L.; Liu, X.; Feng, X. Adv. Synth.
Catal. 2010, 352, 976.
(4) Salvador Gonzꢁalez, A.; Gꢁomez Arrayꢁas, R.; Rodríguez Rivero,
M.; Carretero, J. C. Org. Lett. 2008, 10, 4335.
(5) (a) Deng, H.-P.; Wei, Y.; Shi, M. Adv. Synth. Catal. 2009,
351, 2897. (b) Yuan, Z.-L.; Jiang, J.-J.; Shi, M. Tetrahedron 2009,
65, 6001.
(6) (a) Carswell, E. L.; Snapper, M. L.; Hoveyda, A. H. Angew. Chem.,
Int. Ed. 2006, 45, 7230. (b) Mandai, H.; Mandai, K.; Snapper, M. L.;
Hoveyda, A. H. J. Am. Chem. Soc. 2008, 130, 17961. (c) Wieland, L. C.;
Vieira, E. M.; Snapper, M. L.; Hoveyda, A. H. J. Am. Chem. Soc. 2009,
131, 570.
(7) For a sole example of direct, asymmetric catalytic VMMnR to
aromatic aldimines (not Mukaiyama-type) utilizing R,β-unsaturated γ-
butyrolactam donors, see: Shepherd, N. E.; Tanabe, H.; Xu, Y.;
Matsunaga, S.; Shibasaki, M. J. Am. Chem. Soc. 2010, 132, 3666.
(8) Denmark, S. E.; Heemstra, J. R., Jr. J. Org. Chem. 2007, 72, 5668.
(9) Curti, C.; Ranieri, B.; Battistini, L.; Rassu, G.; Zambrano, V.;
Pelosi, G.; Casiraghi, G.; Zanardi, F. Adv. Synth. Catal. 2010, 352, 2011.
(10) Singh, R. P.; Foxman, B. M.; Deng, L. J. Am. Chem. Soc. 2010,
132, 9558.
2252
dx.doi.org/10.1021/jo1021234 |J. Org. Chem. 2011, 76, 2248–2252