Promotion of a Ti-Mediated Mannich Reaction by a Proton Source

Article abstract of DOI:10.1021/acs.joc.5b02452

Low temperature NMR studies revealed that a diastereoselective Mannich reaction between a phenyl oxazolidone-derived titanium enolate and an aromatic aldimine was found to occur only after introduction of a proton source. While various protic additives could be used to promote the transformation, the best results were obtained using AcOH to afford the corresponding Mannich products in high diastereoselectivities and yields.

Full text of DOI:10.1021/acs.joc.5b02452

Note
pubs.acs.org/joc
Promotion of a Ti-Mediated Mannich Reaction by a Proton Source
John Limanto,* Naoki Yoshikawa, Robert A. Reamer, Lushi Tan, Andrew Brunskill, and Mikhail Reibarkh
Department of Process Chemistry, Merck Research Laboratories Merck & Company, Inc., Rahway, New Jersey 07065, United States
S
* Supporting Information
ABSTRACT: Low temperature NMR studies revealed that a diastereoselective Mannich reaction between a phenyl oxazolidone-
derived titanium enolate and an aromatic aldimine was found to occur only after introduction of a proton source. While various
protic additives could be used to promote the transformation, the best results were obtained using AcOH to afford the
corresponding Mannich products in high diastereoselectivities and yields.
β-Amino acids are ubiquitous in nature1 and they have become
increasingly important building blocks for new drug develop-
ment and peptide modifications.2 While there are many
asymmetric approaches to this class of compounds,3 one of
the more common methods involves a Mannich-type reaction.4
Among these transformations, reactions between chiral
titanium (Ti) enolates and aldimines/ketimines have been
widely investigated and most of these studies involved strongly
activated imines (or iminium ions) as the electrophiles.5
The use of less activated aromatic aldimines in diaster-
eoselective Mannich reactions in conjunction with chiral
oxazolidone-derived Ti-enolates was initially reported by the
Schering-Plough group6 and subsequently investigated by
others.7 The original authors noted that unless the reactions
were quenched at low temperatures, yield variability was often
observed, presumably due to the reversibility of the trans-
formation.6e
In support of a drug development program, we began
evaluating a similar transformation, where we were interested in
accessing β-amino acid derivative 3a as a useful synthetic
intermediate. Our initial experiment showed that treatment of
aldimine 1 with in situ generated Ti-enolate8 of phenyl
oxazolidone 29 (2-enol-Ti) at −40 °C in CH2Cl2, followed
by quenching the reaction with aqueous HCl at −20 °C
afforded the desired Mannich adducts, with concomitant
removal of the TMS group, in a 5:1 anti/syn diastereose-
lectivity10 and a 52% combined yield (Scheme 1). Initial
attempts to optimize the reaction did indeed afford variable
chemical yields and poor reproducibility, even when the
reaction was quenched at low temperatures (<−20 °C).
In order to better understand these findings, we decided to
monitor the transformation more carefully by using low
temperature NMR spectroscopy, for which the data are
presented in Figure 1. As a reference, the spectrum of isolated
anti-Mannich product 3a11 is shown at the top of the figure
(1D). The spectrum at the bottom (1A) represents a mixture
of aldimine 1 (1.2 equiv) and phenyl-oxazolidone 2 (1 equiv)
in CD2Cl2, kept between −30 and −40 °C. Addition of TiCl4
(1.1 equiv) to this mixture produced a new species, presumably
a complex between 2 and TiCl4,12 as evident by the splitting of
the CH2 protons α to the carbonyl group into two separate
peaks (1B). Subsequent exposure to Hunig’s base (1.2 equiv)
generated the corresponding Ti-enolate species (2-enol-Ti),
having a distinct sp2C−H enolate peak with a chemical shift of
3.8 ppm (1C).12 To our surprise, due to the absence of the
product’s signature peaks around δ 6.3 ppm, no Mannich
product(s) appeared to form, even after warming up the NMR
sample to 25 °C.13
We initially thought that the lack of reactivity might have
been due to the low solubility of aldimine 1 and apparent
heterogeneity of the reaction mixture at low temperature. To
eliminate this possibility, we repeated the NMR experiment
using a more soluble aldimine 4, and the results are presented
in Figure 2. Similarly, no desired Mannich reaction was
observed between 2-enol-Ti and aldimine 4 as illustrated in the
NMR spectrum 2A (vs product spectrum in 2D). Furthermore,
the transformation did not appear to be promoted by the
presence of excess TiCl4. However, upon addition of 1 mol
equiv of AcOH at −40 °C, the corresponding Mannich product
5a started to form, as evidenced by appearance of product
signature peaks around δ 6.2 ppm (2B). Treatment with
another equivalent of AcOH drove the reaction further (2C)
with concomitant formation of starting oxazolidone 2,
presumably via quenching of the Ti-enolate species.14 To the
best of our knowledge, this is the first reported observation
where a protic additive such as acetic acid is actually required to
Received: October 23, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.5b02452
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
Scheme 1. Initial Result on Ti-Mediated Mannich Reaction
Figure 1. Low temperature NMR studies of imine 1 and Ti-enolate of 2. The NMR studies were conducted in CD2Cl2 at an initial temperature of
−40 °C under an inert atmosphere and further monitored at rt. 1A: 1H NMR of a mixture of iodo-bromophenyl aldimine 1 and oxazolidone 2. 1B:
1H NMR of the mixture after charging TiCl4. 1C: 1H NMR of the mixture after charging Hunig’s base at −40 °C and allowed to warm up to rt. 1D:
1H NMR of authentic anti-Mannich product 3a.
promote a Mannich reaction between a Ti-derived enolate and
an aromatic aldimine.
was performed in the presence of aldimine at −40 °C in
CH2Cl2 prior to exposure to the proton source, slowly added
over 2 h as a solution in CH2Cl2. The mixture was subsequently
treated with 1 N HCl to effect removal of the TMS protecting
group. Entries 1−3 showed that simple alcohols such as ethanol
or 2-propanol or more acidic alcohols, such as trifluoroethanol,
afforded the Mannich products in decent anti/syn diaster-
eoselectivities and high combined assay yields. Catechol also
appeared to promote the reaction with good diastereoselectivy,
albeit in lower assay yield (entry 4). The use of chiral alcohol
sources, such as menthol or R-/S-BINOL,15 did not improve
the diastereoselectivities, but retained the high assay yields
(entries 5−6). On the other hand, promoting the reaction with
more acidic proton sources, such as PPTS or mandelic acid,
Regeneration of starting oxazolidone 2 during addition of an
acid promoter suggests that, depending on the proton source
and how it is introduced, protonation of the 2-enol-Ti
intermediate could compete favorably with the Mannich
reaction, resulting in overall lower conversions. Supporting
this hypothesis, we found that fast addition of AcOH (i.e., over
5−10 min) led to incomplete conversions (70−80%) with
respect to oxazolidone 2 as the limiting reagent, whereas higher
conversions (>90%) were obtained over 1−2 h of addition rate.
In attempts to further optimize the desired transformation, a
variety of proton sources were also evaluated and the results are
tabulated in Table 1. In all experiments, Ti-enolate formation
B
DOI: 10.1021/acs.joc.5b02452
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
Figure 2. Low temperature NMR studies of imine 4 and Ti-enolate of 2. The NMR studies were conducted in CD2Cl2 at an initial temperature of
−40 °C under an inert atmosphere and further monitored at rt. 2A: 1H NMR of a mixture of aldimine 4 and Ti-enolate of 2. 2B: 1H NMR of the
mixture postaddition of 1 equiv of AcOH at −40 °C and aged for 1 h. 2C: 1H NMR of the mixture postaddition of 2 equiv of AcOH at −40 °C and
1
aged for 1 h. 2D: H NMR of authentic anti-Mannich product 5a (both TMS groups are intact).
gave similar diastereoselectivities, albeit with much lower assay
yields (entries 7−8).
diastereoselectivity, while maintaining a high assay yield
(entry 5). Other tertiary alkyl amines, such as nBu3N,
Cy2NMe, and TMEDA, afforded the Mannich products with
comparable diastereoselectivities (entries 6−8). It is interesting
to note that the use of (−)-sparteine appeared to deteriorate
the selectivity, but retained the good yield (entry 9). Upon
scaling-up the conditions in entry 5, diastereomerically pure
anti-Mannich product 3a could be isolated in 86% yield after
crystallization of the crude product from a mixture of the
CH2Cl2/MTBE/heptane system.
The preference for the anti- and facial selectivity could be
rationalized by a six-member ring chelated transition state TS1
or TS2 (Figure 3),6e involving a (Z)-Ti-enolate16 and the
thermodynamically more stable (E)-imine.17 This proposal,
however, does not take into consideration the essential role of a
proton source as the reaction promoter, as discovered through
our experimental findings. In this regards, an alternative
transition state TS3 is proposed to be in effect, where the
hydroxyl group of the proton source (ROH) coordinates to the
Ti-center, thereby providing an acidic proton for imine
activation.18
By looking more closely at AcOH as a promoter, a
diastereoselectivity as high as 10:1 and an assay yield of 95%
could actually be obtained (entry 9) upon adding the acid
slowly as described above. While addition of a premade slurry
of aldimine 1 and AcOH in CH2Cl2 at −40 °C into a
preformed Ti-enolate solution of 2 in CH2Cl2 over 1−2 h also
effected the Mannich reaction, the best conversion observed
was only 80%, accompanied by a less clean reaction profile. Due
to this reason and the inconvenience of maintaining the slow
addition of a heterogeneous mixture of presumed protonated
imine species, the best practical mode is to slowly add a
solution of AcOH into a mixture of aldimine and 2-enol-Ti in
the corresponding solvent at −50 to −40 °C to effect the
transformation efficiently.
Focusing on AcOH as a reaction promoter, we next
evaluated the potential impact of amines and solvents on
reaction yields and diastereoselectivities. It was found early on
that the formation of Ti-enolate 2-enol-Ti could be generated
more efficiently in CH2Cl2 than in any other solvents listed in
Table 2. Hence, for the remainder of the study, the enolate
formation was always carried out in CH2Cl2 and a cosolvent
was introduced to the reaction mixture after enolization was
complete, prior to exposure to AcOH. Common solvents, such
as toluene, DME, or NMP, afforded lower diastereoselectivities
and assay yields (entries 2−4) relative to THF or no cosolvent
(entry 1). The use of EtOAc slightly improved the
In summary, we have found that an in situ generated Ti-
enolate of oxazolidone such as 2-enol-Ti only reacts with
unactivated aldimines such as 1 in the presence of a proton source.
To the best of our knowledge, this is the first report of this
observation. Specifically, addition of AcOH over 2−3 h to a
mixture of 2-enol-Ti and 1 in a 1:1 mixture of CH2Cl2/EtOAc
at −50 to −40 °C triggered the Mannich reaction to afford the
C
DOI: 10.1021/acs.joc.5b02452
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
a
a
Table 1. Proton Sources Screen for Mannich Reaction
Table 2. Amine and Cosolvent Effect on Mannich Reaction
b
c
d
b
c
d
entry
amine
iPr2NEt
cosolvents
d.r.
assay yield
entry
proton source
EtOH
d.r.
assay yield
1
2
3
4
5
6
7
8
9
THF or none
PhCH3
DME
10:1
9:1
95%
88%
90%
83%
95%
95%
95%
83%
93%
1
2
3
4
5
6
7
8
9
6:1
5:1
93%
99%
99%
75%
88%
95%
34%
75%
95%
iPr2NEt
2-propanol
CF3CH2OH
Catechol
iPr2NEt
6:1
5:1
iPr2NEt
NMP
3:1
10:1
6:1
iPr2NEt
nBu3N
EtOAc
THF
12:1
9:1
Menthol
(R)- or (S)-BINOL
5:1
Cy2NMe
TMEDA
(−)-sparteine
THF
8:1
PPTS
7:1
e
THF
8:1
Mandelic acid
8:1
f
THF
5:1
AcOH
10:1
a
a
Reaction conditions: [Oxazolidone 2] = 0.45 M in CH2Cl2, Aldimine
Reaction conditions: [Oxazolidone 2] = 0.45 M in CH2Cl2, Aldimine
1 (1.2 equiv), TiCl4 (1.1 equiv), Amine (1.2 equiv), aged for 1 h
between −40 °C for enolate formation. AcOH was added over 2−3 h
between −50 and −40 °C, followed by 1 N HCl to affect TMS group
removal. Added after enolate formation at 50% of total volume.
Measured by HPLC. Calculated by HPLC method against a product
1 (1.2 equiv), TiCl4 (1.1 equiv), iPr2NEt (1.2 equiv), aged for 1h at
b
−40 °C for enolate formation. Added at −40 °C over 2 h in CH2Cl2
c
followed by 1 N HCl to affect TMS removal. Measured by HPLC or
b
d
1HNMR analysis of crude reaction mixture. Combined assay yields,
c
d
e
calculated by HPLC method against a product standard. Added as a
solution in anhydrous THF. Added as a solution in anhydrous THF
standard.
f
or CH2Cl2 over 2−3 h between −50 and −40 °C.
corresponding β-amino products in a 12:1 anti/syn diaster-
eoselectivity and in 95% assay yield. The anti diastereomer was
selectively crystallized (after in situ TMS group removal) to give
pure product 3a in 86% isolated yield. Under these reaction
conditions, the competing protonation was kept to a minimum,
while maintaining high conversions and reaction yields.
Figure 3. Possible transition states.
EXPERIMENTAL SECTION
■
General Methods. All reactions were conducted under an inert
(N2) atmosphere. Commercial reagents and commercial HPLC and
anhydrous solvents were used without further purification. Concen-
tration in vacuo refers to removal of the solvent using a rotary
evaporator at reduced pressure (10−20 Torr). Reaction mixtures and
Mannich products were analyzed using reversed phase high perform-
ance liquid chromatography (RP-HPLC) with MeCN and 0.1%v
(−40 °C), via a thin cannula, under air-free conditions. The 1H NMR
spectra were then taken using a 500 MHz instrument, where the probe
was precooled to −40 °C and then warmed up gradually to rt, at which
point no desired Mannich product was observed (Figure 1). For the
experiment involving imine 4 (Figure 2), neat AcOH was
incrementally added (2 × 1 equiv wrt remaining Ti-enolate) to the
mixture of Ti-enolate of 2 and imine 4 (freshly prepared and aged for
1 h at −40 °C) and an aliquot was sampled after 1 h of aging using the
same technique described above. Formation of the desired Mannich
product was observed after an initial equivalent charge of AcOH and
continued to form after an additional charge and aging for another
hour.
(E)-1-(4-Bromophenyl)-N-(4-iodophenyl)methanimine (1).
To a solution of p-bromobenzaldehyde (10 g, 54 mmol, 1 equiv) in
dry MeOH (120 mL) was added p-iodoaniline (12.1 g, 54 mmol, 1
equiv) portionwise as a solid. Within a few minutes, precipitation was
obtained and the mixture was aged at rt overnight. The slurry was
filtered, washed with cold MeOH (50 mL), and dried in vacuo. The
desired aldimine was obtained in 96% isolated yield (20 g) and used as
is in the next step. Mp 172 °C. 1H NMR (CDCl3, 500 MHz) δ 8.37 (s,
1H), 7.76 (dm, J = 8.4 Hz), 7.71 (dm, J = 8.7 Hz), 7.62 (dm, J = 8.4
Hz), 6.97 (dm, J = 8.7 Hz). 13C NMR (CDCl3, 125 MHz) δ 159.3,
1
H3PO4/H2O as the mobile phase. The H NMR experiments were
conducted on either a 500 or 600 MHz instrument, and H and 13C
1
NMR spectra were all measured on a 500 MHz instrument. The high
resolution mass spectroscopy (HRMS) spectra were obtained using a
UPLC H-class system with QTOF and an electrospray ionization
(ESI) ion source. The melting point data were measured by a
differential scanning calorimetry (DSC) instrument. The concen-
tration for specific optical rotations was reported in grams/100 mL of
solvent.
Low Temperature NMR Experiments (Figures 1−2). The
samples for NMR experiments were prepared in small 10 mL vials
according to the general Mannich procedure described below. For
example, a mixture of OTMS-oxazolidone (400 mg scale) and imine 1
or 4 in either CD2Cl2 (4 mL, KF < 100 ppm, for imine 1) or extra dry
CH2Cl2 (4 mL, KF < 20 ppm, for imine 4) was treated with neat TiCl4
followed by Hunig’s base at −40 °C. After aging for 1 h, an aliquot of
sample (0.5−0.6 mL) was then transferred into a precooled NMR tube
D
DOI: 10.1021/acs.joc.5b02452
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
151.3, 138.3, 134.8, 132.1, 130.2, 126.2, 122.9, 90.7. HRMS (ESI) m/
z: calcd for C13H9BrIN + H [M + H]+: 385.9041, found 385.9040.
(S)-3-((S)-5-(4-Fluorophenyl)-5-((trimethylsilyl)oxy)-
pentanoyl)-4-phenyloxazolidin-2-one (2). To a solution of (S)-3-
((S)-5-(4-fluorophenyl)-5-hydroxypentanoyl)-4-phenyloxazolidin-2-
one9 (31.2 g, 36.5 mmol, 1 equiv) in a 1:3 mixture of toluene/MTBE
(130 mL) at −10 °C was added neat TMSCl (5.13 mL, 40.1 mmol, 1.1
equiv), followed by dropwise addition of Et3N (5.59 mL, 40.1 mmol,
1.1 equiv). The resulting thick slurry was allowed to warm up to 20 °C
overnight and then filtered. The wet cake was washed with dry MTBE
(50 mL), and the resulting solution was concentrated to give the
product as a pale yellow oil (15 g, 95% yield), which was used as-is in
the next step. If desired, the product can be precipitated by trituration
from cold hexanes at 5 °C overnight. Upon filtration and drying under
oxazolidone 2 (2 g, 4.14 mmol, 1 equiv) in dry CH2Cl2 (10 mL) was
treated with TiCl4 (0.5 mL, 4.5 mmol, 1.1 equiv) followed by Hunig’s
base (0.9 mL, 5.18 mmol, 1.25 equiv) at −40 °C. The resulting
mixture was aged at this temperature for 1 h and then treated with the
solution of TMS-protected imine. A solution of AcOH (0.47 mL, 8.28
mmol, 2 equiv) in CH2Cl2 (5 mL) was then added over 2 h at −40 °C,
and the mixture was diluted with EtOAc, allowed to warm up to 0 °C,
and treated with 2 N H2SO4 to effect TMS removal and extraction of
titanium species into the aqueous layer. The organic layer was then
separated and washed accordingly. Because isolation of the free alcohol
Mannich product was not as efficient as in compound 3a, the crude
product was resilylated using BSA (2.1 g, 10.35 mmol, 2.5 equiv) in
refluxing CH2Cl2 (25 mL) for 3 h to afford bis-TMS anti-Mannich
product 5a, which was isolated by crystallization from MTBE/hexane
25
25
as a white solid (2.5 g, 73% yield). Mp 180 °C. αD −75.9 (c 1.10,
N2, the product was isolated as a white solid. Mp 43 °C. αD +55 (c
1
1
CDCl3). H NMR (CDCl3, 500 MHz) δ 7.31 (dm, 2H, J = 8.8 Hz),
1.14, CDCl3). H NMR (CDCl3, 500 MHz) δ 7.42−7.33 (m, 3H),
7.22−7.16 (m, 3H), 7.11 (dm, 2H, J = 8.6 Hz), 7.10−7.06 (m, 4H),
6.98 (tm, 2H, J = 8.8 Hz), 6.76 (dm, 2H, J = 8.4 Hz), 6.27 (dm, 2H, J
= 8.8 Hz), 5.40 (dd, 2H, J = 8.2, 2.5 Hz), 5.13 (d, 1H, J = 10.0 Hz),
4.66 (t, 1H, J = 8.7 Hz), 4.50−4.40 (m, 2H), 4.37−4.30 (m, 1H), 4.20
(dd, 1H, J = 8.8, 3.2 Hz), 1.68−1.28 (m, 5H), 0.28 (s, 9H), −0.06 (s,
9H). 13C NMR (CDCl3, 125 MHz) δ 175.3, 162.1 (d, J = 244.6 Hz),
154.9, 154.6, 146.4, 140.9 (d, J = 3.1 Hz), 138.4, 137.8, 133.3, 128.5,
128.2, 127.5 (d, J = 8.0 Hz), 125.4, 120.3, 116.4, 115.2 (d, J = 21.4
Hz), 74.2, 70.2, 60.9, 58.4, 48.6, 38,4, 27.6, 0.5, 0.1. 19F NMR (CDCl3,
470 MHz) δ −115.7. HRMS (ESI) m/z: calcd for C39H46FIN2O5Si2 +
H [M + H]+: 825.2052, found 825.2068.
7.32−7.28 (m, 2H), 7.27−7.22 (m, 2H), 6.99 (mt, 2H, J = 8.7 Hz),
5.42 (dd, 1H, J = 8.7, 3.7 Hz), 4.69 (app t, 1H, J = 8.7 Hz), 4.62 (dd,
1H, J = 7.2, 4.7 Hz), 4.28 (dd, 1H, J = 8.7, 3.7 Hz), 2.96 (t, 2H, J = 7.1
Hz), 1.79−1.53 (m, 4H), 0.02 (s, 9H). 13C NMR (CDCl3, 125 MHz)
δ 172.7, 162.1 (d, J = 245.2 Hz), 153.9, 141.1 (d, J = 3.1 Hz), 139.4,
129.4, 128.9, 127.5 (d, J = 8.0 Hz), 126.1, 115.1 (d, J = 21.1 Hz), 74.2,
70.1, 57.8, 40.0, 35.4, 20.6, 0.3. 19F NMR (CDCl3, 470 MHz) δ
−116.1. HRMS (ESI) m/z: calcd for C23H28FNO4Si + H [M + H]+:
+
430.1850; C20H19FNO3 [M-OTMS]: 340.1343; found 340.1346 [M-
OTMS].
(S)-3-((2R,5S)-2-((S)-(4-Bromophenyl)((4-iodophenyl)amino)-
methyl)-5-(4-fluorophenyl)-5-hydroxypentanoyl)-4-phenyl-
oxazolidin-2-one (3a). General procedure for Mannich reac-
tion. A mixture of OTMS-oxazolidone 2 (2g, 4.65 mmol, 1 equiv) and
I−Br−Imine 1 (2.15g, 5.58 mmol, 1.2 equiv) in dry CH2Cl2 (20 mL)
was cooled to −40 °C and then treated with neat TiCl4 (0.56 mL, 5.12
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.joc.5b02452.
mol, 1.1 equiv), followed by Hunig’s base (0.97 mL, 5.58 mmol, 1.2
̈
equiv), to affect Ti-enolate formation (2-enol-Ti). The resulting
solution was aged at −40 °C for 1 h and then diluted with dry EtOAc
(10 mL). The resulting mixture was then treated with a solution of
AcOH (391 mg, 6.51 mmol, 1.4 equiv) in CH2Cl2 (16 mL) over 2−3
h, while maintaining the temperature between −50 and −40 °C. At the
end of the addition, the reaction mixture was diluted with EtOAc (20
mL), treated with 1 N aqueous HCl (10 mL), while allowing the
temperature to warm to rt and aging until complete TMS removal was
observed. The organic layer was separated and washed successively
with H2O (10 mL), 10 wt % aqueous K2CO3 (10 mL), and brine (20
mL). Assaying the organic layer revealed a 95% combined assay yield
of Mannich products in 12:1 anti/syn diastereoselectivity, or an 88%
assay yield of the desired anti diastereomer. Upon crystallization from
CH2Cl2/MTBE/Heptane (1:4:5), the desired product 3a was isolated
as a single diastereomer in 86% yield (3.2 g, 93.4 wt %) as an MTBE-
containing crystalline solid. Mp 120 °C (desolvation) and 178 °C
(post recrystallization from the melt). αD25 −52.6 (c 1.05, CDCl3). 1H
NMR (CDCl3, 500 MHz) δ 7.35−7.28 (m, 4H), 7.28−7.22 (m, 3H),
7.17 (mt, 2H, J = 7.8 Hz), 7.06−7.01 (m, 4H), 6.96 (dm, 2H, J = 7.5
Hz), 6.23 (dm, 2H, J = 9.8 Hz), 5.39 (dd, 1H, J = 8.4, 3.2 Hz), 5.32 (br
s, 1H), 4.66 (t, 1H, J = 8.8 Hz), 4.64−4.56 (m, 2H), 4.43 (br s, 1H),
4.20 (dd, 1H, J = 8.8, 3.2 Hz), 2.00−1.55 (m, 5H). 13C NMR (CDCl3,
125 MHz) δ 174.9, 162.5 (d, J = 245.9 Hz), 154.3, 146.1, 140 (d, J =
3.4 Hz), 139.4, 138.3, 137.9, 132.1, 129.3, 128.7, 127.7 (d, J = 8.0 Hz),
125.5, 121.6, 116.0, 115.6 (d, J = 21.5 Hz), 73.5, 70.2, 59.3, 58.1, 47.5,
36.3, 27.0. 19F NMR (CDCl3, 470 MHz) δ −114.6. HRMS (ESI) m/z:
calcd for C33H29BrFIN2O4 + H [M + H]+:743.0418, found 743.0415.
(S)-3-((2R,5S)-5-(4-Fluorophenyl)-2-((S)-((4-iodophenyl)-
amino)(4-((trimethylsilyl)oxy)phenyl)methyl)-5-((trimethyl-
silyl)oxy)pentanoyl)-4-phenyloxazolidin-2-one (5a). Same pro-
cedure as above. The imine 4 was obtained by reacting p-hydroxy
benzaldehyde (15 g, 123 mmol, 1 equiv) and p-iodoaniline (27 g, 123
mmol, 1 equiv) in MeOH to afford the desired product in 91% yield
(36 g). To a slurry of the imine (1.6 g, 4.97 mmol, 1.2 equiv) in dry
CH2Cl2 (10 mL) was added neat TMSCl (0.7 mL, 5.47 mmol, 1.3
equiv) followed by Hunig’s base (0.9 mL, 5.18 mmol, 1.25 equiv) at
−10 °C and aged for 1 h. In a separate flask, a solution of OTMS-
NMR spectra and single crystal structure of Mannich
product 3a and 5a (PDF)
Crystallographic data for 5a (CIF)
Crystallographic data for 3a (CIF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: john_limanto@merck.com.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Feng Xu, Jeongchul Kim, Paul Devine, Tim Wright,
Dave Tschaen, and Jerry Murry for useful scientific discussions.
We also thank Alfred Lee for collecting the DSC information,
Dr. Richard Staples from Crystallographic Resources, Inc. for
crystallographic support, and Rebecca Arvery for measuring
specific optical rotations.
REFERENCES
■
(1) Spiteller, P.; Von Nussbaum, F. β-Amino Acids in Natural
Products. In Enantioselective Synthesis of β-Amino Acids, 2nd ed.; Juaristi,
E., Soloshonok, V. A., Eds.; Wiley-VCH: New York, 2005; pp 19−91.
(2) Steer, D. L.; Lew, R. A.; Perlmutter, P.; Smith, A. I.; Aguilar, M. −
I. Curr. Med. Chem. 2002, 9, 811.
(3) See for examples: (a) Seebach, D.; Beck, A. K.; Capone, S.;
̌
Deniau, G.; Groselj, U.; Zass, E. Synthesis 2009, 2009, 1. (b) Liu, M.;
Sibi, M. P. Tetrahedron 2002, 58, 7991. (c) Abele, S.; Seebach, D. Eur.
J. Org. Chem. 2000, 2000, 1.
(4) For recent reviews of enantioselective Mannich reactions, see:
(a) Verkade, J. M. M.; van Hemert, L. J. C.; Quaedflieg, P. J. L. M.;
Rutjes, F. P. J. T. Chem. Soc. Rev. 2008, 37, 29. (b) Friestad, G. K.;
Mathies, A. K. Tetrahedron 2007, 63, 2541.
E
DOI: 10.1021/acs.joc.5b02452
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
(5) See for examples: (a) Tang, T. P.; Ellman, J. A. J. Org. Chem.
1999, 64, 12. (b) Kawakami, T.; Ohtake, H.; Arakawa, H.; Okachi, T.;
Imada, Y.; Murahashi, S. Org. Lett. 1999, 1, 107. (c) Bravo, P.; Fustero,
S.; Guidetti, M.; Volonterio, A.; Zanda, M. J. Org. Chem. 1999, 64,
8731. (d) Fujisawa, T.; Kooriyama, Y.; Shimizu, M. Tetrahedron Lett.
1996, 37, 3881. (e) Abrahams, I.; Motevalli, M.; Robinson, A. J.;
Wyatt, P. B. Tetrahedron 1994, 50, 12755.
(18) For a similar proposed transition state in the proline-catalyzed
Mannich reaction, see: Bahmanyar, S.; Houk, K. N. Org. Lett. 2003, 5,
1249.
(6) (a) Thiruvengadam, T. K.; Fu, X.; Tann, C.-H.; McAllister, T. L.;
Chiu, J. S.; Colon, C. U.S. Patent 6,207,822 B1, 2001.
(b) Thiruvengadam, T. K.; Tann, C.-H.; McAllister, T. L. U.S. Patent
5,561,227, 1996. (c) Thiruvengadam, T. K.; Tann, C.-H.; Lee, J.;
McAllister, T.; Sudhakar, A. U.S. Patent 5,306,817, 1994. (d) Chen, L.-
y.; Zaks, A.; Chackalamannil, S.; Dugar, S. J. Org. Chem. 1996, 61,
8341. (e) Thiruvengadam, T. K.; Sudhakar, A. R.; Wu, G. Practical
Enantio- and Diastereoselective Processes for Azetidones. In Process
Chemistry in the Pharmaceutical Industry; Gadamasetti, K. G., Ed.;
Marcel Dekker: New York, 1999; pp 221−242.
(7) See for examples: (a) Zhu, L.; Lu, C.; Chen, Z.; Yang, G.; Li, Y.;
Nie, J. Tetrahedron: Asymmetry 2015, 26, 6. (b) Ferstl, E. M.;
Venkatesan, H.; Ambhaikar, N. B.; Snyder, J. P.; Liotta, D. C. Synthesis
2002, 2002, 2075. (c) Pinheiro, S.; Greco, S. J.; Veiga, L. S.; de Farias,
F. M. C.; Costa, P. R. R. Tetrahedron: Asymmetry 2002, 13, 1157. For
a review on Gilman−Speeter type reactions, see: (d) Hart, D. J.; Ha,
D.-C. Chem. Rev. 1989, 89, 1447.
(8) For standard Ti-enolate generation conditions, see: (a) Evans, D.
A.; Urpi, F.; Somers, T. C.; Clark, J. S.; Bilodeau, M. T. J. Am. Chem.
Soc. 1990, 112, 8215. For examples of Aldol reactions involving chiral
Ti-enolates, see: (b) Crimmins, M. T.; She, J. Synlett 2004, 8, 1371.
(c) Crimmins, M. T.; McDougall, P. J. Org. Lett. 2003, 5, 591.
(d) Crimmins, M. T.; King, B. W.; Tabet, E. A.; Chaudhary, K. J. Org.
Chem. 2001, 66, 894. (e) Yan, T.-H.; Hung, A.-W.; Lee, H.-C.; Chang,
C.-S.; Liu, W.-H. J. Org. Chem. 1995, 60, 3301.
(9) See ref 6. Also, see: (a) Chen, X.; Liu, Z.-Q.; Huang, J.-F.; Lin, C.-
P.; Zheng, Y.-G. Chem. Commun. 2015, 51, 12328. (b) Bertrand, B.;
Durassier, S.; Frein, S.; Burgos, A. Tetrahedron Lett. 2007, 48, 2123.
(10) The anti-Mannich product being the major diastereomer was
consistent with the previous finding reported by the Schering-Plough
team (see ref 6) and is supported by the proposed transition state
complex shown in Figure 3.
(11) In addition to confirmation from the previous report by the
Schering-Plough team, the absolute and relative stereochemistry
configuration of the desired anti-Mannich product are confirmed by
single crystal structure analysis (see Supporting Information)
(12) The proposed six-member ring chelated complex between the
two carbonyl groups and TiCl4, as well as the corresponding six-
member ring chelated Ti-enolate species, have been widely accepted
based on the significant body of work in this area (see ref 8), for
examples. Furthermore, the presence of such chelated species appears
to be supported by expected downfield shifts of relevant carbon signals
observed in our 13C NMR experiments (see Supporting Information).
(13) Lack of reactivity was also observed using Ti(O-iPr)Cl3 in lieu
of TiCl4.
(14) Subsequent experiments showed that prolonged aging of the
quenched reaction mixture with AcOH at rt resulted in no increase in
the quantity of oxazolidone 2, suggesting irreversibility of the
transformation and that the amount of 2 observed during the quench
was due to competing protonation of the enolate.
(15) For examples on chiral protic acid-promoted reactions, see:
(a) Akiyama, T.; Itoh, J.; Fuchibe, K. Adv. Synth. Catal. 2006, 348, 999.
(b) Taylor, M. S.; Jacobsen, E. N. Angew. Chem., Int. Ed. 2006, 45,
1520. (c) Yamamoto, H.; Futatsugi, K. Angew. Chem., Int. Ed. 2005, 44,
1924. (d) Schreiner, P. R. Chem. Soc. Rev. 2003, 32, 289.
(16) For Ti-enolate-mediated Aldol reactions, see (a) Evans, D. A.;
Rieger, D. L.; Bilodeau, M. T.; Urpi, F. J. Am. Chem. Soc. 1991, 113,
1047. (b) Ha, D.-c.; Hart, D. J.; Yang, T.-k. J. Am. Chem. Soc. 1984,
106, 4819.
(17) Yeh, H. J. C.; Ziffer, H.; Jerina, D. M.; Boyd, D. M. J. Am. Chem.
Soc. 1973, 95, 2741. 1H NMR of imine 2 in CDCl3 showed only one
single isomer.
F
DOI: 10.1021/acs.joc.5b02452
J. Org. Chem. XXXX, XXX, XXX−XXX