Enantioselective, chromatography-free synthesis of β3-Amino acids with natural and unnatural side chains

Article abstract of DOI:10.1021/op300069n

β³-Amino acids are key components of some pharmaceuticals, excellent surrogates for metabolically labile α-amino acids, and building blocks for chiral heterocycles. Unfortunately they are not easily accessible in enantiomerically pure form, especially when possessing unnatural side chains. A flexible, chromatography-free process for the synthesis of enantiopure β³-amino acids possessing natural and unnatural side chains is described. The procedure uses inexpensive starting materials and reagents and offers a good alternative to the hazardous and expensive Arndt-Eistert homologation of enantiopure α-amino acids. Its utility has been demonstrated with the preparative scale synthesis of two valuable β³-amino acids possessing unnatural side chains.

Full text of DOI:10.1021/op300069n

Article
pubs.acs.org/OPRD
Enantioselective, Chromatography-Free Synthesis of β3-Amino Acids
with Natural and Unnatural Side Chains
Thibaud Gerfaud, Ying-Ling Chiang, Imants Kreituss, Justin A. Russak, and Jeffrey W. Bode*
Laboratorium fur Organische Chemie, Department of Chemistry and Applied Biosciences, ETH−Zurich, Wolfgang Pauli Strasse 10,
̈
̈
8093 Zurich, Switzerland
̈
S
* Supporting Information
ABSTRACT: β3-Amino acids are key components of some
pharmaceuticals, excellent surrogates for metabolically labile α-
amino acids, and building blocks for chiral heterocycles.
Unfortunately they are not easily accessible in enantiomerically
pure form, especially when possessing unnatural side chains. A
flexible, chromatography-free process for the synthesis of
enantiopure β3-amino acids possessing natural and unnatural
side chains is described. The procedure uses inexpensive
starting materials and reagents and offers a good alternative to
the hazardous and expensive Arndt−Eistert homologation of
enantiopure α-amino acids. Its utility has been demonstrated
with the preparative scale synthesis of two valuable β3-amino acids possessing unnatural side chains.
requires high hydrogen pressures and proprietary ligands.8
INTRODUCTION
■
Many other methods have been reported, but no general
approach has emerged; the various routes are usually not
amenable to scale-up or diversity in side-chain functionality and
have to be optimized for each substrate.9 For 3-aryl substituted
β3-amino acids, enzymatic resolution provides access to a
number of derivatives, but this approach does not appear to
give satisfactory results for aliphatic or functionalized side
chains.10 In some high volume cases, directed evolution of
enzymes can offer an efficient and economically viable
alternative to classical chemical approaches for the large scale
production of pharmaceuticals containing β-amino acids
residues.11 Unfortunately, this solution is not general and
requires intensive optimization for a given substrate. Other
types of β-amino acids, particularly β2-amino acids, are even
more difficult to prepare.12
β-Amino acids are a fascinating and important class of
unnatural amino acids with diverse applications. They are
present in a wide range of natural products1 (Jasplakinolide)
and are key components of some pharmaceuticals such as
Sitagliptin,2 developed at Merck, or Otamixaban,3 under
development at Sanofi-Aventis (Figure 1). They are also the
basic units of a rapidly growing class of oligomeric peptides and
are widely regarded as excellent surrogates for metabolically
labile α-amino acids.4 Considerable efforts are currently
underway to use β-amino acids for the formation of peptides
with secondary, tertiary, and even quaternary structures and to
design and identify structures with biological and catalytic
properties.5 The individual β-amino acid monomers are also
highly sought as the basic building blocks of chiral heterocycles.
Unlike peptides consisting of natural α-amino acids, which
are widely available in enantiomerically pure form at a nominal
cost, each enantiopure β-amino acid monomer must be
synthesized from a suitable chiral starting material or by
asymmetric synthesis.6 A method able to rapidly and efficiently
provide β3-amino acids with natural and unnatural side chains
from simple and commercially available starting materials
would thus be of primary interest for exploratory work. β3-
Amino acid monomers containing naturally occurring side
chains are now widely available by homologation of the
corresponding natural α-amino acids.7 Although this chemistry
is both expensive and hazardous, as it requires large quantities
of toxic diazomethane and enantiopure, suitably protected
starting materials, it has been executed on large scale with
specialized facilities. For β-amino acids containing unnatural
side chains, asymmetric enamine or enamide transfer hydro-
genation is attractive from an industrial point of view but
We have recently disclosed a unified approach to β2-, β2,3-,
and β3-amino acids featuring a chiral auxiliary mediated
cycloaddition followed by a redox neutral fragmentation of
the resulting cycloadduct to give the enantiopure β-amino acids
(Scheme 1).13 This approach has proven to be useful for the
gram scale preparation of β-amino acids, with the side chain
derived from the corresponding aldehyde. Particularly in cases
where the corresponding α-amino acids are not available, this
route provides a reliable and predictable alternative to the
Arndt−Eistert homologation. A number of β-amino acids
possessing aliphatic or protected functional groups side chains
have been synthesized under one standard set of reaction
conditions. This method is of particular interest because the
Received: March 14, 2012
Published: March 30, 2012
© 2012 American Chemical Society
687
dx.doi.org/10.1021/op300069n | Org. Process Res. Dev. 2012, 16, 687−696

Organic Process Research & Development
Article
Figure 1. Structure and occurrence of β-amino acids.
Scheme 1. Diastereoselective [3 + 2] Cycloaddition/Redox Neutral Fragmentation Approach for the Synthesis of Enantiopure
β-Amino Acids
products are obtained in enantiomerically pure form, both
configurations are accessible by simply switching the chiral
auxiliary for its enantiomer, and the use of dangerous or toxic
reagents is not necessary.
application of these methods to the preparative scale synthesis
of two valuable β3-amino acids that we have not previously
reported is also described.
The successful development of this approach for β3-amino
acid synthesis required a large-scale preparation of L- and D-
gulose-derived chiral auxiliaries and a practical synthesis of
cyclohexylidene acrylate 5. These needs led us to re-evaluate
the chiral auxiliary synthesis, which was difficult to execute on
large scale because of the physical properties of the products,
use of expensive reagents, and requirement for chromatog-
raphy. The acrylate, as well, was both cumbersome to prepare
on scale and subject to degradation during storage. In this
report we provide the full details of our optimized synthesis of
the chiral auxiliaries 1 and an alternative approach to acrylate 5
featuring an in situ generation from chloride 10. The
RESULTS AND DISCUSSION
■
Improved Synthesis of Gulose-Derived Auxiliaries.
Our approach to the synthesis of β3-amino acids relies on a
diastereoselective [3 + 2] nitrone cycloaddition to introduce
and control the stereochemistry of the resulting isoxazolidine.
Our early work in this area was based on the 1977 report by
Vasella and co-workers using the easily prepared D-mannose-
derived auxiliaries.14 This chiral auxiliary, however, had three
insurmountable problems: (1) it gave the pseudounnatural
configuration in the resulting β3-amino acids and the
enantiomer, L-mannose, is available only in small quantities
and at great cost; (2) many of the cycloadditions gave only
688
dx.doi.org/10.1021/op300069n | Org. Process Res. Dev. 2012, 16, 687−696

Organic Process Research & Development
Article
Scheme 2. Preparation of the Chiral Auxiliary by Rohloff (a) and Improved Procedure (b)
Scheme 3. First Approaches to the Required Acrylate via Horner−Wadsworth−Emmons Reaction (a) or Using a Sulfone As
Leaving Group (b)
modest diastereoselectivity; and (3) enrichement of the major
diastereoisomer by recrystallization was difficult.
(+)-3, reduced to the lactol (+)-4 using DIBAL-H, and
converted to a glassy 6:5 E/Z mixture of gulose-derived oximes
in 42% overall yield (Scheme 2a). The use of these gulose-
derived auxiliaries in the nitrone cycloaddition gave excellent
results. Not only were higher diastereoselectivities obtained in
comparison with the D-mannose-derived auxiliaries (even in the
cases of less bulky, aliphatic side chains), but the cycloadducts
were more easily crystallized, facilitating the enrichment and
purification of the major diastereomer.
The synthesis of 1 up to 100 g scale posed several problems:
(1) it involved an expensive DIBAL-H reduction, requiring
cryogenic temperatures, and a time-consuming and solvent-
intensive workup; (2) the final product, a glassy mixture of E
and Z oximes, was difficult to both purify and handle; and (3)
column chromatography was required and residual solvent was
difficult to remove from the resulting syrup.
In the course of their total synthesis of (+)-negamycin,
Kibayashi and co-workers have reported the use of D-gulose-
derived oxime as a convenient surrogate for L-mannose.15 We
became interested in this approach because both D- and L-
gulose are commercially available at a reasonable price in the
form of their gulonic acid-1,4-lactone, which could be
converted to the chiral auxiliaries in three steps.16 Our first
investigations on the use of these chiral auxiliaries in the nitrone
cycloaddition were plagued by difficulties in the crystallization
of both starting chiral auxiliary and obtained isoxazolidines. We
reasoned that the cyclohexylidene protecting groups employed
by Kibayashi could disminish the ability of the isoxazolidine to
crystallize, and by analogy to the mannose-derived products, we
anticipated that switching the cyclohexylidene protecting
groups for acetonide would improve their properties. Following
procedures by Rohloff and Schwartz17 and Tamura and
Sakamoto,14f we prepared 2,3:5,6-O-diisopropylidene-gulose
oxime ((+)-1) in three steps: D-gulonic acid-1,4-lactone
((+)-2) was protected as its diisopropylidene derivative
To improve the synthesis of the chiral auxiliary, we first
developed a reduction of (+)-3 to (+)-4 using NaBH4 in
methanol as an inexpensive alternative to DIBAL-H. This also
simplified the reaction workup and obviated the need for
column chromatography. Additionally, the final crude glassy
689
dx.doi.org/10.1021/op300069n | Org. Process Res. Dev. 2012, 16, 687−696

Organic Process Research & Development
Article
Scheme 4. Synthesis of Acrylate Precursor 10
mixture of E/Z oximes can be converted exclusively into the Z-
isomer (+)-1 by overnight sonication in hexane, providing a
white solid powder which can be further recrystallized into a
white crystalline solid (Scheme 2b). The D-gulose auxiliary (+)-
1 and L-gulose auxiliary (−)-1 are both obtained in the same
range of yields and are stable and storable as solids for long
periods of time. This chromatography-free protocol is scalable,
practical, reliable, and commonly used in our laboratory on a
50−100 g scale.18 On scale, the use of sonication might be
problematic, but the implementation of a dynamic crystal-
lization process to enrich the mixture with the Z-isomer using a
sonication loop to control crystal growth could be envisionned
as an alternative.
Improved Synthesis of the Acrylate. Another obstacle to
the cost-effective scale-up of our approach to β3-amino acids
was the synthesis and stability of the requisite acrylate 5. In our
first generation synthesis, 5 was prepared from the correspond-
ing phosphonate19 via an Horner−Wadsworth−Emmons
reaction, but on scale it had a propensity to polymerize,
rendering its handling and storage difficult (Scheme 3a). Even
stabilized with hydroquinone, 5 could usually not be stored
more than 1 week at −20 °C. To avoid the storage and
handling of 5, we sought to identify a precursor that could be
used to generate the desired acrylate in situ.
hexanone is carried out with a catalytic amount of sulfuric acid
and should not require additional purification. It remained to
be determined, however, if the dioxolanone product 10 would
be stable to storage and amenable to an in situ generation of the
acrylate under conditions compatible with the nitrone cyclo-
addition.
The synthesis of 10 began with the oxidation of α-
chlorohydrin with nitric acid (Scheme 4). The previously
reported procedure23 was found to be convenient on small
scale (<10 g), but upon scale-up, its highly exothermic nature
and the strong gas evolution was difficult to control. A reverse
addition protocol was implemented to avoid those issues (see
the Supporting Information for a detailed procedure). After
quenching, extraction with ethyl acetate and recrystallization
provide 9 as a white crystalline solid. Using this method, the
oxidation could be routinely performed in our laboratory on a
50 g scale (450 mmol) and has been executed by our research
partners on a kilogram scale.
With this material in hand, we turned our attention to the
acetal formation. The use of 5 mol % of concentrated sulfuric
acid proved to be sufficient to catalyze the acetal formation in
refluxing toluene with removal of the water (Dean−Stark trap).
To ensure complete conversion and simple purification of the
product, we decided to use a slight excess of the hydroxy acid 9
(1.2 equiv). Once water had been distilled off (usually 10−12
h), the crude reaction mixture was neutralized and washed with
saturated aqueous sodium bicarbonate to remove the excess 9,
affording the desired acetal 10 in good yield (>80%) and in
sufficient purity to be used as such for the next step (acrylate
formation and [3 + 2] cycloaddition).
In our previous work, the cycloaddition was performed in
toluene at 110 °C. To devise a one-pot procedure, we first
screened different bases in this solvent. No reaction was
observed in toluene using triethylamine or DIPEA, even under
reflux, whereas the use of DBU led to partial cleavage of the
acetal (Table 1). Switching to a more polar solvent such as
chloroform gave superior results. Although no reaction was
observed at room temperature with triethylamine, under reflux
clean formation of the acrylate took place and no hydrolysis or
polymerization of the acrylate was observed. Having identified a
suitable base for acrylate generation, we concentrated our
efforts on finding a procedure where acrylate formation and [3
+ 2] cycloaddition could be telescoped, and which would avoid
the use of chloroform as a solvent, which is prohibitive on large
scale because of its toxicity (suspected carcinogen), environ-
mental impact, and cost. To avoid the use of molecular sieves
or other dehydrating agents (mandatory for clean nitrone
formation), we decided to screen other solvents possessing the
following properties: (1) a boiling point above 100 °C
(required for the cycloaddition with sterically hindered
On the basis of reports from Roush,20 we initially
investigated the use of sulfones as leaving groups for acrylate
formation (Scheme 3b). The requisite starting material could
be prepared by a 5-step sequence: first, the hydroxy acid 6 is
obtained via a low-yielding epoxidation of methyl acrylate,
followed by nucleophilic ring opening with magnesium
thiophenolate and final saponification of the methyl ester.
Second, the dimethyl acetal of cyclohexanone was prepared and
then transacetalized with 6. Oxidation of thioether 7 to the
corresponding sulfone with mCPBA followed by elimination
with DBU yielded the desired acrylate 5. In addition to a long
linear sequence, this protocol suffered from other notable
disadvantages such as expensive reagents (Grignard reagent),
low yields, and several chromatographic purifications. Elimi-
nation of the sulfone with DBU to generate the acrylate always
led to partial cleavage of the acetal, preventing its use as an in
situ precursor to 5.
As an alternative we sought to explore the synthesis and
reactivity of dioxolanone 10, which was known to eliminate to
the corresponding acrylate upon exposure to a mild base.21
This precursor was promising because (1) the necessary 3-
chloro-2-hydroxypropanoic acid (9) can be obtained from α-
chlorohydrin (8), an inexpensive and widely accessible starting
material;22 (2) oxidation of α-chlorohydrin can be performed
with nitric acid (another inexpensive reagent) and the product
purified by recrystallization; (3) acetal formation with cyclo-
690
dx.doi.org/10.1021/op300069n | Org. Process Res. Dev. 2012, 16, 687−696

Organic Process Research & Development
Article
were obtained with good to excellent levels of diastereose-
lectivity, and recrystallization of the crude reaction mixture
afforded the diastereomerically pure cycloadducts 11, avoiding
purification by column chromatography.
Table 1. Optimization of Acrylate Formation
This protocol has been routinely applied for cycloadditions
with many different aldehydes on both milligram and gram
scales.13 The one-pot sequential formation of the acrylate
followed by the cycloaddition proved to be superior; in the
purely one-pot procedure, some hydrolysis of the acetal was
observed, even in the presence of molecular sieves.
temp time
entry base (equiv)
solvent
(°C)
(h)
result
1
2
3
4
Et3N (2)
Et3N (2)
DIPEA (2)
DBU (2)
toluene
toluene
toluene
toluene
23
10
10
10
10
no reaction
no reaction
no reaction
Removal of the chiral auxiliary with aqueous perchloric acid
in CH3CN proceeded cleanly, affording after neutralisation
with NaHCO3 the corresponding isoxazolidines 12 in good
yield and very good purity. As reported previously, these mild
conditions are also tolerant of protected functionalised side
chains.13 The isoxazolidines can be precipitated as their HCl
salt and stored as such. Attempts to store the free isoxazolidines
led to slow decomposition in many cases. Furthermore, the
isoxazolidine·HCl salts can be used directly in the redox neutral
fragmentation. On scale, the use of perchloric acid could
become problematic because of its strong oxidizing nature.24 As
an alternative, TFA/DCM (1:1) is also effective for chiral
auxiliary removal, but these conditions are less tolerant to
protected side chains and thus less interesting from the
perspective of a flexible and general process. After evaporation
of the TFA/DCM mixture, the isoxazolidine·TFA salts
obtained were found to be less stable than their HCl
counterparts. If the isoxazolidines do not need to be stored,
the isoxazolidine free base can be directly submitted to
fragmentation to give the β3-amino acids 13 in high yields
(Scheme 5).25
Simply warming the salts in a 1:1 t-BuOH/H2O mixture at
60 °C for 24−36 h affords the corresponding enantiopure β3-
amino acids salts 13·HCl (Scheme 5). For further scale-up of
this process, the recovery of the chiral auxiliary would be an
advantage. In some cases, we have found that the cleavage of
the auxiliary can be performed using hydroxylamine buffered by
sodium acetate. These mild conditions are generally tolerant to
protected side chains and allowed in many cases >70% recovery
of the chiral auxiliary. The major limitation of this approach is
the necessary chromatographic purification to separate the free
110
110
110
no acrylate formation,
partial acetal cleavage
a
5
6
7
8
Et3N (2)
DIPEA (2)
Et3N (2)
DBU (2)
CHCl3
CHCl3
CHCl3
CHCl3
23
65
65
65
10
10
10
10
no reaction
no reaction
clean acrylate formation
acrylate formation and
partial acetal cleavage
a
9
Et3N (2)
Et3N (2)
CPME
100
100
18
18
45% starting
material:55% acrylate
10
n-propyl
acetate
clean acrylate formation,
100% conversion
a
1
After workup, cyclohexanone was observed by H NMR.
aldehydes), (2) immiscible with water (in order to use a Dean−
Stark trap to remove water), (3) capable of forming an
azeotrope with water, (4) a low heat of vaporization, (5) a
polarity sufficient for chloride elimination, (6) environmentally
more friendly, and (7) low toxicity. Cyclopentyl methyl ether
(CPME) and n-propyl acetate fullfilled these criteria and were
thus tested in this transformation (Table 1, entries 9 and 10).
Both solvents exhibited clean and exclusive formation of the
acrylate upon heating at 100 °C, but n-propyl acetate proved to
be superior in terms of conversion. The telescoped procedure
was thus performed as follows: the acrylate precursor was
heated under reflux in the presence of 2 equiv of triethylamine
in n-propylacetate for 18 h, the mixture was cooled to 23 °C,
the aldehyde and the oxime chiral auxiliary were added, and the
mixture was heated under reflux with removal of the water
(Dean−Stark trap). After workup, the desired cycloadducts
Scheme 5. In Situ Acrylate Generation, [3 + 2] Nitrone Cycloaddition, Chiral Auxiliary Cleavage, and Redox Neutral
Fragmentation Sequence for the Synthesis of β3-Amino Acids
691
dx.doi.org/10.1021/op300069n | Org. Process Res. Dev. 2012, 16, 687−696

Organic Process Research & Development
Article
Scheme 6. Preparation of (R)-3-Amino-3-cyclohexylpropanoic Acid Hydrochloride (13a·HCl) and (S)-3-Amino-5,5-
dimethylhexanoic Acid (13b)
isoxazolidine from the recovered chiral auxiliary. For our
purpose, and regarding the easy access to the chiral auxiliary,
this option was not investigated but could be implemented in
specific cases for large scale applications.
is highly suitable for the multigram-scale synthesis of a wide
range of β3-amino acids possessing unnatural side chains, has a
potential for further scale-up, and offers a good alternative to
the hazardous and expensive Arndt−Eistert homologation of
enantiopure α-amino acids. Its utility has been demonstrated
with the synthesis of (R)-3-amino-3-cyclohexylpropanoic acid
hydrochloride and (S)-3-amino-5,5-dimethylhexanoic acid.
These revised protocols will also be useful for the other β3-
amino acids that we have prepared using this chiral auxiliary
and nitrone cycloaddition.13
To demonstrate the utility and reliability of our method, we
have applied this procedure to the synthesis of two valuable β3-
amino acids possessing unnatural side chains. (Scheme 6)
The one-pot acrylate formation/cycloaddition sequence
using the acrylate precursor 10 and cyclohexylcarboxaldehyde
furnished crude 11a in 24 h (dr 91:9:0:0), which was
recrystallized from hexane to afford diastereomerically pure
11a (as determined by 1H NMR) in 64% yield. Cleavage of the
chiral auxiliary with perchloric acid in acetonitrile afforded the
corresponding free isoxazolidine 12a, which was precipitated as
its HCl salt 12a·HCl in 82% yield. (R)-3-Amino-3-cyclo-
hexylpropanoic acid hydrochloride (13a·HCl) was obtained
after fragmentation in 1:1 t-BuOH/H2O at 60 °C for 36 h and
simple trituration with EtOAc in 90% yield and 99% ee (47%
overall from 10). Crude 11b was obtained by the one-pot
acrylate formation/cycloaddition sequence between the acrylate
precursor 10 and 3,3-dimethylbutanal for 48 h (dr 77:13:10:0,
the longer reaction time is attributed to the increased steric
hindrance of the aldehyde). Two successive recrystallizations
from 9:1 hexane/EtOAc were necessary to give diastereomeri-
EXPERIMENTAL SECTION
■
General. Chemicals were purchased from Acros, Aldrich, or
BioBlocks Inc. and used without further purification unless
otherwise stated. Flash column chromatography was performed
on Silicycle Silica Flash F60 (230−400 Mesh) using a forced
flow of 0.3−0.5 bar. Thin layer chromatography was performed
on aluminium backed plates precoated with silica gel (Merck,
Silica Gel 60 F254) and were visualized by fluorescence
quenching under UV light or by staining with phosphomolyb-
1
dic acid. H NMR and 13C NMR were measured on VARIAN
Mercury 300 MHz, 75 MHz or Bruker Avance 400 MHz, 100
MHz instruments, respectively. Chemical shifts are expressed in
parts per million (ppm) downfield from residual solvent peaks,
and coupling constants are reported in hertz (Hz). Splitting
patterns are indicated as follows: br, broad; s, singlet; d,
doublet; dd, doublet of doublet; t, triplet; q, quartet; m,
multiplet. High-resolution mass spectrometric measurements
were performed by the mass spectrometry service of the LOC
at the ETHZ on Agilent 1200 (LC) and Bruker maXis for ESI-
Q-TOF instruments. Optical rotation measurements were
performed using a Jasco P-2000 Polarimeter, operating at the
sodium D line with a 100 mm path length cell, and were
1
cally pure 11b (as determined by H NMR) in 52% yield.
Cleavage of the chiral auxiliary with perchloric acid in
acetonitrile furnished free isoxazolidine 12b,25 which was
directly submitted to fragmentation in 1:1 t-BuOH/H2O at
60 °C for 36 h and trituration with EtOAc, affording (S)-3-
amino-5,5-dimethylhexanoic acid (13b) in 86% yield and 99%
ee (44% overall from 10).
CONCLUSION
■
We have developed a flexible, enantioselective, and chromatog-
raphy-free protocol for the synthesis of β3-amino acids. This
method features the in situ generation of an acrylate from a
stable and storable precursor, a diastereoselective nitrone [3 +
2] cycloaddition, and a redox-neutral fragmentation. We have
also provided a scalable, chromatography-free synthesis of the
chiral auxiliary from either enantiomer of commerically
available gulonic acid γ-lactone. We believe that this approach
reported as [α]T (concentration (g/100 mL), solvent).
D
Infrared (IR) data was obtained on a JASCO FT-IR-4100
spectrometer with only major peaks being reported. SFC
(supercritical fluid chromatography) was performed on a
JASCO liquid chromatography unit. Daicel Chiralcel columns
(0.46 cm × 25 cm) were used. See the Supporting Information
for details of chromatographic conditions. Melting points were
692
dx.doi.org/10.1021/op300069n | Org. Process Res. Dev. 2012, 16, 687−696

Organic Process Research & Development
Article
measured on an Electrothermal Mel-Temp melting point
apparatus and are uncorrected.
Hz, 1H), 4.12 (dd, J = 3.5, 8.5 Hz, 1H), 3.72 (dd, J = 7.5, 8.5
Hz, 1H), 3.06 (d, J = 2.0 Hz, 1H), 1.44 (s, 6H), 1.38 (s, 3H),
1.28 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 113.0, 109.9,
101.5, 85.8, 82.4, 80.0, 75.7, 66.2, 26.9, 26.1, 25.6, 24.9.
2,3:5,6-O-Diisopropylidene-D-gulono-lactone ((+)-3).
A 1 L round-bottomed flask equipped with a Teflon-coated
magnetic stir bar is charged with acetone (560.0 mL) and
dimethoxypropane (140.0 mL; 1.130 mol; 4.00 equiv). D-
Gulonic acid-1,4-lactone (51.1 g, 0.287 mol; 1.00 equiv) and p-
toluenesulfonic acid monohydrate (5.4 g; 28.8 mmol; 0.10
equiv) are added. After being stirred at 23 °C for 24 h, the
reaction is quenched by adding NaHCO3 (7.2 g; 86.2 mmol;
0.30 equiv) and stirred for an additional 1 h (Caution: CO2
releases). The solvent is concentrated by rotary evaporation;
the resultant brown solid is taken up in water (500 mL) and
extracted with EtOAc (3 × 300 mL). The combined organic
layers are washed with saturated brine (500 mL) and dried with
Na2SO4. The solvent is concentrated by rotary evaporation to
yield an off-white solid (74.3 g, 99% crude yield). The crude
solid is dissolved in a minimal amount of refluxing EtOAc (ca.
65 mL), and hexane (ca. 15 mL) is added dropwise until a
cloudy solution is obtained. An additional portion of EtOAc
(ca. 15 mL) is added until the solution turns clear again. The
solution is cooled to 23 °C while the crystals are formed and
placed at 4 °C overnight. The crystals are collected and washed
with a chilled mixture of EtOAc and hexane (1:1) to afford the
title compound (45.1 g) as white needle-shaped crystals. The
filtrate is concentrated and recrystallized in a similar manner
from EtOAc (ca. 20 mL) and hexane (40 mL) to provide
additional product (6.3 g) as white needle-shaped crystals, for a
2,3:5,6-O-Diisopropylidene-D-gulose Oxime ((+)-1).
Caution! Nonaqueous mixtures of hydroxylamine can be
explosive. A 1 L round-bottomed flask equipped with a Teflon-
coated magnetic stir bar is charged with absolute EtOH (250.0
mL). Crude (+)-4 (44.1 g; 0.170 mol; 1.00 equiv) is added, and
the flask is placed in an oil bath. While the mixture is stirred,
hydroxylamine hydrochloride (17.7 g; 0.255 mol; 1.50 equiv)
and sodium acetate trihydrate (34.7 g; 0.255 mol; 1.50 equiv)
are added in one portion. The suspension is heated to 70 °C
and stirred for 2 h. The solution is cooled to 23 °C, and 600
mL of saturated aqueous NaHCO3 is added over 1 h (Caution:
CO2 releases, upon scale-up the use of phosphate buffers to
quench the reaction might be preferred.) Ethanol is removed
by rotary evaporation, and the residue is extracted with EtOAc
(3 × 400 mL). The combined organic layers are washed with
brine, dried (Na2SO4), and concentrated by rotary evaporation.
The crude oil is treated with a minimal amount of hexane (ca. 4
mL) and then sonicated overnight, turning into a white solid.
The trace amount of hexanes is removed under high vacuum to
afford (+)-1 as a white powder (46.3 g, 99% crude yield) that is
exclusively Z-isomer. The crude solid (46.3 g) is dissolved in a
minimal amount of refluxing EtOAc (ca. 30 mL), and hexane
(ca. 90 mL) is added dropwise until a cloudy solution is
obtained. The solution is cooled to room temperature while the
crystals are formed and placed at 4 °C overnight. The crystals
are collected and washed with a cold mixture of 1:2 EtOAc/
hexane to afford compound (+)-1 (41.3 g, 88% yield) as white
crystals. Mp = 117−119 °C; [α]21D = +172.8 (c 1.1 in CHCl3).
This compound is in equilibrium in solution with its closed
combined mass of 51.4 g (69% yield). Mp = 150−152 °C;
1
[α]21 = −78.4 (c 0.99 in CHCl3); H NMR (400 MHz,
D
CDCl3) δ 4.84 (d, J = 5.5 Hz, 1H), 4.74 (dd, J = 3.0, 5.5 Hz,
1H), 4.46−4.41 (m, 2H), 4.21 (dd, J = 6.5, 9.0 Hz, 1H), 3.82
(dd, J = 6.5, 9.0 Hz, 1H), 1.47 (s, 3H), 1.46 (s, 3H), 1.39 (s,
3H), 1.37 (s, 3H); 13C NMR (100 MHz, CDCl3) δ: 173.1,
114.8, 110.6, 81.0, 76.2, 75.9, 75.4, 65.3, 26.9, 26.8, 26.0, 25.3.
2,3:5,6-O-Diisopropylidene-D-gulofuranose ((+)-4).
Caution! This procedure should be carried out in an
efficient fume hood due to the evolution of hydrogen gas
during the reaction. A 1 L round-bottomed flask equipped
with a Teflon-coated magnetic stir bar and an internal
thermometer is charged with methanol (800.0 mL). (+)-3
(51.0 g, 0.198 mol, 1.00 equiv) is added and the flask is placed
in a sodium chloride/ice bath. When the internal temperature
reaches −5 °C, sodium borohydride (3.75 g, 99.0 mmol, 0.50
equiv) is added in 6 portions each, 15 min apart. After addition
of sodium borohydride, the mixture is stirred for an additional
30 min until TLC shows the completion of the reaction. The
reaction is quenched by the addition of water (20 mL), and
stirred for 30 min until no bubble releases, then allowed to
warm to room temperature. Methanol is removed by rotary
evaporation; the crude residue is taken up in water (500 mL),
and extracted with EtOAc (3 × 400 mL). The combined
organic layers are washed with saturated brine and dried with
Na2SO4. The solvent is concentrated by rotary evaporation to
yield the title compound 3 as a white solid (45.6 g, 89% crude
yield) which is exclusively the β−isomer. The crude product is
sufficiently pure to be used directly for the next step. An
analytical sample is obtained by recrystallization (1.5 g crude
product is recrystallized from EtOAc (ca. 5 mL) and hexane
(ca. 15 mL) to give 1.1 g of pure product). Mp = 119−120 °C;
[α]21D = −2.7 (c 0.97 in CHCl3); 1H NMR (400 MHz, CDCl3)
δ 5.45 (d, J = 2.0 Hz, 1H), 4.69 (dd, J = 4.0, 6.0 Hz, 1H), 4.62
(d, J = 6.0 Hz, 1H), 4.38−4.34 (m, 1H), 4.21 (dd, J = 6.5, 8.5
1
hydroxylamine form. H NMR (400 MHz, CDCl3) δ 9.18 (s,
1H), 7.13 (d, J = 4.0 Hz, 1H), 5.25 (dd, J = 4.0, 7.5 Hz, 1H),
4.32−4.23 (m, 3H), 4.12 (dd, J = 6.5, 8.5 Hz, 1H), 3.58 (t, J =
8.5 Hz, 1H), 3.39 (t, J = 6.5 Hz, 1H), 1.56 (s, 3H), 1.42 (s,
3H), 1.40 (s, 3H), 1.35 (s, 3H); 13C NMR (100 MHz, CDCl3)
δ 152.2, 110.2, 109.8, 78.6, 78.1, 72.8, 70.9, 66.4, 26.5, 26.3,
25.6, 24.8.
3-Chloro-2-hydroxypropanoic Acid [1713-85-5] (9).
Caution! Nitrogen oxides are highly toxic. This procedure
should be carried out in a well-ventilated hood. A 1 L two-
necked round-bottomed flask equipped with a Teflon-coated
magnetic stir bar, a reflux condenser, and an addition funnel is
charged with 3-chloro-1,2-propanediol (50.0 g; 0.452 mol; 1.00
equiv). The top of the condenser is fitted with a vacuum
adapter and some tubing to ensure that the gases formed during
the reaction will be bubbled through 2.0 L of a 3.0 N NaOH
aqueous solution cooled at 0 °C. The addition funnel is charged
with nitric acid (155.0 mL of a 65% aqueous solution) and
sealed with a septum. The flask and its contents are heated to
80 °C with an oil bath, and a small portion (10−15 mL) of
concentrated nitric acid is added with vigorous stirring to
initiate the reaction. As the reaction begins, a large amount of
dark orange-brown vapors appears. Stirring is continued, and
the addition of nitric acid is resumed after the initial exothermic
reaction subsides (usually within 15 min). The remaining nitric
acid is added in 20−30-mL aliquots over 30−60 min so that
nitrogen oxides are continually evolved but the reaction does
not become violent. After completion of the addition, the
temperature of the oil bath is increased to 105 °C, and the
mixture is stirred for 3 h. During this time, nitrogen is flushed
693
dx.doi.org/10.1021/op300069n | Org. Process Res. Dev. 2012, 16, 687−696

Organic Process Research & Development
Article
in the reaction mixture via a needle through the septum of the
addition funnel to ensure that all acidic vapors are bubbled
through the basic NaOH solution. Then the reaction mixture is
cooled to 23 °C, and 35.0 g of NaHCO3 dissolved in 300 mL of
H2O is slowly added to partially neutralize the nitric acid. The
aqueous layer is saturated with NaCl and extracted with EtOAc
(7 × 250 mL). The combined organic layers are dried with
Na2SO4. EtOAc is evaporated by rotary evaporation, the
temperature of the water bath is increased to 60 °C, and the
remaining nitric acid is evaporated to yield a yellow solid. The
crude solid is dissolved in a minimal amount of refluxing
chloroform (ca. 200 mL, 5 mL per gram). At this point, some
solid may not dissolve in chloroform, and in this case the hot
solution is filtered. The filtrate is cooled to room temperature
while the crystals are formed and placed at 4 °C overnight. The
crystals are collected and washed with cold chloroform to afford
9 (25.3 g; 45% yield) as white needle-shaped crystals. Mp 75
°C; IR (neat, cm−1) ν 3443, 3405, 1915, 1613, 1224, 1737,
132 °C; [α]26 = −16.4 (c 1.0 in CHCl3); IR (neat, cm−1) ν
D
2937, 1799, 1449, 1371, 1249, 1207, 1116, 1072, 847; 1H NMR
(400 MHz, CDCl3) δ 4.87 (d, J = 6.1 Hz, 1H), 4.69−4.62 (m,
2H), 4.35 (dt, J = 8.4, 6.6 Hz, 1H), 4.20 (dd, J = 8.5, 6.9 Hz,
1H), 4.05 (dd, J = 8.5, 3.8 Hz, 1H), 3.73 (dd, J = 8.6, 6.4 Hz,
1H), 3.56−3.46 (m, 1H), 2.76 (dd, J = 14.0, 8.0 Hz, 1H), 2.36
(d, J = 15.0 Hz, 1H), 2.09−0.95 (m, 33H); 13C NMR (100
MHz, CDCl3) δ 169.6, 112.9, 111.5, 109.7, 105.9, 96.7, 84.3,
84.2, 80.2, 75.6, 66.0, 65.8, 38.6, 37.6, 36.9, 36.4, 31.1, 30.0,
26.6, 26.4, 26.1, 25.8, 25.6, 25.2, 24.9, 24.3, 23.0, 22.9; HRMS
(m/z, ES) calcd for C28H44NO9 538.3011, found 538.3009.
11b. Obtained according to the general procedure from 10.0
g of 10 (48.8 mmol) with a cycloaddition time of 48 h. Two
successive recrystallizations from 9:1 hexane/EtOAc (100 mL)
afforded 11b as a white solid (13.3 g, 52% yield, single
cycloadduct). Mp = 132−133 °C; [α]26 = −7.4 (c 0.9 in
D
CHCl3); IR (neat, cm−1) ν 2944, 1799, 1452, 1374, 1200, 1155,
1071, 1033, 990, 923, 848; 1H NMR (400 MHz, CDCl3) δ 4.87
(d, J = 6.1 Hz, 1H), 4.69 (s, 1H), 4.65 (dd, J = 6.0, 4.0 Hz, 1H),
4.37 (dd, J = 15.3, 6.9 Hz, 1H), 4.21 (dd, J = 8.4, 6.9 Hz, 1H),
4.06 (dd, J = 8.4, 3.9 Hz, 1H), 3.91 (q, J = 5.8 Hz, 1H), 3.70
(dd, J = 8.4, 7.0 Hz, 1H), 2.96 (dd, J = 13.6, 7.5 Hz, 1H), 2.12
(dd, J = 13.6, 1.7 Hz, 1H), 1.98−1.60 (m, 9H), 1.52−1.42 (m,
1
1427, 1100, 713; H NMR (CD3OD, 400 MHz) δ 4.45 (t, J =
4.2 Hz, 1H), 3.84 (d, J = 4.4 Hz, 2H); 13C NMR (CD3OD, 100
MHz) δ 172.8, 70.4, 46.2; HRMS (m/z, ES) calcd for
C3H4ClO3 122.9854, found 122.9860.
5-Chloromethyl-2,2-pentamethylene-1,3-dioxolan-4-
one (10). A 500 mL round-bottomed flask equipped with a
Teflon-coated magnetic stir bar is charged with 9 (10.0 g; 80.6
mmol; 1.20 equiv), cyclohexanone (7.1 g; 72.6 mmol; 1.00
equiv), and toluene (134.0 mL). Then concentrated sulfuric
acid (200 μL; 3.8 mmol; 0.05 equiv) is added, and the flask
equipped with a Dean−Stark trap and a reflux condenser. The
flask and its contents are heated under reflux for 20 h while
water is removed by azeotropic distillation. The reaction
mixture is cooled to 23 °C, and the solvent is evaporated by
rotary evaporation. The crude oil is dissolved in EtOAc (300
mL), and the organic layer is washed with a saturated aqueous
NaHCO3 solution (2 × 200 mL) and saturated brine (200 mL)
and dried with Na2SO4. The solvent is removed by rotary
evaporation to afford 10 (12.6 g, 76% yield) as pale brown oil.
IR (neat, cm−1) ν 2938, 2862, 1791, 1602, 1556, 1450, 1371,
6H), 1.39 (s, 3H), 1.38 (s, 3H), 1.29 (s, 3H), 0.97 (s, 9H); 13
C
NMR (100 MHz, CDCl3) δ 169.5, 112.8, 111.7, 109.7, 105.7,
97.2, 84.6, 84.3, 80.2, 75.6, 66.0, 58.0, 47.8, 42.8, 37.6, 36.4,
30.5, 29.8 (3C), 26.6, 26.1, 25.2, 25.0, 24.3, 23.0, 22.9; HRMS
(m/z, ES) calcd for C27H44NO9 ([M + H]+) 526.3011, found
526.3005.
General Procedure for Chiral Auxiliary Cleavage
Affording the Stable Isoxazolidine Hydrochloride Salt
12·HCl. Caution! Perchloric acid is a strong oxidizing agent
that reacts violently with metals and is potentially explosive
at high concentrations. A round-bottomed flask equipped with
a Teflon-coated magnetic stir bar is charged with 11 (1.00
equiv) and acetonitrile (0.1 M). Perchloric acid (70% w/w
solution in H2O, 2.50 equiv) is added, and the flask is stirred at
23 °C for 4−8 h. The reaction mixture is quenched by the
careful addition of saturated aqueous NaHCO3 (Caution: CO2
releases), and acetonitrile is evaporated under vacuum. The
aqueous layer is extracted with EtOAc, and the organic layer is
washed with saturated brine and dried over Na2SO4. The
solvent is evaporated by rotary evaporation to afford a pale
yellow oil. This oil is dissolved in MTBE (0.1 M) and cooled at
0 °C, and a solution of HCl in diethyl ether (2 M solution, 1.10
equiv) is slowly added over 1 h. After completion of the
addition, the flask is placed in the refrigerator at 4 °C overnight.
The solid is collected by filtration and washed with MTBE to
afford 12·HCl.
1
1212, 1155, 1117, 1072, 940, 909; H NMR (CDCl3, 400
MHz) δ 4.73 (t, J = 3.5 Hz, 1H), 3.85 (d, J = 3.3 Hz, 2H),
2.01−1.81 (m, 3H), 1.81−1.64 (m, 5H), 1.58−1.39 (m, 2H);
13C NMR (CDCl3, 100 MHz) δ 169.7, 112.5, 74.1, 42.7, 36.1,
36.0, 24.4, 23.0, 22.9; HRMS (m/z, ES) calcd for
C9H13ClNaO3 227.0445, found 227.0444.
General Procedure for the Cycloaddition. A round-
bottomed flask equipped with a Teflon-coated magnetic stir bar
and a reflux condenser is charged with 10 (1.00 equiv) and dry
n-propylacetate (0.5 M). Triethylamine (2.00 equiv) is added,
and the flask is heated under reflux for 18 h. The reaction
mixture is cooled to 23 °C, and (+)-1 (1.00 equiv) and the
requisite aldehyde (1.00 equiv) are added. A Dean−Stark trap
is fitted on the flask, and the reaction mixture heated under
reflux for an additional 24−48 h. The course of the reaction is
monitored by TLC until complete disappearence of the
acrylate. The reaction mixture is cooled to 23 °C and diluted
with EtOAc. The organic layer is washed with aqueous 1 N
HCl and brine, dried over Na2SO4, and evaporated to dryness.
The crude product is recrystallized 2 times to afford the pure
cycloadduct 11.
12a·HCl. Obtained according to the general procedure from
16.0 g of 11a (29.7 mmol) with a reaction time of 6 h. White
solid (8.1 g, 82% yield). Mp 141−142 °C; [α]26D = +29.1 (c 0.9
in CHCl3); IR (neat, cm−1) ν 2921, 2853, 2415, 1808, 1442,
1
1284, 1244, 1194, 1104, 1085, 942; H NMR (400 MHz,
CD3OD) δ 3.89 (dd, J = 17.6, 10.1 Hz, 1H), 3.18 (dd, J = 14.4,
7.5 Hz, 1H), 2.56 (dd, J = 14.4, 10.5 Hz, 1H), 1.97−1.01 (m,
21H); 13C NMR (100 MHz, CD3OD) δ 165.2, 113.2, 107.0,
66.8, 38.5, 38.4, 36.8, 35.2, 30.2, 29.4, 25.2, 25.0, 24.7, 23.6,
22.6, 22.5; HRMS (m/z, ES) calcd for C16H26NO4 296.1856,
found 296.1861.
General Procedure for the Redox Neutral Fragmenta-
tion Affording the β3-Amino Acid HCl Salt 13·HCl. A
round-bottomed flask equipped with a Teflon-coated magnetic
11a. Obtained according to the general procedure 1 from
10.0 g of 10 (48.8 mmol) with a cycloaddition time of 24 h.
Recrystallization from pure hexane (100 mL) afforded 11a as a
white solid (16.8 g, 64% yield, single cycloadduct). Mp = 131−
694
dx.doi.org/10.1021/op300069n | Org. Process Res. Dev. 2012, 16, 687−696

Organic Process Research & Development
Article
stir bar is charged with the isoxazolidine hydrochloride salt
(1.00 equiv) and a 1:1 mixture of t-BuOH/H2O (0.1 M). The
reaction mixture is stirred at 60 °C for 36 h. The reaction
mixture is evaporated to dryness under vacuum to afford a
solid. The solid is triturated with acetone, collected by filtration,
and washed with acetone to afford 13·HCl.
ACKNOWLEDGMENTS
■
This work was supported by the David and Lucille Packard
Foundation (fellowship to J.W.B), the Arnold and Mable
Beckman Foundation, and ETH-Zurich. We thank Dr. Stefan
̈
Abele (Actelion Pharmaceuticals) for helpful suggestions.
(R)-3-Amino-3-cyclohexylpropanoic Acid Hydrochlor-
ide 13a·HCl. Obtained according to the general procedure
from 5.5 g of 12a·HCl (16.7 mmol) with a reaction time of 36
REFERENCES
■
(1) (a) Cardillo, G.; Tomasini, C. Chem. Soc. Rev. 1996, 25, 117−128.
(b) Steer, D. L.; Lew, R. A.; Perlmutter, P.; Smith, A. I.; Aguilar, M. I.
Curr. Med. Chem. 2002, 9, 811−822.
h. White solid (3.1 g, 90% yield). Mp 231−232 °C;26 [α]26
=
D
+49.3 (c 0.1 in H2O); IR (neat, cm−1) ν 2917, 2852, 1709,
(2) Hansen, K. B.; Balsells, J.; Dreher, S.; Hsiao, Y.; Kubryk, M.;
Palucki, M.; Rivera, N.; Steinhuebel, D.; Armstrong, J. D.; Askin, D.;
Grabowski, E. J. J. Org. Process Res. Dev. 2005, 9, 634−639.
(3) Sakai, T.; Kawamoto, Y.; Tomioka, K. J. Org. Chem. 2006, 71,
4706−4709.
1
1586, 1505, 1418, 1386, 1236, 1193, 1164; H NMR (400
MHz, DMSO-d6) δ 12.71 (brs, 1H), 8.02 (brs, 2H), 3.25 (dd, J
= 11.6, 5.6 Hz, 1H), 2.65 (dd, J = 17.3, 5.2 Hz, 1H), 2.53 (dd, J
= 17.3, 7.2 Hz, 1H), 1.77−1.53 (m, 6H), 1.25−0.93 (m, 5H);
13C NMR (100 MHz, DMSO-d6) δ 172.4, 52.3, 34.7, 28.4
(2C), 28.0, 26.0, 25.9 (2C); HRMS (m/z, ES) calcd for
C9H18NO2 ([M − Cl]+) 172.1332, found 172.1329.
(4) (a) Seebach, D.; L. Matthews, J. Chem. Commun. 1997, 21,
2015−2022. (b) Cheng, R. P.; Gellman, S. H.; DeGrado, W. F. Chem.
Rev. 2001, 101, 3219−3232.
(5) (a) Qiu, J. X.; Petersson, E. J.; Matthews, E. E.; Schepartz, A. J.
Am. Chem. Soc. 2006, 128, 11338−11339. (b) Goodman, C. M.; Choi,
S.; Shandler, S.; DeGrado, W. F. Nat. Chem. Biol. 2007, 3, 252−262.
(6) For selected examples, see: (a) Lubell, W. D.; Kitamura, M.;
Noyori, R. Tetrahedron: Asymmetry 1991, 2, 543−554. (b) Hsiao, Y.;
Rivera, N. R.; Rosner, T.; Krska, S. W.; Njolito, E.; Wang, F.; Sun, Y.
K.; Armstrong, J. D.; Grabowski, E. J. J.; Tillyer, R. D.; Spindler, F.;
Malan, C. J. Am. Chem. Soc. 2004, 126, 9918−9919. (c) Zhou, Y.-G.;
Tang, W.; Wang, W.-B.; Li, W.; Zhang, X. J. Am. Chem. Soc. 2002, 124,
4952−4953. (d) Song, J.; Wang, Y.; Deng, L. J. Am. Chem. Soc. 2006,
128, 6048−6049. (e) Palomo, C.; Oiarbide, M.; Halder, R.; Kelso, M.;
Gomez-Bengoa, E.; Garcia, J. S. M. J. Am. Chem. Soc. 2004, 126, 9188−
9189. (f) Chi, Y.; English, E. P.; Pomerantz, W. C.; Horne, W. S.;
Joyce, L. A.; Alexander, L. R.; Fleming, W. S.; Hopkins, E. A.; Gellman,
S. H. J. Am. Chem. Soc. 2007, 129, 6050−6055. (g) Duursma, A.;
Minnaard, A. J.; Feringa, B. L. J. Am. Chem. Soc. 2003, 125, 3700−
3701. (h) Morimoto, H.; Lu, G.; Aoyama, N.; Matsunaga, S.;
Shibasaki, M. J. Am. Chem. Soc. 2007, 129, 9588−9589. (i) Yang, J. W.;
Stadler, M.; List, B. Angew. Chem., Int. Ed. 2007, 46, 609−611.
(j) Matsumura, K.; Zhang, X.; Hori, K.; Murayama, T.; Ohmiya, T.;
Shimizu, H.; Saito, T.; Sayo, N. Org. Process Res. Dev. 2011, 15, 1130−
1137.
General Procedure for Chiral Auxiliary Cleavage and
Redox Neutral Fragmentation Affording the β3-Amino
Acid 13. A round-bottomed flask equipped with a Teflon-
coated magnetic stir bar is charged with 11 (1.00 equiv) and
acetonitrile (0.1 M). Perchloric acid (70% w/w solution in
H2O, 2.5 equiv) is added, and the flask is stirred at 23 °C for
4−8 h. The reaction mixture is quenched by the careful
addition of saturated aqueous NaHCO3, and acetonitrile is
evaporated under vacuum. The aqueous layer is extracted with
EtOAc, and the organic layer is washed with saturated brine
and dried over Na2SO4. The solvent is evaporated by rotary
evaporation to afford an oil. A round-bottomed flask equipped
with a Teflon-coated magnetic stir bar is charged with this oil
and a 1:1 mixture of t-BuOH/H2O (0.1 M). The reaction
mixture is stirred at 60 °C for 36 h. The reaction mixture is
evaporated to dryness under vacuum to afford a solid. The solid
is triturated with EtOAc, collected by filtration, and washed
with EtOAc to afford 13.
(S)-3-Amino-5,5-dimethylhexanoic Acid 13b. Obtained
according to the general procedure from 13.0 g of 11b (24.7
mmol) with a reaction time of 36 h. White solid (3.4 g, 86%
(7) Seebach, D.; Overhand, M.; Kuhnle, F. N. M.; Martinoni, B.;
̈
Oberer, L.; Hommel, U.; Widmer, H. Helv. Chim. Acta 1996, 79, 913−
yield). Mp 203−204 °C; [α]26 = +27.5 (c 1.1 in H2O); IR
941.
D
(neat, cm−1) ν 2954, 1645, 1544, 1466, 1393, 1309, 1251, 1213,
(8) Shimizu, H.; Nagasaki, I.; Matsumura, K.; Sayo, N.; Saito, T. Acc.
Chem. Res. 2007, 40, 1385−1393.
(9) For reviews see: (a) Juaristi, E.; Soloshonok, V. Enantioselective
Synthesis of β-Amino Acids, 2nd ed.; Wiley: New York, 2005. (b) Liu,
M.; Sibi, M. P. Tetrahedron 2002, 58, 7991−8035.
(10) For a recent example, see: Grayson, J. I.; Roos, J.; Osswald, S.
Org. Process Res. Dev. 2011, 15, 1201−1206.
(11) Savile, C. K.; Janey, J. M.; Mundorff, E. C.; Moore, J. C.; Tam,
S.; Jarvis, W. R.; Colbeck, J. C.; Krebber, A.; Fleitz, F. J.; Brands, J.;
Devine, P. N.; Huisman, G. W.; Hughes, G. J. Science 2010, 329, 305−
309.
1
1150, 1089; H NMR (400 MHz, CD3OD) δ 3.48−3.37 (m,
1H), 2.53 (dd, J = 16.9, 3.6 Hz, 1H), 2.31 (dd, J = 16.9, 9.4 Hz,
1H), 1.62−1.45 (m, 2H), 0.99 (s, 9H); 13C NMR (100 MHz,
CD3OD) δ 176.4, 46.7, 46.5, 39.8, 29.6, 28.6 (3C); HRMS (m/
z, ES) calcd for C8H18NO2 ([M + H]+) 160.1332, found
160.1331.
ASSOCIATED CONTENT
■
(12) Lelais, G.; Seebach, D. Biopolymers 2004, 76, 206−243.
(13) Yu, S. Y.; Ishida, H.; Juarez-Garcia, M. E; Bode, J. W. Chem. Sci.
2010, 1, 637−641.
S
* Supporting Information
Copies of 1H and 13C NMR spectra for all synthesized
compounds. Procedures and SFC spectra for ee determination
of the final β3-amino acids. This material is available free of
charge via the Internet at http://pubs.acs.org.
(14) (a) Vasella, A. Helv. Chim. Acta 1977, 60, 1273−1295.
(b) Huber, R.; Vasella, A. Tetrahedron 1990, 46, 33−58.
For
selected applications of carbohydrate-derived chiral auxiliaries for the
synthesis of hydroxylamines and isoxazolidines, see: (c) Basha, A.;
Henry, R.; McLaughlin, M. A.; Ratajczyk, J. D.; Wittenberger, S. J. J.
Org. Chem. 1994, 59, 6103−6106. (d) Lantos, I.; Flisak, J.; Liu, L.;
Matsuoka, R.; W; Stevenson, D.; Tubman, K.; Tucker, L.; Zhang, W.-
Y.; Adams, J.; Sorenson, M; Garigipati, R.; Erhardt, K.; Ross, S. J. Org.
AUTHOR INFORMATION
■
Corresponding Author
bode@org.chem.ethz.ch
Chem. 1997, 62, 5385−5391. (e) Fassler, R.; Frantz, D. E.; Oetiker, J.;
̈
Notes
Carreira, E. M. Angew. Chem., Int. Ed. 2002, 41, 3054−3056. (f) Mita,
N.; Tamura, O.; Ishibashi, H.; Sakamoto, M. Org. Lett. 2002, 4, 1111−
The authors declare no competing financial interest.
695
dx.doi.org/10.1021/op300069n | Org. Process Res. Dev. 2012, 16, 687−696

Organic Process Research & Development
Article
́
1114. (g) Topic, D.; Aschwanden, P.; Fassler, R.; Carreira, E. M. Org.
̈
Lett. 2005, 7, 5329−5330. (h) Ritter, T.; Carreira, E. M. Angew. Chem.,
Int. Ed. 2005, 44, 936−938. (i) Gao, Z.-X.; Wang, M.; Wang, S.; Yao,
Z.-J. Org. Lett. 2009, 11, 3678−3681.
(15) (a) Iida, H.; Kasahara, K.; Kibayashi, C. J. Am. Chem. Soc. 1986,
108, 4647−4648. (b) Kasahara, K.; Iida, H.; Kibayashi, C. J. Org. Chem.
1989, 54, 2225−2233.
(16) D- and L-Gulono lactones are derived, respectively, from D-xylose
and L-ascorbic acid. They are available from Carbosynth for 1200 and
2500 USD/kg (catalog price).
(17) Rohloff, J. C.; Alfredson, T. V.; Schwartz, M. A. Tetrahedron Lett.
1994, 35, 1011−1014.
(18) D- and L-Gulose oxime chiral auxiliaries are commercially
available from BioBlocks, Inc.
(19) Huchler, G.; Rall, W.; Ries, U. U.S. Patent 2008045705, 2008.
(20) Roush, W. R.; Brown, B. B. J. Org. Chem. 1992, 57, 3380−3387.
(21) Likhterov, V. R.; Etlis, V. S.; Ternoskoi, L. A. Khim. Geterotsikl.
Soedin. 1985, 1316−1318.
(22) For example, α-chlorohydrin [CAS 96-24-2] is available from
Acros for 53.70 USD/liter.
(23) (a) Hope, D. B.; Walti, M. J. Chem. Soc. C 1970, 2475−2478.
(b) Wisniewski, K. Org. Prep. Proced. Int. 1999, 31, 211−214.
(24) Harris, E. M. CRC Handbook of Laboratory Safety, 4th ed.; Furr,
A. K., Ed.; CRC Press: Boca Raton, 1995; Section 4.6.2.4, Perchloric
Acid, p 308.
(25) Isoxazolidine·HCl salts can be stored in the absence of water at
room temperature. The free base form is stable in the absence of water
during the evaporation of the solvent used for the workup (EtOAc, 30
°C) and can be stored for 24 h at 4 °C before being submitted to
fragmentation. The free base oxidizes slowly to the isoxazoline.
(26) Nejman, M.; Sliwinska, A.; Zwierzak, A. Tetrahedron 2005, 61,
8536−8541.
696
dx.doi.org/10.1021/op300069n | Org. Process Res. Dev. 2012, 16, 687−696