Threonine aldolases-an emerging tool for organic synthesis

Article abstract of DOI:10.1016/j.tet.2006.11.035

In a systematic study, 21 ring-substituted benzaldehydes were reacted with glycine under catalysis with a l-threonine aldolase (lTA) from Pseudomonas putida and a d-threonine aldolase (dTA) from Alcaligenes xylosoxidans to form the corresponding β-hydroxy-α-amino acids 1-18. dTA proved to be highly selective with ee's >99% (d) and de's up to 99% (syn). Two thiamphenicol precursors were synthesized utilizing dTA on a preparative scale. lTA-catalyzed reactions led to ee's >99% (l) but low to moderate de's (20-50%, syn).

Full text of DOI:10.1016/j.tet.2006.11.035

Tetrahedron 63 (2007) 918–926
Threonine aldolases—an emerging tool for organic synthesis
a
Johannes Steinreiber, Kateryna Fesko, Christoph Reisinger, Martin Schurmann,
b
a
c
€
Friso van Assema,c Michael Wolberg,c,y Daniel Minkc and Herfried Griengla,b,
*
aResearch Centre Applied Biocatalysis, Petersgasse 14, 8010 Graz, Austria
bInstitute of Organic Chemistry, Graz University of Technology, Stremayrgasse 16, 8010 Graz, Austria
cDSM Pharmaceutical Products, Advanced Synthesis, Catalysis and Development, PO Box 18, 6160 MD Geleen, The Netherlands
Received 12 October 2006; revised 7 November 2006; accepted 10 November 2006
Available online 8 December 2006
Abstract—In a systematic study, 21 ring-substituted benzaldehydes were reacted with glycine under catalysis with a L-threonine aldolase
(LTA) from Pseudomonas putida and a D-threonine aldolase (DTA) from Alcaligenes xylosoxidans to form the corresponding b-hydroxy-a-
amino acids 1–18. DTA proved to be highly selective with ee’s >99% (D) and de’s up to 99% (syn). Two thiamphenicol precursors were
synthesized utilizing DTA on a preparative scale. LTA-catalyzed reactions led to ee’s >99% (L) but low to moderate de’s (20–50%, syn).
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
of a-amino aldehydes,31 catalytic glycine-derived silicon
enolates,32 chiral ammonium salts,33,34 ammonium ylides,35
radical bromination of protected amino acids forming trans-
oxazolidinones as intermediates,36 and numerous others.
b-Hydroxy-a-amino acids and the corresponding amino
alcohols are valuable precursors for active ingredients of
pharmaceuticals and agrochemicals.1–4 Furthermore, these
compounds are ubiquitous in nature and important as both
biological amino acids (threonine, serine, and 3-hydroxypro-
line) as well as key structural units in numerous bioactive
natural products like antibiotics (chloramphenicol,5,6 glyco-
peptides, e.g., vancomycin, ristocetin),7,8 antifungal agents
(sphingofungins),9,10 immunosuppressants (mycestericins),11
HIV-inhibitory and cytotoxic depsipeptides (papuamides,12
callipeltin E13), anticancer agents (polyoxypeptin A),14 neu-
rotrophic agents (lactacystin),14,15 and anti-inflammatory
agents (cyclomarins).16–18 In addition, they are useful chiral
intermediates due to their ability to undergoavariety of trans-
formations.
Although each of these approaches represents an elegant
methodology, there still remain some limitations such as in-
sufficient stereoselectivity, expensive precursors or the need
of stoichiometric amounts of chiral auxiliaries. In contrast,
the direct enzymatic synthesis of b-hydroxy-a-amino acids
from glycine and appropriate aldehydes is a onestep process
under mild conditions with minimal protection needs for
sensitive functionalities and hence may complement the
established chemical methods.37,38 This was first shown for
whole cell fermentation back in 197539 and 1987.40 Since
then several threonine aldolases (TAs, EC 4.1.2.5)41 and
serine hydroxymethyl transferases (EC 2.1.2.1),42–45 one
L-phenylserine aldolase (EC 4.1.2.26),46 and one D-3-
hydroxyaspartate aldolase (EC 4.1.3.14)47 have been discov-
ered. Threonine aldolases use pyridoxal-50-phosphate (PLP)
as a co-factor48,49 to promote the physiological degradation
of threonine forming glycine and acetaldehyde. In the reverse
reaction, PLP activates glycine in a non-physiological aldol
reaction (Scheme 1). The low acidity of the Ca-protons
prevents competing chemical background reactions. Genes
encoding L-threonine aldolase (LTA) have been found in
plants,50 vertebrates,51 several bacteria, yeasts, and fungi.41
Two X-ray structures of LTAs are described in the litera-
ture.52,53 In contrast, only a few D-threonine aldolase genes
(DTA) have been reported so far.41
Since the end of the 1980s, tremendous efforts have been de-
voted to the synthesis of optically pure b-hydroxy-a-amino
acids using chemical approaches such as Sharpless di-
hydroxylation,19 Sharpless epoxidation,15 Sharpless amino-
hydroxylation,20 dynamic kinetic resolution via ruthenium/
rhodium catalyzed dehydrogenation of N-substituted a-
amino-b-keto esters,21–27 aziridine ring-opening,28 aza-Claisen
rearrangements of allylic acetimidates,29 addition of imides
to aldehydes,30 organocatalytic asymmetric aldol reactions
Keywords: Lyase; Threonine aldolase; Biocatalysis; Amino acid.
* Corresponding author. Tel.: +43 316 873 8741; fax: +43 316 873 8740;
e-mail: herfried.griengl@a-b.at
Present address: DSM Pharma Chemicals, ResCom, Donaustaufer Str.
378, D-93055 Regensburg, Germany.
y
A decade ago, L- and D-threonine aldolases have been stud-
ied for the biocatalytic synthesis of b-hydroxy-a-amino
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.11.035

J. Steinreiber et al. / Tetrahedron 63 (2007) 918–926
919
OH
OH
CO2H
CO2H
LTA, PLP
+
NH2
NH2
O
X
X
(a)
L-anti-1
L-syn-1
CO2H
NH2
X
OH
(b)
OH
CO2H
CO2H
DTA, PLP
MnCl2
+
NH2
NH2
X
X
D-anti-1
D-syn-1
Scheme 1. Threonine aldolase catalyzed synthesis of (a) L-phenylserine L-1 and (b) D-phenylserine D-1.
acids using various aldehydes and glycine. The reaction pro-
ceeded with complete stereospecificity at the a-carbon.54–56
However, these investigations suffer from modest yields and
low diastereoselectivities. These obstacles were overcome
for b- and g-benzyloxy aldehydes as was shown in the syn-
thesis of monobactams57 and mycestericin D and F.58,59
Recently, L-phenylserine aldolase (from P. putida)60 and
D-3-hydroxyaspartate aldolase (from Paracoccus denitrifi-
cans IFO 13301)61 were characterized with a similar sub-
strate range compared to LTA and DTA, respectively, but
very low diastereoselectivities for the produced b-hydroxy-
a-amino acids. Additionally, the structure of L-phenylserine
aldolase was obtained (pdb-entry: 1v72). Interestingly, the
high structural similarity of PLP enzymes was shown by
a novel engineered DTA, which was obtained by a single
point mutation of alanine racemase (EC 5.1.1.1).62
centrifugation had a specific activity of 20 U/mg total pro-
tein (Biorad). Approximately 4800 U (D-threonine) of over-
expressed DTA was obtained from 1 L cell culture and the
specific activity was determined for the crude enzyme to
be 8 U/mg.
2.2. Assignment of absolute configuration
The analytical yields and diastereoselectivities were moni-
tored by HPLC and NMR. The assignment for syn- and
anti-phenylserine derivatives 1–18 was based on the cou-
pling constant between the a- and b-protons, which is
slightly larger for the syn- than for the anti-isomer.56,69
Further confirmation was obtained from the shift of the
a-proton, which is lower for syn than for anti-products.56
L- and D-threonine aldolases have been shown to be highly se-
lective for the stereoconfiguration at the a-carbon.56,61 This
was demonstrated by HPLC using well-established derivati-
zation methods for the analysis of amino acids on an achiral
RP18 column (o-phthaldialdehyde OPA/2-mercaptoethanol
achiral derivatization for syn/anti determination and OPA/
N-acetyl-cysteine chiral derivatization for the separation of
all four diastereoisomers).70,71 Additionally, chiral HPLC
analytics using CrownpackÒ CR(+) allowed the assignment
for L- and D-isomers due to the fact that the D-amino acid
always elutes first.
Surprisingly, most enzymatic studies have focused on the
kinetic resolution of chemically produced DL-syn mixtures
(max. 50% yield)63–65 and not on the more promising aldol
reaction creating two stereogenic centers (max. 100%
yield).54–58,66 Herein, we report on an improved enzymatic
procedure using D-threonine aldolases from A. xylosoxi-
dans,65 which leads to high enantio- and diastereoselectivity,
whereas LTA from P. putida67 gave only moderate selectiv-
ity. Both enzymes were also tested on benzaldehyde deriva-
tives, which had been described as unfavored substrates.56
The reactions were completed after 1–4 h with generally
high analytical yields, although strongly depending on the
reaction conditions.
2.3. Temperature profile and timescale for the synthesis
reaction
For the synthesis of D-phenylserine D-1 employing DTA, the
rate of the aldol reaction was studied from 5 to 40 ꢀC (Table
1). As it was shown before, a high excess of glycine shifts the
equilibrium toward the phenylserine side and thus 10 equiv
of glycine was utilized (1 M glycine, 100 mM benzalde-
hyde) throughout the entire study.56 Lowering the tempera-
ture to 5 ꢀC resulted in an increase of diastereoselectivity
(entry 1, up to 98%) as well as in a higher analytical yield
2. Results and discussion
2.1. Enzyme overexpression and properties of
LTA and DTA
Recombinant Escherichia coli L- and DTA expression strains
based on the arabinose inducible pBAD plasmid system (In-
vitrogen) were obtained as published.68 Threonine aldolase
activity was measured by an NADH coupled assay with
yeast alcohol dehydrogenase (Sigma). In this assay, L- or
D-threonine is converted into glycine and acetaldehyde by
the action of TA. Then, by the action of yeast alcohol dehy-
drogenase, the released acetaldehyde is reduced to ethanol
with simultaneous oxidation of NADH to NAD+ monitored
by the decrease of absorbance at 340 nm. Recombinant over-
expression of LTAyielded 16,000 units from 1 L cell culture.
The crude enzyme extract prepared by cell disruption and
Table 1. DTA-catalyzed synthesis of D-syn-1 at different temperatures
Entries
T/(ꢀC)
Time/(h)
Analytical yield/(%)
de/(%)
1
2
3
4
5
20
30
40
4
2
1
1
79
72
43
45
98
92
91
73
Conditions: 1 mL buffer solution (pH 8) containing glycine (1 M), benz-
aldehyde (100 mM), PLP (50 mM), MnCl2 (50 mM), and DTA (23 U);
ee >99% (D) for all reactions; all results obtained by HPLC.

920
J. Steinreiber et al. / Tetrahedron 63 (2007) 918–926
(79%) after 4 h. Rising the temperature to 20 and 30 ꢀC re-
sulted in a slightly lower de (entries 2 and 3) whereas at
40 ꢀC, the selectivity is significantly decreased (entry 4). It
is assumed that a lower temperature reduces the reaction
rate and as a consequence, the formation of the kinetic prod-
uct is favored. To the best of our knowledge, our results rep-
resent the first example of D-syn-1 being produced in an
enzyme-catalyzed procedure with high diastereoselectivity
and analytical yield. No similar effects were detected for
LTA at temperatures below 40 ꢀC and after 2 h the maximum
analytical yield of L-syn/anti-1 was obtained (85% yield, de
20% syn).
Table 2. Synthesis of 1: amount (% v/v) of co-solvent used without any loss
of activity and selectivity; LTA: analytical yield 40%, de 20% (syn); DTA:
analytical yield 65%, de 85% (syn)
No.
Co-solvent
Co-solvent
% (v/v) LTA
Co-solvent
% (v/v) DTAa
1
2
3
4
5
6
DMSO
DMF
EtOH
CH3CN
TBME
THF, toluene, Et2O,
cyclohexane
30
20
20
10
30
<10
30
20
<10
<10
<10
<10
Conditions: 1 mL solution containing glycine (1 M), benzaldehyde
(100 mM), PLP (50 mM), and LTA (77 U) or DTA (23 U) at 25 ꢀC;
ee >99% (D) for DTA and ee >99% (L) for LTA; LTA: 1 h; DTA: 2 h; analyt-
ical yield and de determined by HPLC.
2.4. pH dependency
a
Additional: MnCl2 (50 mM).
LTA and DTA reversibly catalyze the formation and cleavage
of 1. The mechanism published by John indicates that the re-
action is sensitive to pH as general base catalysis is in-
volved.48 Hence, the pH dependency of LTA and DTA was
determined in order to obtain information of the possible
range of process optimization (Fig. 1). LTA is stable between
pH 5.5 and 8 but shows a clear decrease of activity at pH 5.0.
On the contrary, DTA shows little activity at pH 6.0. How-
ever, by increasing the pH to 9.5 the analytical yield of D-1
was doubled retaining high diastereoselectivity as compared
to pH 8.0.
2.6. Synthesis of phenylserine derivatives 1–18
Threonine aldolases have been tested on a variety of alde-
hydes for the synthesis of b-hydroxy-a-amino acids in the
past.54–56 Nevertheless, a detailed study on benzaldehyde
derivatives would help to understand the influence of aryl
substituents for further modeling of the active sites of LTA
and DTA. For the enzymes investigated no structural data ex-
ist to date whereas three structures of LTAs from other micro-
organisms are available. Additionally, some industrially
relevant precursors for bioactive compounds, for example,
thiamphenicol were synthesized and isolated.
2.5. Co-solvents
The tolerance of LTA and DTA concerning co-solvents was
tested on the synthesis of 1 (Table 2). Surprisingly, the
solvents employed did not improve the performance of the
reaction as it had been shown for other threonine aldolases.56
Nevertheless, DMSO and DMF were both tolerated by LTA
and DTA up to 20 and 30%, respectively (entries 1 and 2).
LTA was less sensitive toward co-solvents compared to
DTA as demonstrated by CH3CN and EtOH, which were
tolerated up to 10 and 20%, respectively (entries 3 and 4).
TBME (up to 30% v/v) was shown to be the only water-
immiscible co-solvent, which allowed running the LTA-
catalyzed reaction (entry 5) without any loss of performance.
All other co-solvents such as THF, Et2O, toluene, and cyclo-
hexane inactivated the enzymes irreversibly (entry 6).
2.6.1. L-Threonine aldolase. Following the preliminary
studies, the substrate range for LTA investigated was ex-
tended to various benzaldehyde derivatives and the results
were compared with benzaldehyde (Table 3, entry 1). Fluo-
robenzaldehydes showed good analytical yields and no
significant change in selectivity (entries 2–4). 2-Chloro-
benzaldehyde gave high analytical yield and increased de
of 52% (entry 5). However, L-6 and especially L-7 were
only obtained with significant lower analytical yields and
de’s (entries 6 and 7). All bromobenzaldehydes showed a
similar decrease in analytical yield from ortho to meta and
para compared to the chloroderivatives (entries 8–10 vs
5–7). L-9 was obtained with similar de’s compared to L-5
(entry 5 vs 9). Apparently, the steric demand of 2-chloro
and 3-bromobenzaldehyde improves the selectivity signifi-
cantly. The low analytical yields of L-7 and L-10 may be ex-
plained by the low solubility of the corresponding aldehydes.
L-11–13 and the sulfonamide compound L-15 showed high
analytical yield with low selectivity whereas the methylsul-
fonylbenzaldehyde gave an increased selectivity of de 53%
(entries 11–15). L-11–15 are precursors for thiamphenicol
and chloramphenicol isomers7 and L-16 is a precursor for
the amino alcohol norfenefrin.72 L-16 was obtained with
moderate analytical yield and selectivity (entry 16). Simi-
larly, L-17—which can be converted to octopamin73—was
synthesized although with low analytical yield and selectiv-
ity (entry 17). L-3,4-Dihydroxyphenylserine is an utmost im-
portant target. It is used as a drug against Parkinson’s disease
but can also be used for the synthesis of noradrenaline.74,75
Since 4-hydroxybenzaldehyde was a poor substrate, it was
not surprising that neither 3,4-dihydroxy- nor 3,5-dihydroxy-
benzaldehyde or 4-hydroxy-3-(hydroxymethyl)benzalde-
hyde was accepted by LTA (entries 18, 20, and 21). The
100
LTA
DTA
80
60
40
20
0
5
6
7
8
9
10
pH
Figure 1. Synthesis of (a) L-syn/anti-1 utilizing LTA and (b) D-syn/anti-1
employing DTA in different buffers; Conditions: 1-mL solution containing
glycine (1 M), benzaldehyde (100 mM), PLP (50 mM) at 25 ꢀC; (a) LTA
(77 U), ee >99% (L) and de 25% (syn) for all reactions; 2 h; (b) MnCl2
(50 mM) and DTA (23 U); ee >99% (D) and de >90% (syn) for all reactions;
1 h; all results obtained by HPLC.

J. Steinreiber et al. / Tetrahedron 63 (2007) 918–926
Table 3. LTA-catalyzed synthesis of L-phenylserine derivatives L-1–18
921
OH
OH
O
CO2H PLP, LTA
pH 8
CO2H
CO2H
+
NH2
NH2
NH2
x
x
x
L-anti 1-18
L-syn 1-18
Entries
X
Analytical
yield (%)a
de (%)a
Entries
X
Analytical
yield (%)a
de (%)a
1
2
3
4
5
6
7
8
H L-1
2-F L-2
3-F L-3
4-F L-4
2-Cl L-5
3-Cl L-6
4-Cl L-7
2-Br L-8
3-Br L-9
4-Br L-10
2-NO2 L-11
85
68
64
51
90
69
57
79
63
47
99
20
35
27
29
52
30
17
34
55
14
32
12
13
14
15
16
17
18
19
20
21
3-NO2 L-12
4-NO2 L-13
4-MeSO2 L-14
2-NH2SO2 L-15
3-OH L-16
4-OH L-17
3,4-OH
3,4-O(Me) L-18
3-MeOH-4-OH
3,5-OH
74
79
21
24
68
53
92b
56
24b
51
11b
<1b
15
36b
n.d.
26
9
10
11
<1b
<1b
n.d.
n.d.
Conditions: 1 mL solution containing glycine (1 M), benzaldehyde (100 mM), PLP (50 mM), and LTA (77 U) at 25 ꢀC; 1–2 h; ee >99% (L) for all reactions.
Determined by HPLC.
a
Determined by 1H NMR.
b
latter two substrates serve as starting material for the syn-
thesis of levabuterol and terbutaline, respectively.76,77 To
eliminate the unfavorable effects of both hydroxy groups,
piperonal was applied as protected 3,4-dihydroxybenzalde-
hyde and furnished the corresponding product L-18 with
low analytical yield and selectivity (entry 19). This strategy
renders dihydroxybenzaldehyde compounds applicable to
the TA technology.
The meta compounds D-6,9 were gained with highest analyt-
ical yield, the meta D-6,9 and para products D-7,10 gave
good de’s. However, the analytical yields of 2- and 4-haloge-
nated compounds D-5,7,8,10 were modest and the Cb-selec-
tivities for both ortho-substrates were lower compared to
benzaldehyde. Thus, the active site of DTA seems to be
more rigid compared to LTA, complicating the accommoda-
tion of bulky benzaldehydes, especially with ortho-substitu-
ents. This effect can also be seen for 2-nitrobenzaldehyde,
which only gave D-11 with low analytical yield and de-
creased selectivity (entry 11). On the contrary, the other
nitro-compounds were obtained with good (D-12) and very
high (D-13) de’s (entries 12 and 13). Both the methylsulfo-
nium and sulfonamide benzaldehydes were converted with
similar analytical yields and very high syn/anti ratios (en-
tries 14 and 15). 3-Hydroxybenzaldehyde was well accepted
2.6.2. D-Threonine aldolase. Consequently, the same benz-
aldehyde derivatives were tested also on D-threonine aldol-
ase. D-1 was obtained with good analytical yield and
excellent de (Table 4, entry 1). All fluoro-substrates gave
similar analytical yields but only D-2 and D-4 were obtained
with comparable de’s (entries 2–4). Similar effects were
found for chloro- and bromobenzaldehydes (entries 5–10).
Table 4. DTA-catalyzed synthesis of D-phenylserine derivatives D-1–18
OH
OH
O
PLP, DTA
CO2H
CO2H
+
CO2H
MnCl2, pH 9.5
NH2
NH2
NH2
x
x
x
D-anti 1-18
D-syn 1-18
Entries
X
Analytical
yield (%)a
de (%)a
Entries
X
Analytical
yield (%)a
de (%)a
1
2
3
4
5
6
7
8
H D-1
2-F D-2
3-F D-3
4-F D-4
2-Cl D-5
3-Cl D-6
4-Cl D-7
2-Br D-8
3-Br D-9
4-Br D-10
2-NO2 D-11
79
68
54
42
27
60
26
6
98
95
81
91
67
85
86
35
71
74
65
12
13
14
15
16
17
18
19
20
21
3-NO2 D-12
90
31
80
75
4-NO2 D-13
4-MeSO2 D-14
2-NH2SO2 D-15
3-OH D-16
4-OH D-17
3,4-OH
3,4-O(Me) D-18
3-MeOH-4-OH
3,5-OH
63
99
53b
76
>90b
86
15b
<1b
16b
<1b
<1b
70b
n.d.
46b
n.d.
n.d.
9
10
11
43
12
18
Conditions: 1 mL solution containing glycine (1 M), benzaldehyde (100 mM), PLP (50 mM), MnCl2 (50 mM), and DTA (23 U) at 5 ꢀC; 4 h; ee >99% (D) for all
reactions.
Determined by HPLC.
a
Determined by 1H NMR.
b

922
J. Steinreiber et al. / Tetrahedron 63 (2007) 918–926
by DTA forming D-16 whereas 4-hydroxybenzaldehyde
furnished the corresponding product D-17, similar to the
LTA-catalyzed reaction, with low analytical yield (entries
16 and 17). All dihydroxy compounds (entries 18, 20, and
21) remained unconverted. However, piperonal as protected
analog of the unfavored 3,4-dihydroxybenzaldehyde gave
D-18 with low analytical yield and selectivity (entry 19).
enzymes in our hands we investigated whether LTA and
DTA act on donor substrates such as D- and L-alanine, DL-
leucine, glycine ethylester, glycine amide, and ethanolamine
together with benzaldehyde or acetaldehyde. Deplorably,
no conversion to the corresponding b-hydroxy-a-amino
derivatives was detected.
2.7. Separation of derivatized syn/anti-1
3. Conclusion
Unfortunately, the low diastereoselectivity of LTA is a severe
drawback for preparative application. Hence, new strategies
involving selective coupled reactions and genetically
improved enzymes are under investigation. For larger scale,
recrystallization of the syn/anti-mixture has been applied
and pure products were obtained.56 Small amounts of syn/
anti-1 derivatives can be separated by a two-step derivatiza-
tion involving an initial esterification78 and subsequent BOC
protection of the amino group.79 The diprotected phenyl-
serines were separated by flash chromatography yielding
enriched syn- and anti-products.
In summary, both enzymes—LTA from P. putida and DTA
from A. xylosoxidans—were tested on a variety of benzalde-
hyde derivatives as acceptor substrates and the reaction con-
ditions were optimized. DTA accepted all benzaldehydes
excluding the dihydroxy substrates and furnished D-syn
products with ee’s >99% and de’s up to 99% (syn). To out-
line the synthetic applicability of these reactions, a precursor
of a thiamphenicol isomer and an analog were synthesized
and isolated on a preparative scale with high yields. In
case of LTA, the same benzaldehyde derivatives were well-
accepted substrates and gave the corresponding amino acids
with high analytical yields and ee’s.
2.8. Thiamphenicol
Several pharmaceuticals can be found in the literature, which
contain the physiologically essential b-hydroxy-a-amino
functionalities, for example, thiamphenicol (D-syn-2-dichlor-
acetamido-1-(4-methylsulfonylphenyl)-propan-1,3-diol) and
analogs (D-syn-2-dichloracetamido-1-(4-sulfamoylphenyl)-
propan-1,3-diol). Hence, 4-formyl-benzenesulfonamide was
synthesized as precursor using standard reactions.80–82
Finally, the corresponding phenylserine derivatives D-syn-
14,15 were synthesized using DTA and isolated in good
yields, absolute stereocontrol of the a-carbon (ee >99%)
and excellent stereocontrol even for the b-carbon (Scheme
2). Further conversion to yield the desired thiamphenicol
enantiomer can easily be achieved using standard trans-
formations.83 Additionally, the physiologically active enan-
tiomer L-14 was isolated in good yield (75%) but low de
(20% syn).
4. Experimental
4.1. General experimental procedures
All reagents and solvents were obtained from commercial
sources and appropriately purified, if necessary. 1H and
13C NMR spectra were recorded on a Varian INOVA 500
(1H 499.82 MHz, 13C 125.69 MHz) or on a Varian GEMINI
200 (1H 199.98 MHz, 13C 50.29 MHz) using the residual
peaks of CDCl3 (1H: d 7.26, 13C: d 77.0), D2O (1H: d 4.79)
or DMSO-d6 (1H: d 2.50, 13C: d 40.2) as references. H2O/
D2O-NMR samples were taken directly from the aqueous
solution, diluted with D2O (1:1) and recorded using H2O
presaturation.84 Analytical HPLC was carried out with
a Hewlett Packard Series 1100 HPLC using a G1315A diode
array detector. If not otherwise noted, a PurospherÒ STAR
RP18 (250 mm, 5 mm) column was used for analysis. Ana-
lytical yields and de’s of b-hydroxy-a-amino acids were
determined by HPLC after derivatization with ortho-
phthaldialdehyde/2-mercaptoethanol (OPA/MCE, achiral
derivatization, de determination) or OPA/N-acetyl cysteine
2.9. Different donors
Aldolases are known to be flexible for the acceptor but rigid
for the donor substrate. To clarify the situation for the
OH
R
O
CO2H
S
Gly
O
S
O
S
NH2
DTA
O
O
OH
(a)
(b)
yield 81%
de 92%
S
D-syn 14
OH
Cl
S
O
ee >99%
HN
O
S
O
R
Cl
OH
O
L-syn
R
CO2H
Gly
DTA
S
a) R = Me, thiamphenicol isomer
b) R = NH2
O
O
NH2
S
O
S
H2N
H2N
O
yield 61%
de >95%
ee >99%
D-syn 15
Scheme 2. Synthetic route to (a) a thiamphenicol isomer and (b) a thiamphenicol analog.

J. Steinreiber et al. / Tetrahedron 63 (2007) 918–926
923
(OPA/NAC, chiral derivatization, ee determination).70,71 DL-
anti-185 and DL-syn-phenylserine derivatives86 were synthe-
sized as reference materials using published methods. All
phenylserine derivatives are known compounds and thus,
no characterization using high-resolution MS or microana-
lysis was undertaken. Mostly, diastereomeric mixtures were
obtained.
pH 8 (50 mM KH2PO4)/CH3CN¼79:21, 0.8 mL/min,
-syn -syn -anti -anti
tD ¼8.6 min, tL ¼10.4 min, tD ¼12.2 min, tL
¼
13.9 min; NMR data were consistent with those reported.87
4.4.3. L-3-Fluorophenylserine L-3. HPLC: OPA/MCE:
buffer pH 8 (50 mM KH2PO4)/CH3CN¼71:29, 0.8 mL/
min, tsyn¼14.1 min, tanti¼17.5 min; OPA/NAC: buffer pH
8 (50 mM KH2PO4)/CH3CN¼79:21, 0.8 mL/min, tD
¼
-syn
4.2. Overexpression
11.1 min, tL ¼14.1 min, tD ¼15.3 min, tL ¼17.6 min;
-syn
-anti
-anti
1H NMR (500 MHz, D2O): d 3.71 (d, 0.63H, syn, J¼4.5 Hz),
3.88 (d, 0.37H, anti, J¼4.0 Hz), 5.10 (d, 0.63H, syn,
J¼4.5 Hz), 5.16 (d, 0.37H, anti, J¼4.0 Hz), 7.00 (m, 3H),
7.25 (m, 1H).
Both LTA and DTA were expressed from pBAD constructs
containing the genuine coding DNA sequences under the
control of the ara promotor. E. coli Top10F’ cells containing
the pBAD-LTAPp12565 (LTA) or the pBAD-DTAAf (DTA)
plasmids were cultivated on 2xTY media supplemented
with 100 mg/mL ampicilline. Overnight cultures (100 mL)
in 300 mL flasks were inoculated with single colonies and
grown at 37 ꢀC. Main cultures (330 mL) in baffled
1000 mL flasks were inoculated with 3 mL of the preculture
and grown at 37 ꢀC for approximately 4 h to an OD600 of 1.5.
Temperature was then lowered to 28 ꢀC and induced with
L-arabinose at a final concentration of 0.002%. Cultivation
was continued overnight and cells harvested by centrifuga-
tion for 15 min at 5000ꢁg. After resuspension of the pellets
in 0.1 M sodium phosphate buffer the cells were disrupted
by ultrasonic treatment. The crude lysate was cleared by cen-
trifugation at 20,000ꢁg for 1 h and the supernatant (cell-free
extract) stored frozen at ꢂ20 ꢀC until use.
4.4.4. L-4-Fluorophenylserine L-4. HPLC: OPA/MCE:
buffer pH 8 (50 mM KH2PO4)/CH3CN¼71:29, 0.8 mL/
min, tsyn¼13.8 min, tanti¼17.7 min; OPA/NAC: buffer pH
8 (50 mM KH2PO4)/CH3CN¼79:21, 0.8 mL/min, tD
NMR data were consistent with those reported.87
¼
-syn
10.4 min, tL ¼13.4 min, tD ¼15.5 min, tL ¼17.9 min;
-syn
-anti
-anti
4.4.5. L-2-Chlorophenylserine L-5. HLPC: OPA/MCE:
buffer pH 8 (50 mM KH2PO4)/CH3CN¼71:29, 0.8 mL/
min, tsyn¼15.4 min, tanti¼19.3 min; OPA/NAC: buffer pH
8 (50 mM KH2PO4)/CH3CN¼79:21, 0.8 mL/min, tD
1H NMR (500 MHz, D2O): d 3.91 (d, 1H, J¼3.5 Hz), 5.31
(d, 0.24H, anti, J¼3.5 Hz), 5.48 (d, 0.76H, syn, J¼3.5 Hz),
7.30 (m, 4H).
¼
-syn
11.3 min, tL ¼13.3 min, tD ¼19.6 min, tL ¼23.0 min;
-syn
-anti
-anti
4.3. Activity assay
4.4.6. L-3-Chlorophenylserine L-6. HPLC: OPA/MCE:
buffer pH 8 (50 mM KH2PO4)/CH3CN¼71:29, 0.8 mL/
min, tsyn¼22.5 min, tanti¼27.8 min; OPA/NAC: buffer pH
The assay mixture contained D- or L-threonine (50 mM),
HEPES–NaOH buffer (100 mM, pH 8), PLP (50 mM),
MnCl2 (only for DTA, 100 mM), NADH (200 mM), yeast al-
cohol dehydrogenase (30 U, Sigma), and 50 mL of diluted
cell-free extract (1:100 for DTA, 1:1000 for LTA) in a final
volume of 3 mL. The reactions were started by the addition
of the diluted cell-free extract and recorded by the decrease
of absorbance at 340 nm in a spectrophotometer. One unit of
the enzyme is the amount of enzyme that catalyzes the
formation of 1 mmol of acetaldehyde (1 mmol of NADH
oxidized) per minute at room temperature under above-
described conditions.
8 (50 mM KH2PO4)/CH3CN¼79:21, 0.8 mL/min, tD
¼
-syn
20.4 min, tL ¼26.8 min, tD ¼27.4 min, tL ¼32.5 min;
-syn
-anti
-anti
1H NMR (500 MHz, D2O): d 3.70 (d, 0.65H, syn, J¼4.5 Hz),
3.87 (d, 0.35H, anti, J¼4.0 Hz), 5.07 (d, 0.65H,
syn, J¼4.5 Hz), 5.13 (d, 0.35H, anti, J¼4.0 Hz), 7.20
(m, 4H).
4.4.7. L-4-Chlorophenylserine L-7. HPLC: OPA/MCE:
buffer pH 8 (50 mM KH2PO4)/CH3CN¼69:31, 0.8 mL/
min, tsyn¼17.3 min, tanti¼21.4 min; OPA/NAC: buffer pH
8 (50 mM KH2PO4)/CH3CN¼79:21, 0.8 mL/min, tD
NMR data were consistent with those reported.88
¼
-syn
4.4. General procedure for the synthesis of L-phenyl-
serine derivatives
22.9 min, tL ¼30.4 min, tD ¼32.5 min, tL ¼39.1 min;
-syn
-anti
-anti
4.4.1. L-Phenylserine L-1. To a solution of LTA (77 U) and
PLP (13 ng, 50 nmol) in 1 mL buffer (KH2PO4, 50 mM,
pH 8.0) benzaldehyde (10 mg, 0.1 mmol) and glycine
(75 mg, 1.0 mmol) were added. The reaction mixture was
stirred at rt for 1 h to give L-syn/anti-1; analytical yield
85%; de 20% (syn); ee syn>99% (L), anti>99% (L);
HPLC: OPA/MCE: buffer pH 8 (50 mM KH2PO4)/CH3CN¼
73:27, 0.8 mL/min, tsyn¼16.3 min, tanti¼21.6 min; OPA/
NAC: buffer pH 8 (50 mM KH2PO4)/CH3CN¼81:19,
4.4.8. L-2-Bromophenylserine L-8. HPLC: OPA/MCE:
buffer pH 8 (50 mM KH2PO4)/CH3CN¼71:29, 0.8 mL/
min, tsyn¼16.9 min, tanti¼21.7 min; OPA/NAC: buffer pH
8 (50 mM KH2PO4)/CH3CN¼79:21, 0.8 mL/min, tD
¼
-syn
12.6 min, tL ¼14.8 min, tD ¼23.3 min, tL ¼27.4 min;
-syn
-anti
-anti
1H NMR (500 MHz, D2O): d 3.91 (d, 0.67H, syn, J¼3.5 Hz),
3.93 (d, 0.33H, anti, J¼3.0 Hz), 5.23 (d, 0.33H, anti,
J¼3.0 Hz), 5.44 (d, 0.67H, syn, J¼3.5 Hz), 7.08 (t, 0.33H,
anti, J¼8.0 Hz), 7.11 (t, 0.67H, syn, J¼8.0 Hz), 7.23 (t,
0.33H, anti, J¼8.0 Hz), 7.29 (t, 0.67H, syn, J¼8.0 Hz),
7.35 (t, 0.33H, anti, J¼8.0 Hz), 7.44 (d, 1H, J¼8.0 Hz),
7.48 (d, 0.67H, syn, J¼8.0 Hz).
0.8 mL/min, tD ¼11.5 min, tL ¼15.4 min, tD
¼
-syn
-syn
-anti
18.1 min, tL ¼21.4 min; NMR data were consistent with
-anti
those reported.87
4.4.2. L-2-Fluorophenylserine L-2. HPLC: OPA/MCE:
buffer pH 8 (50 mM KH2PO4)/CH3CN¼71:29, 0.8 mL/
min, tsyn¼11.9 min, tanti¼13.9 min; OPA/NAC: buffer
4.4.9. L-3-Bromophenylserine L-9. HPLC: OPA/MCE:
buffer pH 8 (50 mM KH2PO4)/CH3CN¼69:31, 0.8 mL/
min, tsyn¼18.0 min, tanti¼21.9 min; OPA/NAC: buffer pH 8

924
J. Steinreiber et al. / Tetrahedron 63 (2007) 918–926
(50 mM KH2PO4)/CH3CN¼76:24, 0.8 mL/min, tD
¼
1H NMR (500 MHz, D2O): d 3.73 (d, 0.62H, syn,
J¼4.0 Hz), 3.91 (d, 0.38H, anti, J¼3.5 Hz), 5.16 (d,
0.62H, syn, J¼4.0 Hz), 5.22 (d, 0.38H, anti, J¼3.5 Hz),
7.40 (d, 0.76H, anti, J¼8.5 Hz), 7.47 (d, 1.24H, syn,
J¼8.5 Hz), 7.72 (d, 0.76H, anti, J¼8.5 Hz), 7.75 (d,
1.24H, syn, J¼8.5 Hz).
-syn
12.2 min, tD ¼15.1 min, tL ¼15.5 min, tL ¼17.8 min;
-anti
-syn
-anti
1H NMR (500 MHz, D2O): d 3.69 (d, 0.77H, syn, J¼4.5 Hz),
3.87 (d, 0.23H, anti, J¼4.0 Hz), 5.07 (d, 0.77H, syn,
J¼4.5 Hz), 5.12 (d, 0.23H, anti, J¼4.0 Hz), 7.18 (m 2H),
7.40 (m, 2H).
4.4.10. L-4-Bromophenylserine L-10. HPLC: OPA/MCE:
buffer pH 8 (50 mM KH2PO4)/CH3CN¼69:31, 0.8 mL/
min, tsyn¼20.9 min, tanti¼25.6 min; OPA/NAC: buffer pH
4.4.16. L-3-Hydroxyphenylserine L-16. HPLC: LiChros-
pherÒ 100 RP-18 column (250 mm, 5 mm), OPA/MCE:
buffer pH 8 (50 mM KH2PO4)/CH3CN¼81:19, 0.8 mL/
min, tsyn¼8.1 min, tanti¼12.1 min; 1H NMR (500 MHz,
D2O): d 3.70 (d, 0.75H, syn, J¼4.5 Hz), 3.86 (d, 0.25H,
anti, J¼4.0 Hz), 5.06 (d, 0.75H, syn, J¼4.5 Hz), 5.09 (d,
0.25H, anti, J¼4.0 Hz), 6.75 (m, 3H), 7.13 (m, 1H).
8 (50 mM KH2PO4)/CH3CN¼76:24, 0.8 mL/min, tD
¼
-syn
13.8 min, tL ¼18.0 min, tD ¼18.2 min, tL ¼21.4 min;
-syn
-anti
-anti
NMR data were consistent with those reported.89
4.4.11. L-2-Nitrophenylserine L-11. HPLC: OPA/MCE:
buffer pH 8 (50 mM KH2PO4)/CH3CN¼71:29, 0.8 mL/
min, tsyn¼12.9 min, tanti¼13.8 min; OPA/NAC: buffer pH
4.4.17. L-4-Hydroxyphenylserine L-17. (See Section
4.4.15) H NMR (500 MHz, D2O): d 3.66 (d, 0.68H, syn,
1
8 (50 mM KH2PO4)/CH3CN¼81:19, 0.8 mL/min, tD
¼
J¼5.0 Hz), 3.84 (d, 0.32H, anti, J¼3.5 Hz), 5.00 (d,
0.68H, syn, J¼5.0 Hz), 5.07 (d, 0.32H, anti, J¼3.5 Hz),
6.53 (d, 0.64H, anti, J¼9.0 Hz), 6.72 (d, 1.36H, syn,
J¼8.0 Hz), 7.06 (d, 0.64H, anti, J¼9.0 Hz), 7.12 (d,
1.36H, syn, J¼8.0 Hz).40,91
-syn
13.5 min, tL ¼15.2 min, tD ¼18.8 min, tL ¼20.6 min;
-syn
-anti
-anti
1H NMR (500 MHz, D2O): d 3.94 (d, 0.66H, syn, J¼4.0 Hz),
3.98 (d, 0.34H, anti, J¼3.0 Hz), 5.54 (d, 0.34H, anti,
J¼3.0 Hz), 5.71 (d, 0.66H; syn, J¼4.0 Hz), 7.39 (m, 1H),
7.64 (m, 2H), 7.94 (m, 1H).
4.4.18. L-3,4-Methylenedioxyphenylserine L-18. HPLC:
LiChrospherÒ 100 RP-18 column (250 mm, 5 mm), OPA/
MCE: buffer pH 8 (50 mM KH2PO4)/CH3CN¼76:24,
0.8 mL/min, tsyn¼8.9 min, tanti¼11.8 min; 1H NMR
(500 MHz, D2O): d 3.65 (d, 0.63H, syn, J¼4.5 Hz), 3.82
(d, 0.37H, anti, J¼4.5 Hz), 5.05 (m, 1H), 5.79 (d, 1H,
J¼4.5 Hz), 6.70 (m, 3H).91,92
4.4.12. L-3-Nitrophenylserine L-12. HPLC: OPA/MCE:
buffer pH 8 (50 mM KH2PO4)/CH3CN¼71:29, 0.8 mL/
min, tsyn¼15.5 min, tanti¼18.8 min; OPA/NAC: buffer pH
8 (50 mM KH2PO4)/CH3CN¼81:19, 0.8 mL/min, tD
¼
-syn
19.8 min, tL ¼24.3 min, tD ¼26.1 min, tL ¼29.7 min;
-syn
-anti
-anti
1H NMR (500 MHz, D2O): d 3.75 (d, 0.61H, syn, J¼4.5 Hz),
3.91 (d, 0.39H, anti, J¼4.0 Hz), 5.18 (d, 0.61H, syn,
J¼4.5 Hz), 5.25 (d, 0.39H, anti, J¼40 Hz), 7.44 (t, 0.39H,
anti, J¼8.0 Hz), 7.47 (t, 0.61H, syn, J¼8.0 Hz), 7.57 (d,
0.39H, anti, J¼8.0 Hz), 7.65 (d, 0.61H, syn, J¼8.0 Hz),
8.03 (m, 1H), 8.07 (s, 0.39H, anti), 8.14 (s, 0.61H, syn).
4.5. General procedure for the synthesis of D-phenyl-
serine derivatives
4.5.1. D-Phenylserine D-1. To a solution of DTA (23 U),
MnCl2 (6 ng, 50 nmol), and PLP (13 ng, 50 nmol) in 1 mL
buffer (glycine 100 mM/NaCl 50 mM, pH 9.5) benzalde-
hyde (10 mg, 0.1 mmol) and glycine (75 mg, 1.0 mmol)
were added. The reaction mixture was stirred at 5 ꢀC for
3 h to give D-syn-1; analytical yield 79%; de 98% (syn); ee
syn>99% (D), anti>99% (D).
4.4.13. L-4-Nitrophenylserine L-13. HPLC: OPA/MCE:
buffer pH 8 (50 mM KH2PO4)/CH3CN¼71:29, 0.8 mL/
min, tsyn¼16.2 min, tanti¼19.9 min; OPA/NAC: buffer pH
8 (50 mM KH2PO4)/CH3CN¼81:19, 0.8 mL/min, tD
¼
-syn
21.3 min, tL ¼26.6 min, tD ¼28.1 min, tL ¼31.0 min;
-syn
-anti
-anti
1H NMR (500 MHz, D2O): d 3.74 (d, 0.62H, syn, J¼4.5 Hz),
3.92 (d, 0.38H, anti, J¼4.0 Hz), 5.19 (d, 0.62H, syn,
J¼4.5 Hz), 5.24 (d, 0.38H, anti, J¼4.0 Hz), 7.41 (d,
0.76H, anti, J¼8.0 Hz), 7.48 (d, 1.24H, syn, J¼8.0 Hz),
8.05 (d, 0.76H, anti, J¼8.5 Hz), 8.08 (d, 1.24H, syn,
J¼8.0 Hz).88,90,91
4.5.2. D-3-(4-Methylsulfonylphenyl)serine D-14. To a
solution of glycine (2.25 g, 30 mmol), PLP (0.4 mg,
1.5 mmol), MnCl2 (0.18 mg, 1.5 mmol), and DTA (690 U)
in 30 mL buffer (glycine 100 mM/NaCl 50 mM, pH 9.5)
4-(methylsulfonyl)benzaldehyde (500 mg, 2.7 mmol) was
added and stirred at rt. After 2 h, the opaque, slightly
yellow solution was diluted with 300 mL MeOH and
stored at 4 ꢀC overnight. Precipitated glycine was filtered
off (1.8 g, 80% of starting glycine) and the filtrate was
evaporated. The crude product was purified on a silica
column (CH2Cl2/MeOH/NH3 (30% in H2O) gradient
75:20:5–10:10:1) to yield 569 mg D-syn-14; yield 81%;
de 87% (syn); ee syn>99% (D), anti>99% (D); for HPLC
4.4.14. L-3-(4-Methylsulfonylphenyl)serine L-14. HPLC:
OPA/MCE: buffer pH 8 (50 mM KH2PO4)/CH3CN¼71:29,
0.8 mL/min, tsyn¼5.8 min, tanti¼6.7 min; OPA/NAC: buffer
pH 8 (50 mM KH2PO4)/CH3CN¼81:19, 0.8 mL/min,
tD ¼6.1 min, tL ¼6.8 min, tD ¼8.0 min, tL
¼
-syn
-syn
-anti
-anti
1
8.1 min; H NMR (200 MHz, D2O): d 3.19 (s, 3H), 3.86
(d, 0.76H, syn, J¼4.4 Hz), 4.04 (d, 0.24H, anti, J¼3.6 Hz),
5.31 (d, 0.76H, syn, J¼4.4 Hz), 5.38 (d, 0.24H, anti,
J¼3.6 Hz), 7.61 (m, 2H), 7.91 (m, 2H).83,91
1
and H NMR, see Section 4.4.14; 13C NMR (125 MHz,
D2O): syn: d 43.5, 60.7, 71.0, 127.5, 127.9, 138.8,
146.5, 171.8.
4.4.15. L-2-Amino-3-hydroxy-3-(4-sulfamoylphenyl)-
propionic acid L-15. The analytical yield was calculated
by measuring the ratio of the remaining glycine and the
product, and the ratio of syn/anti was determined by integra-
tion of the corresponding Ca-proton of the diastereomers;
4.5.3. D-2-Amino-3-hydroxy-3-(4-sulfamoylphenyl)-
propanoic acid D-15. To a solution of glycine (1.13 g,
15 mmol), PLP (0.2 mg, 0.75 mmol), MnCl2 (0.09 mg,
0.75 mmol), and DTA (345 U) in 15 mL buffer (glycine

J. Steinreiber et al. / Tetrahedron 63 (2007) 918–926
925
11. Brunner, M.; Koskinen, A. M. P. Curr. Org. Chem. 2004, 8,
1629.
100 mM/NaCl 50 mM, pH 9.5) was added 4-formylbenz-
enesulfonamide (250 mg, 1.3 mmol) and stirred at rt. After
4 h, the opaque, slightly yellow solution was diluted with
150 mL MeOH and stored at 4 ꢀC overnight. Precipitated
glycine was filtered off (0.59 g, 52% of starting glycine)
and the filtrate was evaporated. The crude product was puri-
fied on a silica column (CH2Cl2/MeOH/NH3 (30% in H2O)
gradient 75:20:5–10:10:1) to yield 205 mg D-syn-15; yield
12. Ford, P. W.; Gustafson, K. R.; McKee, T. C.; Shigematsu, N.;
Maurizi, L. K.; Pannell, L. K.; Williams, D. E.; Dilip de
Silva, E.; Lassota, P.; Allen, T. M.; Van Soest, R.; Andersen,
R. J.; Boyd, M. R. J. Am. Chem. Soc. 1999, 121, 5899.
13. Calimsiz, S.; Morales Ramos, A. I.; Lipton, M. A. J. Org.
Chem. 2006, 71, 6351.
61%; de >95% (syn); for 1H NMR, see Section 4.4.15; 13
C
NMR (125 MHz, D2O): syn: d 60.9, 71.4, 126.5, 127.2,
141.1, 145.1, 172.6.
14. Chen, Z.; Ye, T. Synlett 2005, 2781.
15. Nagamitsu, T.; Sunazuka, T.; Tanaka, H.; Omura, S.;
Sprengeler, P. A.; Smith, A. B., III. J. Am. Chem. Soc. 1996,
118, 3584.
16. Sugiyama, H.; Shioiri, T.; Yokokawa, F. Tetrahedron Lett.
2002, 43, 3489.
4.6. L-3-(4-Methylsulfonylphenyl)serine L-14
17. Hansen, D. B.; Joullie, M. M. Tetrahedron: Asymmetry 2005,
16, 3963.
18. Hansen, D. B.; Lewis, A. S.; Gavalas, S. J.; Joullie, M. M.
Tetrahedron: Asymmetry 2006, 17, 15.
19. Alonso, M.; Riera, A. Tetrahedron: Asymmetry 2005, 16,
3908.
20. Kim, I. H.; Kirk, K. L. Tetrahedron Lett. 2001, 42, 8401.
21. Noyori, R.; Ikeda, T.; Ohkuma, T.; Widhalm, M.; Kitamura,
M.; Takaya, H.; Akutagawa, S.; Sayo, N.; Saito, T. J. Am.
Chem. Soc. 1989, 111, 9134.
To a solution of glycine (2.25 g, 30 mmol), PLP (0.4 mg,
1.5 mmol), and LTA (2300 U) in 30 mL buffer (KH2PO4,
50 mM, pH 8.0) was added 4-(methylsulfonyl)benzaldehyde
(500 mg, 2.7 mmol) under stirring at rt. The opaque, slightly
yellow solution was diluted with 300 mL MeOH after 2 h
and incubated at 4 ꢀC overnight. Excess glycine precipitated
and was filtered off (1.8 g, 80% of starting glycine) and the
filtrate was evaporated. The crude product was purified on
a silica column (CH2Cl2/MeOH/NH3 (30% in H2O) gradient
75:20:5–10:10:1) to yield 526 mg L-syn/anti-14; yield 75%;
de 2(syn); ee syn>99% (L), anti>99% (L); for HPLC and
1H NMR, see Section 4.4.14; 13C NMR (125 MHz, D2O):
syn+anti: d 43.4, 43.5, 59.6, 60.7, 70.7, 71.0, 127.5 (2),
127.6, 127.9, 138.8 (2), 144.7, 146.5, 170.8, 171.8.
22. Genet, J. P.; Pinel, C.; Mallart, S.; Juge, S.; Thorimbert, S.;
Laffitte, J. A. Tetrahedron: Asymmetry 1991, 2, 555.
23. Mordant, C.; Duenkelmann, P.; Ratovelomanana-Vidal, V.;
Genet, J.-P. Eur. J. Org. Chem. 2004, 3017.
24. Mordant, C.; Duenkelmann, P.; Ratovelomanana-Vidal, V.;
Genet, J.-P. Chem. Commun. 2004, 1296.
25. Makino, K.; Goto, T.; Hiroki, Y.; Hamada, Y. Angew. Chem.,
Int. Ed. 2004, 43, 882.
Acknowledgements
¨
26. Makino, K.; Hiroki, Y.; Hamada, Y. J. Am. Chem. Soc. 2005,
127, 5784.
27. Makino, K.; Fujii, T.; Hamada, Y. Tetrahedron: Asymmetry
2006, 17, 481.
€
The Osterreichische Forschungsforderungsgesellschaft
(FFG), the Province of Styria, the Styrian Business Promo-
tion Agency (SFG), the city of Graz, the Fonds zur
Forderung der wissenschaftlichen Forschung (FWF, project
€
28. Loncaric, C.; Wulff, W. D. Org. Lett. 2001, 3, 3675.
29. Fanning, K. N.; Jamieson, A. G.; Sutherland, A. Org. Biomol.
Chem. 2005, 3, 3749.
30. Willis, M. C.; Cutting, G. A.; Piccio, V. J. D.; Durbin, M. J.;
John, M. P. Angew. Chem., Int. Ed. 2005, 44, 1543.
31. Thayumanavan, R.; Tanaka, F.; Barbas, C. F., III. Org. Lett.
2004, 6, 3541.
W901-B05 DK Molecular Enzymology), and the European
Commission within the framework of a Marie-Curie Indus-
trial Host Fellowship to M.S. and M.W. are acknowl-
edged for financial support. We would like to thank Marcel
Wubbolts and Theo Sonke for stimulating discussions.
References and notes
32. Kobayashi, J.; Nakamura, M.; Mori, Y.; Yamashita, Y.;
Kobayashi, S. J. Am. Chem. Soc. 2004, 126, 9192.
33. Mettath, S.; Srikanth, G. S. C.; Dangerfield, B. S.; Castle, S. L.
J. Org. Chem. 2004, 69, 6489.
34. Ooi, T.; Taniguchi, M.; Doda, K.; Maruoka, K. Adv. Synth.
Catal. 2004, 346, 1073.
35. Wang, Y.; Chen, Z.; Mi, A.; Hu, W. Chem. Commun. 2004,
2486.
36. Crich, D.; Banerjee, A. J. Org. Chem. 2006, 71, 7106.
37. Silvestri, M. G.; Desantis, G.; Mitchell, M.; Wong, C. H. Topics
in Stereochemistry; Denmark, S. E., Ed.; Wiley-VCH:
Weinheim, 2003; Vol. 23, pp 267–342.
38. Fessner, W.-D. Modern Aldol Reactions; Mahrwald, R., Ed.;
Wiley-VCH: Weinheim, 2004; pp 201–272.
39. Kubota, K.; Nakazawa, H.; Enei, H.; Okumura, S. Japanese
Patent 51,067,788, 1976; Chem. Abstr. 1976, 85, 157975.
40. Cable, K. M.; Herbert, R. B.; Mann, J. J. Chem. Soc., Perkin
Trans. 1 1987, 1593.
1. Breuer, M.; Ditrich, K.; Habicher, T.; Hauer, B.; Kesseler, M.;
Stuermer, R.; Zelinski, T. Angew. Chem., Int. Ed. 2004, 43, 788.
2. Panke, S.; Wubbolts, M. Curr. Opin. Chem. Biol. 2005, 9, 188.
3. Panke, S.; Held, M.; Wubbolts, M. Curr. Opin. Biotechnol.
2004, 15, 272.
4. Schoemaker, H. E.; Mink, D.; Wubbolts, M. G. Science 2003,
299, 1694.
5. Hajra, S.; Karmakar, A.; Maji, T.; Medda, A. K. Tetrahedron
2006, 62, 8959.
6. George, S.; Narina, S. V.; Sudalai, A. Tetrahedron 2006, 62,
10202.
7. Fisch, A.; Bryskier, A. Antimicrobial Agents; Bryskier, A., Ed.;
ASM: Washington, DC, 2005; pp 925–929.
€
8. Nicolaou, K. C.; Boddy, C. N. C.; Brase, S.; Winssinger, N.
Angew. Chem., Int. Ed. 1999, 38, 2097.
9. Kang, S. H.; Kang, S. Y.; Lee, H.-S.; Buglass, A. J. Chem. Rev.
2005, 105, 4537.
41. Liu, J. Q.; Dairi, T.; Itoh, N.; Kataoka, M.; Shimizu, S.;
Yamada, H. J. Mol. Catal. B: Enzym. 2000, 10, 107.
10. Ohfune, Y.; Shinada, T. Eur. J. Org. Chem. 2005, 5127.

926
J. Steinreiber et al. / Tetrahedron 63 (2007) 918–926
42. Ogawa, H.; Gomi, T.; Fujioka, M. Int. J. Biochem. Cell Biol.
2000, 32, 289.
68. Reisinger, C.; van Assema, F.; Schuermann, M.; Hussain, Z.;
Remler, P.; Schwab, H. J. Mol. Catal. B: Enzym. 2006, 39, 149.
69. Saeed, A.; Young, D. W. Tetrahedron 1992, 48, 2507.
70. Molnar-Perl, I. J. Chromatogr., A 2001, 913, 283.
71. Zuman, P. Chem. Rev. 2004, 104, 3217.
72. International Drug Directory, 16th ed.; Medpharm: Stuttgart,
1995.
73. Roeder, T. Prog. Neurobiol. 1999, 59, 533.
74. Maruyama, W.; Naoi, M.; Narabayashi, H. J. Neurol. Sci. 1996,
139, 141.
75. Eisenhofer, G.; Kopin Irwin, J.; Goldstein David, S.
Pharmacol. Rev. 2004, 56, 331.
76. Carl, J. C.; Myers, T. R.; Kirchner, H. L.; Kercsmar, C. M.
J. Pediatr. 2003, 143, 731.
77. Waldeck, B. Eur. J. Pharmacol. 2002, 445, 1.
78. Plattner, J. J.; Marcotte, P. A.; Kleinert, H. D.; Stein, H. H.;
Greer, J.; Bolis, G.; Fung, A. K. L.; Bopp, B. A.; Luly, J. R.
J. Med. Chem. 1988, 31, 2277.
43. Szebenyi, D. M. E.; Musayev, F. N.; Di Salvo, M. L.; Safo,
M. K.; Schirch, V. Biochemistry 2004, 43, 6865.
44. Contestabile, R.; Paiardini, A.; Pascarella, S.; Di Salvo, M. L.;
D’Aguanno, S.; Bossa, F. Eur. J. Biochem. 2001, 268, 6508.
45. Schirch, V.; Szebenyi, D. M. E. Curr. Opin. Chem. Biol. 2005,
9, 482.
46. Bruns, F. H.; Fiedler, L. Nature 1958, 181, 1533.
47. Gibbs, R. G.; Morris, J. G. Biochim. Biophys. Acta 1964,
85, 501.
48. John, R. A. Biochim. Biophys. Acta 1995, 1248, 81.
49. Percudani, R.; Peracchi, A. EMBO Rep. 2003, 4, 850.
50. Jander, G.; Norris, S. R.; Joshi, V.; Fraga, M.; Rugg, A.; Yu, S.;
Li, L.; Last, R. L. Plant J. 2004, 39, 465.
51. Edgar, A. J. BMC Genomics 2005, 6, 32.
52. Kielkopf, C. L.; Burley, S. K. Biochemistry 2002, 41, 11711.
53. pdb-code: 1svv: from Leishmania major.
54. Vassilev, V. P.; Uchiyama, T.; Kajimoto, T.; Wong, C.-H.
Tetrahedron Lett. 1995, 36, 4081.
79. Lo, L.-C.; Yang, C.-T.; Tsai, C.-S. J. Org. Chem. 2002, 67,
1368.
55. Vassilev, V. P.; Kajimoto, T.; Wong, C.-H. RIKEN Rev. 1995,
8, 25.
80. Miller, E.; Sprague, J. M.; Kissinger, L. W.; McBurney, L. F.
J. Am. Chem. Soc. 1940, 62, 2099.
56. Kimura, T.; Vassilev, V. P.; Shen, G.-J.; Wong, C.-H. J. Am.
Chem. Soc. 1997, 119, 11734.
81. Yang, L.-M.; Lin, S.-J.; Hsu, F.-L.; Yang, T.-H. Bioorg. Med.
Chem. Lett. 2002, 12, 1013.
57. Vassilev, V. P.; Uchiyama, T.; Kajimoto, T.; Wong, C.-H.
Tetrahedron Lett. 1995, 36, 5063.
58. Shibata, K.; Shingu, K.; Vassilev, V. P.; Nishide, K.; Fujita, T.;
Node, M.; Kajimoto, T.; Wong, C.-H. Tetrahedron Lett. 1996,
37, 2791.
59. Nishide, K.; Shibata, K.; Fujita, T.; Kajimoto, T.; Wong, C.-H.;
Node, M. Heterocycles 2000, 52, 1191.
60. Misono, H.; Maeda, H.; Tuda, K.; Ueshima, S.; Miyazaki, N.;
Nagata, S. Appl. Environ. Microbiol. 2005, 71, 4602.
61. Liu, J. Q.; Dairi, T.; Itoh, N.; Kataoka, M.; Shimizu, S. Appl.
Microbiol. Biotechnol. 2003, 62, 53.
82. Van Es, T.; Staskun, B. Org. Synth. 1971, 51, 20.
83. Kaptein, B.; van Dooren, T. J. G. M.; Boesten, W. H. J.; Sonke,
T.; Duchateau, A. L. L.; Broxterman, Q. B.; Kamphuis, J. Org.
Process Res. Dev. 1998, 2, 10.
84. Weber, H.; Brecker, L. Curr. Opin. Biotechnol. 2000, 11,
572.
85. Chmurny, A. B.; Gross, A. T.; Kupper, R. J.; Woodbury, R. P.;
Roberts, R. L. EP 220923 A2, 1987; Chem. Abstr. 1987, 108,
36502.
86. Mita, R.; Katoh, T.; Higuchi, C.; Yamaguchi, A. U.S. Patent
4,605,759, 1985; Chem. Abstr. 1985, 103, 215800.
87. Soloshonok, V. A.; Kukhar, V. P.; Galushko, S. V.; Svistunova,
N. Y.; Avilov, D. V.; Kuzmina, N. A.; Raevski, N. I.; Struchkov,
Y. T.; Pysarevsky, A. P.; Belokon, Y. N. J. Chem. Soc., Perkin
Trans. 1 1993, 3143.
62. (a) Seebeck, F. P.; Hilvert, D. J. Am. Chem. Soc. 2003, 125,
10158; (b) Seebeck, F. P.; Guainazzi, A.; Amoreira, C.;
Baldridge, K. K.; Hilvert, D. Angew. Chem., Int. Ed. 2006,
45, 6824.
63. Herbert, R. B.; Wilkinson, B.; Ellames, G. J.; Kunec, E. K.
J. Chem. Soc., Chem. Commun. 1993, 205.
88. Jiang, Y.; Wu, S.; Chen, D.; Ma, Y.; Liu, G. Tetrahedron 1988,
44, 5343.
64. Liu, J. Q.; Odani, M.; Dairi, T.; Itoh, N.; Shimizu, S.; Yamada,
H. Appl. Microbiol. Biotechnol. 1999, 51, 586.
65. Liu, J. Q.; Odani, M.; Yasuoka, T.; Dairi, T.; Itoh, N.; Kataoka,
M.; Shimizu, S.; Yamada, H. Appl. Microbiol. Biotechnol.
2000, 54, 44.
66. Lotz, B. T.; Gasparski, C. M.; Peterson, K.; Miller, M. J.
J. Chem. Soc., Chem. Commun. 1990, 1107.
67. Liu, J. Q.; Ito, S.; Dairi, T.; Itoh, N.; Kataoka, M.; Shimizu, S.;
Yamada, H. Appl. Environ. Microbiol. 1998, 64, 549.
89. Herbert, R. B.; Wilkinson, B.; Ellames, G. J. Can. J. Chem.
1994, 72, 114.
90. Blaser, D.; Seebach, D. Liebigs Ann. Chem. 1991, 1067.
91. NMR shifts are pH dependent, and thus published data from
DCl/D2O or CF3COOD/D2O are significantly different.
92. Belokon, Y. N.; Bulychev, A. G.; Vitt, S. V.; Struchkov, Y. T.;
Batsanov, A. S.; Timofeeva, T. V.; Tsyryapkin, V. A.;
Ryzhov, M. G.; Lysova, L. A. J. Am. Chem. Soc. 1985, 107,
4252.