Chemical and enzymatic methodologies for the synthesis of enantiomerically pure glyceraldehyde 3-phosphates

Article abstract of DOI:10.1016/j.carres.2013.12.023

Glyceraldehyde 3-phosphates are important intermediates of many central metabolic pathways in a large number of living organisms. d-Glyceraldehyde 3-phosphate (d-GAP) is a key intermediate during glycolysis and can as well be found in a variety of other metabolic pathways. The opposite enantiomer, l-glyceraldehyde 3-phosphate (l-GAP), has been found in a few exciting new pathways. Here, improved syntheses of enantiomerically pure glyceraldehyde 3-phosphates are reported. While d-GAP was synthesized by periodate cleavage of d-fructose 6-phosphate, l-GAP was obtained by enzymatic phosphorylation of l-glyceraldehyde. ¹H- and ³¹P NMR spectroscopy was applied in order to examine pH-dependent behavior of GAP over time and to identify potential degradation products. It was found that GAP is stable in acidic aqueous solution below pH 4. At pH 7, methylglyoxal is formed, whereas under alkaline conditions, the formation of lactic acid could be observed.

Full text of DOI:10.1016/j.carres.2013.12.023

Carbohydrate Research 389 (2014) 18–24
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Chemical and enzymatic methodologies for the synthesis
of enantiomerically pure glyceraldehyde 3-phosphates
⇑
Dominik Gauss, Bernhard Schoenenberger, Roland Wohlgemuth
Sigma–Aldrich GmbH, Industriestrasse 25, 9471 Buchs, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
Glyceraldehyde 3-phosphates are important intermediates of many central metabolic pathways in a large
Received 14 September 2013
Received in revised form 21 December 2013
Accepted 27 December 2013
Available online 10 January 2014
number of living organisms.
glycolysis and can as well be found in a variety of other metabolic pathways. The opposite enantiomer,
-glyceraldehyde 3-phosphate ( -GAP), has been found in a few exciting new pathways. Here, improved
syntheses of enantiomerically pure glyceraldehyde 3-phosphates are reported. While -GAP was synthe-
sized by periodate cleavage of -fructose 6-phosphate, -GAP was obtained by enzymatic phosphorylation
of
-glyceraldehyde. 1H- and 31P NMR spectroscopy was applied in order to examine pH-dependent
behavior of GAP over time and to identify potential degradation products. It was found that GAP is stable
in acidic aqueous solution below pH 4. At pH 7, methylglyoxal is formed, whereas under alkaline
conditions, the formation of lactic acid could be observed.
D-Glyceraldehyde 3-phosphate (D-GAP) is a key intermediate during
L
L
D
D
L
Keywords:
Glycolysis
Glycerokinase
Enzymatic phosphorylation
L
D-Glyceraldehyde 3-phosphate
Ó 2014 Elsevier Ltd. All rights reserved.
L-Glyceraldehyde 3-phosphate
1. Introduction
The development of biochemistry is very much related to key
role not only in glycolysis and gluconeogenesis, but in a variety
of additional metabolic pathways like for example the meth-
ylerythritol phosphate (MEP) pathway,11,12 photosynthetic carbon
discoveries made in the enzymatic reactions with
D-glucose in
fixation,13,14 the shikimate pathway,15,16 clavulanic acid biosynthe-
yeast extract by Hans and Eduard Buchner,1 Arthur Harden, and
William Young2 as well as in muscle extracts by the classical work
of Gustav Embden,3 Otto Meyerhof,4 Carl Neuberg,5 Jacob Parnas,6
Otto Warburg,7 Gerty and Carl Cori.8 The sequence of enzymatic
sis,17 and thiamine metabolism.18 In contrast, the
L
-glyceraldehyde
-GAP) has not been found to play the corresponding
-glucose catabolic pathway in Paracococcus sp.19 and the
discovery of YghZ in Escherichia coli catalyzing the stereospecific
reduction of -GAP to
-Glycerol 3-phosphate20 has provided the
first evidence of a metabolic bypass for the removal of toxic -GAP.
3-phosphate (
L
role in a
L
reaction steps for the complete anaerobic
D
-glucose catabolic path-
L
L
way in eukaryotes, with the corresponding chemical structure of
the metabolites, has been established with the Embden–Meyer-
hof–Parnas pathway (EMP). The natural variety of alternative
L
Otto Warburg observed already in 1924 that cancer cells metab-
olize glucose into lactate even in the presence of sufficient oxygen
to support mitochondrial oxidative phosphorylation,21–23 a finding
which has obtained renewed attention.24,25
D
-glucose metabolic pathways9 is of continuing interest in pro-
karyotes, with the Entner–Doudoroff pathway (ED) as the most
common, as well as in archaea, since in a phylogenetic analysis
of genomes the genes of 30% of the organisms could not be anno-
tated to either the EMP- or the ED-pathway.10 In order to elucidate
the reaction topologies and protein functions under different con-
ditions, especially when branchpoints are involved, the prepara-
tion of the corresponding enzyme substrates is important. The
availability of all relevant metabolites from the different glycolytic
pathways in pure and stable form as well as comprehensive knowl-
edge about their stabilities has been a challenge up to the present.
The first synthesis of racemic glyceraldehyde 3-phosphate, later
to be known as Fischer–Baer ester, had a considerable effect on the
development of the Embden–Meyerhof scheme and enabled stud-
ies of its properties.26–28 Since
D- and L-glyceraldehyde 3-phos-
phates have different biological properties, the synthesis of pure
enantiomers is highly important.29–31 Whereas there are numerous
papers dealing with syntheses of racemic GAP, the number of pub-
lished protocols leading to enantiomerically pure or enriched
D- or
L-GAP is small and all of them have drawbacks which make them
The central metabolite
shared by the EMP and ED pathways in lower glycolysis, plays a
D
-glyceraldehyde 3-phosphate (
D
-GAP),
unattractive for lab scale production purposes.
The oldest published preparative formation of D-GAP is a lengthy
multistep synthesis using mercury salts in one of the deprotection
steps.29 An attractive one step synthesis31 starting with
-fructose
6-phosphate does not consider the pH lability of the product and
⇑
D
Corresponding author. Tel.: +41 81 7552640; fax: +41 81 7552840.
E-mail address: roland.wohlgemuth@sial.com (R. Wohlgemuth).
http://dx.doi.org/10.1016/j.carres.2013.12.023
0008-6215/Ó 2014 Elsevier Ltd. All rights reserved.

D. Gauss et al. / Carbohydrate Research 389 (2014) 18–24
19
gave in our hands varying yields and product purities. Another oxi-
dative degradation of -fructose 6-phosphate uses toxic and terato-
genic lead-tetraacetate and requires extensive and non-scaleable
purification at near neutral pH implying partial product degrada-
tion.40,43 Alternatively the product was not purified at all but used
directly in a next reaction step.35 A disadvantage of the enzy-
GAP is transformed to methylglyoxal via an enediolate phos-
phate transition product in water at pH 7 (Scheme 1).32,33,41 In
alkaline conditions, lactic acid is formed.26,28,34 Figure 1 shows
the comparison of 1H NMR spectra of lactic acid and DL-GAP at
strongly alkaline conditions in D2O. The chemical shift values are
exactly matching confirming the formation of lactic acid which
has not been described in a previous kinetic analysis.33 However,
the signal at 4.1 ppm, which can be assigned to the CHOH proton,
shows a different coupling pattern. While the signal is split up into
a quartet in case of lactic acid resulting from coupling to the adja-
cent methyl group, it appears as a triplet in the spectrum of the
DL-GAP degradation product under alkaline conditions.
D
matic synthesis of D-GAP using an aldolase-catalyzed cleavage of
D
-fructose 1,6-diphosphate is the often incomplete conversion due
to the thermodynamic equilibrium of this reaction.42 Therefore a
considerable purification effort is necessary in order to obtain a pure
product. The degree of purity of the isolated
type of aldolase used was not specified.42
D-GAP as well as the
In the
L
-GAP synthesis, the disadvantages of using racemic glyc-
Although the mechanism of the conversion from GAP to lactic
acid is not definitely known, a possible explanation for the dif-
ferent coupling patterns is the deuteration of the newly formed
methyl group by the deuterated solvent. This assumption is con-
firmed by the ratio of the observed integral values (only two
protons in the case of the deuterated methyl group) and the
weak H/D-coupling of the deuterated lactic acid methyl group
(upper left in Fig. 1). There are two more interesting observa-
tions worth to note regarding the degradation of DL-GAP: (1) if
the compound is left in neutral, aqueous solution for several
days, methylglyoxal signals are appearing in the 1H NMR spec-
trum, but no formation of lactic acid can be observed and (2)
methylglyoxal itself does not show any formation of lactic acid
even after 15 h under alkaline conditions. Therefore, the decom-
position mechanism of GAP under alkaline conditions is different
from the one under neutral conditions and methylglyoxal does
not seem to be an intermediate in the non-enzymatic degrada-
tion of GAP at high pH-values.
In order to identify the pH-range where GAP is stable and to get
first insights into degradation velocities, NMR spectra of aqueous
solutions of DL-GAP at different pH values were recorded. To quan-
tify the amount of remaining DL-GAP, the percentage of the inte-
grals in the 31P NMR spectra were plotted against time in
solution (Fig. 2). As discussed earlier, in both routes of non-enzy-
matic GAP-degradation, the compound is dephosphorylated and
inorganic phosphate is released.
eraldehyde as starting material for the glycerokinase-catalyzed
phosphorylation with the phospho-enolpyruvate/pyruvate kinase
system for ATP recycling35 will be outlined later in this paper.
Scale-up of this procedure resulted in diminished product yields.
A multi-enzyme procedure directed toward kinetic resolution of
racemic GAP with D-glyceraldehyde-3-phosphate dehydrogenase
suffers from the same drawback.32
A lengthy approach for the preparation of both enantiomers of
GAP was the enzymatic resolution of racemic glycidaldehyde die-
thylacetal via a series of reaction steps. As the resulting (R)- or
(S)-glycidaldehyde diethylacetal respectively was subjected to a
nucleophilic phosphorylation in a pH range which is unfavorable
for the stability of GAP, no further purification was attempted
and the material was used directly for the following reaction step
instead.38,39
2. Results and discussion
2.1. Stability of DL-GAP
The instability of GAP under alkaline or neutral pH-conditions
has been known for a very long time. Nevertheless, aqueous condi-
tions in the neutral pH-range are used in nearly all published pro-
cedures dealing with synthesis, purification, or application of the
compound. Here, 1H and 31P NMR experiments were used in order
to identify degradation products and to examine pH-dependent
behavior of GAP over time. Furthermore, these results could pro-
vide a closer insight in terms of stability and degradation mecha-
nisms of GAP.
It can be concluded that the product is stable at least for days
under strongly acidic conditions. After 4 days at pH 4.2, degrada-
tion can be observed at a low level. Former experiments indicated
full stability at pH 3.5. At pH 5.6 more than 50% of the compound is
Scheme 1. Non-enzymatic degradation of
intermediate is analogous to -GAP.
D
-GAP to methylglyoxal and inorganic phosphate in water at pH 7.33 The formation mechanism of the enediolate phosphate
L

20
D. Gauss et al. / Carbohydrate Research 389 (2014) 18–24
Figure 1. 1H NMR spectra of lactic acid (top) and DL-GAP (bottom) in D2O under alkaline conditions. DL-GAP is obviously converted into lactic acid although the signal at
4.1 ppm shows a different coupling pattern (lower left part of the spectra). The CHOH-signal, split up to a quartet by the methyl group, appears as a triplet in the DL-GAP
spectrum, probably because of the presence of a CH2D-functionality. This is confirmed by the extract on the upper left, where, in contrast to the non-deuterated lactic acid, a
small H/D-coupling can be seen in the methyl group signal of the deuterated lactic acid.
the calcium salt of D-GAP (Scheme 2). As discussed earlier, the
compound is degrading at pH-values above 4. Therefore, the pH-
value was always kept below 4 during the whole procedure. An-
other modification compared to the published procedure was the
use of acetone instead of ethanol for precipitation of the calcium
salt because the latter could not be removed completely under
reduced pressure.
The D-GAP calcium salt is only soluble under strongly acidic
conditions. Enzymatic determination of the content resulted in val-
ues of about 50% (w/w). The calcium salt was transferred to a solu-
tion of its free acid using cation exchange chromatography.
2.3. Enzymatic synthesis of L-GAP
Since no chemical synthesis of
approach for the phosphorylation of
been followed to yield enantiomerically pure
described35 that glycerokinase from Cellulomonas sp. (EC 2.7.1.30)
L
-GAP is known, an enzymatic
-glyceraldehyde ( -GA) has
-GAP. It was
Figure 2. 31P NMR integral values of DL-GAP are plotted as a function of time at
different pH-values. The compound is stable under acidic conditions only and
decomposes above a pH-value of 4. The pH 1.4-curve is nearly completely covered
by the pH 2.6-curve.
L
L
L
has the ability to phosphorylate selectively only the L-isomer of
DL-glyceraldehyde with adenosine 50-triphosphate (ATP) as phos-
phorylation reagent. ATP was continuously rephosphorylated with
phosphoenolpyruvic acid (PEP) and pyruvate kinase from rabbit
muscle (EC 2.7.1.40) (Scheme 3) and used in a molar ratio of
degraded after 4 days. Under alkaline conditions (pH 12.2), less
than half of the DL-GAP amount is remaining after 1 h.
2.2. Chemical synthesis of
D
-GAP
1:100 compared to L-GA and PEP.
Synthesis of enantiomerically pure
modified procedure published by Szewczuk et al.,31 where
-fructose 6-phosphate was cleaved with periodic acid to obtain
D
-GAP was achieved using a
2.3.1. Evaluation of the substrate
-Glyceraldehyde ( -GA) is described to catalyze the dephos-
phorylation of ATP in the presence of glycerokinase.36 We therefore
D
D
D

D. Gauss et al. / Carbohydrate Research 389 (2014) 18–24
21
Scheme 2. Chemical synthesis of
D
-glyceraldehyde 3-phosphate. Step one is a modification of an already published method31 to obtain the calcium salt of
D
-GAP. In a second
step, the calcium salt was converted into the corresponding free acid of D-GAP by ion exchange.
Scheme 3. Phosphorylation of L-glyceraldehyde to enantiomerically pure L-glyceraldehyde 3-phosphate using a coupled enzymatic system.
investigated the use of pure
D
- and
L
-enantiomers as starting mate-
2.3.2. Reaction optimization and scale-up
-Glyceraldehyde is the starting material of choice for the
straight forward preparation of enantiomerically pure -GAP. How-
rials for the enzymatic phosphorylation with glycerokinase by 31P
L
NMR spectroscopy.
L
Using enantiomerically pure
D
-GA as starting material, no phos-
ever, yields of around 30% are not satisfying for converting the
reaction into a preparative scale. Modifying the reaction conditions
like temperature, pH-value, or concentration of the reaction part-
ners is often a simple way to speed up an enzymatic reaction or
to stabilize the desired product.
phorylation can be observed in the 31P NMR (Fig. 3C). However,
there was a significant formation of inorganic phosphate, in agree-
ment with the observations of Hayashi.36 A significant formation of
inorganic phosphate as a result of the dephosphorylation of L-GAP
in neutral solution is unlikely in view of the degradation kinetics of
DL-GAP in neutral solution (see Fig. 2).
The optimum pH-values of the used enzymes are pH 7.5 for the
pyruvate kinase from rabbit muscle37 and pH 7.3–9.8 for the glyc-
erokinase from Cellulomonas sp.36 respectively. Experiments at
slightly acidic conditions (pH 4.8) resulted in no formation of the
desired product. Regarding the instability of GAP at higher pH-val-
ues, neutral conditions seemed to be a good compromise between
activities of the enzymes and the stability of the desired product.
This is confirmed by the experiment where pure
L-GA is used as
starting material (Fig. 3B): -GAP is formed in a significant amount
L
with a much slower generation of inorganic phosphate. The latter
can be explained by the instability of the compound in neutral
solution (see Section 2.1).
With a racemic mixture of
formed in a considerably lower amount compared to the use of
the pure -enantiomer, whereas a higher amount of inorganic
phosphate formation can be observed (Fig. 3A). If the inorganic
phosphate would solely originate from the dephosphorylation of
D
- and
L
-GA, the desired product is
As shown in Figure 3 the formation of L-GAP from L-GA is dimin-
ished to a large extent by its degradation due to the low enzyme
concentrations used. Therefore, higher concentrations of pyruvate
kinase and glycerokinase were used and the reaction was again
monitored by 31P NMR spectroscopy. A tenfold increase in enzyme
L
the
racemic GA compared to pure
experimental results. We therefore conclude that glycerokinase
from Cellulomonas sp. is selectively phosphorylating the -GA
-GA is catalyzing the dephosphorylation
L
-GAP formed, only half of the amount would be expected using
concentration already leads to >90% conversion to
reaction time of about two hours.
In a larger scale preparation, the reaction was quenched by add-
ing aqueous HCl. A pH decrease to pH 2–3 resulted in a reaction
stop and the product is kept under stable conditions. After removal
L-GAP after a
L
-GA. This is in contrast to our
L
to form
of ATP.
L-GAP while D
of the enzymes using ultracentrifugal filter units, the L-GAP was

22
D. Gauss et al. / Carbohydrate Research 389 (2014) 18–24
lactic acid could be observed. Methylglyoxal does not seem to be
an intermediate in non-enzymatic lactic acid formation.
To obtain the two enantiomerically pure isomers of GAP, two
completely different routes for synthesis have been applied: While
D
-GAP has been synthesized using periodate cleavage of
6-phosphate, -GAP has been obtained by enzymatic phosphoryla-
tion of -glyceraldehyde with glycerokinase from Cellulomonas sp.
The enzyme has been shown to phosphorylate specifically the
-isomer, whereas the -isomer remained not just nonphosphory-
lated, but moreover catalyzed the dephosphorylation of ATP.
Therefore, we have established the use of -glyceraldehyde instead
D-fructose
L
L
L
D
L
of racemic DL-glyceraldehyde for the enzymatic synthesis of
as the preferred route regarding yields and purity.
L-GAP
4. Experimental part
4.1. General
All reagents and solvents were from Sigma–Aldrich GmbH
without any further purification. NMR-spectra were recorded on
a Bruker Avance III 600 MHz spectrometer equipped with a BBO
probe head with z-gradient using 600.2 MHz for 1H, 150.9 MHz
for 13C, and 243 MHz for 31P. Spectra were recorded in D2O or
H2O/D2O at 298.2 K. Since common 31P NMR spectra are not suit-
able for quantification, a delay of 10 s between each pulse, a pulse
angle of 90° and inverse gated decoupling has been applied for the
31P NMR measurements conducted herein. Differentiation between
organic and inorganic phosphate was achieved using non proton
decoupled 31P NMR spectra. Chemical shifts were referenced to
3-(trimethyl-silyl)-propionic-2,2,3,3-d4 acid sodium salt (TSP) in
case of 1H and 13C NMR spectra (0.00 ppm) and phosphoric acid
in case of 31P NMR spectra (0.00 ppm), respectively. Optical rota-
tions were recorded on a Perkin–Elmer 241 polarimeter with the
sodium D line at 20 °C and 1 cm path length.
4.2. Stability of DL-GAP at different pH-conditions
Preparation of the NMR-samples was performed as follows:
50
600
l
l
L of aqueous DL-GAP solution (50 mg/mL) was solved in
L D2O and the particular pH value was adjusted by adding
NaOD in D2O. 1H- and 31P NMR spectra were recorded in intervals
of 1 h for the first 11 h and after 16, 22, 39, 48, 60, 72, 84, and 96 h,
respectively. For quantification of remaining DL-GAP and formed
inorganic phosphate, the sum of the two integral values was nor-
malized to 100 and the fraction of the DL-GAP signal was plotted
against time in solution.
Figure 3. 31P NMR monitoring of the enzymatic phosphorylation of different
enantiomers of glyceraldehyde. Only the
L
-isomer is phosphorylated by glyceroki-
nase from Cellulomonas sp., whereas
D-glyceraldehyde catalyzes the dephosphoryl-
ation of ATP.
4.3.
D
-Glyceraldehyde 3-phosphate calcium salt
-fructose 6-phosphate disodium salt
An aqueous solution of
D
precipitated as its calcium salt, while non-phosphorylated com-
pounds remained in solution. The white solid was washed with
acetone and hydrochloric acid, dried under reduced pressure, and
converted into an aqueous solution of its free acid using cation
exchange chromatography.
hydrate (5 g, 15.5 mmol in 155 mL of water) was cooled with ice
and a cold solution of periodic acid (12.6 g, 108.8 mmol in 56 mL
of water) was added dropwise. After stirring for 1 h, ethylene gly-
col (5 mL, 89.7 mmol) was added and the solution was stirred for
another 30 min. The pH was adjusted to 3.5 with 2 N NaOH and
150 mL of acetone was added, resulting in the formation of a white
precipitate. The suspension was stirred for 90 min and the precip-
itate was removed by filtration and washed twice with acetone. To
the clear filtrate, Ca(OAc)2 (7.9 g, 43.5 mmol in 45 mL of water)
was added. After stirring the formed suspension in an ice bath
for 90 min, the white solid was filtrated and washed with small
3. Conclusions
The pH-dependent stability of glyceraldehyde 3-phosphate
(GAP) and the syntheses of the enantiomerically pure D-GAP and
L
-GAP are reported. The stability in aqueous solution was examined
using 1H- and 31P NMR spectroscopy. GAP has been shown to be
stable below pH 4, whereas two different degradation products
appeared at higher pH-values: under neutral conditions, methylgly-
oxal is formed, whereas under alkaline conditions, formation of
portions of acetone (2ꢀ), 10 M HCl (1ꢀ), 1 mM HCl (1ꢀ), and ace-
l
tone (2ꢀ). After drying overnight under reduced pressure, 1.97 g of
a white solid was obtained which contained only a low amount of
D-GAP.

D. Gauss et al. / Carbohydrate Research 389 (2014) 18–24
23
To the clear filtrate, acetone was added and the white suspen-
sion was left at +4 °C overnight. The precipitate was filtrated and
phospho(enol)pyruvic acid monopotassium salt (1.34 g, 6.5 mmol),
and MgCl2ꢃ6H2O (13.5 mg, 66.5 mol) in 50 mL H2O/D2O (95:5) at
l
washed with small portions of acetone (2ꢀ), 10
l
M HCl (1ꢀ),
pH = 6.9, 2kU pyruvate kinase from rabbit muscle and 1kU glycero-
kinase from Cellulomonas sp. were added. The solution was stirred
at room temperature and monitored with 31P NMR spectroscopy.
After 130 min and a product content of >90%, the reaction was
stopped by adding 15 mL of 1 M aqueous HCl. After 2 h at +4 °C,
the white precipitate was separated with Amicon ultracentrifugal
filter units (molecular weight cutoff mass: 10 kDa). The clear solu-
tion was cooled to 0 °C in an ice bath and 25 mL of a 0.8 M aqueous
solution of calcium acetate monohydrate (3.5 g, 19.87 mmol) was
added. Afterward, 100 mL of acetone was added dropwise at 0 °C
to the stirred solution and a white solid precipitated.
1 mM HCl (1ꢀ), and acetone (2ꢀ). After drying overnight under
reduced pressure, 2.26 g (10.9 mmol, 70%) of a white solid was
obtained and converted into the free acid without any further
purification.
4.4.
D
-Glyceraldehyde 3-phosphate free acid solution
D-Glyceraldehyde 3-phosphate calcium salt (33 mg, 0.16 mmol)
was stirred with excess Dowex 50WX8 resin (hydrogen form) in
2 mL H2O for 10 min until the white solid was dissolved com-
pletely. The mixture was filtered over some more resin, the latter
washed with distilled water until the pH of the eluent was >4
and then concentrated under reduced pressure to yield 1.7 mL of
The mixture was stirred for an additional hour under cooling in
an ice bath and the white precipitate was filtered with suction and
washed with small portions of acetone (2ꢀ), 10 M HCl (1ꢀ),
l
a 10.9 mg/mL aqueous solution of
D-GAP (18.5 mg, 0.11 mol,
1 mM HCl (1ꢀ) and acetone (2ꢀ). After drying overnight under
69%). The low yield presumably results from a less than 100%
reduced pressure, 1.35 g of a white solid was obtained.
D
-GAP content in the starting calcium salt. ½a D20
ꢁ
+15.5° (c 1.1,
The white solid was stirred with 50 meq Dowex 50WX8 (hydro-
gen form) in 80 mL for 1 h and filtered over another 120 meq of ion
exchange resin, the latter washed with distilled water until the pH
of the eluent was >4 and concentrated under reduced pressure to
H2O); 1H NMR (600 MHz, D2O, pH 1.7): d ppm 3.70 (m, 1H, CHOH),
3.93 (m, 1H, CH2), 4.02 (m, 1H, CH2), 4.99 (d, J = 5.89 Hz, 1H, CHO);
13C NMR (151 MHz, D2O, pH 1.7): d ppm 68.87 (d, J = 5.7 Hz, CH2),
75.95 (d, J = 7.8 Hz, CHOH), 92.31 (s, CHO); 31P NMR (243 MHz,
D2O, pH 1.7, decoupled): d ppm 0.58 (s, H2PO4); 31P NMR (243
MHz, D2O, pH 1.7, non-decoupled): d ppm 0.58 (t, J = 5.8 Hz,
H2PO4).
yield 21.7 mL of a 39 mg/mL aqueous solution of L-GAP (0.85 g,
5 mmol, 77%). For storage and analysis, the solution was diluted
to a concentration of 8.6 mg/mL. ½a D20
ꢁ
ꢄ12.8° (c 0.86, H2O); 1H
NMR (600 MHz, D2O, pH 1.75): d ppm 3.70 (m, 1H, CHOH), 3.92
(m, 1H, CH2), 4.02 (m, 1H, CH2), 4.99 (d, J = 5.66 Hz, 1H, CHO);
13C NMR (151 MHz, D2O, pH 1.75): d ppm 68.84 (d, J = 5.7 Hz,
CH2), 75.95 (d, J = 7.88 Hz, CHOH), 92.30 (s, CHO); 31P NMR
(243 MHz, D2O, pH 1.75, decoupled): d ppm 0.06 (s, H2PO4); 31P
NMR (243 MHz, D2O, pH 1.75, non-decoupled): d ppm 0.06 (t,
J = 5.8 Hz, H2PO4).
The product content in solution was determined using the
following enzymatic procedure:
The aqueous solution was diluted in 0.1 M KH2PO4 buffer (pH
5.5) and 20
(pH 6.5) containing EDTA (20 mM) and b-NADH (0.15 mM).
Absorption at 340 nm was measured and after addition of 3 L of
-glycerophosphate dehydrogenase (75–200 U/mg) and triose-
lL thereof was added to 1 mL of a 0.2 M KH2PO4 buffer
l
a
phosphate isomerase (750–2000 U/mg) from rabbit muscle (sus-
pension in 3.2 M ammonium sulfate, pH = 6, EC 1.1.1.8), the
absorption at 340 nm was measured again.
Acknowledgements
We thank Wolfgang Hass for the improvement in the enzymatic
assay of D-GAP. The support of the BMBF Project Biokatalyse2021
P28 and Sigma–Aldrich is gratefully acknowledged.
The concentration was calculated according to c [mg/mL] =
(A V F M)/(eꢂv) with A = Difference of the two absorption values
ꢂ
ꢂ ꢂ
at 340 nm, V = total Volume [
weight of -GAP [mg/ mol],
mol), and = volume of added D
l
e
L], F = dilution factor, M = molecular
= extinction coefficient (6.22 cm2/
-GAP-solution.
D
l
Supplementary data
l
v
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.carres.
2013.12.023.
4.5. Comparison of DL-,
materials for the enzymatic synthesis of
L
- and
D
-Glyceraldehyde as starting
L-GAP
To prepare the NMR-samples for the quantitative examination
of the enzymatic phosphorylation of glyceraldehyde at the 3-posi-
tion, aqueous solutions of the appropriate enantiomer of glyceral-
dehyde (100 mM final concentration) were mixed with aqueous
solutions of phospho(enol)pyruvic acid monopotassium salt
(100 mM final concentration), adenosine 50-triphosphate disodium
salt hydrate (1 mM final concentration), MgCl2 (1 mM final concen-
tration), and triethylphosphine oxide as 31P NMR reference (20 mM
final concentration) in a mixture of H2O/D2O (final ratio: 9:1). The
pH was adjusted to 7.0 and glycerokinase from Cellulomonas sp.
(1U) and pyruvate kinase from rabbit muscle (2U) were added.
31P NMR-spectra were recorded in hourly intervals for the first
twelve hours and after 24 h. For quantification, the sum of the inte-
grals of the arising glyceraldehyde 3-phosphate, phospho(enol)
pyruvic acid, and inorganic phosphate were normalized to 100
and the particular values were plotted against the reaction time.
References
1. Buchner, E.; Rapp, R. Ber. Dtsch. Chem. Ges. 1897, 30, 2668–2678.
2. Harden, A.; Young, W. J. Proc. R. Soc. (London) 1906, 77, 405–420.
3. Embden, G.; Deuticke, H. J.; Kraft, G. J. Mol. Med. 1933, 12, 213–215.
4. Meyerhof, O. Experientia 1948, 4, 169–176.
5. Neuberg, C.; Kobel, M. J. Bacteriol. 1934, 28, 461–471.
6. Parnas, J. K. Nature 1943, 151, 577–580.
7. Warburg, O.; Walter, C. Naturwissenschaften 1941, 29, 589–590.
8. Cori, C. T.; Cori, G. T. Ann. Rev. Biochem. 1941, 10, 151–180.
9. Bar-Even, A.; Flamholz, A.; Noor, E.; Milo, R. Nat. Chem. Biol. 2012, 8, 509–517.
10. Flamholz, A.; Noor, E.; Bar-Even, A.; Liebermeister, W.; Milo, R. Proc. Natl. Acad.
Sci. 2013, 110, 10039–10044.
11. Rohmer, M. Pure Appl. Chem. 2007, 79, 739–751.
12. Zhao, L.; Chang, W.; Xiao, Y.; Liu, H. W.; Liu, P. Ann. Rev. Biochem. 2013, 82, 497–
530.
13. Dijkhuisen, L.; Harder, W. Antonie van Leeuwenhoek 1984, 50, 473–487.
14. Hasunuma, T.; Harada, K.; Miyazawa, S.-I.; Kondo, A.; Fukusaki, E.; Miyake, C. J.
Exp. Bot. 2010, 61, 1041–1051.
15. Floss, H. G. Nat. Prod. Rep. 1997, 14, 433–452.
16. Herrmann, K. M.; Weaver, L. M. Ann. Rev. Plant Physiol. Plant. Mol. Biol. 1999, 50,
473–503.
17. Kershaw, N. J.; Caines, M. E. C.; Sleeman, M. C.; Schofield, C. J. Chem. Commun.
2005, 4251–4263.
18. Begley, T. P.; Ealick, S. E.; McLafferty, F. W. Biochem. Soc. Trans. 2012, 40, 555–
560.
4.6. L-Glyceraldehyde 3-phosphate free acid solution
To a solution of
L-glyceraldehyde (0.6 g, 6.65 mmol), adenosine
50-triphosphate disodium salt hydrate (0.04 g, 66.5
lmol),

24
D. Gauss et al. / Carbohydrate Research 389 (2014) 18–24
19. Shimizu, T.; Takaya, N.; Nakamura, A. J. Biol. Chem. 2012, 287, 40448–40456.
20. Desai, K. K.; Miller, B. G. Biochemistry 2008, 47, 7983–7985.
21. Warburg, O.; Posener, K.; Negelein, F. Biochem. Z. 1924, 152, 319–344.
22. Warburg, O. The Metabolism of Tumors; Costable: London, 1930.
23. Warburg, O. Biochem. Z. 1956, 123, 309–314.
24. Vander Heiden, M. G.; Cantley, L. C.; Thompson, C. B. Science 2009, 324, 1029–
1033.
25. Cairns, R. A.; Harris, I. S.; Mak, T. W. Nat. Rev. Cancer 2011, 11, 85–95.
26. Barrenscheen, H.; Beneschovsky, H. Biochem. Z. 1933, 265, 159–168.
27. Fischer, H. O. L.; Baer, E. Berichte 1932, 65, 1040.
33. Richard, J. P. Chem. Biol. 2008, 3, 605–607.
34. Trigalo, F.; Jachymczyk, W.; Young, J. C.; Szabo, L. J. Chem. Soc., Perkin Trans. 1
1975, 593–598.
35. Wong, C. H.; Whitesides, G. M. J. Org. Chem. 1983, 48, 3199–3205.
36. Hayashi, S.-I.; Lin, E. C. C. J. Biol. Chem. 1967, 242, 1030–1035.
37. Bucher, T.; Pfleiderer, G. Methods Enzymol. 1955, 1, 435–440.
38. Pedersen, R. L.; Liu, K. K. C.; Rutan, J. F.; Chen, L.; Wong, C. H. J. Am. Chem. Soc.
1990, 55, 4897–4901.
39. Barbas, C. F., III; Wang, Y. F.; Wong, C. H. J. Am. Chem. Soc. 1990, 112, 2013–
2014.
28. Kiessling, W. Berichte 1934, 67, 869–874.
40. Li, H.; Tian, J.; Wang, H.; Yang, S. Q.; Gao, W. Y. Helv. Chim. Acta 2010, 93, 1745–
1750.
41. Phillips, S. A.; Thornalley, P. J. Eur. J. Biochem. 1993, 212, 101–105.
42. Rosen, O. M.; Hoffee, P.; Horecker, B. L. J. Biol. Chem. 1965, 240,
1517–1524.
29. Ballou, C. E.; Fischer, H. O. L. J. Am. Chem. Soc. 1955, 77, 3329–3331.
30. Oesper, P. Arch. Biochem. Biophys. 1961, 95, 237–241.
31. Szewczuk, A.; Wolny, E.; Wolny, M.; Baranowsky, T. Acta Biochim. Pol. 1961, 8,
201–207.
32. Richard, J. P. J. Am. Chem. Soc. 1984, 106, 4926–4936.
43. Serianni, A. S.; Pierce, J.; Barker, R. Biochemistry 1979, 18, 1192–1199.