Communications
Given that DHAP is a key substrate for DHAP-dependent al-
dolases to prepare various 1-phosphorylated sugars,[17,25,26] the
efficiency of the PKGstea–DHAK@Mg2Al biohybrid was evaluated
for four reactions by using commercially available rabbit
muscle aldolase (RAMA, 3S,4R stereoselective) and rhamnulose
aldolase (RhuA, 3R,4S stereoselective) produced and purified as
already described.[27] These aldolases were used with two ac-
ceptors, glycolaldehyde and d-glyceraldehyde (Scheme 2, see
loss in activity in the LDH. Typically, 45 mg of DHAK (182 U),
2.5 mg of FSAA129S (13 U), 1 mg of PKGstea (92 U), and 0.1 mg of
TPI (1638 U) were used, corresponding to an enzyme/LDH
ratios of 1:1, 1:20, 1:50, and 1:500 for DHAK, FSAA129S, PKGstea
,
and TPI, respectively. Note that owing to its instability, a larger
amount of DHAK was involved in the biohybrid. As a result,
100% of the proteins were immobilized, and the retained ac-
tivities were in the same order of magnitude as those obtained
if the individually immobilized enzymes were consid-
ered: 50, 50, 90, and 92% for DHAK, PKGstea, TPI, and
FSAA129S, respectively. As observed for single enzymes,
the XRD pattern of the biohybrid traduced the forma-
tion of an ill-defined LDH structure (Figure 2 f), which
confirmed for the first time the possibility to extend
such a strategy to multi-enzymatic LDH preparation.
The efficiency of this bio-nanoreactor was compared
with that of the free enzymes. The rate of d-F6P for-
mation was the same as that if the immobilized
system or the free enzyme were used (see the Sup-
porting Information). Owing to the high quantity of
DHAK, DHA phosphorylation was faster than the al-
dolization, as indicated by the higher rate of pyruvate
Scheme 2. Enzymatic cascade reactions with the PKGstea–DHAK@Mg2Al biohybrid for the
preparation of phosphorylated sugars.
also Scheme S3). The three matrices give similar results with
the two kinases (encapsulation efficiency of 100% and immo-
bilized activities of 50%; Table 1), and to limit DHAP isomeriza-
tion into d-G3P, the co-immobilization was preferentially con-
ducted in Mg2Al–LDH to minimize TPI activity (36% of retained
activity, see Table 1), often present as a contaminant enzyme.
Indeed, TPI can afford d-G3P, which can react with DHAP in
the presence of RAMA to produce d-fructose-1,6-bisphosphate
as a byproduct. Under these conditions, RAMA provides d-xylu-
lose-1-phosphate[28] and d-fructose-1-phosphate,[29] whereas
RhuA leads to l-xylulose-1-phosphate[28] and d-sorbose-1-phos-
phate.[30] To completely shift the equilibrium to sugar forma-
tion, 2 equivalents of acceptor aldehydes was added and no
more DHAP was detected. All the products were successfully
purified by barium precipitation (see the Supporting Informa-
tion), and yields of 75, 80, 79, and 78% were obtained for l-xy-
lulose-1-phosphate, d-fructose-1-phosphate, d-sorbose-1-phos-
phate, and d-xylulose-1-phosphate, respectively.
formation relative to that of d-F6P. The d-F6P generated by
the biohybrid was purified and isolated in 90% yield, as con-
firmed by NMR spectroscopy (see the Supporting Information).
The biohybrid was reusable over three cycles (Figure 4, see
also Figure S5). After each cycle, the activity of DHAK dimin-
ished as showed in Table 2, whereas the activities of FSAA129S
,
PKGstea, and TPI remained stable at approximately 85% over the
three cycles. As a consequence, this decreased activity of
DHAK led to increased reaction times from 60 min for the
second cycle to 70 min for the third one. In this latter case, the
rates of formation of DHAP and d-F6P were the same, and con-
sequently, the determinant step was probably the formation of
Encouraged by these very good results, we decided to
design, in the second part of the work, a four-enzyme bio-
nanoreactor involving PKGstea, DHAK, TPI, and FSAA129S to syn-
thesize d-F6P as a model reaction (Scheme 1). This sequence
was performed in a one-pot cascade reaction. First, DHAP is
formed by phosphorylation of DHA by using ATP-dependent
DHAK. Then, isomerization of DHAP into d-G3P occurs under
catalysis of TPI.[17] Finally, aldolization between DHA and d-G3P
by using FSAA129S shifts the equilibrium (Scheme 1) to the for-
mation of d-F6P.
According to the results obtained for the individual enzyme
immobilizations, the MgZnAl-LDH matrix was selected to pre-
pare such a biohybrid, as it displayed higher enzyme loading
and retained higher enzyme activity. Direct coprecipitation was
performed in a medium containing a determined quantity of
each enzyme, following the previous results obtained for the
free enzyme system[17] and taking into account the observed
Figure 4. Rate of d-F6P formation by using the PKGstea–DHAK–TPI–
FSAA129S@MgZnAl biohybrid. Conditions: DHA (0.8 mmol, 2 equiv.), PEP
(0.32 mmol), ATP (36 mmol), MgCl2 (10 mm); pH adjusted to 7.5; DHAK
(20 U), PKGstea (11 U), FSAA129S (3 U), and TPI (288 U). The concentrations of
pyruvate, d-F6P, and PEP were measured by spectrophotometric assays.
ChemCatChem 2015, 7, 3110 – 3115
3113
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim