Exploiting Acid Phosphatases in the Synthesis of Phosphorylated Monoalcohols and Diols

Article abstract of DOI:10.1002/ejoc.201501306

A set of phosphatases was evaluated for their potential to catalyze the regio- and stereoselective phosphorylation of alcohols using a high-energy inorganic phosphate donor, such as di-, tri- and polyphosphate. Parameters such as type and amount of phosphate donor and pH of the reaction were investigated in order to minimize the thermodynamically favored hydrolysis of the phosphate donor and the formed phosphate ester. Diols were monophosphorylated with high selectivities. This biocatalytic phosphorylation method provides selectively activated and/or protected synthetic intermediates for further chemical and/or enzymatic transformations and is applicable to a large scale (6.86 g) in a flow setup with immobilized phosphatase.

Full text of DOI:10.1002/ejoc.201501306

DOI: 10.1002/ejoc.201501306
Communication
Phosphorylation
Exploiting Acid Phosphatases in the Synthesis of
Phosphorylated Monoalcohols and Diols
Gábor Tasnádi,[a] Michael Lukesch,[b] Michaela Zechner,[b] Wolfgang Jud,[b] Mélanie Hall,[b]
Klaus Ditrich,[c] Kai Baldenius,[c] Aloysius F. Hartog,[d] Ron Wever,[d] and Kurt Faber*[b]
Abstract: A set of phosphatases was evaluated for their poten- phate ester. Diols were monophosphorylated with high selectiv-
tial to catalyze the regio- and stereoselective phosphorylation
of alcohols using a high-energy inorganic phosphate donor,
such as di-, tri- and polyphosphate. Parameters such as type
and amount of phosphate donor and pH of the reaction were
investigated in order to minimize the thermodynamically fa-
vored hydrolysis of the phosphate donor and the formed phos-
ities. This biocatalytic phosphorylation method provides select-
ively activated and/or protected synthetic intermediates for fur-
ther chemical and/or enzymatic transformations and is applica-
ble to a large scale (6.86 g) in a flow setup with immobilized
phosphatase.
Introduction
perspective. Although enzymatic ATP recycling is feasible with
reasonable efficiency (up to several hundred TTNs),[4a,5] PAPS
cannot be recycled efficiently so far.[4] Most importantly, ATP-
dependent kinases and PAPS-dependent sulfotransferases are
typically involved in biosynthetic pathways and signal messag-
ing and thus are often highly substrate-specific and hence do
not allow broad synthetic applications.
An efficient alternative to kinases is the use of ATP-independ-
ent phosphatases using cheap inorganic polyphosphates (pyro-
phosphate, PPi; triphosphate, PPPi; polyphosphoric acid, polyP)
as activated phosphate donor (P-donor), which circumvents the
need for cofactor recycling.[6] Additionally, due to their occur-
rence in biodegradation pathways, phosphatases usually display
a broad substrate spectrum, and inorganic phosphate formed
as by-product is environmentally benign. A major drawback re-
sides in the reversibility of phosphate transfer, causing the un-
desired hydrolysis of the newly formed phosphate ester, which
requires a kinetically tightly controlled system (Figure 1).
Functional groups often require activation in order to enhance
their reactivity in synthetic protocols. The OH functionality is
commonly activated by attachment of a good leaving group,
e.g., by conversion into an ester derivative of a strong acid
(tosylate, mesylate, triflate), usually employing Mitsunobu or
Appel conditions.[1] This classic methodology suffers from poor
atom economy and creates large amounts of waste, thus ren-
dering these processes inefficient and unattractive, although re-
markable progress towards catalytic processes was made in the
past decade.[2] In nature, activation of hydroxy groups often
proceeds through enzymatic phosphorylation by kinases using
activated phosphate (adenosine triphosphate, ATP) or sulfation
catalyzed by mammalian sulfotransferases in conjunction with
the respective cofactor, i.e., 3′-phosphoadenosine-5′-phospho-
sulfate (PAPS).[3] Nonspecific bacterial arylsulfotransferases are
also able to sulfate hydroxy groups by using organic sulfate
donors.[4] The requirement for molar equivalents of these cofac-
tors in synthetic applications is problematic from a scale-up
[a] Austrian Centre of Industrial Biotechnology,
c/o Department of Chemistry, Organic & Bioorganic Chemistry, University
of Graz,
Heinrichstrasse 28, 8010 Graz, Austria
[b] Department of Chemistry, Organic & Bioorganic Chemistry, University of
Graz,
Heinrichstrasse 28, 8010 Graz, Austria
E-mail: Kurt.Faber@Uni-Graz.at
http://biocatalysis.uni-graz.at
[c] White Biotechnology Research Biocatalysis, BASF SE,
Carl-Bosch-Strasse 38, 67056 Ludwigshafen, Germany
[d] Van't Hoff Institute for Molecular Sciences, University of Amsterdam,
Science Park 904, 1098 XH Amsterdam, The Netherlands
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/ejoc.201501306.
Figure 1. Mechanism of (a) competing hydrolysis and (b) transphosphoryla-
tion involving a covalent phospho-enzyme intermediate catalyzed by PhoN-
Se.[26]
© 2015 The Autors. Published by Wiley-VCH Verlag GmbH & Co. KGaA.
This is an open access article under the terms of the Creative Commons
Attribution License, which permits use, distribution and reproduction in
any medium, provided the original work is properly cited.
The use of phosphatases for synthetic applications dates
back to the late 1940s, when Axelrod and Appleyard achieved
Eur. J. Org. Chem. 2016, 45–50
45
© 2016 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
transphosphorylation using phosphatase-enriched extracts and niques were investigated to enhance product titers of phos-
p-nitrophenyl phosphate (p-NPP) or phenolphthalein phos- phorylated alcohols and to minimize the thermodynamically fa-
phate as donor.[7] Later, Nordlie identified glucose-pyrophos- vored hydrolytic reaction.
photransferase activity of E. coli alkaline phosphatase;[8] how-
ever, the method was applied on a preparative-scale only dec-
ades later.[9] Asano et al. applied acid phosphatases from enteric
Results and Discussion
bacteria to the phosphorylation of nucleosides and designed
enzyme mutants with reduced hydrolytic activity.[10] Acid phos-
Enzyme candidates were selected from the family of acid phos-
phatases from Shigella flexneri (PhoN-Sf) and Salmonella enter-
ica ser. typhimurium LT2 (PhoN-Se) were found to have a broad
substrate spectrum[11] and were applied in combination with
aldolases to synthesize natural and unnatural sugar molecules
on a gram-scale.[12]
phatases, which have been shown to transfer efficiently a phos-
phate moiety from a high-energy inorganic di/tri/polyphos-
phate onto various hydroxylated compounds. Nonspecific acid
phosphatase PhoN-Se from Salmonella typhimurium and acid
phosphatase PhoN-Sf from Shigella flexneri have been reported
to convert various primary and secondary alcohols and sug-
ars[11] and were implemented in cascade reactions involving
glucose-6-phosphate[19] and dihydroxyacetone phosphate as
intermediates.[12] Acid phosphatase phytase catalyzes the hy-
drolysis of phytate, a phosphorylated natural cyclitol by liberat-
ing inorganic phosphate and is therefore commonly used as
animal feedstock additive.[20] Acid phosphatase AphA-St from
Salmonella typhimurium LT2[21] was selected due to its phos-
photransferase activity on alcohols using p-NPP as donor.[22]
PiACP from Prevotella intermedia[23] was selected based on high
sequence similarity with robust PhoC-Mm developed for nucle-
oside phosphorylation.[24] All enzymes except phytase were
overexpressed in E. coli and subsequently purified to homoge-
neity by using His-tag or Strep-tag techniques (Figure S1).
A common mechanistic feature of phosphate-transferring
phosphatases is the formation of a covalent phospho-enzyme
In drug discovery, phosphorylation is a common prodrug
strategy to enhance the accessibility of drug candidates within
the body by enhancing their polarity and solubility.[13] In some
cases, the phosphorylated form may be the active species.[14]
In synthetic chemistry, phosphate di- and triesters are widely
applied as Brønsted acid catalysts for various asymmetric trans-
formations[15] and as leaving groups in transition-metal-cata-
lyzed coupling reactions[16] as well as in organolithium[17] and
alkene synthesis.[18] In contrast, phosphate monoesters have at-
tracted so far little attention as tools for functional-group modi-
fication in classic chemical processes, most likely as a conse-
quence of their insolubility in organic solvents resulting from
their negative charge. Current chemical routes to phosphate
monoesters often proceed via the corresponding phosphate di-
or triesters and require harsh reagents and conditions leading
to hardly separable product mixtures.
In this study, we aimed to evaluate a set of phosphatases as intermediate, which undergoes either nucleophilic attack by
potential candidates for synthetic applications. Various tech- water (hydrolysis) or accepts a suitable nucleophile (transphos-
Table 1. Substrate scope of PhoN-Sf, PhoN-Se and PiACP in transphosphorylation.
[a] Reaction conditions: pH = 4.2, 50 μg mL–1 enzyme, 1 mL of volume, 1 % DMSO as internal standard, 30 °C, 600 rpm shaking. PPi: pyrophosphate; PPPi:
triphosphate; polyP: polyphosphoric acid. n.d. = not determined, n.a. = not applicable. Product levels were determined by consumption of substrate using
HPLC with RI detector (control experiments used as reference). [b] Time [min] needed to reach maximal product concentration in brackets. [c] In g L–1. [d]
Limit of detectability of bisphosphate is ca. 5–10 %. [e] 300 m
observed.
M
substrate used. [f] Monitored by 31P NMR spectroscopy. [g] Only hydrolysis of P-donor
Eur. J. Org. Chem. 2016, 45–50
www.eurjoc.org
46
© 2016 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
phorylation) (Figure 1). To this end, most enzymes use histidine only P-donor hydrolysis was observed (data not shown), prov-
as catalytic residue (PhoN-Sf, PhoN-Se, PiACP and phytase) ex- ing that these enzymes do not accept amines in place of alco-
cept AphA-St which uses aspartic acid.[25] Due to the competi- hols. Sterically demanding secondary alcohol 7 was trans-
tion between hydrolysis and transphosphorylation in aqueous formed only with PhoN-Se, which is in line with the previously
medium, high substrate loads are required to achieve high observed preference of this enzyme for sec-alcohols.[11] Similar
product titers, while kinetic control is crucial to keep the phos- observations were made with 1,2-propanediol (8), which was
phorylation/hydrolysis ratio as high as possible.
exclusively phosphorylated at the prim-alcohol by PhoN-Sf and
PiACP, while PhoN-Se showed reduced selectivity (ca. 90:10
prim/sec-phosphate, determined by 31P NMR spectroscopy; Fig-
ure S6).
In order to verify the activity of all recombinant proteins, a
spectrophotometric assay based on the hydrolysis of p-NPP was
used. A range of alcohols was investigated, including primary
and secondary mono- and dihydroxy compounds, as well as
cyclic substrates (Table 1). Products were identified by 31P NMR
spectroscopy and HPLC-MS (see Supporting Information). Al-
though phytase has been successfully employed in a cascade
reaction to convert glycerol into glycerol-1-phosphate in a non-
stereoselective fashion,[27] under our assay conditions, the en-
zyme showed no transphosphorylation activity at pH = 2.5 and
4.2 on any substrate. Strong hydrolytic activity on inorganic
phosphate donors was still observed, confirming active protein
(data not shown). AphA-St also contains Mg2+ (in addition to
the catalytically active nucleophile aspartate[21]), whose activity
is affected by inorganic polyphosphates by metal chelation. Al-
In order to elucidate the enantioselectivity of phosphatases,
the two enantiomers of 8 were tested separately. In case of
PhoN-Sf and PiACP, similar activities on both enantiomers were
observed indicating non-enantioselective phosphorylation.
With PhoN-Se, a marginal enantiopreference for (R)-8 was de-
tected (Figure S7). To further investigate the enantioselectivity
of PhoN-Se, both enantiomers of 2-butanol (9) were subjected
to screening conditions; however, no significant difference in
the reaction course was observed. Furthermore, racemic cyclic
trans-diol 10 was tested, and beside a moderate amount of
monophosphorylated product, no stereoselectivity could be
observed (as determined by measurement of the ee value of
remaining 10 by GC after derivatization of both OH functionali-
ties; see Supporting Information).
though the addition of 10 m
hydrolytic activity on p-NPP in the presence of 250 m
M
MgCl2 could partly restore the
PPi (Fig-
M
Substrates containing N- or S-nucleophiles or charged
groups were not converted (Figure 2), while PPi was still hydro-
lyzed indicating that the enzymes were still active (data not
shown). The absolute chemoselectivity of O-phosphorylation vs.
N- and S-nucleophiles was corroborated by the non-acceptance
of substrates 11 and 12. Mevalonolactone (rac-13) and charged
substrates, like lactic (rac-14) and tartaric acid (15) (both
enantiomers tested separately) were not converted. Although
ure S2), only traces of product and marginal hydrolytic activity
were obtained in the transphosphorylation of 1,4-butanediol
(500 m
M
) by using 100 m
M
PPi as donor at pH = 4.2 and 5.5,
) (data
independent of the MgCl2 concentration (10, 50, 100 m
M
not shown). This suggests that inorganic oligophosphates (as
opposed to p-NPP) are not ideal phosphate donors for metal-
containing phosphatases, such as AphA-St. On the contrary,
PhoN-Sf, PhoN-Se and PiACP all proved to be efficient catalysts
in the phosphorylation of various alcohols (Table 1).
O-phospho-D,L-serine could be stereoselectively hydrolyzed by
a mutant of PhoN-Se,[29] the wild-type enzyme was not active
in the phosphorylation of rac-serine (16). Finally, oximes 17–19
did not undergo phosphorylation.
Phosphorylation of prim-diols proceeded to various degrees
depending on the chain length: While short-chain ethylene
glycol (1) was selectively monophosphorylated, 1,4-butanediol
(2) and the unsaturated (Z) analog 4 yielded the corresponding
mono/bisphosphorylated products in a ratio of 90:10. Extend-
ing the chain length even further (1,6-hexanediol, 3) gave 85:15
of mono/diphosphate as determined by LC-RI chromatography
(see Supporting Information). It is obvious, that the presence of
a (negatively charged) phosphate group in the proximity of the
reacting hydroxy moiety determines whether an α,ω-diol un-
dergoes bisphosphorylation: While short-chain ethylene glycol
is exclusively monophosphorylated, long-chain 1,6-hexanediol
yields the bisphosphate in up to 15 %.
Figure 2. Substrates not converted by PhoN-Sf, PhoN-Se, PiACP and phytase
[300 mM substrate concentration except for rac-14 (200 mM) and 17 (15 mM)].
Haloalcohol 5 was also converted: PhoN-Se tolerated higher
substrate concentrations than PhoN-Sf and PiACP (500 m
M vs.
300 m ) resulting in the formation of 113 m product in
M
M
Due to the reversibility of the reaction (Figure 1), kinetic con-
trol of the reaction is necessary to ensure optimal product titers
by avoiding product depletion by (reverse) hydrolysis and to
ensure optimal product titers. This may be achieved by adjust-
ment of reaction parameters such as pH or selection of oligo-
phosphate donor.[30] Therefore, 1,4-butanediol (2) was selected
60 min. To probe the chemoselectivity of the enzymes for O-
vs. N-nucleophiles, 6-amino-1-hexanol (6) was tested. The three
enzymes exclusively furnished the O-phosphate monoester at
very similar product levels (ca. 71–85 m
M), while the amino moi-
ety remained untouched. This was proven by 31P NMR spectro-
scopy resulting in a chemical shift typical for signals of phos- as a model substrate, and enzymes were tested with tri- and
phorylated primary hydroxy groups (Figure S5).[28] In line with polyphosphate as alternative to PPi (Table 1 and Figures S8–
this observation, 1,6-diaminohexane (12) was not converted, 10).
Eur. J. Org. Chem. 2016, 45–50
www.eurjoc.org
47
© 2016 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
The amount of 4-hydroxybutyl phosphate formed could be ity of the enzymes and the pH. High product levels were found
significantly increased by replacing PPi by high amounts of PPPi at pH = 3.5–4.2. Above this pH range, hydrolysis usually domi-
and polyP in the reaction catalyzed by PhoN-Sf and PiACP, al- nated (Table S2). Although identical amounts of transferable
lowing up to 61 % conversion (305 m
the theoretical maximum. The difference in the slope in product PPPi, significant differences in product formation were obtained
disappearance clearly shows that product hydrolysis is de- after 16.5 h reaction time. Using 500 m PPi as a donor, product
creased in the presence of tri- and polyphosphate compared to yields of 2 for PhoN-Sf and PiACP were similar at the three pH
PPi. In contrast, these donors had a negative influence on the values. Using PPi, up to 292 m product (up to 58 % conver-
M product) with respect to phosphate moieties are available from 500 mM PPi and 250 mM
M
M
performance of PhoN-Se, where a product maximum was ob- sion) could be obtained accompanied by nearly total consump-
tained with PPi. Moreover, HPLC monitoring of the reaction tion of PPi (> 90 %), and the product/Pi ratio was ca. 1:2 (the
indicated incomplete consumption of polyphosphoric acid. ideal ratio is 1:1 in the absence of product hydrolysis and com-
This suggests that this enzyme accepts only short chain oligo- plete phosphate transfer). However, under these conditions
phosphates as reported for calf intestine alkaline phosphat- PhoN-Se furnished only inorganic monophosphate, indicating
ase.[9]
Since PhoN-Sf generally delivered the highest product titers
total hydrolysis of PPi and product. In the presence of 250 mM
PPPi the picture changed: The transphosporylation activity at
pH = 3.5 of PhoN-Sf dropped considerably, that of PiACP also
dropped, but surprisingly considerable phosphorylating activity
with primary alcohols as acceptors, the enzyme was selected
for further investigations with 1,4-butanediol concerning the
effect of pH (range 3.8–4.8) by using different P-donors (Fig-
ure 3). The pH had no significant impact on the reaction using
PPi. On the other hand, PPPi and polyP usually yielded higher
product concentrations at lower pH, due to slower product hy-
drolysis. However, the poorly defined oligomeric composition
of the polyphosphate donor renders precise analysis as well as
product isolation challenging, and polyP was not investigated
further.
of PhoN-Se was maintained (164 mM, 33 % conversion and 54 %
P-donor consumption). At higher pH values this enzyme was
still active, but only hydrolysis was observed, while PhoN-Sf and
PiACP delivered moderate to good amounts of product at full
consumption of the P-donor. At 400 m
both PhoN-Sf and PiACP showed the highest phosphorylating
activity resulting in up to 393 m product (79 % conversion,
M PPPi and pH = 4.2
M
product/Pi ratio ca. 4:5) at 75 % P-donor consumption.
The results depicted in Figure 3 using PhoN-Sf suggest that
In general, higher PPPi concentrations translate into higher
a high number of transferable high-energy phosphate moieties product concentration and lower Pi concentration, indicating
as well as low pH values are crucial to achieve and maintain more efficient phosphate transfer due to reduced product hy-
high product titers. Inspired by this observation, reactions were drolysis. From the data it is clear that using these phosphorylat-
conducted at variuos pH values using high amounts of PPi and ing enzymes in synthesis effectively requires a careful study of
PPPi overnight (Table 2). It should be noted that within the the donor concentrations, the type of donor, and the pH; in
incubation time of 16.5 h both transphosphorylation and hy- batch processes also the time of incubation should be carefully
drolysis of product and phosphate donor will occur, and the chosen. A long incubation time will result in low yields or near
balance between the two processes is determined by the activ- complete hydrolysis of the phosphorylated product.
Figure 3. Effect of P-donor type and pH on the phosphorylation of 2 (product formation over time). Conditions: 500 mM 2, 250 mM
PPi or PPPi or 100 g L–1
polyP, 50 μg mL–1 PhoN-Sf, 1 mL volume, 1 % DMSO as internal standard, 30 °C, 600 rpm shaking.
Eur. J. Org. Chem. 2016, 45–50
www.eurjoc.org
48
© 2016 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Table 2. Amount of product and Pi formed in acid phosphatase catalyzed phosphorylation of 2 with PPi or PPPi as donor, at three pH values with PhoN-Sf,
PiACP and PhoN-S.
PPi (500 m
PhoN-Sf
M)
PPPi (250 m
PhoN-Sf
M)
PPPi (400 m
PiACP
M)
PhoN-Se
PiACP
PhoN-Se
PiACP
PhoN-Se
PhoN-Sf
pH = 3.5
pH = 3.8
pH = 4.2
product [m
Pi [m
M
]
236
743
0.32
98
253
655
0.39
91
0
15
22
0.68
5
3
190
540
0.35
97
131
106
1.24
32
164
244
0.67
54
33
4
747
0.01
100
1
8
13
0.62
2
2
276
238
1.16
43
24
13
1.85
3
5
393
507
0.78
75
48
46
1.04
8
10
M]
940
0.00
94
0
product/Pi[a]
Cons.donor [%][b]
Conv. [%][c]
47
51
26
product [m
Pi [m
M]
226
760
0.30
99
292
611
0.48
90
0
122
613
0.20
98
24
31
669
0.05
93
83
M]
934
0.00
93
0
1099
0.08
99
17
21
1154
0.02
98
product/Pi[a]
Cons.donor [%][b]
Conv. [%][c]
45
58
38
55
79
product [m
Pi [m
M]
110
893
0.12
100
22
105
818
0.13
92
0
101
603
0.17
94
3
315
805
0.39
93
265
768
0.35
86
M]
924
0.00
92
0
703
0.00
94
product/Pi[a]
Cons.donor [%][b]
Conv. [%][c]
21
20
6
1
63
53
4
[a] Reaction conditions: 500 mM
2, phosphate donor as indicated, 50 μg mL–1 PhoN-Sf, pH as indicated, 1 mL volume, 1 % DMSO as internal standard, 30 °C,
600 rpm shaking, 16.5 h. Ideal (maximum obtainable) product/Pi ratio is 1:1 with PPi and 2:1 with PPPi. [b] Cons.donor: consumption of P-donor calculated as
follows: (cprod + cPi)/ctheoretical max. Pi × 100. [c] Conv.: conversion of substrate with respect to theoretical maximum.
Consequently, kinetic control of the phosphorylation is very action parameters, thus allowing the use of phosphatases in
important, and physical separation of the phosphatase by em- the synthetic direction. Scale-up was demonstrated by the syn-
ploying a continuous flow-reactor has proven to be an efficient thesis of 4-hydroxybutyl phosphate (6.86 g, 41 % yield) using
tool to maintain high product levels (by minimizing hydrolysis) an immobilized enzyme in flow mode.
and scale-up the reaction.[12c,31] Under optimized conditions,
the contact time between enzyme and substrate(s) is controlled
by the flow rate to ensure that (rapid) phosphorylation is com-
plete, but (slow) product hydrolysis cannot take place after the
product has left the enzyme-containing column reactor. This
Experimental Section
For experimental details, see the Supporting Information.
requires the enzyme to be immobilized on a suitable support.
We evaluated several epoxy- and amino-functionalized poly-
meric beads for the covalent immobilization of PhoN-Sf and
Acknowledgments
Funding by the Austrian Science Fund (FWF) within the DK Mo-
PiACP (Table S3). The activity of the beads was tested in the p-
lecular Enzymology (project W9). This work has been supported
NPP assay. Relizyme HA403/M, an amino-functionalized resin,
by the Federal Ministry of Science, Research and Economy
resulted in the highest specific activity for both enzymes (159
(BMWFW), the Federal Ministry of Traffic, Innovation and Tech-
and 215 U g–1 dry beads for PhoN-Sf and PiACP, respectively).
nology (BMVIT), the Styrian Business Promotion Agency SFG,
the Standortagentur Tirol, the Government of Lower Austria
PhoN-Sf (300 U) immobilized on Relizyme HA (300 mg) was
applied to the preparative-scale phosphorylation of 2, yielding
and Business Agency Vienna through the COMET-Funding Pro-
6.86 g (22.5 mmol) of the corresponding barium salt of 4-
gram managed by the Austrian Research Promotion Agency
hydroxybutyl phosphate (41 % isolated yield, for details see
FFG. Dr. Michael Breuer (BASF SE) is kindly acknowledged for
Supporting Information).
providing plasmids of AphA-St and PiACP.
Keywords: Phosphorylation · Phosphatase · Biocatalysis ·
Regioselectivity · Inorganic phosphate donors
Conclusions
The enzymatic phosphorylation of primary and secondary alco-
hols in aqueous medium using cheap inorganic di-, tri- and
polyphosphate as P-donors catalyzed by three acid phosphat-
ases proceeded with absolute chemoselectivity for O- vs. N- and
S-nucleophiles and with high regioselectivity for prim- over sec-
hydroxy groups. The use of mild reaction conditions allowed
the formation of phosphate monoesters as sole products, yield-
ing environmentally innocuous inorganic monophosphate and
unreacted phosphate donor as by-products. PhoN-Se could
phosphorylate secondary alcohols, although no chiral induction
was observed. The thermodynamically favored hydrolysis of
phosphate esters could be minimized by controlling various re-
[1] R. Appel, Angew. Chem. Int. Ed. Engl. 1975, 14, 801–811; Angew. Chem.
1975, 87, 863–874.
[2] a) S. Fletcher, Org. Chem. Front. 2015, 2, 739–752; b) H. A. van Kalkeren,
S. H. A. M. Leenders, C. R. A. Hommersom, F. P. J. T. Rutjes, F. L. van Delft,
Chem. Eur. J. 2011, 17, 11290–11295; c) R. M. Denton, J. An, B. Adeniran,
A. J. Blake, W. Lewis, A. M. Poulton, J. Org. Chem. 2011, 76, 6749–6767.
[3] a) K. Faber, Biotransformations in Organic Chemistry, 6th ed., Springer
Verlag, Berlin, Heidelberg, 2011, pp. 26–27; b) F. Westheimer, Science
1987, 235, 1173–1178.
[4] a) I. Ayuso-Fernández, M. A. Galmés, A. Bastida, E. García-Junceda, Chem-
CatChem 2014, 6, 1059–1065; b) M. A. van der Horst, J. F. T. van Lieshout,
A. Bury, A. F. Hartog, R. Wever, Adv. Synth. Catal. 2012, 354, 3501–3508;
Eur. J. Org. Chem. 2016, 45–50
www.eurjoc.org
49
© 2016 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
c) M. A. van der Horst, A. F. Hartog, R. El Morabet, A. Marais, M. Kircz, R.
Wever, Eur. J. Org. Chem. 2015, 534–541.
[18] H. Quast, T. Dietz, Synthesis 1995, 1300–1304.
[19] A. F. Hartog, T. van Herk, R. Wever, Adv. Synth. Catal. 2011, 353, 2339–
2344.
[20] D. E. C. S. Rao, K. V. Rao, T. P. Reddy, V. D. Reddy, Crit. Rev. Biotechnol.
2009, 29, 182–198.
[5]
[6]
J. N. Andexer, M. Richter, ChemBioChem 2015, 16, 380–386.
R. Wever, T. van Herk, in Enzyme Catalysis in Organic Synthesis (Eds.: K.
Drauz, H. Gröger, O. May), Wiley-VCH, Weinheim, 2012, pp. 1001–1033.
a) B. Axelrod, J. Biol. Chem. 1948, 172, 1–13; b) J. Appleyard, Biochem. J.
1948, 42, 596–590.
R. C. Nordlie, W. J. Arion, J. Biol. Chem. 1964, 239, 1680–1685.
A. Pradines, A. Klaebe, J. Perie, F. Paul, P. Monsan, Tetrahedron 1988, 44,
[21] R. D. Makde, V. Kumar, G. D. Gupta, J. Jasti, T. P. Singh, S. K. Mahajan,
Acta Crystallogr., Sect. D Biol. Crystallogr. 2003, 59, 1849–1852.
[22] K. Dissing, W. Uerkvitz, Enzyme Microb. Technol. 2006, 38, 683–688.
[23] X. Chen, T. Ansai, S. Awano, T. Iida, S. Barik, T. Takehara, J. Bacteriol. 1999,
181, 7107–7114.
[24] Y. Asano, Y. Mihara, H. Yamada, J. Mol. Catal. B 1999, 6, 271–277.
[25] Other phosphatases use a hydroxy (Ser or Thr) or thiol moiety (Cys) as
chemical operator, which forms a covalent enzyme–phosphate interme-
diate during catalysis, see: S. C. Bihani, A. Das, K. S. Nilgiriwala, V. Prashar,
M. Pirocchi, S. K. Apte, J.-L. Ferrer, M. V. Hosur, PLoS ONE 2011, 6, e22767;
P. J. O'Brien, D. Herschlag, Biochemistry 2002, 41, 3207–3225; K. Kolmo-
din, J. Åqvist, FEBS Lett. 2001, 498, 208–213.
[26] R. D. Makde, S. K. Mahajan, V. Kumar, Biochemistry 2007, 46, 2079–2090.
[27] R. Schoevaart, F. van Rantwijk, R. A. Sheldon, J. Org. Chem. 2000, 65,
6940–6943.
[28] L. D. Quin, A. J. Williams, Practical interpretation of P-31 NMR spectra and
computer assisted structure verification, Advanced Chemistry Develop-
ment, Inc., Toronto, Canada, 2004.
[7]
[8]
[9]
6373–6386.
[10]
E. Suzuki, K. Ishikawa, Y. Mihara, N. Shimba, Y. Asano, Bull. Chem. Soc.
Jpn. 2007, 80, 276–286.
[11] a) N. Tanaka, Z. Hasan, A. F. Hartog, T. van Herk, R. Wever, Org. Biomol.
Chem. 2003, 1, 2833–2839; b) T. van Herk, A. F. Hartog, A. M.
van der Burg, R. Wever, Adv. Synth. Catal. 2005, 347, 1155–1162.
[12] a) T. van Herk, A. F. Hartog, H. E. Schoemaker, R. Wever, J. Org. Chem.
2006, 71, 6244–6247; b) L. Babich, L. J. C. van Hemert, A. Bury, A. F.
Hartog, P. Falcicchio, J. van der Oost, T. van Herk, R. Wever, F. P. J. T.
Rutjes, Green Chem. 2011, 13, 2895–2900; c) L. Babich, A. F. Hartog, L. J. C.
van Hemert, F. P. J. T. Rutjes, R. Wever, ChemSusChem 2012, 5, 2348–
2353.
[13] a) J. Rautio, H. Kumpulainen, T. Heimbach, R. Oliyai, D. Oh, T. Jarvinen, J.
Savolainen, Nat. Rev. Drug Discovery 2008, 7, 255–270; b) V. J. Stella, K. W.
Nti-Addae, Adv. Drug Delivery Rev. 2007, 59, 677–694.
[29] T. van Herk, A. F. Hartog, H. J. Ruijssenaars, R. Kerkman, H. E. Schoemaker,
R. Wever, Adv. Synth. Catal. 2007, 349, 1349–1352.
[30] L. Babich, PhD thesis (“Enzymatic cascade reactions involving phosphor-
ylated intermediates: immobilization and process optimization”), Univer-
sity of Amsterdam, 2013.
[14] T. Sanchez, T. Estrada-Hernandez, J.-H. Paik, M.-T. Wu, K. Venkataraman,
V. Brinkmann, K. Claffey, T. Hla, J. Biol. Chem. 2003, 278, 47281–47290.
[15] M. Terada, Synthesis 2010, 1929–1982.
[31] L. Babich, A. F. Hartog, M. A. van der Horst, R. Wever, Chem. Eur. J. 2012,
18, 6604–6609.
[16] a) B. M. Trost, L. C. Czabaniuk, J. Am. Chem. Soc. 2012, 134, 5778–5781;
b) A. Yanagisawa, N. Nomura, H. Yamamoto, Tetrahedron 1994, 50, 6017–
6028; c) S.-I. Murahashi, Y. Tanigawa, Y. Imada, Y. Taniguchi, Tetrahedron
Lett. 1986, 27, 227–230.
Received: October 12, 2015
[17] D. Guijarro, B. Mancheño, M. Yus, Tetrahedron 1994, 50, 8551–8558.
Published Online: December 15, 2015
Eur. J. Org. Chem. 2016, 45–50
www.eurjoc.org
50
© 2016 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim