Regiospecific phosphohydrolases from Dictyostelium as tools for the chemoenzymatic synthesis of the enantiomers D-myo-inositol 1,2,4-trisphosphate and D-myo-inositol 2,3,6-trisphosphate: Non-physiological, potential analogues of biologically active D-myo-

Article abstract of DOI:10.1016/S0960-894X(01)00536-4

A new de novo synthesis of the enantiomeric pair D-myo-inositol 1,2,4-trisphosphate and D-myo-inositol 2,3,6-trisphosphate is described. Starting from enantiopure dibromocyclohexenediol, several C₂ symmetrical building blocks were synthesized w

Full text of DOI:10.1016/S0960-894X(01)00536-4

Bioorganic & Medicinal Chemistry Letters 11 (2001) 2705–2708
Regiospecific Phosphohydrolases from Dictyostelium as Tools for
the Chemoenzymatic Synthesis of the Enantiomers D-myo-Inositol
1,2,4-Trisphosphate and D-myo-Inositol 2,3,6-Trisphosphate:
Non-physiological, Potential Analogues of Biologically
Active D-myo-Inositol 1,3,4-Trisphosphate
Stephan Adelt,a Oliver Plettenburg,a,y Guido Dallmann,a Frank P. Ritter,a
Stephen B. Shears,b Hans-Josef Altenbacha and Gunter Vogela,*
aInstitute fur Biochemie und Organische Chemie, Bergische Universitat Wuppertal, Gaussstrasse 20, 42097Wuppertal, FRG
¨
¨
bLaboratory of Signal Transduction, NIEHS, National Institutes of Health, Research Triangle Park, NC 27709, USA
Received 26 April 2001; revised 9 July 2001; accepted 2 August 2001
Abstract—A new de novo synthesis of the enantiomeric pair d-myo-inositol 1,2,4-trisphosphate and d-myo-inositol 2,3,6-trisphos-
phate is described. Starting from enantiopure dibromocyclohexenediol, several C2 symmetrical building blocks were synthesized
which gave access to d-myo-inositol 1,2,4,5-tetrakisphosphate and d-myo-inositol 1,2,3,6-tetrakisphosphate. Exploiting the high
regiospecificity of two partially purified phosphohydrolases from Dictyostelium, a 5-phosphatase and a phytase, the inositol tetra-
kisphosphates were converted enzymatically to the target compounds. Their potential to modulate the activity of Ins(3,4,5,6)P4 1-
kinase was investigated and compared with the effects of d-myo-inositol 1,3,4-trisphosphate. # 2001 Elsevier Science Ltd. All rights
reserved.
Introduction
Ins(1,3,4)P3, a downstream-product of the phosphoino-
sitide signaling cascade, is involved in the regulation of
Ins(3,4,5,6)P4 levels.4 Ins(1,3,4)P3 is a competing sub-
strate for the Ins(3,4,5,6)P4 1-kinase and is phosphory-
lated by the enzyme at 5- and 6-positions.5 Certain
diseases, such as cystic fibrosis, may be treated by either
up- or down-regulation of calcium activated chloride
secretion,6 and for that reason the design of effective
Ins(3,4,5,6)P4 agonists and antagonists for pharmaco-
logical intervention in the signaling actions of
Ins(3,4,5,6)P4 is of great interest. In this context it was
myo-Inositol phosphates are involved in essential pro-
cesses in eucaryotic cells.1 Interdependent kinases and
phosphohydrolases construct a complex metabolic net-
work that allows a fine-tuning of cellular inositol phos-
phate levels. This is especially important for isomers
participating in cellular signaling.
Ins(3,4,5,6)P4 is an inhibitor of calcium activated chlo-
ride channels in the plasma membrane.2 These ion
channels contribute to the homoeostatic control of salt
and fluid secretion with impact on osmoregulation and
pH balance. The cellular accumulation of Ins(3,4,5,6)P4
is known to correlate well with receptor-dependent
changes in phospholipase C activity, but the molecular
mechanisms that link these two events have only
recently been elucidated.3 It was shown that
found that
a
racemic mixture of Ins(1,2,4)P3/
Ins(2,3,6)P3 (both non-physiological isomers) was only
4-fold less potent than Ins(1,3,4)P3 as an inhibitor of
Ins(3,4,5,6)P4 phosphorylation by the kinase. We there-
fore set out to clarify the relative potencies of
Ins(1,2,4)P3 (8a) and Ins(2,3,6)P3 (8b) as 1-kinase inhi-
bitors.
Their first total synthesis described here combines a
straightforward chemical synthesis of Ins(1,2,4,5)P4 (5)
and Ins(1,2,3,6)P4 (7) with subsequent regiospecific
enzymatic dephosphorylations to obtain Ins(1,2,4)P3
*Corresponding author. Tel.:+49-202-439-2791; fax:+49-202-439-
3399; e-mail: vogel@uni-wuppertal.de
yPresent address: Department of Chemistry, The Scripps Research
Institute, 10550 N. Torrey Pines Rd., La Jolla, CA 92037, USA.
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00536-4

2706
S. Adelt et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2705–2708
(8a) and Ins(2,3,6)P3 (8b). The target compounds were
tested for their ability to influence the activity of
recombinant human Ins(3,4,5,6)P4 1-kinase and the
reason for this effect was investigated.
Synthetic Application of Phosphohydrolases
One focus of our research is the establishment of new,
preparatively applicable enzymatic reactions in inositol
phosphate synthesis.7,9 We direct our closest attention
to conversions producing enantiomers of current biolo-
gical interest which are only difficult to obtain by con-
ventional organic synthetic methods.
Chemistry
Enantiopure dibromocyclohexenediol (+)-1 serves as
starting material. It can be prepared on a large scale in
only four steps from p-benzoquinone.7 Treatment of
dibromodiol 1 with sodium benzylate in benzyl alcohol
leads in a stereocontrolled transformation to con-
duritol-B derivative 2 (Scheme 1). A fast cis-dihydroxy-
lation step with RuCl3/NaIO4 gives compound 3 with
myo-inositol stereochemistry, which can be phosphory-
lated by reaction with 3-diethylamino-2,3,4-benzodioxa-
phosphepane in the presence of 1H-tetrazole and
subsequent oxidation with mCPBA. Complete depro-
tection of the Ins(1,2,4,5)P4 derivative 4 can be achieved
in one step by a Pd-catalyzed hydrogenation. The pro-
duct Ins(1,2,4,5)P4 (5) was subjected to enzymatic con-
version as described below.8
Previous work on the substrate specificity of two par-
tially purified enzymes from the cellular slime mold
Dictyostelium discoideum encouraged us to utilize them
for preparation of the target compounds. An easily to
cultivate axenic mutant of this microorganism (strain AX-
2) was used as enzyme source.10 Cells were lysed by freez-
ing/thawing and the homogenate was separated into a
cytosolic fraction and a membrane-enriched fraction by
ultracentrifugation. The supernatant contains an enzyme
of the Ins(1,4,5)P3 5-phosphatase family and the sediment
an enzyme with characteristics of a phytase (InsP6-phos-
phohydrolase) that could be solubilized with nonionic
detergent. Both protein solutions were subjected to column
chromatography to gain enriched enzyme preparations,
which are devoid of interfering phosphate, InsPx meta-
bolites and unspecific phosphohydrolase activities.11,12
As published previously starting from the same pre-
cursor 1 it is possible to synthesize the isomer
Ins(1,2,3,6)P4 (7) via a different conduritol-B derivative
6 (Scheme 1).9 It has to be mentioned that the
enantiomers of these inositol tetrakisphosphates are
also accessible, illustrating the versatility of this syn-
thetic strategy.
The Mg2+-dependent Ins(1,4,5)P3 5-phosphatase pos-
sesses a pH optimum at about 7. We tested the substrate
specificity of the 5-phosphatase with an extensive num-
ber of inositol phosphates and phosphoinositides. A
detailed list of the compounds and the results of this
study will be published elsewhere. Relevant for the
intended application are examined isomers modified at
their functional groups in position C-1 and C-2.
The tetrakisphosphates 5 and 7 were already pure by
NMR spectroscopic standards. Purification of the pro-
ducts by HPLC ensures purities suitable for biological
experiments. Total phosphate determination after sul-
furic acid hydrolysis allows a reliable quantification of
the hygroscopic substances.7
Dephosphorylation
of
Ins(1,2,4,5)P4
(5)
and
dideoxy(1,2)Ins(4,5)P2 (the double bond of compound 2
was hydrogenated for its synthesis, see Scheme 1) takes
place with approximately 20 and 10% of the rate the
Scheme 1.

S. Adelt et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2705–2708
2707
physiological substrate Ins(1,4,5)P3 is converted. Enzy-
matic activity was measured in each case under sub-
strate saturation of the enzyme. Considering that the
reaction occurs highly specifically at the 5-position and
that it stops after removal of one phosphate group this
is still fast enough to be of practical value in enzyme-
assisted synthesis. Working with the total enzyme
activity obtained from 1010 cells (10 g wet weight) it is
possible to dephosphorylate 25 mmol Ins(1,2,4,5)P4 (5)
in about 4 h (Scheme 2).13 After HPLC purification and
quantification (see above) we calculated an excellent
overall yield of 95% for the enantiomer Ins(1,2,4)P3
(8a).14
Figure 1. Effects of InsP3 isomers upon Ins(3,4,5,6)P4 1-kinase activ-
ity. Enzyme activity was assayed as previously described,4 except that
recombinant human kinase5 and a higher substrate concentration of
5 mM [3H]Ins(3,4,5,6)P4 (approx. 2000 dpm/assay) were used. Activity
remaining in the presence of non-radiolabeled Ins(1,3,4)P3 (squares,
solid line), Ins(1,2,4)P3 (8a; diamonds, broken line) and Ins(2,3,6)P3
(8b; circles, dotted line) is expressed as a percentage in relation to
control.
It only recently emerged that the actions of Ins(1,3,4)P3
upon Ins(3,4,5,6)P4 1-kinase activity result from a com-
petition for phosphorylation and the enzyme is now
reclassified as Ins(3,4,5,6)P4 1-kinase/Ins(1,3,4)P3 5/6-
kinase.5 Therefore the question arises whether
Ins(1,2,4)P3 (8a) is indeed a competitive inhibitor or a
competing substrate. To clarify this aspect we have
identified potential phosphorylation products by
HPLC-MDD and evaluated the kinetics of the reaction.
Ins(1,3,4)P3 is phosphorylated at the expected positions
(Fig. 2B) leading to Ins(1,3,4,6)P4 (65%) and
Scheme 2.
In contrast to the 5-phosphatase, the phytase reaction
possesses a pH optimum in the range 4–5. From the
substrate Ins(1,2,3,6)P4 (7) phosphate is released pre-
dominantly in position C-1 (Scheme 2), but there is a
tendency toward further hydrolyzation of Ins(2,3,6)P3
(8b) to unidentified InsP2 isomers. Nevertheless under
optimized experimental conditions acceptable yields of
about 45% for 8b have been achieved.14,15
As described previously, the isolated products were
checked for isomeric impurities by HPLC-MDD.7 We
were unable to detect any contaminating inositol tris-
phosphate isomer in 8a (purityꢀ99%). Less than 2% of
byproducts were found in 8b. They were identified as
Ins(1,2,3)P3 and Ins(1,2,6)P3.
Effects on Kinase Activity
The relative potencies of non-physiological 8a and 8b to
influence the activity of recombinant human Ins(3,4,5,6)P4
1-kinase were determined and compared with the data for
Ins(1,3,4)P3 (Fig. 1). [3H]Ins(3,4,5,6)P4 was used as sub-
strate and the amount of [3H]Ins(1,3,4,5,6)P5 formed in the
presence and absence of each particular inositol
trisphosphate was analyzed.4 This study clearly demon-
strates that it is the Ins(1,2,4)P3 (8a, IC50ꢁ9 mM) which
imitates the inhibitory actions of Ins(1,3,4)P3 (IC50ꢁ4
mM) whereas the optical antipode Ins(2,3,6)P3 is almost
ineffective (8b, IC50ꢀ300 mM). Previous work with
racemic Ins(1,2,4)P3/Ins(2,3,6)P3 could not answer this
important issue.4
Figure 2. Identification of the kinase products by HPLC-MDD ana-
lysis after complete conversion of Ins(1,3,4)P3 (B) and Ins(1,2,4)P3 (8a,
C). Chromatogram A shows the separation of authentic standards.

2708
S. Adelt et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2705–2708
Ins(1,3,4,5)P4 (35%). We also observed rapid phos-
phorylation of Ins(1,2,4)P3 (8a), but with a remarkable
difference in regiospecificity [7% Ins(1,2,4,6)P4 and 93%
Ins(1,2,4,5)P4; see Figure 2C]. Consistent with the
experiment presented above, Ins(2,3,6)P3 (8b) is not a
substrate. Inspection of the molecular models and first
calculations confirm a structural relationship between
Ins(1,3,4)P3 and Ins(1,2,4)P3 (8a). According to mole-
cular modeling analysis the axial phosphate group in 2-
position of Ins(1,2,4)P3 (8a) can be nearly superimposed
on the equatorially orientated 3-phosphate group of
Ins(1,3,4)P3 by rotating the corresponding phosphate
groups around the C–O bonds. The structural resem-
blance may explain why both compounds are converted
with nearly equal rates [specific activities at substrate
saturation: 36.9 mU/mg for Ins(1,3,4)P3 and 39.8 mU/
mg for Ins(1,2,4)P3 (8a)]. Further progress in under-
standing substrate recognition and varying specificities
of the kinase for different substrates will be best addres-
sed by determination of the structure of the active site.
5. Yang, X.; Shears, S. B. Biochem. J. 2000, 351, 551.
6. Schultz, C.; Roemer, S.; Stadler, C.; Rudolf, M. T.; Wolf-
son, E.; Traynor-Kaplan, A. E.. Gastroenterology 1997, 112,
A401.
7. Adelt, S.; Plettenburg, O.; Stricker, R.; Reiser, G.; Alten-
bach, H.-J.; Vogel, G. J. Med. Chem. 1999, 42, 1262.
8. 1H NMR data for compound 5 [D2O, 400 MHz, pH adjus-
ted to 6.0 (ND4OD)]: d ppm 3.56 (dd, J=2, 10.2 Hz, H-C3);
3.78 (ꢀt, Jꢁ9.6 Hz, H-C6); 3.86 (ꢀq, Jꢁ8.7 Hz, H-C5); 3.92
(dꢀt, J=2, 7.4 Hz, H-C1); 4.15 (ꢀq, Jꢁ9.2 Hz, H-C4); 4.68
(H-C2, partially under HDO-signal); 13C NMR (101 MHz): d
ppm 70.61 (s, C3); 71.77 (dd, C6); 74.39 (dd, C1); 74.86 (d,
20
C4); 77.01 (dd, C2); 78.04 (m, C5); ½ꢀꢂD ꢃ26.3 ꢄ (c=0.7, H2O,
free acid). This agrees with published data: Mills, St. J.; Potter,
B. V. L. J. Chem. Soc., Perkin Trans. 1 1997, 1279.
9. Plettenburg, O.; Adelt, S.; Vogel, G.; Altenbach, H.-J. Tet-
rahedron: Asymmetry 2000, 11, 1057.
10. Watts, D. J.; Ashworth, J. M. Biochem. J. 1970, 119, 171.
11. The cytosolic 5-phosphatase was enriched by chromato-
graphy on heparin agarose (type II, Sigma) and DEAE-
Sepharose ff (Amersham-Pharmacia). The fraction with high-
est activity was concentrated by ultrafiltration (Centriprep-10,
Millipore).
12. Membrane-associated phytase was solubilized with 1%
hydrogenated Triton X-100 (Calbiochem). The enzyme was
partially purified by sequential biochromatography (Q-
Sepharose ff and Source 15Q, both from Amersham-Pharma-
cia).
Concluding Remarks
In vitro Ins(1,2,4)P3 (8a) delays the off-switch reaction
of the signaling molecule Ins(3,4,5,6)P4 catalyzed by
Ins(3,4,5,6)P4 1-kinase/Ins(1,3,4)P3 5/6-kinase. Unex-
pectedly, the non-physiological trisphosphate 8a func-
tions as a substrate and not as an inhibitor of the
enzyme. Phosphorylation of 8a leads predominantly to
Ins(1,2,4,5)P4, a compound with potential to act as an
Ins(1,4,5)P3-receptor agonist.16 In vivo Ins(1,2,4,5)P4 is
most likely dephosphorylated by a member of the 5-
phosphatase family, thereby regenerating Ins(1,2,4)P3.
The synthesis of a cell-permeant, bioactivatable analo-
gue of Ins(1,2,4)P3 (8a) for studies with intact cells could
be a promising new starting point for the development
of drugs that might intervene in the signal transduction
pathways with therapeutic benefit.4
13. The conversion of Ins(1,2,4,5)P4 (5) catalyzed by the 5-
phosphatase was carried out at room temperature in a volume
of 12 mL [40 mM Bis-Tris (pH 7.0), 200 mM sucrose, 5 mM
MgCl2, 0.25 mM EDTA] on a reciprocal shaker (60 rpm). At
an initial substrate concentration of 400 mM 5 and a volume
activity of about 10 mU/mL (total activity obtained from 1010
cells), it took approximately 0.5 h for a complete conversion.
New substrate was added four times from a 30 mM stock
solution. After additional 1.5 h of incubation the reaction was
stopped with 2 mL 2.5 M HCl. The pH of the mixture was
adjusted to 7, and the denaturated protein was removed by
centrifugation (6000g, 15 min, rt). Overall yield after purifica-
tion by HPLC was 95%, equivalent to 22.8 mmol 8a. For
details about reaction control, purification by HPLC and
quantification of inositol phosphates see ref 7.
14. 1H NMR data for compound 8a (D2O, 400 MHz, pH
adjusted to 6.0 (ND4OD)): d ppm: 3.5 (ꢀt, Jꢁ9.2 Hz, H-C5);
3.65 (dd, J=1.5, 9.5 Hz, H-C3); 3.84 (ꢀt, Jꢁ9.6 Hz, H-C6);
4.0 (dꢀt, J=1.8, 9.7 Hz, H-C1); 4.18 (ꢀq, Jꢁ9.3 Hz, H-C4);
4.73 (H-C2, partially covered by HDO-signal); 13C NMR
(101 MHz): d ppm: 70.68 (m, C3); 71.92 (d, C6); 74.04 (s, C5);
Acknowledgements
This work was supported by the Deutsche For-
schungsgemeinschaft (Grant VO 348/3-1) and the
Fonds der Chemischen Industrie.
20
74.6 (m, C1); 75.23 (d, C2); 77.54 (d, C4); ½ꢀꢂD (8a) ꢃ15.7 ꢄ
20
(c=3.78, H2O, free acid); ½ꢀꢂD (8b) +19.6 ꢄ (c=1.23, H2O,
free acid).
15. The dephosphorylation of Ins(1,2,3,6)P4 (7) catalyzed by
the phytase was carried out at room temperature in a volume
of 13 mL (50 mM Mes/Na+ (pH 5.1), 0.05% Triton X-100
hydr) with an initial substrate concentration of 310 mM and a
volume activity of 1.4 mU/mL (total activity obtained from
5ꢅ109 cells) on a reciprocal shaker (40 rpm). Every 1.5 h new
substrate was added from a 20 mM stock solution (three
times). After 9 h of incubation the reaction was stopped with
3 mL 2.5 N HCl. The pH of the reaction-mixture was adjusted
to 7, and the denaturated protein was removed by centrifuga-
tion (6000g, 15 min, rt). Overall yield after purification by
HPLC was 45%, equivalent to 7.2 mmol Ins(2,3,6)P3 (8b).
16. Chung, S.-K.; Shin, B.-G.; Chang, Y.-T.; Suh, B.-C.; Kim,
K.-T. Bioorg. Med. Chem. Lett. 1998, 8, 659.
References and Notes
1. Shears, S. B. Biochim. Biophys. Acta 1998, 1436, 49.
2. Xie, W.; Kaetzel, M. A.; Bruzik, K. S.; Dedman, J. R.;
Shears, S. B.; Nelson, D. J. J. Biol. Chem. 1996, 271, 14092.
3. Ho, M. W. Y.; Carew, M. A.; Yang, X.; Shears, S. B. In
Cockroft, S., Ed.; Frontiers in Molecular Biology: Biology of
Phosphoinositides; Oxford University Press, Oxford, UK,
2000; p 298.
4. Yang, X.; Rudolf, M.; Carew, M. A.; Yoshida, M.; Nerr-
eter, V.; Riley, A. M.; Chung, S. K.; Bruzik, K. S.; Potter,
B. V. L.; Schultz, C.; Shears, S. B. J. Biol. Chem. 1999, 274,
18973.