Tethered phytic acid as a probe for measuring phytase activity

Article abstract of DOI:10.1016/j.bmcl.2005.04.009

A novel approach for measuring phytase activity is presented. We have developed a new chromophoric substrate analog of phytic acid, 5-O-[6-(benzoylamino)hexyl]-d-myo-inositol-1,2,3,4,6-pentakisphosphate that permits direct measurement of the phosphate est

Full text of DOI:10.1016/j.bmcl.2005.04.009

Bioorganic & Medicinal Chemistry Letters 15 (2005) 3157–3161
Tethered phytic acid as a probe for measuring phytase activity
Duane F. Berrya,* and David A. Berryb
aVirginia Polytechnic Institute and State University, Department of Crop and Soil Environmental Sciences,
Blacksburg, VA 24061, USA
bBerry & Associates, Inc., 2434 Bishop Circle East, Dexter, MI 48130, USA
Received 19 January 2005; revised 4 April 2005; accepted 5 April 2005
Available online 4 May 2005
Abstract—A novel approach for measuring phytase activity is presented. We have developed a new chromophoric substrate analog
of phytic acid, 5-O-[6-(benzoylamino)hexyl]-D-myo-inositol-1,2,3,4,6-pentakisphosphate that permits direct measurement of the
phosphate ester bond-cleavage reaction using HPLC. This compound, along with its dephosphorylated T-phosphatidylinositol
intermediates, are quantified using reversed phase chromatography with UV detection.
Ó 2005 Elsevier Ltd. All rights reserved.
The majority of phosphorous (P) in farm animal feed
grains is present as a mixed salt of myo-inositol hexakis-
phosphate,1 more commonly referred to as phytic acid
(myo-IP6 (1)). Because grain-consuming swine and poul-
try do not produce phytase, myo-IP6 (1) is largely
unavailable as a phosphorus-containing nutrient source
and thus excreted in waste in high concentrations.1,2
Such manures are applied to cropland to enhance soil
fertility or as a means of disposal. Little information ex-
ists regarding the process of how myo-IP6 (1) is trans-
myo-IP6 (1) to buffered cell lysate and subsequent mea-
surement of the released orthophosphate (ortho-P) by
colorimetric analysis.4,5
While conducting similar experiments, we have observed
a substantial ortho-P release from bacterial cell lysate
resulting from Ôcell freeÕ phosphatase-mediated hydroly-
sis (e.g., exopolyphosphatase, non-specific acid phos-
phatases) of cell-associated phosphate compounds
including accumulated (intracellular) polyphosphate.
Interestingly, bacteria, archaea and fungi are all capable
of producing polyphosphate.6,7 It would appear that the
ortho-P release assay is not a specific measure of phytase
activity in lysed cell preparations and its use under these
conditions could result in either an exaggerated estimate
of phytase activity or possibly even a false positive test
result.
formed into
a
crop-available nutrient. Recent
experimental evidence suggests that bacterial phytase
plays an important role in these (bio)chemical
transformations.3
Phytase assays conducted on bacterial cell wall-free
lysate or whole cell lysate routinely call for addition of
Phytases catalyze the sequential hydrolysis of myo-IP6
(1), forming ortho-P and a series of partially dephospho-
rylated phosphoric esters of myo-inositol.8 In some
cases, hydrolysis may go to completion yielding the par-
ent compound myo-inositol.8 Based on biochemical
properties and amino acid sequence alignment, Oh
et al.9 have categorized phytases into two major classes,
the histidine acid phytases (containing the PhyA, PhyB,
and PhyC groups), to which most of the bacterial and
fungal phytases belong, and the alkaline phytases
(PhyD). The phytate-degrading enzyme, 3-phytase
(myo-inositol hexakisphosphate 3-phosphohydrolase,
EC 3.1.3.8; PhyA and PhyB groups) hydrolyzes myo-
IP6 (1) preferentially at the C-3 position, while 6-phytase
*
Corresponding author. E-mail: duberry@vt.edu
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.04.009

3158
D. F. Berry, D. A. Berry / Bioorg. Med. Chem. Lett. 15 (2005) 3157–3161
(myo-inositol hexakisphosphate 6-phosphohydrolase,
EC 3.1.3.26; PhyC) hydrolyzes myo-IP6 (1) preferentially
at the C-6 position.8
Our overall goal is to develop a specific and sensitive
quantitative enzyme assay capable of measuring phytase
activity in either cell culture filtrate, cell-lysate prepara-
tions, or soils using HPLC with UV detection. Develop-
ment of a method for determining soil phytase activity is
critical to understanding the fate of myo-IP6 (1) in soil
and water–sediment environments. As pointed out
recently by Turner et al.,10 Ô. . . there has been little
research on phytase in soils and waters because no
artificial substrates are presently available that can be
used for convenient measurement of activityÕ. We now
report the synthesis of 5-O-[6-(benzoylamino)hexyl]-D-
myo-inositol-1,2,3,4,6-pentakisphosphate, i.e, T-IP5 (2)
(Scheme 1) where T represents the tethered chromo-
phore, for use in the development of a phytase assay.
We chose 4-O-benzyl-1,6:2,3-di-O-cyclohexylidene-
myo-inositol (3) as the starting point for our synthesis
since it is a convenient intermediate that allows for ether
derivatization at the 5-position (phytic acid number-
ing).11 An ether linkage would be hydrolytically stable
and resistant to phosphatase activity. We also felt that
derivatization at the 5-position would be desirable for
two reasons. First, it positions the linker ÔmetaÕ to the
initial site of reaction, the 3-position, which should
minimize the chances that the linker (plus chromophore)
would interfere with the preferential active site of a
3-phytase. Second, the placement of the linker in the
meso-plane (the 5-position) simplifies the stereochemical
issues.
Following the literature procedure,12 we were able to
prepare 100 g of the pure isomer 3 with flash chromato-
graphy followed by recrystallization. The protected myo-
inositol 3 was alkylated at the open 5-position using the
mesylate of 6-azidohexan-1-ol13 and sodium hydride in
DMF. It was found that moderate heat was necessary
to ensure a complete reaction. The resinous product 4
was isolated in 55% yield. Simple hydrogenation of this
material over 10% Pd/C reduced the azide to the amine
with concomitant removal of the benzyl group at the 6-
position affording a 55% yield of 5. Selective benzoyla-
tion of the amino group in 5 was accomplished with
benzoyl cyanide14 in dichloromethane giving a 77% yield
of 6. When 6 was heated in an acetic acid/water mixture
at 100 ° C for 45 min, the 1,6:2,3-cyclohexylidene groups
were selectively removed, giving the 5-O-derivetized
myo-inositol 7 in 94% yield. Initial attempts to multi-
phosphorylate 7 by reacting the lithium salt of 7 with
tetrabenzyl pyrophosphate15,16 were unsuccessful. It
has been reported that multi-phosphorylation of inositol
derivatives can be problematic.17,18 Recent procedures
using phosphoramidites for multi-phosphorylation of
inositol derivatives offered alternatives.19–21 Through
the reaction of 7 with N,N-diethyl-1,5-dihydro-2,4,3-
benzodioxaphosphepin-3-amine22 in the presence of tet-
razole, followed by oxidation with m-CPBA, we were
able to prepare 8, in 88% yield. The 31P NMR of 8
showed three singlets in a 2:1:2 ratio, consistent with a
molecule containing a meso-plane and equilibration of
Scheme 1. Synthesis of T-IP5 (2). Reagents and conditions: (a)
MsO(CH2)6N3 (2.5 equiv), NaH (2 equiv), DMF, 55 °C, 16 h (97%);
(b) 10% Pd/C, THF–MeOH (1:1), H2, 50 psi, 8 h (55%); (c) BzCN
(1.1 equiv), CH2Cl2, 16 h, rt (76.6%); (d) HOAc, H2O, 100 °C, 24 h
(94%); (e) N,N-diethyl-1,5-dihydro-2,4,3-benzodioxaphosphepin-3-amine
(7 equiv), tetrazole (10.5 equiv), MeCN, 24 h; (f) 11 equiv MCPBA,
ꢀ50 to 25 °C (88%, two steps); (g) 5% Pd/C, MeOH, THF, H2O,
H2, 50 psi; (h) 10 equiv NH3 (95.8%, two steps).
chair conformations. These results are in contrast to
those reported for 2-O-(6-carbobenzyloxyaminohexyl)-
perbenzyl phytate (with the linker esterified to the phos-
phate group at the C-2 position) where the 31P NMR
indicated that the cyclohexane ring was twisted, report-
edly due to steric crowding, and each non-equivalent
phosphorous gave rise to a singlet.19 Removal of the
protecting groups from 8 was accomplished by hydroge-
nation over 5% Pd/C, giving 2 as a penta-dihydrogen
phosphate, which was then directly, without isolation,
converted to the decaammonium salt 2 via treatment
with ammonium hydroxide. The overall yield for these
last two steps was 96%. The 31P NMR for 2 consisted
of three singlets at 0.56, 1.19, and 3.12 ppm (integration
2:1:2), consistent with the phosphates at C-1 and C-3
being equivalent, as with the phosphates at C-4 and
C-6, and with the singlet at 1.19 ppm (integration one)
attributable to the C-2 meso-plane phosphate group.
The 1H NMR spectrum confirmed the structural

D. F. Berry, D. A. Berry / Bioorg. Med. Chem. Lett. 15 (2005) 3157–3161
3159
components, that is, linker, benzamido, and ring meth-
ines, for structure 2. Further definitive evidence for 2
was obtained from the mass spectrum. Though a nega-
tive ion Maldi mass spectrum did give an [MꢀH]ꢀ ion
at m/z 782, determination of an exact mass was not pos-
sible. However, the exact mass of 782.0168 (predicted
782.0183), consistent with the desired product 2 anion,
was achieved by negative ion electrospray.
signed T-IP4 the retention time of 7.7 min in the HPLC
chromatogram (Fig. 1). HPLC analysis of the time
course experiment revealed that as T-IP5 (2) and T-IP4
concentrations diminished, a new T-phosphatidylinosi-
tol intermediate with a retention time of 5.8 min
increased, reaching a maximum of 0.857 mM following
2.5 h of incubation (Figs. 1 and 2b). We have tentatively
identified this intermediate as T-IP3 based on the ob-
served (simultaneous) decrease in the concentration of
T-IP5 (2) and T-IP4 along with the concomitant release
of 2.57 nM ortho-P (Fig. 2a). The loss of two P groups
from T-IP5 (2) (starting concentration 1.2 mM), yielding
T-IP3, would result in a release of 2.4 mM ortho-P.
HPLC analysis of reaction intermediates also shows that
as the concentration of T-IP3 decreased, the amounts of
two T-phosphatidylinositol intermediates, with reten-
tion times of 3.9 and 4.3 min (shoulder on 3.9 min),
increased (Fig. 1). Based on the results of a recent bio-
chemical characterization study of five fungal phytases
(three isolated from Asperigillus species), which revealed
that the P group located at (five) equatorial position of
myo-IP6 (1) was preferentially removed with a tendency
for accumulation of myo-inositol 2-monophosphate (P-
group in axial position)25 along with our observations
that the concentration of T-myo-inositol (7) (0.21 mM)
formed (Fig. 3), and the amounts of T-IP3 remaining
(0.67 mM) and of ortho-P released (3.28 mM) after
5 h, we designated the intermediate with the retention
time of 3.9 min as T-myo-inositol 2-monophosphate
(T-IP1). Further investigation will be required to
confirm the identity of the T-phos- phatidylinositol
intermediates.
The 3-phytase assay consisted of 3 mL 0.2 M glucine-
HCl buffer (pH 2.6) solution containing 0.27 U of
Asperigillus ficuum 3-phytase (EC 3.1.3.8, Sigma) and
1.2 mM T-IP5 (2) (ammonium salt form). Similarly,
the 6-phytase assay consisted of 3 mL 0.2 M sodium ace-
tate buffer (pH 5.2) solution containing 0.27 U of wheat
6-phytase (EC 3.1.3.26, Sigma) and 1.2 mM T-IP5 (2).
Control assays were set up in a similar manner but with-
out phytase. Reactions were carried out in closed tinted
reaction vessels at an incubation temperature of 37 °C
(3-phytase) or 50 °C (6-phytase) with continuous stir-
ring. Aliquot samples of the 3-phytase reaction mixture
were taken (t = 0, 10, 20, 30, 40, 150, 300, and 1440 min)
for HPLC23 and/or ortho-P colorimetric24 analysis.
Aliquot samples of the 6-phytase reaction mixture were
taken at t = 0, 1, 2.5, and 6 h for HPLC23 analysis.
The chromatographic analysis of the products formed
from the 3- and 6-phytase catalyzed reaction of T-IP5
(2) was performed on an analytical HPLC system using
a reversed phase column, buffered tetrabutylammonium
hydroxide eluent23 and a PEEK tubing system through-
out to minimize adsorption of the T-IP5 (2). Since myo-
IP6 (1) readily adsorbs to stainless steel surfaces,23 it
seemed plausible that T-IP5 (2) would also. Compounds
were quantified by the external standards method.
Accumulation of the phosphatidylinositol intermediate,
T-IP3 (Figs. 1 and 2b) indicates that removal of a
phosphate group from T-IP3 to give T-IP2 is a rate-
controlling step for the 3-phytase catalyzed conversion
of T-IP5 (2). Ullah and Phillippy26 observed a similar
dephosphorylation pattern while working with myo-IP6
(1) and immobilized A. ficuum 3-phytase. In their exper-
iments, myo-IP4 tended to accumulate, indicating that
dephosphorylation of myo-IP4 and concomitant produc-
tion of myo-IP3 was a rate-controlling step. Subsequent
to observing nearly complete conversion of T-IP5 (2)
(1.2 mM) to T-myo-inositol (7) (0.96 mM) within 24 h
(Fig. 3), we did observe a slight buildup of T-IP1
(Fig. 1). By comparison, myo-IP6 (1) was converted to
T-myo-inositol pentakisphosphate (T-IP5 (2); tR =
8.5 min) was dephosphorylated in the presence of 3-phy-
tase as evidenced by appearance of T-myo-inositol phos-
phate intermediates (Fig. 1) and concomitant release of
ortho-P (Fig. 2a). In the initial reaction, 3-phytase
removed a P group, presumably from the 3-position,
yielding T-myo-inositol tetrakisphosphate (T-IP4).
Based on hydrophobicity considerations (i.e., a decrease
in the number of P groups on T-myo-inositol results in
fewer lipophilic tetrabutylammonium counterion groups
associated with the molecule), we have tentatively as-
Figure 1. Separation of T-IP5 (2) and the phosphatidylinositol intermediates, T-IP4, T-IP3, and T-IP1, produced during the 3-phytase catalyzed
dephosphorylation of T-IP5, was achieved by reversed phase ion pair chromatography. Where (a) = T-IP1 (tR 3.9 min); (b) = T-IP3 (tR 5.8 min);
(c) = T-IP4 (tR 7.7 min), and (d) = T-IP5 (2) (tR 8.5 min).

3160
D. F. Berry, D. A. Berry / Bioorg. Med. Chem. Lett. 15 (2005) 3157–3161
Figure 2. Dephosphorylation kinetic studies. (a) Progress curve showing release of ortho-P during the 3-phytase catalyzed dephosphorylation of T-
IP5 (2). (b) Progress curve showing accumulation of T-IP3 subsequent to its disappearance. The concentration of T-IP3 was determined by using T-
IP5 (2) as the standard. (c) Progress curve showing ortho-P release for 3-phytase catalyzed dephosphorylation of myo-IP6 (1) (initial concentration
1.2 mM)—included for comparative purposes.
phytase isolated from A. ficuum is capable of converting
myo-inositol 2-monophosphate to myo-inositol, albeit at
an extremely slow rate. It is also possible that conver-
sion of myo-inositol 2-monophosphate to myo-inositol
in our assays was catalyzed by a phosphatase (i.e., the
A. ficuum 3-phytase preparation from Sigma may have
contained some phosphatase). In addition to producing
an extracellular 3-phytase, A. ficuum is known to pro-
duce extracellular phosphomonoesterases capable of
dephosphorylating myo-inositol 2-monophosphate.26
These phosphomonoesterases apparently do not exhibit
activity toward myo-IP6 (1).27 Following 24 h of incuba-
tion, there was no discernable reduction in T-IP5 (2)
concentration in the control reaction as determined by
HPLC analysis. Furthermore, there was no indication
that the amide bond, which secures the benzamido chro-
mophore to the linker, was hydrolyzed. In the unlikely
event that the amide bond were to be cleaved during
an assay, the product, benzoic acid would be readily
Figure 3. HPLC chromatograms illustrating the appearance of T-myo-
inositol (7) from the 3-phytase dephosphorylation assay of T-IP5 (2).
quantifiable by HPLC. No benzoic acid was detected
during the course of the reaction of the 3-phytase with
T-IP5 (2).
T-myo-inositol was separated on a PRP-1 reversed phase column using
MeOH–H2O. The concentration of the T-myo-inositol (7) was deter-
mined by use of an authentic standard.
myo-inositol (based on ortho-P analysis) within 3 h (Fig.
2c). The complete conversion of T-IP5 (2) and myo-IP6
(1) to the T-myo-inositol and myo-inositol, respectively,
was not anticipated because phosphohydrolases classi-
fied as 3-phytase generally do not catalyze the removal
of the phosphate group from myo-inositol 2-monophos-
phate, leaving this compound as an end product.25 How-
ever, Ullah and Phillippy26 have reported that a 3-
The T-IP5 (2) probe can also serve as a substrate for 6-
phytase (Fig. 4). Based on the dephosphorylation pat-
tern, 6-phytase catalyzed phosphorus removal from T-
IP5 (tR = 11.2 min) appears to result in the formation
of T-IP4 (tR = 8.6 min) with the subsequent buildup of
T-IP3 (tR = 6.9 min). In the control reaction, following
6 h of incubation, there was no discernable reduction
in T-IP5 (2) concentration as determined by HPLC anal-

D. F. Berry, D. A. Berry / Bioorg. Med. Chem. Lett. 15 (2005) 3157–3161
3161
References and notes
1. Ertl, D. S.; Young, K. A.; Raboy, V. J. Environ. Qual.
1998, 27, 299.
2. Peperzak, P.; Caldwell, A. G.; Hunziker, R. R.; Black, C.
A. Soil Sci. 1959, 87, 293.
3. Richardson, A. E.; Hadobas, P. A.; Hayes, J. E.; OÕHara,
C. P.; Simpson, R. J. Plant Soil 2001, 229, 47.
4. Richardson, A. E.; Hadobas, P. A. Can. J. Microbiol.
1997, 43, 509.
5. Jareonkitmongkol, S.; Ohya, M.; Watanabe, R.; Takagi,
H.; Nakamori, S. J. Ferment. Bioeng. 1997, 83, 393.
6. Kulaev, I. S. The Biochemistry of Inorganic Polyphos-
phates; Wiley: New York, 1979.
7. Kornberg, A.; Rao, N. N.; Ault-Riche, D. Annu. Rev.
Biochem. 1999, 68, 89.
8. Cosgrove, D. J. Studies in Organic Chemistry 4. Inositol
Phosphates: Their Chemistry, Biochemistry, and Physiol-
ogy; Elsevier Scientific Pub.: Amsterdam, 1980, Chapter
10.
Figure 4. Separation of tethered (T)-IP5 (2) and the phosphatidylin-
ositol intermediates produced during the 6-phytase catalyzed dephos-
phorylation of T-IP5 was achieved by reversed phase ion pair
chromatography. Where (b0) = T-IP3 (tR 6.9 min); (c0) = T-IP4 (tR
8.6 min), and (d) = T-IP5 (2) (tR 11.2 min). Note that primes are used,
since T-IP3 and T-IP4 may be different isomeric forms of the
compounds referred to in Figure 1. The ratio of the MeOH–H2O
component of the mobile phase was 57:43.
ysis. Once again, there was no indication that the amide
bond, securing the benzamido chromophore to the lin-
ker, was hydrolyzed. T-myo-inositol was not detected
in the 6-phytase reaction mixture (data not shown).
9. Oh, B.-C.; Choi, W.-C.; Park, S.; Kim, Y.-o.; Oh, T.-K.
Appl. Microbiol. Biotechnol. 2004, 63, 362.
10. Turner, B. L.; Paphazy, M. J.; Haygrath, P. M.; McKel-
vie, I. D. Philos. Trans. R. Soc. London B. 2002, 357, 449.
11. Two numbering schemes are found in the literature for
myo-inositol derivatives. The numbering scheme for com-
pound 3 reflects the numbering scheme for the Chemical
Abstracts Index Name for this compound.
12. Garegg, P. J.; Iversen, T.; Johansson, R.; Lindberg, B.
Carbohydr. Res. 1984, 130, 322.
13. Hirschmann, R.; Hynes, J.; Cichy-Knight, M. A.; van
Rijn, R. D.; Sprengeler, P. A.; Spoors, P. G.; Shakespeare,
W. C.; Pietranico-Cole, S.; Barbosa, J.; Liu, J.; Yao, W.
Q.; Rohrer, S.; Smith, A. B., III J. Med. Chem. 1998, 41,
1382.
14. Murahashi, S.-I.; Naota, T.; Nakajima, N. Tetrahedron
Lett. 1985, 26, 925.
15. Ozaki, S.; Kondo, Y.; Nakahira, H.; Yamaoka, S.;
Watanabe, Y. Tetrahedron Lett. 1987, 28, 4691.
16. Kozikowski, A. P.; Fauq, A. H.; Wilcox, R. A.; Nahorski,
S. R. J. Org. Chem. 1994, 59, 2279.
In summary, we have synthesized a novel tethered phy-
tic acid probe that can serve as a chromophoric sub-
strate for phytase. Dephosphorylation of T-IP5 (2) is
readily quantified using reversed phase HPLC with
UV detection. Detection of the T-phosphatidylinositol
intermediates, T-IP4 and T-IP3, in an HPLC analysis
of assay samples containing uncharacterized phospho-
hydrolases would be indicative of 3-phytase or possibly
6-phytase. It is important to realize, however, bacterial
cells contain numerous phosphohydrolases with over-
lapping substrate specifities.28 We cannot yet rule out
that non-specific phosphohydrolases may be able to
dephosphorylate T-IP5 (2), but if this were the case,
we suspect the dephosphorylation would be at a much
slower rate, suggesting that reaction kinetics could also
play a useful role in identifying a 3- or 6-phytase.
17. Kilgour, G. L.; Ballou, C. E. J. Am. Chem. Soc. 1958, 80,
3956.
18. Krylova, V. N.; Gornaeva, N. P.; Oleinik, G. F.; Shvets,
V. I. Zh. Org. Khim. 1980, 16, 315.
19. Marecek, J. F.; Prestwich, G. D. Tetrahedron Lett. 1991,
32, 1863.
20. Riley, A. M.; Laude, A. J.; Taylor, C. W.; Potter, B. V. L.
Bioconjugate Chem. 2004, 15, 278.
21. Pietrusiewicz, K. M.; Salamonczyk, G. M.; Bruzik, K. S.;
Wieczorek, W. Tetrahedron 1992, 48, 5523.
22. Ozaki, S.; Kondo, Y.; Shiotani, N.; Ogasawara, T.;
Watanabe, Y. J. Chem. Soc., Perkin Trans. 1 1992,
729.
We are currently exploring the derivatization of T-myo-
inositol pentakisphosphate (T-IP5 (2)) with fluorescent
markers (at the amine position of the linker) for probing
the surface of minerals (e.g., goethite) or perhaps for use
in biological studies where phytate presence has impor-
tant biological implications.29
Acknowledgements
23. Lehrfeld, J. J. Agric. Food Chem. 1994, 42, 2726.
24. Bencini, D. A.; Wild, J. R.; OÕDonovan, G. A. Anal.
Biochem. 1983, 132, 254.
We thank Hubert L. Walker, Jr. for his assistance in
conducting the enzymatic studies.
25. Wyss, M.; Brugger, R.; Kronenberger, A.; Remy, R.;
Fimbel, R.; Oesterhelt, G.; Lehmann, M.; van Loon, A. P.
G. M. Appl. Environ. Microbiol. 1999, 65, 367.
26. Ullah, A. H. J.; Phillippy, B. Q. Prep. Biochem. 1988, 18,
483.
27. Ullah, A. H. J. Prep. Biochem. 1988, 18, 459.
28. Rossolini, G. M.; Schippa, S.; Riccio, M. L.; Berlutti, F.;
Macaskie, L. E.; Thaller, M. C. Cell. Mol. Life Sci. 1998,
54, 833.
Supplementary data
Selected characterization data for compounds 4, 6, 7, 8,
and 2 along with HPLC protocol. The supplementary
data is available online with the paper in ScienceDirect.
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmcl.2005.04.009.
29. Morey, J.; Orell, M.; Barcelo, M. A.; Deya, P. M.; Costa,
A.; Ballester, P. Tetrahedron Lett. 2004, 45, 1261.