Study of a cyclopamine glucuronide prodrug for the selective chemotherapy of glioblastoma

Article abstract of DOI:10.1016/j.ejmech.2009.12.067

The first glucuronide prodrug of the hedgehog signaling inhibitor cyclopamine was synthesized. The carbamoyl derivatisation of cyclopamine significantly decreased its toxicity towards the U87 human glioblastoma cell line. However, when the prodrug was incubated with β-glucuronidase in the culture media, the active drug was efficiently released thereby restoring its anti-proliferative activity.

Full text of DOI:10.1016/j.ejmech.2009.12.067

European Journal of Medicinal Chemistry 45 (2010) 1678–1682
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Preliminary communication
Study of a cyclopamine glucuronide prodrug for the selective chemotherapy
of glioblastoma
a
a
b
b
a,
*
´
´
Florian Hamon , Brigitte Renoux , Corinne Chadeneau , Jean-Marc Muller , Sebastien Papot
a
´
Universite de Poitiers, UMR-CNRS 6514, 40 avenue du Recteur Pineau, 86022 Poitiers, France
Universite de Poitiers, UMR CNRS 6187, 40 Avenue du Recteur Pineau, 86022 Poitiers, France
b
´
a r t i c l e i n f o
a b s t r a c t
Article history:
The first glucuronide prodrug of the hedgehog signaling inhibitor cyclopamine was synthesized. The
carbamoyl derivatisation of cyclopamine significantly decreased its toxicity towards the U87 human
glioblastoma cell line. However, when the prodrug was incubated with b-glucuronidase in the culture
Received 17 July 2009
Received in revised form
21 December 2009
Accepted 23 December 2009
Available online 13 January 2010
media, the active drug was efficiently released thereby restoring its anti-proliferative activity.
Ó 2010 Elsevier Masson SAS. All rights reserved.
Keywords:
Cyclopamine
Prodrug
Chemotherapy
Glioblastoma
Glucuronide
1. Introduction
family of transcription factors which play prominent roles in this
pathway. In vertebrates, three Gli proteins (Gli1, 2, 3) have been
Glioblastoma multiforme (GBM) is the most common and
aggressive primary brain tumor in adults with a very poor prog-
nosis of patients (median survival ranges from 9 to 12 months),
linked to its highly invasive characteristics and to the lack of effi-
cient treatment of these tumors [1,2]. Hence, there is an urgent
need to develop novel therapeutic strategies against GBM. These
cancers, also referred as type IV astrocytoma, derive from brain
astrocytes or neural progenitor cells [3,4]. Astrocytic commitment
and growth lie on a number of cytokines, trophic and develop-
mental factors. Among them are the hedgehog (Hh) family of
proteins which control cell proliferation and differentiation in
processes that range from insect segmentation and limb formation
to vertebrate neural tube development and bone differentiation [5].
Hh factors act on target cells through binding to a transmembrane
glycoprotein receptor called patched (Ptc), associated to a co-
receptor, called smoothened (Smo), a member of the G-protein
coupled receptors (GPCR) superfamily. In the absence of Hh ligand,
Ptc represses Smo activation. Hh binding to Ptc alleviates its
repression on Smo, leading to activation of the Smo/Ptc complex
and of its associated downstream effectors, particularly the Gli
characterized [6,7]. Overexpression of Gli2 appears to be sufficient
to generate and maintain the malignant process in some cancer
types such as basal cell carcinoma [8].
Alteration of function and/or expression of components of the
Hh pathway has been demonstrated in several cancer types. Ptc
mutations have been associated to medulloblastoma and generally,
dysfunction of the Hh pathway is involved in development and
progression of several human tumors resembling primitive
precursor cells, deriving from the skin, lung, prostate, gastrointes-
tinal tract and brain, including GBM [8–12]. Interestingly, Gli1 was
originally identified long ago as a gene amplified in a human
malignant glioma [13]. Blockade of the Hh pathway in all these
cancer types thus appears like a promising therapeutic strategy.
Prominent molecules to target this pathway are represented by
the alkaloid substance cyclopamine and its derivatives (Fig. 1).
Cyclopamine was initially discovered as the teratogenic agent from
Veratrum californicum (the hellebore or corn-lily) causing hol-
oprosencephaly (or cyclopia) in lambs, a syndrome associated with
severe alterations in the development of the central nervous
system and of the facial skeleton. Several studies suggest that
cyclopamine specifically inhibits Smoothened (Smo) activity and
thus interrupts the activated Hh–Gli pathway in cancer cells [14–
17]. Recent reports confirm the effectiveness of cyclopamine to
eradicate GBM in experimental models in vitro or in vivo. Viable
* Corresponding author.
E-mail address: sebastien.papot@univ-poitiers.fr (S. Papot).
0223-5234/$ – see front matter Ó 2010 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2009.12.067

F. Hamon et al. / European Journal of Medicinal Chemistry 45 (2010) 1678–1682
1679
OH
NO2
NO2
H
O
O
O
N
H
O
N
H
HO
O
O
-Glucuronidase
O
H
O
H
HO
OH
O
O
H
H
H
1
H
2
HN
H
O
H
OH
OH
H
OH
H
O
O
+
CO2
HO
OH
HO
NO2
OH
+
3
NO2
HO
+
H2O
OH
Cyclopamine
O
OH
Cyclopamine
+
Fig. 1.
GBM cells injected intracranially following Hedgehog blockade by
cyclopamine were no longer able to form tumors in athymic mice.
The data also indicate that a cancer stem cell population critical for
ongoing growth had been removed [18,19]. The specificity of
cyclopamine for the Hh pathway is also its main pitfall, due to its
potential ‘‘on-target’’ toxicity in somatic stem cells of the patient
which are also Hh-dependent.
In order to avoid this drawback, we propose the study of the
glucuronide prodrug 1 designed to deliver cyclopamine predomi-
nantly in the vicinity of the tumor (Fig. 1) [20]. Indeed, this prodrug
solvolysis of the ester functional groups. Moreover, as pointed out
earlier by Schmidt and co-workers, the one step deprotection of
2,3,4-tri-O-acetyl-glucuronyl methyl esters is problematic under
basic conditions due to the formation of the 4,5-unsaturated
byproduct [30]. Consequently, cleavage of the protecting groups is
usually achieved in two steps including acetates transesterification
followed by hydrolysis of the methyl ester. Recently, we discov-
ered a mild and efficient one step procedure for the deprotection
of 2,3,4-tri-O-acetyl-glucuronyl methyl esters with 3 equivalents
of dibutyltin oxide (Bu2SnO) in methanol [31]. This method
appeared to be general and tolerant of several functional groups
such as carbamates, acetals or allyl ethers. Thus, when applied to
could be selectively activated by b-glucuronidase present in high
concentration in necrotic areas of neoplasia [21]. The selectivity of
such an approach has already been demonstrated in vivo with
several glucuronide prodrugs which exhibited superior therapeutic
efficacy in various tumour xenograft models compared to standard
O
NO2
O
chemotherapy [22–24]. Thus, since elevated level of b-glucuroni-
O
4
O
O
AcO
O
dase has been observed in human GBM [25], the use of 1 should
increase cyclopamine concentration in the tumor while reducing its
delivery in normal tissues.
AcO
OAc
O
NO2
Prodrug 1 includes a nitrobenzylphenoxy carbamate linker
between the enzyme substrate and the drug [26,27]. With this
design, the carbohydrate moiety will be substantially far away from
a
cyclopamine to allow an easy recognition of 1 by b-glucuronidase.
Enzymatic hydrolysis of the glycosidic bond should generate the
phenol intermediate 2 which may then induce the release of the
free drug through a 1,6 elimination process as illustrated in Fig. 1.
O
NO2
O
5
O
N
H
O
AcO
O
O
H
AcO
OAc
O
2. Results and discussion
H
2.1. Chemistry
H
b
Prodrug 1 was prepared starting from the well-known acti-
vated carbonate 4 (Scheme 1) [28]. The latter was first coupled
with commercially available cyclopamine in the presence of
pyridine to afford the carbamate 5 in 76% yield. At this stage,
deprotection of the glucuronide moiety remains to be carried out
to obtain the expected compound 1. However, instability of
cyclopamine in acidic media [29] rules out acid-catalysed
1
OH
Scheme 1. Reagents and conditions: (a) Cyclopamine, pyridine (3 equiv.), DMF, rt, 16 h,
76% (b) Bu2SnO (3 equiv.), MeOH, reflux, 24 h, 81%.

1680
F. Hamon et al. / European Journal of Medicinal Chemistry 45 (2010) 1678–1682
the deprotection of the glucuronide derivative 5 this new proce-
dure allowed the synthesis of prodrug 1 in 81% yield after puri-
fication by flash column chromatography.
2.2. Stability and enzymatic hydrolysis
The stability of 1 was examined in both phosphate buffer
(0.02 M, pH ¼ 7.2) and cell culture media supplemented with 10%
foetal calf serum at 37 ꢀC. No decomposition of the prodrug was
detected after 48 h under these conditions. Enzymatic hydrolysis
was then carried out with an excess of
b-glucuronidase and
monitored by HPLC (Fig. 2). As expected after addition of the
enzyme, rapid disappearance of 1 was observed generating the
phenol intermediate 2. This was followed by the clean decompo-
sition of 2 which led to the release of cyclopamine together with
the formation of the 4-hydroxy-3-nitrobenzyl alcohol 3 resulting
from nucleophilic addition of water to the corresponding methy-
lene quinone (Fig. 1).
Fig. 3. Viability of U87 cells treated during 5 days with prodrug 1 in the absence
) or presence ( ) of -glucuronidase or with cyclopamine ( ).
Values were obtained from two independent experiments, each performed in hex-
aplicate and are expressed as mean ꢂ SD.
(
b
2.3. Biological evaluations
released efficiently cyclopamine thereby restoring its initial anti-
proliferative activity. In the light of these results, this prodrug
seems to be a good candidate for further in vivo investigations.
The decrease of the growth rate of U87 human GBM cells by
cyclopamine has been previously reported in vitro and in vivo
[18,19]. Prodrug 1 was then evaluated for its anti-proliferative
activity on U87 cells after a 5-day treatment (Fig. 3). When incu-
bated alone in the culture media, prodrug 1 did not affect viability
4. Experimental
of cells, except at the highest tested dose (200 mM). This result
4.1. General chemistry methods
demonstrated that the carbamoyl derivatisation of the amino group
of cyclopamine significantly reduced its toxicity. However, in the
All reactions were performed under N2 atmosphere. Solvents
used were of HPLC quality and chemicals were of analytical grade.
1H and 13C NMR were performed on an Avance 400 Bruker. The
chemical shifts are expressed in part per million (ppm) relative to
presence of
the viability of U87 cells with an IC50 value of 21
M). This can be unambigu-
b
-glucuronidase (40 U/ml), 1 dramatically decreased
mM similar to that
obtained for cyclopamine (IC50 ¼ 15.5
m
ously attributed to an efficient release of the drug in the culture
media as previously observed in the course of enzymatic hydrolysis
experiments.
TMS (
d
¼ 0 ppm) and the coupling constant J in hertz (Hz). NMR
multiplicities are reported using the following abbreviations:
b ¼ broad, s ¼ singulet, d ¼ doublet, t ¼ triplet, q ¼ quadruplet,
m ¼ multiplet. The reaction progress was monitored on precoated
3. Conclusion
silica gel TLC plates MACHEREY-NAGEL ALUGRAMÒ SIL G/UV254
.
(0.2 mm silica gel 60 Å). Spots were visualized under 254 nm UV
light and/or by dipping the TLC plate into a solution of 3 g of
phosphomolibdic acid in 100 mL of ethanol followed by heating
with a heat gun. Flash column chromatography was performed
In summary, we have prepared the first glucuronide prodrug of
cyclopamine. This prodrug exhibited reduced toxicity compared to
the free drug on U87 glioblastoma cells. On the other hand, enzy-
matic cleavage of the glucuronide moiety by
b
-glucuronidase
using MACHEREY-NAGEL silica gel 60 (15–40 mm).
100%
80%
60%
40%
20%
0%
Prodrug 1
2
3
Cyclopamine
-2
0
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
32
Time (hours)
Fig. 2. Enzymatic cleavage of prodrug 1 (0.1 mg.mLꢁ1);
b
-glucuronidase (133 U.mLꢁ1) was introduced at t ¼ 0 h.

F. Hamon et al. / European Journal of Medicinal Chemistry 45 (2010) 1678–1682
1681
4.2. Synthesis and characterization of described compounds
(0.02 M, pH 7) at 37 ꢀC. Aliquots of 20
mL were taken at indicated
time and analyzed by HPLC.
4.2.1. Synthesis and characterization of compound 5
Carbonate 4 (0.077 g, 0.12 mmol) and cyclopamine 1 (0.050 g,
0.12 mmol) were dissolved in DMF (3.7 mL). Pyridine (0.03 mL,
0.36 mmol) was added and the solution was stirred at room
temperature for 12 h. The mixture was then hydrolyzed with a satu-
rated NH4Cl solution and extracted three times with EtOAc. The
organic layer was dried over MgSO4 and evaporated to dryness. The
residue was purified by flash chromatography (30/70 and 50/50
EtOAc/PE). Product 5 was isolated as a white powder in 76% yield
4.5. Cell culture
The U87 human glioblastoma cell line was maintained in Dul-
becco’s modified Eagle’s medium (DMEM) with GlutaMAXÔ I and
sodium pyruvate (Invitrogen), supplemented with 10% fetal calf
serum and 100 U/mL penicillin and 100 mg/mL streptomycin (Invi-
trogen). Cells were incubated in a humidified 95% air/5% CO2
controlled atmosphere at 37 ꢀC.
(0.085 g, 0.092 mmol). 1H NMR (CDCl3)
d (ppm): 0.90 (m, 3H); 0.95
(m, 3H); 1.21 (m, 8H); 1.48 (m, 2H); 1.70 (m,12H); 2.05–2.03 (2s, 6H);
2.08(s, 3H);2.15 (m, 5H);2.30(m,1H); 2.80(m, 2H); 3.14 (m,1H);3.46
(m, 2H); 3.50 (m,1H); 3.70 (s, 3H); 4.14 (d,1H, J ¼ 7.5 Hz); 5.12 (s, 2H);
5.28 (m, 5H); 7.39 (d,1H, J ¼ 8.7 Hz); 7.54 (dd,1H, J ¼ 8.7 Hz, J ¼ 2 Hz);
4.6. Cell proliferation
Cell viability was evaluated using the CellTiter 96Ò Aqueous One
Solution Cell Proliferation Assay (Promega). U87 cells were plated in
7.81 (d, 1H, J ¼ 2 Hz). 13C NMR (CDCl3)
d (ppm): 10.47, 13.52, 14.20,
96-well plates at a density of 400 cells/well in 100
a 24 h incubation, medium was replaced by medium containing the
prodrug ꢂ -glucuronidase (40 U/mL) or cyclopamine. Control cells
were incubated in the presence of DMSO (used for prodrug or
cyclopamine solubilization) ꢂ -glucuronidase. Cell viability was
determined after5 daysof treatment byadding 20
L of CellTiter 96Ò
mL medium. After
18.67, 20.44, 20.52, 20.59, 21.01, 24.58, 28.31, 29.03, 31.36, 31.44,
32.59,35.11, 36.53, 38.13, 41.77, 49.20, 50.96,51.98,53.07, 53.44, 60.40,
65.33, 68.68, 70.16, 71.01, 71.77, 72.58, 85.02, 99.71, 120.11, 121.82,
124.56, 126.36, 133.05, 133.33, 141.22, 141.64, 143.41, 148.73, 157.3,
162.55, 166.71, 169.26, 169.32, 170.02. ESI-MS: m/z [M þ H]þ 923.
b
b
m
Aqueous One Solution Reagent into each well 3 h before measuring
the optical density. Metabolically active cells convert 3-(4,5-dime-
thylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-
2H-tetrazolium (MTS) into a coloured formazan product that was
measured in a spectrophotometric microplate reader at 490 nm.
The OD of control cells was considered as the 100 value.
4.2.2. Synthesis and characterization of prodrug 1
Compound 5 (0.085 g, 0.092 mmol) was dissolved in MeOH
(2.5 mL). After the addition of dibutyltin oxide (0.067 g, 0.28 mmol)
the mixture was heated at reflux overnight. The mixture was then
evaporated and subjected to flash column chromatography (10/90
and 30/70 MeOH/CH2Cl2) to furnish the expected prodrug 1 as
a white solid in 81% yield (0.057 g, 0.075 mmol). 1H NMR (CD3OD/
Acknowledgments
(CD3)2CO 3/2) d (ppm): 0.90 (m, 3H); 1.00 (m, 8H); 1.04 (m,1H); 1.24
(m, 5H); 1.43 (m,1H); 1.55 (m, 2H); 1.65 (s, 3H); 1.77 (m, 3H); 1.91 (m,
2H); 2.13 (m, 3H); 2.26 (m,4H); 2.83 (m, 1H); 3.00 (m, 1H); 3.24 (m,
1H); 3.30 (m,1H); 3.43 (m,1H); 3.61 (m, 5H); 3.91 (d,1H, J ¼ 7.5 Hz);
5.17 (bs, 3H); 5.37 (bs, 1H); 7.55 (d, 1H, J ¼ 8.7 Hz); 7.65 (dd, 1H,
J ¼ 8.7 Hz, J ¼ 2 Hz); 7.87 (d,1H, J ¼ 2 Hz).13C NMR (CD3OD/(CD3)2CO
The authors thank CNRS and La Ligue Nationale contre le Cancer
´
(Comite de Charentes-Maritime) for financial support of this study.
References
[1] T. Demuth, M.E. Berens, J. Neurooncol. 70 (2004) 217–228.
[2] H. Ohgaki, P. Kleihues, J. Neuropathol. Exp. Neurol. 64 (2005) 479–489.
[3] A.L. Vescovi, R. Galli, B.A. Reynolds, Nat. Rev. Cancer 6 (2006) 425–436.
[4] M.R. Quigley, C. Post, G. Ehrlich, Neurosurg. Rev. 30 (2007) 16–20.
[5] M.F. Mehler, Results Probl. Cell Differ. 39 (2002) 27–52.
3/2) d (ppm): 177.9,158.9,151.1,144.0,143.2,141.6,134.9,132.4,127.8,
125.6, 122.5, 119.0, 102.4, 86.2, 77.6, 76.1, 74.4, 73.6, 73.2, 72.3, 66.6,
64.4, 53.3, 51.6, 50.0, 49.2, 43.3, 42.8, 42.6, 39.4, 37.6, 35.8, 33.4, 32.1,
32.0, 29.8, 25.6, 20.9, 19.0, 13.8, 10.8. ESI-MS: m/z [M ꢁ H]ꢁ 781.
[6] J. Briscoe, P. Therond, Dev. Cell 8 (2005) 143–151.
[7] J.-M. Muller, L. Chevrier, S. Cochaud, A.-C. Meunier, C. Chadeneau, Drug Discov.
Today Dis. Mech. 4 (2007) 285–291.
4.3. HPLC analysis
[8] M.E. Hutchin, M.S. Kariapper, M. Grachtchouk, A. Wang, L. Wei, D. Cummings,
J. Liu, L.E. Michael, A. Glick, A.A. Dlugosz, Genes Dev. 19 (2005) 214–223.
[9] C.D. Peacock, Q. Wang, G.S. Gesell, I.M. Corcoran-Schwartz, E. Jones, J. Kim,
W.L. Devereux, J.T. Rhodes, C.A. Huff, P.A. Beachy, D.N. Watkins, W. Matsui,
Proc. Natl. Acad. Sci. U.S.A. 104 (2007) 4048–4053.
[10] A. Ruiz i Altaba, C. Mas, B. Stecca, Trends Cell Biol. 17 (2007) 438–447.
[11] B. Stecca, A. Ruiz i Altaba, J. Neurobiol. 64 (2005) 476–490.
[12] S. Sun, J.H. Schiller, M. Spinola, J.D. Minna, J. Clin. Invest. 117 (2007) 2740–2750.
[13] K.W. Kinzler, S.H. Bigner, D.D. Bigner, J.M. Trent, M.L. Law, S.J. O’Brien,
A.J. Wong, B. Vogelstein, Science 236 (1987) 70–73.
Analytical HPLC was carried out using a DIONEX P680 System with
UV variable wavelength detector and DEDL PLS-ELS 2100 detector.
Enzymatic hydrolysis analysis were performed on a reverse phase
column chromatography (Chromolite Performance, RP18E,
100 ꢃ 4.6 mm) using
a mobile phase (1 mL/min) of CH3CN/
H2O þ 0.1%TFA (gradient from 1:9 to 10:0). Retention time for
compounds 1, 2, 3 and cyclopamine were 9.02, 11.36, 5.27 and
6.44 min respectively. Peak area and calibration curves were
obtained with DIONEX Chromeleon software.
[14] A.S. Kiselyov, Anticancer Agents Med. Chem. 6 (2006) 445–449.
[15] A.S. Kiselyov, S.E. Tkachenko, K.V. Balakin, A.V. Ivachtchenko, Expert Opin.
Ther. Targets 11 (2007) 1087–1101.
[16] M. Lauth, R. Toftgård, Curr. Opin. Investig. Drugs 8 (2007) 457–461.
[17] N. Mahindroo, C. Punchihewa, N. Fujii, J. Med. Chem. 52 (2009) 3829–3845.
[18] E.E. Bar, A. Chaudhry, A. Lin, X. Fan, K. Schreck, W. Matsui, S. Piccirillo,
A.L. Vescovi, F. DiMeco, A. Olivi, C.G. Eberhart, Stem Cells 10 (2007) 2524–2533.
[19] V. Clement, P. Sanchez, N. de Tribolet, I. Radovanovic, A. Ruiz i Altaba, Curr.
Biol. 17 (2007) 165–172.
4.4. Stability and enzymatic hydrolysis
4.4.1. Compound stability
Prodrugs 1 was placed in phosphate buffer (0.02 M, pH 2.1 or 7)
and 10% foetal calf serum at 37 ꢀC for a period of 48 h. HPLC analysis
showed no detectable degradation of this compound in these
conditions.
[20] For a review see M. de Graaf, E. Boven, H.W. Scheeren, H.J. Haisma, H. Pinedo.
Curr. Pharm. Des. 8 (2002) 1391–1403.
[21] K. Bosslet, J. Czech, D. Hoffmann, Tumor Target. 1 (1995) 45–50.
[22] K. Bosslet, R. Straub, M. Blumrich, J. Czech, M. Gerken, B. Sperker, H.K. Kroemer,
J.-P. Gesson, M. Koch, C. Monneret, Cancer Res. 58 (1998) 1195–1201.
[23] P.H. Houba, E. Boven, C.A. Erkenlens, R.G. Leenders, J.W. Scheeren,
H.M. Pinedo, H.J. Haisma, Br. J. Cancer 78 (1998) 1600–1606.
[24] P.H. Houba, E. Boven, I.H. Van der Meulen-Muileman, R.G. Leenders,
J.W. Scheeren, H.M. Pinedo, H.J. Haisma, Br. J. Cancer 78 (2001) 550–557.
[25] C. Nygren, H. von Holst, J.-E. Mansson, P. Fredman, Acta Neurochir 139 (1997)
146–150.
4.4.2. Enzymatic cleavage of prodrug 1
Escherichia coli
Aldrich (reference: G8162). Prodrug 1 (0.1 mg/mL) were incubated
with E. coli -glucuronidase (133 U/mL) in phosphate buffer
b-glucuronidase was purchased from Sigma–
b

1682
F. Hamon et al. / European Journal of Medicinal Chemistry 45 (2010) 1678–1682
[26] S. Papot, I. Tranoy, F. Tillequin, J.-C. Florent, J.-P. Gesson, Curr. Med. Chem.
Anticancer Agents 2 (2002) 155–185.
[27] I. Tranoy-Opalinski, A. Fernandes, M. Thomas, J.-P. Gesson, S. Papot, Anticancer
Agents Med. Chem. 8 (2008) 618–637.
[28] J.-C. Florent, X. Dong, G. Gaudel, S. Mitaku, C. Monneret, J.-P. Gesson,
J.-C. Jacquesy, M. Mondon, B. Renoux, S. Andrianomenjanahary, S. Michel,
M. Koch, F. Tillequin, M. Gerken, J. Czech, R. Straub, K. Bosslet, J. Med. Chem. 41
(1998) 3572–3581.
[29] M.R. Tremblay, M. Nevalainen, S.J. Nair, J.R. Porter, A.C. Castro, M.L. Behnke,
L.-C. Yu, M. Hagel, K. White, K. Faia, L. Grenier, M.J. Campbell, J. Cushing,
C.N. Woodward, J. Hoyt, M.A. Foley, M.A. Read, J.R. Sydor, J.K. Tong,
V.J. Palombella, K. McGovern, J. Adams, J. Med. Chem. 51 (2008)
6646–6649.
[30] A. El Alaoui, F. Schmidt, C. Monneret, J.-C. Florent, J. Org. Chem. 71 (2006)
9628–9636.
[31] The full study concerning this procedure will be published elsewhere.