Efforts toward elucidating Thalidomide's molecular target: An expedient synthesis of the first Thalidomide biotin analogue

Article abstract of DOI:10.1039/c0ob00060d

Herein we describe the synthesis of the first Thalidomide-biotin analogue in order to initiate investigations into the unknown molecular mode of action of Thalidomide. In this manner we describe the attachment of biotin tether through the Huisgen 1,3-dipo

Full text of DOI:10.1039/c0ob00060d

View Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/obc | Organic & Biomolecular Chemistry
Efforts toward elucidating Thalidomide’s molecular target: an expedient
synthesis of the first Thalidomide biotin analogue†
Scott G. Stewart,* Carlos J. Braun, Marta E. Polomska, Mahdad Karimi, Lawrence J. Abraham and
Keith A. Stubbs
Received 29th April 2010, Accepted 25th June 2010
First published as an Advance Article on the web 13th July 2010
DOI: 10.1039/c0ob00060d
Herein we describe the synthesis of the first Thalidomide–
biotin analogue in order to initiate investigations into the
unknown molecular mode of action of Thalidomide. In this
manner we describe the attachment of biotin tether through
the Huisgen 1,3-dipolar cycloaddition or “click” synthetic
methodology.
to use Thalidomide (1), (ThalomidTM ), and analogues Revlimid
and Actimid, (both containing an aromatic amino functionality),
for the treatment of multiple myeloma and erythema nodosum
leprosum (ENL).4
Despite being known for over 50 years, determining the precise
mechanism of action of Thalidomide (1) has remained elusive.5
There have been many reports of Thalidomide (1) and the influence
of several cellular processes, for example the binding with al-acid
glycoprotein (AGP), the interaction with g-rich proteins in FGF-
2 gene, cyclooxygenase (COX-1 and 2) inhibition and possible
protein kinase inhibition.6 Another current hypothesis reported
in the literature is that Thalidomide (1) inhibits in some manner
the expression of the pro- inflammatory cytokine, tumour necrosis
factor (TNF).3,7 More specifically, the action of Thalidomide (1)
is thought to involve the inflammatory NFkB signalling pathway,
specifically inhibiting the activity of an IkB kinase, IKKa.8 In
an effort to elucidate the precise mode of action of Thalidomide
(1), we envisaged the preparation of activity-based probe (Fig. 1),
that if prepared, would allow us to gain insight and characterize
Thalidomide’s molecular target(s). This would also allow us
to elaborate on our initial Thalidomide analogue studies9,10 in
being able to develop more potent compounds against a known
biological target.
We felt that a biotinylated Thalidomide derivative based on 2
would suit the purpose of determining the precise mechanism of
action of Thalidomide (1) as it contains the general design features
that should be incorporated into molecules of this type. The
Thalidomide fragment would act as a recognition element which
would specifically bind to the molecular target(s) of interest. Biotin
would act as a reporter that would permit rapid and sensitive
detection of the cellular location of the probe and finally the
polyethylene glycol (PEG) unit would act as a linker molecule
between the reporter and the recognition element to increase
solubility of the overall molecule in water and to ensure that the
reporter does not interfere with binding to the target(s).
Introduction
The history behind the pharmaceutical development of
Thalidomide, [(R,S)-2-(2,6-dioxo-3-piperidinyl)-1H-isoindole-
1,3(2H)-dione (1) (Fig. 1)] is perhaps one of the most recognized
accounts of a pharmaceutical for chemists and biochemists
alike. This drug, as a racemic mixture, was administered in the
1950’s to pregnant women as a treatment for insomnia and an
antiemetic agent for morning sickness. It was later discovered
however to have side effects in that the R-isomer (at C3¢) (R)-1 was
responsible for the sedative response and the S-isomer (S)-1 had
teratogenic properties.1,2,3 As a result of these findings, in 1962, this
commercially popular drug was withdrawn, but not before 10,000
infants with various limb malformations were born. Currently,
both academia and the pharmaceutical industry have highlighted
new and exciting areas in Thalidomide (1) research with the most
notable being the use of (1) as a treatment for multiple myeloma,
an as yet incurable form of bone marrow cancer. To that end, the
pharmaceutical company Celgene have received FDA approval
The placement of the PEG-linker was also chosen because of
earlier analogue studies3 by our group9,10 and the pioneering work
of Muller11 and Hashimoto,12 which suggest that functionalization
of the left side phthalimide ring (phthaloyl) of Thalidomide results
in compounds with improved biological activity.
Fig. 1 Thalidomide and a generic target biotinylated Thalidomide
analogue.
Results and Discussion
School of Biomedical, Biomolecular and Chemical Science, University of
Western Australia, 35 Stirling Hwy, Crawley, 6009, WA, Australia. E-mail:
sgs@cyllene.uwa.edu.au
† Electronic supplementary information (ESI) available: General ex-
perimental procedures and 1H NMR comparison study. See DOI:
10.1039/c0ob00060d
In the retrosynthetic analysis of our biotin–Thalidomide analogue
2 we could foresee two methods to attach the PEG-biotin
fragment, the first through a simple nucleophilic displacement and
the second through “click” chemistry.13 The synthesis of the three
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4059–4062 | 4059
©

View Online
different Thalidomide analogues that could be putatively used to
construct 2 are illustrated in Scheme 1. Following the procedure
of Hess,14 Thalidomide (1) was treated with an aqueous solution
of formaldehyde to afford 3 in 74% yield and thus introducing a
new hydroxymethylene group to the piperidine-2,6-dione ring.
Scheme 1 Synthesis of Thalidomide fragments
Scheme 2 Synthesis of biotin precursors
Treatment of alcohol 3 with thionyl chloride furnished the
alkylchloride 4 in 70% yield. A Finkelstein reaction of this latter
compound then produced the iodoalkyl compound 5 (56%).14
Thalidomide derivative 5 served as one of the substrates for our
key biotin–PEG attachment sequence.
symmetrical amine 4,7,10-trioxa-1,13-tridecanediamine (71%).19
Formation of the azide 9 from 8 proceeded in excellent yield
(83%) using the novel diazotransfer reagent imidazole-1-sulfonyl
azide.20 The final biotin analogue, 10 was prepared according to
the literature.17
The attachment of the two fragments (Scheme 3) through a
simple SN2 process proved problematic. When alkyl iodide 5 was
treated with the biotin fragment 8 no reaction was observed even
over long periods. Subjecting a similar substrate, compound 6, to
standard Huisgen 1,3-dipolar cycloaddition conditions (CuSO4,
ascorbic acid and alkyne 10), also failed to provide any of the
desired biotinylated product. It is assumed that the azide was
not remote enough for the alkyne to approach in this instance.
The final Thalidomide substrate 7, containing a longer alkyne
tether, was subjected to reaction with azido-PEG-biotin 9. Under
the “click” reaction conditions, CuI, diisopropylethyl amine, the
Alternatively, one of the Huisgen cycloaddition precursors, alkyl
azide 6, was also synthesised following the procedure of Hess (see
ESI†).14 The last of our key Thalidomide fragments, compound
7, was prepared following treatment of Thalidomide (1) with
NaH, NaI and 5-chloro-1-pentyne. This secondary compound
was devised to provide an alternative for the Huisgen 1,3-dipolar
cycloaddition dienophile, where the point of attachment was
considered more remote from the Thalidomide ring system.
The synthesis of the biotin fragments 8,15 9 and 1016,17,18
(Scheme 2) were achieved using a divergent synthesis utilizing
a known activated biotin ester 11, which is readily prepared
from biotin 12.16 Following this process the amine derivatized
PEG linker 8 was prepared through amidation of 11 with the
4060 | Org. Biomol. Chem., 2010, 8, 4059–4062
This journal is
The Royal Society of Chemistry 2010
©

View Online
Scheme 3 Synthesis of Thalidomide biotin analogue
desired 1,4-triazole 13 was produced with complete regioselectivity
and excellent yields (99%) as a mixture of diastereoisomers.13f
In many instances in the literature we found that many
synthetically prepared biotinylated analogues were not rigorously
identified spectroscopically. This may be in part due to their
formation being carried out on a small scale or only in a reaction
in vivo. In this described procedure we sought full characterisation
prior to any further biological studies.
1H, H1∧), 4.42 (t, J = 7.2 Hz, 2H, H1¢¢¢), 4.47 (m, 1H, H4∧), 5.02
(m, 1H, C3¢), 5.37 (s, 1H, H1∧), 6.11 (s, 1H, H3∧), 6.61 (m, 1H,
NHCO), 7.44 (d, J = 3 Hz, 1H, H5¢¢), 7.76 (m, 2H, Ar–CH), 7.87
(m, 2H, Ar–CH); 13C NMR (150.9 MHz, CDCl3): 22.12 (C4¢/C5¢),
23.2 (C2¢¢¢/C3*), 25.7 (C9¢¢¢/C2*/C5*), 27.3 (C2¢¢), 28.2 and 28.3
(C3¢¢/C4*/C9¢¢¢/C2*/C5*), 29.1 (C9¢¢¢/C2*/C5*), 30.4 (C9¢¢¢/C2*/C5*),
32.1 (C4¢/C5¢), 36.1 (C9¢¢¢/C2*/C5*), 37.8 (C10¢¢¢), 40.2 (C1¢¢), 40.7 (C7∧),
47.1 (C1¢¢¢), 50.3 (C3¢), 55.7 (C5∧), 60.2 (C8∧), 61.9 (C4∧), 67.4 (C3¢¢¢),
70.6, 70.5, 70.3, 70.2, 70.11 (C4¢¢¢/C5¢¢¢/C6¢¢¢/C7¢¢¢/C8¢¢¢), 121.6 (C5¢¢),
123.8 (2 x Ar–CH), 131.0 (2 x Ar–C), 134.6 (2 x Ar–CH), 147.2
(C4¢¢), 163.6 (C2∧), 167.5 (C1 and C3), 168.8 (C2¢), 171.1 (C1¢), 173.1
(NHCO, C1*); IR (NaCl, cm-1) u: 3285 (N–H), 2925, 1716 (C O),
1680 (C O), 1391, 1151 (C–O), 722; MS FAB, m/z (%): 797 (100)
[M+H]∑+, 796 (9) [M]∑+, 767 (12), 514 (45); FAB HRMS calculated
for C38H52N8O9S [M+H]∑+ = 797.3578; found [M+H]∑+ = 797.3633.
1
A comparison of the H NMR of each of the cycloaddition
substrates and the biotinylated Thalidomide indicates a signal
attributed to the triazolic proton was calculated to be equal
to the integral of the signal assigned to the H3¢ proton of the
1
Thalidomide backbone. Other key H NMR resonances can be
clearly associated with this new biotin Thalidomide derivative 13,
while the 1,5-triazole regioisomer is not observed. Additionally,
a [M+H]+ peak was observed at m/z 797 in the Fast Atom
Bombardment (FAB) mass spectrum of compound 13, whilst a
lack of an absorption peak at 2115 cm-1 in the IR spectrum
indicated the absence of any staring material (see supporting
information and illustrated 1H NMR spectra comparison, ESI†).
In this procedure, Azido-PEG-biotin (9) (15 mg, 0.0360 mmol)
was added in one portion to stirred solution of pentynyl-
Thalidomide 7 (12 mg, 0.0396 mmol), CuI (1 mg, 0.00396 mmol),
diisopropylethylamine (7 mL, 0.0396 mmol) in acetonitrile (2 mL)
and the ensuing mixture stirred at room temperature for 18 h. Fol-
lowing TLC analysis indicated starting material remaining, thus
the reaction mixture was heated to 40 ◦C and stirred for a further
48 h. The resulting mixture was then fused to reverse phase silica
and purified using gravity reverse-phase column chromatography
(elution with 7 : 3 methanol–water) to afford the 1,2,3-triazole
product (13) as a waxy white solid (27 mg, 99%); Rf 0.31 (7 : 3
methanol–water); 1H NMR (600 MHz, CDCl3): 1.41 (m, 2H, H4*)
1.56–1.78 (m, 6H, H9¢¢¢/H2*/H5*), 1.95 (m, 2H, H2¢¢), 2.08–2.17
(m, 5H, (H4¢/H5¢/H2¢¢¢/H3*), 2.68–2.89 (m, 5H, H7∧/H4¢/H5¢/H3¢¢),
2.87 (m, 1H, H7∧), 2.95 (m, 1H, H4¢/H5¢), 3.12 (m, 1H, H5∧), 3.33
(m, 2H, H10¢¢¢), 3.41 (td, J = 6 and 2.4 Hz, 2H, H3¢¢¢), 3.52–3.63
(m, 10H, H4¢¢¢/H5¢¢¢/H6¢¢¢/H7¢¢¢/H8¢¢¢), 3.86 (m, 2H, H1¢¢), 4.29 (m,
TNF Reporter Gene Biological Assay
The biotin-linked Thalidomide analogue 13 was initially assayed
in the Jurkat T-cell line containing the GFP reporter construct
(see ESI†) to determine whether the biotinylation has affected
the parent drug’s activity. At 10 and 100 mM concentrations, the
biotin-linked Thalidomide analogue (13) showed no significant
difference in TNF inhibition levels to that of (R,S)-Thalidomide
(1) (Fig. 2). At 100 mM concentration, the biotin-linked analogue
13 displayed a 65% inhibition of TNF expression, comparable
to 67% inhibition by that of (R,S)-Thalidomide (1). Despite
these promising results, the observed DMSO-induced TNF in-
hibition is higher than previously observed. Some of the increased
solvent-based inhibition is due to a 1% cell volume of DMSO
being used in this assay. Increasing volumes of DMSO have
been shown to cause inhibition of NFkB signalling and TNF
formation in rat liver studies, which may explain why DMSO-
induced TNF inhibition is observed.21 Despite these DMSO-based
discrepancies, the comparison of activity between the biotin-linked
Thalidomide analogue 13 and (R,S)-Thalidomide (1) remains
relatively unaffected. Thus, these results show that attachment of
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4059–4062 | 4061
©

View Online
Fig. 2 Biological activity of the biotin-linked Thalidomide analogue (13) compared to (R,S)-Thalidomide (1).
a biotin fragment (13) to the piperidine ring does not significantly
affect the biological activity of Thalidomide (1).
Reviews Of Immunology, 2008, 27, 111; (d) R. Garc´ıa-Sanz, Expert
Opin. Pharmacother., 2006, 7, 195.
7 (a) S. Niwayama, C. Loh, B. E. Turk, J. O. Liu, H. Miyachi and Y.
Hashimoto, Bioorg. Med. Chem. Lett., 1998, 8, 1071; (b) M. Melchert
and A. List, Int. J. Biochem. Cell Biol., 2007, 39, 1489–1499.
8 J. Keifer, D. C. Guttridge, B. P. Ashburner and A. S. Baldwin, J. Biol.
Chem., 2001, 276, 22382.
In conclusion, we have devised a short elegant route to a
biotin derived Thalidomide analogue 13. In this four step process
we have used the a novel diazotransfer reagent to generate the
biotin-azide fragment and a Huisgen 1,3-dipolar cycloaddition or
click chemistry as the key reaction. Given that the production
of 13 as a putative activity-based probe is viable it gives us
the opportunity to investigate the localisation of Thalidomide
and thus identify, in principle, this molecules cellular targets.
Using fluorescein isothiocyanate (FITC)-conjugated streptavidin
and confocal microscopy we envisage that we will visualise the
cellular location of 13 and thus give insight into Thalidomide’s (1)
mode of action. What is also of interest to us, and currently under
development, is an analogue of 13 incorporating a photoaffinity
label.22 Molecules such as these6a,23 could be used in conjunction
with streptavidin affinity-based chromatography24 to elucidate
whether Thalidomide (1) has multiple cellular targets.
9 S. G. Stewart, C. J. Braun, S-L. Ng, M. E. Polomska, M. Karimi and
L. J. Abraham, Bioorg. Med. Chem., 2010, 18, 650.
10 S. G. Stewart, D. Spagnolo, M. E. Polomska, M. Sin, M. Karimi and
L. J. Abraham, Bioorg. Med. Chem. Lett., 2007, 17, 5819.
11 (a) G. W. Muller, R. Chen, S. Huang, L. G. Corral, L. Wong, R. T.
Patterson, Y. Chen, G. Kaplan and D. I. Stirling, Bioorg. Med. Chem.
Lett., 1999, 9, 1625; (b) J. B. Marriott, G. W. Muller, D. Stirling and
A. G. Dalgleish, Expert Opin. Biol. Ther., 2001, 1, 675.
12 (a) Y. Hashimoto, Arch. Pharm., 2008, 341, 536; (b) T. Nakamura, T.
Noguchi, H. Kobayashi, H. Miyachi and Y. Hashimoto, Chem. Pharm.
Bull., 2006, 54, 1709.
13 (a) R. Huisgen, 1,3-Dipolar Cycloaddition ChemistryA Padwa, Ed.;
Wiley, New York, New York, 1984; pp 1–176; (b) V. V. Rostovtsev,
L. G. Green, V. V. Fokin and K. B. Sharpless, Angew. Chem., Int. Ed.,
2002, 41, 2596; (c) S. N. Lam, P. Acharya, R. Wyatt, P. D. Kwong and
C. A. Bewley, Bioorg. Med. Chem., 2008, 16, 10113; (d) R. Huisgen,
Angew. Chem. Int. Ed., 1963, 75, 604; (e) W. G. Lewis, L. G. Green,
F. Grynszpan, Z. Radicacute, P. R. Carlier, P. Taylor, M. G. Finn and
K. B. Sharpless, Angew. Chem. Int. Ed., 2002, 41, 1053; (f) C. W. Tornoe,
C. Christensen and M. Meldal, J. Org. Chem., 2002, 67, 3057.
14 S. Hess, M. A. Ackermann, S. Wnendt, K. Zwingenberger and K. Eger,
Bioorg. Med. Chem., 2001, 9, 1279.
Acknowledgements
The Authors would like to thank the Cancer Council of Western
Australia for the award of a Cancer Council Grant. The authors
would also like to thank Dr Lindsay Byrne for NMR assistance
and Dr Tony Reeder for mass spectra acquisition.
15 D. S. Wilbur, D. K. Hamlin, R. L Vessella, J. E. Stray, K. R. Buhler,
P. S. Stayton, L. A. Klumb, P. M. Pathare and S. A. Weerawarna,
Bioconjugate Chem., 1996, 7, 689.
16 K. A. Stubbs, A. Scaffidi, A. W. Debowski, B. L. Mark, R. V. Stick and
D. J. Vocadlo, J. Am. Chem. Soc., 2008, 130, 327.
17 A. J. Link and D. A. Tirrell, J. Am. Chem. Soc., 2003, 125, 11164.
18 A. J. Link, M. K. S. Vink, N. J. Agard, J. A. Prescher, C. R. Bertozzi
and D. A. Tirrell, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 10180.
19 S. J. Williams, O. Hekmat and S. G. Withers, ChemBioChem, 2006, 7,
116.
20 E. D. Goddard-Borger and R. V. Stick, Org. Lett., 2007, 9, 3797.
21 C. K. Chang, S. Llanes and W. Schumer, J. Surg. Res., 1999, 82, 294.
22 T. Kan, Y. Tominari, Y. Morohashi, H. Natsugari, T. Tomita, T.
Iwatsubo and T. Fukuyama, Chem. Commun., 2003, 2244.
23 S. M. Capitosi, T. P. Hansen and M. L. Brown, Org. Lett., 2003, 5,
2865.
References
1 R. Brynner, T. Stephens, Dark Remedy, The impact of thalidomide and
its revival as a vital medicine; Perseus New York, NY, 2001.
2 D. Ribatti and A. Vacca, Leukemia, 2005, 19, 1525.
3 J. B. Bartlett, K. Dredge and A. G. Dalgleish, Nat. Rev. Cancer, 2004,
4, 314.
4 Celgene pharmaceutical company, http://www.celgene.com.
5 (a) S-C. Mei and R-T. Wu, Mol. Cancer Ther., 2008, 7, 2405; (b) H.
Sano, T. Noguchi, A. Tanatani, Y. Hashimoto and H. Miyachi, Bioorg.
Med. Chem., 2005, 13, 3079; (c) C. Meierhofer, S. Dunzendorfer and
C. J. Wiedermann, J. Infect. Dis., 1999, 180, 216.
6 (a) B. E. Turk, H. Jiang and J. O. Lu, Proc. Natl. Acad. Sci. U. S. A.,
1996, 93, 7552; (b) C. Meierhofer, S. Dunzendorfer and C. Wiedermann,
BioDrugs, 2001, 15, 681; (c) T. Paravar and D. J. Lee, International
24 (a) T. V. Updyke and G. L. Nicolson, J. Immunol. Methods, 1984, 73,
83; (b) J. Buckie and W. G. M. W. Cook, Anal. Biochem., 1986, 156,
463.
4062 | Org. Biomol. Chem., 2010, 8, 4059–4062
This journal is
The Royal Society of Chemistry 2010
©