Thalidomide derivatives for the treatment of neuroinflammation

Article abstract of DOI:10.1002/cmdc.201000326

The precise mechanism-of-action of thalidomide remains uncertain and might differ between diseases and under different clinical condition. With implications in the treatment of a variety of inflammatory and autoimmune diseases, as well as for use as an anticancer agent, alone or in combination with established therapeutics, it is clear that thalidomide and its derivatives deserve further scrutiny. In particular, thalidomide was shown to be effective in a mouse model of multiple sclerosis (MS), an autoimmune inflammatory disorder, called experimental autoimmune encephalomyelitis (EAE). Herein, we describe the synthesis and preliminary biological evaluation of new macromolecular prodrugs of thalidomide bearing an aminoalkyl group on the phthalimide ring. The effectiveness of these compounds to limit EAE was investigated, and it was shown that, at 100 mgkg^-1 thalidomide-equivalent dose, they abrogated the clinical and pathological features of EAE.

Full text of DOI:10.1002/cmdc.201000326

DOI: 10.1002/cmdc.201000326
Thalidomide Derivatives for the Treatment of
Neuroinflammation
Christiane Contino-Pꢀpin,*[a] Audrey Parat,[a] Cindy Patinote,[a] Wendi A. Roscoe,[b]
Stephen J. Karlik,[b] and Bernard Pucci*[a]
The precise mechanism-of-action of thalidomide remains un-
certain and might differ between diseases and under different
clinical condition. With implications in the treatment of a varie-
ty of inflammatory and autoimmune diseases, as well as for
use as an anticancer agent, alone or in combination with es-
tablished therapeutics, it is clear that thalidomide and its deriv-
atives deserve further scrutiny. In particular, thalidomide was
shown to be effective in a mouse model of multiple sclerosis
(MS), an autoimmune inflammatory disorder, called experimen-
tal autoimmune encephalomyelitis (EAE). Herein, we describe
the synthesis and preliminary biological evaluation of new
macromolecular prodrugs of thalidomide bearing an aminoalk-
yl group on the phthalimide ring. The effectiveness of these
compounds to limit EAE was investigated, and it was shown
that, at 100 mgkgꢀ1 thalidomide-equivalent dose, they abro-
gated the clinical and pathological features of EAE.
Introduction
Thalidomide first became known worldwide because of its
tragic teratogenic effects. Subsequently, it was re-evaluated for
use in the treatment of erythema leprosum nodosum (ENL).[1]
Recently, thalidomide has been investigated for its promising
therapeutic potential in various disorders.[2–4] Today, thalido-
mide is being evaluated for the treatment of a variety of in-
flammatory and autoimmune diseases,[5,6] and numerous hem-
atological and solid or nonsolid malignancies.[7,8] More than
twenty clinical trials include thalidomide in their regimen,
alone or in combination with other antineoplastic drugs.[9] In
2006, thalidomide and its structural analogue lenalidomide
(CC-5013, Revlimid) were approved by the Food and Drug Ad-
ministration (FDA, USA) for combination treatment with dexa-
methasone for relapsed or refractory multiple myeloma.[10,11]
The precise mechanism-of-action of thalidomide remains un-
certain and might differ between diseases and under different
clinical condition.[12] Among the numerous modes-of-action as-
cribed to thalidomide, it has been postulated that the effect of
the drug involves modulation of several cytokines associated
with neovascular growth, such as vascular endothelial growth
factor (VEGF), basic fibroblast growth factor (bFGF), hepatocyte
growth factor (HGF), as well as tumor necrosis factor-alpha
(TNF-a) and several interleukins.[4,5,13] Recent studies have pro-
posed that the anti-inflammatory and antiangiogenic effects of
thalidomide might be due to a blockade of nuclear factor-
kappa B (NF-kB) activation,[14] or an inhibitory effect on cyclo-
oxygenase enzymes (COX-1/2).[15,16] Thalidomide was able to
restore blood–brain barrier (BBB) permeability and provide
neuroprotection in amyloid b1ꢀ42 or quinolinic acid-injected
brain models.[17,18] In addition, D’Amato et al.[19] demonstrated
significant antiangiogenic activity of thalidomide on corneal
angiogenesis induced by VEGF.
occurs in both normal physiology and disease.[20] Although this
phenomenon is a critical event for the maintenance, prolifera-
tion and metastasis of tumors,[21–23] angiogenesis has been im-
plicated in the evolution of autoimmune inflammatory disor-
ders, such as rheumatoid arthritis and multiple sclerosis (MS).
Karlik and co-workers reported altered spinal cord vasculariza-
tion in the C57L6 mouse model for MS called experimental au-
toimmune encephalomyelitis (EAE).[24,25]
We have been engaged in the development and synthesis
of thalidomide analogues as potential therapeutic tools. For
angiogenesis inhibition, we designed macromolecular amphi-
pathic carriers endowed with multiple thalidomide units. These
compounds exhibited a significant ability to inhibit angiogene-
sis in a mouse model of corneal neovascularization when com-
pared to thalidomide.[26] We recently described a novel class of
thalidomide analogues substituted on the phthalimide ring
system. Among the several structural modifications on the
phthalimide ring, the introduction of either a carboxylic func-
tion or an N-(aminopropyl)amino group at the C4 position led
to a reduction of clinical signs of EAE.[27] These results suggest-
ed that angiogenesis could be an interesting target for inter-
fering with inflammatory MS-like pathophysiology. In the pres-
ent report, our strategy was to enhance the activity of N-(ami-
[a] Dr. C. Contino-Pꢀpin, Dr. A. Parat, C. Patinote, Prof. B. Pucci
Laboratoire de Chimie Bioorganique et des Systꢁmes Molꢀculaires Vector-
iels, Facultꢀ des sciences, Universitꢀ d’Avignon et des Pays de Vaucluse
33 rue Louis Pasteur, 84000 Avignon (France)
Fax: (+33)49014-4449
E-mail: christine.pepin@univ-avignon.fr
bernard.pucci@univ-avignon.fr
[b] Dr. W. A. Roscoe, Prof. S. J. Karlik
University of Western Ontario, Department of Medical Imaging, LHSC-UC
339 Windermere Road, London, Ontario, N6A 5A5 (Canada)
Angiogenesis, defined as the development of new blood
vessels from pre-existing vasculature, is a natural process that
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201000326.
ChemMedChem 2010, 5, 2057 – 2064
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2057

C. Contino-Pꢀpin, B. Pucci et al.
MED
nopropyl)-4-amino thalidomide by appending oligomeric mul-
tivalent carriers, called telomers.
For many years, we have studied the biomedical potential of
low-weight polymers derived from tris(hydroxymethyl)amino-
methane (Tris).[28] These compounds are obtained by free-radi-
cal polymerization of an acryloyl monomer derived from tris(h-
ydroxymethyl)acrylamidomethane (THAM) in the presence of
an alkane or a perfluoroalkanethiol as a transfer reagent called
telogen. The physicochemical parameters of these telomers
(molecular weight, hydrophilic–lipophilic balance, electric
charge) can be adjusted through both the starting material
and the experimental conditions.[29,30] With regard to their bio-
logical properties, we previously showed that these multifunc-
tional carriers are able to cross physiological membranes and
diffuse through all tissue types.[31] A whole-body autoradiogra-
phy performed on rat and mouse allowed us to confirm the
ubiquitous biodistribution of THAM-derived telomers in all bio-
logical compartments except the brain.[32] Moreover, after intra-
venous (iv) or oral (po) administration to rat, these compounds
were slowly released within 100 h, and no toxicity was ob-
served. Finally, the anchorage of various antitumor agents,
such as cytosine arabinoside (Ara-C) or 5-fluorouracil (5-Fu), to
the polymeric backbone gave macromolecular prodrugs with
improved bioavailability and a better therapeutic index than
the parent drugs.[28,33]
Scheme 1. Reagents and conditions: a) CF3CONH2, Et3N, HOBt, EDC·Cl, CH2Cl2,
08C!RT, 24 h, 95%; b) CF3CO2H/CH2Cl2 (7:3), 08C!RT, 1 h, 100%; c) 4-nitro-
phthalic anhydride, AcOH, molecular sieves (4 ꢂ), reflux, 48 h, 70%;
d) (Boc)2O, Et3N, dioxane, 08C!RT, o/n, 90%; e) AcOH, H2O, RT, 4 h, 98%;
f) H2, Pd/C, DMF/THF (9:1), RT, 16 h, 70%. EDC, 1-ethyl-3-(3-dimethylamino-
propyl)carbodiimide; HOBt, hydroxybenzotriazole.
Herein, we describe the synthesis of a series of telomers
with N-(aminopropyl)-4-amino thalidomide 1 appended. The
synthesis of
1
was recently reported and is show in
amino-3,3-diethoxypropane,[36] provided compound 6 in satis-
factory yield. Notably, compound 6 is a yellow fluorophore;
after removal of the N-Boc protecting groups, its maximum ex-
citation and emission in water are 276 nm and 552 nm, respec-
tively (data not shown). Thalidomide 6 was used as the starting
material for the subsequent attachment of each telomer.
Telogen N-((trisbenzyloxymethyl)-methyl)-3-mercapto-propio-
namide 9, was synthesized in four consecutive steps from 3-
mercaptopropionic acid (Scheme 2).[37] Co-telomerization reac-
tions (Scheme 3) between tris(acetoxymethyl)acrylamidome-
thane monomer 10 and g-amino propionic acid-derived acryl-
amide monomer 11,[31] in the presence of telogen 9 as a trans-
fer reagent and azobisisobutyronitrile (AIBN) as a radical initia-
tor, were carried out under an inert atmosphere in refluxing
THF. Telomerization was continued until complete consump-
tion of starting monomers (monitored by TLC). Telomers 12–14
Scheme 1.[27] According to the hydrophobic nature of thalido-
mide, and to the structure–activity relationships gained from
previous biological experiments, this thalidomide derivative
was conjugated to variable cotelomers of Tris end-capped with
a hydrophilic tail.
Results and Discussion
Chemistry
Synthesis of thalidomide analogue 6 was performed as previ-
ously reported (Scheme 1). Briefly, the preparation of nitro de-
rivative 3 started by cyclization of N-tert-butyloxycarbonyl-l-
glutamic acid in the presence of trifluoroacetamide following
the method of Galons et al.[34] Subsequent deprotection of the
resulting N-Boc glutarimide 2 with trifluoroacetic acid (TFA)
gave the corresponding TFA salt. Subsequent condensation of
this intermediate with 4-nitrophthalic anhydride by refluxing
overnight in glacial acetic acid[35] led to 4-nitro-substituted tha-
lidomide 3 in 70% yield. Finally, the reductive amination of 3
in the presence of 5, previously obtained in two steps from 1-
Scheme 2. Reagents and conditions: a) triphenylmethyl chloride (1.1 equiv),
CH2Cl2, RT, 16 h, 94%; b) Tris (1.1 equiv), EEDQ (1.2 equiv), EtOH, reflux, 12 h,
85%; c) C6H5COCl (4.5 equiv), Et3N (9 equiv), CH2Cl2, RT, 12 h, 82%; d) TFA/
CH2Cl2 (2:8), 08C!RT, o/n, 100%. EEDQ, 2-ethoxy-1-ethoxycarbonyl-1,2-dihy-
droquinoline; TFA, trifluoroacetic acid; Tris, tris(hydroxymethyl)aminome-
thane.
2058
www.chemmedchem.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 2057 – 2064

Treating Neuroinflammation
been reported in the literature
and ascribed to both instrument
and sample preparation parame-
ters used for the MALDI-MS anal-
ysis.[38] In such conditions, these
two techniques were ruled out
for telomer characterization.
The use of compound 9 as a
telogen agent in this synthetic
route is relevant for two reasons:
1) to allow facile NMR estimation
of the DPn by comparing the
total area of typical signals for
each monomer to well-identified
aromatic signals ascribed to the
telogen part; and 2) to support
the NMR estimation by UV cali-
bration, since in this case poly-
meric units do not interfere with
UV absorbance of telogen end-
caps on the polymer backbone.
1H NMR analysis permitted the
determination of x and y values
(Scheme 3) by comparing the
area of the peaks assigned to six
ortho-aromatic protons in the
telogen moiety (doublet; d=
8 ppm, integral 6H) to the signal
ascribed to methylene protons
of monomer 10 (singlet; d=
4.4 ppm, integral 6·x·H) and to
succinimidyl protons of 11 (sin-
glet; d=2.9 ppm, integral 4·y·H).
These 1H NMR data were sup-
ported by UV measurements
(see Supporting Information).
Scheme 3. Reagents and conditions: a) AIBN, THF, N2, reflux, 16 h, 56% for 12, 70% for 13 and 55% for 14; b) TFA,
CH2Cl2 (5:5), 08C!RT, 1 h, 100%; c) Et3N, THF, RT, N2 atmosphere, 4 days, 80% for 15, 83% for 16 and 99% for 17;
d) MeONa (cat.), MeOH, RT, o/n, 100%. AIBN, azobisisobutyronitrile; TFA, trifluoroacetic acid.
NMR determinations, although
suitable for short telomers,
become inaccurate as the size of
telomers increases. For this
reason, we used the UV absorb-
were prepared by using different ratio of monomers 10 and
11, and purified by size-exclusion chromatography. The degree
of polymerization (DPn) is defined as the average number of
repeating units in the polymeric backbone plus one (the telo-
gen moiety). The DPn depends on the ratio of telogen/mono-
mers, adjusted through both starting materials and experimen-
tal conditions.[29] The DPn value was determined by proton
NMR in combination with UV analysis.
ance of benzoyl groups from the telogen moiety in combina-
tion with the NMR data for more precise DPn estimation. The
three benzoyl groups appended on the Tris moiety of the telo-
gen agent strongly absorb at l=230 nm. A calibration curve
of the absorbance versus the molar concentration of telogen 9
was first established using standard solutions (0–70 mm) in a
1:1 mixture of dichloromethane and methanol. Triplicate sam-
ples of telomer were analyzed; in each case, a precise weight
of telomers 12–14 was dissolved in the solvent system, defin-
ing the weight concentration of telomer 12–14 in gLꢀ1 ([telo-
mer]W). This solution was diluted to a concentration suitable
for UV measurements. The value of the UV absorbance provid-
ed the molar concentration of the telogen 9 (molLꢀ1) in solu-
tion ([telogen]M). The average molecular weight of the co-
telomer (hMWi) was then calculated using Equation (1), with
errors within ꢁ3%:
Previously, we demonstrated that Tris-derived polymers
cannot be characterized using gel permeation chromatography
(GPC) measurements.[37] A study performed by matrix-assisted
laser desorption/ionization time-of-flight mass spectrometry
(MALDI-TOF-MS) also yielded inconsistent results (not pub-
lished), probably due to the variable fragmentation behavior of
the telomers during the MALDI experiments. This disparity be-
tween results observed for synthetic polymers has already
ChemMedChem 2010, 5, 2057 – 2064
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
2059

C. Contino-Pꢀpin, B. Pucci et al.
MED
W
M
ments. Control mice were also treated on the first day they
showed clinical signs with daily ip injections of sterile phos-
phate-buffered saline (PBS). The mice were sacrificed on the
eighth day, and the spinal cord was dissected, fixed in 10%
neutral-buffered formalin, embedded in paraffin wax, and sec-
tioned into 5 mm sections. Histological slides were prepared
using hematoxylin–eosin and solochrome cyanine R. Slides
were scored blindly according to a published pathological
scoring scale.[25] Differences between clinical and pathological
scores were determined by a Mann–Whitney U test for non-
parametric data, with p <0.05 considered to be significant.
In parallel to this study, the effect of free thalidomide on
EAE mice was investigated at two different doses (100 and
200 mgKgꢀ1 po). Since free thalidomide was effective at inhib-
iting angiogenesis and inflammation in other studies, it was
used here as a positive control.[39,40] Treatment also began on
the first day of clinical signs of the disease. Mice were also
treated for seven days and sacrificed on the eighth day. Spinal
cords were dissected and processed as described above.
ð1Þ
hMWitelomer ¼ ½telomerꢂ =½telogenꢂ
Since each co-telomer backbone contains a single telogen
moiety, the average molecular mass of the co-telomer can be
derived. The DPn of the co-telomer was determined by com-
1
bining H NMR and UV data through Equation (2):
hMWitelomer ¼ MWTAþ½ðx ꢃ MWTrisÞþðy ꢃ MWAA
Þ
ð2Þ
where MWTA is the molecular mass of the telogen agent 9,
MWTris and MWAA those of monomers 10 and 11, respectively,
and x+y is the total number of monomers in the backbone.
The average degree of polymerization (DPn) is then easily ob-
tained from Equation (3):
DPn ¼ ðxþyÞþ1
ð3Þ
At the same concentration used for the UV calibration
(~0.5 gLꢀ1), no absorbance was observed at l=230 nm for a
co-telomer devoid of phenyl groups. No degradation of the
benzoyl groups was observed by proton NMR when the pure
telogen agent was reacted with AIBN in refluxing THF. There-
fore, the ester bonds of the telogen agent are stable during te-
lomerization. We combined the x/y ratio from the proton NMR
data with this to determine the DPn of the telomers. At low
DPn, the molecular weights obtained by proton NMR alone
were found to be in good agreement with that obtained by
the combination of NMR and UV data.
Mice immunized with MOG35–55 to induce EAE began show-
ing clinical signs 10–15 days post-immunization. They were
randomized into PBS control or treatment groups once they
showed clinical signs of illness and were then treated for
seven days. The mean clinical scores for compounds 18 and 19
are illustrated in Figure 1a (1 mgkgꢀ1
)
and Figure 1b
(100 mgkgꢀ1). The mean clinical scores for free thalidomide
given orally at 100 mgkgꢀ1 or 200 mgkgꢀ1 are shown in Fig-
ure 2a. Pathological evaluations for both experiments are
shown in Figure 1c and Figure 2b.
Subsequent removal of the protecting group of thalidomide
amino derivative 6 with TFA and condensation to telomers 12–
14 produced telomers 15–17. Final treatment under Zemplꢀn
conditions yielded the desired telomers 18–20 in satisfactory
yields; NMR analysis confirmed the total disappearance of aro-
matic groups of telogen.
At enrolment, all mice were at a clinical score of at least 1.
During the experimental period, the vehicle controls showed
an increase in clinical signs of disease, which decreased partial-
ly by the end of the observation period. Compound 18 yielded
decreased disease even when given at a thalidomide-equiva-
lent dose of 1 mgkgꢀ1 (p <0.05). The disease was very mild at
100 mgkgꢀ1 thalidomide-equivalent dose of compounds 18
and 19 (p <0.001 for 18; p <0.026 for 19), and some mice
achieved a score of 0, or complete reversal of all clinical signs.
Figure 1c shows the scores for the blinded pathological assess-
ments. The mean scores for inflammation and demyelination
for all treatments were somewhat lower than the saline con-
trols and reached statistical significance for compound 19
when dosed at 100 mgkgꢀ1. The combination of decreased
clinical score and pathological score for compound 19 is con-
sistent with anti-inflammatory activity. As a consequence, the
decrease in demyelination score is an indication that myelin
damage was decreased, resulting from reduced inflammation
compared to controls.
Multiple sclerosis animal model
To induce experimental autoimmune encephalomyelitis (EAE),
C57L6 mice were immunized with MOG35–55 peptide. Each im-
munization contained a mixture of MOG peptide in complete
Freund’s adjuvant (CFA) supplemented with Myobacterium tu-
berculosis. Mice were given intraperitoneal (ip) injections of
Bordella pertussis toxin on the day of immunization, and on
day two post-immunization.
The clinical signs of EAE include a limp tail, and affected or
paralyzed limbs. The clinical scoring scale was: 0.0: healthy; 1:
limp tail; 1.5: poor righting reflex but tail is no longer limp (re-
covery only); 2.0: poor righting reflex with limp tail; 3.0: above
signs as well as one hind limb affected; 3.5: above signs as
well as one hind limb paralyzed; 4.0: above signs as well as
both hind limbs affected; 4.5: above signs as well as both hind
limbs paralyzed; 5.0: moribund. On the first day of clinical
signs (clinical score=1), the mice were treated with com-
pounds 18 or 19 (100 mgkgꢀ1 or 1 mgkgꢀ1 thalidomide-equiv-
alent dose) via ip injections, and treatments were continued
daily for seven days. Compound 20, with a low number of tha-
lidomide pendants, was not tested in these preliminary experi-
In a previous experiment, 100 mgkgꢀ1 of free thalidomide
was ineffective in this model when administrated by ip injec-
tion.[27] Although thalidomide administered orally was effective
in decreasing clinical and pathological signs of EAE, it was not
as effective as compounds 18 and 19. Therefore, we have ach-
ieved disease reversal with these new compounds at lower
doses.
However, on the basis of the ability of telomers to reach the
bloodstream after oral administration,[32] and considering that
2060
www.chemmedchem.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 2057 – 2064

Treating Neuroinflammation
Figure 2. a) Mean clinical scores and b) pathological scores for EAE mice
treated orally with vehicle (*; &), or free thalidomide (100 mgkgꢀ1: *; &;
200 mgkgꢀ1: !; &). Mice treated with 100 mgkgꢀ1 of free thalidomide
showed fewer clinical signs of EAE. Mice treated with 200 mgkgꢀ1 of free
thalidomide showed fewer pathological signs of EAE ( p <0.05). Error bars
*
indicate the standard error of the mean.
availability due to structural modification of the parent thalido-
mide, and/or multiple appending of the active component on
the same carrier. Considering the hydrophobic character of
thalidomide, conjugation to a hydrophilic moiety might be ex-
pected to enhance activity. Thus, N-(aminopropyl)-4-amino tha-
lidomide appears to be a promising thalidomide derivative
when appended to a hydrophilic macromolecular carrier, com-
pared to the parent drug. In addition to a possible synergistic
effect gained from multiplicity, the efficacy against EAE might
arise through enhanced permeability and retention at sites of
BBB impairment.[41]
Figure 1. a) and b) Mean clinical scores and b) pathological scores for EAE
mice treated with vehicle (*), or compounds 18 (*) and 19 (!) at
1 mgkgꢀ1 (panel a) and 100 mgkgꢀ1 (panel b); ( p <0.05). Error bars indi-
*
cate the standard error of the mean.
N-(aminopropyl)amino-thalidomide telomers 18 and 19 ach-
ieved substantial success in the reversal of clinical signs of
acute EAE in mice. All animals exhibited definite clinical signs
on entry into the study, so this was a treatment study and not
a prevention study. Our findings can be compared to recent re-
sults obtained by McLarnon et al., who demonstrated thalido-
mide-mediated neuroprotection and reduced vascular remod-
eling in an excitotoxin-injected brain model.[18] We expect a
similar effect with the use of telomers endowed with N-(ami-
free thalidomide is more active when given orally, further in-
vestigation of N-(aminopropyl)amino-thalidomide telomers is
necessary to specify both the best dose and mode of adminis-
tration in the EAE mouse model of MS.
Two factors could account for increased activity with com-
pounds 18 and 19 after ip injection: enhanced general bio-
ChemMedChem 2010, 5, 2057 – 2064
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
2061

C. Contino-Pꢀpin, B. Pucci et al.
MED
(d, J=8.65 Hz, 1H), 4.22 (m, 1H), 2.69 (m, 1H), 2.47 (m, 2H), 1.92
(m, 1H), 1.39 ppm (s, 9H); 13C NMR (63 MHz, [D6]DMSO): d=24.5,
28.2, 31.0, 50.4, 78.2, 172.5, 173.0 ppm.
nopropyl)amino-thalidomide in the EAE model, particularly as
VEGF-induced vascular changes have been observed by inhibi-
tion experiments using SU5416 (Semaxinib).[25] To ascertain
whether the mechanism of action is that of antiangiogenesis,
further investigation is required, in particular to identify the
molecular targets of these agents. The optical properties of
these new fluorescent derivatives will permit future histological
assays.
2’-(4-Ntrophthaloyl)glutarimide (3): Compound
2
(1.7 g,
7.4 mmol) was treated with a mixture of TFA and CH2Cl2 (7:3;
10 mL) and stirred at RT for 1 h. The corresponding TFA salt pre-
cipitated from the reaction mixture, and excess TFA was removed
in vacuo (quantitative yield). The TFA salt (1.8 g) was added to a
solution of 4-nitrophthalic anhydride (1.62 g, 8.41 mmol) and mo-
lecular sieves (4 ꢂ) in acetic acid (25 mL). The mixture was heated
at reflux for 2 d, and then filtered through Celite and washed with
EtOAc. The organic solvents were removed in vacuo, and the crude
compound was purified by flash chromatography (cyclohexane/
EtOAc; 20!0%). Recrystallization from EtOAc/n-hexane gave 3 as
a white powder (1.57 g, 70%): Rf =0.49 (EtOAc/cyclohexane; 6:4);
mp: 226.8–227.88C; [a]D20 =+1.47 (c=1, DMF); 1H NMR (250 MHz,
[D6]DMSO): d=11.20 (s, 1H), 8.68 (d, J=8.1 Hz, 1H), 8.57 (s, 1H),
8.20 (d, J=8.1 Hz, 1H), 5.24 (dd, J=5.2, 12.7 Hz, 1H), 2.89 (m, 1H),
2.50 (m, 2H), 2.13 ppm (m, 1H); 13C NMR (63 MHz, [D6]DMSO): d=
22.3, 31.3, 49.9, 118.9, 125.5, 130.6, 133.0, 136.2, 152.2, 165.7, 166.0,
173.2 ppm; HRMS-ESI+: m/z [M+H]+ calcd for C13H9N3O6:
304.0564, found: 304.0565.
Conclusions
The aim of this work was to synthesize and evaluate thalido-
mide analogues as agents against neuroinflammation. Previ-
ously, we demonstrated the therapeutic potency of thalido-
mide derivatives in angiogenesis inhibition[26] and EAE.[27] Al-
though we cannot yet provide details on the precise mecha-
nism of action of thalidomide and its derivatives, herein we
have shown enhanced activity from molecular modifications of
thalidomide and proposed new therapeutic applications for
this “re-discovered” old drug.
2’-(4-(N-(N-tert-Butyloxycarbonyl)propylamino)phthaloyl)glutari-
mide (6): 3-N-(tert-butyloxycarbonyl)aminopropanal (0.342 g,
1.98 mmol), obtained from 1-amino-3,3-diethoxypropane following
the method of Bi et al.,[34] was dissolved in anhydrous THF/DMF
(9:1; 10 mL). 2’-(4-Nitrophthaloyl)glutarimide (0.5 g, 1.65 mmol) and
catalytic Pd/C (10% in weight) were then added to the solution.
The mixture was stirred at RT under H2 (8 bars) for 16 h. After con-
sumption of the starting material, the reaction was filtered through
Celite and concentrated in vacuo. Purification by flash chromatog-
raphy (EtOAc/cyclohexane; 3:7) gave 6 as a yellow powder (0.71 g,
70%): Rf =0.71 (EtOAc/cyclohexane; 8:2); mp: 89.38C; [a]2D0 =+5.7
Experimental Section
Chemistry
Reactions and the homogeneity of the compounds were moni-
tored by thin layer chromatography (TLC; Merck F254 silica plates)
visualized under UV light (l=254 and 366 nm) or by spraying with
a 5% H2SO4 solution in EtOH, and/or a 5% ninhydrin solution in
EtOH followed by heating at ~1508C to detect characteristic
groups. Flash chromatography was carried out on Merck silica gel
Gerduran Si 60 (40–63 mm). Size-exclusion chromatography was
carried out using Sephadex LH-20 resin or Sephadex G-25 resin (GE
Healthcare Bio-Sciences AB). Final products were purified by high-
pressure liquid chromatography (HPLC) on a Varian Pro Star Micro-
sorb C18 column (5 mm granulometry, 4.6ꢃ250 mm or 21.4ꢃ
250 mm). Melting points were determined in open capillary tubes
1
(c=1, CH2Cl2); H NMR (250 MHz, CDCl3): d=8.14 (s, 1H), 7.61 (d,
J=8.3 Hz, 1H), 7.02 (s, 1H), 6.79 (d, J=8.7 Hz, 1H), 5.49 (m, 1H),
4.95 (dd, J=4.8, 11.8 Hz, 1H), 4.71 (m, 2H), 3.23–3.33 (m, 4H),
2.78–2.98 (m, 3H), 2.21 (m, 1H), 1.79 (qt, J=6.2 Hz, 2H), 1.48 ppm
(s, 9H); 13C NMR (63 MHz, CDCl3): d=22.8, 28.4, 29.1, 37.5, 40.1,
49.0, 79.8, 105.9, 116.6, 117.8, 125.5, 134.6, 153.7, 156.6, 167.5,
168.0, 168.9, 171.4 ppm; HRMS-ESI+: m/z [M+H]+ calcd for
C21H26N4O6: 431.1925, found: 431.1925.
1
and are uncorrected. H, 13C and DEPT NMR spectra were recorded
on a Bruker AC-250 spectrometer (1H, 250 MHz; 13C, 62.86 MHz).
Chemical shifts (d) are given in ppm relative to the solvent residual
peak as a heteronuclear reference. Abbreviations used for signal
patterns are: s (singlet), d (doublet), t (triplet), q (quartet), m (mul-
tiplet). UV-Vis spectra were recorded on a Cary Win Varian spectro-
photometer with a double-compartment quartz (Suprasil) cell
(length=10 mm). High resolution mass spectrometry (HRMS) was
recorded on a QStar Elite (Applied Biosystems SCIEX) spectrometer
equipped with an atmospheric pressure ionization (API) source
(ESI+, ToF). HRMS data was collected by the Spectropole, Facultꢀ
des Sciences et Techniques de Saint-Jꢀrꢄme (Marseille, France).
Co-telomers bearing peracetylated Tris and succinimidyl moiet-
ies (12–14): Monomers 10 and 11 (see Table 1 for ratios) were dis-
solved in anhydrous THF under N2. Appropriate amounts of telo-
gen 9 and radical initiator AIBN were added, and the mixture was
stirred at reflux under N2 until complete consumption of mono-
mers (~12–24 h). The mixture was concentrated in vacuo and puri-
fied by size-exclusion chromatography (CH2Cl2/MeOH; 1:1) to give
co-telomers 12–14 as white powders (56% for 12, 70% for 13, and
55% for 14). Their DPn values (see Table 2) were calculated using
1H NMR data and confirmed by UV measurements (see Supporting
Information).
2’-N-(tert-Butyloxycarbonyl)aminoglutarimide (2): Trifluoroaceta-
mide (2.7 g, 24.2 mmol) was added to a solution of N-tert-(butylox-
ycarbonyl)glutamic acid (5 g, 20.2 mmol), HOBt (6 g, 44.0 mmol),
and Et3N (8.5 mL, 60.5 mmol) in CH2Cl2 (50 mL) at 08C. EDC·Cl
(8.14 g, 42.5 mmol) was added, and the mixture was allowed to
reach RT. After consumption of glutamic acid (24 h), water (30 mL)
was added. The organic layer was washed with saturated aq
Na2CO3 (2ꢃ100 mL) and brine (2ꢃ100 mL), and dried (Na2SO4), fil-
tered and concentrated in vacuo. Crystallization (EtOAc) gave 2 as
a white powder (4.4 g, 95%), no further purification was necessary
for the next step: Rf =0.67 (EtOAc); mp: 193.7–194.48C; [a]2D0 =ꢀ63
Co-telomers bearing peracetylated Tris and N-(aminopropyl)-4-
aminothalidomide moieties (15–17): Co-telomer (12–14) was dis-
solved in anhydrous THF. Deprotected compound 6 (deprotected
using TFA/CH2Cl2 (3:7); see HRMS data in the Supporting Informa-
tion) was added to the solution in excess (relative to active ester
groups; 1.3 equiv per succinimidyl moiety). Et3N was used to
adjust and maintain pH 8. The mixture was stirred at RT for 4 d
under N2 and protected from light. Concentration in vacuo and pu-
rification by size-exclusion chromatography (MeOH/CH2Cl2, 5:5)
1
(c=1, DMF); H NMR (250 MHz, [D6]DMSO): d=10.70 (s, 1H), 7.14
2062
www.chemmedchem.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 2057 – 2064

Treating Neuroinflammation
tussis toxin (List Biological Laboratories, Cedarlane; Hornby,
Canada) on the day of immunization and on day two post-immuni-
zation. Mice were housed in a light- and temperature-controlled
environment, with a 12 h day/night cycle at 21–238C, and received
food and water ad libitium.
Table 1. Initial conditions for the synthesis of co-telomers 12–14.
Co-telomer
Molar ratio (amount [mg])
10 11
9
AIBN
(15.7)
(31.4)
(15.7)
12
13
14
1
1
1
(100)
(200)
(100)
15
45
15
(865)
(5190)
(865)
5
15
5
(243)
(1460)
(243)
0.5
0.5
0.5
On the first day of clinical symptoms, mice were treated with com-
pounds 18 and 19 at 100 or 1 mgkgꢀ1 thalidomide-equivalent
dose (100 mgkgꢀ1: 18, n=7; 19, n=6; 1 mgkgꢀ1: 18, n=6; 19, n=
6). Treatments were carried out daily for seven days. Control mice
(n=5 and n=4 for the two dose groups) were also treated on the
first day they showed clinical signs with daily ip injections of
0.1 mL sterile PBS. The mice were sacrificed on the eighth day,
their spinal cord was dissected, fixed in 10% neutral-buffered for-
malin, embedded in paraffin wax, and sectioned into 5 mm sec-
tions. Histological slides were prepared using hematoxylin-eosin
and solochrome R cyanin. Slides were scored blindly according to
a published pathological scoring scale.[25]
Table 2. Experimental data for co-telomers 12–14.
Co-telomer
Structure[a]
DPn[b]
Yield
[%]
Mw
x
y
[gmolꢀ1
]
12
13
14
23
40
12
8
15
4
32
56
17
56
70
55
9476
16371
5149
[a] Average number of peracetylated Tris (x) and thalidomide derivative
(y) monomer units. [b] DPn: degree of polymerization. The number of
monomer units (x and y) in the polymer backbone + 1 (DPn=x+y+1).
As a positive control, free thalidomide was administered to EAE
mice at 100 (n=8) and 200 mgKgꢀ (n=9); a control using vehicle
was also conducted (n=8). Treatment began on the first day of
clinical signs of disease. Mice were treated for seven days and sac-
rificed on the eighth day. Spinal cords were dissected and pro-
cessed as described above.
gave co-telomers 15–17 as yellow powders (80% for 15, 83% for
16, and 99% for 17). At this point, the coupling ratio of N-(amino-
propyl)-4-aminothalidomide was specified as described in the main
body of the manuscript. Notably, for each telomer, NMR structural
evaluation showed that the proportion of thalidomide moieties
was similar to the succinimidyl groups in the precursor telomers
12–14 (see Table 3).
All animals used in this study were maintained and handled in ac-
cordance with the International Guiding Principles for Biomedical
Research Involving Animals and using protocols approved by the
Animal Use Subcommittee of the University of Western Ontario
(Canada).
Keywords: angiogenesis
neuroinflammation · telomers · thalidomide
·
multiple
sclerosis
·
Table 3. Experimental data for co-telomers 15–17.
Co-telomer
Structure[a]
DPn[b]
Yield
[%]
Mw
[gmolꢀ1
]
x
y
[1] S. K. Teo, K. E. Resztak, M. A. Scheffler, K. A. Kook, J. B. Zeldis, D. I. Stir-
ling, S. D. Thomas, Microbes Infect. 2002, 4, 1193–1202.
[2] J. J. Wu, D. B. Huang, K. R. Pang, S. Hsu, S. K. Tyring, Br. J. Dermatol.
2005, 153, 254–273.
[3] S. M. Capitosti, T. P. Hansen, M. L. Brown, Bioorg. Med. Chem. 2004, 12,
327–336.
[4] L. RosiÇol, M. T. Cibeira, M. Segarra, M. C. Cid, X. Filella, M. Aymerich, M.
Rozman, L. Arenillas, J. Esteve, J. Bladꢀ, E. Montserrat, Cytokine 2004, 26,
145–148.
15
16
17
23
40
12
8
15
4
32
56
17
80
83
99
11196
19596
6009
[a] Average number of peracetylated Tris (x) and thalidomide derivative
(y) monomer units. [b] DPn: degree of polymerization. The number of
monomer units (x and y) in the polymer backbone + 1 (DPn=x+y+1).
[5] A. Thiele, R. Bang, M. Gꢅtschow, M. Rossol, S. Loos, K. Eger, G. Tiegs, S.
Hauschildt, Eur. J. Pharmacol. 2002, 453, 325–334.
[6] H. W. Man, L. G. Corral, D. I. Stirling, G. W. Muller, Bioorg. Med. Chem.
Lett. 2003, 13, 3415–3417.
Co-telomers bearing Tris and N-(aminopropyl)-4-aminothalido-
mide moieties (18–20): Co-telomer (15–17) was dissolved in
MeOH under N2 and treated with catalytic NaOMe (0.5 mol% per
telomer). After stirring at RT for 12 h, IRC-50 resin was added to
neutralize the solution. The mixture was shaken for 15 min, filtered
and concentrated in vacuo, and purified by size-exclusion chroma-
tography on a Sephadex G-25 gel (H2O). The appropriate fractions
were lyophilized to give co-telomers 18–20 as yellow powders
(quantitative yield).
[7] P. Musto, Leukemia 2004, 18, 325–332.
[8] S. R. Porter, J. Jorge Jr., Oral Oncol. 2002, 38, 527–531.
[9] S. Sleijfer, W. H. J. Kruit, G. Stoter, Eur. J. Cancer 2004, 40, 2377–2382.
[10] M. Melchert, A. List, Int. J. Biochem. Cell Biol. 2007, 39, 1489–1499.
[11] C. Hulin, La revue de mꢀdecine interne 2007, 28, 682–688.
[12] T. D. Stephens, C. J. W. Brunde, B. J. Fillmore, Biochem. Pharmacol. 2000,
59, 1489–1499.
[13] F. E. Kruse, A. M. Joussen, K. Rohrschneider, M. D. Becker, H. E. Vçlcker,
Graefe’s Arch. Clin. Exp. Ophthalmol. 1998, 236, 461–466.
[14] J. A. Keifer, D. C. Guttridge, B. P. Ashburner, A. S. Baldwin Jr., J. Biol.
Chem. 2001, 276, 22382–22387.
[15] T. Noguchi, R. Shimazawa, K. Nagasawa, Y. Hashimoto, Bioorg. Med.
Chem. Lett. 2002, 12, 1043–1046.
[16] G. J. Du, H. H. Lin, Q. T. Xu, M. W. Wang, Aesculape Vascular Pharm.
2005, 43, 112–119.
[17] J. K. Ryu, J. G. McLarnon, Neurobiol. Dis. 2008, 29, 254–266.
[18] J. K. Ryu, N. Jantaratnotai, J. G. McLarnon, Neuroscience 2009, 163, 601–
608.
Multiple Sclerosis Animal Model
Six-week-old female C57L6 mice (Jackson Laboratories, Bar Harbor,
ME) were immunized by subcutaneous injection in the hind flank
with 200 mg of MOG35–55 peptide (Sheldon Biochemical; Montreal,
Canada). Each immunization contained a 1 mgmLꢀ mixture of
100 mg of MOG peptide in CFA (Difco Laboratories; Michigan,
USA) supplemented with 400 ng of M. tuberculosis (Difco Laborato-
ries; Michigan, USA). Mice were given 200 ng ip injections of B. per-
ChemMedChem 2010, 5, 2057 – 2064
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
2063

C. Contino-Pꢀpin, B. Pucci et al.
MED
[19] R. J. D’Amato, M. S. Loughnan, E. Flynn, J. Folkman, Proc. Natl. Acad. Sci.
USA 1994, 91, 4082–4085.
[33] S. Jasseron, C. Contino-Pꢀpin, J. C. Maurizis, M. Ollier, B. Pucci, Eur. J.
Med. Chem. 2003, 38, 825–836.
[20] S. Liekens, E. De Clercq, J. Neyts, Biochem. Pharmacol. 2001, 61, 253–
270.
[21] J. Folkman, New Engl. J. Med. 1971, 285, 1182–1186.
[22] F. A. Shepherd, S. S. Sridhar, Lung Cancer 2003, 41 (Suppl. 1), 63–72.
[23] J. Folkman, J. Natl. Canc. Inst. 1990, 82, 4–6.
[24] S Kirk, J. A. Frank, S. Karlik, J. Neurol. Sci. 2004, 217, 125–130.
[25] W. A. Roscoe, M. E. Welsh, D. E. Carter, S. J. Karlik, J. Neuroimmunol.
2009, 209, 6–15.
[26] S. Pꢀrino, C. Contino-Pꢀpin, R. Satchi-Fainaro, C. Butterfield, B. Pucci,
Bioorg. Med. Chem. Lett. 2004, 14, 421–425.
[27] C. Contino-Pꢀpin, A. Parat, S. Pꢀrino, C. Lenoir, M. Vidal, H. Galons, S.
Karlik, B. Pucci, Bioorg. Med. Chem. Lett. 2009, 19, 878–881.
[28] C. Contino-Pꢀpin, J. C. Maurizis, B. Pucci, Curr. Med. Chem. 2002, 9, 645–
665.
[29] B. Pucci, E. Guy, F. Vial-Reveillon, A. A. Pavia, Eur. Polym. J. 1988, 24,
1077–1086.
[34] N. Flaih, C. Pham-Huy, H. Galons, Tetrahedron Lett. 1999, 40, 3697–3698.
[35] G. W. Muller, R. Chen, S. Y. Huang, L. G. Corral, L. M. Wong, R. T. Patter-
son, Y. Chen, G. Kaplan, D. I. Stirling, Bioorg. Med. Chem. Lett. 1999, 9,
1625–1630.
[36] W. Bi, J. Cai, S. Liu, M. Baudy-Floc’h, L. Bi, Bioorg. Med. Chem. 2007, 15,
6909–6919.
[37] K. S. Sharma, G. Durand, F. Giusti, B. Olivier, A. S. Fabiano, P. Bazzacco, T.
Dahmane, C. Ebel, J. L. Popot, B. Pucci, Langmuir 2008, 24, 13581–
13590.
[38] S. J. Wetzel, C. M. Guttman, J. E. Girard, Int. J. Mass Spectrom. 2004, 238,
215–225.
[39] T. Asano, H. Kume, F. Taki, S. Ito, Y. Hasegawa, Biol. Pharm. Bull. 2010,
33, 1028–1032.
[40] T A. Faivre-Chauvet, P. Thomare, G. Deleris, K. Estieu-Gionnet, A. Bikfalvi,
J. Barbet, F. Paris, J. Nucl. Med. 2010, 51, 624–631.
[41] A. V. Kabanov, H. E. Gendelman, Prog. Polym. Sci. 2007, 32, 1054–1082.
[30] B. Pucci, J. C. Maurizis, A. A. Pavia, Eur. Polym. J. 1991, 27, 1101–1106.
[31] F. Chehade, J. C. Maurizis, B. Pucci, A. A. Pavia, M. Ollier, A. Veyre, F.
Escaig, C. Jeanguillaume, R. Dennebouy, G. Slodzian, E. Hindie, Cell. Mol.
Biol. 1996, 42, 335–342.
Received: August 3, 2010
Revised: September 13, 2010
Published online on October 8, 2010
[32] J. C. Maurizis, M. Azim, M. Rapp, B. Pucci, A. A. Pavia, J. C. Madelmont,
A. Veyre, Xenobiotica 1994, 24, 535–541.
2064
www.chemmedchem.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 2057 – 2064