Structural modifications of thalidomide produce analogs with enhanced tumor necrosis factor inhibitory activity

Full text of DOI:10.1021/jm9603328

3238
J . Med. Chem. 1996, 39, 3238-3240
Sch em e 1a
Str u ctu r a l Mod ifica tion s of Th a lid om id e
P r od u ce An a logs w ith En h a n ced Tu m or
Necr osis F a ctor In h ibitor y Activity1
George W. Muller,*,† Laura G. Corral,†
Mary G. Shire,† Hua Wang,† Andre Moreira,‡
Gilla Kaplan,‡ and David I. Stirling†
Celgene Corporation, Warren, New J ersey 07059, and
Department of Cellular Immunology, The Rockefeller
University, New York, New York 10021
a
Received May 3, 1996
Reagents: (a) NH4OAc, CH2(CO2H), EtOH, reflux; (b) N-
carbethoxyphthalimide, Na2CO3, CH3CN/H2O; (c) (1) CDI/THF, (2)
concentrated NH4OH.
In tr od u ction . Thalidomide (1) was developed ini-
tially as a “safe” sedative without the toxicity and
addictive potential of barbiturates. The molecule, a
synthetic derivative of glutamic acid, consists of a glu-
tarimide ring and a phthaloyl ring. At physiologic
pH, hydrolysis of thalidomide occurs at both the phthal-
imido and glutarimide rings.2 It has therefore been
postulated that thalidomide may act as a prodrug for
one of its hydrolysis products or metabolites.
or N-phthaloylisoglutamine. Analogs were therefore
prepared based on the hydrolysis of the glutarimide ring
of thalidomide.
Ch em istr y. Phthalimido â-amino amide derivatives
were prepared as shown in Scheme 1. The â-amino
â-aryl acids were prepared as previously described12 by
the treatment of aryl aldehydes with ammonium acetate
and malonic acid in refluxing ethanol. The N-phthaloyl
group was introduced by treating the â-amino acid with
N-carbethoxyphthalimide in the presence of sodium
carbonate.13 Phthalimido amides were prepared by CDI
activation of the carboxylic acid in THF at ambient
temperature followed by treatment with excess am-
monium hydroxide. The N-methyl amide derivative was
prepared by a similar procedure using aqueous meth-
ylamine in place of ammonium hydroxide. Ester de-
rivative 21 was prepared by direct condensation of
methyl 3-amino-3-(3′,4′-dimethoxyphenyl)propionate hy-
drochloride with N-carbethoxyphthalimide in the pres-
ence of sodium carbonate. Substituted phthalimido
groups were introduced by condensation of a substituted
phthalic anhydride with methyl 3-amino-3-(3′,4′-dimeth-
oxyphenyl)propionate hydrochloride in acetic acid or in
acetic acid in the presence of sodium acetate. The
aminophthaloyl-substituted compounds 23 and 25 were
prepared by hydrogenation of the corresponding nitro
compounds.
The enantiomers of methyl 3-amino-3-(3′,4′-dimeth-
oxyphenyl)propionate were prepared by the method of
Davies.14 The phthaloyl group was then introduced as
described above. Chiral purity was determined by
chiral HPLC analysis on a Daicel Crown-Pak R(+)
column, and both isomers were found to be optically
pure (>95% ee).
Resu lts a n d Discu ssion . Inhibition of TNFR was
measured in the supernatant of human PBMCs stimu-
lated with LPS.15 In this assay, thalidomide has an IC50
of 194 µM for inhibition of TNFR synthesis with a dose-
response curve which maximizes at 60-70% inhibition
of TNFR synthesis.16 Simple phthalimidoalkyl amides
which might mimic glutarimide hydrolysis products of
thalidomide were evaluated for their ability to inhibit
TNFR. Phthalimidoalkyl amides (2-4) derived from
glycine, â-alanine, and γ-aminobutyric acid, respec-
tively, were prepared and found to have minimal TNFR
inhibition activity (0% inhibition at 980 µM, 22%
inhibition at 458 µM, and 23% inhibition at 403 µM,
respectively). In exploring the effects of substitution of
simple phthalimidoalkyl amides, the phthalimido amide
of 3-phenylpropionic acid, 5, was prepared and found
to be nearly equipotent to thalidomide (Table 1).
After a few years of use as a sedative, thalidomide
was withdrawn from the market when its potent and
tragic teratogenic properties became apparent.3 How-
ever, during this time, it was noted that thalidomide
was remarkably effective for the treatment of erythema
nodusum leprosum (ENL), an acute inflammatory mani-
festation of lepromatous leprosy.4,5 More recently, thal-
idomide was found to exert immunomodulatory and
anti-inflammatory effects in a variety of disease states.
These include graft-versus-host disease following bone
marrow transplantation, rheumatoid arthritis, inflam-
matory bowel disease (IBD), cachexia in AIDS, and
opportunic infections in AIDS.6 In studies to define the
physiological targets of thalidomide, the drug was found
to have a wide variety of biological activities exclusive
of its sedative effect including neurotoxicity,7 teratoge-
nicity,8 suppression of tumor necrosis factor-R (TNFR)
production by monocytes/macrophages9 and the ac-
companying inflammatory toxicities associated with
high levels of TNFR,10 and inhibition of angiogenesis
and neovascularization.11 It is as yet unclear what the
mechanism of action of the drug is and whether all the
biological activities are mediated through the same
pathway.
Because of the demonstrated effect of thalidomide on
the production of TNFR, and the central role that TNFR
plays in the immune response and the inflammatory
cascade, we decided to focus on improving the TNFR-
inhibiting properties of thalidomide by structure modi-
fication. We have designed and synthesized analogs of
thalidomide optimized for their ability to control TNFR
synthesis. Initial studies demonstrated the importance
of an intact phthaloyl ring. Hydrolysis of the glutarim-
ide ring affords either
N-phthaloylglutamine
† Celgene Corp.
‡ The Rockefeller University.
S0022-2623(96)00332-9 CCC: $12.00 © 1996 American Chemical Society

Communications to the Editor
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 17 3239
Ta ble 1. TNFR Inhibition by N-Phthaloyl â-Amino â-Aryl
Amides
Ta ble 3. TNFR Inhibition by Ring-Substituted Phthaloyl
Analogs
compd
X1
X2
X3
X4
IC50 (µM) TNFRa
compd
X1
X2
X3
IC50 (µM) TNFRa
22
23
24
25
26
27
28
29
30
H
H
H
H
H
H
H
Cl
H
H
H
H
H
H
NO2
NH2
H
H
H
H
H
NO2
NH2
OH
H
H
Cl
H
34
0.38
64
0.45
15
4.7
13
5
6
7
8
9
10
11
12
13
14
15
H
H
H
H
H
H
H
H
H
H
H
H
CN
H
CN
H
OCH3
H
OCH3
OEt
OPr
Cl
260
89
150
120
62
12
5.6
55
H
OCH3
OCH3
OEt
OPr
Cl
Ph
Cl
H
H
Cl
H
tBu
7.9
4.2
70
51
39
a
Ph
IC50 for TNFR inhibition in LPS-stimulated human PBMCs.
OCH3
H
OCH3
a
IC50 for TNFR inhibition in LPS-stimulated human PBMCs.
than the N-methyl analog 16 resulted in decreased
activity. The free carboxylic acid 19 was found to be
5-fold less active. The primary alcohol 20 afforded a
slight increase in activity. Replacement of the amide
moiety with a carboxymethyl group (compound 21)
produced a near 5-fold increase in activity.
Ta ble 2. TNFR Inhibition by Amide Isosteres
To improve the activity of 21, the effect of substitution
of the aromatic ring of the phthaloyl ring was explored
(Table 3). Dihalo substitution of the aromatic ring (28
and 29) decreased activity. Alkyl substitution had only
a minor effect. Nitro group substitution in the 3- or
4-position decreased activity by more than 1 order of
magnitude. However, 3- or 4-amino substitution (23
and 25, respectively) yielded compounds with submi-
cromolar IC50’s.
compd
R
IC50 (µM) TNFRa
16
17
18
19
20
21
CONHCH3
CONHCH2CH3
CONHBn
CO2H
CH2OH
CO2CH3
12
53
84
60
7.5
2.9
All the analogs described above were prepared and
tested as racemic mixtures. Thalidomide has always
been used clinically as a racemic mixture. The S-isomer
is generally considered to be teratogenic with the
R-isomer being a nonteratogenic sedative; however, this
view is controversial.17 Eriksson and co-workers re-
cently reported that both isomers of thalidomide are
rapidly racemized in plasma and in vivo.18 This report
suggests that even if only one isomer were teratogenic,
there would be no difference in teratogenicity between
the isomers in vivo. We prepared both isomers of
thalidomide by a known procedure19 and evaluated
them for TNFR inhibition. Dose-response curves for
the racemate and single isomers were similar (Figure
1) showing a very flat dose-response curve. The
a
IC50 for TNFR inhibition in LPS-stimulated PBMCs.
The activity of 5 was optimized by exploring substitu-
tion patterns on the 3-phenyl ring. Substitution with
an electron-withdrawing group or an electron-donating
group in the meta or para position resulted in increased
activity (Table 1). Substitution with a 3-cyano group
(6) or a 4-cyano group (7) afforded compounds which are
more active than thalidomide. Substitution with an
electron-donating group such as 3-methoxy (8) or 4-meth-
oxy (9) also afforded increases in activity. Thus, sub-
stitution effects appear to be mediated by steric effects.
Disubstitution as illustrated by the 3,4-dimethoxy ana-
log 10 afforded a synergistic effect with a compound
15-fold more potent than thalidomide. The 3,5-dimeth-
oxyphenyl analog 15 was 3 times less active than 10.
Homologs of 3,4-dimethoxy substitution of 10 were
prepared and evalutated. Substituents larger than
diethoxy had decreased activity. The diethoxy analog
11 was found to be 2 times as active as 10. Unlike
thalidomide, these compounds can inhibit 100% of the
TNFR formed by LPS stimulation.
R-isomer 32 (IC50 ) 3.1 µM) and S-isomer 33 (IC50
)
4.5 µM) of 21 were found to have TNFR synthesis
inhibition activity similar to that of the racemate. This
series of compounds, unlike thalidomide, do not have
the acidic chiral hydrogen and would be expected to be
chirally stable; thus differences in teratogenicity be-
tween the isomers could be of significance in vivo.
Con clu sion s. Using thalidomide as a lead structure,
we have designed a new series of drugs which inhibit
TNFR production to varying degrees in LPS-stimulated
human PBMCs. Replacement of the amino glutarimide
portion of thalidomide with â-amino â-aryl amino acid
derivatives and substitution of the phthaloyl ring have
resulted in analogs having TNFR inhibition potencies
approaching 500 times that found for thalidomide.
Phthalimido amides such as 10 were initially explored
because of their relationship to thalidomide hydrolysis
products. Isosteric replacement of the amide moiety
revealed that the amide moiety was not optimal (Table
2). Compound 10 was chosen as the parent compound
because of its high activity. Substituted amides larger

3240 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 17
Communications to the Editor
plantation Tolerance. Trans. Proc. 1987, XIX (1), 2658-2661.
(e) Klausner, J .; Makonkawkeyoon, S.; Akarasewi, P.; Nakata,
K.; Kasinrerk, W.; Corral, L.; Dewar, R.; Lane, C.; Freedman,
V.; Kaplan, G. The Effect of Thalidomide on the Pathogenesis
of Human Immunodeficiency Virus Type 1 and M. tuberculosis
Infection. J . Acquired Immune Defic. Syndr. 1996, 11, 247-257.
(7) Crawford, C. L. Thalidomide Neuropathy. Lepr. Rev. 1969, 40,
126-128.
(8) (a) Smithells, R. W.; Newman, C. G. H. Recognition of thalido-
mide defects. J . Med. Genet. 1992, 29, 716-723.
(9) Sampaio, E. P.; Sarno, E. N.; Galilly, R.; Cohn, Z. A.; Kaplan,
G. Thalidomide Selectively Inhibits Tumor Necrosis Factor R
Production by Stimulated Human Monocytes. J . Exp. Med. 1991,
173, 699-703.
(10) (a) Koch, H. P. Thalidomide and Congeners as Anti-inflammatory
Agents. Prog. Med. Chem. 1985, 22, 166-242. (b) Gunzler, V.
Thalidomide- A Therapy for the Immunological Consequences
of HIV Infection? Med. Hypotheses 1989, 30, 105-109.
(11) D’amato, R. J .; Loughnan, M. S.; Flynn, E.; Folkman, J .
Thalidomide is an inhibitor of angiogenesis. Proc. Natl. Acad.
Sci. U.S.A. 1994, 91, 4082-4085.
F igu r e 1. Inhibition of LPS-induced TNFR in LPS-stimulated
human PBMC (error bars represent (SEM).
(12) (a) Kalvin, D.; Woodard, R. Synthesis of (4R)-D,L-[4-2H]- and
(4S)-D,L-[4-2H]Homoserine Lactones. J . Org. Chem. 1985, 50,
2259-2263. (b) Madigan, D.; Muller, G.; Walters, D.; Culberson,
J .; DuBois, G.; Carter, J .; Nagarajan, S.; Klix, R.; Ager, D.; Klade,
C. Eur. Patent Appl. 344819 A1, Aug. 23, 1989.
Unlike thalidomide, analogs in this series inhibit 100%
of the TNFR synthesis in LPS-stimulated PBMCs. The
difference in dose-response curves is suggestive of a
possible change in mechanism of action which is being
investigated. Other studies are in progress to (a)
examine the effect of these analogs on leukocyte cytok-
ine production and their associated toxicities, (b) further
the understanding of the mechanism of action of these
analogs,20 and (c) expand the structure-activity rela-
tionship. The toxicology of compounds in this series is
currently being evaluated.
(13) Shealy, Y. F.; Opliger, C. E.; Montgomery, J . A. Synthesis of D-
and L- Thalidomide and Related Studies. J . Pharm. Sci. 1968,
57, 757-764.
(14) Davies, S. G.; Ichihar, O. Asymmetric Synthesis of R-â-Amino
butanoic acid and S-â-Tyrosine: Homochiral Lithium Amide
Equivalent for Michael Additions to R,â-Unsaturated Esters.
Tetrahedron Asymmetry 1991, 2, 183.
(15) Assay for inhibition of TNF synthesis by human PBMCs: Human
PBMCs from normal donors were obtained by Ficoll-Hypaque
(Pharmacia Fine Chemicals, Piscataway, NJ ) density centrifuga-
tion. Cells (106 cells/mL) were cultured in RPMI (Gibco Labo-
ratories, Grand Island, NY) supplemented with 10 AB+ serum
(Biocell, Rancho Dominguez, CA), 2 mM L-glutamine, 100 U/mL
penicillin, and 100 µg/mL streptomycin (Gibco). Test compounds
were dissolved in DMSO (Sigma Chemical, St. Louis, MO) at
20 mg/mL; further dilution was done with culture medium. The
final DMSO concentration in all assays including the controls
was 0.25%. Test compounds were added to cells 1 h prior to the
addition of LPS. PBMCs (106 cells/mL) were stimulated with 1
µg/mL of LPS from Salmonella minnesota R595 (List Biological
Labs, Campbell, CA). Cells, in triplicate, were incubated with
LPS for 18-20 h at 37 °C in 5% CO2. Supernatants were then
harvested and assayed for cytokine levels. In some experiments,
supernatants were kept frozen at -70 °C until use. Cell viability
was assayed by Trypan blue exclusion dye method. The concen-
tration of TNFR in the culture supernatants was determined
by ELISA (ENDOGEN, Boston, MA) according to the manufac-
turer’s directions. All compounds were assayed in a minimum
of three separate experiments. Percent inhibition was deter-
mined as 100 × [1 - (cytokine(experimental)/cytokine(control))].
(16) Thalidomide was routinely tested at 194 µM as a positive control
and found to inhibit TNF production by 50%. All other IC50’s
were calculated by nonlinear regression analysis (variable slope)
using Prism by GraphPad Software, Inc.
(17) Wintersk, W.; Frankus, E. Thalidomide enantiomers. Lancet
1992, 339, 365.
(18) Eriksson, T.; Bjorkman, S.; Roth, B.; Fyge, A.; Hoglund, P.
Stereospecific Determination, Chiral Inversion In Vitro and
Pharmacokinetics in Human of the Enantiomers of Thalidomide.
Chirality 1995, 7, 44-52.
(19) Reepmeyer, J . Separation of R- and S-Thalidomide by Reversed-
Phase HPLC with â-Cyclodextrin in the Mobile Phase. Chirality
1996, 8, 11-17.
(20) Corral, L.; Muller, G.; Moreira, A.; Chen, Y.; Wu, M.; Stirling,
D.; Kaplan, G. Selection of novel analogues of thalidomide with
enhanced TNF-R inhibitory activity. Mol. Med., in press.
Ack n ow led gm en t. Funding in part was provided
by U.S. Public Health Service Grant AI33124 (G.K.,
A.M.).
Refer en ces
(1) Muller, G.; Corral, L.; Shire, M.; Kaplan, G.; Stirling, D. Novel
inhibitors of tumor necrosis factor-R. Presented in part at the
210th American Chemical Society National Meeting, Chicago, IL,
Fall 1995; Med. Chem., Abstract 84.
(2) (a) Schumacher, H.; Smith, R. L.; Williams, R. T. The metabolism
of thalidomide: the spontaneous hydrolysis of thalidomide in
solution. Brit. J . Pharmacol. 1965, 25, 324. (b) Georgopoulos,
A.; Koch, H. P.; Czejka, M. J . In-vitro kinetics and metabolism
of R-phthalimido-glutarimide. Int. J . Exp. Clin. Chemother. 1990,
3 (3), 135-141.
(3) D’Arcy, P. F.; Griffin, J . P. Thalidomide Revisited. Adverse Drug
React. Toxicol. Rev. 1994, 13 (2), 65-76.
(4) Sheskin, J . Thalidomide in the treatment of lepra reactions. Clin.
Pharmacol. Ther. 1965, 6, 303.
(5) (a) Hendler, S. S.; McCarthy, M. F. Thalidomide for Autoimmune
Disease. Med. Hypotheses 1983, 10, 437-443. (b) Crawford, C.
L. Use of Thalidomide in Leprosy. Adverse Drug React. Toxicol.
Rev. 1994, 13 (4), 177-192.
(6) (a) Gutierrez-Rodriguez, O. Thalidomide, A Promising New
Treatment for Rheumatoid Arthritis. Arthritis Rheum. 1984, 27
(10), 1118-1121. (b) Peterson, D. L.; Georghiou, P. R.; Allworth,
A. M.; Kemp, R. J . Thalidomide as Treatment of Refactory
Apthous Ulceration Related to Human Immunodeficiency Virus.
Infection Clin. Infect. Dis. 1995, 20, 250-254. (c) Schuler, U.;
Ehninger, G. Thalidomide: Rationale for Renewed Use in
Immunological Disorders. Drug Safety 1995, 12 (6), 364-369.
(d) Vogelsang, G. B.; Hess, A. D.; Gordon, G.; Brundrette, R.;
Santos, G. W. Thalidomide Induction of Bone Marrow Trans-
J M9603328