Design and Stereoselective Synthesis of Retinoids with Ferrocene or N-Butylcarbazole Pharmacophores that Induce Post-Differentiation Apoptosis in Acute Promyelocytic Leukemia Cells

Article abstract of DOI:10.1002/cmdc.201100065

New ferrocene and N-alkylcarbazole retinoids were designed and synthesized stereoselectively in good yields. A number of these synthesized ligands, in particular 2, 3, and 11, were found to exhibit a high RARα activation potential and to effectively induce post-differentiation apoptosis in NB4 acute promyelocytic leukemia (APL) cells. Increasing the length of the side chain attached to the heterocycle of the carbazole arotinoids creates new opportunities for altered compound catabolism and for fine-tuning of the apoptosis-inducing potential of the ligand. In the carbazole series of new retinoids, maximal activity was established for N-butylcarbazole analogue 11 in all assays (i.e., RARα activation, differentiation induction, and apoptosis induction). Study of the mechanism of apoptosis revealed an activation of initiator caspases-8 and -9, followed by efficient cleavage of effector caspase-3 on day6 of treatment. Subsequent induction of a caspase cascade in NB4 cells triggered ultimate leukemic cell death. The selected ligands 2, 3, and 11 may provide alternate options for the treatment of APL in cases of life-threatening ATRA syndrome, resistance, and high toxicity to conventionally used retinoids.

Full text of DOI:10.1002/cmdc.201100065

MED
DOI: 10.1002/cmdc.201100065
Design and Stereoselective Synthesis of Retinoids with
Ferrocene or N-Butylcarbazole Pharmacophores that
Induce Post-Differentiation Apoptosis in Acute
Promyelocytic Leukemia Cells
Diana Ivanova,[a, b] Hinrich Gronemeyer,*[b] and Angel R. de Lera*[a]
New ferrocene and N-alkylcarbazole retinoids were designed
and synthesized stereoselectively in good yields. A number of
these synthesized ligands, in particular 2, 3, and 11, were
found to exhibit a high RARa activation potential and to effec-
tively induce post-differentiation apoptosis in NB4 acute pro-
myelocytic leukemia (APL) cells. Increasing the length of the
side chain attached to the heterocycle of the carbazole aroti-
noids creates new opportunities for altered compound catabo-
lism and for fine-tuning of the apoptosis-inducing potential of
the ligand. In the carbazole series of new retinoids, maximal
activity was established for N-butylcarbazole analogue 11 in all
assays (i.e., RARa activation, differentiation induction, and
apoptosis induction). Study of the mechanism of apoptosis re-
vealed an activation of initiator caspases-8 and -9, followed by
efficient cleavage of effector caspase-3 on day 6 of treatment.
Subsequent induction of a caspase cascade in NB4 cells trig-
gered ultimate leukemic cell death. The selected ligands 2, 3,
and 11 may provide alternate options for the treatment of APL
in cases of life-threatening ATRA syndrome, resistance, and
high toxicity to conventionally used retinoids.
Introduction
Leukemia is a hematological disease characterized by an ab-
normal proliferation of immature white blood cells. The exces-
sive number of leukocytes leads also to an imbalance in the
count of other blood cells such as platelets and erythrocytes.
In acute promyelocytic leukemia (APL), the proliferating cells
are promyelocytes whose differentiation into granulocytes is
blocked. According to the French–American–British classifica-
tion system, APL is the M3 subtype of acute myelogenous leu-
kemia (AML); this subtype represents 5–8% of all AML cases.
In 95–98% of cases, APL is caused by a balanced reciprocal
chromosomal translocation t(15;17)(q22;q12–21) that fuses the
retinoic acid receptor alpha gene (RARa), located on chromo-
some 17, with the promyelocytic leukemia gene (PML), initially
located on chromosome 15.[1] RARa, b, and g are ligand-depen-
dent transcription factors that regulate gene expression by
direct binding to DNA. Like other nuclear receptors, RARs have
a modular structure with six domains (A–F), which include the
DNA binding domain (C) and the ligand binding domain (E).
Oncogenic properties are attributed only to the PML–RARa(do-
mains B–F) fusion protein. The reciprocal fusion protein RARa-
(domain A)–PML can be found in 80% of APL cases, but no on-
cogenic functions can be associated with it.[2] According to the
’two-hit’ model, fusion proteins induce a ’pre-leukemic’ state,
while secondary genetic lesions that favor the survival of ge-
netically modified cells lead to leukemogenesis.[3]
functions’ as well as ’gain of new oncogenic functions’ of the
fusion partners. Oncoproteins form multiprotein co-repressor
complexes with histone deacetylases (HDACs), methyl-CpG
binding proteins, and DNA methyltransferases that cause tran-
scriptional repression of adjacent genes.
Patients with PML–RARa, NPM–RARa, and NuMA–RARa are
sensitive to all-trans-retinoic acid (ATRA), which in combination
with anthracycline antibiotics and consolidation therapy leads
to remission of up to 80–90% of APL patients. The problem is
that about one quarter of the newly diagnosed patients re-
lapse, become resistant to ATRA, and should be subjected to
bone marrow transplantation and treatment with highly toxic
arsenic trioxide. Patients with PLZF–RARa or STAT5b–RARa are
not sensitive to ATRA due to stronger affinity of the co-repress-
ors that bind not only to the RARa part but also to its partner
part in the fusion protein.[4]
[a] Dr. D. Ivanova,+ Prof. A. R. de Lera
Department of Organic Chemistry, University of Vigo, 36310 Vigo (Spain)
Fax: (+34)986-812-316
E-mail: qolera@uvigo.es
[b] Dr. D. Ivanova,+ Prof. H. Gronemeyer
Institute of Genetics and Molecular and Cellular Biology
BP 10142, Strasbourg, 67404 Illkirch Cedex (France)
Fax: (+33)388-653-437
E-mail: hg@igbmc.fr
[+] Present address: ChimOrbel Ltd.
In rare cases of APL, other genes are fused to the RARa
gene such as promyelocytic leukemia zinc finger (PLZF), nucle-
ophosmin (NPM), nuclear mitotic apparatus protein (NuMA), or
signal transducer and activator of transcription 5b (STAT5b)
genes. The fusion oncoproteins cause both ’loss of normal
83 blvd. Nicola Petkov, 1614 Sofia (Bulgaria)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201100065: characterization data for vari-
ous intermediate compounds in the synthesis of ligands 1–14.
1518
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 1518 – 1529

Ferrocene and N-Butylcarbazole Retinoids
The therapy of APL with ATRA or RARa agonists is based on
their binding with RARa and the induction of a major allosteric
change in the protein.[5] This conformational effect leads to dis-
sociation of the co-repressors and association of co-activators
(with histone acetyltransferase activity) with the receptor. Bind-
ing of co-repressors and co-activators to the receptor is mutu-
ally excluded. Histone acetylation leads to an open chromatin
conformation at the promoter site of the genes that regulate
the differentiation process. Then, general transcription factors
(TFIIB, TFIIH, etc.), Mediator, and RNA polymerase II are recruit-
ed, and the transcription is started. Thus, ATRA or RARa ago-
nists reverse the balance in NB4 APL cells from the oncogenic
PML–RARa to the normal RARa. In cancer cells, these process-
Ferrocenyl compounds have been analyzed as chromo-
phores with potential technical applications.[12,13] The ethyl
ester of ferrocenyl retinoic acid has been studied with regard
to papilloma regression in mouse models.[14] Ferrocene ana-
logues of the breast cancer drug tamoxifen and its active me-
tabolite hydroxytamoxifen, analogously termed ferrocifens,
have been synthesized and showed antiproliferative properties
in both hormone-dependent and hormone-independent
breast cancer cells.[15] Therefore, ferrocene compounds re-
vealed new perspectives in pharmacology and particularly in
oncology.[16]
The compounds with strongest anticancer activity, i.e., 2 and
3 from the ferrocene series and 11 from the N-alkylcarbazole
series that were identified here as new apoptogenic molecules
in NB4 APL cells, have potential application in cases of either
increased catabolism, resistance to conventionally used reti-
noid drugs, or retinoic acid syndrome, of which the latter rep-
resents a life-threatening complication in APL patients that re-
quires holding ATRA for a period of time.
es occur at pharmacological concentrations of ATRA (~10À8
–
10À6 m) that are higher than its normal physiological concentra-
tion (~10À9 m).
ATRA is the natural ligand of RARs, whereas 9-cis-retinoic
acid is a pan-agonist that activates both RARs and retinoid X
receptors (RXRs).[6] ATRA-like compounds are excluded by the
sharper bend in the RXR ligand binding pocket (RXR-LBP),
which accommodates conformationally flexible analogues of 9-
cis-RA. RXR agonists (rexinoids) activate RXR in permissive het-
erodimers (such as PPARa–RXR, PPARg–RXR, and LXR–RXR)
that can also be activated by ligands of the RXR partner pro-
tein. Non-permissive heterodimers (such as RAR–RXR, TR–RXR,
and VDR–RXR)[7] are activated only by ligands of the RXR part-
ner protein; however, rexinoids can act in synergy with them.
RXR agonists can cause undesirable therapeutic side effects
such as an increase in the level of triglycerides (by LXR–RXR),
an imbalance in glucose metabolism (by PPARg–RXR), or a de-
crease in thyroid hormone levels (by TR–RXR).[8] In contrast,
RAR class-selective agonists act preferably through RAR path-
ways and therefore do not result in therapeutic side effects
caused by activation of RXR pathways.
Results and Discussion
The new anticancer retinoids are the result of several series of
experiments: 1) rational design and stereoselective synthesis of
RAR class-selective retinoids with ferrocene-, phenanthrene-, or
N-alkylcarbazole-containing pharmacophores; 2) analysis of the
RAR/RXR selectivity of the ligands; 3) correlation of the RARa
activation potential with the induction of post-differentiation
apoptosis in NB4 cells, according to the mechanism of APL;
and 4) identification of molecular targets of the ligands in ex-
trinsic and intrinsic apoptotic pathways.
RARa activation is the target process in the induction of
post-differentiation apoptosis in NB4 cells harboring PML–
RARa fusion protein. RARb is a known tumor suppressor in-
duced by retinoids, whereas RARg is not involved in the differ-
entiation of NB4 cells.
Herein we report the stereoselective synthesis of a series of
ferrocene, phenanthrene, and N-alkylcarbazole retinoids 1–14.
New RAR class-selective (i.e., RAR versus RXR) agonists (com-
pounds 2, 3, and 8–13) were identified, and their potential to
induce post-differentiation apoptosis in NB4 APL cells was es-
tablished.[9] Details of the caspase cascade that triggered apop-
tosis in APL cells after treatment with the new potential drugs
were also investigated.
Design and synthesis of new RAR class-selective agonists
New ferrocene-, phenanthrene-, and carbazole-containing
pharmacophores were designed to contain hydrophobic cyclic
groups and all-trans side chains in order to bind to the I-
shaped RAR ligand binding domain (LBD) and to be discrimi-
nated by the L-shaped RXR ligand binding pocket (LBP). A ter-
minal carboxylic group in the structures of all studied retinoids
guaranteed successful anchoring of the ligand in the LBP of
the corresponding retinoic acid receptor.
Ferrocene derivatives 2 and 3 are the first organometallic
compounds determined to be RAR class-selective agonists. An
alkyl side chain at the nitrogen atom of the carbazole aroti-
noids (polyaromatic retinoid) 8–13 was envisaged to possibly
alter the catabolism, pharmacokinetics, and pharmacodynam-
ics of the ligand in order to decrease its toxicity and to im-
prove its therapeutic index and efficacy.
Our previous study on the synthesis and biological activity
of phenanthrene retinoids revealed that the phenanthrene
group can be accommodated in the LBPs of RARa and
RARb.[17] These results prompted us to synthesize other phen-
anthrene retinoids in order to study their differential RAR/RXR
activation potential and to identify new selective retinoid ago-
nists in the phenanthrene group of retinoids. The promising
results obtained in the investigations of ferrocene analogues
of anticancer compounds in vitro were the rational back-
ground to generate ferrocene retinoic acid analogues that may
N-alkylcarbazole derivatives have been studied as electro-
photographic photoreceptors.[10] A very interesting and atypi-
cal retinoid, AGN193198 (Allergan Inc., Irvine, CA, USA), which
is structurally related to 11, was extremely effective in the in-
duction of apoptosis in cancer cells (prostate, breast, and gas-
trointestinal carcinoma cells) by caspase (-3, -8, -9, and -10)
cleavage, but neither RAR/RXR activation nor cell differentia-
tion were registered.[11]
ChemMedChem 2011, 6, 1518 – 1529
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
1519

D. Ivanova, H. Gronemeyer, A. R. de Lera
MED
be able to induce post-differentiation apoptosis in NB4 APL
cells.
Retinoids 1–14 with all-trans side chains were designed by
replacement of the natural b-ionone ring of ATRA with ferro-
cene, phenanthrene, or alkylated carbazole groups. These
modifications provided an opportunity to study experimentally
the adaptation of both the ligand and the hydrophobic LBP to
trigger transactivation of the receptor. Ligand docking is envis-
aged in the future based on the recent determination of the
crystal structure of RARa[18]—the target receptor for induction
of post-differentiation apoptosis in NB4 cells according to the
accepted mechanism of APL. While this work started, only the
structures of ligand binding domains of RARb,[19] RARg[20], and
an antagonist-bound RARa[21] as a heterodimer with RXR had
been resolved by X-ray crystallography.
Modifications of the side chain (in length and volume) were
made in order to study empirically the restrictions and the flex-
ibility of all RAR LBPs. For example, a phenyl group was intro-
duced at different position of the side chain in compounds 1,
4, 5, and 7. Compounds 3 and 8 were modified to contain two
or one isoprene units in the side chain, respectively. In reti-
noids 2, 6, and 9–14, the presence of a terminal benzoic acid
group in a trans-configured side chain was envisaged to elicit
pan-RAR activation analogously to TTNPB {p-[(E)-2-(5,6,7,8-tet-
rahydro-5,5,8,8-tetramethyl-2-naphthyl)propen-1-yl]benzoic
acid}—a pan-RAR agonist, the predecessor of the family of syn-
thetic arotinoids (retinobenzoic acids).
Scheme 1. Synthesis of ferrocene, phenanthrene, and N-alkylcarbazole reti-
noids 1–14.[24] a) (EtO)2P(O)CH2C6H4CO2Me, nBuLi, DMPU, THF, À708C,
30 min, then À408C, 2 h; b) DIBAL-H, THF, À708C, 1–2 h; c) MnO2, CH2Cl2, RT,
2–6 h; d) (EtO)2P(O)CH2CO2Et, nBuLi, DMPU, THF, À708C, 30 min, then
À408C, 2 h; e) (EtO)2P(O)CH2C(CH3)=CHCO2Et, nBuLi, DMPU, THF, À708C,
30 min, then À408C, 2 h; f) MeOH/5n NaOH, reflux, 2–14 h; 6n HCl. See Ex-
perimental Section for reaction times of various analogues for steps c) and
f).
An alkyl side chain of increasing length, attached to the het-
erocycle of the carbazole arotinoids, was foreseen to fine-tune
the potentials with regard to RARa activation and induction of
post-differentiation apoptosis of the new analogues 8–13. The
molecular basis for the maximal RARa-agonist activity of com-
pound 11 from the carbazole group of arotinoids will be eluci-
dated in the future by ligand docking in an agonist-bound
RARa LBD.
dehydes with C2- or benzyl phosphonoesters. This result was
expected because the reaction mechanism involves the forma-
tion of a phosphonoester carbanion intermediate that is stabi-
lized by delocalization of the negative charge. The nucleophilic
attack of the carbanion on the carbon atom of the aldehyde
group proceeds under thermodynamic control, thus leading to
stereoselective synthesis of the thermodynamically more stable
product with trans configuration at the newly formed double
bond. Minor 2E/2Z isomerization (~10%) was registered by
NMR analysis of the reaction mixture only in the HWE reactions
with C5-phosphonate. This effect was due to the low energy
barrier to isomerization of the terminal double bond and did
not depend on the initial 2E/2Z ratio in the C5-phosphonate.[23]
In the final step, the purified all-trans isomer of the ester from
the HWE reaction was held at reflux in methanolic sodium hy-
droxide to obtain the desired acid in high yields (Table 1, col-
umn f). The new potential anticancer agents were character-
ized by NMR, IR, HRMS, and/or elemental analysis (see Experi-
mental Section below).
In addition, the alkylated backbone of the heterocycle was
foreseen to create an opportunity for an altered catabolism
and decreased toxicity of the ligand (pharmacologically orient-
ed drug design). It was taken into consideration that heteroar-
otinoids have been found to be less toxic[22] than TTNPB be-
cause of the metabolic stability of its tetramethylated cyclo-
hexenyl ring.
Stereoselective synthesis of the ligands
Horner–Wadsworth–Emmons (HWE) olefination was selected as
a key reaction for the synthesis of new N-alkylcarbazole, ferro-
cene, and phenanthrene retinoid ligands because of the high
yields and stereoselectivity in the reactions of suitably con-
structed C2-, C5-, or benzyl phosphonates with corresponding
aromatic aldehydes (Scheme 1). Environmentally hazardous
metals such as lead and palladium were not used in the syn-
thesis of our compounds, considering their potential applica-
tion in the pharmaceutical industry.
Differential RAR/RXR transactivation by the new retinoids
Identification of new transcriptionally competent ligands for
nuclear receptors was carried out using the ’domain-swap’
method, which is based on transcription of the protein-coding
genes of interest followed by ligand-induced receptor activa-
tion in genetically engineered ’reporter’ cells (baculovirus-in-
All-trans configurations of the compounds were maintained
along the reaction sequences. Only trans isomers were ob-
tained in the HWE reactions of the corresponding aromatic al-
1520
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 1518 – 1529

Ferrocene and N-Butylcarbazole Retinoids
Table 1. Yields of compounds in the multistep synthesis of retinoids 1–
14.
Table 2. Differential RAR/RXR activation by new RAR class-selective reti-
noids.
Compd
Group[a]
a[b]
b[b]
c[b]
d[b]
e[b,c]
f[b]
Compd
Group RARa RARb RARg
RXRb
Selectivity[a]
vehicle ctrl
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Fc
Fc
Fc
51
68
–
51
75
100
75
–
100
74
75
57
76
51
100
100
100
100
73
100
73
–
–
–
–
–
63
83
67
63
98
92
98
–
–
–
–
–
96
96
–
–
–
100
82
–
77
100
100
94
99
83
100
90
94
vehicle
ATRA
9-cis-RA
ctrl
ctrl
ctrl
ctrl
ctrl
ctrl
ctrl
Fc
1.00
17.62
ND
46.06 25.03
45.49 ND
ND 17.52
ND ND
1.26 12.66
1.00
9.61
ND
1.00
4.17
ND 30.03
6.40
ND
ND
1.00
ND[b] RAR
pan-RAR/RXR
RAR
RARa agonist
RARb agonist
RARg agonist
RARb,g
Fc
–
TTNPB
ND
ND
ND
PHT
PHT
PHT
NAC
NAC
NAC
NAC
NAC
NAC
Fc
71
80
–
–
–
–
–
–
–
BMS753[c]
–
BMS641[c]
75
90
–
–
–
–
–
–
BMS961[c]
6.50
6.17
ND
1
2
3
4
6
8
9
10
11
12
13
1.69
2.05
2.58
1.77
1.35
1.70
2.17
2.39
2.21
2.17
1.88
Fc
20.78 19.60 10.22
RAR
RAR
99
Fc
Fc
PHT
13.51
3.28 11.92
2.48 4.62
9.33
5.28
9.65
1.32
6.85
8.81
9.73
9.18
100
100
100
99
RARb,g; weak RARa
partial RARb
RAR
RAR
RAR
RAR
RAR
RAR
–
–
–
–
NAC 13.37 20.56
NAC 20.60 19.89
NAC 26.68 19.94
NAC 31.36 19.55
–
[a] Fc=ferrocene, PHT=phenanthrene, NAC=N-alkylcarbazole. [b] Stages
of the synthesis according to Scheme 1. [c] 2E/2Z ratio is ~90:10 (see text
for details).
NAC 21.46 19.30 10.74
NAC 13.66 18.87 9.26
[a] RAR/RXR activation is presented as fold induction, defined as the ratio
of the relative luciferase units (RLU) detected in the presence of the cor-
responding ligand (at 10À6 m) over the RLU of the vehicle control in HeLa
cells. Fold induction of 2.0–3.0 was considered very low activation, a
value of 3.0–3.5 as weak activation, and a value >3.5 as considerable ac-
tivation. All values are compared with the data of the corresponding pos-
itive control. The values represent the average of four measurements
from two separate experiments with duplicate determinations. [b] ND:
not detected. [c] Selective agonists BMS753, BMS641, and BMS961 were
kindly provided by Bristol-Myers Squibb.
fected insect cells, CV-1 cells, human HeLa cells, etc.).[25] This
was the method of choice for identification of new RARa ago-
nists instead of computer-aided ligand docking considering
the fact that no crystal structure of an agonist-bound RARa
LBD is at present available. In our experiments, HeLa cells were
transfected with two types of recombinant DNA: 1) a chimeric
gene that generates a fusion protein DBD(Gal4)–LBD(RAR) con-
sisting of the DNA binding domain (DBD) of Gal4 (a yeast tran-
scription factor) fused to the ligand binding domain (LBD) of
the nuclear receptor of interest; and 2) a second expression
vector, (17m)5-bG-luc, that contains a luciferase (luc) reporter
gene under the control of five copies of the cognate GAL4 re-
sponse element [(17m)5] in front of a b-globin promoter.[26] Ac-
tivation of RAR LBD by an agonist in the presence of luciferase
[EC 1.13.12.7], coenzyme A, and luciferin generates lumines-
cence, which can be detected using a luminometer.
the RARa LBP may reveal the structural basis for the maximal
RARa activation elicited by the N-butyl derivative 11 from the
RAR class-selective carbazole series 8–13.
We found previously that a phenanthrene group in the hy-
drophobic moiety of the molecule, as in compound 6, creates
steric hindrance with RARg Met272.[17] Retinoid 6 is discriminat-
ed also by RARa because of the presence of a phenyl ring with
two double bonds in the central part of the side chain, as dis-
cussed above. Being discriminated by RARa, RARg, and RXR,
retinoid 6 is partial RARb agonist because maximal RARb acti-
vation could not be reached by an increase in ligand concen-
tration. The binding of long molecules such as 5 and 7 and
short molecules such as 14 was abolished by the length re-
strictions for successful anchoring of the ligand in all RAR LBPs.
The new RAR class-selective compounds were obtained to
broaden the spectrum of potentially less toxic drugs in APL
therapy. The RAR class selectivity of the new ligands 2, 3, and
8–13 is promising for further preclinical trials, as these com-
pounds may avoid undesirable side effects (imbalance in trigly-
ceride, glucose, and thyroid hormone levels) due to RXR-per-
missive heterodimer activation. Therefore, we examined
whether the high RARa activation potential of these com-
pounds correlates with the induction of post-differentiation
apoptosis in NB4 cells according to the mechanism of APL.
Using fluorescence-activated cell sorting (FACS) analysis, we se-
lected compounds 2 and 3 from the ferrocene series and N-
butyl derivative 11 from the carbazole series as strongest
This method is insensitive to endogenous retinoic acid re-
ceptors because they cannot recognize the Gal4 response ele-
ments. A mutant of RXRb (RXRbDAB) was used to eliminate
false-positive signals.[27] RXRb transactivation was used to eval-
uate the transactivation at all three RXRs (a, b, and g) because
of the similar structure of the corresponding RXR LBPs.
Structure–activity relationship of ligand-induced RAR trans-
activation
All-trans-retinoids 2, 3, and 8–13 were found to be RAR class-
selective ligands (Table 2). All compounds studied here were
discriminated by the L-shaped RXR LBP because of the flexure
limit of all-trans side chains.
Ferrocenyl derivatives 1 and 4 with a phenyl group in the
central region of the side chain activated RARb and RARg. The
prevention of RARa transactivation due to steric hindrance,
which is imposed by the presence of a phenyl group with two
adjacent double bonds in the middle region of the polyenic
side chain, was empirically established here. Ligand docking in
ChemMedChem 2011, 6, 1518 – 1529
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
1521

D. Ivanova, H. Gronemeyer, A. R. de Lera
MED
apoptogenic substances in the corresponding groups of li-
gands.
propidium iodide (PI) to discriminate necrotic from apoptotic
cells. Ferrocene ligands 2 and 3 and N-butylcarbazole arotinoid
11 with high RARa activation potential and consequently high
differentiation-inducing potential from the corresponding
series of new pharmacophores triggered effectively apoptosis
in NB4 cells (Figure 1).
Correlation of RARa activation with differentiation and
apoptosis induction in NB4 cells by new anticancer agents
Both ferrocene ligands 2 and 3 have a high potential to
induce efficiently post-differentiation apoptosis in NB4 APL
cells. As the ethyl ester of ferrocene retinoic acid 3 was former-
ly investigated in a papilloma regression study,[14] we selected
the new compound 2 as a representative RAR agonist from
the ferrocene series of retinoids for analysis of its mechanism
of activity. Thus, highly efficient retinoid 2 from the ferrocene
series and the most active compound 11 from the N-alkylcar-
bazole series were selected for Western blot analysis to exam-
ine their mechanism of apoptosis induction.
NB4 cells undergo differentiation in response to treatment
with ATRA or RARa agonists at pharmacological concentra-
tions. In the ferrocene series, only ligands 2 and 3 (at 10À6 m),
which have a high RARa activation potential, demonstrated
high differentiation- and apoptosis-inducing activities, accord-
ing to the mechanism of APL. Arotinoid 11 triggered maximal
post-differentiation apoptosis in NB4 cells because of its maxi-
mal potential in the carbazole series of ligands to activate
RARa.
As detected by FACS analysis, NB4 cells differentiated into
granulocytes and expressed CD11c glycoproteins on their sur-
face in response to treatment with new RAR class-selective ret-
inoids 2, 3, and 8–13 at 10À6 m concentration (Figure 1A). Ma-
turation of NB4 cells into monocytes and cell-surface expres-
sion of CD14 was not induced (data not shown).
Western blot analysis pertaining to apoptosis induction
Induction of tumor necrosis factor-related apoptosis- inducing
ligand (TRAIL) by the extrinsic apoptosis pathway was the first
cellular target, detected on day 3 after treatment of NB4 cells
with ferrocene retinoid 2 (Figure 2A). Weak TRAIL expression
was registered at treatment of NB4 cells with N-butylcarbazole
arotinoid 11 (Figure 2B). Constitutive expression of at least one
TRAIL receptor (DR5) in NB4 cells was demonstrated by West-
ern blotting.
Retinoid-induced apoptosis of leukemic cells (HL60, NB4,
etc.) is a post-differentiation process.[28] Retinoids that trig-
gered maturation in a high percentage of NB4 cells but at low
intensity of the labeling were possibly not apoptogenic be-
cause terminal differentiation of the leukemic cells has not
been reached. Apoptosis was examined by FACS analysis using
double labeling of treated NB4 cells with Annexin V–FITC and
TRAIL triggered formation of the death-inducing signaling
complex (DISC) that effectively induced subsequently the acti-
vation of initiator caspase-8 by cleavage, followed by cleavage
of effector caspase-3. Western blot analysis proved activation
of caspase-8 by formation of p18 and p43/41 fragments that
induced efficient cleavage of caspase-3 to fragment p17. Minor
detection of fragment p17 from caspase-3 in the vehicle con-
trol was due to spontaneous death of minimal amounts of
NB4 cells from the non-treated cells. Simultaneous action of
caspases-3, -8, and -9 accelerated apoptosis and led to massive
leukemic cell death on the day 6 of treatment.
Expression of the BH3-interacting domain death agonist
(Bid), a protein that connects the intrinsic and extrinsic apopto-
sis pathways, was not induced. Nevertheless, upregulation and
cleavage of initiator caspase-9 by the intrinsic apoptosis path-
way was triggered at the same time as caspase-8 by the extrin-
sic apoptosis pathway. The casp9 gene was detected to harbor
a 2-bp-spaced direct repeat of a retinoic acid-responsive ele-
ment (DR-2 RARE) in its second intron and was shown to be
under the direct control of RAR in a previous study of carcino-
ma cells (MCF-7).[29] This finding supports the direct activation
of caspase-9 by retinoids.
An interesting protein with a molecular mass of ~25 kDa
was detected by the antibody for caspase-9 at treatment with
all studied synthetic or natural retinoids. This protein appeared
on days 2–3 after beginning the treatment with retinoids even
before TRAIL activation. The identification of this molecular
target is of great interest to clarify the general mechanism un-
derlying retinoid anticancer activity in APL in the future.
Figure 1. FACS analysis of A) differentiation induction (fluorescence intensity
of CD11c+/PIÀ cells) and B) apoptosis induction (percent Annexin V+ cells) in
NB4 APL cells after six days of treatment with the new retinoid agonists 2, 3,
and 8–13 (all at 1 mm). Values represent the average ÆSD of two independ-
ent experiments.
1522
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 1518 – 1529

Ferrocene and N-Butylcarbazole Retinoids
Figure 2. Western blot analysis of expression of caspase cascade proteins triggered by A) ferrocene retinoid 2 or B) N-butylcarbazole arotinoid 11. Abbrevia-
tions 1D, 2D, … denote day 1, day 2, … of treatment. The total lysate of HeLa cells (cultured 2 days) was used as a positive control for DR5 expression, while
the total lysate of NB4 cells treated with Etoposide (1 mm) was used as a positive control for cleavage of caspases. The cells for vehicle control were main-
tained in culture during the treatment and showed identical immunoreactive proteins by Western blot for days 1, 3, and 6.
In summary, Western blot analysis to determine the mecha-
nism of apoptosis revealed that the new ferrocene or N-butyl-
carbazole agonists with high RARa activation potential induce
post-differentiation apoptosis by early expression of TRAIL,
which is followed by activation of initiator caspase-8 from
DISC and activation of effector caspase-3. Beginning at the cell
surface by the extrinsic pathway, cell death was accelerated by
the intrinsic inducers of apoptosis. Activation of caspase-9,
possibly under the direct control of RAR, amplified caspase-3
cleavage, and massive NB4 cell death occurred on day 6 of
treatment.
rations of the double bonds were maintained along the reac-
tion sequences.
An alkyl side chain with increasing number of carbon atoms
was attached to the heterocycle of the carbazole arotinoids in
order to fine-tune the ligand agonist activity. In the carbazole
series of new pharmacophores a maximum of the activity was
found for the N-butyl analogue 11 in all assays—RARa activa-
tion, induction of differentiation, and induction of apoptosis in
NB4 cells.
Western blot analysis of treated NB4 APL cells revealed that
ferrocene retinoid 2 induced TRAIL expression and caspase-8
activation from the DISC by the extrinsic apoptosis pathway.
TRAIL was detected on day 3 of treatment of NB4 cells with
ferrocene retinoid 2; however, it was weakly induced by the N-
butylcarbazole arotinoid 11. At least one TRAIL receptor (DR5)
was constitutively expressed in NB4 APL cells.
Conclusions
We established an RAR class-selective (RAR versus RXR) agonist
activity of the newly synthesized ferrocene retinoids 2 and 3
and N-butylcarbazole arotinoid 11, all of which induce differen-
tiation and apoptosis in NB4 cells. In accordance with the dis-
cussed mechanism of APL, post-differentiation apoptosis is a
consequence of the high RARa activation potential of the new
pharmacophores.
At the same time, the intrinsic apoptosis pathway accelerat-
ed the retinoid-induced caspase cascade. Western blot analysis
proved upregulation and cleavage of caspase-9, which is nec-
essary for the assembly of the ’apoptosome’—the wheel-
shaped heptameric death machine, consisting of caspase-9,
cytochrome c, and Apaf-1 (apoptotic protease-activating
factor 1).
The compounds were obtained stereoselectively in good
yields by multistep chemical syntheses, using Horner–Wads-
worth–Emmons olefination as a key reaction. All-trans configu-
ChemMedChem 2011, 6, 1518 – 1529
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
1523

D. Ivanova, H. Gronemeyer, A. R. de Lera
MED
tained in an analogous procedure from the methyl ester of p-toluic
acid.
BH3-interacting domain death agonist (Bid), a protein that
connects the extrinsic and intrinsic apoptosis pathways, was
not activated. This result was interpreted in terms of the find-
ing that casp9 was directly activated through the retinoic acid
response element.[29]
General procedures: All manipulations with isomerizable products
were performed under low light in amber glassware. The synthe-
sized retinoids were stored for five years at À808C in the dark; all
biological assays were performed with compounds from deep
frozen (À808C) storage. HWE reactions and reductions with DIBAL-
H were carried out in flame-dried flasks.
Coordinated action of caspases-3, -8, and -9 triggered a
ligand-induced caspase cascade. FACS analysis of post-differen-
tiation apoptosis and Western blot analysis proved massive
death of NB4 cells on day 6 of treatment with ferrocene 2 or
N-butylcarbazole ligand 11.
The new apoptogenic ferrocene (2 and 3) and N-butylcarb-
azole (11) RAR agonists provide alternative options to study
the treatment of APL in cases of life-threatening ATRA syn-
drome, high toxicity, or resistance to conventionally used reti-
noids in order to broaden the spectrum of available effective
and less toxic drugs.
Horner–Wadsworth–Emmons (HWE) reactions: To a stirred solu-
tion of the corresponding phosphonate (1.75 mmol) and DMPU[30]
(3.45 mmol) in dry THF (4 mL) was added nBuLi (1.75 mmol) at 08C
under argon atmosphere. This temperature was maintained for
30 min. After that, the temperature was decreased to about
À708C, and a solution of the corresponding aldehyde (1.17 mmol)
in THF (2 mL) was added dropwise. The reaction mixture was
stirred for 30 min at À708C and then at À408C for 2 h. Subse-
quently, the reaction was left to proceed in the dark at room tem-
perature (RT) overnight. Complete conversion of the starting alde-
hyde was verified by TLC (EtOAc/hexane 20:80). Then, an equal
volume of saturated NH4Cl(aq) was added to the reaction mixture
and the organic layer was extracted with Et2O (for ferrocenyl reti-
noids) or EtOAc (for phenanthryl and N-alkylcarbazolyl retinoids).
The combined organic extracts were washed with H2O, dried over
Na2SO4, and the solvent was evaporated in vacuo. The residue was
purified by CC.
Perspectives
The RARa activation by arotinoid 11 from the carbazole series
of RAR agonists will be explained by ligand docking in an ago-
nist-bound RARa LBD.[18]
Ferrocene retinoids 2 and 3 and N-butylcarbazole arotinoid
11, which were evaluated as compounds with highest activity
in our study, have promise for further preclinical studies of
acute promyelocytic leukemia to determine their therapeutic
index and efficacy. The N-butyl side chain of arotinoid 11 cre-
ates an opportunity for an altered catabolism of the com-
pound and a decrease in drug toxicity compared with conven-
tionally used retinoids.
DIBAL-H reductions: To a stirred solution of the corresponding
ester (2.6 mmol) in dry THF (18 mL) was added dropwise DIBAL-H
(7.8 mmol, 1.2m in hexane) at À708C under argon atmosphere.
The mixture was treated with 6n HCl until the suspension dis-
solved (at pH 1–2), and the organic layer was extracted with
EtOAc. The extracts were washed with H2O, dried over Na2SO4, and
the solvent was evaporated in vacuo.
MnO2 oxidations: To a stirred solution of the corresponding alco-
hol (2.6 mmol) in dry CH2Cl2 (25 mL) was added a fivefold excess
(w/w relative to the alcohol) of dried MnO2.[31] The suspension was
stirred at RT until the reaction was completed for 2 h (1–4), 4 h, (5,
7), or 6 h (6) (monitored by TLC). The reaction mixture was filtered
through Celite, the solids were washed with EtOAc, and the sol-
vent from the organic layer was evaporated in vacuo. The residue
was filtered through a short silica gel layer (eluent EtOAc/hexane
50:50) to remove residual MnO2 and was employed directly in a
subsequent HWE reaction.
Experimental Section
Chemistry
Materials and methods: Column chromatography (CC) was carried
out on Kieselgel 60 (230–400 mesh, Merck), using EtOAc/hexane as
eluent with an increasing percentage of EtOAc; thin-layer chroma-
tography (TLC) was performed on silica gel-coated plates with fluo-
rescence indicator (Merck), with visualization at 254 nm. NMR spec-
tra were recorded on a Bruker-AMX400 instrument at 400.13 MHz
(1H NMR) or 100.61 MHz (13C NMR). Chemical shifts are given in d
[ppm], and coupling constants (J) are given in Hz. Assignments of
resonances in 13C NMR spectra, which were supported by DEPT135
spectra, are: CH3(q), CH2(t), CH(d), C(s). IR spectra were registered
on a JASCO FTIR-4200 spectrometer; peaks are given in n˜ [cmÀ1
with intensities s (strong), m (medium), or w (weak). MS (EI, 70 eV)
spectra were recorded on a VG-Autospec M apparatus. Melting
points (mp) were determined on a Stuart Scientific apparatus.
Ester hydrolysis: Pure all-trans isomer of the corresponding ester
1
(0.7 mmol, isomeric purity determined by H NMR), obtained after
HWE olefination, was suspended in MeOH (110 mL) and 5n NaOH
(55 mL). The mixture was held at reflux for 2 h (1, 14), 4 h (2–9),
7 h (10–12), or 14 h (13). Complete hydrolysis of the ester was veri-
fied by TLC (20:80 EtOAc/hexane; 10% MeOH in CH2Cl2). The reac-
tion mixture was then cooled to RT, neutralized to pH 7 with 6n
HCl, and subsequently an equal volume of H2O was added. The or-
ganic layer was extracted with EtOAc and the extracts were
washed with H2O. Note that the organic layers should not be dried
over Na2SO4 because of considerable irreversible sorption of the
desired acid. The solvents were evaporated and the residue was
dried in vacuo. The resulting oil was crystallized from hexane or 5–
10% EtOAc in hexane to obtain the desired acid.
]
Carbazole, triethyl phosphonoacetate, ferrocene aldehyde, and
phenanthrene-9-carbaldehyde were purchased from Sigma–Al-
drich. Other phosphonates were obtained by standard procedures:
triethyl 3-methyl-4-phosphono-2-butenoate (C5-phosphonate) was
synthesized by bromination of the methyl ester of 3,3’-dimethyla-
crylic acid with an equimolar amount of N-bromosuccinimide in
the presence of traces of dibenzoyl peroxide in CCl4 (reflux, 6 h),
followed by an Arbuzov reaction of the resulting bromomethylderi-
vative with a slight excess (1.1-fold) of P(OEt)3 (reflux, 4 h). p-[(Di-
ethylphosphono)methyl]benzoate (aromatic phosphonate) was ob-
Synthesis of N-alkylcarbazoles:[32] Powdered KOH (0.53 g,
9.5 mmol) was dissolved in DMSO (10 mL) at RT. A solution of car-
bazole (1.05 g, 6.3 mmol) in DMSO (6 mL) was added dropwise
(H2O bath cooling) and the corresponding alkyl iodide (9.5 mmol)
1524
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 1518 – 1529

Ferrocene and N-Butylcarbazole Retinoids
was added. The reaction mixture was stirred for 30 min, poured in
H2O (10 mL)/ice, and extracted with CH2Cl2. The combined organic
extracts were washed with H2O and brine, dried over Na2SO4, and
the solvents were evaporated in vacuo. The residue was purified
by CC on silica gel (1–5% EtOAc in hexane).
1280, (s), 1179 (m), 984 (s), 937 (m), 807 (m), 766 cmÀ1 (m); MS (EI,
70 eV) m/z (%): 358 (100) [M]+; 191 (27); HRMS: m/z calcd for
C21H18FeO2: 358.0656, found: 358.0653.
(all-E)-3,7-dimethyl-9-ferrocenylnona-2,4,6,8-tetraenoic acid 3:[34]
Red oil (67% total yield, Scheme 1 steps e,b,c,e,f): 1H NMR
(400 MHz, CDCl3): d=2.02 (s, 3H, CH3), 2.37 (s, 3H, CH3), 4.13 (s,
5H, H-Fc), 4.32 (s, 2H, H-Fc), 4.45 (s, 2H, H-Fc), 5.81 (s, 1H, H-2),
6.19 (d, J=11.0 Hz, 1H, H-6), 6.32 (d, J=14.5 Hz, 1H, H-4), 6.46 (s,
2H, H-8+H-9), 6.90–7.10 ppm (m, 1H, H-5); IR (film): n˜ =2750–
3250 (br, OH), 2923 (m), 1673 (CO, s), 1603 (C=C, m), 1572 (s), 1254
(s), 1186 (s), 1153 (m), 956 cmÀ1 (m); MS (EI, 70 eV) m/z (%): 362
(100) [M]+, 318 (25), 256 (46), 186 (25), 121 (17); HRMS: m/z [M]+
calcd for C21H22FeO2: 362.0969, found: 362.0967.
Synthesis of N-alkylcarbazol-3-aldehydes by Vilsmeier–Haack
formylation:[33] To a stirred solution of the corresponding N-alkyl-
carbazole (5.8 mmol) in anhydrous DMF (10 mL) was added drop-
wise POCl3 (40.6 mmol) under argon atmosphere and cooling (ice/
H2O). Subsequently, the reaction mixture was allowed to warm to
RT for 1 h. It was then carefully heated and held at reflux (808C)
for an additional 5 h. After that, the solution was poured on ice/
H2O and extracted with CH2Cl2. The combined extracts were
washed with H2O, dried over Na2SO4, and the solvents were evapo-
rated in vacuo. The residue was purified by CC on silica gel (15%
EtOAc in hexane).
Ethyl (all-E)-5-[p-(2’-ferrocenylethen-1’-yl)phenyl]-3-methylpenta-
2,4-dienoate: Red oil (82% yield, Scheme 1 step e): 1H NMR
(400 MHz, CDCl3): d=1.31 (t, J=7.1 Hz, 3H, CH3-ester), 2.41 (s, 3H,
CH3), 4.16 (s, 5H, H-Fc), 4.20 (q, J=7.1 Hz, 2H, CH2-ester), 4.32 (s,
2H, H-Fc), 4.49 (s, 2H, H-Fc), 5.91 (s, 1H, H-2), 6.67 (d, J=16.1 Hz,
1H, CH=), 6.81 (d, J=16.1 Hz, 1H, CH=), 6.91 (d, J=16.1 Hz, 1H,
CH=), 6.93 (d, J=16.1 Hz, 1H, CH=), 7.41 (s, 2H, Har), 7.42 ppm
(s, 2H, Har); IR (film): n˜ =3096 (m), 3030 (m), 2958 (m), 2924 (m),
2853 (m), 1706 (CO, s), 1592 (C=C, s), 1239 (m), 1152 (s), 1044 (m),
963 (m), 814 cmÀ1 (m); MS (EI, 70 eV) m/z (%): 426 (100) [M]+;
HRMS: m/z [M]+ calcd for C26H26FeO2: 426.1282, found: 426.1262.
Ethyl (all-E)-3-[p-(2’-ferrocenylethen-1’-yl)phenyl]prop-2-enoate:
Red oil (96% yield, Scheme 1 step d): 1H NMR (400 MHz, CDCl3):
d=1.34 (t, J=7.1 Hz, 3H, CH3-ester), 4.15 (s, 5H, H-Fc), 4.27 (q, J=
7.1 Hz, 2H, CH2-ester), 4.32 (s, 2H, H-Fc), 4.49 (s, 2H, H-Fc), 6.42 (d,
J=16.1 Hz, 1H, CH=), 6.68 (d, J=16.1 Hz, 1H, CH=), 6.95 (d, J=
16.1 Hz, 1H, CH=), 7.43 (d, J=8.3 Hz, 2H, Har), 7.49 (d, J=8.3 Hz,
2H, Har), 7.67 ppm (d, J=16.1 Hz, 1H, CH=); IR (film): n˜ =3089
(m), 3027 (m), 2958 (m), 2925 (m), 2854 (m), 1708 (CO, s), 1630 (C=
C, s), 1597 (s), 1313 (m), 1264 (m), 1206 (m), 1172 (s), 1039 (m),
814 cmÀ1 (m); MS (EI, 70 eV) m/z (%): 386 (100) [M]+; HRMS: m/z
[M]+ calcd for C23H22FeO2: 386.0969, found: 386.0952.
(all-E)-5-[p-(2’-ferrocenylethen-1’-yl)phenyl]-3-methylpenta-2,4-
dienoic acid 4: Red solid (25% total yield, Scheme 1 steps a,b,c,e,f):
1
mp: >2908C (10% EtOAc/hexane); H NMR (400 MHz, [D6]DMSO):
(all-E)-3-[p-(2’-ferrocenylethen-1’-yl)phenyl]prop-2-enoic acid (1):
Red solid (24% total yield, Scheme 1 steps a,b,c,d,f): mp: >2908C
(10% EtOAc/hexane); H NMR (400 MHz, CDCl3): d=4.15 (s, 5H, H-
Fc), 4.33 (s, 2H, H-Fc), 4.49 (s, 2H, H-Fc), 6.44 (d, J=16.0 Hz, 1H,
CH=), 6.69 (d, J=16.0 Hz, 1H, CH=), 6.97 (d, J=16.0 Hz, 1H, CH=),
7.45 (d, J=8.3 Hz, 2H, Har), 7.51 (d, J=8.3 Hz, 2H, Har), 7.75 ppm
(d, J=16.0 Hz, 1H, CH=); IR (film): n˜ =2500–3000 (br, OÀH+CÀH),
1677 (CO, s), 1623 (C=C, s), 1594 (s), 1426 (m), 1325 (m), 1223 (m),
1176 (m), 955 cmÀ1 (m); MS (EI, 70 eV) m/z (%): 358 (100) [M]+;
HRMS: m/z [M]+ calcd for C21H18FeO2: 358.0656, found: 358.0654.
d=2.32 (s, 3H, CH3), 4.15 (s, 5H, H-Fc), 4.34 (s, 2H, H-Fc), 4.57 (s,
2H, H-Fc), 5.96 (s, 1H, H-2), 6.77 (d, J=15.8 Hz, 1H, CH=), 7.00 (d,
J=15.8 Hz, 1H, CH=), 7.04 (d, J=15.8 Hz, 1H, CH=), 7.06 (d, J=
15.8 Hz, 1H, CH=), 7.49 (d, J=8.2 Hz, 2H, Har), 7.56 ppm (d, J=
8.2 Hz, 2H, Har); IR (film): n˜ =2750–3250 (br, OÀH+CÀH), 1667
(CO, s), 1587 (C=C, s), 1288 (s), 1250 (s), 1173 (s), 957 (s), 806 cmÀ1
(s); MS (EI, 70 eV) m/z (%): 398 (100) [M]+; HRMS: m/z [M]+ calcd
for C24H22FeO2: 398.0969, found: 398.0965.
1
Ethyl (all-E)-3-{p-[2’-(9’’-phenanthryl)ethen-1’-yl]phenyl}prop-2-
enoate: Yellow solid (71% yield, Scheme 1 step d): mp: 1598C;
1H NMR (400 MHz, CDCl3): d=1.36 (t, J=7.1 Hz, 3H, CH3), 4.29 (q,
J=7.1 Hz, 2H, CH2), 6.47 (d, J=16.0 Hz, 1H, CH=), 7.20 (d, J=
16.0 Hz, 1H, CH=), 7.50–8.00 (m, 12H, Har+2CH=), 8.23 (d, J=
7.9 Hz, 1H, Har), 8.66 (d, J=7.9 Hz, 1H, Har), 8.73 ppm (d, J=
7.9 Hz, 1H, Har); 13C NMR (100 MHz, CDCl3): d=14.3 (q), 60.5 (t),
118.0 (d), 122.5 (d), 123.2 (d), 124.5 (d), 124.8 (d), 126.6 (d), 126.7
(d, 2ꢁ), 126.8 (d), 127.1 (d, 2ꢁ), 127.6 (d), 128.5 (d, 2ꢁ), 128.7 (d),
130.3 (s), 130.5 (s), 130.6 (s), 131.2 (d), 131.7 (s), 133.6 (s), 133.9 (s),
139.5 (s), 144.0 (d), 167.0 ppm (s); IR (film): n˜ =3028 (m), 2980 (CÀ
H, m), 1707 (CO, s), 1632 (C=C, s), 1600 (m), 1309 (m), 1262 (m),
1205 (m), 1175 (s), 1038 (m), 963 (m), 747 (m), 723 (m) cmÀ1; MS
(EI, 70 eV) m/z (%): 378 (100) [M]+, 349 (22), 303 (20); HRMS: m/z
[M]+ calcd for C27H22O2: 378.1620, found: 378.1621.
Methyl (1E,3E)-p-(4-ferrocenylbuta-1,3-dien-1-yl)benzoate: Red
solid (68% yield, Scheme 1 step a): mp: 188–1898C (dec., 5%
EtOAc in hexane); 1H NMR (400 MHz, CDCl3): d=3.91 (s, 3H,
COOCH3), 4.14 (s, 5H, H-Fc), 4.31 (s, 2H, H-Fc), 4.43 (s, 2H, H-Fc),
6.45–6.65 (m, 3H, CH=), 6.94 (dd, J=15.6 Hz, J=9 Hz, 1H, CH=),
7.45 (d, J=8.4 Hz, 2H, Har), 7.98 ppm (d, J=8.4 Hz, 2H, Har);
13C NMR (100 MHz, CDCl3): d=52.0 (q), 67.1 (d, 2ꢁ), 69.4 (d, 7ꢁ),
82.6 (s), 125.8 (d, 2ꢁ), 126.5 (d), 128.2 (s), 128.6 (d), 130.0 (d, 2ꢁ),
132.3 (d), 133.7 (d), 142.4 (s), 167.0 ppm (s, COOH); IR (film): n˜ =
3095 (m), 2946 (m), 1711 (CO, s), 1598 (C=C, m), 1280 (s), 1107 (m),
993 cmÀ1 (m); MS (EI, 70 eV) m/z (%): 372 (100) [M]+, 191 (17), 170
(16); HRMS: m/z [M]+ calcd for C22H20FeO2: 372.0813, found:
372.0811.
(1E,3E)-p-(4-ferrocenylbuta-1,3-dien-1-yl)benzoic acid 2: Red
solid (54% total yield, Scheme 1 steps d,b,c,a,f): mp: >2908C (10%
EtOAc in hexane); 1H NMR (400 MHz, [D6]DMSO): d=4.14 (s, 5H,
H-Fc), 4.35 (s, 2H, H-Fc), 4.54 (s, 2H, H-Fc), 6.59 (d, J=15.4 Hz, 1H,
CH=), 6.64 (dd, J=15.4 Hz, J=9.0 Hz, 1H, CH=), 6.68 (d, J=
15.4 Hz, 1H, CH=), 7.12 (dd, J=15.4 Hz, J=9.0 Hz, 1H, CH=), 7.57
(d, J=8.3 Hz, 2H, Har), 7.88 ppm (d, J=8.3 Hz, 2H, Har); 13C NMR
(100 MHz, [D6]DMSO): d=66.8 (d, 2ꢁ), 69.0 (d, 5ꢁ), 69.1 (d, 2ꢁ),
82.4 (s), 125.7 (d, 2ꢁ), 126.4 (d), 128.2 (d), 128.7 (s), 129.6 (d, 2ꢁ),
132.3 (d), 133.7 (d), 141.8 (s), 166.9 ppm (s, COOH); IR (film): n˜ =
2500–3250 (br, OÀH+CÀH), 1675 (CO, s), 1593 (C=C, s), 1418 (s),
(all-E)-3-{p-[2’-(9’’-phenanthryl)ethen-1’-yl]phenyl}prop-2-enoic
acid 5: Yellow solid (38% total yield, Scheme 1 steps a,b,c,d,f): mp:
259–2658C (dec.); 1H NMR (400 MHz, [D6]DMSO): d=6.59 (d, J=
16.2 Hz, 1H, CH=), 7.42 (d, J=16.2 Hz, 1H, CH=), 7.63 (d, J=
16.2 Hz, 1H, CH=), 7.65–7.80 (m, 6H, Har), 7.86 (d, J=8.3 Hz, 2H,
Har), 8.05 (dd, J=8.2 Hz, J=2 Hz, 1H, Har), 8.19 (d, J=16.2 Hz, 1H,
CH=), 8.22 (s, 1H, Har), 8.48 (dd, J=8.2 Hz, J=2.1 Hz, 1H, Har),
8.83 (d, J=8.2 Hz, 1H, Har), 8.90 ppm (dd, J=8.2 Hz, J=2.1 Hz, 1H,
Har); 13C NMR (100 MHz, [D6]DMSO): d=118.9 (d), 122.7 (d), 123.3
(d), 123.9 (d), 124.6 (d), 126.7 (d), 126.8 (d), 126.9 (d), 126.9 (d),
127.0 (d), 127.3 (d, 2ꢁ), 128.6 (d, 3ꢁ), 129.6 (s), 129.9 (s), 130.0 (s),
ChemMedChem 2011, 6, 1518 – 1529
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
1525

D. Ivanova, H. Gronemeyer, A. R. de Lera
MED
131.0 (d), 131.2 (s), 132.9 (s), 133.6 (s), 139.0 (s), 143.4 (d),
167.6 ppm (s); IR (film): n˜ =2500–3500 (br, OÀH+CÀH), 1678 (CO,
s), 1623 (C=C, s), 1597 (s), 1422 (s), 1311 (s), 1282 (s), 1214 (s), 948
(s), 741 (s), 717 cmÀ (s)1; MS (EI, 70 eV) m/z (%): 350 (100) [M]+, 303
(19), 202 (16); HRMS: m/z [M]+ calcd for C25H18O2: 350.1307, found:
350.1305.
7.14 (d, J=16.1 Hz, 1H, CH=), 7.40 (d, J=16.1 Hz, 1H, CH=),
7.60–7.80 (m, 6H, Har), 7.82 (d, J=8.3 Hz, 2H, Har), 8.05 (dd, J=
6.9 Hz, J=2.1 Hz, 1H, Har), 8.15 (d, J=16.1 Hz, 1H, CH=), 8.21 (s,
1H, Har), 8.48 (dd, J=7.3 Hz, J=1.8 Hz, 1H, Har), 8.83 (dd, J=
7.3 Hz, J=1.8 Hz, 1H, Har), 8.90 ppm (dd, J=7.3 Hz, J=1.8 Hz, 1H,
Har); 13C NMR (100 MHz, [D6]DMSO): d=13.2 (q), 120.6 (d), 122.7
(d), 123.3 (d), 123.8 (d), 124.6 (d), 125.8 (d), 126.8 (d, 2ꢁ), 126.9 (d),
127.0 (d), 127.3 (d, 2ꢁ), 127.4 (d, 2ꢁ), 128.6 (d), 129.6 (s), 129.9 (s),
130.0 (s), 131.3 (s+d), 131.8 (d), 133.1 (s), 133.3 (d), 135.7 (s), 137.4
(s), 151.2 (s), 167.8 ppm (s); IR (film): n˜ =2500–3500 (br, OÀH+CÀ
H), 1692 (CO,s), 1588 (C=C, s), 1415 (s), 1259 (s), 1173 (s), 957 (s),
852 (s), 740 (s), 719 cmÀ1 (s); MS (EI, 70 eV) m/z (%): 390 (100) [M]+,
203 (34); HRMS: m/z [M]+ calcd for C28H22O2: 390.1620, found:
390.1624; Anal. calcd for C28H22O2: C 86.13, H 5.68, found: C 85.98,
H 5.68.
Methyl (1E,3E)-p-[4-(9’-phenanthryl)buta-1,3-dien-1-yl]benzoate:
Orange solid (100% yield, Scheme 1 step a): mp: 179–1808C;
1H NMR (400 MHz, CDCl3): d=3.93 (s, 3H, COOCH3), 6.79 (d, J=
15.4 Hz, 1H, CH=), 7.13 (dd, J=15.4 Hz, J=10.3 Hz, 1H, CH=),
7.25 (dd, J=15.4 Hz, J=10.3 Hz, 1H, CH=), 7.50–7.80 (m, 7H,
Har+CH=), 7.91 (dd, J=7.8 Hz, J=1.7 Hz, 1H, Har), 7.96 (s, 1H,
Har), 8.03 (d, J=8.3 Hz, 2H, Har), 8.24 (dd, J=7.8 Hz, J=1.7 Hz, 1H,
Har), 8.67 (d, J=7.8 Hz, 1H, Har), 8.75 ppm (d, J=7.8 Hz, 1H, Har);
13C NMR (100 MHz, CDCl3): d=52.0 (q), 122.5 (d), 123.2 (d), 124.3
(d), 124.5 (d), 126.2 (d, 2ꢁ), 126.6 (d), 126.7 (d, 2ꢁ), 126.9 (d), 128.7
(d), 128.9 (s), 130.0 (d, 2ꢁ), 130.3 (s), 130.4 (s), 130.5 (s), 131.6 (d),
131.7 (s), 131.9 (d, 2ꢁ), 131.9 (d), 133.2 (s), 141.8 (s), 166.9 ppm (s);
IR (film): n˜ =3028 (w), 2949 (CÀH, w), 1713 (CO, s), 1600 (C=C, m),
1434 (m), 1413 (m), 1275 (s), 1175 (m), 1106 (m), 984 (m) cmÀ1; MS
(EI, 70 eV) m/z (%): 364 (100) [M]+, 305 (31), 215 (66), 69 (17);
HRMS: m/z [M]+ calcd for C26H20O2: 364.1463, found: 364.1463;
Anal. calcd for C26H20O2: C 85.69, H 5.53, found: C 85.63, H 5.59.
Ethyl (2E,4E)-5-(N-ethylcarbazol-3’-yl)-3-methyl-penta-2,4-dien-
oate: Yellow oil (90% yield, Scheme 1 step e): 1H NMR (400 MHz,
CDCl3): d=1.31 (t, J=7.2 Hz, 3H, CH3), 1.42 (t, J=7.2 Hz, 3H, CH3),
2.46 (s, 3H, 3-CH3), 4.20 (q, J=7.2 Hz, 2H, CH2), 4.34 (q, J=7.2 Hz,
2H, CH2), 5.91 (s, 1H, H-2), 6.86 (d, J=16.0 Hz, 1H, CH=), 7.14 (d,
J=16.0 Hz, 1H, CH=), 7.24 (t, J=7.9 Hz, 1H, Har), 7.35 (d, J=
7.9 Hz, 1H, Har), 7.39 (d, J=7.9 Hz, 1H, Har), 7.47 (t, J=7.9 Hz, 1H,
Har), 7.60 (d, J=7.9 Hz, 1H, Har), 8.09 (d, J=7.9 Hz, 1H, Har),
8.17 ppm (s, 1H, Har); IR (film): n˜ =3046 (m), 2978 (m), 2934 (m),
2898 (m), 1703 (CO,s), 1594 (C=C, s), 1473 (m), 1384 (m), 1347 (m),
1235 (s), 1150 (s), 960 (m), 748 (m) cmÀ1; MS (EI, 70 eV) m/z (%):
333 (44) [M]+, 260 (100), 245 (16), 230 (16), 205 (26); HRMS: m/z
[M]+ calcd for C22H23NO2: 333.1729, found: 333.1728.
(1E,3E)-p-[4-(9’-phenanthryl)buta-1,3-dien-1-yl]benzoic acid 6:
Orange solid (61% total yield, Scheme 1 steps d,b,c,a,f): mp: 262–
1
2668C; H NMR (400 MHz, [D6]DMSO): d=6.93 (d, J=15.3 Hz, 1H,
CH=), 7.31 (dd, J=15.3 Hz, J=10.5 Hz, 1H, CH=), 7.49 (dd, J=
15.3 Hz, J=10.5 Hz, 1H, CH=), 7.60–7.80 (m, 7H, 6Har+CH=),
7.95 (d, J=8.3 Hz, 2H, Har), 8.04 (dd, J=7.2 Hz, J=2.0 Hz, 1H, Har),
8.20 (s, 1H, Har), 8.30–8.40 (m, 1H, Har), 8.82 (dd, J=7.2 Hz, J=
2.0 Hz, 1H, Har), 8.86–8.94 ppm (m, 1H, Har); 13C NMR (100 MHz,
[D6]DMSO): d=122.7 (d), 123.4 (d), 123.7 (d), 124.0 (d), 126.2 (d,
2ꢁ), 126.8 (d), 126.9 (d), 127.0 (d), 127.0 (d), 128.7 (d), 129.4 (s),
129.6 (s), 129.7 (s), 129.8 (d, 2ꢁ), 129.9 (s), 130.8 (d), 131.2 (s), 131.7
(d), 132.0 (d), 132.2 (d), 132.4 (s), 141.3 (s), 167.0 ppm (s); IR (film):
n˜ =2500–3250 (br, OÀH+CÀH), 1731 (w), 1682 (CO, s), 1595 (C=C,
m), 1418 (m), 1285 (s), 970 (m), 756 (s) cmÀ1; MS (EI, 70 eV) m/z (%):
350 (100) [M]+, 305 (22), 215 (68); HRMS: m/z [M]+ calcd for
C25H18O2: 350.1307, found: 350.1308.
(2E,4E)-5-(N-ethylcarbazol-3’-yl)-3-methylpenta-2,4-dienoic acid
8: Yellow solid (81% total yield, Scheme 1 steps e,f): mp: 198–
1998C; 1H NMR (400 MHz, [D6]DMSO): d=1.32 (t, J=7.1 Hz, 3H,
CH3), 2.38 (s, 3H, 3-CH3), 4.45 (q, J=7.1 Hz, 2H, CH2), 5.94 (s, 1H, H-
2), 7.09 (d, J=16.2 Hz, 1H, CH=), 7.15–7.30 (m, 2H, Har+CH=),
7.47 (t, J=8.1 Hz, 1H, Har), 7.61 (d, J=8.1 Hz, 2H, Har), 7.74 (dd,
J=8.1 Hz, J=1.4 Hz, 1H, Har), 8.16 (d, J=8.1 Hz, 1H, Har),
8.41 ppm (s, 1H, Har); 13C NMR (100 MHz, [D6]DMSO): d=13.3 ( q),
13.6 ( q), 37.0 ( t), 109.3 (d), 109.3 (d), 118.9 (d), 119.1 (d), 119.5 (d),
120.4 (d), 122.1 (s), 122.5 (s), 125.1 (d), 126.0 (d), 127.2 (s), 128.9 (d),
135.0 (d), 139.7 (s), 139.9 (s), 151.9 (s), 167.9 ppm (s); IR (film): n˜ =
2250–3250 (br, OÀH+CÀH), 1675 (CO, m), 1570 (C=C, s), 1340 (m),
1281 (m), 1252 (m), 1228 (m), 1171 (s), 922 cmÀ1 (m), 720 (s); MS
(EI, 70 eV) m/z (%): 305 (45) [M]+, 260 (100), 245 (24), 230 (28);
HRMS: m/z [M]+ calcd for C20H19NO2: 305.1416, found: 305.1416;
Anal. calcd for C20H19NO2: C 78.66, H 6.27, N 4.59, found: C 78.62, H
6.33, N 4.40.
Ethyl (all-E)-5-{p-[2’-(9’’-phenanthryl)ethen-1’-yl]phenyl}-3-meth-
ylpenta-2,4-dienoate: Yellow solid (75% yield, Scheme 1 step e):
mp: 1468C; 1H NMR (400 MHz, CDCl3): d=1.32 (t, J=7.1 Hz, 3H,
CH3 ester), 2.44 (d, J=0.9 Hz, 3H, CH3), 4.22 (q, J=7.1 Hz, 2H, CH2
ester), 5.94 (s, 1H, H-2), 6.86 (d, J=16.0 Hz, 1H, CH=), 6.97 (d, J=
16.0 Hz, 1H, CH=), 7.23 (d, J=16.0 Hz, 1H, CH=), 7.52 (d, J=
8.3 Hz, 2H, Har), 7.55–7.75 (m, 6H, Har), 7.85–7.95 (m, 2H, Har+
CH=), 7.98 (s, 1H, Har), 8.26 (dd, J=8.0 Hz, J=1.6 Hz, 1H, Har),
8.68 (d, J=8.0 Hz, 1H, Har), 8.75 ppm (dd, J=8.0 Hz, J=1.6 Hz, 1H,
Har); 13C NMR (100 MHz, CDCl3): d=13.8 (q), 14.4 (q), 59.8 (t),
120.0 (d), 122.5 (d), 123.2 (d), 124.5 (d), 124.6 (d), 126.6 (d), 126.6
(d), 126.7 (d), 126.7 (d), 126.8 (d), 127.1 (d, 2ꢁ), 127.5 (d, 2ꢁ), 128.7
(d), 130.3 (s), 130.5 (s), 130.7 (s), 131.4 (d), 131.8 (s), 131.9 (d), 133.6
(d), 133.8 (s), 135.9 (s), 137.9 (s), 152.0 (s), 167.0 ppm (s); IR (film):
n˜ =3028 (w), 2980 (w), 1704 (CO, s), 1597 (C=C, m), 1238 (m), 1152
(s), 959 cmÀ1 (m); MS (EI, 70 eV) m/z (%): 418 (100) [M]+, 345 (17),
203 (44); HRMS: m/z [M]+calcd for C30H26O2: 418.1933, found:
418.1925.
Methyl
(E)-p-[2-(N-ethylcarbazol-3’-yl)ethen-1-yl]benzoate:
Yellow solid (100% yield, Scheme 1 step a): mp: 1238C; 1H NMR
(400 MHz, CDCl3): d=1.46 (t, J=7.2 Hz, 3H, CH3), 3.93 (s, 3H,
COOCH3), 4.38 (q, J=7.2 Hz, 2H, CH2), 7.17 (d, J=15.4 Hz, 1H,
CH=), 7.26 (d, J=15.4 Hz, 1H, CH=), 7.35–7.55 (m, 4H, Har), 7.61
(d, J=8.3 Hz, 2H, Har), 7.70 (dd, J=8.3 Hz, J=1.2 Hz, 1H, Har), 8.04
(d, J=8.3 Hz, 2H, Har), 8.14 (d, J=8.3 Hz, 1H, Har), 8.26 ppm (s,
1H, Har); 13C NMR (100 MHz, CDCl3): d=13.8 (q), 37.7 (t), 52.0 (q),
108.7 (d), 108.7 (d), 119.2 (d), 119.2 (d), 120.5 (d), 122.9 (s), 123.4 (s),
124.6 (d), 124.8 (d), 125.9 (d, 2ꢁ), 126.0 (d), 127.9 (s), 128.2 (s),
130.0 (d, 2ꢁ), 132.3 (d), 140.1 (s), 140.4 (s), 142.6 (s), 167.0 ppm (s);
IR (film): n˜ =3022 (w), 2976 (w), 2949 (w), 2889 (w), 1713 (CO, s),
1596 (C=C, s), 1473 (m), 1280 (s), 1233 (m), 1178 (m), 1107 (m), 960
(m), 747 cmÀ1 (m); MS (EI, 70 eV) m/z (%): 355 (100) [M]+, 154 (14);
HRMS: m/z [M]+ calcd for C24H21NO2: 355.1572, found: 355.1583;
Anal. calcd for C24H21NO2: C 81.10, H 5.96, N 3.94, found: C 81.33, H
6.00, N 3.87.
(all-E)-5-{p-[2’-(9’’-phenanthryl)ethen-1’-yl]phenyl}-3-methylpen-
ta-2,4-dienoic acid 7: Yellow solid (40% total yield, Scheme 1
1
steps a,b,c,e,f): mp: 226–2288C; H NMR (400 MHz, [D6]DMSO): d=
2.35 (s, 3H, CH3), 6.00 (s, 1H, H-2), 7.07 (d, J=16.1 Hz, 1H, CH=),
1526
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 1518 – 1529

Ferrocene and N-Butylcarbazole Retinoids
(E)-p-[2-(N-ethylcarbazol-3’-yl)ethen-1-yl]benzoic acid 9. Yellow
solid (94% total yield, Scheme 1 steps a,f): mp: 274–2778C; 1H NMR
(400 MHz, [D6]DMSO): d=1.33 (t, J=7.1 Hz, 3H, CH3), 4.46 (q, J=
7.1 Hz, 2H, CH2), 7.24 (t, J=7.3 Hz, 1H, Har), 7.35 (d, J=16.4 Hz,
1H, CH=), 7.48 (t, J=7.3 Hz, 1H, Har), 7.59 (d, J=16.4 Hz, 1H,
CH=), 7.62 (d, J=8.2 Hz, 1H, Har), 7.64 (d, J=8.2 Hz, 1H, Har), 7.73
(d, J=8.3 Hz, 2H, Har), 7.79 (dd, J=8.1 Hz, J=1.3 Hz, 1H, Har), 7.95
(d, J=8.3 Hz, 2H, Har), 8.19 (d, J=8.1 Hz, 1H, Har), 8.45 ppm (s,
1H, Har); 13C NMR (100 MHz, [D6]DMSO): d=13.7 (q), 37.0 (t), 109.3
(d), 109.3 (d), 119.0 (d, 2ꢁ), 120.4 (d), 122.1 (s), 122.5 (s), 124.4 (d),
124.9 (d), 125.9 (d, 3ꢁ), 127.7 (s), 128.7 (s), 129.7 (d, 2ꢁ), 132.0 (d),
139.5 (s), 139.9 (s), 142.1 (s), 167.1 ppm (s); IR (film): n˜ =2250–3250
(br, OÀH+CÀH), 1679 (CO, s), 1595 (C=C, s), 1473 (m), 1424 (m),
1294 (s), 1231 (s), 1178 (s), 961 (m), 864 (m), 799 (m), 741 cmÀ1 (s);
MS (EI, 70 eV) m/z (%): 341 (100) [M]+, 326 (64), 97 (26), 83 (31), 69
(40); HRMS: m/z [M]+ calcd for C23H19NO2: 341.1416, found:
341.1409.
140.9 (s), 142.6 (s), 167.0 ppm (s); IR (film): n˜ =3051 (m), 3022 (m),
2954 (m), 2926 (m), 2870 (m), 1710 (CO, s), 1591 (s), 1466 (m), 1335
(m), 1278 (s), 1176 (s), 1102 (s), 960 (s), 802 (s), 741 cmÀ1 (s); MS (EI,
70 eV) m/z (%): 383 (100) [M]+, 340 (87), 280 (16), 154 (48); HRMS:
m/z [M]+ calcd for C26H25NO2: 383.1885, found: 383.1881; Anal.
calcd for C26H25NO2: C 81.43, H 6.57, N 3.65, found: C 81.46, H 6.55,
N 3.62.
(E)-p-[2-(N-butylcarbazol-3’-yl)ethen-1-yl]benzoic acid 11: Yellow
solid (75% total yield, Scheme 1 steps a,f): mp: 2538C (dec.);
1H NMR (400 MHz, [D6]DMSO): d=0.89 (t, J=7.1 Hz, 3H, CH3),
1.25–1.40 (m, 2H, CH2), 1.70–1.85 (m, 2H, CH2), 4.41 (t, J=7.1 Hz,
2H, CH2), 7.23 (t, J=7.5 Hz, H, Har), 7.35 (d, J=16.6 Hz, H, CH=),
7.47 (t, J=7.5 Hz, H, Har), 7.52–7.67 (m, 3H, 2Har+CH=), 7.72 (d,
J=8.2 Hz, 2H, Har), 7.78 (d, J=8.2 Hz, H, Har), 7.95 (d, J=8.2 Hz,
2H, Har), 8.18 (d, J=8.2 Hz, H, Har), 8.44 ppm (s, H, H-4’); 13C NMR
(100 MHz, [D6]DMSO): d=13.6 (q), 19.67 (t), 30.6 (t), 42.0 (t), 109.4
(d), 109.5 (d), 118.9 (d), 118.9 (d), 120.3 (d), 122.0 (s), 122.4 (s), 124.4
(d), 124.9 (d), 125.9 (d, 3ꢁ), 127.6 (s), 128.7 (s), 129.7 (d, 2ꢁ), 132.0
(d), 140.0 (s), 140.4 (s), 142.0 (s), 167.1 ppm (s); IR (film): n˜ =2400–
3200 (br, OÀH+CÀH), 1676 (CO, s), 1590 (C=C, s), 1488 (m), 1420
(s), 1315 (s), 1289 (s), 1176 (s), 942 (s), 863 (m), 800 (s), 739 cmÀ1 (s);
MS (EI, 70 eV) m/z (%): 369 (86) [M]+, 326 (68), 148 (100), 83 (15),
69 (18); HRMS: m/z [M]+ calcd for C25H23NO2: 369.1729, found:
369.1746; Anal. calcd for C25H23NO2: C 81.27, H 6.27, N 3.79, found:
C 81.01, H 6.45, N 3.64.
Methyl
(E)-p-[2-(N-propylcarbazol-3’-yl)ethen-1-yl]benzoate:
1
Yellow solid (74% yield, Scheme 1 step a): mp: 144–1458C; H NMR
(400 MHz, CDCl3): d=0.99 (t, J=7.3 Hz, 3H, CH3), 1.85–2.00 (m,
2H, CH2), 3.93 (s, 3H, COOCH3), 4.29 (t, J=7.3 Hz, 2H, CH2), 7.17 (d,
J=16.3 Hz, H, CH=), 7.22–7.30 (m, H, Har), 7.35–7.55 (m, 4H,
3Har+CH=), 7.60 (d, J=8.2 Hz, 2H, Har), 7.69 (dd, J=8.2 Hz, J=
2 Hz, H, Har), 8.04 (d, J=8.2 Hz, 2H, Har), 8.13 (d, J=8.2 Hz, H,
Har), 8.26 ppm (s, H, H-4’); 13C NMR (100 MHz, CDCl3): d=11.8 (q),
22.3 (t), 44.7 (t), 52.0 (q), 109.0 (d), 109.0 (d), 119.1 (d), 119.1 (d),
120.4 (d), 122.8 (s), 123.2 (s), 124.6 (d), 124.7 (d), 125.9 (d, 3ꢁ),
127.9 (s), 128.2 (s), 130.0 (d, 2ꢁ), 132.3 (d), 140.6 (s), 140.9 (s), 142.6
(s), 167.0 ppm (s); IR (film): n˜ =3052 (m), 2959 (m), 2872 (m), 1703
(CO, s), 1589 (s), 1465 (s), 1331 (s), 1276 (s), 1221 (s), 1175 (s), 1103
(s), 963 (s), 742 cmÀ1 (s); MS (EI, 70 eV) m/z (%): 369 (100) [M]+, 340
(61); HRMS m/z [M]+ calcd for C25H23NO2: 369.1729, found:
369.1720; Anal. calcd for C25H23NO2: C 81.27, H 6.27, N 3.79, found:
C 81.01, H 6.27, N 3.81.
Methyl
(E)-p-[2-(N-pentylcarbazol-3’-yl)ethen-1-yl]benzoate:
1
Yellow solid (57% yield, Scheme 1 step a): mp: 104–1058C; H NMR
(400 MHz, CDCl3): d=1.85–1.95 (m, 3H, CH3), 1.30–1.45 (m, 4H,
CH2-CH2), 1.85–1.97 (m, 2H, CH2), 3.93 (s, 3H, COOCH3), 4.31 (t, J=
7.2 Hz, 2H, CH2), 7.17 (d, J=16.3 Hz, H, CH=), 7.22–7.30 (m, H,
Har), 7.35–7.55 (m, 4H, 3Har+CH=), 7.61 (d, J=8.2 Hz, 2H, Har),
7.69 (dd, J=8.2 Hz, J=1.7 Hz, H, Har), 8.04 (dd, J=8.2 Hz, J=
1.7 Hz, 2H, Har), 8.13 (d, J=8.2 Hz, H, Har), 8.26 ppm (s, H, H-4’);
13C NMR (100 MHz, CDCl3): d=13.9 (q), 22.4 (t), 28.7 (t), 29.4 (t),
43.2 (t), 52.0 (q), 108.9 (d), 109.0 (d), 119.1 (d), 120.4 (d), 122.8 (s),
123.2 (s), 124.6 (d), 124.7 (d), 124.8 (d), 125.9 (d, 2ꢁ), 127.9 (s),
128.2 (s), 130.0 (d, 2ꢁ), 132.2 (d), 132.3 (d), 140.6 (s), 140.9 (s),
142.6 (s), 167.0 ppm (s); IR (film): n˜ =3023 (m), 2950 (m), 2929 (m),
2865 (w), 1716 (CO, s), 1593 (s), 1469 (m), 1276 (s), 1175 (s), 1102
(s), 945 (m), 795 (m), 738 cmÀ1 (s); MS (EI, 70 eV) m/z (%): 397 (100)
[M]+, 340 (59); HRMS: m/z [M]+ calcd for C27H27NO2: 397.5088,
found: 397.2051; Anal. calcd for C27H27NO2: C 81.58, H 6.85, N 3.52,
found: C 81.44, H 6.91, N 3.50.
(E)-p-[2-(N-propylcarbazol-3’-yl)ethen-1-yl]benzoic
acid
10:
Yellow solid (73% total yield, Scheme 1 steps a,f): mp: 263–48C;
1H NMR (400 MHz, [D6]DMSO): d=0.89 (t, J=7.1 Hz, 3H, CH3),
1.70–1.90 (m, 2H, CH2), 4.38 (t, J=7.1 Hz, 2H, CH2), 7.23 (t, J=
7.5 Hz, H, Har), 7.35 (d, J=16.4 Hz, H, CH=), 7.47 (t, J=7.5 Hz, H,
Har), 7.53–7.68 (m, 3H, 2Har+CH=), 7.72 (d, J=8.2 Hz, 2H, Har),
7.78 (d, J=8.2 Hz, H, Har), 7.95 (d, J=8.2 Hz, 2H, Har), 8.18 (d, J=
8.2 Hz, H, Har), 8.44 ppm (s, H, H-4’); 13C NMR (100 MHz,
[D6]DMSO): d=11.3 (q), 21.8 (t), 43.7 (t), 109.5 (d), 109.6 (d), 118.9
(d), 118.9 (d), 120.3 (d), 122.0 (s), 122.3 (s), 124.4 (d), 124.8 (d), 125.9
(d, 3ꢁ), 127.6 (s), 128.7 (s), 129.7 (d, 2ꢁ), 132.0 (d), 140.1 (s), 140.5
(s), 142.0 (s), 167.1 ppm (s); IR (film): n˜ =2250–3250 (br, OÀH+CÀ
H), 1674 (CO, s), 1593 (C=C, s), 1480 (m), 1422 (m), 1290 (s), 1223
(m), 1174 (m), 1141 (m), 943 (m), 738 cmÀ1 (m); MS (EI, 70 eV) m/z
(%): 355 (100) [M]+, 326 (95); HRMS: m/z [M]+ calcd for C24H21NO2:
355.1572, found: 355.1558; Anal. calcd for C24H21NO2: C 81.10, H
5.96, N 3.94, found: C 81.26, H 6.01, N 3.92.
(E)-p-[2-(N-pentylcarbazol-3’-yl)ethen-1-yl]benzoic
acid
12:
Yellow solid (57% total yield, Scheme 1 steps a,f): mp: 226–2278C
(dec.); 1H NMR (400 MHz, [D6]DMSO): d=0.82 (t, 3H, J=6.8 Hz,
CH3), 1.20–1.40 (m, 4H, 2CH2), 1.70–1.85 (m, 2H, CH2), 4.40 (t, J=
6.8 Hz, 2H, CH2), 7.23 (t, J=7.5 Hz, H, Har), 7.35 (d, J=16.4 Hz, H,
CH=), 7.47 (t, J=7.5 Hz, H, Har), 7.53–7.68 (m, 3H, 2Har+CH=),
7.72 (d, J=8.2 Hz, 2H, Har), 7.78 (d, J=8.2 Hz, H, Har), 7.94 (d, J=
8.2 Hz, 2H, Har), 8.18 (d, J=8.2 Hz, H, Har), 8.44 ppm (s, H, H-4’);
13C NMR (100 MHz, [D6]DMSO): d=13.8 (q), 21.8 (t), 28.2 (t), 28.5
(t), 42.2 (t), 109.4 (d), 109.5 (d), 118.9 (d, 2ꢁ), 120.3 (d), 122.0 (s),
122.4 (s), 124.4 (d), 124.9 (d), 125.9 (d, 3ꢁ), 127.6 (s), 128.7 (s),
129.7 (d, 2ꢁ), 132.0 (d), 140.0 (s), 140.4 (s), 142.0 (s), 167.1 ppm (s);
IR (film): n˜ =2350–3200 (br, OÀH+CÀH), 1675 (CO, s), 1592 (C=C,
s), 1486 (m), 1420 (m), 1289 (s), 1225 (m), 1174 (m), 941 (m),
736 cmÀ1 (s). MS (EI, 70 eV) m/z (%): 383 (100) [M]+, 355 (15), 326
(92); HRMS: m/z [M]+calcd for C26H25NO2: 383.1885, found:
383.1878; Anal. calcd for C26H25NO2: C 81.43, H 6.57, N 3.65, found:
C 81.28, H 6.60, N 3.63.
Methyl
(E)-p-[2-(N-butylcarbazol-3’-yl)ethen-1-yl]benzoate:
1
Yellow solid (75% yield, Scheme 1 step a): mp: 127–1288C; H NMR
(400 MHz,CDCl3): d=0.96 (t, J=7.2 Hz, 3H, CH3), 1.35–1.50 (m, 2H,
CH2), 1.80–1.95 (m, 2H, CH2), 3.93 (s, 3H, COOCH3), 4.32 (t, J=
7.2 Hz, 2H, CH2), 7.17 (d, J=16.3 Hz, 1H,CH=), 7.22–7.30 (m, H,
Har), 7.35–7.55 (m, 4H, 3Har+CH=), 7.60 (d, J=8.1 Hz, 2H, Har),
7.69 (d, J=8.1 Hz, H, Har), 8.04 (d, J=8.1 Hz, 2H, Har), 8.13 (d, J=
8.1 Hz, H, Har), 8.25 ppm (s, H, H-4’); 13C NMR (100 MHz, CDCl3):
d=13.8 (q), 20.5 (t), 31.1 (t), 43.0 (t), 52.0 (q), 108.9 (d), 109.0 (d),
119.1 (d), 119.1 (d), 120.4 (d), 122.8 (s), 123.2 (s), 124.6 (d), 124.8 (d),
125.9 (d, 3ꢁ), 127.9 (s), 128.2 (s), 130.0 (d, 2ꢁ), 132.3 (d), 140.6 (s),
ChemMedChem 2011, 6, 1518 – 1529
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
1527

D. Ivanova, H. Gronemeyer, A. R. de Lera
MED
Methyl
(E)-p-[2-(N-hexylcarbazol-3’-yl)ethen-1-yl]benzoate:
cells. Antibodies for caspases-8, -9, -10, cleaved caspase-3 and BID
were purchased from Cell Signaling Technology, DR5 antibody was
obtained from Sigma, DcR1and DcR2 antibodies were from Chemi-
con International Inc., and TRAIL antibody was obtained from R&D
Systems. All antibodies were used at 1:1000 dilution.
Yellow solid (76% yield, Scheme 1 step a): mp: 98–998C; 1H NMR
(400 MHz, CDCl3): d=0.87 (t, J=7.0 Hz, 3H, CH3), 1.20–1.47 (m,
6H, 3xCH2), 1.80–1.95 (m, 2H, CH2), 3.93 (s, 3H, COOCH3), 4.31 (t,
J=7.0 Hz, 2H, CH2), 7.17 (d, J=16.2 Hz, H, CH=), 7.22–7.30 (m, H,
Har), 7.35–7.55 (m, 4H, 3Har+CH=), 7.61 (d, J=8.2 Hz, 2H, Har),
7.69 (d, J=8.2 Hz, H, Har), 8.04 (d, J=8.2 Hz, 2H, Har), 8.13 (d, J=
8.2 Hz, H, Har), 8.25 ppm (s, H, H-4’); 13C NMR (100 MHz, CDCl3):
d=14.0 (q), 22.5 (t), 26.9 (t), 29.0 (t), 31.6 (t), 43.2 (t), 52.0 (q), 108.9
(d), 109.0 (d), 119.1 (d, 2ꢁ), 120.4 (d), 122.8 (s), 123.3 (s), 124.6 (d),
124.8 (d), 125.9 (d, 3ꢁ), 127.9 (s), 128.2 (s), 130.0 (d, 2ꢁ), 132.3 (d),
140.6 (s), 140.9 (s), 142.6 (s), 167.0 ppm (s); IR (film): n˜ =3045 (m),
2952 (m), 2921 (m), 2853 (w), 1709 (s), 1625 (w), 1594 (s), 1464 (m),
1277 (s), 1178 (s), 1105 (s), 954 (s), 797 (s), 742 (s), 690 (s), 607 cmÀ1
(s); MS (EI, 70 eV) m/z (%): 411 (100) [M]+, 340 (58), 155 (16);
HRMS: m/z [M]+ calcd for C28H29NO2: 411.2198, found: 411.2195;
Anal. calcd for C28H29NO2: C 81.72, H 7.10, N 3.40, found: C 81.71, H
7.13, N 3.41.
NB4 cells[35] were purchased from the DSMZ, Germany. Cells were
cultured in RPMI 1640 supplemented with 10% fetal calf serum
(FCS), glutamine (2 mm), HEPES buffer (25 mm), and gentamycin
(40 mgmLÀ1), in a humidified incubator at 378C and 5% CO2. Every
two days the cells were diluted to 1ꢁ105 cellsmLÀ1 and treated
with the corresponding ligand (at 1 mm final concentration).
HeLa cells, stably transfected with (17m)5-bG-Luc-Neo reporter[26]
and with Gal4-mRARa (or b, g) or Gal4-hRXRb plasmids, were main-
tained in DMEM with 5% FCS, geneticin (G418, 0.8 mgmLÀ1), puro-
mycin (0.3 mgmLÀ1), and gentamycin (40 mgmLÀ1); in case of the
Gal4-hRXRb HeLa cell line, hygromycin (0.2 mgmLÀ1) was also
added. The ligand binding assays were carried out in DMEM with-
out red phenol with 5% charcoal-treated FCS. Luminescence was
determined on a Berthold MicroLumat LB96P luminometer.
(E)-p-[2-(N-hexylcarbazol-3’-yl)ethen-1-yl]benzoic acid 13: Yellow
solid (76% total yield, Scheme 1 steps a,f): mp: 225–2268C (dec.);
1H NMR (400 MHz, [D6]DMSO): d=0.81 (t, J=6.9 Hz, 3H, CH3),
1.10–1.40 (m, 6H, 3CH2), 1.70–1.85 (m, 2H, CH2), 4.40 (t, J=6.9 Hz,
2H, CH2), 7.23 (t, J=7.5 Hz, H, Har), 7.35 (d, J=16.4 Hz, H, CH=),
7.47 (t, J=7.5 Hz, H, Har), 7.52–7.67 (m, 3H, 2Har+CH=), 7.72 (d,
J=8.2 Hz, 2H, Har), 7.78 (d, J=8.2 Hz, H, Har), 7.94 (d, J=8.2 Hz,
2H, Har), 8.18 (d, J=8.2 Hz, H, Har), 8.44 ppm (s, H, H-4’); 13C NMR
(100 MHz, [D6]DMSO): d=13.7 (q), 21.9 (t), 26.0 (t), 28.4 (t), 30.8 (t),
42.2 (t), 109.4 (d), 109.5 (d), 118.9 (d, 2ꢁ), 120.3 (d), 122.0 (s), 122.3
(s), 124.4 (d), 124.8 (d), 125.9 (d, 3ꢁ), 127.6 (s), 128.7 (s), 129.7 (d,
2ꢁ), 132.0 (d), 140.0 (s), 140.4 (s), 142.0 (s), 167.1 ppm (s); IR (film):
n˜ =2300–3200 (br, OÀH+CÀH), 1674 (CO, s), 1590 (C=C, s), 1469
(s), 1416 (m), 1283 (s), 1178 (m), 957 (m), 930 (m), 868 (m), 795 (m),
736 cmÀ1 (s); MS (EI, 70 eV) m/z (%): 397 (100) [M]+, 326 (73);
HRMS: m/z [M]+ calcd for C27H27NO2: 397.2042, found: 397.2036;
Anal. calcd for C27H27NO2: C 81.58, H 6.85, N 3.52, found: C 81.52, H
6.85, N 3.50.
Western blot analysis: Cells (50ꢁ105) were collected by centrifuga-
tion (500 g, 10 min), washed (PBS) and lysed in ice-cold buffer
(10 mm Tris-HCl pH 7.5, 1 mm EDTA, 150 mm NaCl, 1% Nonidet P-
40, 1 mm DTT) containing 1ꢁ protease inhibitor cocktail (PIC tab-
lets complete Roche Diagnostic GmbH, stock solution 25ꢁ concen-
tration), 0.1 mm phenylmethylsulfonyl fluoride (PMSF), and 100 mm
Na3VO4. Cell lysates were clarified (12000 g, 15 min at 48C), and
protein concentrations were determined by Quick StartTM Bradford
Protein Assay Kit (Bio-Rad Laboratories, GmbH).
Cell lysates (60 mg protein per lane) were fractionated by sodium
dodecyl sulfate polyacrylamide gel (12 or 15%) electrophoresis
(SDS-PAGE) and then transferred on nitrocellulose membrane (Pro-
tran BA 85, 0.45 mm pore size, Whatman Schleicher & Schuell,
GmbH). After that, the membranes were blocked with 5% nonfat
dry milk in 1ꢁ PBS for 1 h, incubated with primary antibody (over-
night at 48C) and with secondary horseradish peroxidase (HRP)-
linked antibody in 5% nonfat dry milk in 1ꢁ PBS (1 h at RT). The
membranes were washed (1% Tween in 1ꢁ PBS), dried, and immu-
noreactive proteins were visualized using the Amersham Biosci-
ences enhanced chemiluminescence (ECL) Western blotting detec-
tion system.
(E)-p-(2-ferrocenyl-ethen-1-yl)benzoic acid 14: Red solid (50%
total yield, Scheme 1 steps a,f): mp: >2908C (10% EtOAc in
hexane); H NMR (400 MHz, [D6]DMSO): d=4.15 (s, 5H, H-Fc), 4.36
1
(s, 2H, H-Fc), 4.60 (s, 2H, H-Fc), 6.83 (d, J=16.5 Hz, 1H, CH=), 7.15
(d, J=16.5 Hz, 1H, CH=), 7.59 (d, J=8.2 Hz, 2H, Har), 7.89 ppm (d,
J=8.2 Hz, 2H, Har); 13C NMR (100 MHz, [D6]DMSO): d=67.1 (d, 3ꢁ),
69.0 (d, 5ꢁ), 69.2 (d), 82.4 (s), 124.4 (d), 125.5 (d, 2ꢁ), 128.4 (s),
129.7 (d, 2ꢁ), 130.2 (d), 141.9 (s), 167.1 ppm (s, COOH); IR(film): n˜ =
3250–2750 (br, OH), 3089, 2981, 2672, 2548, 1676 (CO, s), 1596 (C=
C, s), 1423 (s), 1288 (s), 1177 (m), 915 (s), 817 (s), 766 cmÀ1 (s); MS
(EI, 70 eV) m/z (%): 332 (100) [M]+, 165 (32); Anal. calcd for
C19H16FeO2: C 68.70, H 4.85, found: C 68.67, H 4.85.
The following assays were performed as reported in Reference [36]:
RAR/RXR transactivation analysis, differentiation induction assays,
and apoptosis induction assays.
Acknowledgements
Supporting Information contains characterization data for other
intermediate compounds in the synthesis of ligands 1–14.
The work was supported by the European Commission ANTI-
CANCER RETINOIDS, FP5 contract QLK3-CT-2002-02029. The Or-
ganizing Committee and Cold Spring Harbor Laboratory honored
this work with a financial aid award at the “Nuclear Receptors:
Bench to Bedside” International Conference, New York, USA
(2006). The authors thank C. Gaudon, A. Bindler, and M. Lieb for
excellent technical assistance.
Biological assays
Materials and methods: Differentiation and apoptosis induction
assays were performed on a FACScan flow cytometer (Becton Dick-
inson). Antibodies used for differentiation analyses were as follows
(all from BD Pharmingen): PE anti-human CD11c or FITC anti-
human CD14 in the test sample and PE mouse IgG1 or FITC mouse
IgG2a antibodies, respectively, as isotypic control to determine the
nonspecific background staining. Recombinant human Annexin V–
FITC (Caltag Laboratories) was used as marker for apoptotic cells
and PI (propidium iodide, Sigma–Aldrich) as a marker for necrotic
Keywords: antitumor agents · apoptosis · drug discovery ·
receptors · structure–activity relationships
1528
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 1518 – 1529

Ferrocene and N-Butylcarbazole Retinoids
[1] A. Melnick, J. D. Licht, Blood 1999, 93, 3167–3215.
[20] J.-P. Renaud, N. Rochel, M. Ruff, V. Vivat, P. Chambon, H. Gronemeyer, D.
Moras, Nature 1995, 378, 681–689.
[21] W. Bourguet, V. Vivat, J.-M. Wurtz, P. Chambon, H. Gronemeyer, D.
Moras, Mol. Cell 2000, 5, 289–298.
[2] The RARa–PML but not the PML–RARa transcript was detected in one
APL patient: M. Lafage-Pochitaloff, M. Alcalay, V. Brunel, L. Longo, D.
Sainty, J. Simonetti, F. Birg, P. G. Pelicci, Blood 1995, 85, 1169–1174.
[3] L. He, M. Bhaumik, C. Tribioli, E. M. Rego, S. Ivins, A. Zelent, P. P. Pandolfi,
Mol. Cell 2000, 6, 1131–1141.
[4] D. R. Soprano, P. Qin, K. J. Soprano, Annu. Rev. Nutr. 2004, 24, 201–221.
[5] W. Bourguet, P. Germain, H. Gronemeyer, Trends Pharmacol. Sci. 2000,
21, 381–388.
[6] 9-cis-Retinoic acid is thought to be the putative natural pan-RAR/RXR
ligand obtained in vivo by isomerization of its ’pro-hormone’ ATRA.
[7] PPAR: peroxisome proliferator-activated receptor, TR: thyroid hormone
receptor, VDR: vitamin D receptor, LXR: liver X receptor.
[8] Selective RXR modulators decreased glucose levels in mouse models of
non-insulin-dependent diabetes mellitus without suppression of the
thyroid hormone axis and without increase in blood triglycerides: P. Y.
Michellys, R. J. Ardecky, J. H. Chen, D. L. Crombie, G. J. Etgen, M. M. Faul,
A. L. Faulkner, T. A. Grese, R. A. Heyman, D. S. Karanewsky, K. Klausing,
M. D. Leibowitz, S. Liu, D. A. Mais, C. M. Mapes, K. B. Marschke, A. Reifel-
Miller, K. M. Ogilvie, D. Rungta, A. W. Thompson, J. S. Tyhonas, M. F.
Boehm, J. Med. Chem. 2003, 46, 2683–2696.
[22] D. Benbrook, M. M. Madler, L. W. Spruce, P. J. Birckbichler, E. C. Nelson, S.
Subramanian, G. M. Weerasekare, J. B. Gale, M. K. Patterson Jr. , B. Wang,
W. Wang, S. Lu, T. C. Rowland, P. DiSivestro, C. Lindamood III, D. L. Hill,
K. D. Berlin, J. Med. Chem. 1997, 40, 3567–3583.
[23] G. V. Kryshtal, E. P. Serebryakov, Izv. Akad. Nauk. Ser. Khim. 1995, 10,
1867–1885; [Russ. Chem. Bull. 1995, 44, 1785–1804 (Engl. Transl.)].
[24] Compound 3 (Ar=PHT), published in our previous work,[17] activated
RARa and RARb, but was discriminated by RARg (Met272) and RXR. It
triggered differentiation of NB4 cells at a high percentage but with a
low level of CD11c expression, and therefore apoptosis was not in-
duced.
[25] a) S. Green, P. Chambon, Nature 1987, 325, 75–78; b) R. M. Evans, Mol.
Endocrinol. 2005, 19, 1429–1438; c) W. Bourguet, A. R. de Lera, H. Gro-
nemeyer, Meth. Enzymol. 2010, 485, 161–195.
[26] J.-Y. Chen, S. Penco, J. Ostrowski, P. Balaguer, M. Pons, J. E. Starrett, P.
Reczek, P. Chambon, H. Gronemeyer, EMBO J. 1995, 14, 1187–1197.
[27] a) H. Te Riele, E. R. Maandag, A. Clarke, M. Hooper, A. Berns, Nature
1990, 348, 649–651; b) S. Nagpal, S. Friant, H. Nakshatri, P. Chambon,
EMBO J. 1993, 12, 2349–2360.
[9] D. Ivanova, H. Gronemeyer, A. R. de Lera, abstract at the international
conference “Nuclear Receptors: Bench to Bedside”, section: Nuclear Re-
ceptors and Cancer, Cold Spring Harbor Laboratory, New York (USA)
2006, p. 30.
[28] S. J. Martin, J. G. Bradley, T. G. Cotter, Clin. Exp. Immunol. 1990, 79, 448–
453.
[29] L. J. Donato, N. Noy, Cancer Res. 2005, 65, 8193–8199.
[30] B. Domꢆnguez, Y. Pazos, A. R. de Lera, J. Org. Chem. 2000, 65, 5917–
5925.
[31] MnO2 was obtained by a standard procedure: a mixture of MnSO4
(0.200 mol) and KMnO4 (0.133 mol) in 600 mL H2O was stirred for 1 h at
RT, filtered, and washed until the filtrate remained colorless. The solid
precipitate was then dried for two days at RT.
[32] C. W. Brown, S. Liu, J. Klucik, K. D. Berlin, P. J. Brennan, D. Kaur, D. M.
Benbrook, J. Med. Chem. 2004, 47, 1008–1017 (with modifications).
[33] Reactions and Syntheses in the Organic Chemistry Laboratory (Eds.: L. F.
Tietze, T. Eicher), University Science Books, Mill Valley, California, 1989,
Chapter 3.3, p. 330 (with modifications).
[34] The 13C NMR spectrum of this ferrocene compound was not recorded
because of ligand instability and rapid partial isomerization in solution.
[35] a) M. J. S. Roussel, M. Lanotte, Oncogene 2001, 20, 7287–7291; b) G.
Benoit, M. Roussel, F. Pendino, E. Sꢄgal-Bendirdjian, M. Lanotte, Onco-
gene 2001, 20, 7161–7177.
[10] A. Konishi, E. Kawahara, M. Yoshikawa, T. Suzuki, M. Shimoda, M. Shoji,
A. Kojima, Patent JP06059474A2, 1994, 0304.
[11] R. G. Keedwell, Y. Zhao, L. A. Hammond, S. Qin, K.-Y. Tsang, A. Reitmair, Y.
Molina, Y. Okawa, L. I. Atangan, D.-L. Shurland, K. Wen, D. Michael, A.
Wallace, R. Bird, R. A. S. Chandraratna, G. Brown, Cancer Res. 2004, 64,
3302–3312.
[12] J. A. Bandy, H. E. Bunting, M. L. H. Green, S. R. Marder, M. E. Thompson,
D. Bloor, P. V. Kolinsky, R. J. Jones in Organic Materials for Nonlinear
Optics (Eds.: R. A. Hann, D. Bloor), Royal Society of Chemistry, Special
Publication No. 69, Burlington House, London, 1989, pp. 219–224.
[13] E. V. Mironova, A. Yu. Lukin, S. V. Shevyakov, S. G. Alexeeva, V. I. Shvets,
O. V. Demina, A. A. Khodonov, L. V. Khitrina, Biochemistry (Moscow)
2001, 66, 1323–1333.
[14] H. Mayer, W. Bollag, R. Hꢂnni, R. Rꢃegg, Experientia 1978, 34, 1105–
1119.
[15] a) M. Gçrmen, P. Pigeon, S. Top, E. A. Hillard, M. Huchꢄ, C. G. Hartinger,
F. de Montigny, M. A. Plamont, A. Vessiꢅres, G. Joauen, ChemMedChem
2010, 5, 2039–2050; b) A. Nguyen, S. Top, P. Pigeon, A. Vessiꢅres, E. A.
Hillard, M.-A. Plamont, M. Huchꢄ, C. Rigamonti, G. Joauen, Chem. Eur. J.
2009, 15, 684–696; c) S. Top, B. Dauer, J. Vaissermann, G. Joauen, J. Or-
ganomet. Chem. 1997, 541, 355–361.
[36] M. W. Bꢃttner, Ch. Burschka, J. O. Daiss, D. Ivanova, N. Rochel, S. Kam-
merer, C. Peluso-Iltis, A. Bindler, C. Gaudon, P. Germain, D. Moras, H.
Gronemeyer, R. Tacke, ChemBioChem 2007, 8, 1688–1699.
[16] S. R. S. Rudrangi, V. K. Bontha, S. Bethi, M. V. Reddy, IJPRD (International
Journal of Pharmaceutical Research and Development) 2010, 2, 53–60.
[17] D. Ivanova, C. Gaudon, A. Rossin, W. Bourget, H. Gronemeyer, Bioorg.
Med. Chem. 2002, 10, 2099–2102.
Received: February 5, 2011
Revised: March 4, 2011
Published online on April 19, 2011
[18] A. le Maire, C. Teyssier, C. Erb, M. Grimaldi, S. Alvarez, A. R. de Lera, P. Ba-
laguer, H. Gronemeyer, C. A. Royer, P. Germain, W. Bourguet, Nat. Struct.
Mol. Biol. 2010, 17, 801–807.
[19] P. Germain, S. Kammerer, E. Perez, C. Peluso-Iltis, D. Tortolani, F. C. Zusi,
J. Starrett, P. Lapointe, J.-P. Daris, A. Marinier, A. R. de Lera, N. Rochel, H.
Gronemeyer, EMBO Rep. 2004, 5, 877–882.
Please note: Minor changes to this article in ChemMedChem were made
after publication in EarlyView. In particular, Prof. Angel R. de Lera was moved
from the Acknowledgements to become a co-author of the paper. Also, the
addition of a reference necessitated the removal of two sentences from the
discussion. The scientific results and conclusions remain unaffected.
The Editor.
ChemMedChem 2011, 6, 1518 – 1529
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
1529