Higher Carbon Analogues of 1,4-Dihydropyridines as Potent TGFβ/Smad Inhibitors

Article abstract of DOI:10.1002/ejic.201901223

The C to Si and Ge exchange in bioactive compounds has often led to positive changes in the molecular properties, whereby Ge analogues are underrepresented. This is only possible at tetrahedral positions, and it is necessary for the analogue building bloc

Full text of DOI:10.1002/ejic.201901223

A Journal of
Accepted Article
Title: Higher carbon analogues of 1,4-dihydropyridines as potent
TGFβ/Smad inhibitors
Authors: Eva Rebecca Barth, Daniel Längle, Fabian Wesseler,
Christopher Golz, Anna Krupp, Dennis Schade, and Carsten
Strohmann
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Inorg. Chem. 10.1002/ejic.201901223
Link to VoR: http://dx.doi.org/10.1002/ejic.201901223

10.1002/ejic.201901223
European Journal of Inorganic Chemistry
FULL PAPER
Higher carbon analogues of 1,4-dihydropyridines as potent
TGFβ/Smad inhibitors
Eva R. Barth,[a] Daniel Längle,[b,c] Fabian Wesseler, [b,c] Christopher Golz,[a] Anna Krupp,[a] Dennis
Schade[b,c] and Carsten Strohmann*[a]
Abstract: The C to Si and Ge exchange in bioactive compounds has
often led to positive changes in the molecular properties, whereby Ge
analogues are underrepresented. This is only possible at tetrahedral
positions, and it is necessary for the analogue building blocks to
withstand the synthetic conditions. Here, we present the synthesis of
Si and Ge analogues of a TGF inhibiting 1,4-dihydropyridine, which
has a tetrahedral carbon in an important position for its activity. The
molecular properties are compared by the discussion of the single
X-ray crystal structures and the electrostatic potential on the molecule
surface which could play a role in the target interaction. Biological
activities show that the C to Si and Ge exchange goes with full
preservation of the potency, thus representing good starting points to
modify physicochemical features and pharmacokinetics of these
TGF inhibitors.
receptor kinase domains.[2] However, using
a phenotypic
screening in stem cells, the substance class of 1,4-dihydro-
pyridines (DHPs) was identified as TGF inhibitors.[3] Importantly,
these compounds inhibit TGF signaling via stimulation of
proteasomal degradation of the type II TGF receptor. Following
our initial studies on the structure activity relationships of the
identified hit compound 1 (see Fig. 1),[3,4] we substantially
explored the structure activity relationships for this unique class
of TGF inhibitors in a seminal series of reports.[5] Among several
SAR-optimized derivatives, 4’-(tert-butyl) DHP analogue 2 (see
Fig. 1) was highly potent (i.e., IC50 = 0.28 µM).
Introduction
Many processes in the human body are controlled by ligand-
receptor interactions on the surface of cells, which then activate
or inhibit specific signalling pathways to produce a biochemical
response. Pathways involved in the onset and progression of
diseases such as cancer, diabetes and cardiovascular diseases
are of particular interest for medical research. In this context, the
TGF/Smad (Transforming Growth Factor) signalling pathway
often plays a key role as it is involved in cell differentiation, cell
migration and proliferation.[1] To date, several small molecule
TGF inhibitors have been described which mostly target the
Figure 1. Chemical structures of the 1,4-dihydropyridine-class of TGF
inhibitors, with an early hit compound 1 and the potent 4’-tert-butyl analogue 2.[6]
Here, we report on interesting derivatives of 2 containing the
higher carbon analogues silicon and germanium at the 4’ position.
We investigated the influence of these elements to the molecular
structure, its electronic properties (especially the resulting partial
charges) that are crucial for the target interaction, as well as their
biological activity as TGF inhibitors. In addition to fine-tuning of
the DHP’s physicochemical features by the altered electro-
negativity, the possibility of introducing further interesting element
properties into the compounds was also explored. In the past,
silicon analogues of active substances have often shown positive
effects such as increased lipophilicity or blocking of certain
metabolic pathways.[7] In comparison to C/Si isosterism there are
only a few reports on Ge analogues for which increased
lipophilicity and low toxicity were achieved.[8] Here it seems to be
interesting to insert germanium so that the follow-up studies on
metabolism and the detection of degradation products via mass
spectrometry can be facilitated due to its characteristic isotope
pattern.[8f]
[a]
Dr. E. R. Barth, Dr. C. Golz, A. Krupp, Prof. Dr. C. Strohmann*
Department of Inorganic Chemistry, Faculty of Chemistry and
Chemical Biology, Dortmund University of Technology, Otto-Hahn-
Straße 6/6a, 44227 Dortmund, Germany
E-mail: carsten.strohmann@tu-dortmund.de
http://www.carbanion.de
[b]
[c]
D. Längle,+ F. Wesseler,+ Prof. Dr. D. Schade+
Department of Chemistry and Chemical Biology
Technical University Dortmund
Otto-Hahn-Str. 4, 44227 Dortmund (Germany)
D. Längle, Prof. Dr. D. Schade
Department of Pharmaceutical and Medicinal Chemistry
Christian-Albrechts-University Kiel
Gutenbergstr. 76, 24118 Kiel (Germany)
E-mail: schade@pharmazie.uni-kiel.de
https://www.pharmazie.uni-kiel.de/de/pharmazeutische-chemie/prof-
dr-dennis-schade
[+]
Former affiliation and address during this project
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/ejic.201901223
European Journal of Inorganic Chemistry
FULL PAPER
Results and Discussion
To prepare the b-annelated 1,4-dihydropyridines, the general one
pot Hantzsch concept was used. As shown in scheme 1, the
silicon and germanium containing benzaldehyde derivatives 5a
and 5b were prepared as building blocks in two steps starting from
1,4-dibromobenzene and then subjected to the multicomponent
reaction to furnish the desired 1,4-dihydropyridine compounds 6a,
6b and 2, respectively.
Figure 2. Crystal structure of C-DHP 2. The compound crystallizes from ethanol
in the space group P21/c as colorless blocks. Selected bond lengths [Å], angles
[°] and torsion angles [°]: C44–C47 1.533(3), C47–C48 1.535(4), C47–C49
1.537(4), C47–C50 1.531(4), C41–C42 1.388(3), C41–C46 1.396(3), C43–C44
1.390(3), C44–C45 1.399(3), C42–C43 1.389(3),C45–C46 1.381(3), C41–C42–
C43 121.4(2), C41–C42–C43–C44 –0.5(4), C29–C40–C41–C46 –82.1(3), O5–
C28–C29–C30 –170.5(2).
The bond lengths and angles of the tert-butyl group correspond
to the typical values for C−C single bonds.[7] The phenyl ring is
also an ordinary aromatic sp2 system. Two typical structural
features for 1,4-DHPs can also be observed: the flattened boat
conformation of the annelated ring system, and the pseudo axial
position of the substituent at C40, which is thus almost
perpendicular to the ring plane.[5] The angle (N2–C40–C41 107°)
here is even steeper than the corresponding angle in the biphenyl
derivative 1 (118°).[5] The ethyl ester linked to C29 is also
arranged in this plane. In addition, the corresponding ethyl group
is present in a disorder caused by the two cis and trans rotation
isomers of C26 related to C28.
Scheme 1. Synthesis of the 4’-(tert-butyl) bioisosteric 1,4-dihydropyridines 6a,
6b, and 2.
After purifying of the DHP compounds via column chromato-
graphy, the resulting solid substances were crystallized from
ethanol to be used for single crystal X-ray crystallography. The
elementary cell of all three crystal structures contains both
enantiomers because the synthesis was racemic. However, since
they only differ in the absolute configuration, the discussion of the
structures is limited to the (R)-enantiomer. In figure 2, the
molecular structure of the carbon DHP compound 2 is shown.
Crystals were also obtained for the corresponding silicon
compound 6a, the structure of which is shown in figure 3.
This article is protected by copyright. All rights reserved.

10.1002/ejic.201901223
European Journal of Inorganic Chemistry
FULL PAPER
Figure 3. Crystal Structure of Si-DHP 6a. The compound crystallizes from
ethanol in the space group P21/c as colorless blocks. Selected bond lengths [Å],
angles [°] and torsion angles [°]: Si1–C19 1.8751(15), Si1–C22 1.8599(18), Si1–
C23 1.8661(18), Si1–C24 1.8562(19), C16–C17 1.389(2), C16–C21 1.394(2),
C17–C18 1.389(2), C18–C19 1.394(2), C19–C20 1.402(2), C20–C21 1.388(2),
C18–C17–C16 120.9(1), C16–C17–C18–C19 –0.6(2), C4–C15–C16–C21 –
83.4(2), O2–C3–C4–C5 –171.6(1), C3–O2–C2–C1 –79.6(2).
Figure 4. Crystal structure of Ge-DHP 6b. The compound crystallizes from
ethanol in the space group Pbca as colorless blocks. Selected bond lenghts [Å],
angles [°] and torsion angles [°]: Ge1–C19 1.955(3), Ge1–C22 1.936(4), Ge1–
C23 1.945(4), Ge1–C24 1.957(4), C16–C17 1.386(4), C16–C21 1.395(4), C17–
C18 1.388(4), C18–C19 1.385(4), C19–C20 1.399(4), C20–C21 1.378(4), C16–
C17–C18 121.2(3), C16–C17–C18–C19 –0.1(5), C4–C15–C16–C21 –61.0(3),
O2–C3–C4–C5 –174.1(3).
The Si–C distances within the trimethylsilyl group are in the range
of 1.8562(19) (Si1–C24) and 1.8661(18) Å (Si1–C23), which, like
the distance to the phenyl group [1.8751(15) Å (Si1–C19)],
corresponds to typical values of Si–C single bonds.[9a] The angle
between the DHP plane and the substituent at position 4 amounts
to 106° (N1–C15–C16). The disorder of the ethyl ester group here
leads to an almost vertical arrangement of C1 in relation to the
ring plane. The other structural characteristics are the same as
those of the C-DHP compound described above.
The general structural characteristics of the germanium DHP
compound correspond to the lighter homologues. The Ge–C
distances within the trimethylgermyl group are in the range of
1.936(4) (Ge1–C22) and 1.957(4) Å (Ge1–C24), which, as well as
the distance to the phenyl group [1.955(3) Å (Ge1–C19)]
corresponds to typical values of Ge–C single bonds.[9] The angle
between the DHP plane and the substituent at position 4 is slightly
flattened to 114° (N1–C15–C16) and therefore more similar to 1.
The crystal contained a solvent molecule which, however, was
disordered. A cavity was found which matched the volume and
electron density of ethanol and was treated with a solvent mask.
The germanium DHP structure differs from the silicon and carbon
structure in the arrangement of the aromatic plane relative to the
vertical center plane of the dihydropyridine ring. However, these
are probably packing effects, since germanium compound also
crystallizes in another space group and under inclusion of a
solvent molecule.
At last, the molecular structure of germanium compound 6b was
determined, which is shown in figure 4.
On the basis of quantum chemical calculations on the B3LYP/6-
31+G(d) level a free geometry optimization of the three DHP
derivatives was carried out. The Conolly surfaces with a sample
radius of 1.4 A were created and assigned with the electrostatic
potential to detect the areas of the molecules that are able to
interact with the target protein. The results are shown in figure 5.
This article is protected by copyright. All rights reserved.

10.1002/ejic.201901223
European Journal of Inorganic Chemistry
FULL PAPER
Figure 6. Extract from the gas chromatogram of a mixture of compounds 6a, 2
and 6b. Column material: silica, 0.25 mm; eluent: He.
In corresponding HPLC experiments, the same unexpected order
of the three compounds is observed (see Supplementary
Information for further details).
Figure 5. Conolly surfaces (sample radius 1.4 Å), occupied with the electrostatic
potential of carbon, silicon and germanium dihydropyridines 6a, 6b, and 2 in
comparison.
The views of the molecule surfaces show two aspects that were
expected due to the crystal structures and the different
electronegativities of the elements: on the one hand, the
increasing steric demand in the area of the trimethyl element
group from carbon via silicon to germanium. On the other hand, a
slight polarization of the Si-methyl bonds with formation of a
positive partial charge on the silicon, which is less pronounced in
the germanium compound and therefore more similar to the
carbon-containing one. This effect can be explained by the
electronegativities: The EN of carbon and germanium are more
similar to each other than the EN of Silicon.
Additionally, this similarity of carbon and germanium was
observed in chromatographic experiments. Following usual
principles, especially regarding the size of the trimethyl element
groups, a development of chromatographic properties from
carbon via silicon to germanium could be expected. As figure 6
reveals, the silicon compound 6a differs strongly from the carbon
and germanium analogues 2 and 6b, respectively.
Figure 7. Conolly surfaces (sample radius 1.4 Å), occupied with the electrostatic
potential of carbon, silicon and germanium dihydropyridines 6a, 6b, and 2 in
comparison.
This article is protected by copyright. All rights reserved.

10.1002/ejic.201901223
European Journal of Inorganic Chemistry
FULL PAPER
Additionally, there is one aspect that has already been discovered
in our previous studies: a localized, relatively strongly marked
negative potential at C8 of the fused ring system.[5] This charge
can also be seen on the flat underside of the molecule, as shown
in figure 7. The more electronegative and electron-rich oxygen
atoms do not contribute to the development of a negative
electrostatic potential because their electrons are delocalized due
to mesomerism. Furthermore, it is noticeable that a very strong
potential gradient is present on the surface. In the direction of the
center of the molecule, the partial charge changes very rapidly
from negative to clearly positive values. This positive surface is
most pronounced with silicon DHP. This again could be explained
by the electronegativities of the elements.
Table 1. TGF inhibition profiles (IC50 values and maximum inhibition at 5
µM) of 4-substituted, bioisosteric 4’-(tert-butyl) DHP derivatives.
Compound[a]
Ph-DHP 1
C-DHP 2
IC50 (M)
Emax (% inhibition at 5 M )[a]
0.85 ± 0.30[5]
0.28 ± 0.12[5]
0.28 ± 0.05
0.28 ± 0.04
95 ± 2[5]
95 ± 4[5]
98 ± 1
Si-DHP 6a
Ge-DHP 6b
98 ± 1
[a] Data from N = 3-4 independent experiments in a Smad 4-binding element
(SBE4) dual luciferase assay in HEK293T cells (mean ± SD, normalized to
DMSO = 100%).[5]
The biological activity of silicon and germanium DHP was tested
in a Smad4-binding element (SBE4) reporter gene assay in
HEK293T as reported previously (Fig. 8).[5] For direct comparison,
a potent non-selective ALK4,5,7 inhibitor (SB-431542)[10] and the
original DHP 1 were tested. The new silicon (6a) and germanium
(6b) compounds both show almost identical dose-response
curves and behave perfectly bioisosteric to their carbon congener
as they exhibit the same TGF inhibition potency and efficacy, i.e.
with an IC50 of 0.28 µM and maximum pathway inhibition >95%
(Table 1). With that profile they also outperform the original 4’-Ph-
substituted DHP 1.
Conclusions
In conclusion, we present the synthesis of Si and Ge analogues
6a and 6b of the active TGF Inhibitor 2. The investigation of all
three compounds by X-ray analysis showed that, except the
element-carbon bond lengths, the differences are due to packing
effects, and they are located at molecule areas that are flexible in
solution. As the pseudo axial arrangement of the aromatic
substituent at 4-positions is steeper for the C and Si analogues
(compared to 1), but not for the Ge compound, this seems to be
also a packing effect and no indication for activity. Quantum
chemical calculations showed that the carbon and germanium
compound differ less, while the silicon one has a slightly positive
charge at the Si-center and more negative ones at the methyl
groups than the other compounds. This effect is probably based
on the element-specific electronegativities. All three of the
molecules show a negatively charged area at C8, which has also
been detected in active substances of our previous studies.[5]
Importantly, all DHP compounds (2, 6a, 6b) exhibited the same
TGF inhibition profile, underlining the possibility for bioisosteric
replacement. Since Si bioisosters are typically more lipophilic and
at the same time could protect from oxidative metabolism, such
analogues might exhibit a beneficial pharmacokinetic profile. The
same could hold true for germanium analogues. Interestingly,
chromatographic experiments as well as electronic examinations
revealed a greater difference of silicon and germanium compared
to carbon and germanium. This might have influence on a wide
range of bioisosteric replacement considerations, underlining that
germanium should not be generally neglected in such processes.
Figure 8. Representative TGF inhibition curves of higher carbon containing
1,4-DHPs 6a and 6b in comparison to the original DHP 1 and a the ALK4,5,7
inhibitor SB-431542. (Smad 4-binding element (SBE4) dual luciferase assay in
HEK293T cells).[4,5]
Experimental Section
Materials and Methods: All reactions were performed under an Argon
atmosphere in flame-dried Schlenk-type glassware on a vacuum line. The
solvents diethyl ether (Et2O) and tetrahydrofuran (THF) were dried with
Na/benzophenone and distilled under argon. 1H-NMR (internal standard
CDCl3, C6D6), 13C-NMR (internal standard CDCl3, C6D6) and 29Si-NMR
spectra were recorded on a Bruker DRX 300 MHz at room temperature.
The 13C-NMR and 29Si-NMR spectra were recorded 1H-broadband
decoupled ({1H}). The 29Si-NMR are referred to TMS as external standard
and measured via the INEPT pulse sequence. For the assignment of the
This article is protected by copyright. All rights reserved.

10.1002/ejic.201901223
European Journal of Inorganic Chemistry
FULL PAPER
multiplicities, the following abbreviations were used: s = singlet, brs =
broad singlet, d = doublet, t = triplet, m = multiplet. Data were processed
using ACD/NMR Processor Academic Edition (Product Version 12.01).
obtained as a colourless, viscous liquid (yield: 0.87 g, 4.88 mmol, 75%).
1H-NMR (300 MHz, CDCl3):
δ = 0.32 [s, 9H, Si(CH3)3], 7.70 [d,
J(H,H)=8.05 Hz, 2H, Si(CH3)3CCH], 7.85 [d, J(H,H)=8.42 Hz, 2H,
(HCO)CCH], 10.03 (s, 1H, HCO) ppm; 13C{1H}-NMR (75 MHz, CDCl3): δ =
1 [3C, Si(CH3)3], 129 [2C, (HCO)CCH], 134 [2C, Si(CH3)3CCH], 136 [1C,
(HCO)C], 149 [1C, Si(CH3)3C], 193 (1C, HCO) ppm; 29Si{1H}-NMR
(60 MHz, CDCl3): δ = –2.7 ppm; GC/EI-MS [80 °C (1 min) – 300 °C
(5.5 min)] tR = 3.91 min; m/z (%): 178 (22) (M+), 163 (100) [(M-Me)+].
Single-crystal X-ray diffraction analysis was performed on an Oxford
Diffraction Xcalibur S diffractometer using graphite-monochromated Mo-
K radiation ( = 0.71073 Å). Determination of the unit cells and data
collection was carried out at 100 K. The crystal structure were solved with
direct methods (SHELXS-97) and refine against F2 with the full-matrix
least-squares method (SHELXL-97). A multi-scan adsorption correction
using the implemented CrysAlis RED program was employed. The non-
hydrogen atoms were refined anistropicallly.
4-(Trimethylgermyl)benzaldehyde (5b): To
a
solution of (4-
bromophenyl)trimethylgermane (1.67 g, 6.12 mmol) in THF (38 mL) n-
BuLi (2.69 mL, 2.50 M, 6.73 mmol) was added dropwise at –78 °C. The
reaction mixture was stirred at –78 °C for 15 min. After adding
Dimethylformamid (1.34 g, 18.4 mmol) the reaction mixture was slowly
warmed to room temperature and stirred overnight at room temperature.
Then the reaction mixture was quenched by the addition of aqueous
saturated NH4Cl solution. The aqueous phase was extracted with Et2O and
the combined organic phases were washed with aqueous saturated NaCl
solution and dried over Na2SO4. After removal of all volatile components
and column chromatographic purifying (pentane, pentane/Et2O = 100:1 →
50:1) the product could be obtained as a colourless, viscous liquid (yield:
Synthetic Procedures:
(4-Bromophenyl)trimethylsilane (4a): To the solution of 1,4-
dibrombenzene (1.50 g, 6.36 mmol) in Et2O (13 mL) n-BuLi (2.54 mL,
2.50 M, 6.36 mmol) was added dropwise at –78 °C. The reaction mixture
was stirred at –78 °C for 4 h. After adding chlorotrimethylsilane (0.76 g,
7.00 mmol) the reaction mixture was slowly warmed to room temperature
and stirred overnight at room temperature. Then the reaction mixture was
quenched by the addition of H2O. The aqueous phase was extracted with
Et2O and the combined organic phases were washed with aqueous
saturated NaCl solution and dried over Na2SO4. After removal of all volatile
components the product could be obtained as a colourless, viscous liquid
(yield: quantitative). The reaction product was used without further
1
1.05 g, 4.70 mmol, 77%). H-NMR (300 MHz, CDCl3): δ = 0.44 [s, 9H,
Ge(CH3)3], 7.66 [d, J(H,H)=8.05 Hz, 2H, Ge(CH3)3CCH], 7.84 [d,
J(H,H)=8.05 Hz, 2H, (HCO)CCH], 10.01 (s, 1H, HCO) ppm; 13C{1H}-NMR
(75 MHz, CDCl3): δ = –2 [3C, Ge(CH3)3], 129 [2C, (HCO)CCH], 134 [2C,
Ge(CH3)3CCH], 136 [1C, (HCO)C], 152 [1C, Ge(CH3)3C], 193 (1C, HCO)
ppm; GC/EI-MS [80 °C (1 min) – 300 °C (5.5 min)] tR = 4.16 min, m/z (%):
224 (2) (M+), 209 (100) [(M-Me)+], 179 (13) [(M-3Me)+].
1
purification in the following synthesis step. H-NMR (300 MHz, CDCl3): δ
= 0.30 [s, 9H, Si(CH3)3], 7.39-7.43 [m, 2H, Si(CH3)3CCH], 7.50-7.54 (m,
2H, BrCCH) ppm; 13C{1H}-NMR (75 MHz, CDCl3): δ = –1 [3C, Si(CH3)3],
124 (1C, BrC), 131 (2C, BrCCH), 135 [2C, Si(CH3)3CCH], 139 [1C,
Si(CH3)3C] ppm; 29Si{1H}-NMR (60 MHz, CDCl3): δ = –3.4 ppm; GC/EI-MS
[80 °C (1 min) – 300 °C (5.5 min)] tR = 3.82 min, m/z (%): 228 (13) (M+),
213 (99) [(M-Me)+].
General Procedure for the Synthesis of 1,4-dihydropyridine(DHP): To
a solution of dimedone, ammonium acetate and acetylacetone in ethanol
the benzaldehyde and iodine was added. The reaction mixture was stirred
over night at room temperature. After removal of all volatile components
the residue was dissolved in EtOAc. The organic phase was washed with
aqueous saturated Na2S2O3 and aqueous saturated NaCl solution and
dried over Na2SO4. After removal of all volatile components the products
were purified per individual techniques as noted below.
(4-Bromophenyl)trimethylgermane (4b): To the solution of 1,4-dibrom-
benzene (1.50 g, 6.36 mmol) in Et2O (13 mL) n-BuLi (2.54 mL, 2.5 M,
6.36 mmol) was added dropwise at –78 °C. The reaction mixture was
stirred at –78 °C for 4 h. After adding chlorotrimethylgermane (1.10 g,
7.00 mmol) the reaction mixture was slowly warmed to room temperature
and stirred overnight at room temperature. Then the reaction mixture was
quenched by the addition of H2O. The aqueous phase was extracted with
Et2O and the combined organic phases were washed with aqueous
saturated NaCl solution and dried over Na2SO4. After removal of all volatile
components the product could be obtained as a colourless, viscous liquid
(yield: 1.67 g, 6.12 mmol, 96%). The reaction product was used without
further purification in the following synthesis step. 1H-NMR (300 MHz,
CDCl3): δ = 0.39 [s, 9H, Ge(CH3)3], 7.34 [d, J(H,H)=8.05 Hz, 2H,
Ge(CH3)3CCH], 7.49 [d, J(H,H)=8.05 Hz, 2H, BrCCH] ppm; 13C{1H}-NMR
(75 MHz, CDCl3): δ=2 [3C, Ge(CH3)3], 123 (1C, BrC), 131 (2C, BrCCH),
135 [2C, Ge(CH3)3CCH], 141 [1C, Ge(CH3)3C] ppm; GC/EI-MS [80 °C
(1 min) – 300 °C (5.5 min)] tR = 7.71 min, m/z (%): 274 (7) (M+), 259 (100)
[(M-Me)+], 229 (12) [(M-3Me)+].
C-DHP 2: The reaction of dimedone (86.9 mg, 0.62 mmol), ammonium
acetate (67.1, 0.87 mml), acetylacetone (80.7 mg, 0.62 mmol), 4-tert-
butylbenzaldehyde (100 mg, 0.62 mmol), iodine (45.5 mg, 0.18 mmol) in
EtOH (1.24 mL) gave after column chromatographic purification
(pentane/Et2O = 5:1 → 2:1 → 1:1 → 1:2 → 1:5 →1:10) yellowish crystals
(yield: 21 mg, 0.053 mmol, 9%). 1H-NMR (300 MHz, C6D6): δ = 0.76 [s, 3H,
(CO)CH2CCH3], 0.79 [s, 3H, (CO)CH2CCH3], 0.93 [t, J(H,H) = 6.95 Hz, 3H,
OCH2CH3], 1.10 - 1.24 [s, 9H, C(CH3)3], 1.88 (s, 2H, NCCH2), 2.07 [d,
J(H,H) = 4.39 Hz, 2H, (CO)CH2], 2.35 (s, 3H, NCCH3), 3.95 (q, J(H,H)=
7.20 Hz, 2H, OCH2), 5.64 (s, 1H, CH), 6.93 (s, 1H, NH), 7.35 [d, J(H,H) =
7.68 Hz, 2H, C(CH3)3CCHortho], 7.67 [d, J(H,H)
= 7.32 Hz, 2H;
C(CH3)3CCHCHmeta] ppm; 13C{1H}-NMR (75 MHz, C6D6): δ = 14 (1C,
OCH2CH3), 19 (1C, NCCH3), 27 [1C, (CO)CH2CCH3], 29 [1C,
(CO)CH2CCH3], 32 [3C, C(CH3)3], 33 [1C, (CO)CH2C(CH3)2], 34 [1C,
C(CH3)3], 37 [1C, (CO)CCH], 41 (1C, NCCH2), 51 [1C, (CO)CH2], 60 (1C,
OCH2CH3), 107 [1C, NCC(COOEt)], 112 [1C, NCC(CO)], 125 [2C,
(CH3)3CCCH], 128 [2C, (CH3)3CCCHCH], 144 [2C, (CH3)3CCCHCHC],
145 [1C, (CH3)3CC], 149 (1C, NCCH2), 149 (1C, NCCH3), 168 (1C,
COOEt), 196 (1C, CO) ppm; GC-EI/MS [80 °C (1 min) – 300 °C (5.5 min)]
tR = 9.11 min, m/z (%): 395 (17) (M+), 262 (100) [(M–Me3CPh)+].
4-(Trimethylsilyl)benzaldehyde (5a): To a solution of (4-bromophenyl)tri-
methylsilane (1.50 g, 6.54 mmol) in THF (40 mL) n-BuLi (2.88 mL, 2.50 M,
7.19 mmol) was added dropwise at –78 °C. The reaction mixture was
stirred at –78 °C for 15 min. After adding dimethylformamid (1.43 g,
19.6 mmol), the reaction mixture was slowly warmed to room temperature
and stirred overnight at room temperature. Then the reaction mixture was
quenched by the addition of aqueous saturated NH4Cl solution. The
aqueous phase was extracted with Et2O and the combined organic phases
were washed with aqueous saturated NaCl solution and dried over Na2SO4.
After removal of all volatile components and column chromatographic
purifying (pentane, pentane/Et2O = 100:1 → 50:1) the product could be
Si-DHP 6a: The reaction of dimedone (78.6 mg, 0.56 mmol), ammonium
acetate (43.2 mg, 0.56 mml), acetylacetone (72.9 mg, 0.17 mmol), 4-
(trimethylsilyl)benzaldehyde (42.6 mg, 0.56 mmol), iodine (45.5 mg,
0.17 mmol) in EtOH (1.40 mL) gave after column chromatographic
purification (pentane/Et2O = 5:1 → 2:1 → 1:1 → 1:2 → 1:5 →1:10) and
This article is protected by copyright. All rights reserved.

10.1002/ejic.201901223
European Journal of Inorganic Chemistry
FULL PAPER
recrystallization in EtOH yellowish crystals (yield: 108 mg, 0.26 mmol,
46%). 1H-NMR (300 MHz, CDCl3): δ = 0.18 [s, 9H, Si(CH3)3], 0.94 [s, 3H,
(CO)CH2CCH3], 1.05 (s, 3H, (CO)CH2CCH3 ), 1.20 (t, J(H,H) = 7.14 Hz,
3H, OCH2CH3), 2.06-2.42 [m, 7H, (CO)CH2, NCCH2, NCCH3], 4.06 (q, J =
6.95 Hz, 2H, OCH2), 5.02 (s, 1H, CH), 6.80 (br. s., 1H, NH), 7.15-7.50 (m,
4H, CHaromatic) ppm; 13C{1H}-NMR (75 MHz, CDCl3): δ = –1 [3C, Si(CH3)3],
14 (1C, OCH2CH3), 19 (1C, NCCH3), 27 [1C, (CO)CH2CCH3], 29 [1C,
(CO)CH2CCH3], 33 [1C, (CO)CH2C(CH3)2], 37 [1C, (CO)CCH], 41 (1C,
NCCH2), 51 [1C, (CO)CH2], 60 (1C, OCH2CH3), 106 [1C, NCC(COOEt)],
112 [1C, NCC(CO)], 127 [2C, (CH3)3SiCCH], 133 [2C, (CH3)3SiCCHCH],
137 [1C, (CH3)3SiC], 144 [2C, (CH3)3SiCCHCHC], 1478 (1C, NCCH2), 149
(1C, NCCH3), 168 (1C, COOEt), 196 (1C, CO) ppm; 19Si{1H}-NMR
(60 MHz, CDCl3): δ = –4.4 (1Si) ppm; GC/EI-MS [80 °C (1 min) – 300 °C
(5.5 min)] tR = 8.95 min, m/z (%): 411 (11) (M+), 262 (100) [(M–Me3SiPh)+].
elemental analysis calcd (%) for C24H33NO3Si: C 70.03, H 8.08, N 3.40;
found: C 69.7, H 8.1, N 3.1.
Acknowledgements
We thank Stephanie Schulze for her efforts in the HPLC
experiments and the Deutsche Forschungsgemeinschaft (DFG)
and the Fonds der Chemischen Industrie for financial support.
Keywords: Bioisosters • silicon chemistry • germanium
chemistry • 1,4-dihydropyridines • TGF inhibitors
[1]
[2]
a) C. H. Heldin, K. Miyazono, P. ten Dijke, Nature 1997, 390, 465–471.;
b) K. Wharton, R. Derynck, Development 2009, 136, 3691–3697.
a) J. M. Yingling, K. L. Blanchard, J. S. Sawyer, Nat. Rev. Drug Discov.
2004, 3, 1011–1022; b) S. Colak, P. Ten Dijke, Trends Cancer 2017, 3,
56–71.
[3]
[4]
[5]
E. Willems, J. Cabral-Teixeira, D. Schade, W. Cai, P. Reeves, P. J.
Bushway, M. Lanier, C. Walsh, T. Kirchhausen, J. C. I. Belmonte, J.
Cashman, M. Mercola, Cell Stem Cell 2012, 11, 242–252.
Ge-DHP 6b: The reaction of dimedone (63.1 mg, 0.45 mmol), ammonium
acetate (34.7, 0.45 mml), acetylacetone (58.6 mg, 0.45 mmol), 4-
(trimethylgermyl)benzaldehyde (100 mg, 0.45 mmol), iodine (33.0 mg,
0.13 mmol) in EtOH (1.40 mL) gave after column chromatographic
purification (pentane/Et2O = 5:1 → 2:1 → 1:1 → 1:2 → 1:5 →1:10) and
recrystallization in EtOH yellowish crystals (yield: 72.0 mg, 0.16 mmol,
35%). 1H-NMR (300 MHz, CDCl3): δ = 0.30 [s, 9H, Ge(CH3)3], 0.94 [s, 3H,
(CO)CH2CCH3], 1.05 (s, 3H; (CO)CH2CCH3), 1.21 (t, J = 7.14 Hz, 3H;
OCH2CH3), 2.07-2.42 [m, 7H; (CO)CH2, NCCH2, NCCH3], 4.05 („q“, J =
6.95 Hz, 2H; OCH2), 5.02 (s, 1H; CH), 6.61 (br. s., 1H; NH) 7.19-7.40 (m,
4 H; CHaromatic) ppm; 13C{1H}-NMR (75 MHz, C6D6): δ = –2 [3C, Ge(CH3)3],
14 (1C, OCH2CH3), 19 (1C, NCCH3), 27 [1C, (CO)CH2CCH3], 29 [1C,
(CO)CH2CCH3], 33 [1C, (CO)CH2C(CH3)2], 36 [1C, (CO)CCH], 41 (1C,
NCCH2), 51 [1C, (CO)CH2], 60 (1C, OCH2CH3), 106 [1C, NCC(COOEt)],
112 [1C, NCC(CO)], 128 [2C, (CH3)3GeCCH], 133 [2C, (CH3)3GeCCHCH],
140 [1C, (CH3)3GeC], 144 [2C, (CH3)3GeCCHCHC], 147 (1C, NCCH2),
149 (1C, NCCH3), 168 (1, COOEt), 196 (1C, CO) ppm; GC/EI-MS [80 °C
(1 min) – 300 °C (5.5 min)] tR = 9.33 min, m/z (%): 457 (4) (M+), 262 (100)
[(M–Me3GePh)+]. elemental analysis calcd (%) for C24H33NO3Ge: C 63.19,
H 7.29, N 3.07; found: C 63.2, H 7.3, N 2.9.
D. Schade, M. Lanier, E. Willems, K. Okolotowicz, P. Bushway, C.
Wahlquist, C. Gilley, M. Mercola, J.R. Cashman, J. Med. Chem. 2012,
55, 9946-9957.
a) D. Längle, V. Marquardt, E. Heider, B. Vigante, G. Duburs, I. Luntena,
D. Flötgen, C. Golz, C. Strohmann, O. Koch, D. Schade, Eur. J. Med.
Chem. 2015, 95, 249-266; b) D. Längle, T. R. Werner, F. Wesseler, E.
Reckzeh, N. Schaumann, L. Drowley, M. Polla, A. T. Plowright, M. N. Hirt,
T. Eschenhagen, D. Schade, Chem. Med. Chem. 2019, 14, 1-14.
D. Schade, M. Lanier, E. Willems, K. Okolotowicz, P. Bushway, C.
Wahlquist, C. Gilley, M. Mercola, J. R. Cashman, J. Med. Chem. 2012,
55, 9946–9957.
[6]
[7]
a) N. R. Perestelo, G. G. Llanos, C. P. Reyes, A. Amesty, K. Sooda, S.
Afshinjavid, I. A. Jiménez, F. Javid, I. L. Bazzocchi, J. Med. Chem., 2019,
62, 4571–4585; b) E. Eryilmaz, Eur. J. Pharm. Sci. 2019, 134, 266–273;
c) M. F. Khan, L. Zepeda-Velazquez, M. A. Brook, Colloids Surf. B 2015,
132, 216–224; d) S. Fujii, Y. Hashimoto, Future Med. Chem. 2017, 9,
485–202; e) R. Ramesh, D. S. Reddy, J. Med. Chem. 2018, 61, 3779–
3798.
[8]
a) B. Célariès, M. Gielen, D. de Vos, G. Rima, Appl. Organomet. Chem.
2003, 17, 191–193; b) S. Fujii, Y. Miyajima, H. Masuno, H. Kagechika, J.
Med. Chem. 2013, 56, 160–166; c) E. Lukevics, L. Ignatovich, Appl.
Organomet. Chem. 1992, 6, 113–126; d) E. Lukevics, L. Ignatovich, I.
Shestakova, Appl. Organomet. Chem. 2003, 17, 898–905; e) P. G. R.
Ortega, M. Montejo, J. J. L. González, J. Mol. Model. 2014, 20, 2430; f)
R. Tacke , T. Heinrich , T. Kornek , M. Merget , S. A. Wagner , J. Gross ,
C. Keim , G. Lambrecht , E. Mutschler , T. Beckerss , M. Bernd, T.
Reissmann, Phosphorus, Sulfur Silicon Relat. Elem. 1999, 150, 69–87;
g) T. Yamakawa, H. Kagechika, E. Kawachi, Y. Hashimoto, K. Shudo, J.
Med. Chem. 1990, 33, 1430–1437; h) 1. R. Tacke, D. Reichel, P.G.
Jones, X. Hou, M. Waelbroeck, J. Gross, E. Mutschler, G. Lambrecht, J.
Organomet. Chem. 1996, 521, 305–323; i) R. Tacke, M. Merget, R.
Bertermann, M. Bernd, T. Beckers, T. Reissmann, Organometallics 2000,
19, 3486–3497; j) R. Tacke, B. Becker, D. Berg, W. Brandes, S.
Dutzmann, S. Schaller, J. Organomet. Chem. 1992, 438, 45–55.
a) A. F. Hollemann, E. Wiberg, N. Wiberg in Anorganische Chemie, Vol.
103, De Gruyter, Berlin, 2017, p. 1551; b) K. M. Baines, W. G. Stibbs,
Coord. Chem. Rev., 1995, 145, 157–200.
CCDC 1943895 (for 6a), 1943893 (for 6b) and 1943894 (for 2) contain the
supplementary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic Data Centre.
TGF/Smad assay: A SMAD-4 binding element (SBE-4)-based transient
luciferase reporter gene assay (in 293T cells) was performed as previously
described in detail. In brief, cells were batch-transfected with a SBE4-firefly
luciferase and TK-driven renilla luciferase plasmid, replated after ca. 12 h
on 96-well plates and incubated for 2 hours before addition of compounds,
DMSO and TGF-2 (10 ng/mL). Each condition was done in triplicate. After
overnight incubation, firefly and renilla luciferase activities were quantified
on a Tecan Infinite M1000 (Crailsheim, Germany) following the instructions
of the DualGlo Assay-Kit (Promega, Madison, USA). GraphPad Prism 5
was used for data evaluation.
[9]
[10] G. J. Inman, F. J. Nicolas, J. F. Callahan, J. D. Harling, L. M. Gaster, A.
D. Reith, N. J. Laping, C. S. Hill, Mol. Pharmacol. 2002, 62, 65−74.
This article is protected by copyright. All rights reserved.