Biologically active trifluoromethyl-substituted metallocene triazoles: Characterization, electrochemistry, lipophilicity, and cytotoxicity

Article abstract of DOI:10.1002/ejic.201200798

We report the facile synthesis of four different trifluoromethylated metallocene triazoles and their biological evaluation (M = Fe and Ru). The cytotoxicity of all compounds was evaluated using MCF-7, HT-29, PT-45 and GM5657 cells and IC₅₀ values as low as 33 μM were found. It was shown that the metallocene moiety is crucial for the cytotoxic effect. The electrochemical behavior of triazoles 3-6 was investigated by cyclic voltammetry. These electrochemical measurements revealed that all triazoles are less prone to oxidation than ferrocene. Moreover, Log P determination by RP-HPLC showed increased lipophilicity for the fluorinated derivatives with Log P values up to 3.8. Furthermore, successful bioconjugation could be achieved by coupling triazole 7 to the amino acid L-leucine. A series of new trifluoromethylated metallocene triazoles was synthesized using click chemistry (copper-catalyzed azide-alkyne cycloaddition). The new triazoles display suitable lipophilic character and exert promising antiproliferative activities against a range of cancer cell lines. Copyright

Full text of DOI:10.1002/ejic.201200798

Job/Unit: I20798
/KAP1
Date: 29-10-12 16:30:43
Pages: 8
FULL PAPER
DOI: 10.1002/ejic.201200798
Biologically Active Trifluoromethyl-Substituted Metallocene Triazoles:
Characterization, Electrochemistry, Lipophilicity, and Cytotoxicity
Marcus Maschke,[a] Max Lieb,[a] and Nils Metzler-Nolte*[a]
Keywords: Bioinorganic chemistry / Medicinal chemistry / Cytotoxicity / Click chemistry / Fluorinated ligands / Metallo-
cenes / Sandwich complexes / N ligands / Cycloaddition
We report the facile synthesis of four different trifluorometh-
ylated metallocene triazoles and their biological evaluation
(M = Fe and Ru). The cytotoxicity of all compounds was
evaluated using MCF-7, HT-29, PT-45 and GM5657 cells and
cyclic voltammetry. These electrochemical measurements re-
vealed that all triazoles are less prone to oxidation than ferro-
cene. Moreover, LogP determination by RP-HPLC showed
increased lipophilicity for the fluorinated derivatives with
LogP values up to 3.8. Furthermore, successful bioconjug-
ation could be achieved by coupling triazole 7 to the amino
IC50 values as low as 33 μM were found. It was shown that
the metallocene moiety is crucial for the cytotoxic effect. The
electrochemical behavior of triazoles 3–6 was investigated by
acid L-leucine.
suitable reaction is the azide–alkyne cycloaddition, other-
wise known as a “click” reaction. This reaction can be per-
formed under mild, non-oxidizing conditions, is well estab-
lished, reliable and high-yielding.[6] Notably, azide–alkyne
cycloadditions between metallocenes and terminal alkynes
have been previously described by Astruc and co-workers.[7]
Herein, we report the synthesis, characterization and bio-
logical evaluation of several novel trifluoromethylthio- and
trifluoromethyl-substituted metallocene triazoles. Applica-
tion of well established “click”-chemistry enables easy ac-
cess to these new compounds. In subsequent reactions the
synthesized compounds could be coupled to natural amino
acids to yield the first bioconjugate with these new tri-
fluoromethylated metallocene triazoles. Moreover, all new
trifluoromethylated metallocene triazoles were tested for
cytotoxicity on various cancer cell lines, the effects of elec-
tron-withdrawing substituents was revealed using cyclic
voltammetry and lipophilicities were determined by RP-
HPLC.[8]
Introduction
During the last several decades metallocene derivatives,
particularly those based on ferrocene, have attracted con-
siderable attention as chemotherapeutic agents and anti-
biotics.[1] However, among these compounds, fluorinated
species are nearly unknown. On the other hand, fluorinated
compounds such as 5-fluorouracil or -cytosine are among
the oldest commercially available chemotherapeutic
agents.[2] Consequently, the prospect of combining the vir-
tues of metallocenes and fluorinated substituents seemed to
us a promising venture by which to identify new biolo-
gically active entities. Such compounds are likely to provide
interesting properties such as enhanced lipophilicity, higher
acidity and altered electrostatic interactions.[3] Further-
more, the strong electron-withdrawing effect of the fluori-
nated substituents is likely to influence the redox character-
istics of the metal center. Electron-withdrawing groups
(EWGs) reduce the electron density at the ferrocene moiety
inducing a higher resistance to oxidation.[4]
The fluorination of ferrocene, on the other hand, poses
considerable challenges; direct introduction of fluorine or Results
trifluoromethyl groups often implies the use of strong oxid-
izing agents and acidic conditions.[5] In this respect a careful
choice of reagents and conditions is required to create ferro-
cene moieties bearing trifluoromethylated substituents. One
Synthesis and Characterization
As outlined in Scheme 1 all four triazoles 6–9 were syn-
thesized using copper-catalyzed azide–alkyne cycloaddition
(CuAAC) in degassed solvents. Application of a 1:1.3 ratio
of azide:alkyne resulted in quantitative yields. Complete-
ness of reaction was monitored by TLC. Purification was,
typical for “click” reactions, easily done by column
chromatography. All novel triazoles were characterized by
elemental analysis, NMR, IR and GC–MS. The solid-state
structure of 6, 7 and 9 could be determined by X-ray crys-
tallography.
[a] Fakultät für Chemie und Biochemie, Ruhr-Universität
Bochum,
Universitätsstraße 150, 44801 Bochum, Germany
Fax: +49-234-3214378
E-mail: nils.metzler-nolte@rub.de
Homepage: www.chemie.rub.de/ac1
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201200798
Eur. J. Inorg. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1

Job/Unit: I20798
/KAP1
Date: 29-10-12 16:30:43
Pages: 8
M. Maschke, M. Lieb, N. Metzler-Nolte
FULL PAPER
Scheme 1. Syntheses of metallocene triazoles 6–9. Conditions: (i)
CuSO4, sodium ascorbate, THF/H2O (10:1), 3 d, N2, room temp.
Azidoferrocene (1),[9] (azidomethyl)ferrocene (2),[10] (azi-
domethyl)ruthenocene (3),[11] 1,1,1-trifluoro-2-(trifluorome-
thyl)-3-butyn-2-ol (4)[12] and 3-(trifluoromethylthio)-1-pro-
pyne (5)[13] were chosen as starting materials for the click
reactions.
Figure 2. Intermolecular hydrogen bonding in triazole 6; N–(H)O
bond length: 2.805 Å.
Reaction of azidoferrocene (1) with 1,1,1-trifluoro-2-(tri-
fluoromethyl)-3-butyn-2-ol (4) at room temp. gave the first
trifluoromethyl-substituted ferrocene triazole 6 in 36%
The reaction of (azidomethyl)ferrocene (2) with 1,1,1-tri-
fluoro-2-(trifluoromethyl)-3-butyn-2-ol (4) at room temp.
gave triazole 7 in 95% yield. Unlike the case for compound
6, an X-ray analysis showed that the Cp and triazole rings
are not coplanar at about an angle of 53.1° due to the CH2
unit between the triazole ring and the ferrocene moiety.
Bond lengths of the ferrocene and triazole moiety are nearly
the same as for 6 (Figure 3).
1
yield. Characterization by H-NMR showed the aromatic
protons of the substituted Cp ring as a singlet. Moreover,
the OH signal at about 5.40 ppm disappeared upon ad-
dition of MeOD. In 19F-NMR a singlet for the symmetric
CF3 groups was observed at –77.8 ppm. Suitable crystals
for an X-ray analysis could be grown by slow evaporation
of a CHCl3 solution. The solid-state structure of 6 was de-
termined by X-ray diffraction. The ORTEP plot of the mo-
lecule is shown in Figure 1. The structure was found to pos-
sess the classical “sandwich”-like appearance with an
average Fe–C bond length of 2.037(6) Å for the substituted
and 2.036(4) Å for the unsubstituted Cp ring. These bond
lengths are fully in line with values reported in the literature
for ferrocene derivatives.[14] Moreover, the Cp ring and the
triazole ring are almost in plane, with a small distortion of
11.2°. Compound 6 forms dimers in the solid state through
a weak hydrogen bridge between two molecules [N–(H)O
distance of 2.805 Å] (Figure 2).
Figure 3. ORTEP plot of 7 (top) and 9 (bottom). Selected bond
lengths of 7 [Å]: Fe1–Ccentroid 1.65, Fe1–Caverage 2.04, N1–N2 1.338,
N2–N3 1.310, N1–C11 1.488, C–Faverage 1.33°Å. Selected bond
lengths of 9 [Å]: Ru1–Ccentroid 1.81, Ru1–Caverage 2.19, N1–N2
1.344, N2–N3 1.311, N1–C1 1.481 C–Faverage 1.33°Å.
Reaction of azidoferrocene (1) with 3-(trifluoromethyl-
thio)-1-propyne (5) at room temp. gave triazole 8 in 55%
yield. In the 19F-NMR a singlet for the SCF3-group was
observed at –44.5 ppm. Work-up of triazole 8 by column
Figure 1. ORTEP plot of 6. Selected bond lengths [Å]: Fe1–Ccentroid
1.64, Fe1–Caverage 2.04, N1–N2 1.339, N2–N3 1.309, N1–C6 1.422
C–Faverage 1.32.
2
www.eurjic.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 0000, 0–0

Job/Unit: I20798
/KAP1
Date: 29-10-12 16:30:43
Pages: 8
Biologically Active Metallocene Triazoles
chromatography however was more difficult than for 6 and
Cyclic voltammograms of all trifluoromethylated ferro-
7. The difficulty in purification of 8 may explain the rela- cene triazoles revealed one-electron waves with quasi-re-
tively low yield for this triazole.
versible Nernst behavior (ΔEp = 70–76 mV) and peak cur-
Red
The first trifluoromethylated triazole with a ruthenocene rent ratios IpOx/Ip
moiety is compound 9, which was synthesized from recently
of about 1 (Table 1, Figure 5).
reported (azidomethyl)ruthenocene (3)[11] and 1,1,1-tri-
Table 1. Cyclic voltammetry data for compounds 6–8 (1 mm) in
CH3CN with TBAPF6 as the supporting electrolyte (0.1 m) and
ferrocene (1 mm) as standard.
fluoro-2-(trifluoromethyl)-3-butyn-2-ol (4) at room temp.
(60% yield). Unlike previous observations for the ferrocene
triazoles, 1H NMR spectroscopy of 9 revealed the aromatic
protons of the substituted Cp ring as a pair of pseudo-trip-
lets. X-ray analysis revealed nearly an identical structure
compared to 7, except for the fact that the central atom is
ruthenium. Ru–C bonds have an average length of 2.191(4)
Å [2.179(4) Å for the unsubstituted Cp ring], which is in
accordance with literature data (Figure 3).[15]
f
Ox
Red
Ox
Scan rate ΔE0 [mV] ΔEp
IP
IP
IPRed
/
[mVs–1]
vs. Fc0/+
[mV]
[μA]
[μA]
IP
6
7
8
50
50
50
251
126
217
70
74
76
14.8
17.9
3.76
14.5
17.4
3.24
1.02
1.03
1.16
In order to test a possible bioconjugation of one of the
newly synthesized triazoles we chose Boc-Leu-OH as a
model system (Scheme 2). Coupling of triazole 7 with the
Boc protected natural amino acid l-leucine gave bioconjug-
ate 10. A Steglich esterification was applied for the coupling
reaction.[16] Characterization by 19F-NMR showed a split-
ting of the signal for the fluorine atoms due to the asym-
metric center at the leucine α-CH (Figure 4). Progress of
the reaction could thus be conveniently monitored by 19F-
NMR spectroscopy. After purification by column
chromatography a yield of 61% was obtained. However, it
appeared during later studies that 10 hydrolyzes very easily
even in slightly basic conditions and therefore, no further
biological studies were carried out on this compound.
Figure 5. Cyclic voltammogram of 6–8 (1 mm) in CH3CN with
TBAPF6 as supporting electrolyte (0.1 m) with a scan rate of
250 mV/s (6, 7) and 1000 mV/s (8).
Triazole 6 has a half-wave potential at about 250 mV,
which is in line with literature data for other ferrocene tri-
azoles with electron-withdrawing groups (190–270 mV).[17]
Triazole 7 is more easily oxidized than 6, as reflected by
the lower half-wave potential of 126 mV, which is 112 mV
higher than in literature reports for similar ferrocenyl meth-
yltriazoles.[17a] This could be due to the C(CF3)2OH group
of the triazole ring. Nevertheless, the methylene group be-
tween the ferrocene moiety and the triazole ring shields the
metal center from the EWG compared to triazole 6.
Scheme 2. Coupling of 7 to Boc-Leu-OH by use of Steglich esterifi-
cation.
For triazole 8 a half-wave potential of 217 mV was ob-
served, which is 23 mV lower than that observed for 6. The
difference can clearly be attributed to the varied fluorinated
substituents of the triazole rings in 6 and 8 (Table 1).
Triazole 9 showed non-reversible behavior at all five scan
Ox
rates. Therefore, only the peak potential Ep (≈ 456 mV)
and the peak current IPOx (6.2–20.7 μA) was determined. It
is known that ruthenocene possesses a non-reversible redox
system with a two-electron transfer in acetonitrile.[8b]
Figure 4. 19F NMR of compound 10 with characteristic splitting
of the CF3 signal (lower trace) and 7 (above).
Electrochemical Studies
Cytotoxicity
An electrochemical micro volume cell with a stationary
glassy carbon working electrode (Ø = 3 mm), Ag/AgCl in
The cytotoxicities of trifluoromethyl-substituted ferro-
aq. KCl (3 m) as reference electrode and a platinum wire (Ø cene triazoles 6–9 were evaluated using MCF-7 (breast can-
= 2 mm) as counter electrode were used. cer), HT-29 (colon cancer), and PT-45 (pancreas cancer)
Eur. J. Inorg. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3

Job/Unit: I20798
/KAP1
Date: 29-10-12 16:30:43
Pages: 8
M. Maschke, M. Lieb, N. Metzler-Nolte
FULL PAPER
cells. As a comparison, cytotoxicity was also checked
against a non-cancerous fibroblast cell line (GM5657). To
assess the importance of the metallocene moiety for the
cytotoxic effect, we synthesized the metal-free compound
11 (Scheme 3). All cancer cell lines were found to be moder-
Conclusions
In summary, the facile, high-yielding, and reliable prepa-
ration of new metallocene-containing trifluoromethylated
triazoles was accomplished using “click” chemistry. We re-
ported the synthesis, characterization and biological evalu-
ation of four trifluoromethylated metallocene triazoles. The
electrochemical behavior of triazoles 6–9 was investigated
by cyclic voltammetry and revealed these agents to possess
a greater resistance to oxidation than their non-fluorinated
counterparts. Determination of LogP by PR-HPLC re-
vealed higher lipophilicity for all novel compounds relative
to ferrocene. Cytotoxicities of the new compounds were
evaluated using MCF-7, HT-29, PT-45 and GM5657 cells
and IC50 Ն 33 μm were found for all compounds. These
assays revealed that the metallocene moiety is crucial for
cytotoxicity. Because all novel trifluoromethylated metallo-
cene triazoles show similar cytotoxicity, electrochemical be-
havior, and lipophilicity, a structure–activity relationship
can be established. The cytotoxic activity displayed and
lack of cytotoxicity against fibroblasts by these novel com-
pounds is perhaps rationalized by formation of reactive
oxygen species (ROS) inside the cell; cancer cells appear
more prone to the effects of these new compounds, consis-
tent with an ROS hypothesis. Furthermore, successful bi-
oconjugation was demonstrated by the effective coupling of
triazole 7 to l-leucine.
ately sensitive to the novel metallocene triazoles (IC50
Ն
33 μm), whereas normal cells (GM5657) were unaffected at
concentrations of the triazoles of up to 100 μm. Thus, some
selectivity for cancer cells over normal cells seems to exist
for the ferrocene triazoles investigated herein. The IC50 val-
ues of 11 indicate that the metallocene must have some cru-
cial function in the cytoxicity on cancer cells (Table 2).
Scheme 3. A metal-free compound for comparison of the cytotoxi-
city data.
Table 2. IC50 values (in μm) of triazoles 6–11.
MCF-7
HT-29
PT-45
GM5657
[a]
[a]
[a]
[a]
[a]
[a]
[a]
6
58Ϯ7.8
38Ϯ0.5
49Ϯ14
7
67Ϯ0.1
55Ϯ5
not tested
not tested
8
9
11
33Ϯ5
41Ϯ15
55Ϯ25
[a]
[a]
[a]
In future projects the introduction of fluorinated substit-
uents will be extended to other metallocenes. Moreover, fur-
ther research will focus on the bioconjugation of these novel
fluorinated metallocene derivatives to larger biomolecules.
[a] Inactive up to 100 μm.
Lipophilicity
Experimental Section
Lipophilicty is one of four key properties of potential
drugs according to Lipinski’s “rule of five”.[18] Although
the ferrocene moiety already is a rather lipophilic group,
incorporation of trifluoromethyl groups in the metallocene
scaffold should shift the LogP to even higher values; en-
hanced lipophilicity would likely improve cell membrane
permeability. To test these expectations, the water/octanol
partition coefficients (LogP values) of 6, 7 and 9 were de-
termined by the reverse-phase HPLC (RP-HPLC) by the
method of Minick and co-workers.[19]
The results, as shown in Table 3, indicate an increase in
lipophilicity for all tested triazoles. LogP values were found
to increase by about 0.4 units compared to ferrocene for all
compounds. Comparison of 6 and the similar structures of
7 and 9 reveals a slight increase of lipophilicity due to the
methylene group.
General Remarks: Unless noted otherwise, all preparations were
carried out under an inert atmosphere of argon or N2 using stan-
dard Schlenk techniques. All reagents and anhydrous solvents were
purchased from commercial sources and used as received. The rea-
gents azidoferrocene (1),[9] (azidomethyl)ferrocene (2),[10] (azi-
domethyl)ruthenocene (3),[11] 1,1,1-trifluoro-2-(trifluoromethyl)-3-
butyn-2-ol (4),[12] 3-(trifluoromethylthio)-1-propyne (5)[13] were
prepared by literature procedures. NMR spectra were recorded at
ambient temperature with Bruker DPX 200, DPX 250, DRX 400
and DRX 600 spectrometers. The chemical shifts (δ) are reported
in parts-per-million (ppm) relative to the residual proton chemical
shifts of the deuterated solvent set relative to external TMS. Cou-
pling constants (J) are quoted in Hertz. IR spectra were recorded
with a Bruker Tensor 27 spectrometer with an ATR unit as solid
samples, wavelengths of absorption are given in cm–1. Electrospray
ionization mass spectra (ESI-MS) were recorded with a Bruker Es-
quire 6000 spectrometer. GC–MS spectra were recorded using a
Shimadzu GC–MS-QP2010. Elemental analyses of ruthenium con-
taining compounds were carried out at the laboratory for microan-
alytics and thermal analyses, University of Essen (Inorganic Chem-
istry Department), all others were carried out at the RUBiospek
Biospectroscopy Department, Ruhr-Universität Bochum.
Table 3. Octanol/water partition coefficients of ferrocene and the
triazoles 6, 7, and 9.
LogKW
LogP
R2
Ferrocene
2.95
3.31
3.43
3.38
3.40
3.75
3.86
3.81
0.998
0.999
0.999
0.996
X-ray Structure Determination of 6, 7 and 9: Single crystals were
obtained by slow evaporation of a CHCl3 solution. The crystals (6,
7: orange crystals, 9: white needles) were placed on a glass capillary
in perfluorinated oil and measured in a cold gas flow. The intensity
6
7
9
4
www.eurjic.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 0000, 0–0

Job/Unit: I20798
/KAP1
Date: 29-10-12 16:30:43
Pages: 8
Biologically Active Metallocene Triazoles
Table 4. Selected crystallographic data for triazoles 6, 7, and 9.
6
7
9
Empirical formula
Formula weight
Temperature
Crystal system, space group
Unit cell dimensions
C15H11F6FeN3O
419.02 g/mol
298 K
monoclinic, P21/c
a = 10.350(8) Å, α = 90°
b = 20.934(15) Å,
β = 97.140(6)°
c = 7.614(6) Å, γ = 90°
1637(2) Å3
C16H13F6FeN3O
433.14 g/mol
173(2) K
C16H13F6RuN3O
478.36 g/mol
173 K
¯
triclinic, P1
monoclinic, P21/c
a = 11.209(11) Å, α = 71.42(3)° a = 16.49(2) Å, α = 90°
b = 13.153(11), Å
β = 82.47(4)°
b = 11.338(16) Å,
β = 90.684(16)°
c = 11.310(17) Å, γ = 90°
2114(5) Å3
c = 13.412(12) Å, γ = 67.54(2)°
Volume
1732(3) Å3
Z
4
4
4
Theta range of data collection 2.78 to 24.99°
1.97 to 25.00°
2.47 to 25.00°
Reflections collected/unique
Completeness of θ = 24.99
Data/restraints/parameters
Goodness of fit on F2
Final R indices (I Ͼ 2σI)
R indices (all data)
13826/2869 (Rint = 0.0698)
99.7%
2869/0/235
1.051
R1 = 0.0533, wR2 = 0.1301
R1 = 0.0686, wR2 = 0.1402
0.559 and –0.374 eÅ–3
14513/5991 (Rint = 0.0591)
98.1%
5991/0/495
17408/3658 (Rint = 0.0906)
98.3%
3658/0/248
1.083
1.055
R1 = 0.0836, wR2 = 0.2272
R1 = 0.1051, wR2 = 0.2423
1.642 and –0.415eÅ–3
R1 = 0.0454, wR2 = 0.1168
R1 = 0.0527, wR2 = 0.1230
1.309 and –0.613e. Å–3
Largest diff. peak and hole
data were measured with a Rigaku Mercury 375 R/M CCD
(XtaLAB mini) diffractometer. The structures were solved by direct
methods (SHELXS 97[20]) and refined against F2 with all measured
reflections (SHELXL97[20] and Platon/Squeeze[21]). Table 4 con-
tains relevant crystallographic data for all crystal structures.
acetate, 4:1). C15H11F6FeN3O (419.11): calcd. C 42.99, H 2.65, N
10.03; found C 43.86, H 2.47, N 9.8. H NMR (CDCl3): δ = 7.91
(s, 1 H, CHtriazole), 5.40 (s, 1 H, OH), 4.88 (s, 2 H, CpH2), 4.34 (s,
2 H, CpH2), 4.24 (s, 5 H, Cp) ppm. 19F NMR (CDCl3): δ = –77.8
(s, 6 F) ppm. 13C NMR (CDCl3): δ = 137.2, 123.2, 122.4, 120.5,
1
117.6, 110.0, 93.0, 70.4, 67.3, 62.7, 29.7 ppm. IR (solid): ν = 157
˜
CCDC-878370 (for 9), -878371 (for 6) and -878372 (for 7) contain
the supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
ν(C–H); 1777–1629 ν(C–Caromatic); 1400–1322 ν(C–Caromatic); 1150
ν(C–F) cm–1. ESI-MS (pos.): m/z = 419.08 ([M]+, exact mass of
C15H11F6FeN3O = 419.02), 441.86 ([M + Na]+). MS (EI): m/z =
419 [M]+, 252 [M – (CF3)2COH]+, 185 [Fc – H]+, 121 [CpFe]+, 56
[Fe]+.
1,1,1-Trifluoro-2-(trifluoromethyl)-3-butyn-2-ol (4): Alkyne 4 was
prepared by literature procedures, but with a modified workup
method.[12,22]
2-(1-Ferrocenylmethyl-1H-1,2,3-triazol-4-yl)-1,1,1,3,3,3-hexafluoro-
propan-2-ol (7): Column chromatography: n-hexane/ethyl acetate
(4:1). Orange solid (0.33 g, 0.076 mmol, 95%); Rf = 0.4 (n-hexane/
ethyl acetate, 4:1). C16H13F6FeN3O (433.13): calcd. C 44.37, H
An oven-dried Carius tube was purged with nitrogen while cooling,
charged with a stirring bar and ethynylmagnesium bromide (20 mL,
0.5 m in THF) was added. After this, gaseous hexafluoroacetone
(1 equiv.) was added to the heterogeneous mixture and was stirred
vigorously for 2 d at room temp. The crude mixture was poured
onto ice and concd. aqueous HCl (2 mL) was added subsequently.
After distillation (boiling range 64–66 °C) the THF/alcohol azeo-
trope was obtained. The free alcohol was obtained from the un-
avoidable THF/alcohol azeotrope by addition of the azeotrope to
concentrated sulfuric acid at 100 °C. Volatile material was then
evaporated into a dry ice trap. Alkyne 4 was obtained as a colorless
liquid. 1H NMR (CDCl3): δ = 3.50 (s, 1 H, OH), 2.80 (s, 1 H,
CH) ppm. 19F NMR (CDCl3): δ = –77 (s, 6 F) ppm.
1
3.03, N 9.70; found C 44.98, H 3.30, N 9.4. H NMR (CDCl3): δ
= 7.62 (s, 1 H, CHtriazole), 5.59 (s, 1 H, OH), 5.35 (s, 2 H, CH2)
4.29 (s, 4 H, Cp), 4.19 (s, 5 H, Cp) ppm. 19F NMR (CDCl3): δ =
–77.9 (s, 6 F) ppm. 13C NMR (CDCl3): δ = 137.0, 122.1 (CHtriazole),
79.8, 74.1 (quart, CF3), 69.4 (unsubstituted Cp-Ring), 69.0, 51.9
(CH ) ppm. IR (solid): ν = 3134 ν(C–H); 2925 ν(C–Caromatic);
˜
2
1538–1248 ν(C–Caromatic); 1175 ν(C–F), 1105 ν(C–F) cm–1. MS
(EI): m/z = 433 [M]+, 199 [FcCH2]+, 121 [CpFe]+, 56 [Fe]+.
1-Ferrocenyl-4-[(trifluoromethylthio)methyl]-1H-1,2,3-triazole (8):
Column chromatography: n-hexane/ethyl acetate (2:1) Orange solid
(0.02 g, 0.05 mmol, 55%); Rf = 0.4 (n-hexane/ethyl acetate, 2:1).
C14H12F3FeN3S (367.17): calcd. C 45.80, H 3.29, N 11.44, S 8.73;
found C 49.51, H 3.47, N 9.46, S 5.36. 1H NMR (CDCl3): δ = 7.74
(s, 1 H, CHtriazole), 4.84 (s, 2 H, CH2), 4.27 (s, 4 H, Cp), 4.22 (s, 5
H, Cp) ppm. 19F NMR (CDCl3): δ = –41.5 (s, 3 F) ppm. 13C NMR
(CDCl3): δ = 141.9, 131.2, 121.0, 92.5, 69.2, 65.8, 61.2, 28.7,
General Synthesis of Triazoles 6–9 and 11: An oven-dried 50 mL
Schlenk flask was purged with nitrogen while cooling, charged with
a stirring bar and the azide (1 equiv.) in THF (20 mL). CuSO4
(0.05 equiv.) and sodium ascorbate (0.3 equiv.) in H2O (2 mL) were
added consecutively. The solution was degassed for 10 min. After
this, the corresponding alkyne (1.3 equiv.) was added to the hetero-
geneous mixture and was stirred vigorously for 3 d at room temp.
in the dark. TLC analysis showed the complete consumption of the
azide. Dichloromethane was added to the reaction mixture, and it
was washed three times with water, dried with MgSO4 and the sol-
vents evaporated in vacuo. The crude product was purified by col-
umn chromatography (silica Merck 60) using a mixture of n-hex-
ane/ethyl acetate.
23.7 ppm. IR (solid): ν = 2956, 2922, 2853, 2744 ν(C–H); 1583–
˜
1458 ν(C–Caromatic); 1106 ν(C–F) cm–1. MS (EI): m/z = 367 [M]+,
338, 238, 199 [FcCH2]+, 185 [Fc – H]+, 121 [CpFe]+, 56 [Fe]+.
1,1,1,3,3,3-Hexafluoro-2-(1-ruthenocenylmethyl-1H-1,2,3-triazol-4-
yl)propan-2-ol (9): Column chromatography: n-hexane/ethyl acetate
(4:1). Colorless solid (0.06 g, 0.13 mmol, 60%); Rf = 0.5 (n-hexane/
ethyl acetate, 4:1). C16H13F6N3ORu + 0.5, C6H14: calcd. C 43.76,
H 3.87, N 8.06; found C 43.48, H 3.55, N 8.03. 1H NMR (CDCl3):
δ = 7.77 (s, 1 H, CHtriazole), 5.55 (s, 1 H, OH), 5.20 (s, 2 H, CH2)
4.66 (t, 2 H, Cp), 4.59 (t, 2 H, Cp), 4.51 (s, 5 H, Cp) ppm. 19F
NMR (CDCl3): δ = –77.8 (s, 6 F) ppm. 13C NMR (CDCl3): δ =
2-(1-Ferrocenyl-1H-1,2,3-triazol-4-yl)-1,1,1,3,3,3-hexafluoropropan-
2-ol (6): Column chromatography: n-hexane/ethyl acetate (4:1).
Orange solid (0.13 g, 0.31 mmol, 36%); Rf = 0.53 (n-hexane/ethyl
Eur. J. Inorg. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
5

Job/Unit: I20798
/KAP1
Date: 29-10-12 16:30:43
Pages: 8
M. Maschke, M. Lieb, N. Metzler-Nolte
FULL PAPER
137.1, 122.4 (CHtriazole), 122.1, 84.2, 74.1 (CF3), 71.4 (unsubstituted
Cp-Ring), 71.3 (substituted Cp-Ring), 71.1 (substituted Cp-Ring),
was set to 0 mV as the reference potential for all measurements.
Ferrocene was used as an internal reference for the scan rate at
50 mVs–1 and used as an external reference for all other scan rates
to avoid an overlap of the corresponding redox processes.
70.9 (substituted Cp-Ring), 70.9, 50.7 (CH ) ppm. IR (solid): ν =
˜
2
2922 ν(C–H); 1735 ν(C–Caromatic); 1376–1206, 1100 ν(C–F) cm–1.
MS (EI): m/z = 479 [M]+ , 245[Cp2 RuCH2 ]+ , 185 [Fc –
H]+, 167 [CpRu]+.
The electrochemical behavior of all four triazoles 6–9 was examined
by cyclic voltammetry (CV) at five different scan rates (50, 100,
250, 500 and 1000 mVs–1).
Compound 10: An oven-dried 50 mL flask was charged with a stir
bar and triazole 7 (0.022 g, 0.051 mmol) in dichloromethane
(20 mL). Boc-Leu-OH (0.016 g, 0.069 mmol), EDC (0.018 g,
0.094 mmol) and DMAP (0.0087 g, 0.094 mmol) were added con-
secutively to the heterogeneous mixture and stirred vigorously for
1 d at room temp. The reaction mixture was washed three times
with water, dried with MgSO4 and the solvents evaporated in
vacuo. The crude product was purified by column chromatography
(silica Merck 60) using a mixture of n-hexane/ethyl acetate (4:1).
Complex 10 was obtained as an orange oil (0.02 g, 0.031 mmol,
61%). 1H NMR (CDCl3): δ = 7.72 (s, 1 H, CHtriazole), 5.08–4.29
(m, 9 H, CpH9), 4.66 (t, 2 H, CpH2), 1.70 (m, 3 H), 1.44 (s, 9 H,
3 CH3), 0.98 (s, 3 H, CH3), 0.96 (s, 3 H, CH3) ppm. 19F NMR
(CDCl3): δ = –71 (t, 3 F, CF3), –72 (t, 3 F, CF3) ppm. 13C NMR
(CDCl3): δ = 137.1, 122.4, (CHtriazole) 84.2, 74.1 (CF3), 71.4 (unsub-
stituted Cp-Ring), 71.3 (substituted Cp-Ring), 71.1 (substituted
Cp-Ring), 70.9 (substituted Cp-Ring), 51.7 (CH2) ppm. ESI-MS
(pos. mode, 70 eV): m/z = 767.9 [M + H]+, 790.9 [M + Na]+. Exact
mass of complex cation: 766.54.
Cytotoxicity Assays: MCF-7, HT-29, PT-45 and GM5657 cells
were cultured in DMEM medium supplemented with 10% fetal
calf serum (FCS), 2 mm l-glutamine, penicillin (100 U/mL), and
streptomycin (100 μg/mL) in a 5% CO2 atmosphere. Crystal violet
assay was applied to determine the absolute cell numbers. All cells
were seeded in 96-well cell-culture treated microtiter plates (MTP)
and grown for 24 h under standard conditions. All tested com-
pounds were dissolved in cell culture medium with 0.5% dimethyl
sulfoxide (DMSO) and applied to the cells in 1, 5, 20, 50, 100,
500 μm concentrations for 72 h. Cells were fixed with 4% glutaral-
dehyde in 2% water for 25 min at room temp. Membranes were
permeabilized by Triton X-100 (0.1%) in PBS for 10 min. and then
aqueous crystal violet solution (0.04%) was added to the cells. Af-
ter 30 min of mechanical shaking, the cells were washed five times
with water and the crystal violet was eluted with 70% EtOH for
3 h. Absorbance was detected at 570 nm (Tecan Sapphire 2 micro-
plate reader). The cell mass was plotted against the concentration.
IC50 were calculated from the sigmoidal function.
1,1,1,3,3,3-Hexafluoro-2-(1-phenyl-1H-1,2,3-triazol-4-yl)propan-2-
ol (11): Column chromatography: n-hexane/ethyl acetate (4:1). Col-
orless solid (0.1 g, 0.32 mmol, 77%); Rf = 0.5 (n-hexane/ethyl acet-
Supporting Information (see footnote on the first page of this arti-
cle): GC–MS spectra of compounds 6–9.
1
ate, 4:1). H NMR (CDCl3): δ = 8.15 (s, 1 H, CHtriazole), 7.77 (d,
J = 7.49 Hz, 2 H), 7.61–7.53 (m, 3 H) 5.35 (s, 1 H, OH) ppm. 19F
NMR (CDCl3): δ = –77.7 (s, 6 F) ppm. 13C NMR (CDCl3): δ =
Acknowledgments
136.9, 135.3, 129.1, 128.9, 123.1, 120.0 ppm. IR (solid): ν = 3154,
˜
This work was supported by the Deutsche Forschungsgemeinschaft
(DFG), Research Unit FOR630, “Biological Function of Organo-
metallic Compounds”; www.rub.de/for630). M. M. wishes to thank
the Ruhr-University Research School for financial support. The
authors are grateful to Dr. Klaus Merz for solving the X-ray crystal
structures and to Dr. Malay Patra for providing (azidomethyl)-
ruthenocene. Thanks also go to Annegret Knüfer for the cytotoxi-
city test and to Dr. Nicolas Plumere for providing the electrochemi-
cal setup.
2932, 2853 ν(C–H); 1597–1257, 1059 ν(C–F) cm–1. MS (EI): m/z =
311 [M]+, 214 [M – (CF)3COH]+.
Determination of LogP Values by RP-HPLC: The RP-HPLC
method of Minick and co-workers was used for LogP determi-
nation.[19] All values were determined on a Reprosil-Par C 18-AQ,
5 μm ϫ250 mmϫ4.6 mm column. The water layer was saturated
with octanol. As eluents, 3-(N-morpholino)propanesulfonic acid
buffer (MOPS, 0.02 m, pH 7.4) with n-decylamine (0.15% v/v) and
MeOH with n-octanol (0.25% v/v) were used. In order to calibrate
the system 4-methoxyaniline, 4-bromoaniline, naphthalene and
tert-butylbenzene were eluted at six isocratic eluent concentrations
(60% to 85% of MeOH in 5% increments). The retention times,
together with the dead times obtained from uracil, were used to
calculate the corresponding capacity factors kЈ H2O/MeOH ratio.
The extrapolation to 100% H2O (0% MeOH) gave the logkw value
for each standard. All four logkw values of the reference com-
pounds were plotted against the literature logP values of the refer-
ence compounds to obtain a linear function that correlates both
parameters. This function was used to determine all logP values
from their logkw values, which were measured in the same way as
those for the references.
[1] a) G. Gasser, I. Ott, N. Metzler-Nolte, J. Med. Chem. 2011, 54,
3–25; b) D. R. van Staveren, N. Metzler-Nolte, Chem. Rev.
2004, 104, 5931–5986; c) A. Nguyen, A. Vessires, E. A. Hillard,
S. Top, P. Pigeon, G. Jaouen, Chimia 2007, 61, 716–724; d) M.
Salmain, N. Metzler-Nolte, in Ferrocenes (Ed.: P. Stepnicka),
John Wiley and Sons, Chinester, 2008, pp. 499–639; e) G. Gas-
ser, N. Metzler-Nolte, Curr. Opinion Chem. Biol. 2012, 16, 84–
91.
[2] a) C. Heidelberger, N. K. Chaudhuri, P. Danneberg, D. Mo-
oren, L. Griesbach, R. Duschinsky, R. J. Schnitzer, E. Pleven,
J. Scheiner, Nature 1957, 179, 663–666; b) A. Vermes, H.-J. Gu-
chelaar, J. Dankert, J. Antimicrob. Chemother. 2000, 46, 171–
179.
[3] G.-V. Röschenthaler, Nachr. Chem. 2005, 53, 743–746.
[4] C. Rhode, J. Lemke, M. Lieb, N. Metzler-Nolte, Synthesis
2009, 12, 2015–2018.
[5] A. E. Feiring, J. Fluorine Chem. 1977, 10, 375–386.
[6] H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. 2001,
113, 2056; Angew. Chem. Int. Ed. 2001, 40, 2004–2021.
[7] A. K. Diallo, S. Menuel, E. Monflier, J. Ruiz, D. Astruc, Tetra-
hedron Lett. 2010, 51, 4617–4619.
Electrochemical Measurements: An electrochemical micro volume
cell with a stationary glassy carbon working electrode (Ø = 3 mm),
Ag/AgCl in aq. KCl (3 m) as reference electrode and a platinum
wire (Ø = 2 mm) as counter electrode were used. All electrochemi-
cal measurements were undertaken with 5 mL of 0.1 mm solutions
of the respective trifluoromethylated metallocene triazole in aceto-
nitrile. CV of ferrocene was performed in the same electrolyte solu-
tion, and the electrochemical half-wave potential E1/2 of the redox
couple ferrocene/ferrocenium FcH0/+ vs. Ag/AgCl (E1/2) (270 mV)
[8] a) A. L. C. Hansch, Substituent Constants for Correlation
Analysis in Chemistry and Biology, John Wiley & Sons, New
6
www.eurjic.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 0000, 0–0

Job/Unit: I20798
/KAP1
Date: 29-10-12 16:30:43
Pages: 8
Biologically Active Metallocene Triazoles
York, 1979; b) S. P. Gubin, S. A. Smirnova, L. I. Denisovich,
A. A. Lubovich, J. Organomet. Chem. 1971, 30, 243–255.
[9] A. Shafir, M. P. Power, G. D. Whitener, J. Arnold, Organome-
tallics 2000, 19, 3978–3982.
[10] J. M. Casas-Solvas, E. Ortiz-Salmerón, J. J. Giménez-Martínez,
L. García-Fuentes, L. F. Capitán-Vallvey, F. Santoyo-
González, A. Vargas-Berenguel, Chem. Eur. J. 2009, 15, 710–
725.
[17] a) N. Huesken, G. Gasser, S. D. Koester, N. Metzler-Nolte, Bi-
oconjugate Chem. 2009, 20, 1578–1586; b) S. D. Koester, J. Dit-
trich, G. Gasser, N. Huesken, I. C. Henao Castaneda, J. L. Jios,
C. O. Della Védova, N. Metzler-Nolte, Organometallics 2008,
27, 6326–6332; c) J. Zagermann, K. Klein, K. Merz, M. Mo-
lon, N. Metzler-Nolte, Eur. J. Inorg. Chem. 2011, 30, 4212–
4219.
[18] C. A. Lipinski, F. Lombardo, B. W. Dominy, P. J. Feeney, Adv.
Drug Delivery Rev. 1997, 23, 3–25.
[19] D. J. Minick, J. H. Frenz, M. A. Patrick, D. A. Brent, J. Med.
Chem. 1988, 31, 1923–1933.
[11] M. Patra, N. Metzler-Nolte, Chem. Commun. 2011, 47, 11444–
11446.
[12] H. Abele, A. Haas, M. Lieb, Chem. Ber. 1986, 119, 3502–3506.
[13] A. Haas, H.-U. Krächter, Chem. Ber. 1988, 121, 1833–1840.
[14] J. D. Dunitz, L. E. Orgel, A. Rich, Acta Crystallogr. 1956, 9,
373–375.
[15] A. O. Borissova, M. Y. Antipin, D. S. Perekalin, K. A. Lys-
senko, CrystEngComm 2008, 10, 827–832.
[20] G. M. Sheldrick, shelxtl-97, University of Göttingen, Germany
1997.
[21] P. van der Sluis, A. L. Spek, Acta Crystallogr., Sect. A 1990,
46, 194–201.
[22] R. E. A. Dear, E. E. Gilbert, J. Org. Chem. 1968, 33, 819–823.
Received: July 14, 2012
[16] B. Neises, W. Steglich, Angew. Chem. 1978, 90, 556; Angew.
Chem. Int. Ed. Engl. 1978, 17, 522–524.
Published Online: ᭿
Eur. J. Inorg. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
7

Job/Unit: I20798
/KAP1
Date: 29-10-12 16:30:43
Pages: 8
M. Maschke, M. Lieb, N. Metzler-Nolte
FULL PAPER
Heterocyclic Sandwich Complexes
A series of new trifluoromethylated metall-
ocene triazoles was synthesized using
“click” chemistry (copper-catalyzed azide–
alkyne cycloaddition). The new triazoles
display suitable lipophilic character and ex-
ert promising antiproliferative activities
against a range of cancer cell lines.
M. Maschke, M. Lieb,
N. Metzler-Nolte* ............................. 1–8
Biologically Active Trifluoromethyl-Substi-
tuted Metallocene Triazoles: Characteri-
zation, Electrochemistry, Lipophilicity, and
Cytotoxicity
Keywords: Bioinorganic chemistry / Med-
icinal chemistry
/ Cytotoxicity / Click
chemistry / Fluorinated ligands / Metallo-
cenes / Sandwich complexes / N ligands /
Cycloaddition
8
www.eurjic.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 0000, 0–0