Grzegorz Mlostoń1, Mateusz Kowalczyk1,2, Małgorzata Celeda1, Katarzyna Gach-Janczak3, Anna Janecka3, Marcin Jasiński1. 1. Faculty of Chemistry, University of Lodz, 91403 Łódź, Poland. 2. The Bio-Med-Chem Doctoral School of the University of Lodz and Lodz Institutes of the Polish Academy of Sciences, Faculty of Biology and Environmental Protection, University of Lodz, 90237 Łódź, Poland. 3. Department of Biomolecular Chemistry, Medical University of Lodz, 92215 Łódź, Poland.
Abstract
A straightforward access to 2-unsubstituted imidazole N-oxides with subsequent deoxygenation by treatment with Raney-nickel followed by N-benzylation opens up a convenient route to lepidilines A and C. Both imidazolium salts were used to generate in situ the corresponding imidazol-2-ylidenes, which smoothly reacted with elemental sulfur, yielding imidazole-2-thiones. These reactions were performed either under classical conditions in pyridine solutions or mechanochemically using solid Cs2CO3 as a base. The structure of lepidiline C was unambiguously confirmed by X-ray analysis of its hexafluorophosphate. An analogous protocol toward lepidilines B and D and their 4,5-diphenyl analogues is less efficient due to observed instability of the key precursors, i.e., the respective 2-methylimidazole N-oxides. Comparison of cytotoxic activity against HL-60 and MCF-7 cell lines of all lepidilines, as well as their selected structural analogues (e.g., 4,5-diphenyl derivatives and PF6 salts), revealed slightly more potent activity of the 2-methylated series, irrespectively of the type of counterion present in the imidazolium salt. Remarkably, the well-known 1,3-diadamantylimidazolium bromide (the "Arduengo salt"), known as the precursor of the first, shelf-stable NHC representative, and its adamantyloxy analogue displayed the most significant cytotoxic activity in the studied series.
A straightforward access to 2-unsubstituted imidazole N-oxides with subsequent deoxygenation by treatment with Raney-nickel followed by N-benzylation opens up a convenient route to lepidilines A and C. Both imidazolium salts were used to generate in situ the corresponding imidazol-2-ylidenes, which smoothly reacted with elemental sulfur, yielding imidazole-2-thiones. These reactions were performed either under classical conditions in pyridine solutions or mechanochemically using solid Cs2CO3 as a base. The structure of lepidiline C was unambiguously confirmed by X-ray analysis of its hexafluorophosphate. An analogous protocol toward lepidilines B and D and their 4,5-diphenyl analogues is less efficient due to observed instability of the key precursors, i.e., the respective 2-methylimidazole N-oxides. Comparison of cytotoxic activity against HL-60 and MCF-7 cell lines of all lepidilines, as well as their selected structural analogues (e.g., 4,5-diphenyl derivatives and PF6 salts), revealed slightly more potent activity of the 2-methylated series, irrespectively of the type of counterion present in the imidazolium salt. Remarkably, the well-known 1,3-diadamantylimidazolium bromide (the "Arduengo salt"), known as the precursor of the first, shelf-stable NHC representative, and its adamantyloxy analogue displayed the most significant cytotoxic activity in the studied series.
Lepidilines A–D (1a–1d) belong to the class of imidazolium
alkaloids found in extracts prepared from roots of Lepidium
meyenii (so-called Maca), a South American plant known in
folk medicine of Peruvian Indian tribes for more than a thousand years.
Over centuries, aqueous extracts as well as dried roots of Maca were
used as a natural drug and as a food additive. Currently it is widely
explored as a popular dietary supplement easily available not only
in the pharmacy but also in the food markets.[1] Moreover, lepidilines A and C have been used as convenient precursors
of nucleophilic carbenes (NHCs) applied for the synthesis of bioactive
metal complexes with gold(I), silver(I), and iridium(I) ions.[2]Alkaloids 1a and 1b were isolated and
identified for the first time in 2003, and the structure of lepidiline
A was unambiguously proved by the X-ray analysis.[3] More recently, two nonsymmetric imidazolium salts, 1c and 1d (lepidilines C and D, respectively),
were also isolated from Maca extracts, and their structures were elucidated
on the basis of spectroscopic data analysis.[4] Noteworthy, whereas promising anticancer activity of lepidilines
A and B was discussed in the original report by Cui et al.,[3] no information is available on biological properties
of lepidilines C and D. Although lepidilines A and B are easily available
via double benzylation of 4,5-dimethyl- and 2,4,5-trimethylimidazole,[5] respectively, for the synthesis of their unsymmetrically
substituted analogues C and D the above standard alkylation procedure
cannot be applied. Thus, due to the current interest in the chemistry
and application of imidazolium salts, the development of general methods
for multigram scale synthesis of lepidilines A–D is of practical
importance.In our continuous research on the synthesis and
reactivity of imidazole N-oxides of type 2, we demonstrated that they
are superior building blocks for the preparation of diverse imidazole
derivatives.[6] As shown in Scheme , condensation of α-hydroxyiminoketones 3 with imines 4 followed by deoxygenation of
the first formed imidazole N-oxides 2 offers a convenient access to polysubstituted imidazoles 5 with excellent control on the substitution pattern. Thus, the method
allows for preparation of more complex imidazole derivatives bearing
either functionalized alkyl or aryl groups located at the N(1), C(2),
C(4), and C(5) atoms of the core heterocycle. In the quest for designed
lepidiline precursors, diacetyl monoxime (3a, R3 = R4 = Me) and benzyl formaldimines (R2 =
H) and acetimines (R2 = Me) of type 4 are
indicated as convenient starting materials. Benzylation of the key
4,5-dimethylimidazoles 5 should lead to desired alkaloids 1 (Scheme ).
Scheme 1
Multistep Synthesis of Lepidilines 1 Starting
with Condensation
of α-Hydroxyiminoketones 3 with Imines 4, Followed by Deoxygenation of the Initially Formed Imidazole N-Oxides 2 and N-Benzylation
of the Resulting Imidazoles 5
The goal of the present study was to elaborate a general method
for the preparation of the title compounds and their structural analogues,
such as imidazole-2-thiones available via intermediate nucleophilic
carbenes. In addition, an unambiguous confirmation of the structure
of a representative nonsymmetric alkaloid (lepidiline C or D) is also
of interest. Finally, cytotoxic activity of the selected imidazole-based
products was tested against two cancer cell lines, HL-60 and MCF-7,
and for comparison on normal HUVEC cells.
Results and Discussion
In our recent publication, 2-unsubstituted imidazole N-oxides such as 2 were used as key building blocks for
the preparation of a series of benzyloxy analogues of lepidiline A.[7] In the presented study, deoxygenation of N-oxides 2 was required to get the desired
1,4,5-tri- and 1,2,4,5-tetrasubstituted imidazoles. In an initial
experiment, the known 1-benzyl-4,5-dimethylimidazole N-oxide (2a)[8] was treated
with freshly prepared Raney-nickel, in EtOH, and the obtained 1-benzyl-4,5-dimethylimidazole
(5a) was N-alkylated with benzyl chloride
under microwave (MW) irradiation. The reaction was complete after
5 min, yielding the expected lepidiline A (1a) in nearly
quantitative yield (Scheme ). The same method was applied for the synthesis of the 4,5-diphenyl
analogue of 1a (i.e., compound 6a), but
in the case of benzylation with BnCl a very low conversion of ca.
5% was observed after 60 min of heating, while application of BnBr
as a more reactive electrophile provided the expected imidazolium
bromide 6a[Br] in 87% yield after 45 min. Furthermore,
the anion exchange aimed at preparation of hexafluorophosphates derived
from 1a and 6a[Br] was easily achieved by
treatment of the starting salts with NH4PF6 in
aqueous EtOH.
Scheme 2
Synthesis of Lepidiline A (1a) and Its
4,5-Diphenyl
Analogue 6a
To the best of our knowledge, synthesis of lepidiline C has not
yet been elaborated and reported. In our hands, imidazole 5a was successfully alkylated with m-methoxybenzyl
chloride under MW conditions to afford the expected imidazolium alkaloid 1c in 84% yield (Scheme ). The product was isolated as a viscous oil, which
after recrystallization from an i-Pr2O/CH2Cl2 mixture gave a colorless solid with a melting
point of 94–96 °C. The measured temperature of the Cr
→ I phase transition in 1c was clearly different
from that reported for lepidiline C isolated from natural sources
(mp 225–228 °C). Nevertheless the 1H and 13C NMR spectra of the obtained material corresponded well
with the reported chemical shifts.[4] Fortunately,
the anion exchange in 1c for PF6– enabled growth of fine single crystals suitable for X-ray analysis,
which unambiguously confirmed the expected structure of the imidazolium
cation in 1c[PF] (Figure ).[9] By analogy, the 4,5-diphenylimidazolium analogues
of lepidiline C, 6c and 6c[PF], were prepared using imidazole 5b as the starting material (Scheme ). Whereas replacement of Cl– by Br– has practically no impact on the chemical
shifts of signals in the 1H NMR spectra of imidazolium
salts of type 1 and 6, introduction of PF6– in the 2-unsubstituted series resulted
in a remarkable high-field shift of the diagnostic signals attributed
to C(2)-H. For example, the aforementioned singlets
for 1c, 1c[Br], and 1c[PF] were found at δ 10.80, 10.76,
and 8.62, respectively.
Scheme 3
Syntheses of Lepidiline C (1c), Its 4,5-Diphenylimidazolium
Analogue 6c, and the Anion Exchange Reactions Leading
to the Corresponding Hexafluorophosphates 1c[PF] and 6c[PF]
Figure 1
X-ray
analysis of imidazolium hexafluorophosphate 1c[PF] derived from lepidiline C.
X-ray
analysis of imidazolium hexafluorophosphate 1c[PF] derived from lepidiline C.In order to demonstrate flexibility of the presented
synthetic
method for the preparation of unsymmetric imidazolium salts, the same
lepidiline C and its diphenyl analogue 6c were prepared
in an alternative protocol using 1-(3-methoxybenzyl)-functionalized N-oxides 2c and 2d (Scheme ). For example, after smooth
deoxygenation of 2c followed by treatment of the resulting
imidazole 5c with BnCl, the expected 1c was
isolated in fair 54% overall yield (for two steps). These experiments
indicate that the preparation of the target unsymmetric imidazolium
salts can be readily achieved by using different sets of starting
materials, i.e., α-hydroxyiminoketones, benzylamines, and benzyl
halides, and thereby, the applied protocol enhances the utility of
this method for the preparation of differently substituted imidazolium
salts.The replacement of N-benzyl formaldimine
(4a) by the respective acetimine 4c in the
reaction
with α-hydroxyiminoketone 3a opened up a straightforward
access to imidazole N-oxide 2e considered
as a suitable precursor of lepidilines B and D (Scheme ). Indeed, reduction of 2e followed
by N-benzylation of 5e with benzyl chloride
or m-methoxybenzyl chloride provided desired alkaloids 1b and 1d, respectively. In contrast to 2e, attempted preparation of its 4,5-diphenyl analogue using
benzil monoxime (3b) and acetimine 4c was
unsuccessful, as the initially formed N-oxide suffered
decomposition during workup. We assume that the observed decomposition
of this imidazole N-oxide results from the anticipated
limited stability of its 3-hydroxy-2-methylidene tautomer. Apparently,
the presence of two Ph substituents located at C(4) and C(5) enables
the tautomeric rearrangement and in the presence of moisture leads
to unidentified, deeply red-colored product(s). For that reason we
waved on the synthesis of 1-benzyl-2-methyl-4,5-diphenylimidazole
(5f) via the respective imidazole N-oxide,
and instead the required imidazole 5f was obtained following
a known procedure based on multicomponent condensation of benzil,
acetaldehyde, benzylamine, and ammonium acetate in the presence of
InCl3 used as a catalyst.[10] Thus,
the synthesis of the target 4,5-diphenyl analogue of lepidiline D
(i.e., compound 6d) was achieved using the latter heterocycle 5f and m-methoxybenzyl chloride under MW
irradiation, which efficiently accelerated the quaternization process.
Scheme 4
Three-Step Synthesis of Lepidilines B (1b) and D (1d) Starting with N-Benzyl Acetimine (4c) and Diacetyl Monoxime (3a) (Above) and Benzylation
of 5f Leading to the 4,5-Diphenyl Analogue of Lepidiline
D, i.e., Imidazolium Salt 6d (Below)
Imidazole-2-thiones are known as biologically
active compounds,
which display diverse biological activity, and many representatives
are recognized as potent antimicrobial, antithyroid, anti-HIV, and
anticancer agents.[11] One of the relatively
new and attractive methods for the synthesis of non-enolizable imidazole-2-thiones
comprises sulfurization of transient imidazol-2-ylidenes with elemental
sulfur.[12] In a recent report, this method
was successfully applied for the conversion of numerous benzyloxy-functionalized
imidazolium salts into the corresponding imidazole-2-thiones.[7] Thus, lepidilines A and C seem to be attractive
substrates for further functionalization via the respective intermediate
carbenes (NHCs). In a typical experiment, imidazolium chloride 1a was treated with Et3N and S8 in dry
pyridine solution at room temperature (Scheme , Method A). After overnight stirring the
expected imidazole-2-thione 7a was isolated as a crystalline
product, although in low 34% yield. In the search for a more efficient
protocol, the ball-mill approach was checked by using Cs2CO3 as a base and a 2-fold excess of elemental sulfur,
in the presence of butanone as liquid-assisted grinding solvent (LAGs)
(Method B). To our delight, the expected imidazole-2-thione 7a was formed solely, and the product was isolated in an excellent
yield of 97%. The 13C NMR spectrum of 7a confirmed
the presence of the thiourea unit in the molecule, as the typical
resonance of this diagnostic group was found at δ 162.7. Analogous
procedures applied for lepidiline C afforded the respective imidazole-2-thione 7c isolated as a colorless solid in 53% (Method A) and 73%
(Method B) yield. Two more products of that type (i.e., compounds 8a and 8c) with a 4,5-diphenylimidazole motif
were also obtained in an analogous manner.
Scheme 5
Sulfurization of
the in Situ Generated Imidazol-2-ylidenes
Derived from Lepidilines 1a,c and Their
4,5-Diphenylimidazolium Analogues 6a,c Leading
to Non-enolizable Imidazole-2-thiones 7 and 8
Potential biological activity
of all four lepidilines A–D
(1a–1d), as well as their 4,5-diphenyl
analogues 6a[PF], 6c[PF], and 6d[PF] (as hexafluorophosphate
salts) and a series of imidazole-2-thiones 7 and 8 obtained therefrom, was evaluated in vitro against two human
cancer cell lines: promyelocytic leukemia HL-60 and breast cancer
adenocarcinoma MCF-7. Selected analogues were also tested against
human umbilical vein endothelial cells (HUVECs) using the MTT cytotoxicity
assay. Concentration–response analysis was performed to determine
drug concentrations required to inhibit the growth of cells by 50%
(IC50) after 48 h of incubation. The obtained results are
summarized in Table and compared with activity of the known anticancer agent doxorubicin
(DXR) used as a positive control.[13]
Table 1
Cell Growth Inhibitory Activity of
the Selected Imidazolium Salts 1, 6, 9, and 10 (Ad = Adamantan-1-yl), Imidazole-2-thiones 7 and 8, and the Reference Doxorubicin (DXR)
IC50 [μM]a
entry
cmpd
R
R′
Ar
Ar′
X
HL-60
MCF-7
HUVEC
1
1a
Me
H
Ph
Ph
Cl
32.3 ± 3.5
>100
>100
2
1b
Me
Me
Ph
Ph
Cl
3.8 ± 0.2
>100
>100
3
1c
Me
H
m-MeOC6H4
Ph
Cl
27.7 ± 2.5
75.0 ± 5.0
>100
4
1d
Me
Me
m-MeOC6H4
Ph
Cl
1.1 ± 0.1
>100
>100
5
1a[PF6]
Me
H
Ph
Ph
PF6
32.0 ± 2.0
>100
6
1c[PF6]
Me
H
m-MeOC6H4
Ph
PF6
19.9 ± 0.2
69.9 ± 0.2
7
6a[PF6]
Ph
H
Ph
Ph
PF6
1.2 ± 0.1
8.2 ± 0.2
8
6c[PF6]
Ph
H
m-MeOC6H4
Ph
PF6
1.2 ± 0.2
7.9 ± 0.3
8.8 ± 0.1
9
6d[PF6]
Ph
Me
m-MeOC6H4
Ph
PF6
1.2 ± 0.1
6.5 ± 0.2
10
7a
Me
H
Ph
>100
>100
11
7c
Me
H
m-MeOC6H4
20.2 ± 1.9
>100
12
8a
Ph
H
Ph
8.1 ± 0.3
>100
13
8c
Ph
H
m-MeOC6H4
55 ± 5
>100
14
9
1.8 ± 0.1
51.5 ± 1.5
15
10
0.3 ± 0.1
4.5 ± 0.1
9.1 ± 0.3
16
DXR
0.12 ± 0.01
0.9 ± 0.1
1.4 ± 0.1
Compound concentration
required
to inhibit metabolic activity by 50%. The cells were incubated with
the analogues for 48 h. Values are expressed as mean ± SEM from
the concentration–response curves of at least three experiments
using a nonlinear estimation (quasi-Newton algorithm) method.
Compound concentration
required
to inhibit metabolic activity by 50%. The cells were incubated with
the analogues for 48 h. Values are expressed as mean ± SEM from
the concentration–response curves of at least three experiments
using a nonlinear estimation (quasi-Newton algorithm) method.Lepidilines B and D were found to
be significantly cytotoxic against
HL-60 cells with IC50 values in the low micromolar range
of 3.8 and 1.1 μM, respectively, which were an order of magnitude
lower than those obtained for lepidilines A and C (i.e., 32.3 and
27.7 μM, respectively). None of the lepidilines showed similarly
high cytotoxicity on the MCF-7 cell line, and lepidiline D turned
out to be over 100-fold more cytotoxic for leukemia than for breast
cancer cells. In the experiments performed on normal (HUVEC) cells
the IC50 values for all four lepidilines A–D were
over 100 μM, which indicates a large safety margin for these
compounds.Replacement of the chloride counterion in lepidilines
A and C by
hexafluorophosphate (i.e., compounds 1a[PF] and 1c[PF], respectively) did not result in increased activity
(Table , entries 5
and 6).It is worth stressing that this is the first report
in which the
cytotoxicity of all four representatives of the lepidiline family
has been checked against selected cell lines, which made it possible
to compare their activity under identical conditions. In a previous
report by Zheng only lepidilines A and B were examined against eight
human cancer cell lines.[3] The mentioned
work indicated lepidiline A was inactive against all those cell lines
with IC50 values above 10 μM (the values reported
in the paper are expressed in μg/mL). In contrast, lepidiline
B showed cytotoxic activity against several cell lines, especially
high against pancreatic adenocarcinoma PACA2 (IC50 = 1.38
μg/mL, i.e., 4.2 μM) and against breast carcinoma MDA-231
(IC50 = 1.66 μg/mL, i.e., 5.1 μM). Unfortunately,
it is difficult to compare those results with our data obtained on
different cell lines. However, the common observation from both studies
is that lepidiline B is more cytotoxic than lepidiline A.Analysis
of the structure–activity relationship of 4,5-diphenyl
analogues of lepidilines A, C, and D (Table , entries 7–9) revealed that the replacement
of methyl by phenyl at C(4) and C(5) of the imidazole ring caused
a large increase of cytotoxicity against the MCF-7 cell line and in
the case of 6c[PF] also against HUVECs, making these analogues much less selective
as compared with natural lepidilines.Introduction of a sulfur
atom at C(2) in a series of imidazole-2-thiones 7 and 8 derived from lepidilines A and C and
their analogues bearing Ph substituents at C(4) and C(5) atoms of
the imidazole ring was not advantageous for activity, especially against
MCF-7 cells (Table , entries 10–13). This decreased cytotoxicity may result from
lower bioavailability of imidazole-2-thiones, which in contrast to
lepidilines are not charged molecules. It is well known that the neutral
organic compounds interact with membranes only through hydrophobic
bonds, whereas charged substances can additionally benefit from electrostatic
interactions.[14]In contrast to imidazole-2-thiones,
two highly oleophilic imidazolium
salts 9 and 10 bearing at N(1) and N(3)
atoms either adamantan-1-yl (the “Arduengo salt”)[15] or adamantan-1-yloxy groups,[16] respectively, were found to be significantly cytotoxic
for HL-60 cells. Notably, imidazolium salt 10 was the
most active among all tested compounds (IC50 = 0.3 and
4.5 μM on HL-60 and MCF-7 cell lines, respectively) and also
displayed some selectivity.As a positive control in the MTT
assay a well-known anticancer
drug, DXR, widely used to treat breast cancer and acute lymphocytic
leukemia among other cancer types,[13] has
been used. The IC50 values for DXR against HL-60 and MCF-7
cells were below 1 μM (0.12 and 0.9 μM against HL-60 and
MCF-7 cells, respectively), and the value obtained for normal HUVEC
cells was 1.4 μM under the same experimental conditions. Thus,
DXR was 10-fold more cytotoxic than lepidiline D, but the HUVEC/HL-60
IC50 ratio for DXR was about 10 and that for lepidiline
D over 100.In summary, the present study showed that the title
lepidilines
A–D can be conveniently prepared using imidazole N-oxides as key intermediates, which are readily available via cyclocondensation
of diacetyl monoxime with N-benzyl aldimines. Initial
deoxygenation with Raney-nickel followed by microwave-assisted benzylation
leads to both symmetric and nonsymmetric imidazolium alkaloids in
excellent yield and purity. X-ray analysis of the imidazolium salt
derived from synthetic lepidiline C was presented for the first time
and confirmed the postulated structure of the naturally occurring
material. In addition, sulfurization of nucleophilic carbenes derived
from lepidilines A and C enables convenient preparation of corresponding,
hitherto unknown, non-enolizable imidazole-2-thiones.Our results
contribute to the development of methods useful for
synthesis of naturally occurring, biologically active imidazole derivatives
relevant for medicinal chemistry and related applications.[17,18] Progress in this attractive field was summarized in a very recent,
comprehensive review.[19] Further applications
of imidazolium salts of “lepidiline type” in coordination,
organometallic, and materials chemistry are also possible.[18,19]The biological results presented here revealed significant
cytotoxicity
of lepidilines B and D in the tested series of naturally occurring
alkaloids and, most importantly, their remarkable selectivity against
leukemia HL-60 versus normal HUVEC cells. These compounds as well
as their 4,5-diphenyl imidazolium derivatives and the presented adamantyloxy
analogue of the “Arduengo salt”[15] can be considered not only as readily available precursors of nucleophilic
carbenes[2] but also as useful probes in
the search for new leads in antileukemic drug discovery.
Experimental Section
General Experimental Procedures
Melting points were
determined in capillaries with a MEL-TEMP apparatus (Aldrich) and
are uncorrected. The IR spectra were measured neat with an Agilent
Cary 630 FTIR spectrometer. NMR spectra were measured on a Bruker
AVIII instrument (1H at 600 MHz, 13C at 151
MHz). Chemical shifts are reported relative to residual nondeuterated
solvent signals (for CDCl3: 1H NMR: δ
7.26, 13C NMR: δ 77.16; for acetone-d6: 1H NMR: δ 2.09, 13C NMR:
δ 30.60).[20] MS (ESI) was performed
with a Varian 500-MS LC ion trap; high-resolution MS (ESI-TOF) measurements
were performed with a Synapt G2-Si mass spectrometer (Waters). Nonionic
products were purified by standard column chromatography (CC) on silica
gel (230–400 mesh) or by preparative thin-layer chromatography
(PTLC); organic salts were purified by recrystallization from i-Pr2O or hexanes/CH2Cl2 mixtures. Mechanochemical reactions were performed by using a Retsch
Mixer Mill MM400. Elemental analyses were obtained with a Vario EL
III (Elementar Analysensysteme GmbH) instrument. Commercially available
solvents and starting materials were used as received. Starting materials,
i.e., diacetyl monoxime (3a),[21] benzil monoxime (3b),[22] formaldimine 4a(23) (in the form of a trimer,
i.e., 1,3,5-tribenzylhexahydro-1,3,5-triazine), and acetimine 4c(24) (monomeric) were prepared
following the general literature procedures.
Synthesis of Imidazolium
Chlorides and Bromides of Type 1 and 6
To a deoxygenated solution of
imidazole 5 (1.0 mmol) in MeCN (10 mL) was added benzyl
halide (1.5 mmol), and the resulting mixture was MW-irradiated at
110 °C until the starting imidazole was fully consumed (TLC monitoring,
usually up to 60 min). The solvent was removed under reduced pressure,
and the crude product was washed with several portions of dry Et2O. Solid products were recrystallized from a CH2Cl2/hexanes mixture.
The crude product was purified
by preparative thin-layer chromatography (SiO2, CH2Cl2/MeOH, 92:8); spectroscopically pure sample
of 6d (178 mg, 37%) was isolated as a colorless oil and
used for the next step without further purification. 1H
NMR (600 MHz, CDCl3) δ 7.39–7.36 (m, 2H),
7.34–7.22 (m, 12H), 7.04–7.01 (m, 2H), 6.83–6.80
(m, 1H), 6.60–6.58 (m, 1H), 6.53 (mc, 1H), 5.54
(s, 2H), 5.52 (s, 2H), 3.74 (s, 3H), 2.82 (s, 3H); 13C
NMR (151 MHz, CDCl3) δ 160.3, 136.0, 135.6, 133.9,
132.3, 132.2, 131.0 (4C), 130.6, 130.5 (2C), 129.5 (2C), 129.2 (4C),
128.7, 126.4 (2C), 125.17, 125.15, 118.3, 113.9, 112.2, 55.5, 49.74,
49.67, 12.8.
Synthesis of Imidazolium Hexafluorophosphates 1[PF] and 6[PF]
To a solution of
imidazolium
chloride or bromide of type 1 or 6 (0.30
mmol) in EtOH (1.0 mL) was added dropwise a solution of NH4PF6 (54 mg, 0.33 mmol) in H2O (1.0 mL), and
the mixture was stirred for 30 min. The crude oily or crystalline
product was isolated, washed with dry Et2O (3 × 4
mL), and recrystallized from a CH2Cl2/i-Pr2O mixture (by slow evaporation of the solvents
at room temperature).
To
a solution of appropriate α-hydroxyiminoketone
of type 3 (9.0 mmol) in glacial acetic acid (20 mL) was
added a portion of formaldimine 4 (10 mmol), and the resulting mixture
was stirred at room temperature overnight. Then, excess concentrated
HCl was added (4 mL), the solvents were removed under reduced pressure,
and the resulting product was dissolved in MeOH. After excess solid
NaHCO3 was added the stirring was continued for 2 h until
the evolution of CO2 ceased. The solvent was removed in
vacuo, the residue was triturated with CH2Cl2, the precipitate salts were filtered off, the solvent was evaporated,
and the residue was washed with a few portions of Et2O
to give imidazole N-oxides 2, which
were used for the next step without further purification.
This compound was prepared by a modified procedure as follows:
To a solution of diacetyl monooxime (3a, 203 mg, 2.0
mmol) in EtOH (4 mL) was added an excess of freshly prepared formaldimine 4c in two portions (first portion: 532 mg, 4.0 mmol; the second
portion of 133 mg, 1.0 mmol was added after 24 h), and the resulting
mixture was stirred at room temperature for 48 h. The solvent was
removed and the crude 2e was used for next step without
purification. Colorless oil, 350 mg (∼81%); 1H NMR
(600 MHz, CDCl3) δ 7.33–7.28 (m, 3H), 6.95–6.92
(m, 2H), 5.00 (s, 2H), 2.39 (s, 3H), 2.21 (s, 3H), 2.06 (s, 3H); 13C NMR (151 MHz, CDCl3) δ 135.1, 129.3 (2C),
128.3, 127.3, 125.7 (2C), 125.6, 119.2, 47.4, 8.82, 8.44, 7.71.
Synthesis of Formaldimine 4b
A mixture
of 3-methoxybenzylamine (5.0 g, 36.5 mmol) and aqueous formaldehyde
(37%, 3.26 g, 40.0 mmol) in benzene (60 mL) was refluxed in a Dean–Stark
apparatus for 1.5 h. The solvent was removed under reduced pressure
to give a thick yellow oil (5.3 g, 97%) identified as the trimeric
form of 4b, which was used for the next step without
further purification.
1,3,5-Tri(3-methoxybenzyl)hexahydro-1,3,5-triazine
(trimer of 4b):
1H NMR (600 MHz,
CDCl3) δ 7.21–7.18 (m, 3H), 6.94–6.91
(m, 6H), 6.79–6.77
(m, 3H), 3.80 (s, 9H), 3.67 (s, 6H), 3.43 (sbr, 6H); 13C NMR (151 MHz, CDCl3) δ 157.9, 140.3, 129.3,
121.2, 114.2, 112.7, 73.9, 57.1, 55.3; anal. calcd for C27H33N3O3 (447.3) C 72.46, H 7.43,
N 9.39; found C 72.48, H 7.38, N 9.45.
Synthesis of Imidazoles 5
To a solution
of imidazole N-oxide 2 (2.0 mmol) in
MeOH (5.0 mL) was added portionwise an excess of freshly prepared
suspension of Raney-nickel in MeOH. The resulting mixture was stirred
at room temperature until the starting N-oxide was
fully consumed (monitored by TLC; typically ca. 1 h). The solids were
filtered off and the solvent was removed in vacuo to give spectroscopically
pure imidazole 5.
This compound was
prepared following
a general literature protocol[10] in a 4.0
mmol scale; the crude product 5f was isolated by preparative
thin-layer chromatography (SiO2, CH2Cl2/MeOH, 98:2): yellow oil, 143 mg (11%); 1H NMR (600 MHz,
CDCl3) δ 7.54–7.52 (m, 2H), 7.40–7.21
(m, 10H), 7.17–7.14 (m, 1H), 6.97–6.95 (m, 2H), 4.98
(s, 2H), 2.42 (s, 3H).
General Procedures for Synthesis of Imidazole-2-thiones 7 and 8
Method A: Imidazolium chloride 1 or 6 (0.19 mmol) and elemental sulfur (0.46
mmol) in a pyridine/Et3N mixture (1:1, 3.0 mL) were stirred
at room temperature overnight. The solvents were removed in vacuo,
and the obtained residue was purified by PTLC (using CH2Cl2 as an eluent) to give products isolated as solid materials.Method B: Imidazolium salt 1 or 6 (0.3
mmol), solid Cs2CO3 (0.45 mmol), elemental sulfur
(S8, 0.75 mmol), and zircon grinding balls were placed
in a mechanochemical reactor, and the mixture was ground for 60 min.
The resulting crude mixture was treated with CH2Cl2 (10 mL), and the solids were filtered off and washed with
three portions of CH2Cl2 (5 mL each). After
the solvent was removed under reduced pressure the crude product was
purified by PTLC using CH2Cl2 as an eluent.
Pale yellow crystals of compound 1c[PF] were obtained
from a CH2Cl2/i-Pr2O mixture by slow evaporation of the solvents at room temperature.
A suitable crystal was measured on an XtaLAB Synergy, Dualflex, Pilatus
300 K diffractometer. The crystal was mounted in inert oil on nylon
loops and kept at 100.00(10) K during data collection. Measurements
for compound 1c[PF] were performed using mirror-focused Cu Kα radiation, λ
= 1.541 84 Å. Absorption corrections were implemented
on the basis of multiscans. Using Olex2,[27] the structure was solved with the XT[28] structure solution program using intrinsic phasing and refined with
the XL[29] refinement package using least
squares minimization. Hydrogen atoms were included using rigid methyl
groups or a riding model starting from calculated positions. Complete
data have been deposited with the Cambridge Crystallographic Data
Centre under the number CCDC-2059690. Copies of the data can be obtained
free of charge from www.ccdc.cam.ac.uk/structures/.C20H23F6N2OP, M = 452.37: triclinic, space group P1̅ (no. 2), a = 9.8930(3) Å, b = 11.1547(4) Å, c = 11.1640(4) Å,
α = 60.849(4)°, β = 84.409(3)°, γ = 71.453(3)°, V = 1017.60(7) Å3, Z =
2, T = 100.00(10) K, μ(Cu Kα) = 1.834
mm–1, Dc = 1.476 g cm–3, 23 177 reflections measured (9.092°
≤ 2θ ≤ 157.666°), 4125 unique (Rint = 0.0370, Rsigma = 0.0187),
which were used in all calculations. The final R1 was 0.0319 (I > 2σ(I)) and wR2 was 0.0824 (all data).
Cell Culture
and Treatment
Human promyelocytic leukemia
(HL-60) and human breast cancer adenocarcinoma (MCF-7) cell lines
were obtained from the European Collection of Cell Cultures, while
HUVECs were purchased from the American Type Culture Collection. HL-60
cells were cultured in RPMI 1640 plus GlutaMax I medium (Gibco/Life
Technologies, Carlsbad, CA, USA). MCF-7 cells were maintained in minimum
essential medium Eagle (Sigma-Aldrich, St. Louis, MO, USA) supplemented
with 2 mM glutamine and Men nonessential amino acid solution (Sigma-Aldrich).
Both media were supplemented with 10% heat-inactivated fetal bovine
serum (Biological Industries, Beit-Haemek, Israel) and antibiotics
(100 U/mL penicillin and 100 μg/mL streptomycin) (Sigma-Aldrich).
HUVEC cells were grown in EGM-2 endothelial medium BulletKit (Lonza,
Basel, Switzerland). Cells were maintained at 37 °C in a 5% CO2 atmosphere and were grown until 80% confluent.The
tested compounds were dissolved in sterile dimethyl sulfoxide (DMSO)
and further diluted with culture medium. The final concentration of
DMSO in cell cultures was less than 0.1% v/v. Controls without and
with 0.1% DMSO were performed in each experiment. At the used concentration
DMSO had no effect on the observed parameters.
Cytotoxicity
Determination (MTT Assay)
The MTT assay
was performed according to the known procedure.[30] The cells were incubated with the analogues for 48 h. The
absorbance of the blue formazan product was measured at 560 nm using
a FlexStation 3 multi-mode microplate reader (Molecular Devices, LLC,
CA, USA) and compared with control (untreated cells). All experiments
were performed in triplicate.
Authors: Rubén M Carballo; José M Padrón; Israel Fernández; Daniel A Cruz; Luana Grmuša; Víctor S Martín; Juan I Padrón Journal: J Org Chem Date: 2022-08-03 Impact factor: 4.198