An unusual substitution reaction directed by an intramolecular re-arrangement

Article abstract of DOI:10.1016/j.tet.2005.06.074

Secondary amines and thiols undertake a substitution reaction on the side chain of 2-bromoethyl-pyridinium derivatives 'directed' by an intramolecular re-arrangement. Experimental investigations strongly indicate that the reaction is initiated by an alpha

Full text of DOI:10.1016/j.tet.2005.06.074

Tetrahedron 61 (2005) 8410–8418
An unusual substitution reaction directed by an
intramolecular re-arrangement
Alexis D.C. Parenty, Louise V. Smith and Leroy Cronin
*
Department of Chemistry, University of Glasgow, Joseph Black Building, Glasgow G12 8QQ, UK
Received 27 April 2005; revised 9 June 2005; accepted 23 June 2005
Available online 18 July 2005
Abstract—Secondary amines and thiols undertake a substitution reaction on the side chain of 2-bromoethyl-pyridinium derivatives
‘
directed’ by an intramolecular re-arrangement. Experimental investigations strongly indicate that the reaction is initiated by an alpha
addition of the nucleophile onto the iminium moiety of the N-heteroaromatic cation, followed by a cyclisation and an oxidative ring opening.
This novel substitution process is able to occur with less reactive nucleophiles that would not undergo conventional substitution with
‘isolated’ bromoethyl moieties.
q 2005 Elsevier Ltd. All rights reserved.
1
. Introduction
alpha addition step. The alternative pathway, (Path B,
Scheme 1), was ruled out when experiments highlighted the
poor reactivity of the 2-bromoethyl side chain, as well the
It is almost impossible to imagine a biological process that
does not involve a heterocyclic compound. Heterocycles
1
12,13
necessity for a disfavoured 5-endo-trig-cyclisation.
also form the building blocks of many important pharma-
ceuticals. One of the reasons for the widespread use of
2
heterocyclic compounds in nature as well as in the
pharmaceutical industry is their implication in a wide
range of reaction types allowing subtle structural modifi-
3
–5
cation.
Therefore, investigations working towards the
understanding of the reactivity of heterocyclic compounds
is of great practical significance. Efficient syntheses leading
to N-heteroaromatic cations are especially interesting as
they often form the framework of DNA intercalating agents
6
,7
and have anticancer properties. In this respect, molecules
containing a phenanthridinium core are one important
subset of heteroaromatic cations.
8
–11
Recently, we have discovered a flexible methodology
1
2
allowing the ring extension of N-heterocyclic cations.
Specifically, we found that a primary amine reacted with
-bromoethyl-phenanthridinium leading to the correspond-
2
ing dihydro-imidazo-phenanthridinium derivative (DIP) via
a nucleophilic addition followed by a cyclisation and an
in-situ oxidation step (Path A, Scheme 1).
Evidence strongly suggesting Path A as a mechanism was
confirmed by isolation of intermediates demonstrating the
Keywords: N-Heterocyclic; Substitution; Intramolecular re-arrangement;
Mechanism; Phenanthridinium.
Scheme 1. The cascade reaction leading to the Dihydro-Imidazo-
Phenanthridinium DIP derivative. Note that the hydride loss could also
be viewed as two consecutive electron transfers and a proton loss.
12
*
Corresponding author. Tel.: C44 141 330 6650; fax: C44 141 330 4888;
e-mail: l.cronin@chem.gla.ac.uk
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.06.074

A. D. C. Parenty et al. / Tetrahedron 61 (2005) 8410–8418
8411
Table 1. Products and yields
Entry
Structure 5
Secondary amine 2
Yield %
71
5a
Piperidine
5
b
Piperazine
73
93
5c
1,5,9-Triaza-cyclododecane
5d
p-Methoxy methyl aniline
77
Following the discovery of this useful one-pot-method-
ology, we were interested in utilising nucleophiles other
than primary amines: for example, what would be the result
if a nucleophile with only one available valence is used with
the same reaction conditions? Herein, we report a surprising
reaction that allows substitution of the bromo-ethyl side
chain via a ‘directed’ cyclisation process with secondary
amines and thiol-derivatives.
Scheme 3). This intricate mechanism would involve an
alpha addition on the iminium moiety of molecule 1, followed
by a ring cyclisation and a retro-5-endo-trig ring opening
leading to re-oxidation of the heterocyclic middle ring.
Strong arguments in favour of Path A mechanism of
Scheme 3 were obtained in the course of our experiments.
Firstly, the reactivity of the iminium moiety of starting
material 1 was assessed by reacting piperidine with the
fluoro- and hydroxyl analogues of molecule 1 (respectively,
1F and 1OH, Scheme 4). These two analogues, with their
iminium moiety assumed to be as reactive as the one of
molecule 1 (this assumption was made on the basis of DFT
2
. Results and discussion
Under mild reaction conditions, the starting material
-bromoethyl-phenanthridinium bromide 1 reacts with a
series of secondary amines 2 to give the corresponding
-aminoethyl-phenanthridinium bromide derivatives 5a–d
Scheme 2 and Table 1).
2
2
(
3 2
Scheme 2. Reagents and conditions: (a) DMF, Et N, N , rt, 48 h.
Upon initial inspection, this result seems to be in direct
contradiction with the previously published Path A
mechanism shown in Scheme 1, which states that direct
bromine substitution on the side chain of molecule 1, that is,
1
2
Path B, does not take place. Nevertheless, another
pathway, avoiding direct substitution could also explain
the formation of substituted products 5a–d (Path A,
Scheme 3. The two possible mechanistic pathways leading to 5a–d.
Reagents and conditions: (a) DMF, Et N, N , rt, 48 h.
3
2

8412
A. D. C. Parenty et al. / Tetrahedron 61 (2005) 8410–8418
Scheme 4. NMR phase transfer experiment with fluoro and hydroxyl
analogue of molecule 1.
1
4
calculations), have the advantage of being unable to
undertake the cyclisation of the hypothetic second step of
Path A (Scheme 3). For reason of simplicity, and in order to
evaluate how quickly the alpha addition step takes place, the
Figure 1. Absorbance at 320 nm versus time showing the formation of 5
from the reaction of 1 with piperidine highlighting the de-aromatisation/re-
aromatisation process expected for Path A.
reaction was undertaken in a biphasic system with D O/
2
CDCl ; this was done in practice in a NMR tube, where 1 or
F
3
1OH were partitioned between the organic and aqueous layers.
Upon addition of piperidine, a white precipitate appears
instantaneously in the D O layer and shifts towards the bottom
2
1
CDCl layer where H and C NMR measurements wereused
13
3
to characterize the corresponding alpha adduct 3 or 3
F
OH
(
Scheme 4), analogues of intermediate 3 (Scheme 3).
The efficiency of the alpha addition step was also realised by
isolating and analysing the D O layer during an identical
2
NMR phase transfer reaction: following the addition of
piperidine to the biphasic system, no starting material could
be detected after only 1 min of reaction time.
Figure 2. The alpha adduct intermediate(s) 3 and/or 4 can be observed by
UV spectroscopy. lmax for the starting and final material is ca. 320 nm.
This NMR phase transfer experiment highlights the high
reactivity of the iminium moiety of starting material 1.
Moreover, other examples of alpha addition of secondary
amines onto N-heteroaromatic cations like phenanthridi-
synthesized. This is because, under the mild reaction
conditions implemented (Scheme 3), it seems unlikely that
the weakly nucleophilic aromatic amine p-methoxy-methyl-
aniline is capable of performing a direct substitution on the
ethyl-bromine side chain of starting material 1 (Path B). On
the contrary, in the synthesis of 5d via Path A, the ethyl-
bromine side chain substitution of the corresponding
aromatic amine intermediate 3 (Scheme 3) would be
favoured entropically by the formation of a five-membered
ring. Therefore, to test the reactivity of the ethyl-bromine
side chain of 1 on its own, p-methoxy-methylaniline was
added to a DMF solution of 1,2-dibromoethane, using the
same reaction conditions as for the synthesis of 5a–d. Under
these mild conditions, no substitution on this bromoethyl-
analogue of 1 could be detected by TLC. The same test was
repeated with 2-bromoethyl-trimethyl-ammonium bromide,
with identical outcome. Therefore, assuming that those two
bromoethyl derivatives have reactivity similar to the
bromoethyl side chain of staring material 1, the isolation
of final molecule 5d is not likely to occur via Path B. This
result also highlights the more effective substitution reaction
of Path A allowing somewhat weaker nucleophiles to
ultimately undergo this substitution reaction. Consequently,
the alpha addition step of Path A can be seen as a ‘directing
step’ helping the side chain substitution leading to 5.
1
5,16
17
18,19
nium, isoquinolinium and pyridinium
also reported to lead to stable products.
systems are
Secondly, further support for the involvement of the central
aromatic ring in the substitution reaction of Scheme 3 was
obtained by UV absorption spectroscopy measurements.
The absorption was measured between 200 and 500 nm and
the lmax determined to be at ca. 320 nm. If the reaction
proceeds via Path A, the UV absorption of the de-aromatized
intermediates (3 and 4), at this wavelength, should be different
from both the starting material 1 and the final product 5.
Subsequently, the UV absorption was continuously mon-
itored at 320 nm over a period of 12 h. On addition of the
piperidine to a solution of starting material 1, the
absorbance drops instantly from 0.75 to 0.50 (formation of
intermediates 3 and 4) and rises back slowly to 0.75 due to
the formation of final product 5 (Fig. 1).
This data allow us to conclude that the intermediate is present for
at least 2 h. Therefore, in a subsequent reaction, a UV spectrum of
the starting material was taken between 200 and 500 nm, and
compared with both the UV spectrum of the intermediate after
3
0 min and the final product after 12 h reaction time (Fig. 2). The
In order to examine the generality of the reaction to other
N-heteroaromatic cations with an available alpha position,
the reaction was also attempted using 2-bromoethyl-quino-
line 7 to obtain with good yield the corresponding
difference in absorbance between the intermediate and the
starting material confirms the de-aromatisation process
involved in the proposed mechanism of Path A.
2
-aminoethyl-quinolinium derivative 8 (Scheme 5).
Lastly, the hypothetical Path B depicted in Scheme 3, is not
consistent with the observation that compound 5d can be
In this context, it is interesting to note that the

A. D. C. Parenty et al. / Tetrahedron 61 (2005) 8410–8418
8413
functionalization of N-heteroaromatic systems with a
-aminoethyl side chain is often encountered in DNA
intercalators where the resulting tertiary amine is either used
2
2
0–24
as a spacer in bis-intercalating agents,
5–29
structural/solubility/charge reasons.
or used for
2
In addition, the substitution reaction of 2-bromoethyl-
phenanthridinium 1 was also successful with thiol-deriva-
tives. The nucleophile p-methoxybenzyl mercaptan leads to
the corresponding substituted 2-mercaptoethyl-phenanthri-
dinium derivative 9 in good yield (Scheme 6).
Scheme 5. Reaction on quinolinium system. Reagents and conditions:
a) DMF, Et N, N , rt, 48 h.
(
3
2
Figure 3. Phase transfer reaction in a ‘U-Tube’ highlighting the existence of an organic soluble intermediate. RZp-methoxybenzyl.

8414
A. D. C. Parenty et al. / Tetrahedron 61 (2005) 8410–8418
4. Experimental
4
.1. General
All reactions were carried out using oven-dried glassware
under a nitrogen atmosphere using standard Schlenk line
techniques. Commercial starting materials and solvents
1
were used as supplied, without further purification. H and
C NMR were recorded using a Bruker DPX 400
1
3
spectrometer operating at 400 and 100 MHz, respectively.
Chemical shifts (d) are given in ppm relative to residual
solvent peak. Coupling constants (J) are given in Hz. Infra-
red spectral analysis were performed on a JASCO 410
spectrophotometer, using a KBr disc unless otherwise
Scheme 6. Substitution reaction with a thiol derivative. Reagents and
conditions: DMF, Et N, N , rt, 48 h.
3
2
K1
Moreover, indirect proof of the alpha adduct intermediates
could be obtained, using a phase transfer reaction in a
stated; peaks are quoted in wave numbers (cm ) and
their relative intensity are reported as follows: sZstrong,
mZmedium, wZweak. Mass spectra were obtained using a
JEOL JMS 700 spectrometer operating, in FAB, EI, CI or
ES mode. Microanalyses were performed on a CE-440
elemental analyzer. Melting points were determined on a
digital IA9000 series melting point apparatus, using
capillary tubes. UV spectrums were recorded on a Shimadzu
UV-310PC UV–VIS–NIR Scanning Spectrophotometer.
‘
CCl in the bottom (compartment 2, Fig. 3), was filled with
U-tube’ (Fig. 3). In this experiment, a ‘U-tube’ containing
4
water in one side (compartment 1, Fig. 3) and D O in the
2
other (compartment 3, Fig. 3). Molecule 1 was added to
compartment 1, followed by p-methoxybenzyl mercaptan
and triethylamine. The organic layer of compartment 2 was
gently stirred with a small magnet, taking care not to
exchange any water from compartment 1 with D O from
2
compartment 3. After 48 h, the deuterated solution of
Definitions of abbreviations: DMF, dimethylformamide;
DIP, dihydro-imidazo-phenanthridinium bromide; TEA,
triethylamine; rt, room temperature.
1
13
compartment 3 was analyzed by H and C NMR
spectroscopy, and final molecule 9 was characterized. It is
unlikely that the cationic starting material 1 nor the cationic
final product 9 could diffuse from one aqueous medium to
the other through the organic medium. Therefore, the
occurrence of molecule 9 in compartment 3 can be best
explained by the formation of an intermediate soluble in the
organic phase, being able to diffuse between the two
aqueous layers. Furthermore, the exchange of water
between the two opposite layers (i.e., D O with H O and
4.2. Synthesis and analytical data
4.2.1. General procedure for the synthesis of
2-bromoethyl-pyridinium derivative (1 and 7). Phenan-
thridine or quinoline (30.3 mmol) was dissolved in 1,2-
dibromoethane (114.2 g; 52 mL; 608 mmol) and stirred at
90 8C for a week. During that time, any precipitate formed
was recovered by filtration. After each filtration, the
precipitate was rinsed with an additional 5 mL of 1,2-
dibromoethane and the mother liquor stirred at 90 8C until
the next filtration. The reaction was deemed to be complete
when no more precipitate formed. The precipitates were
combined and washed thoroughly with ethyl acetate to give
1 or 7.
2
2
vice versa) was not observed during the NMR experiments
and acts as a further control.
Consequently, the substitution of 1 by thiol-derivatives is
also believed to be ‘directed’ by a similar re-arrangement as
Path A of Scheme 3, that is: an alpha addition forming an
intermediate soluble in organic solvent, followed by a five-
membered ring cyclisation and an oxidative ring opening
leading to the final material 9, which is soluble in aqueous
solution.
4.2.1.1. 2-Bromoethyl-phenanthridinium bromide (1).
Product 1 (10.55 g; 28.8 mmol) was recovered as a beige
1
powder in a 95% yield; mp: 234–235 8C (dec); H NMR
(
D O, 400 MHz): d 9.81 (s, 1H), 8.72 (d, 1H, JZ7.2 Hz),
2
3
. Conclusions
8.63 (d, 1H, JZ7.2 Hz), 8.37 (d, 1H, JZ7.2 Hz), 8.26 (d,
H, JZ7.2 Hz), 8.18 (t, 1H, JZ7.2 Hz), 7.98 (t, 1H, JZ
7.2 Hz), 7.90 (m, 2H), 5.37 (t, 2H, JZ5.8 Hz), 4.05 (t, 2H,
1
We have described the mechanism of a new type of
nucleophilic substitution on 2-bromoethyl-pyridinium-like-
cations. The substitution appears to be ‘directed’ by an
initial alpha addition step on the iminium moiety, followed
by a five-membered ring cyclisation and a re-aromatisation
process via an oxidative C–N bond breakage. This intricate
substitution was shown to be more effective than conven-
tional types of nucleophilic substitution. In further work,
we intend to examine the reaction of secondary amines
with the 3-bromopropyl-pyridinium framework whereby
any substitution taking place is also likely to occur via a
similar re-arrangement requiring the formation of a six-
membered ring intermediate. We will also examine the
reaction pathways outlined here using theoretical methods.
1
3
JZ5.8 Hz); C NMR (D O, 100 MHz): d 155.27 (CH),
2
139.03 (CH), 135.59 (C), 133.18 (CH), 132.78 (C), 132.58
(CH), 130.85 (CH), 130.72 (CH), 126.57 (C), 125.13 (CH),
123.32 (C), 123.00 (CH), 118.91 (CH), 58.87 (CH
), 29.41
); IR (KBr, cm ): 2947 (w), 1620 (m), 763 (s), 717
(m); MS (ES): 288.1 (M–Br) (100), 206.2 (8). Anal. Calcd
for C15 13NBr : C, 49.32; H, 3.59; N, 3.84. Found: C,
2
K1
(CH
2
H
2
49.15; H, 3.48; N, 3.76.
4.2.1.2. 2-Bromoethylquinolinium bromide (7).
Product 4 (9.10 g; 28.7 mmol) was recovered as a beige
1
powder in a 95% yield; mp: 289–290 8C (dec); H NMR
(D O, 400 MHz): d 9.24 (d, 1H, JZ8.4 Hz), 9.12 (d, 1H,
2

A. D. C. Parenty et al. / Tetrahedron 61 (2005) 8410–8418
8415
JZ8.4 Hz), 8.87 (d, 1H, JZ8.4 Hz), 8.33 (d, 1H, JZ
.4 Hz), 8.20 (t, 1H, JZ8.4 Hz), 7.98 (m, 2H), 5.41 (t, 2H,
(FAB): 498.4 (M–2Br) (60), 318.2 (30), 292.1 (50), 249.1
(80), 206.1 (70), 154.0 (100), 136.0 (80), 112.3 (35), 56.9
(30). Anal. Calcd for C H N Br : C, 62.01; H, 5.20; N,
8
1
3
JZ5.8 Hz), 4.05 (t, 2H, JZ5.8 Hz); C NMR (D O,
2
34 34
4
2
1
00 MHz): d 149.80 (C), 149.20 (CH), 138.08 (C), 136.66
8.51. Found: C, 62.30; H, 5.45; N, 8.51.
(
(
CH), 131.43 (CH), 130.68 (CH), 130.55 (CH), 121.59
CH), 118.08 (CH), 58.59 (CH ), 29.28 (CH ); IR (KBr,
2
2
4.2.2.3. Triazacyclododecane triethyl phenanthridi-
nium bromide (5c). Product 5c was obtained as a yellow-
K1
cm ): 3437 (s), 3045 (w), 2981 (m), 2947 (m), 1624 (s),
1
orange powder in a 93% yield; H NMR (CD OD,
1
1
8
599 (m), 1585 (m), 1525 (s), 1489 (w), 1450 (m), 1400 (m),
363 (s), 1242 (s), 1126 (m), 1161 (m), 1144 (m), 1049 (w),
74 (w), 816 (m), 800 (m), 775 (s); MS (FAB): 237 (M–Br)
3
400 MHz): d 9.93 (s, 3H), 8.99 (t, 6H, JZ8.8 Hz), 8.45
(d, 3H, JZ8.0 Hz), 8.42 (d, 3H, JZ6.8 Hz), 8.30 (t, 3H, JZ
7.6 Hz), 8.00 (m, 6H), 7.85 (t, 3H, JZ7.6 Hz), 5.02 (m, 6H),
(
(
98), 236 (100), 209.9 (2), 172 (2), 156 (12), 129.1 (6), 107.2
2), 89.5 (2), 72.7 (1), 59.9 (1). Anal. Calcd for C H NBr :
1
3
1
1
11
2
2.57 (m, 6H), 1.41 (m, 12H), 0.05 (m, 6H); C NMR
(CD OD, 100 MHz): d 156.53 (CH), 140.11 (CH), 136.93
(C), 135.00 (C), 134.23 (CH), 133.95 (CH), 132.28 (CH),
C, 41.67; H, 3.49; N, 4.42. Found: C, 41.75; H, 3.50; N,
4
3
.51.
132.19 (CH), 128.13 (C), 126.71 (CH), 125.14 (C), 124.85
(CH), 121.43 (CH), 57.57 (CH ), 53.34 (CH ), 49.39 (CH ),
4
2
2
.2.2. General procedure for the synthesis of
-aminoethyl-pyridinium derivative (5a–d; 8) and
-thioethyl-pyridinium derivative (9). 2-Bromoethyl-pyr-
2
2
2
K1
23.25 (CH ); IR (KBr, cm ): 3430.74 (s), 2923 (w), 2360
2
(w), 1626 (s), 1456 (m), 1261 (w), 1026 (w), 758 (w); MS
(FAB): 498.4 (M–2Br) (60), 318.2 (30), 292.1 (50), 249.1
(80), 206.1 (70), 154.0 (100), 136.0 (80), 112.3 (35), 56.9
(30). Anal. Calcd for C H N Br : C, 62.98; H, 5.58; Br,
idinium 1 or 7 (1.9 mmol) was dissolved in 20 mL DMF.
Secondary amine or thiol derivative (2.1 mmol) and TEA
(
0.576 mg; 795 mL; 5.7 mmol) were added successively to
5
4
57
6
3
the stirred solution. After stirring for 48 h at rt under
nitrogen, the final product and TEA hydrobromide salt were
precipitated from the solution by adding diethyl ether
23.28; N, 8.16. Found: C, 63.08; H, 4.51; N, 8.10.
4.2.2.4. 5-{2-[(4-Methoxy-phenyl)-methyl-amino]-
ethyl}phenanthridinium bromide (5d). Product 5d was
(
40 mL) and were recovered by filtration. The precipitate
1
obtained as brown powder in a 77% yield; mp: 66–67 8C; H
was washed thoroughly with ethyl acetate and then
triturated twice with 0.5 mL of water to get rid of the
TEA salt. The residue was dried by successive diethyl ether
addition/suction cycles to obtain 5a–d; 8 or 9.
NMR (D O, 400 MHz): d 9.42 (s, 1H), d 8.71 (d, 1H, JZ
2
7.6 Hz), d 8.58 (d, 1H, JZ7.6 Hz), d 8.34 (d, 1H, JZ
7.6 Hz), d 8.13 (t, 1H, JZ7.6 Hz), d 8.03 (m, 2H), d 7.95 (t,
1H, JZ7.6 Hz), d 7.82 (t, 1H, JZ7.6 Hz), d 6.10 (d, 2H, JZ
4.2.2.1. 5-(2-Piperidin-1-yl-ethyl)-phenanthridinium
bromide (5a). Product 5a was obtained as a pale yellow
8.8 Hz), d 5.89 (d, 2H, JZ8.8 Hz), d 5.17 (m, 2H), d 3.97
(m, 2H), d 3.20 (s, 3H), d 2.87 (s, 3H); C NMR (D O,
2
100 MHz):d155.16(CH),d151.02(C),d142.00(C),d138.21
(CH), d 134.85 (C), d 132.31 (CH), d 132.11 (C), d 130.45
(CH), d 129.94 (CH), d 126.11 (C), d 124.61 (CH), d 122.92
(C), d 122.29 (CH), d 119.07 (CH), d 115.46 (CH), d 113.90
1
3
1
powder in a 71% yield; mp: 167–168 8C (dec); H NMR
(
D O, 400 MHz): d 9.80 (s, 1H), 8.90 (d, 1H, JZ7.2 Hz),
2
8
2
7
.83 (d, 1H, JZ8.4 Hz), 8.41 (d, 1H, JZ8 Hz), 8.28 (m,
H), 7.99 (m, 3H), 5.15 (t, 2H, JZ7.2 Hz), 3.04 (t, 2H, JZ
.2 Hz), 2.56 (m, 4H), 1.50 (m, 4H), 1.41 (m, 2H); C NMR
1
3
(CH), d 55.59 (CH
(CH ); MS (FAB): 342.2 (M–Br) (100), 327.2 (10), 218.1 (5),
206.1 (13), 164.1 (75), 133.1 (30), 120.2 (4). Anal. Calcd for
23BrN O: C, 65.25; H, 5.48; N, 6.62. Found: C, 65.28;
), d 55.09 (CH ), d 50.00 (CH ), d 37.59
2 3 2
(
D O, 100 MHz): d 154.63 (CH), 147.71 (C), 138.61 (CH),
2
3
1
(
1
36.45 (C), 135.35 (C), 132.72 (CH), 132.47 (CH), 130.67
CH), 126.56 (CH), 125.11 (CH), 123.83 (C), 123.04 (CH),
19.06 (CH), 56.40 (CH ), 54.87 (CH ), 54.18 (CH ), 25.11
C
H
23
2
H, 5.60; N, 6.58.
2
2
2
K1
(
CH ), 23.42 (CH ); IR (KBr, cm ): 3448 (s), 2923 (m),
2 2
2
1
852 (w), 2794 (w), 2360 (w), 1628 (s), 1535 (w), 1506 (w),
454 (m), 1352 (w), 1257 (w), 1161 (w), 1122 (w), 1036
4.2.2.5. 1-(2-Piperidin-1-yl-ethyl)-quinolinium bro-
mide (8). Product 8 was obtained as a brown powder in
1
an 80% yield; mp: 220–221 8C (dec); H NMR (D O,
(
w), 769 (s); MS (FAB): 291.2 (M–Br) (100); 273.1 (4),
2
2
9
06.1 (7), 193 (7), 154 (92), 137 (60), 136 (60), 112.3 (45),
8.4 (16), 89.5 (11), 77.6 (5), 56.9 (2), 52 (2). Anal. Calcd
400 MHz): d 9.20 (d, 1H, JZ7.6 Hz), 9.06 (d, 1H, JZ
8.4 Hz), 8.32 (t, 2H, JZ7.6 Hz), 8.20 (t, 1H, JZ8.2 Hz),
7.97 (d, 1H, JZ8.2 Hz), 7.94 (d, 1H, JZ8.4 Hz), 5.14 (t,
2H, JZ7.4 Hz), 3.03 (t, 2H, JZ7.4 Hz), 2.58 (m, 4H), 1.53
for C H N Br: C, 64.69; H, 6.24; N, 7.54. Found: C,
2
6
0 23 2
4.17; H, 6.10; N, 7.58.
1
3
(
m, 4H), 1.41 (m, 2H); C NMR (D O, 100 MHz): d 149.40
2
4.2.2.2. Piperazine diethyl-phenanthridinium bro-
mide (5b). Product 5b was obtained as a yellow powder
(CH), 148.57 (CH), 138.42 (C), 136.55 (CH), 131.37 (CH),
130.69 (C), 130.45 (CH), 122.03 (CH), 118.17 (CH), 56.57
1
in a 73% yield; mp: 260–261 8C (dec); H NMR (D O,
(CH
(CH
), 54.57 (CH
); IR (KBr, cm ): 3448 (s), 2923 (m), 2852 (w), 2794
), 54.14 (CH
), 25.04 (CH
), 23.39
2
2
2
2
2
2
K1
4
00 MHz): d 9.80 (s, 2H), d 8.95 (d, 2H, JZ8.0 Hz), d 8.88
d, 2H, JZ8.0 Hz), d 8.42 (d, 2H, JZ8.0 Hz), d 8.33 (d, 2H,
JZ8.0 Hz), d 8.29 (t, 2H, JZ8.0 Hz), d 8.01 (m, 6H), d 5.14
t, 4H, JZ6.8 Hz), d 3.05 (t, 4H, JZ6.8 Hz), d 2.57 (s, 8H);
(
(w), 2360 (w), 1628 (s), 1535 (w), 1506 (w), 1454 (m), 1352
(w), 1257 (w), 1161 (w), 1122 (w), 1036 (w), 769 (s); MS
(FAB): 241.1 (M–Br) (100), 156 (73), 140 (15), 112.3 (33).
Anal. Calcd for C16H N Br: C, 59.82; H, 6.59; N, 8.72.
21 2
Found: C, 59.70; H, 6.62; N, 8.70.
(
1
3
C NMR (D O, 100 MHz): d 155.94 (CH), d 138.46 (CH),
d 134.63 (C), d 133.28 (CH), d 133.07 (C), d 132.41 (CH), d
2
1
30.89 (CH), d 130.54 (CH), d 126.03 (C), d 125.48 (CH), d
23.66 (CH), d 120.19 (CH), d 55.43 (CH ), d 55.08 (CH ),
1
d 52.95 (CH ); IR (KBr, cm ): 3430.74 (s), 2923 (w), 2360
4.2.2.6. 5-[2-(4-Methoxy-benzylsulfanyl)-ethyl]phe-
nanthridinium bromide (9). Product 9 was obtained as a
pale yellow powder in a 76% yield; mp: 182–183 8C (dec);
2
2
K1
2
(
w), 1626 (s), 1456 (m), 1261 (w), 1026 (w), 758 (w); MS

8
1
416
A. D. C. Parenty et al. / Tetrahedron 61 (2005) 8410–8418
H NMR (CD OD, 400 MHz): d 9.91 (s, 1H), d 9.08 (t, 2H,
3
7.94 (m, 2H), 5.40 (t, 1H, JZ4.4 Hz), 5.33 (t, 1H, JZ
4.4 Hz), 5.12 (t, 1H, JZ4.4 Hz), 5.00 (t, 1H, JZ4.4 Hz);
JZ8.0 Hz), d 8.85 (d, 1H, JZ8.0 Hz), d 8.47 (t, 1H, JZ
1
3
8
2
5
3
.0 Hz), d 8.37 (m, 1H), d 8.15 (t, 1H, JZ8.0 Hz), d 8.11 (m,
H), d 6.80 (d, 2H, JZ8.8 Hz), d 6.33 (d, 2H, JZ8.8 Hz), d
.17 (t, 2H, JZ6.0 Hz), d 4.90 (t, 2H, JZ6.0 Hz), d 3.69 (s,
C NMR (D O, 100 MHz): d 155.22 (CH), 138.89 (CH),
2
135.09 (C), 133.08 (C), 132.96 (CH), 132.60 (CH), 130.85
(CH), 130.70 (CH), 126.20 (C), 124.88 (CH), 123.37 (C),
122.79 (CH), 119.18 (CH), 82.08 (CH ), 80.39 (CH ), 58.17
1
3
H), d 3.56 (s, 2H); C NMR (CD OD, 100 MHz): d 161.8
3
2
2
K1
(
(
C), d 156.84 (C), d 140.00 (CH), d 137.21 (C), d 134.41
CH), d 133.77 (CH), d 132.07 (CH), d 131.96 (CH), d
(CH ), 57.98 (CH ); IR (KBr, cm ): 3404 (m), 1534 (m),
2 2
1460 (m), 1447 (m), 1371 (m), 1265 (s), 1165 (s), 1048 (s),
1031 (s), 764 (s), 718 (m); MS (FAB): 226.2 (M–Br) (100),
199.2 (10), 180.1 (25), 101.4 (20); MS (FAB): 226.2 (M–Br)
(100), 199.2 (10), 180.1 (25), 101.4 (20). Anal. Calcd for
C H BrFN: C, 58.84; H, 4.28; N, 4.57. Found: C, 59.00;
1
1
31.31 (C), d 130.96 (CH), d 128.00 (C), d 126.72 (CH), d
25.21 (CH), d 124.74 (CH), d 120.76 (CH), d 114.96 (CH),
d 58.90 (CH ), d 55.92 (CH ), d 36.97 (CH ), d 31.70 (CH );
2
2
3
2
K1
IR (KBr, cm ): 3435 (s), 1626 (s), 1533 (w), 1510 (s),
1
5 13
1
7
450 (m), 1304 (w), 1248 (s), 1174 (w), 1030 (s), 829 (s),
64 (s); MS (FAB): 360.0 (M–Br) (70), 309.0 (20), 290.0
H, 4.23; N, 4.45.
(
1
15), 238.0 (5), 206.0 (10), 179 (7), 155.0 (100), 136.0 (50),
21.1 (50), 108.2 (20), 89.5 (12). Anal. Calcd for C H -
NOSBr: C, 62.72; H, 5.03; N, 3.18. Found: C, 62.72; H,
4.2.4. 2-Hydroxyethyl-phenanthridinium bromide (1OH).
Phenanthridine (2 g; 11.17 mmol) was added to a solution of
2-bromoethanol (3.2 mL; 44.8 mmol). The reaction mixture
2
3 22
5
.01; N, 3.78.
was refluxed for 4 h under N . After cooling, crystallisation
2
was aided by the addition of ether. After 2 h, the crystals
were recovered by filtration and washed with ether to
produce 1OH (3.10 g; 10.13 mmol) as a beige powder in a
4
.2.3. 2-Fluoroethyl-phenanthridinium bromide (1 ). 2-
F
Fluoroethyltosylate was prepared: 2-Fluoroethanol (1 g;
5.6 mmol) dissolved in dry pyridine (15 mL) under
1
1
90% yield; mp: 239–240 8C (dec); H NMR (D
2
O,
nitrogen. The solution was stirred at 0 8C and p-toluene
sulfonyl chloride (6.5 g; 34.1 mmol) was added slowly to
the solution over a period of 30 min, keeping the
temperature below 5 8C. The solution was then stirred at
400 MHz): d 9.71 (s, 1H), 8.81 (d, 1H, JZ8.0 Hz), 8.74
(d, 1H, JZ8.0 Hz), 8.37 (d, 1H, JZ8.0 Hz), 3.32 (d, 1H,
JZ8.0 Hz), 8.22 (t, 1H, JZ8.0 Hz), 7.97 (m, 3H), 5.11 (t,
1
3
2H, JZ5.0 Hz), 4.14 (t, 2H, JZ5.0 Hz); C NMR (D O,
2
0
8C for another 4 h before quenching by slow addition of
100 MHz): d 156.71 (CH), 140.70 (CH), 136.62 (C), 134.88
(CH), 134.68 (CH), 132.81 (CH), 127.93 (C), 126.75 (C),
ice (15 g) then water (20 mL). Ethyl acetate (50 mL) was
added, the organic layer separated and washed with water.
Excess pyridine was removed by washing the organic layer
with a 1 M HCl solution until the aqueous layer became
acidic. The excess tosyl chloride was removed by washing
the organic layer with an aqueous solution of Na CO
125.09 (C), 124.65 (CH), 121.33 (CH), 62.31 (CH
(CH
), 61.19
); IR (KBr, cm ): 3232 (s), 1624 (s), 1581 (w), 1535
2
2
K1
(m), 1446 (s), 1423 (m), 1346 (m), 1261 (s), 1157 (s), 1084
(s), 1033 (s), 899 (s), 867 (m), 756 (s), 717 (s), 609 (s); MS
(ES): 224.1 (M–Br) (100), 222.1 (25), 210.1 (10), 206.1
(40), 194.1 (8), 193.1 (18), 182.1 (10), 181.1 (80), 180.1
(100). Anal. Calcd for C15H14BrNO: C, 59.23; H, 4.64; N,
4.60. Found: C, 59.10; H, 4.73; N, 4.52.
2
3
(pHz10). The organic layer was then washed with brine,
dried over MgSO and concentrated under vacuum to obtain
4
2
-fluoroethyltosylate (3.29 g; 15 mmol) as an oil in a 96%
1
yield; H NMR (CDCl , 400 MHz): d 7.81 (d, 2H, JZ
3
8
4
.0 Hz), 7.38 (d, 2H, JZ8.0 Hz), 4.65 (t, 1H, JZ4.2 Hz),
.53 (t, 1H, JZ4.2 Hz), 4.32 (t, 1H, JZ4.2 Hz), 4.25 (t, 1H,
4.2.5. 5-(2-Fluoroethyl)-6-piperidin-1-yl-5,6-dihydro-
phenanthridine (3
). In an NMR tube, 1
O (0.6 mL). CDCl
3
(10 mg;
F
F
1
3
JZ4.2 Hz), 2.48 (s, 3H); C NMR (CDCl , 100 MHz): d
0.032 mmol) was dissolved in D
2
3
1
45.53 (C), 133.05 (C), 130.32 (CH), 128.37 (CH), 81.78
CH ), 80.06 (CH ), 68.91 (CH ), 68.70 (CH ), 22.05
(0.6 mL) was added followed by 2 equiv of piperidine
(6.30 mL; 5.4 mg; 0.064 mmol) used both as a reactant and
as a base. The NMR tube was energetically shaken for 1 min
(
(
2
2
2
2
CH ); MS (EI): 218.2 (MC1) (50), 185.2 (8), 155.1 (100),
3
1
13
1
39.1 (5), 107.1 (8), 91.1 (100), 65.1 (30), 63.1 (10).
to allow the phase transfer process to occur. A H and
C
NMR spectrum as well as an MS of the bottom organic layer
Phenanthridine (690 mg; 3.84 mmol) was added to a
solution of 2-fluoroethyltosylate (1.68 g; 7.68 mmol) in
DMF (10 mL) and stirred at 100 8C for 48 h. The solution
was then concentrated to a brown oil and precipitated by
addition of acetone (10 mL) followed by diethyl ether
(60 mL). The precipitate was recovered by filtration,
washed with diethyl ether and dried under vacuum to obtain
the tosylate salt of 1 (1.52 g; 3.84 mmol) as an off white
F
1
was taken, characterising 3 ; H NMR (CDCl , 400 MHz): d
F 3
7.79 (d, 1H, JZ8.0 Hz), 7.76 (d, 1H, JZ8.0 Hz), 7.31 (t,
1H, JZ7.2 Hz), 7.22 (t, 1H, JZ7.2 Hz), 7.15 (t, 1H, JZ
8.0 Hz), 7.05 (d, 1H, JZ8.0 Hz), 6.75 (t, 2H, JZ7.2 Hz),
4.97 (s, 1H), 4.75 (m, 1H), 4.60 (m, 1H), 3.90 (m, 1H), 3.80
1
3
(m, 1H), 2.20 (m, 4H), 1.30 (m, 4H), 1.15 (m, 2H);
NMR (CDCl , 100 MHz): d 143.35 (C), 139.72 (C),
C
3
powder in a quantitative yield. Finally, the tosylate salt was
passed through an anionic exchange column (Dowex 1X-
131.65 (C), 129.14 (CH), 127.21 (CH), 126.35 (CH),
123.76 (CH), 123.26 (CH), 122.44 (CH), 121.70 (CH),
8
50) preloaded with a saturated NaBr solution and flushed
120.51 (C), 111.16 (CH), 83.47 (CH), 82.70 (CH
82.19 (CH ), 81.03 (CH ), 80.51 (CH ), 50.06 (CH
48.63 (CH ), 26.29 (CH
(3), 292.38 (M–F) (4), 241.2 (15), 226.3 (100), 195.2 (20),
179.2 (40), 152.2 (15), 84 (100).
2
2
),
),
with distilled water. The compound was eluted with distilled
water. The resulting aqueous solution was washed twice
with ethyl acetate before being concentrated under vacuum
2
2
2
); MS (EIC): 311.37 (MC1)
2
2
to obtain 1 (1.12 g; 3.68 mmol) as a pale yellow powder in
F
1
a 96% yield; mp: 239–240 8C (dec); H NMR (D O,
2
400 MHz): d 9.80 (s, 1H), 8.79 (d, 1H, JZ8.0 Hz), 8.71 (d,
1H, JZ8.0 Hz), 8.36 (d, 1H, JZ8.0 Hz), 8.29 (d, 1H, JZ
8.4 Hz), 8.21 (t, 1H, JZ7.2 Hz), 8.01 (t, 1H, JZ7.2 Hz),
4.2.6. 2-(6-Piperidin-1-yl-6H-phenanthridin-5-yl)-etha-
nol (3OH). In an NMR tube, 1OH (10 mg; 0.033 mmol)
was dissolved in D O (0.6 mL). CDCl (0.6 mL) was added
2
3

A. D. C. Parenty et al. / Tetrahedron 61 (2005) 8410–8418
8417
followed by 2 equiv of piperidine (6.50 mL; 0.066 mmol)
used both as a reactant and as a base. The NMR tube was
energetically shaken for 1 min to allow the phase transfer
molecule 9 was cleanly characterized, along with some
triethylamine.
1
13
process to occur. A H and C NMR spectrum as well as an
MS of the bottom organic layer was taken, characterising
Acknowledgements
1
3OH; H NMR (CDCl , 400 MHz): d 7.85 (d, 1H, JZ
3
7.6 Hz), 7.81 (dd, 1H, JZ7.6, 1.2 Hz), 7.37 (dt, 1H, JZ7.6,
1.2 Hz), 7.26 (dt, 1H, JZ7.6, 1.6 Hz), 7.16 (dt, 1H, JZ7.6,
1.6 Hz), 7.02 (d, 1H, JZ7.6 Hz), 6.77 (dt, 1H, JZ7.6,
1.2 Hz), 6.68 (d, 1H, JZ7.6 Hz), 4.84 (s, 1H), 4.17 (dd, 1H,
This work was supported by EPSRC. We thank Dr. Jesus
M. de la Fuente and Dr. Richard Hartley for fruitful
discussions.
JZ15.6, 2.0 Hz), 4.01 (t, 1H, JZ11.2 Hz), 3.60 (d, 1H, JZ
1
1
1
1.2 Hz), 3.31 (ddd, 1H, JZ15.6, 11.2, 2 Hz), 2.25 (m, 4H),
13
.45 (m, 4H), 1.18 (m, 2H); C NMR (CDCl , 100 MHz): d
43.14 (C), 131.73 (C), 129.20 (CH), 128.23 (CH), 127.69
3
References and notes
(
1
CH), 126.21 (CH), 125.30 (C), 123.14 (CH), 122.13 (CH),
19.63 (C), 117.16 (CH), 111.27 (CH), 80.35 (CH), 59.28
CH ), 57.54 (CH ), 48.10 (CH ), 25.01 (CH ), 23.87
1. Pozharskii, A. F.; Soldatenkov, A. T.; Katritzky, A. R.
Heterocycles in Life and Society, Wiley: Chichester, UK,
1997.
(
(
2
2
2
2
CH ); MS (CIC): 307.41 (M–1) (2), 281.39 (4), 240.29 (1),
2
2
8
26.30 (2), 224.29 (2), 196.25 (1), 180.22 (3), 103.2 (10),
6.2 (100), 52.1 (35).
2. Czarnik, A. W. Acc. Chem. Res. 1996, 29, 112–113.
3. Fagnoni, M. Heterocycles 2003, 60, 1921–1958.
4
5
6
7
8
9
. Hajos, G.; Riedl, Z.; Kollenz, G. Eur. J. Org. Chem. 2001,
3
405–3414.
. Majumdar, K. C.; Basu, P. K.; Mukhopadhyay, P. P.
Tetrahedron 2004, 60, 6239–6278.
4
.2.7. UV spectroscopy determination of the substitution
reaction intermediate(s) 6-(2-bromoethyl)-5-(2-piperi-
din-1-yl-ethyl)-5,6-dihydro-phenanthridine (3) and/or
piperidinium-1,2,3,12b-tetrahydro-imidazo[1,2-f]phen-
anthridine bromide (4). Starting material 1 (37 mg;
. Brana, M. F.; Cacho, M.; Gradillas, A.; Pascual-Teresa, B.;
Ramos, A. Curr. Pharm. Des. 2001, 7, 1745–1780.
. Martinez, R.; Chacon-Garcia, L. Curr. Med. Chem. 2005, 12,
0
.1 mmol) was dissolved in 100 mL water. Ten millilitre
1
27–151.
aliquot was diluted by 10 to obtain a 0.1 mM stock solution.
One millilitre of the stock solution was transferred to a UV
cell and the UV spectrum of the starting material 1
measured between 200 and 500 nm. Two lmax can be
observed at 320 and 361 nm. At 320 nm, the absorption of
the starting material 1 was recorded every 40 s, and after
. Nakanishi, T.; Suzuki, M.; Saimoto, A.; Kabasawa, T. J. Nat.
Prod. 1999, 62, 864–867.
. Nakanishi, T.; Masuda, A.; Suwa, M.; Akiyama, Y.; Hoshino-
Abe, N.; Suzuki, M. Bioorg. Med. Chem. Lett. 2000, 10,
2
321–2323.
1
1
1
1
1
1
1
0. Nakanishi, T.; Suzuki, M.; Mashiba, A.; Ishikawa, K.;
Yokotsuka, T. J. Org. Chem. 1998, 63, 4235–4239.
1. Geen, G. R.; Mann, I. S.; Mullane, M. V.; McKillop, A.
Tetrahedron 1998, 54, 9875–9894.
1
h, piperidine (10 mL; 0.01 mmol) added. UV absorption
measurements were carried on for 12 h, characterizing the
slow transformation of the intermediate(s) into the final
material 3a. After this reaction time, a UV scan was also
undertaken between 200 and 500 nm, characterizing the
final product 3a without intermediate. Subsequently, using
the same reaction conditions, a UV spectrum of the
intermediate was measured after 30 min reaction time
between 200 and 500 nm and compared with the UV
spectrum of both the starting material 1 and the final
material 3a.
2. Parenty, A. D. C.; Smith, L. V.; Pickering, A. L.; Long, D. L.;
Cronin, L. J. Org. Chem. 2004, 69, 5934–5946.
3. Baldwin, J. E.; Tomas, R. C.; Kruse, L. I.; Silberman, L. J.
Org. Chem. 1977, 42, 3846–3852.
4. DFT calculations were carried out using HyperChem with the
6
-31G** basis set and the B2-LPY combination potential.
5. Dostal, J.; Potacek, M.; Nechvatal, M. Collect. Czech. Chem.
Commun. 1993, 58, 395–403.
6. Anne, A.; Moiroux, J. J. Org. Chem. 1990, 55, 4608–4614.
4.2.8. U-tube experiment: formation of 5-[2-(4-methoxy-
benzylsulfanyl)-ethyl] phenanthridinium bromide (9)
17. Krohnke, F.; Vogt, I. Justus Liebigs Ann. Chim. 1956, 600,
211–224.
via a necessary organic soluble intermediate. In a
U-tube, containing a layer of CCl (1 mL) with 10 mg of
18. Lavilla, R.; Gotsens, T.; Guerrero, M.; Masdeu, C.; Santano,
M. C.; Minguillon, C.; Bosch, J. Tetrahedron 1997, 53,
13959–13968.
4
MgSO , was added a layer of H O (1 mL) in one harm and a
4
2
layer of D O (1 mL) in the other harm. In the H O
2
19. Moracci, F. M.; Rienzo, B. D.; Tortorella, S.; Liberatore, F.
Tetrahedron 1980, 36, 785–789.
2
compartment, was added starting material 1 (12.6 mg;
.034 mmol) followed by p-methoxybenzyl mercaptan
4.7 mL; 0.03 mmol) and triethylamine (4.7 mL;
.04 mmol). A white precipitate appears instantaneously
0
(
0
20. Berge, T.; Jenkins, N. S.; Hopkirk, R. B.; Waring, M. J.;
Edwardson, J. M.; Henderson, R. M. Nucl. Acids Res. 2002,
30, 2980–2986.
in the H O compartment. The phase transfer was assisted
2
21. Leon, P.; Garbayjaureguiberry, C.; Lepecq, J. B.; Roques, B.
P. Tetrahedron Lett. 1985, 26, 4929–4932.
by gently stirring the inter-phase with the tip of a Pasteur
pipette before sealing the two extremity of the U-tube with
parafilm. The underneath organic layer was gently stirred
with a small magnet for 48 h, taking care not to
contaminate the D O compartment with H O. The D O
22. Pelaprat, D.; Delbarre, A.; Le Guen, I.; Roques, B. P. J. Med.
Chem. 1980, 23, 1336–1343.
23. Leon, P.; Garbayjaureguiberry, C.; Lambert, B.; Lepecq, J. B.;
Roques, B. P. J. Med. Chem. 1988, 31, 1021–1026.
24. Molina, A.; Vaquero, J. J.; GarciaNavio, J. L.; AlvarezBuilla,
2
2
2
1
layer was analyzed by H NMR spectroscopy and final

8418
A. D. C. Parenty et al. / Tetrahedron 61 (2005) 8410–8418
J.; dePascualTeresa, B.; Gago, F.; Rodrigo, M. M.; Ballesteros,
M. J. Org. Chem. 1996, 61, 5587–5599.
28. Acton, E. M.; Narayanan, V. L.; Risbood, P. A.; Shoemaker,
R. H.; Vistica, D. T.; Boyd, M. R. J. Med. Chem. 1994, 37,
2185–2189.
2
2
2
5. Pelaprat, D.; Oberlin, R.; Le Guen, I.; Roques, B. P.; Le Pecq,
J. B. J. Med. Chem. 1980, 23, 1330–1335.
29. Scala, S.; Akhmed, N.; Rao, U. S.; Paull, K.; Lan, L. B.;
Dickstein, B.; Lee, J. S.; Elgemeie, G. H.; Stein, W. D.; Bates,
S. E. Mol. Pharmacol. 1997, 51, 1024–1033.
6. Werbel, L. M.; Angelo, M.; Fry, D. W.; Worth, D. F. J. Med.
Chem. 1986, 29, 1321–1322.
7. Leon, P.; Garbayjaureguiberry, C.; Barsi, M. C.; Lepecq, J. B.;
Roques, B. P. J. Med. Chem. 1987, 30, 2074–2080.