Synthesis and anti-Pneumocystis carinii activity of conformationally restricted analogues of pentamidine

Article abstract of DOI:10.1016/S0223-5234(99)80102-0

A series of conformationally restricted analogues of pentamidine in which the flexible central bridge has been replaced by trans-cyclopropyl, phenyl, pyridinyl, piperazinyl or homopiperazinyl groups as conformationally restricted linkers have been synthesized. The anti-Pneumocystis carinii activity of these compounds was evaluated in a cell culture model and the DNA binding affinity was determined by thermal denaturation measurements. At 1 μM, compounds 2, 3, 5, 7, 9 and pentamidine were highly effective and caused total inhibition of P. carinii growth in culture. At 0.1 μM, compounds 2, 5, 7 and 10 were more active than pentamidine with N, N'- bis(4amidinophenyl)piperazine 7 being approximately 15-fold more effective than pentamidine. The most active compounds, 7 and 10, showed strong binding affinities for calf thymus DNA and poly(dA-dT); however, a clear correlation between DNA binding affinity and the in vitro anti-P, carinii activity of these compounds was not observed. The results suggest that the nature of the central linker influences the biological actions of these compounds.

Full text of DOI:10.1016/S0223-5234(99)80102-0

Eur. J. Med. Chem. 34 (1999) 531−538
© Elsevier, Paris
531
Laboratory note
Synthesis and anti-Pneumocystis carinii activity of conformationally
restricted analogues of pentamidine
Bin Taoa, Tien L Huanga*, Qian Zhanga, Latasha Jacksona, Sherry F. Queenerb, Isaac O. Donkorc
aCollege of Pharmacy, Xavier University of Louisiana, 7325 Palmetto St., New Orleans LA, 70125, USA
bDepartment of Pharmacology and Toxicology, Indiana University School of Medicine, Indianapolis, IN 46202, USA
cCollege of Pharmacy, The University of Tennessee, Memphis, TN 38163, USA
(Received 25 August 1998; accepted 25 November 1998)
Abstract – A series of conformationally restricted analogues of pentamidine in which the flexible central bridge has been replaced by
trans-cyclopropyl, phenyl, pyridinyl, piperazinyl or homopiperazinyl groups as conformationally restricted linkers have been synthesized. The
anti-Pneumocystis carinii activity of these compounds was evaluated in a cell culture model and the DNA binding affinity was determined
by thermal denaturation measurements. At 1 µM, compounds 2, 3, 5, 7, 9 and pentamidine were highly effective and caused total inhibition
of P. carinii growth in culture. At 0.1 µM, compounds 2, 5, 7 and 10 were more active than pentamidine with N, N≠-bis(4-
amidinophenyl)piperazine 7 being approximately 15-fold more effective than pentamidine. The most active compounds, 7 and 10, showed
strong binding affinities for calf thymus DNA and poly(dA-dT); however, a clear correlation between DNA binding affinity and the in vitro
anti-P. carinii activity of these compounds was not observed. The results suggest that the nature of the central linker influences the biological
actions of these compounds. © Elsevier, Paris
Pneumocystis carinii pneumonia / pentamidine analogues / conformationally restricted / DNA / poly(dA-dT)
1. Introduction
idines [8] as conformationally restricted analogues of
pentamidine. Several of these compounds were found to
be more active and less toxic than pentamidine when
evaluated against PCP in the immunosuppressed rat
model [3, 4, 6, 7].
Pentamidine, 1,5-bis(4-amidinophenoxy)pentane, is an
effective agent widely used for the treatment and prophy-
laxis of Pneumocystis carinii pneumonia (PCP), an op-
portunistic infection commonly seen in patients with
acquired immunodeficiency syndrome (AIDS). Despite
its clinical usefulness against PCP, the drug is plagued
with an extraordinary number of serious adverse reac-
tions [1, 2]. We hypothesize that the multiple pharmaco-
logical actions of the drug might be due to its conforma-
tional flexibility resulting in indiscriminate binding to
both target and non-target macromolecules. Therefore,
we are interested in studying the effect of restricting the
conformational flexibility of pentamidine analogues on
their anti-P. carinii activity. We [3–5] and others [6–8]
have recently reported on the synthesis and promising
anti-P. carinii activities of a series of cis- and trans-
butamidine analogues [3, 4], piperidine-linked aromatic
diimidazolines [5], furans [6, 7], and 2,4-diarylpyrim-
Although the precise mechanism of action of pentami-
dine and its analogues as anti-PCP agents is unknown,
recent evidence suggests that these compounds bind in
the minor groove of DNA and interfere with the normal
functions of DNA-dependent enzymes (e.g. topoisome-
rases) that regulate transcription in the pathogen [9].
Using crystallographic [6, 10, 11], foot-printing [12], and
other biophysical techniques [13], it has been shown that
pentamidine and its analogues interact with DNA in the
minor groove and prefer AT to GC rich DNA sequences.
The interactions involve hydrogen-bonding between the
amidine groups of pentamidine analogues and the bases
and sugars of DNA. In addition, Van der Waals contacts
also occur between the drug and the walls of the narrow
AT-rich minor groove [6, 10, 11]. Despite investigations
into the DNA-binding mode of pentamidine analogues, a
*Correspondence and reprints

532
Figure 1. Reagents used for the synthesis of compounds 1 and 2: a) LiAlH4/ether, reflux; b) SOCl2/pyridine/THF; c) 4-cyanophenol,
Na/EtOH; d) HCl (g)/EtOH/THF; NH3 (g)/MeOH, reflux; e) HCl (g)/EtOH/THF; ethylenediamine/MeOH, reflux.
clear correlation between the DNA binding affinity and
their in vivo anti-PCP actions has not been established.
More recently, inhibition of an endo/exonuclease in P.
carinii has been implicated as a mode of anti-P. carinii
action for several dicationic furans [7].
In this paper we describe the synthesis and in vitro
anti-P. carinii activity of a series of conformationally
restricted analogues of pentamidine (table I). The binding
affinity of these compounds to calf thymus DNA and
poly(dA-dT) was also determined. Several compounds
with strong binding affinity to DNA effectively inhibited
the growth of P. carinii in culture.
dinitrile 17. Similar reactions of acid dichlorides 18 and
19 with 4-cyanoaniline gave the corresponding dicyano
compounds 20 and 21 (figure 3). The dicyano compounds
17, 20 and 21 were converted to the diamidines and
diimidazolines 3 and 4 (figure 2), 5 and 6 (figure 3) in a
manner similar to 1 and 2. In figure 4, an aromatic
nucleophilic displacement reaction between 4-fluro-
benzonitrile 24 and piperazine 22 or homopiperazine 23
afforded the corresponding dinitrile 25 or 26. The dini-
triles 25 and 26 were easily converted to the correspond-
ing diamidines 7 and 9, or the diimidazolines 8 and 10.
3. Biological results and discussion
2. Chemistry
The in vitro anti-P. carinii activity and DNA binding
properties of the synthesized compounds 1–10 are shown
in table II. The compounds were tested in a short-term
culture assay system using human embryonic lung cells
with P. carinii from infected rat lung as described
previously [15, 16]. This culture model has been used
effectively as a rapid screen to predict the anti-P. carinii
activities of experimental compounds [16, 17].
In this study, pentamidine and the untreated control
were used as the positive and negative controls, respec-
tively. The compounds were tested in the culture model at
1.0 and 0.1 µM concentrations in two separate experi-
ments (indicated as exp. 1 and exp. 2 in table II).
Compounds 2, 3, 4, 6, 7, 9 and 10 were evaluated in
experiment 1 and compounds 1, 5, and 8 were evaluated
in experiment 2. Of the nine compounds 1–9 tested at 1.0
µM, five compounds (i.e. 2, 3, 5, 7 and 9) and pentami-
dine were highly effective and totally inhibited the
growth of P. carinii in culture. Compound 1 also showed
The synthesis of compounds 1–10 was accomplished
according to the procedures shown in figures 1–4. For the
synthesis of 1 and 2 (figure 1), reduction of diethyl
trans-1,2-cyclopropanedicarboxylate 11 by LiAlH4 in
ether gave the diol 12 which was heated with SOCl2 to
give the dichloride 13, which was then reacted with the
anion of 4-cyanophenol to give the dinitrile 14. Treatment
of 14 with dry HCl gas in anhydrous EtOH and anhy-
drous THF gave the imidate ester, which was refluxed
either in methanolic ammonia to give the diamidine 1 or
with excess ethylenediamine in anhydrous EtOH to give
the diimidazoline 2 [14]. Similar procedures were used to
convert the other dinitriles to the corresponding dia-
midines or diimidazolines 3–10. In figure 2, 11 was
hydrolysed to the corresponding dicarboxylic acid 15
which was reacted with SOCl2 to give the acid dichloride
16. Treatment of 16 with two moles of 4-cyanoaniline in
the presence of N,N-diisopropylethylamine gave the

533
Table I. Structures and physical properties of pentamidine analogues 1–10.
or
Cpd
R
X
Yield %a
66
M.p. (°C)b 1H NMRc
1
AM
257–259
201–203
> 360
9.23 (s, 4H), 9.02 (s, 4H), 7.83 (d, 4H), 7.12 (d, 4H),
4.14 (m, 4H), 1.88 (m, 4H).
2
3
IM
42
64
10.64 (s, 4H), 8.03 (d, 4H), 7.14 (d, 4H), 4.14 (m,
4H), 3.91 (s, 8H), 1.86 (m, 4H).
AM
11.10 (s, 2H), 9.15 (s, 8H), 7.79 (m, 8H), 2.46 (m,
2H), 1.38 (m, 2H).
4
5
IM
63
63
> 360
> 360
11.16 (s, 2H), 10.55 (bs, 4H), 8.00 (d, 4H), 7.87 (d,
4H), 3.97 (s, 8H), 2.45 (m, 2H), 1.40 (m, 2H).
AM
11.23 (s, 2H), 9.30 (s, 4H), 9.02 (s, 4H), 8.94 (s, 1H),
8.22 (d, 2H), 8.20 (d, 4H), 7.88 (d, 4H), 7.74 (t, 1H).
6
AM
58
318
11.72 (s, 2H), 9.33 (s, 4H), 9.04 (s, 4H), 8.44 (d,
4H), 8.32 (m, 3H), 7.93 (d, 4H).
7
8
9
AM
IM
81
65
56
356
9.02 (s, 4H), 8.69 (s, 4H), 7.79 (d, 4H), 7.15 (d, 4H),
3.58 (s, 8H).
> 360
340
10.17 (s, 4H), 7.87 (d, 4H), 7.09 (d, 4H), 3.91 (s,
8H), 3.61 (s, 8H).
AM
8.91 (s, 4H), 8.61 (s, 4H), 7.73 (d, 4H), 6.92 (d, 4H),
3.76 (s, 4H), 3.53 (m, 4H), 1.96 (m, 2H).
10
IM
71
326
10.14 (s, 4H), 7.85 (d, 4H), 6.94 (d, 4H), 3.89 (s,
8H), 3.78 (s, 4H), 3.55 (m, 4H), 1.93 (m, 2H).
b
c
aYield from the corresponding nitrile. Dihydrochloride except for 2, which is the free base. Dihydrochloride in DMSO-d6.

534
Figure 2. Reagents used for the synthesis of compounds 3 and 4: a) 10% NaOH, reflux; b) SOCl2; c) 4-cyanoaniline/DIEA/THF; d)
HCl (g)/EtOH/THF/CHCl3, NH3 (g)/EtOH, reflux; e) HCl (g)/EtOH/THF; ethylenediamine/EtOH, reflux.
strong inhibitory activity (93% inhibition) at 1.0 µM.
When all ten compounds were tested at 0.1 µM, four
compounds (i.e. 2, 5, 7 and 10) were significantly more
active than pentamidine. Compound 7 with a piperazine
ring as the central linker connecting the benzamidine
groups, was the most active and was 15-fold more
effective than pentamidine. Among the bisbenzamidines,
1, 3, 5, 6, 7 and 9, the rank order of potency against
Figure 3. Reagents used for the synthesis of compounds 5 and 6: a) 4-cyanoaniline/DIEA/THF; b) HCl (g)/EtOH/THF, NH3
(g)/EtOH, reflux.
Figure 4. Reagents used for the synthesis of compounds 7–10: a) K2CO3/DMSO, 120 °C; b) HCl(g)/EtOH/CHCl3/THF;
NH3(g)/EtOH, reflux; c) HCl (g)/EtOH/CHCl3, ethylenediamine/EtOH, reflux.

535
Table II. Anti-P. carinii activity and DNA binding affinity of pentamidine analogues.
Exp., compound
concentration
(µM)
Change in numbers of
P. carinii trophozoites from
day 1 through day 5 in
culture (× 10–5)a
% of control growth
DNA binding (∆Tm, °C)
Calf thymus DNAb
Poly(dA-dT)c
Exp. 1, Control
24.10 ± 4.02
Pentamidine
11.1
20.6
1.0
0.1
–3.98 ± 1.64
11.00 ± 5.23
25.47 ± 3.08
< 0
46
Exp. 2, Control
Pentamidine
1.0
0.1
–2.22 ± 0.70
5.85 ± 2.42
< 0
23
Exp. 2, Compd 1
11.1
10.0
7.1
20.3
20.1
11.1
9.6
1.0
0.1
1.83 ± 0.70
7.61 ± 1.17
7
30
Exp. 1, Compd 2
1.0
0.1
–2.93 ± 1.87
2.15 ± 2.34
< 0
9
Exp. 1, Compd 3
1.0
0.1
–0.12 ± 2.03
25.23 ± 3.43
< 0
105
Exp. 1, Compd 4
6.0
1.0
0.1
12.67 ± 2.85
17.74 ± 2.46
53
74
Exp. 2, Compd 5
8.0
9.9
1.0
0.1
–1.29 ± 1.36
3.43 ± 1.56
< 0
13
Exp. 1, Compd 6
7.1
15.9
23.8
19.4
23.1
24.0
1.0
0.1
9.36 ± 2.22
23.40 ± 3.59
39
97
Exp. 1, Compd 7
17.0
15.0
15.0
15.0
1.0
0.1
–5.97 ± 0.66
0.78 ± 1.72
< 0
3
Exp. 2, Compd 8
1.0
0.1
4.99 ± 2.26
21.20 ± 5.46
20
83
Exp. 1, Compd 9
1.0
0.1
–3.32 ± 1.60
14.62 ± 3.35
< 0
61
Exp. 1, Compd 10
0.1
2.73 ± 1.79
11
aValues are reported as means ± SEMs. Values are calculated by subtracting the numbers of organisms on day 1 of culture from the numbers
detected on day 5; thus, positive values denote growth, numbers near zero denote little change, and negative numbers denote decreasing
b
c
organisms in the culture. Increase in thermal melting of calf thymus DNA. Increase in thermal melting of Poly(dA-dT).
P. carinii at 0.1 µM was 7 > 5 > 1 > 9 > 6 ≈ 3. Compound
7 with a piperazine linker was 35-fold more potent than 3
which contains the trans-1,2-cyclopropanediamide moi-
ety as the central linker. Among the bisbenzimidazolines,
2, 4, 8 and 10, the rank order of potency against P. carinii
at 0.1 µM was 2 ≈ 10 > 4 ≈ 8. In this series, compound
2 was about 9-fold more potent than 8. Substitution of the
amidino groups in 1, 3, 7 and 9, with the cyclic
imidazolino groups resulted in either, a slight to modest
increase (1.4 to 5.5-fold) in anti-P. carinii activity at 0.1
µM for 2, 4, and 10, or a dramatic 28-fold reduction in
anti-P. carinii activity for 8. These results indicate that the
piperazine linker connecting the benzamidine groups in 7
appears to present the optimal conformation for potent
anti-P. carinii activity in this series of compounds.
Compound 7 is also significantly more active than a
series of piperidine-linked aromatic diimidazolines re-
ported earlier [5].

536
The most active compounds, 7 and 10, were also the
5.1.1. General procedure for the synthesis of diamidines
and diimidazolines 1–10 from the corresponding
dinitriles
strongest DNA binders (∆Tm: 17.0 °C and 15.0 °C for
calf thymus DNA, 23.8 °C and 24.0 °C for poly(dA-dT),
respectively) in this series. All ten compounds were
found to bind to both calf thymus DNA and poly(dA-dT).
Binding to poly(dA-dT) was consistently stronger, in
agreement with the observation that pentamidine ana-
logues prefer AT-rich DNA sequences [7, 10, 11]. Com-
pounds containing the more basic amidino groups (e.g. 1,
3, and 7) generally showed stronger binding affinity for
calf thymus DNA and poly(dA-dT) than their correspond-
ing counterparts bearing the imidazolino groups (e.g. 2, 4,
and 8). The rank orders of binding affinity to poly(dA-dT)
by the bisamidines and bisimidazolines were 7 > 9 > 1 >
6 > 3 > 5, and 10 > 2 > 8 > 4, respectively. Although the
most active compounds, 7 and 10, have strong binding
affinity for DNA, a clear correlation between the in vitro
anti-P. carinii activity and DNA binding affinity for the
compounds in this series was not observed.
To a cooled, stirred solution of a dinitrile compound in
EtOH, anhydrous THF and/or CHCl3, anhydrous HCl gas
was bubbled through for 1 h. The solution was then
allowed to reach room temperature and stirred for 3 d.
The solvent was concentrated to near dryness below
40 °C under reduced pressure and ether was added to
precipitate the imidate salt. The solid, after filtration, was
immediately dissolved in MeOH or EtOH. For the
synthesis of amidines, anhydrous ammonia gas was
bubbled through the solution for 1 h, during which time
the solution was heated and then refluxed for 4 h. The
solvent was then removed under reduced pressure and
ether was added to precipitate the crude product after
cooling. Recrystallization from the appropriate solvent(s)
gave the pure amidine as the dihydrochloride salt. For the
synthesis of imidazolines, excess ethylenediamine was
added to the solution of the imidate salt in MeOH or
EtOH. The mixture was refluxed for 2–4 h before the
solvent was evaporated. Ether was added to the residue to
afford the imidazoline free base, which was refluxed in
saturated HCl-EtOH solution for 2–4 h. The solvent was
then removed under reduced pressure and the residue was
purified by recrystallization in the appropriate solvent(s)
to give the desired imidazoline as the dihydrochloride
salt, with the exception of 2 which was obtained as a free
base without refluxing in HCl-EtOH.
4. Conclusions
Ten conformationally restricted analogues of pentami-
dine with different central linkers have been synthesized
and evaluated for anti-P. carinii activity. The binding
affinity of these compounds to calf thymus DNA and
poly(dA-dT) was also measured. Of the ten compounds,
the diamidine 7 with a piperazine linker and diimidazo-
line 10 with a homopiperazine linker were found to be the
most potent (15- and 4-fold more active than pentamidine
at 0.1 µM, respectively) and had the strongest affinity for
DNA. The results indicate that the central linker influ-
ences the biological actions of these compounds.
5.1.2.1.
Trans-1,2-bis(4-cyanophenoxymethylene)
cyclopropane 14
Anhydrous ether (60 mL) was added to LiAlH4 powder
(1.20 g, 30.0 mmol) and the mixture was stirred under a
N2 atmosphere. After 10 min, a solution of diethyl
5. Experimental protocols
trans-1,2-cyclopropanedicarboxylate
11
(4.66 g,
25.0 mmol) in ether (20 mL) was added at such a rate as
to keep the ether refluxing gently. The suspension was
stirred for an additional 30 min, during which time an
extra portion of ether (60 mL) was added to dilute the
suspension. Water (1 mL), 15% NaOH (1 mL) and water
(3 mL) were added dropwise in sequence to the stirred
reaction system under cooling. The resulting mixture was
then filtered and the filtrate was dried over Na2SO4.
Evaporation of the solvent gave 12 (1.40 g, 55%) as a
white liquid. It was used without further purification. H
NMR (CDCl3): δ 4.94 (s, 2H), 3.46–3.37 (m, 4H),
0.99–0.92 (m, 2H), 0.45–0.40 (m, 2H).
To a stirred solution of diol 12 (1.22 g, 12.0 mmol) and
pyridine (0.25 mL) in anhydrous THF (40 mL), cooled in
an ice-bath, was slowly added a solution of SOCl2
(4.5 mL, 20.6 mmol) in anhydrous THF (20 mL). The
5.1. Chemistry
Melting points were determined on a Haake-Buchler
melting point apparatus and are uncorrected. The 1H
NMR spectra were recorded on a Varian Gemini-300 or a
GE Omega-500 spectrometer and d6-dimethylsulfoxide
was used as the solvent unless otherwise noted. The
chemical shifts (δ) are reported in ppm relative to TMS,
0.00 ppm. Elemental analyses (C, H, N) were performed
by M-H-W Laboratories, Phoenix, AZ and are within ±
0.4% of the theoretical values. All chemicals and solvents
were purchased from Aldrich Chemical Company. THF
was distilled from Na and benzophenone before use and
other chemicals were used as received. Anhydrous etha-
nol and methanol were used throughout this study unless
stated otherwise.
1

537
mixture was stirred overnight and then brought to reflux
for 4 h. After cooling, cold water was added and the
mixture was extracted with ether (2 × 50 mL). The
extracts were combined and washed with 10% NaHCO3
until neutral pH and then with water (2 × 20 mL) and
dried over MgSO4. Distillation gave 13 (1.23 g, 67%) as
a pale yellow oil. It was used without further purification.
1H NMR (CDCl3): δ 4.04 (dd, 4H, J = 14.5 and 7.0 Hz),
1.98 (t, 2H, J = 7.0 Hz), 1.30 (t, 2H, J = 7.0 Hz).
DIEA (2 eq) in THF was added and the mixture stirred
overnight. The solvent was removed under reduced pres-
sure. Recrystallization of the residue in the appropriate
solvent(s) gave the corresponding dinitrile.
5.1.3.1.
propanediamide 17
solution
N,N′-Bis(4-cyanophenyl)trans-1,2-cyclo-
A
of
diethyl
trans-1,2-cyclopro-
panedicarboxylate 11 (10.0 g, 11.4 mmol) in 100 mL of 2
M NaOH was refluxed for 1 h and then stirred overnight
at room temperature. The mixture was acidified with 2 M
HCl to pH 3 and extracted with EtOAc (2 × 60 mL) after
saturation with NaCl. The combined organic layers were
washed with brine and dried over Na2SO4. Evaporation
of the solvent afforded 15 (5.15 g, 81%) as a white solid.
1H NMR: δ 12.60 (s, 2H), 1.86–1.83 (m, 2H), 1.24–1.20
(m, 2H).
Under a N2 atmosphere, a mixture of 15 (3.54 g,
27.2 mmol) and SOCl2 (10 mL) was heated at 50–60 °C
for 30 min. Removal of the excess of SOCl2 gave 3.2 g
(80%) of 16 which was used without further purification.
This compound 16 was immediately used in the next
reaction because of its instability.
A solution of 4-cyanophenol (1.58 g, 13.3 mmol) in
EtOH (10 mL) was slowly added to a solution of sodium
(0.33 g, 14.4 mmol) and EtOH (10 mL). To this mixture
was slowly added a solution of 13 (0.92 g, 6.62 mmol) in
EtOH (5 mL). The mixture was then heated and refluxed
for 5 d and filtered after cooling. The solid was washed
with copious amounts of water and dried at 40 °C for 4 h
1
to give 14 (1.52 g, 75%) as a white solid. H NMR
(CDCl3): δ 7.58–7.55 (m, 4H), 6.92–6.90 (m, 4H),
4.07–4.05 (m, 4H), 2.00–1.98 (m, 4H).
5.1.2.2.
Trans-1,2-bis(4-amidinophenoxymethylene)
cyclopropane 1
Following the general procedure (5.1.1.), a reaction of
14 (0.76 g, 2.50 mmol) in EtOH (10 mL) and anhydrous
THF (60 mL) with HCl (g) gave the imidate salt, which
was treated with anhydrous ammonia in MeOH (60 mL)
to give 1 (0.68 g, 66%) as a white solid after re-
crystallization (MeOH/EtOAc: 1/5). Anal. C19H22N4O2
2HCl 2H2O (C, H, N).
Following the general procedure (5.1.3.), 17 was pre-
pared from 16 in 62% yield after recrystallization
(DMSO/H2O: 1/1). 1H NMR: δ 10.80 (s, 2H), 7.75–7.71
(m, 8H), 2.31 (t, 2H, J = 7.0 Hz), 1.34 (t, 2H, J = 7.0 Hz).
5.1.3.2. N,N′-Bis(4-cyanophenyl)1,3-benzodiamide 20
Yield 77%, after recrystallization (DMSO/MeOH:
1/1). 1H NMR: δ 10.80 (s, 2H), 8.51(s, 1H), 8.16 (dd, 2H,
J = 10.0 and 8.0 Hz), 7.98 (d, 4H, J = 8.5 Hz), 7.82 (d,
4H, J = 8.5 Hz), 7.72 (t, 1H, J = 8.0 Hz).
5.1.2.3.
lene)cyclopropane 2
Trans-1,2-bis(4-imidazolinophenoxymethy-
Following the general procedure (5.1.1.), a reaction of
14 (0.186 g, 0.61 mmol) in EtOH (5 mL) and anhydrous
THF (30 mL) with HCl (g) gave the imidate salt, which
was treated with ethylenediamine (1 mL) in MeOH
(30 mL) to give 2 (0.10 g, 42%) as a white solid after
recrystallization from methanol, m.p. 201–203 °C. For
5.1.3.3. N,N′-Bis(4-cyanophenyl)2,6-pyridinediamide 21
Yield 72%, after recrystallization (DMSO/MeOH:
1/1). 1H NMR: δ 11.30 (s, 2H), 8.41 (d, 2H, J = 7.5 Hz),
8.32 (t, 1H, J = 6.0 Hz), 8.16 (d, 4H, J = 8.5 Hz), 7.91 (d,
4H, J = 8.0 Hz).
1
the determination of H NMR, the free base was con-
verted to its dihydrochloride form by refluxing in HCl-
EtOH. The elemental analysis was determined for the free
base form. Anal. C23H26N4O2 (C, H, N).
Compounds 3–10 were prepared according to the
above procedures, and the physical data of compounds
1–10 are listed in table I.
5.1.3.4. N,N′-Bis(4-cyanophenyl)piperazine 25
Under a N2 atmosphere, a mixture of piperazine 22
(3.4 g, 40.0 mmol), 4-fluorobenzonitrile 24 (9.7 g,
80.0 mmol) and anhydrous K2CO3 (16.5 g, 120.0 mmol)
in anhydrous DMSO (100 mL) was heated at 120 °C for
6 h. After it was cooled to room temperature, an excess of
water was added to the reaction mixture. The precipitate
was filtered, washed with water and dried to give 25
(9.6 g, 83%) as a pale yellow solid. 1H NMR (CDCl3): δ
7.53 (d, 4H, J = 8.0 Hz), 6.87 (d, 4H, J = 9.5 Hz), 3.52
(s, 8H).
5.1.3. General procedure for the synthesis of dinitriles
17, 20 and 21
Following a known procedure [18], a solution of the
appropriate acid dichloride in THF was added to a stirred
solution of 4-aminobenzonitrile (2 eq) and N,
N-diisopropylethylamine (DIEA, 2 eq) in THF at 0 °C
under a N2 atmosphere. After 30 min, another portion of

538
5.1.3.5. N,N′-Bis(4-cyanophenyl)homopiperazine 26
and 36.1 °C, respectively. Each ∆Tm value reported in
table II represents the mean of at least two experimental
determinations.
Following the procedure described for the synthesis of
1
25, compound 26 was obtained in 80% yield. H NMR
(CDCl3): δ 7.48 (d, 4H, J = 7.0 Hz), 6.72 (d, 4H, J = 7.0
Hz), 3.71 (s, 4H), 3.49 (t, 4H, J = 7.0 Hz), 2.13-2.10 (m,
2H).
Acknowledgements
This research was supported by grant R24-DA07970
from the National Institute on Drug Abuse (T.L.H) and
grant 1 R15 AI/OD39683 (I.O.D). The compounds were
tested in culture against P. carinii in the laboratories of
Professors Marilyn Bartlett and James Smith at Indiana
University School of Medicine.
5.2. Pharmacology
5.2.1. Screening of compounds for anti-P. carinii activity
in culture
The method used to evaluate the anti-P. carinii activity
of the prepared compounds has been described previ-
ously [15, 16]. All the compounds were dissolved in
DMSO for testing. The data for day 7 are not included for
analysis because the controls began to lose viability at
that time point.
References
[1] Wharton J.M., Coleman D.L., Wofsy C.B., Ann. Intern. Med. 105
(1986) 37–44.
[2] Amsden G.W., Kowalsky S.F., Ann. Pharmacother. 26 (1992)
218–226.
5.2.2. Thermal denaturation studies
Pentamidine, EDTA, Tris-HCl, calf thymus DNA and
poly(dA-dT) used in this study were purchased from
Sigma Chemical Company. The method used for the
determination of the binding affinity (∆Tm) of the com-
pounds 1–10 to calf thymus DNA and the nucleic acid
homopolymer poly(dA-dT) has been described previ-
ously [19, 20]. The binding affinity was measured by
determining the change in midpoint (Tm) of the thermal
denaturation curves of calf thymus DNA as well as
poly(dA-dT) at a 1:5 compound to base pair ratio. The
thermal denaturation temperatures of calf thymus DNA or
poly(dA-dT) were determined in 5.0 mM, pH 7.55, Tris-
HCl buffer containing 50 µM of pure EDTA, 5% DMSO,
and the appropriate concentration of the test compounds
dissolved in DMSO. Calf thymus DNA and poly(dA-dT)
were used at an initial absorbance of about 0.3 at 260 nm.
The melting curves were recorded on a Beckman DU
Series 640 spectrophotometer, with a jacketed six cuvette
Tm cell holder. The cell holder was heated by the Peltier
Temperature Controller that was programmed at a con-
stant rate of 0.5 °C/min. The cell temperature was simul-
taneously recorded by a temperature probe embedded
into the cell block. Data were collected by using the Tm
analysis accessory software installed on the spectropho-
tometer and the program recorded absorbance, tempera-
ture, and cuvette number. Each sample was read for
0.10 s, and 5.0 mM, pH 7.55, Tris-HCl buffer containing
50 µM of pure EDTA and 5% DMSO was used as the
blank. At the end of each experiment, Tm results were
calculated from either the first derivative or the 2-points
fit calculation. Under these conditions, the typical Tm
values for calf thymus DNA and poly(dA-dT) were 60.1
[3] Donkor I.O., Jones S.K., Tidwell R.R., Bioorg. Med. Chem. Lett. 3
(1993) 1137–1140.
[4] Donker I.O., Tidwell R.R., Jones S.K., J. Med. Chem. 37 (1994)
4554–4557.
[5] Huang T.L., Zhang Q., White A.T., Queener S.F., Bartlett M.S.,
Smith J.W., Donkor I.O., Bioorg. Med. Chem. Lett. 6 (1996)
2087–2090.
[6] Boykin D.W., Kumar A., Spychala J., Zhou M., Lombardy R.J.,
Wilson W.D. et al., J. Med. Chem. 38 (1995) 912–916.
[7] Boykin D.W., Kumar A., Xiao G., Wilson W.D., Bender B.C.,
McCardy D.R., Hall J.E., Tidwell R.R., J. Med. Chem. 41 (1998)
124–129.
[8] Kumar A., Boykin D.W., Wilson W.D., Jones S.K., Bender B.K.,
Dykstra C.C., Hall J.E., Tidwell R.R., Eur. J. Med. Chem. 31 (1996)
767–773.
[9] Dykstra C.C., McClerson D.R., Elwell L.P., Tidwell R.R.,
Antimicrob. Agents Chemother. 38 (1994) 1890–1898.
[10] Nunn C.M., Neidle S., J. Med. Chem. 38 (1995) 2317–2325.
[11] Trent J.O., Clark G.R., Humar A., Wilson W.D., Boykin D.W., Hall
J.E. et al., J. Med. Chem. 39 (1996) 4554–4562.
[12] Bailly C., Perrine D., Lancelot J.C., Saturnino C., Robba M., Waring
M.J., Biochemistry 323 (1997) 23–31.
[13] Lombardy R.L., Tanious F.A., Ramachandran K., Tidwell R.R.,
Wilson W.D., J. Med. Chem. 39 (1996) 1452–1462.
[14] Tidwell R.R., Jones S.K., Geratz J.D., Ohmeng K.A., Cory M., Hall
J.E., J. Med. Chem. 33 (1990) 1252–1257.
[15] Queener S.F., Bartlett M.S., Nasr M., Smith J.W., Antimicrob.
Agents Chemother. 37 (1993) 2166–2172.
[16] Bartlett M.S., Shaw M.M., Navaran P., Smith J.W., Queener S.F.,
Antimicrob. Agents Chemother. 39 (1995) 2436–2441.
[17] Queener S.F., J. Med. Chem. 38 (1995) 4739–4759.
[18] Wang W., Lown J.W., J. Med. Chem. 35 (1992) 2890–2897.
[19] Cory M., McKee D.D., Kagan J., Henry D.W., Miller J.A., J. Am.
Chem. Soc. 107 (1985) 2528–2536.
[20] Fairley T.A., Tidwell R.R., Donkor I.O., Naiman N.A., Ohemeng
K.A., Lombardy R.J., Bentley J.A., Cory M., J. Med. Chem. 36
(1993) 1746–1753.