Design, synthesis, and evaluation of new chemosensitizers in multi-drug-resistant Plasmodium falciparum

Article abstract of DOI:10.1021/jm010549o

A series of new chemosensitizers (modulators) against chloroquine-resistant Plasmodium falciparum were designed and synthesized in an attempt to fabricate modulators with enhancing drug-resistant reversing efficacy and minimal side effects. Four aromatic amine ring systems phenothiazine, iminodibenzyl, iminostilbene, and diphenylamine-were examined. Various tertiary amino groups including either noncyclic or cyclic aliphatic amines were introduced to explore the steric tolerance at the end of the side chain. The new compounds showed better drug-resistant reversing activity in chloroquine-resistant than in mefloquine-resistant cell lines and were generally more effective against chloroquine-resistant P. falciparum isolates from Southeast Asian (W2 and TM91C235) than those from South America (PC49 and RCS). Structure-activity relationship studies revealed that elongation of the alkyl side chain of the molecule retained the chemosensitizing activity, and analogues with four-carbon side chains showed superior activity. Furthermore, new modulators with phenothiazine ring exhibited the best chemosensitizing activity among the four different ring systems examined. Terminal amino function has limited steric tolerance as evidenced by the dramatic lose of the modulating activity, when the size of substituent at the amino group increases. The best new modulator synthesized in this study possesses all three optimized structural features, which consist of a phenothiazine ring and a pyrrolidinyl group joined by a four-carbon alkyl bridge. The fractional inhibitory concentration (FIC) index of the best compound is 0.21, which is superior to that of verapamil (0.51), one of the best-known multi-drug-resistant reversing agents. Some of the analogues displayed moderate intrinsic in vitro antimalarial activity against a W-2 clone of P. falciparum.

Full text of DOI:10.1021/jm010549o

J . Med. Chem. 2002, 45, 2741-2748
2741
Design , Syn th esis, a n d Eva lu a tion of New Ch em osen sitizer s in
Mu lti-Dr u g-Resista n t P la sm od iu m fa lcipa r u m
J ian Guan, Dennis E. Kyle, Lucia Gerena, Quan Zhang, Wilbur K. Milhous, and Ai J . Lin*
Division of Experimental Therapeutics, Walter Reed Army Institute of Research, 503 Robert Grant Avenue,
Silver Spring, Maryland 20910
Received December 4, 2001
A series of new chemosensitizers (modulators) against chloroquine-resistant Plasmodium
falciparum were designed and synthesized in an attempt to fabricate modulators with enhancing
drug-resistant reversing efficacy and minimal side effects. Four aromatic amine ring systemss
phenothiazine, iminodibenzyl, iminostilbene, and diphenylamineswere examined. Various
tertiary amino groups including either noncyclic or cyclic aliphatic amines were introduced to
explore the steric tolerance at the end of the side chain. The new compounds showed better
drug-resistant reversing activity in chloroquine-resistant than in mefloquine-resistant cell lines
and were generally more effective against chloroquine-resistant P. falciparum isolates from
Southeast Asian (W2 and TM91C235) than those from South America (PC49 and RCS).
Structure-activity relationship studies revealed that elongation of the alkyl side chain of the
molecule retained the chemosensitizing activity, and analogues with four-carbon side chains
showed superior activity. Furthermore, new modulators with phenothiazine ring exhibited the
best chemosensitizing activity among the four different ring systems examined. Terminal amino
function has limited steric tolerance as evidenced by the dramatic lose of the modulating activity,
when the size of substituent at the amino group increases. The best new modulator synthesized
in this study possesses all three optimized structural features, which consist of a phenothiazine
ring and a pyrrolidinyl group joined by a four-carbon alkyl bridge. The fractional inhibitory
concentration (FIC) index of the best compound is 0.21, which is superior to that of verapamil
(0.51), one of the best-known multi-drug-resistant reversing agents. Some of the analogues
displayed moderate intrinsic in vitro antimalarial activity against a W-2 clone of P. falciparum.
In tr od u ction
antihistaminic, or cardiovascular effects. The effective
dose of these compounds as chemosensitizers is gener-
ally close to or higher than the therapeutic dose for other
clinical applications. Therefore, the principal goal of this
study is to fabricate novel chemosensitizing agents with
improved anti-MDR efficacy and reduced side effects,
and eventually to restore the clinical efficacy of the first-
line antimalarial drugs.
The increasing prevalence of multiple-drug-resistant
(MDR) strains of Plasmodium falciparum in most
malaria-endemic areas has significantly reduced the
efficacy of current antimalarial drugs for treating or
preventing malaria. For instance, resistance to the
inexpensive antimalarial mainstays, such as chloro-
quine, is worldwide. Similarly, resistance to mefloquine,
which was proposed as the drug of choice for chloro-
quine-resistant malaria, has been reported from Africa
and Southeast Asia.1,2 Although drug resistance is a
common problem in the treatment of most microbial
infections, malaria, and many neoplasms, the impact
is more acute for malaria chemotherapy because of the
limited number of clinically useful antimalarial drugs.
Recently, considerable public attention has been di-
rected to address this increasingly serious problem.
A wide variety of drugs representing different
drug classes and diverse chemical structures have
been shown to reverse chloroquine resistance in P.
falciparum.3-5 These include calcium channel blockers
(verapamil) and calmodulin antagonists (trifluoperazine
and phenothiazines), which could be coadministered
with chloroquine to effectively potentiate its efficacy
against chloroquine-resistant cell lines.6,7 However, the
clinical value of these modulators as MDR reversing
agents was impaired by their profound antipsychotic,
Extensive structure-activity relationship studies of
neoplastic MDR modulators with diverse chemical
structures have established two essential features of the
molecule for drug resistance reversal activity: a hydro-
phobic tricyclic aromatic ring and an alkyl side chain
with two amino groups separated by two or three
carbons. Tertiary substituted terminal amines were
more potent than secondary or primary amines. How-
ever, good antipsychotic and antihistaminic agents also
shared similar structural features with anti-MDR agents,
except for the length of side chain.8 Phenothiazines with
ring nitrogen and side-chain nitrogen separated by two
carbons showed the best antihistaminic activity, and
those separated by three carbons exhibited the strongest
antipsychotic activity. Thus, the length of side chain
plays a crucial role in determining the pharmacological
properties of the molecules. Because of the availability,
the reported phenothiazines or related anti-MDR agents
were compounds with side-chain length limited to 2-3
carbons.
While the optimal side-chain length of phenothiazines
and related compounds for antihistaminic and antipsy-
* Corresponding author: Tel 301-319-9084; fax 301-319-9449; e-mail
ai.lin@na.amedd.army.mil.
10.1021/jm010549o This article not subject to U.S. Copyright. Published 2002 by the American Chemical Society
Published on Web 05/18/2002

2742 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 13
Guan et al.
Sch em e 1
clic ring nitrogen failed to react with alkyl halides due
to poor basicity of the heterocyclic ring nitrogen. How-
ever, N-alkylation of the ring nitrogen can be ac-
complished with preformed heterocyclic amine salt
instead of free amine. The general procedure for the
coupling of the heterocyclic rings and the alkyl side
chains involved salt formation of the heterocyclic ring
nitrogen with NaNH2, followed by condensation of the
sodium salt 3 with requisite alkyl halides (2a -c) in
anhydrous xylene under reflux to afford the conjugates
4a -f in 20-80% yields. The THP (tetrahydropyran)
protecting group was readily removed under the cataly-
sis of pyridinium p-toluenesulfonate (PPTS) in MeOH/
THF (1:1) solution to give the desired alcohols 5a -f. The
hydroxyl groups of 5a -f were converted to the corre-
sponding chlorides 6a ,b by thionyl chloride. In some
instances, thionyl chloride failed to provide satisfactory
yields of the desired alkyl chlorides. In such cases, an
alternative chlorinating agent, hexachloroacetone (HCA)/
triphenylphosphine (Ph3P), was employed to prepare
halides 6c-f in yields ranging from 60% to 90%.9
Treatment of 6a -f with appropriate amines in the
presence of sodium carbonate gave the final products
7a -x in 25-93% yields. The identities of the final
products and the intermediates were confirmed by NMR
and mass spectra and elemental analyses.
Resu lts a n d Discu ssion
The in vitro anti-MDR effects of the new chemosen-
sitizers were evaluated in chloroquine- and mefloquine-
resistant TM91C235 cell lines, and the results are
shown in Table 1. Each drug was tested at three
different concentrations: 5000, 500, and 50 ng/mL in
the presence of 10 ng/mL chloroquine or 5 ng/mL meflo-
quine. Chloroquine and mefloquine at above dose level
alone showed no cell growth inhibition to TM91C235.
Most of the new compounds exhibited moderate to good
anti-MDR activity. For example, compounds 7e, 7i, 7j,
and 7m at a concentration of 50 ng/mL completely
restored the sensitivity of TM91C235 to chloroquine as
observed by 99% cell suppression. Unlike chloroquine,
coadministration of the test compounds 7e, 7i, 7j, and
7m with mefloquine (5 mg/mL) did not improve the cell
growth inhibitory activity against TM91C235. However,
compounds 7e, 7i, 7j, 7k , and 7m exhibited moderate
antimalarial activity (30%-62% cell growth suppres-
sion) in the absence of chloroquine or mefloquine,
indicating that these modulators possess intrinsic an-
timalarial activity at <50 ng/mL concentration when
used alone. Furthermore, the drug-resistant reversing
efficacy of the new compounds decreased as the size of
the substituents on the terminal amino group increased
from dimethyl and diethyl (7e, 7i, and 7j) to methyl-
benzyl or dibenzyl (7c, 7d , 7k , and 7l).
chotic activities were reported to be limited to 2-3
carbons, the optimal side-chain length for anti-MDR
activity has yet to be discovered. To explore the optimal
length of side chain and the aromatic ring system for
optimal anti-MDR activity and minimal side effects, we
synthesized a series of new chemosensitizers, which
included derivatives of four different heterocyclic aro-
matic ring systemssphenothiazine, iminodibenzyl, diphe-
nylamine, and iminostilbeneswith side-chain length
from four to six carbons. Various tertiary amino groups
including both cyclic and noncyclic aliphatic amines
were introduced to explore steric tolerance at the
terminal of the side chain. The anti-MDR activity of the
target compounds were evaluated in several chloro-
quine- and/or mefloquine-resistant P. falciparum clones
in vitro.
Ch em istr y
The chemical syntheses of the new chemosensitizing
agents are shown in the synthetic Scheme 1. The
hydroxyl groups of the starting materials, 4-chloro-1-
butanol (1a ) and 6-chloro-1-hexanol (1c), were protected
by forming tetrahydropyran acetal (THP, 2), with py-
ridium p-toluenesulfonate (PPTS) as catalyst (Scheme
1). Five-carbon side chain 5-chloro-1-pentanol (1b) was
not commercially available and was prepared from its
corresponding ester, methyl 5-chlorovalerate (1b′), by
LiAlH4 reduction of the ester group to give the starting
alcohol 1b.
To extensively study the magnitude of chemosensi-
tizing potentiation of chloroquine by the new chemosen-
sitizers in chloroquine-resistant cell lines, the combined
effects of chloroquine and modulators were studied by
isobologram analysis in Asian isolate W2 clone. Vera-
pamil was used as a reference drug in this study. The
x-axis in the isobologram graphs (Figures
The key step to the synthesis of the target compounds
was the conjugation of aliphatic amine side chain with
heterocyclic rings. Phenothiazine and related heterocy-
1-3) represents the fractional inhibitory concentra-
tion (FIC) of chloroquine, and the y-axis is the FIC of

New Chemosensitizers in MDR Plasmodium falciparum
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 13 2743
Ta ble 1. In Vitro Reversal Activity of New Modulators in TM91C235 Cellsa
controlb (% sup)
chloroquinec (% sup)
mefloquined (% sup)
5000
500
50
5000
500
50
5000
500
50
compd
X
R1, R2
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
7a
7b
7c
7d
7e
7i
7j
7k
7l
S
S
S
S
S
CH3, CH3
92
100
35
22
96
96
100
60
13
100
27
38
18
10
70
73
95
32
17
98
17
26
13
7
38
59
62
30
11
58
100
100
62
97
99
27
16
99
99
100
20
8
100
30
31
14
12
99
99
99
10
6
92
100
13
9
95
95
100
41
19
100
31
30
0
2
17
0
C2H5, C2H5
CH3, benzyl
benzyl, benzyl
pyrrolidinyl
CH3, CH3
C2H5, C2H5
CH3, benzyl
benzyl, benzyl
pyrrolidinyl
25
3
0
100
100
100
33
13
100
67
70
95
22
22
96
33
43
64
20
14
64
C2H4
C2H4
C2H4
C2H4
C2H4
7m
99
a
b
Resistant to both chloroquine and mefloquine. Test compounds only. c Combination of 10 ng/mL of chloroquine (no effect on cell
d
growth inhibition at this concentration alone) and test compounds. Combination of 5 ng/mL of mefloquine (no effect on cell growth
inhibition at this concentration alone) and test compounds. % suppression of cell growth.
#
Ta ble 2. FICa Indices of New Modulators in Plasmodium
falciparum W2 Cloneb
compd
7a
7b
7e
7f
7g
7h
7i
7j
7m
7n
7o
7p
7q
7r
7s
X
n
R1, R2
CH3, CH3
C2H5, C2H5
pyrrolidinyl
piperidinyl
morpholinyl
4-methylpiperazinyl
CH3, CH3
C2H5, C2H5
pyrrolidinyl
piperidinyl
morpholinyl
4-methylpiperazinyl
C2H5, C2H5
pyrrolidinyl
C2H5, C2H5
pyrrolidinyl
C2H5, C2H5
pyrrolidinyl
C2H5, C2H5
FIC
S
S
S
S
S
S
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
5
5
6
6
0.23
0.23
0.21
0.49
0.39
0.4
C2H4
C2H4
C2H4
C2H4
C2H4
C2H4
C2H2
C2H2
N/A
0.23
0.39
0.32
0.45
0.31
0.52
0.53
0.33
0.48
0.45
0.44
0.48
0.48
0.57
0.51
7t
7u
7v
7w
N/A
C2H4
C2H4
C2H4
C2H4
F igu r e 1. Isobologram of the interaction of chloroquine with
modulators againts W2 clone.
7x
pyrrolidinyl
tion, less than 1 represents synergism, and greater than
1 means antagonism.
verapamil
a
FIC ) fractional inhibitory concentration (1:1 combination of
To study the effects of polycyclic aromatic moiety on
the anti-MDR activity, modulators with four different
aromatic ringssphenothiazine, iminodibenzyl, imino-
stilbene, and the flexible diphenylamineswere pre-
pared. The isobolograms of the interaction of chloro-
quine with modulators against a W-2 clone are pre-
sented in Figures 1-3. Phenothiazine analogues 7b and
7e, which are represented by the most concave curves
in Figure 1, showed the best MDR-reversing activity.
FIC indices of these two compounds are 0.23 and 0.21
respectively (Table 2). Compounds containing saturated
(7j and 7m ) and unsaturated seven-membered central
rings (7q and 7r ) possessed similar activity, with FIC
indices in the range of 0.32-0.53, but were less active
than phenothiazine analogues. Diphenylamine ana-
logues 7s and 7t were not as potent as their tricyclic
b
drug and chloroquine). Resistant to chloroquine.
the modulator. The FIC is the actual IC50 of one drug
in the presence of the second drug but is expressed as
a fraction of its IC50 when used alone. If the isobologram
graph is a straight line, it represents an additive effect
of the two drugs. If the graph forms a concave curve
below the line, it indicates synergy or potentiation of
the combination. If the curve is above the line, it
indicates an antagonism between the two drugs. Thus,
the more concave the curve exhibited, the more effective
the modulator. For easy comparison, the FIC index of
each tested compound is listed in Table 2. This index is
a mathematical representation of an isobologram. FIC
index is a combination of the FIC of both drugs. An FIC
index of 1 represents the additive effect of the combina-

2744 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 13
Guan et al.
F igu r e 2. Isobologram of the interaction of chloroquine with
imipramine derivatives against W2 clone.
F igu r e 3. Isobologram of the interaction of chloroquine with
phenothiazine derivatives against W2 clone.
ring counterparts and were 2-fold less active than
phenothiazines 7b and 7e as compared by their FIC
indices (Table 2). Generally, all of these compounds
displayed better modulating activity than verapamil in
chloroquine-resistant cell lines.
of half a helical turn.10b A hydrophobic pocket containing
two aromatic phenylalanine residues interacts with the
tricyclic nucleus, and a hydrophilic region, which is
composed of three glutamic acid residues, interacts with
the positively charged nitrogen atom of the side chain
in an electrostatic manner. Similar to the proposal of
Reid on the interaction between phenothiazines and
calmodulin, our results indicated that the optimal
length of alkyl linker probably dictated by the distance
of the two binding sites in target protein. Elongation of
the alkyl bridge from four carbons to 5-6 carbons may
affect the binding of the ring nucleus and the amino
group of the modulators to their corresponding binding
sites, resulting in less active compounds. In addition,
the compromised activity of 7s and 7t as MDR modula-
tors suggested that the structure of diphenylamine is
too flexible to fit snugly in the hydrophobic binding
pocket. Thus, compounds with a restricted tricyclic ring
system were better modulators than those with flexible
diphenylamines. Compounds with bulky substituent at
the side-chain amino group interacted poorly with acidic
residues aligned in the hydrophilic pocket as indicated
by the loss of MDR modulating activity in compounds
containing N-methyl-N-benzylamino (7k ) or N,N-diben-
zylamino (7l) groups.
Drug-resistant parasites from different regions of the
world responded differently to the test compounds as
shown in Table 3, which listed the FIC indices of the
new modulators against TM91C235 (from Thailand),
RCS (from Brazil), and Peruvian PC49 (from South
America). In general, the new modulators exhibited
better anti-MDR activity in isolates from Southeast Asia
(W2 and TM91C235) than in isolates from South
America (RCS and PC49). In addition, the new test
compounds also demonstrated clear differences in
chemosensitizing activity observed in the two South
American chloroquine-resistant isolates, RCS and PC49.
The test compounds exhibited better reversing effects
in the Brazil isolate (RCS) than in the Peru isolate (PC
49). While a number of compoundss7a , 7b, 7e, 7g, 7h ,
7m , 7q, and 7u sdisplayed significant reversing activity
The MDR-reversing activity is also affected by the
length of the alkyl bridge, which connects the hydro-
phobic aromatic ring and the terminal amino group.
Elongation of the alkyl side chain of the modulators
from three carbons to 4-6 carbons retained the chemo-
sensitizing activity. However, the anti-MDR efficacy
decreased as the length of alkyl bridge increased from
four to six carbons. This trend was clearly demonstrated
in Figure 2. For example, the curve of compound 7m ,
which has a four-carbon alkyl linker, was more concave
toward the left of the central line as compared to its
five-carbon (7v) and six-carbon (7x) analogues. The
results indicated that 7m was a more potent anti-MDR
agent than 7v and 7x (FIC ) 0.32 vs 0.48 and 0.57,
respectively). A similar relationship was observed with
compounds 7j, 7u , and 7w .
Figure 3 showed the effects of a series of phenothi-
azine derivatives in which the size of terminal amino
functions was varied. Phenothiazine derivatives having
noncyclic amines 7a (dimethylamino) and 7b (diethy-
lamino) and five-membered cyclic amine 7e (FIC values
0.23, 0.23, and 0.21, respectively) were more active
chemosensitizers than were those with six-membered
cyclic amines, 7f, 7g, and 7h (FIC values 0.49, 0.39, and
0.4, respectively).
The anti-MDR activity and structural requirements
of compounds synthesized in this study are reminiscent
of those found to be important for interactions between
phenothiazines and CaM (calmodulin).10a The relation-
ship between structure and hydrophobicity for anti-
MDR activities suggests that, similar to CaM, chemosen-
sitizers interact in both a hydrophobic and an electro-
static manner with a protein target. Reid et al., who
studied molecular modeling of phenothiazines and their
interaction with the target protein, calmodulin, have
proposed that there are two binding sites at a distance

New Chemosensitizers in MDR Plasmodium falciparum
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 13 2745
Ta ble 3. FIC Indices of New Modulators in Parasites from
Different Regions
sulfate, concentrated, and chromatrographed on a silica gel
column. Elution of the column with 10% ethyl acetate/hexanes
afforded 146 mg (100%) of colorless liquid 1b. 1H NMR (CDCl3,
600 Hz) δ 3.68 (2H, t), 3.57 (2H, t), 1.84 (2H, m), 1.61 (2H, m),
1.54 (2H, m); MS (m/z) 121, 105. Anal. (C5H11OCl) C, H.
Gen er a l P r oced u r e for th e P r ep a r a tion of Com p ou n d s
2a -c: 2-[(4-Ch lor obu tyl)oxy]tetr a h yd r op yr a n (2a ). To a
mixture of 1a (0.54 g, 5 mmol) and dihydropyran (0.76 g, 6
mmol) in 15 mL of CH2Cl2 was added pyridinium p-toluene-
sulfonate (PPTS, 110 mg, 0.4 mmol). After being stirred at
room temperature overnight, the mixture was washed with
water, dried over Na2SO4, concentrated, and chromatro-
graphed on a silica gel column. Elution of the column with
10% ethyl acetate/hexanes afforded 0.96 g (99%) of colorless
liquid 2a : 1H NMR (CDCl3, 600 Hz) δ 4.57 (1H, s), 3.86-3.75
(2H, m), 3.59-3.41 (4H, m), 1.91-1.51 (10H, m); MS (m/z) 193,
157, 103. Anal. (C9H17O2Cl) C, H.
2-[(5-Ch lor open tyl)oxy]tetr ah ydr opyr an (2b): yield 98%;
1H NMR (CDCl3, 300 Hz) δ 4.57 (1H, s), 3.89-3.71 (2H, m),
3.56-3.35 (4H, m), 1.86-1.47 (12H, m); MS (m/z) 207, 105.
Anal. (C10H19O2Cl‚0.25H2O) C, H.
2-[(6-Ch lor oh exyl)oxy]tetr a h yd r op yr a n (2c): yield 97%;
1H NMR (CDCl3, 600 Hz) δ 4.60 (1H, s), 3.89-3.75 (2H, m),
3.57-3.40 (4H, m), 1.84-1.43 (14H, m); MS (m/z) 221, 119,
101. Anal. (C11H21O2Cl) C, H.
compd
TM91C235a
RCSb
Peruvian PC49c
verapamil
7a
7b
7e
7f
7g
7h
7i
7j
7m
7n
7o
7p
7q
7r
7s
7t
7u
0.46
0.28
0.29
0.25
0.52
0.32
0.48
0.26
0.41
0.42
0.54
0.54
0.54
0.33
0.49
0.3
0.63
0.51
0.41
0.51
0.68
0.37
0.49
0.52
0.57
0.51
0.36
0.89
0.66
0.39
0.59
0.33
0.37
0.36
0.37
0.72
0.60
0.61
NTd
0.41
0.5
0.83
0.81
0.98
0.62
0.57
0.66
0.73
0.78
NT
0.6
0.74
NT
0.32
0.48
0.55
0.59
0.54
NT
0.82
0.59
NT
7v
7w
7x
0.92
a
b
Isolate from Southeast Asia. Isolate form Brazil. c Isolate
d
form South America. NT, not tested.
Gen er a l P r oced u r e for th e P r ep a r a tion of Com p ou n d s
4a -f: 10-[4-[(Tetr a h yd r op yr a n -2-yl)oxy]bu tyl]-10H-p h e-
n oth ia zin e (4a ). A reaction mixture containing phenothiazine
(2.1 g, 10.5 mmol) and sodium amide (0.46 g, 12 mmol) in
anhydrous xylenes was refluxed for 2 h. After the mixture was
cooled to room temperature, compound 2a (2.26 g, 12 mmol)
was added and the mixture was refluxed for an additional 6
h. To the reaction mixture was added ice water, dropwise, to
quench excess sodium amide. The mixture was filtered, washed
with water, dried over Na2SO4 and concentrated. The residue
was purified with a silica gel column and was eluted with 5%
ethyl acetate/hexanes to afford 1.46 g (46%) of 4a . 1H NMR
(CDCl3, 600 Hz) δ 7.17 (4H, m), 6.92 (4H, m), 4.56 (1H, s),
3.94 (2H, t), 3.88-3.76 (2H, m), 3.53-3.42 (2H, m), 1.94-1.52
(10H, m); MS (m/z) 355 (M+), 272, 200. Anal. (C21H25O2NS) C,
H, N.
N-[4-[(Tet r a h yd r op yr a n -2-yl)oxy]b u t yl]im in od ib en -
zyl (4b): yield 31%; 1H NMR (CDCl3, 600 Hz) δ 7.15 (6H, m),
6.95 (2H, t), 4.55 (1H, s), 3.84-3.71 (4H, m), 3.50-3.39 (2H,
m), 3.20 (4H, s), 1.71-1.51 (10H, m); MS (m/z) 351 (M+), 268,
250, 196. Anal. (C23H29O2N) C, H, N.
N-[5-[(Tetr a h yd r op yr a n -2-yl)oxy]p en tyl]im in od iben -
zyl (4c): yield 64%; 1H NMR (CDCl3, 600 Hz) δ 7.14 (6H, m),
6.94 (2H, t), 4.57 (1H, s), 3.86-3.71 (4H, m), 3.87-3.70 (2H,
m), 3.20 (4H, s), 1.71-1.44 (12H, m); MS (m/z) 366 (MH+), 282,
264, 208. Anal. (C24H31O2N) C, H, N.
in W2, TM91C235, and RCS (FIC ) 0.5), the same test
compounds exhibited weak to marginal anti-MDR activ-
ity in PC49 (FIC > 0.6).
The assessment of anti-MDR activity of the most
active compound 7e in Aotus monkey against chloro-
quine-resistant P. falciparum is currently in progress.
Con clu sion s
Structure-activity relationship studies of the new
chemosensitizers synthesized in this study indicated
that elongation of the alkyl side chain from three to six
carbons retained the modulating activity, with four-
carbon bridge analogues being the most active. The
phenothiazine ring exhibited the best anti-MDR activity
among the four different ring systems studied. Steric
tolerance at the terminal amino function is limited, as
evidenced by the dramatic loss of the modulating
activity when the size of the substituent on the amino
group increased. The most active compound 7e combines
all three optimized structural features, which consist
of a phenothiazine ring and a pyrrolidinyl group joined
by a four-carbon alkyl bridge. Compound 7e (FIC )
0.21) is more than twice as active as verapamil (FIC )
0.51), one of the best-known chemosensitizers.
N-[6-[(Tet r a h yd r op yr a n -2-yl)oxy]h exyl]im in od ib en -
zyl (4d ): yield 81%; 1H NMR (CDCl3, 600 Hz) δ 7.14 (6H, m),
6.94 (2H, t), 4.57 (1H, s), 3.89-3.70 (4H, m), 3.52-3.35 (2H,
m), 3.19 (4H, s), 1.85-1.34 (14H, m); MS (m/z) 380 (MH+), 296,
208. Anal. (C25H33O2N‚0.5H2O) C, H, N.
Exp er im en ta l Section
N -[4-[(T e t r a h y d r o p y r a n -2-y l)o x y ]b u t y l]im in o s t il-
1
Ch em istr y. Proton nuclear magnetic resonance (1H NMR)
spectra were measured on a Bruker AC-300 or Avance-600
spectrometer with Me4Si(TMS) as the internal reference.
Elemental analyses were performed by Atlantic Microlab Inc.,
Norcross, GA, where analyses are indicated by symbols of the
elements; the analytical results obtained were within (0.4%
of the theoretical values. Mass spectra were recorded on a
Finnigan LCQ mass spectrometer. Silica gel (70-230 mesh),
from EM, was used for column chromatography.
ben e (4e): yield 60%; H NMR (CDCl3, 300 Hz) δ 7.24 (2H,
m), 7.06-6.94 (6H, m), 6.71 (2H, s), 4.49 (1H, s), 3.82-3.63
(4H, m), 3.47-3.32 (2H, m), 1.77-1.46 (10H, m); MS (m/z) 349
(M+), 266, 248, 194. Anal. (C23H27O2N) C, H, N.
Dip h en yl[4-[(t et r a h yd r op yr a n -2-yl)oxy]b u t yl]a m in e
1
(4f): yield 55%; H NMR (CDCl3, 600 Hz) δ 7.29 (4H, t), 7.0
(4H, d), 6.97 (2H, t), 4.59 (1H, s), 3.87-3.75 (4H, m), 3.52-
3.42 (2H, m), 1.80-1.54 (10H, m); MS (m/z) 325 (M+), 270, 242,
224. Anal. (C21H27O2N) C, H, N.
P r ep a r a tion of 5-Ch lor o-1-p en ta n ol (1b). To a solution
of methyl 5-chlorovalerate (180 mg, 1.2 mmol) in 4 mL of
anhydrous diethyl ether was added dropwise LiAlH4 (1.2 mL
of 1 M LAH solution in THF, 1.3 mmol). After 1 h, the excess
of LiAlH4 was decomposed by addition of ice-cold diluted
sulfuric acid and the white solid was removed by filtration.
The filtrate was extracted with ethyl ether three times. The
ether extracts were combined, dried over anhydrous sodium
Gen er a l P r oced u r e for th e P r ep a r a tion of Com p ou n d s
5a -f: 4-(P h en oth ia zin -10-yl)bu ta n -1-ol (5a ): PPTS (14 mg,
0.056 mmol) was added to the solution of 4a (214 mg, 0.56
mmol) in 10 mL of a mixture of methanol/THF (1:1). The
reaction mixture was heated to 60 °C for 4 h. The solvent was
evaporated to dryness under reduced pressure. The residue
was dissolved in 10 mL of CH2Cl2, washed with water, dried
over Na2SO4, and concentrated. The crude oil was purified with

2746 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 13
Guan et al.
a silica gel column and eluted with 2% CH3OH/CH2Cl2 to give
153 mg (100%) of 5a . 1H NMR (CDCl3, 300 Hz) δ 7.16 (4H,
m), 6.89 (4H, m), 3.90 (2H, t), 3.63 (2H, t), 1.86 (2H, m), 1.68
(2H, m); MS (m/z) 271 (M+), 254, 200. Anal. (C16H17ONS‚
0.25H2O) C, H, N.
sealed tube for 24 h. The reaction mixture was successively
filtered, diluted with water, extracted with ethyl acetate, dried
over Na2SO4, and concentrated. The crude oil was chromato-
graphed on a silica gel column and eluted with 2% CH3OH/
1
CH2Cl2 to yield 56 mg (44%) of 7a . H NMR (CDCl3, 300 Hz)
δ 7.13 (4H, m), 6.88 (4H, m), 3.87 (2H, t), 2.26 (2H, t), 2.17-
(6H, s), 1.83 (2H, m), 1.59 (2H, m); MS (m/z) 298 (M+), 254,
200, 100. Anal. (C18H22N2S‚0.25CH3CO2C2H5) C, H, N.
10-[4-(Dieth yla m in o)bu tyl]p h en oth ia zin e (7b): yield
4-(10,11-Dih ydr odiben zo[b,f]azepin -5-yl)bu tan -1-ol (5b):
yield 73%; 1H NMR (CDCl3, 300 Hz) δ 7.09 (6H, m), 6.92 (2H,
m), 3.75 (2H, t), 3.60 (2H, t), 3.17 (4H, s), 1.70-1.54 (4H, m);
MS (m/z) 268 (MH+), 250, 196. Anal. (C18H21ON) C, H, N.
5-(10,11-Dih yd r od ib en zo[b,f]a zep in -5-yl)p en t a n -1-ol
(5c): yield 91%; 1H NMR (CDCl3, 300 Hz) δ 7.09 (6H, m), 6.89
(2H, m), 3.74 (2H, t), 3.57 (2H, t), 3.17 (4H, s), 1.69-1.32 (6H,
m); MS (m/z) 282 (MH+), 264, 208. Anal. (C19H23ON) C, H, N.
6-(10,11-Dih ydr odiben zo[b,f]azepin -5-yl)h exan -1-ol (5d):
yield 83%; 1H NMR (CDCl3, 300 Hz) δ 7.08 (6H, m), 6.90 (2H,
m), 3.72 (2H, t), 3.58 (2H, t), 3.16 (4H, s), 1.62-1.26 (8H, m);
MS (m/z) 296 (MH+), 278, 208. Anal. (C20H25ON‚0.25H2O) C,
H, N.
1
53%; H NMR (CDCl3, 300 Hz) δ 7.13 (4H, m), 6.88 (4H, m),
3.87 (2H, t), 2.47 (6H, m), 1.81 (2H, m), 1.59 (2H, m), 0.98 (6H,
s); MS (m/z) 326 (M+), 311, 254, 128. Anal. (C20H26N2S‚
0.25H2O) C, H, N.
10-[4-(Met h ylben zyla m in o)bu t yl]p h en ot h ia zin e (7c):
yield 86%; 1H NMR (CDCl3, 300 Hz) δ 7.26 (5H, m), 7.14 (4H,
m), 6.88 (4H, m), 3.86 (2H, t), 3.50 (2H, s), 2.43 (2H, t), 2.18
(3H, s), 1.85 (2H, m), 1.68 (2H, m); MS (m/z) 375 (MH+), 176.
Anal. (C24H26N2S‚0.25H2O) C, H, N.
4-(Diben zo[b,f]a zep in -5-yl)bu ta n -1-ol (5e): yield 68%; 1H
NMR (CDCl3, 300 Hz) δ 7.26 (2H, m), 7.03 (6H, m), 6.75 (2H,
s), 3.76 (2H, t), 3.57 (2H, t), 1.78-1.57 (4H, m); MS (m/z) 266
(MH+), 248, 194. Anal. (C18H19ON‚0.25CH3CO2C2H5) C, H, N.
4-(Dip h en yla m in o)bu ta n -1-ol (5f): yield 78%; 1H NMR
(CDCl3, 300 Hz) δ 7.26 (4H, m), 6.98 (6H, m), 3.78 (2H, t), 3.67
(2H, t), 1.80-1.55 (4H, m); MS (m/z) 242 (MH+), 224, 170. Anal.
(C16H19ON) C, H, N.
Gen er a l P r oced u r e for th e P r ep a r a tion of Com p ou n d s
6a , 6b, a n d 6e: 10-(4-Ch lor obu tyl)p h en oth ia zin e (6a ). To
a solution of 5a (2.03 g, 7.5 mmol) in 100 mL of dry benzene
was added, dropwise, thionyl chloride (2 mL, 10 mmol). The
mixture was stirred at room temperature overnight. The
solvent was evaporated to dryness under reduced pressure.
The residue was applied on a silica gel column and eluted with
10% ethyl acetate/hexanes to yield 2.17 g (62%) of 6a . 1H NMR
(CDCl3, 300 Hz) δ 7.16 (4H, m), 6.89 (4H, m), 3.89 (2H, t), 3.54
(2H, t), 2.02-1.85 (4H, m); MS (m/z) 290 (M+), 254, 212, 199.
Anal. (C16H16NS) C, H, N.
10-[4-(Diben zyla m in o)bu tyl]p h en oth ia zin e (7d ): yield
92%; 1H NMR (CDCl3, 300 Hz) δ 7.26 (10H, m), 7.10 (4H, m),
6.88 (2H, t), 6.76 (2H, t), 3.73 (2H, t), 3.50 (4H, s), 2.41 (2H,
t), 1.79 (2H, m), 1.63 (2H, m); MS (m/z) 451 (MH+), 252. Anal.
(C30H30N2S‚0.25CH3CO2C2H5) C, H, N.
10-[4-(P yr r olid in -1-yl)bu tyl]p h en oth ia zin e (7e): yield
1
34%; H NMR (CDCl3, 300 Hz) δ 7.13 (4H, m), 6.88 (4H, m),
3.87 (2H, t), 2.45 (6H, m), 1.85 (2H, m), 1.74 (4H, m), 1.64 (2H,
m); MS (m/z) 324 (M+), 126. Anal. (C20H24N2S) C, H, N.
10-[4-(P iper idin -1-yl)bu tyl]ph en oth iazin e (7f): yield 75%;
1H NMR (CDCl3, 300 Hz) δ 7.13 (4H, m), 6.88 (4H, m), 3.85
(2H, t), 2.30 (6H, t), 1.83 (2H, m), 1.64 (2H, m), 1.58 (4H, m),
1.42 (2H, m); MS (m/z) 338 (M+), 140. Anal. (C21H26N2S‚0.25CH3-
CO2C2H5) C, H, N.
10-[4-(Mor p h olin -4-yl)bu tyl]p h en oth ia zin e (7g): yield
1
73%; H NMR (CDCl3, 300 Hz) δ 7.14 (4H, m), 6.89 (4H, m),
3.88 (2H, t), 3.64 (4H, t), 2.33 (6H, t), 1.85 (2H, m), 1.61 (2H,
m); MS (m/z) 340 (M+), 142. Anal. (C20H24N2SO) C, H, N.
10-[4-(4-Meth ylpiper azin -1-yl)bu tyl]ph en oth iazin e (7h ):
yield 31%; 1H NMR (CDCl3, 300 Hz) δ 7.14 (4H, m), 6.89 (4H,
m), 3.87 (2H, t), 2.35 (10H, m), 2.27 (3H, s), 1.84 (2H, m), 1.62
(2H, m); MS (m/z) 354 (MH+), 155. Anal. (C21H27N3S‚1H2O)
C, H, N.
5-(4-Ch lor obu tyl)im in od iben zyl (6b): yield 62%;1H NMR
(CDCl3, 300 Hz) δ 7.15 (6H, m), 6.98 (2H, m), 3.81 (2H, t), 3.53
(2H, t), 3.21 (4H, s), 1.90-1.67 (4H, m); MS (m/z) 286 (M+),
250, 208. Anal. (C18H20NCl) C, H, N.
[4-(10,11-Dih yd r od iben zo[b,f]a zep in -5-yl)bu tyl]d im e-
th yla m in e (7i): yield 54%; H NMR (CDCl3, 300 Hz) δ 7.10
(6H, m), 6.91 (2H, m), 3.74 (2H, t), 3.15 (4H, s), 2.22 (2H, t),
2.18 (6H, s), 1.55 (4H, m); MS (m/z) 294 (M+), 100. Anal.
(C20H26N2‚0.25H2O) C, H, N.
5-(4-C h lo r o b u t y l)-10,11-d ih y d r o -5H -d ib e n zo [b ,f]-
a zep in e (6e): yield 80%; 1H NMR (CDCl3, 300 Hz) δ 7.26 (2H,
m), 7.02 (6H, m), 6.72 (2H, s), 3.75 (2H, t), 3.48 (2H, t), 1.85
(2H, m), 1.70 (2H, m); MS (m/z) 284 (M+), 248, 206. Anal.
(C18H18NCl) C, H, N.
1
[4-(10,11-Dih yd r od iben zo[b,f]a zep in -5-yl)bu tyl]d ieth y-
la m in e (7j): yield 42%; 1H NMR (CDCl3, 300 Hz) δ 7.10 (6H,
m), 6.90 (2H, m), 3.74 (2H, t), 3.16 (4H, s), 2.45 (4H, q), 2.34
(2H, t), 1.52 (4H, m), 0.95 (6H, t); MS (m/z) 322 (M+), 128.
Anal. (C22H30N2) C, H, N.
Gen er a l P r oced u r e for th e P r ep a r a tion of Com p ou n d s
6c, 6d , a n d 6f:
5-(5-Ch lor op en tyl)-5H-d iben zo[b,f]a zep in e (6c). To a
cooled 25 mL round-bottom flask containing 5c (50 mg, 0.16
mmol) and Ph3P (51 mg, 0.2 mmol) was added hexachloroac-
etone (HCA, 0.05 mL, 0.3 mmol) in 1 mL of CH2Cl2. The
reaction mixture was allowed to come to room temperature
and stirred overnight. The mixture was subjected to purifica-
tion by flash silica gel column chromatography, with elution
first by hexane to remove the HCA and then by 5% ethyl
[4-(10,11-Dih yd r od iben zo[b,f]a zep in -5-yl)bu tyl]m eth -
1
ylben zyla m in e (7k ): yield 93%; H NMR (CDCl3, 300 Hz) δ
7.23 (5H, m), 7.05 (6H, m), 6.89 (2H, t), 3.65 (2H, t), 3.41 (2H,
s), 3.20 (4H, s), 2.28 (2H, t), 2.16 (3H, s), 1.52 (4H, m); MS
(m/z) 370 (M+), 176. Anal. (C26H30N2‚0.25CH3CO2C2H5) C, H,
N.
1
acetate/hexanes to give 49 mg (93%) of 6c. H NMR (CDCl3,
600 Hz) δ 7.14 (6H, m), 6.96 (2H, m), 3.78 (2H, t), 3.51 (2H, t),
3.20 (4H, s), 1.76 (2H, m), 1.62 (2H, m), 1.52 (2H, m); MS (m/
z) 300 (M+), 264, 208. Anal. (C19H22NCl) C, H, N.
[4-(10,11-Dih yd r od iben zo[b,f]a zep in -5-yl)bu tyl]d iben -
zyla m in e (7l): yield 46%; 1H NMR (CDCl3, 300 Hz) δ 7.23
(10H, m), 7.05 (6H, m), 6.89 (2H, t), 3.62 (2H, t), 3.46 (4H, s),
3.13 (4H, s), 2.33 (2H, t), 1.52 (4H, m); MS (m/z) 447 (MH+),
355, 252. Anal. (C32H34N2‚0.5H2O) C, H, N.
5-[4-(P yr r olid in -1-yl)bu tyl]-10,11-d ih yd r o-5H-d iben zo-
[b,f]a zep in e (7m ): yield 44%; 1H NMR (CDCl3, 300 Hz) δ 7.09
(6H, m), 6.90 (2H, m), 3.74 (2H, t), 3.16 (4H, s), 2.39 (6H, m),
1.73 (4H, m), 1.57 (4H, m); MS (m/z) 320 (M+), 126. Anal.
(C22H28N2‚0.25H2O) C, H, N.
5-[4-(P ip er id in -1-yl)bu tyl]-10,11-d ih yd r o-5H-d iben zo-
[b,f]a zep in e (7n ): yield 75%; 1H NMR (CDCl3, 600 Hz) δ 7.13
(6H, m), 6.94 (2H, m), 3.77 (2H, t), 3.20 (4H, s), 2.28 (6H, m),
1.58 (10H, m); MS (m/z) 334 (M+), 140. Anal. (C23H30N2) C, H,
N.
5-(6-Ch lor oh exyl)-5H-d iben zo[b,f]a zep in e (6d ): yield
100%; 1H NMR (CDCl3, 600 Hz) δ 7.14 (6H, m), 6.95 (2H, m),
3.77 (2H, t), 3.51 (2H, t), 3.20 (4H, s), 1.73 (2H, m), 1.61 (2H,
m), 1.40 (4H, m); MS (m/z) 314 (M+), 278, 208. Anal. (C20H24
-
NCl) C, H, N.
(4-Ch lor obu tyl)d ip h en yla m in e (6f): yield 65%; 1H NMR
(CDCl3, 300 Hz) δ 7.26 (4H, m), 6.97 (6H, m), 3.74 (2H, t), 3.54
(2H, t), 1.82 (4H, m); MS (m/z) 260 (M+), 224, 182. Anal.
(C16H18NCl‚0.25H2O) C, H, N.
Gen er a l P r oced u r e for th e P r ep a r a tion of Com p ou n d s
7a -x: 10-[4-(Dim eth yla m in o)bu tyl]p h en oth ia zin e (7a ).
To a solution of 6a (125 mg, 0.4 mmol) in 1 mL of THF were
added 1 mL of dimethylamine (30% in H2O) and a catalytic
amount of Na2CO3. The mixture was heated to 100 °C in a
5-[4-(Mor p h olin -4-yl)bu tyl]-10,11-d ih yd r o-5H-d iben zo-
[b,f]a zep in e (7o): yield 88%; 1H NMR (CDCl3, 300 Hz) δ 7.09

New Chemosensitizers in MDR Plasmodium falciparum
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 13 2747
(6H, m), 6.90 (2H, m), 3.74 (2H, t), 3.64 (4H, t), 3.15 (4H, s),
2.32 (4H, t), 2.25 (2H, t), 1.55 (4H, m); MS (m/z) 336 (M+),
142. Anal. (C22H28N2O) C, H, N.
fold. These dilutions were performed automatically by a
Biomek 1000 or 2000 liquid handling system into 96-well
microtiter plates. Aliquots (25 µL) of each diluted candidate
compound were then transferred to test plates, 200 µL of
parasitized erythrocytes (0.2% parasitemia and 1% hematocrit)
was added to each aliquot, and the plates were incubated at
37 °C in a controlled environment of 5% CO2, 5% O2, and 90%
5-[4-(4-Meth ylp ip er a zin -1-yl)bu tyl]-10,11-d ih yd r o-5H-
1
d iben zo[b,f]a zep in e (7p ): yield 72%; H NMR (CDCl3, 300
Hz) δ 7.09 (6H, m), 6.90 (2H, m), 3.73 (2H, t), 3.15 (4H, s),
2.36 (10H, m), 2.26 (3H, s), 1.54 (4H, m); MS (m/z) 349 (M+),
155. Anal. (C23H31N3‚0.25H2O) C, H, N.
3
N2. After 42 h, 25 µL of H-hypoxanthine was added and the
plates were incubated for an additional 24 h. At the end of
the 66 h incubation period, the plates were frozen at -70 °C
to lyse the red cells and later thawed and harvested onto glass
fiber filter mats by using a 96-well cell harvester. The filter
mats were then counted in a scintillation counter and the
data were analyzed. For each drug, the concentration-
response profile is determined and 50% inhibitory concentra-
tions (IC50) and 90% inhibitory concentrations (IC90) are
determined by use of a nonlinear, logistic dose-response
analysis program.
[4-(Diben zo[b,f]a zep in -5-yl)bu t yl]d iet h yla m in e (7q):
yield 61%; 1H NMR (CDCl3, 300 Hz) δ 7.24 (2H, m), 7.00 (6H,
m), 6.71 (2H, s), 3.72 (2H, t), 2.45 (4H, q), 2.35 (2H, t), 1.52
(4H, m), 0.95 (6H, t); MS (m/z) 320 (M+), 128. Anal. (C22H28N2)
C, H, N.
5-[4-(P yr r olid in -1-yl)bu tyl]-10,11-d ih yd r o-5H-d iben zo-
[b,f]a zep in e (7r ): yield 48%; 1H NMR (CDCl3, 300 Hz) δ 7.24
(2H, m), 7.00 (6H, m), 6.71 (2H, s), 3.72 (2H, t), 2.40 (6H, m),
1.73 (4H, m), 1.60 (4H, m); MS (m/z) 319 (MH+), 126. Anal.
(C22H26N2‚0.25H2O) C, H, N.
N,N-Dieth yl-N′,N′-diph en ylbu tan e-1,4-diam in e (7s): Yield
40%; H NMR (CDCl3, 300 Hz) δ 7.25 (4H, m), 6.96 (6H, m),
Eva lu a tion of Ch em osen sitizin g Activity of th e New
1
Mod u la tor s. Fifty percent inhibitory concentrations (IC50s
)
3.71 (2H, t), 2.51 (4H, q), 2.42 (2H, t), 1.65 (2H, m), 1.50 (2H,
m), 1.00 (6H, t); MS (m/z) 297 (MH+), 224, 128. Anal.
(C20H28N2‚0.25H2O) C, H, N.
Diph en yl[(4-(pyr r olidin -1-yl)bu tyl]am in e (7t): yield 25%;
1H NMR (CDCl3, 300 Hz) δ 7.25 (4H, m), 6.96 (6H, m), 3.71
(2H, t), 2.51 (4H, m), 2.47 (2H, t), 1.79 (4H, m), 1.70 (2H, m),
1.58 (2H, m); MS (m/z) 294 (M+), 126. Anal. (C20H26N2‚0.25CH3-
CO2C2H5) C, H, N.
were determined for each candidate compound alone and for
candidate compounds in fixed combinations of their respective
IC50s (1:1, 1:3, 3:1). These data were used to calculate fractional
inhibitory concentration (FICs).13 The FIC is the actual IC50
of one compound in the presence of a second compound but is
expressed as a fraction of its IC50 when used alone. This index
is a mathematical representation of the isobologram such that
an FIC index of 1.0 represents the line of additive on the
isobologram. An FIC index of less than 1.0 represents synergy
or potentiation, and an FIC index of greater than 1.0 repre-
sents antagonism.
5-[5-(Dieth yla m in o)p en tyl]-10,11-d ih yd r o-5H-d iben zo-
[b,f]a zep in e (7u ): yield 29%; 1H NMR (CDCl3, 300 Hz) δ 7.08
(6H, m), 6.90 (2H, m), 3.72 (2H, t), 3.15 (4H, s), 2.47 (4H, q),
2.33 (2H, t), ∼1.30-1.62 (6H, m), 0.97 (6H, t); MS (m/z) 337
(MH+). Anal. (C23H32N2‚0.25H2O) C, H, N.
Refer en ces
5-[5-(P yr r olid in -1-yl)p en tyl]-10,11-d ih yd r o-5H-d iben -
zo[b,f]a zep in e (7v): yield 50%; H NMR (CDCl3, 300 Hz) δ
1
(1) Nosten, F.; Ter Kuile, F.; Chongsuphajaisiddhi, T.; Luxemburger,
C.; Webster, H. K.; Edstein, M.; Phaipun, L.; Thew, K. L.; White,
N. J . Mefloquine-resistant Falciparum Malaria on the Thai-
Burmese Border. Lancet 1982, 337, 1140-1143.
(2) Oduola, A. M.; Milhous, W. K.; Salako, L, A.; Walker, O.;
Desjardins, R. E. Reduced In Vitro Susceptibility to Mefloquine
in West African Isolates of Plasmodium falciparum. Lancet 1987,
2, 1304-1305.
(3) Gerena, L.; Bass, G, T.; Kyle, D. E.; Oduola, A. M. J .; Milhous,
W. K.; Martin, R. K. Fluoxetine Hydrochloride Enhances In
Vitro Susceptibility to Chloroquine in Resistant Plasmodium
falciparum. Antimicrob. Agents Chemother. 1992, 36, 2761-
2765.
(4) Bitonti, A. J .; Sjoerdsma, A.; McCann, P. P.; Kyle, D. E.; Oduola,
A. M. J .; Rossan, R. N.; Milhous, W. K.; Davidson, D. E. Reversal
of Chloroquine Resistance in Malaria Parasite Plasmodium
falciparum by Desipramine. Science 1988, 242, 1301-1303.
(5) Ye, Z, G.; Van Dyke, K. V. Selective Antimalaria Activity of
Tetrandrine Against Chloroquine Resistant Plasmodium falci-
parum. Biochem. Biophys. Res. Commun. 1989, 159, 242-248.
(6) (a) Kyle, D. E.; Oduola, A. M.; Martin, S. K.; Milhous, W. K.
Plasmodium falciparum: Modulation by Calcium Antaganists
of Resistance to Chloroquine, Desethylchloroquine, Quinine, and
Quinidine In Vitro. Trans. R. Soc. Trop. Med. Hyg. 1990, 84,
474-478. (b) Martin, S. K.; Oduola, A. M.; Milhous, W. K.
Reversal of chloroquine resistance in Plasmodium falciparum
by verapamil. Science 1987, 235, 899-901.
(7) Ford, J . M.; Prozialeck, W. C.; Hait, W. N. Structure Features
Determining Activity of Phenothiazines and Related Drugs for
Inhibition of Cell Growth and Reversal of Multidrug Resistance.
Mol. Pharmacol. 1989, 35, 105-115.
(8) Ford, J . M.; Hait, W. N. Pharmacology of Drugs That Alter
Multidrug Resistance in Cancer. Pharmacol. Rev. 1990, 42, 155-
199.
7.08 (6H, m), 6.89 (2H, m), 3.72 (2H, t), 3.15 (4H, s), 2.44 (4H,
m), 2.36 (2H, t), 1.74 (4H, m), ∼1.28-1.62 (6H, m); MS (m/z)
335 (MH+), 140. Anal. (C23H30N2‚0.25CH3CO2C2H5) C, H, N.
5-[6-(Dieth yla m in o)h exyl]-10,11-d ih yd r o-5H-d iben zo-
[b,f]a zep in e (7w ): yield 36%; 1H NMR (CDCl3, 300 Hz) δ 7.10
(6H, m), 6.90 (2H, m), 3.71 (2H, t), 3.15 (4H, s), 2.62 (4H, q),
2.47 (2H, t), ∼1.20-1.60 (8H, m), 1.05 (6H, t); MS (m/z) 351
(MH+). Anal. (C24H34N2‚0.25H2O) C, H, N.
5-[6-(P yr r olid in -1-yl)h exyl]-10,11-d ih yd r o-5H-d iben zo-
[b,f]a zep in e (7x): yield 45%; 1H NMR (CDCl3, 300 Hz) δ 7.09
(6H, m), 6.89 (2H, m), 3.71 (2H, t), 3.15 (4H, s), 2.44 (4H, m),
2.34 (2H, t), 1.75 (4H, m), ∼1.20-1.60 (8H, m); MS (m/z) 349
(MH+). Anal. (C24H32N2‚0.25H2O) C, H, N.
Biology: In Vitr o Dr u g Su scep tibility Meth od s. The
in vitro antimalarial drug susceptibility assay used was a
modification of the procedures first published by Desjardins
et al. with modifications developed by Milhous et al.11,12 In
brief, the assay is based on the incorporation of radiolabeled
hypoxanthine by the parasites, and inhibition of isotope
incorporation is attributed to activity of known or candidate
antimalarial drugs. For each assay, proven antimalarials,
such as chloroquine, mefloquine, quinine, artemisinin, py-
rimethamine, and sulfadoxine, were used as controls. The
incubation period was 66 h and the starting parasitemia was
0.2% with a 1% hematocrit. The medium used was RPMI-1640
culture medium with no folate or p-aminobenzoic acid (PABA).
Albumax may be used instead of 10% normal heat-inactivated
human plasma. The primary difference in Albumax versus
human plasma is less protein binding of the drug and, hence,
many compounds are slightly more active in this model.
(9) Magid, R. M.; Talley, B. G.; Souther, S. K. Improvements in the
Hexachloroacetone/Triphenylphosphine Procedure for the Con-
version of Allylic Alcohols into Chlorides. J . Org. Chem. 1981,
46, 824-825.
(10) (a) Prozialeck, W. C.; Weiss, B. Interaction of Calmodulin by
Phenothiazines and Related Drugs: Structure-Activity Rela-
tionships. J . Pharmacol. Exp. Ther. 1982, 222, 509-516. (b) Reid,
R. E. Drug Interactions with Calmodulin: the Binding Site. J .
Theor. Biol. 1983, 105, 63-76.
If a candidate compound was tested without any prior
knowledge of its activity or solubility, the compound was
dissolved directly in dimethyl sulfoxide (DMSO) and diluted
400-fold with complete culture medium. These unknown
compounds were normally started at a highest concentration
of about 50 000 ng/mL. The compounds were then diluted
2-fold, 11 times, to give a concentration range of about 1048-

2748 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 13
Guan et al.
(11) (a) Desjardins, R. E.; Canfield, C. J .; Haynes, D. E.; Chulay, J .
D. Quantitative Assessment of Antimalarial Activity In Vitro
Antimalarial Drugs, Antimicrob. Agents Chemother. 1985, 27,
525-530.
(13) Elion, G. V.; Singer, S.; Hitchings, G. H. Antagonists of nucleic
acid derivatives. III. Synergism in combinations of biochemically
related antimetabolites. J . Biol. Chem. 1954, 208, 477-488.
by
a Semiautomated Microdilution Technique. Antimicrob.
Agents Chemother. 1979, 16, 710-718. (b) Chulay, J . D.; Haynes,
J . D.; Diggs, C. L. Plasmodium falciparum: Assessment of In
Vitro Growth by [3H]hypoxanthine Incorporation. Exp. Parasitol.
1983, 55, 138-146.
(12) Milhous, W. K.; Weatherly, N. F.; Bowdre, J . H.; Desjardins, R.
E. In Vitro Activities of and Mechanisms of Resistance to Antifol
J M010549O