Pentacyclo-undecane derived cyclic tetra-amines: Synthesis and evaluation as potent anti-tuberculosis agents

Article abstract of DOI:10.1016/j.ejmech.2009.07.015

As part of an ongoing effort to develop highly potent anti-tuberculosis agents, fourteen pentacyclo-undecane (PCU) tetra-amine compounds were synthesized and screened for their in vitro anti-mycobacterial activity against two TB strains, H37Rv and XDR 194 [an extensively drug-resistant strain of tuberculosis]. Using the broth macrodilution method, nitrofuranylamide based compounds (6a and 6b) showed almost similar activities against the H37Rv strain of Mycobacterium tuberculosis when compared with the control drug, ethambutol. N-Geranyl piperazine PCU (8a) and trans-trans farnesyl piperazine PCU (8b) were 3.2 and 3.7 times more potent than commercially available ethambutol. Both isoprenyl PCU tetra-amine derivatives and N-decyl piperazine PCU (9a) were highly active against the XDR 194 strain of tuberculosis with MICs in the range of 0.63-3.02 μM. Cytotoxicities (IC₅₀) of isoprenyl based compounds (8a, 8b) and compound 9a were tested on a mammalian cell line [MDBK (Madin Darby bovine kidney epithelium)] with values of 30, 24 and 25 μM respectively.

Full text of DOI:10.1016/j.ejmech.2009.07.015

European Journal of Medicinal Chemistry 44 (2009) 4297–4305
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Pentacyclo-undecane derived cyclic tetra-amines: Synthesis and evaluation
as potent anti-tuberculosis agents
Oluseye K. Onajole a, Karnishree Govender b, Patrick Govender c, Paul D. van Helden d, Hendrik G. Kruger a,
,
Glenn E.M. Maguire a, Karen Muthusamy c, Manormoney Pillay b, Ian Wiid d, Thavendran Govender e
*
a School of Chemistry, University of KwaZulu-Natal, Durban, South Africa
b Department of Medical Microbiology, Nelson R Mandela School of Medicine, University of KwaZulu-Natal, Durban, South Africa
c Department of Biochemistry, University of KwaZulu-Natal, Durban, South Africa
d Department of Biomedical Sciences, Faculty of Health Sciences, University of Stellenbosch, Tygerberg, South Africa
e School of Pharmacy and Pharmacology, University of KwaZulu-Natal, Durban, South Africa
a r t i c l e i n f o
a b s t r a c t
Article history:
As part of an ongoing effort to develop highly potent anti-tuberculosis agents, fourteen pentacyclo-
undecane (PCU) tetra-amine compounds were synthesized and screened for their in vitro anti-myco-
bacterial activity against two TB strains, H37Rv and XDR 194 [an extensively drug-resistant strain of
tuberculosis]. Using the broth macrodilution method, nitrofuranylamide based compounds (6a and 6b)
showed almost similar activities against the H37Rv strain of Mycobacterium tuberculosis when compared
with the control drug, ethambutol. N-Geranyl piperazine PCU (8a) and trans–trans farnesyl piperazine
PCU (8b) were 3.2 and 3.7 times more potent than commercially available ethambutol. Both isoprenyl
PCU tetra-amine derivatives and N-decyl piperazine PCU (9a) were highly active against the XDR 194
Received 11 March 2009
Received in revised form
15 July 2009
Accepted 16 July 2009
Available online 22 July 2009
Keywords:
XDR-tuberculosis (TB)
Pentacycloundecane
Isoprenyl
strain of tuberculosis with MICs in the range of 0.63–3.02
compounds (8a, 8b) and compound 9a were tested on a mammalian cell line [MDBK (Madin Darby
bovine kidney epithelium)] with values of 30, 24 and 25 M respectively.
Ó 2009 Elsevier Masson SAS. All rights reserved.
mM. Cytotoxicities (IC50) of isoprenyl based
SQ109
m
parts of the world [3] is almost untreatable and is accompanied by
higher mortality rates than MDR-TB [5].
1. Introduction
The difficulties associated with the long duration of therapy, such
as patient non-compliance following improvement after the inten-
sive phase and ingestionof multiple doses consistingof many tablets,
have also led to a need to develop simple drug regimen. The ensuing
limited treatment options for MDR and XDR-TB have created
a renewed interest in the development of novel anti-TB drug candi-
dates. These new drugs must effectively shorten treatment time and
act against the subpopulation of slowly metabolising bacilli.
Among the promising potential anti-TB drugs currently under-
going human clinical trials are the third generation fluo-
roquinolones, gatifloxacin and moxifloxacin, diarylquinolone TMC
207, nitroimidazole PA-824 and nitroimidazo-oxazole OPC-67683
[6]. Other drugs that are in the preclinical phase include the
diamine SQ109 (2), dipiperidines SQ609, nitroimidazo-oxazole
back-up, synthase inhibitor FAS20013, translocase I inhibitors and
non-fluorinated quinolones [6].
The WHO estimate of 9.2 million new tuberculosis cases with 1.7
million deaths in 2006, indicates that tuberculosis (TB) still accounts
for a significantly large proportion of the global disease burden and
mortality [1]. India, China, Indonesia, South Africa and Nigeria
featured as the top five of the highest affected countries. The highest
incidence rates were found in Africa, with twelve of the fifteen
countries in this continent having the most severe infection rates.
The worldwide increase in prevalence of Mycobacterium.
tuberculosis and the emergence of multidrug-resistant (MDR) TB
and extensively drug-resistant (XDR) TB has imposed a serious
setback on global TB control [2]. There were approximately 0.5
million reported cases of MDR-TB throughout the world in 2006 [2].
Aside from protracted treatment periods [2] and the use of more
toxic and expensive second line drugs [3], MDR-TB is associated
with low cure rates and high mortality [4]. XDR-TB found in most
SQ109
(N-geranyl-N0-(2-adamantyl)ethane-1,2-diamine),
developed by Sequella Inc. was first synthesized by Lee et al.[7]
using solid phase synthesis of 1,2 diamine analogues of ethambutol
(EMB) (1) (Fig. 1). From the several active 1,2 diamine derivatives
* Corresponding author. Tel.: þ27 31 2608212; fax: þ27 31 2603091.
E-mail address: govenderthav@ukzn.ac.za (T. Govender).
0223-5234/$ – see front matter Ó 2009 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2009.07.015

4298
O.K. Onajole et al. / European Journal of Medicinal Chemistry 44 (2009) 4297–4305
H
afford the endo-8,endo-11 diol (15) which upon dehydration under
N
CH2CH3
Dean–Stark conditions, yielded the corresponding 3,5-dially-4-
oxahexacyclo [5.4.1.02,6.03,10.05,9.08,11] dodecane (16) [20]. Ozonol-
ysis of the hexacyclic ether (16) followed by reductive work up
yielded the PCU diol (17) [18] which upon tosylation gave the PCU
ditosylate (18) [18]. The overall yield of 18 from 14 was 60%.
PCU ditosylate (18) was reacted with excess piperazine/homo-
piperazine in dry dichloromethane at ꢀ78 ꢁC to afford the tetra-
amines 5a and 5b respectively (62%) and these were reacted with 5-
nitrofuran-2-carbonyl chloride (19) to obtain compounds 6a and 6b
respectively (65%). PCU ditosylate (18) was reacted with N-benzyl
homopiperazine (20) at a 1:2.2 ratio with triethylamine under
reflux conditions to obtain compound 7 (Scheme 2).
HN
H3CH2C
OH
OH
Ethambutol (1)
NH HN
N
N
SQ109 (2)
SQ775 (3)
Fig. 1. Structures of ethambutol, SQ109 and SQ 775.
Compound 5a was successfully reacted with geranyl bromide
(21a) to afford compound 8a but its purification proved difficult via
column chromatography. To solve this problem, a new synthetic
route was introduced. Applying the methodology employed to
synthesize compound 5a, geranyl bromide (21a) was reacted with
excess piperazine/homopiperazine to obtain N-geranyl piperazine
(22a) and N-geranyl homopiperazine (23a) while the reaction of
trans–trans farnesyl bromide (21b) with piperazine/homopiper-
azine afforded compounds 22b and 23b. Piperazine was also
reacted with three linear alkane chains (C10, C15 and C20) to obtain
compounds 25a, 25b and 25c (Scheme 3).
Reaction of PCU ditosylate (18) with compound 22a and 22b in
the presence of K2CO3 under reflux resulted in compounds 8a and
8b respectively (yield 71%) which were easily purified via column
chromatography. Attempts to react PCU ditosylate (19) with 23a
and 23b were unsuccessful. PCU ditosylate (18) was reacted with
25a, 25b and 25c to yield linear alkane derivatives of PCU dipi-
perazine 9a, 9b and 9c respectively. Reaction of the PCU ditosylate
(18) with N-benzoyl piperazine (26) yielded compound 10, fol-
lowed by the reduction of the carbonyl group using lithium
aluminium hydride to obtain the corresponding benzyl tetra-amine
11 in 55% yield (Scheme 4).
2-(Aminomethyl) pyridine (27) and ethanolamine (28) were
reacted with benzaldehyde (29) via reductive amination to obtain
their corresponding secondary amines, N-benzyl-N-{(pyridin-2-
yl)methyl}amine (30) [21,22] and 2-(benzylamino)ethanol (31) [23]
(Scheme 5).
N-Benzyl-N-{(pyridin-2-yl)methyl}amine (30) was reacted with
PCU ditosylate (18) to obtain compound 32 after which the benzyl
protecting groups were removed by catalytic hydrogenation using
ammonium formate with 10% palladium on carbon to obtain
compound 12 (yield 67%). 2-(Benzylamino) ethanol (31) was
treated with PCU ditosylate (18) to obtain compound 13 in an
overall yield from 18 of 70% (Scheme 6).
obtained using a combinatorial approach, SQ109 (2) was found to
be the most potent new anti-TB lead. SQ109 (2) is an effective
compound against EMB resistant strains [8]. Its activity against
XDR-TB strains and a 25% reduction in the time to cure mice by
enhancing the activity of isoniazid and rifampicin was also repor-
ted. SQ 775 (3) is an example of a cyclic diamine compound with an
activity similar to the control drug EMB [9].
Non-TB related research showed that polycyclic ‘‘cage’’
compounds (such as adamantane and pentacyclo-undecane) can
improve drug lipophilicity, thus serving as a transport aid in
carrying such drugs across the Blood Brain Barrier (BBB) or Central
Nervous System (CNS) [10,11]. It can also reduce the bio-degrada-
tion of the drug thereby prolonging the pharmaceutical effect of
such agents in the body [11–14]. SQ109 and SQ775 both have an
adamantane moiety in common suggesting that this skeleton
might be contributing to its lipophilicity thereby enhancing their
anti-TB activities.
In a similar development, Tangallapally et al. [15,16] recently
reported novel nitrofuranylamide compounds with promising in
vitro anti-tuberculosis activities. However, this activity could not be
duplicated in in vivo analysis due to a short serum half life and rapid
elimination of the compound. This led to the introduction of
a bicyclic (tetrahydroisoquinoline) moiety in order to make it more
resistant to proteolysis [17]. This particular modification showed an
improvement in serum half life while improving its MIC activity
(Fig. 2) for compound 4 [17].
For our work it was proposed that the replacement of the
bicyclic tetrahydroisoquinoline with
a more rigid lipophilic
compound such as the pentacycloundecane (PCU) moiety, a poly-
cyclic ‘‘cage’’, might further enhance the activity by facilitating the
movement of these molecules across the lipid-enriched bacterial
cell membrane.
Inspired by the work of Lee et al. [7] and Tangallapally et al. [15–
17] a series of novel amine based compounds (5–13) bearing the
lipophilic PCU molecule were designed and synthesized (Fig. 3).
3. Results and discussion
All synthesized novel PCU amine derivatives 5–13 excluding 9
were screened in vitro against M. tuberculosis H37Rv (ATCC 25618)
using a broth macrodilution method (BMM). The Clog P of each
novel compound was obtained using the ACDLABS LogP 11.0
program.1 The results are depicted in Table 1.
MIC results were obtained after 21 days of incubation.
Compounds 5a and 5b differ in structure only by the ring sizes of
the diamines. Compound 5b shows low activity against the H37Rv
strain, while 5a did not show any activity at the highest concen-
tration tested. Modification to these compounds with the intro-
duction of a nitrofuran group to form compounds 6a and 6b gave
2. Chemistry
The starting material, PCU ditosylate [18] was synthesized from
pentacyclo-undecane-8,11-dione as illustrated in Scheme 1. The
starting material, Cookson’s dione (14) [19] was reacted with
freshly prepared allyl magnesium bromide (Grignard reaction) to
O
O
O2N
N
N
N
CH3
promising activity with MIC values of 24.1 and 23.1 mM respectively,
4
1
Fig. 2. Lead compound 4.
Downloaded from www.acdlabs.com.

O.K. Onajole et al. / European Journal of Medicinal Chemistry 44 (2009) 4297–4305
4299
O
O
O
O
O
N'
N'
N
N'
N'
N
N'
N
N'
N
N'
N
N'
N
N'
N'
n
n
n
n
N
H
N
H
N
R
N
R
O
O
O
O
9a: R = C10H21
9b: R = C15H31
9c: R = C20H41
n
n
7
NO2 O2N
5a: n = 1
5b: n = 2
6a: n = 1
6b: n = 2
8a: n = 2
8b: n = 3
O
O
O
O
N'
N'
N
N'H
N
N'
N
HN'
N
N'
N
N
HO
OH
N
N
O
O
13
12
11
10
Fig. 3. PCU amine derivatives 5–13.
which are similar to the control drug, EMB. Compounds 6a and 6b
did not show any significant difference in calculated Clog P values
when compared to 5a and 5b.
Compound 7 (the benzyl derivative of 5b) exhibits similar
activity to 5b (mg/mL) while compound 11 (the benzyl derivative of
5a) demonstrates an improved activity against H37Rv, whilst 5a did
not show comparable activity. The benzoyl derivative 10 (of 5a)
showed a weaker potency compared to 11. Compounds 12 and 13
did not show any significant activity.
Based on results obtained by Lee et al. [7,8] in the development
of SQ109 (2), it was decided to synthesize the geranyl and trans–
trans farnesyl derivative of 5a and 5b. As discussed above, the
isoprenyl derivatives of 5b proved to be elusive. Compounds 8a and
Based on the results for compounds 6a, 6b, 8a and 8b we
decided to test the molecules with a more accurate method (BAC-
TEC 460 TB system) against the H37Rv strain. Compounds 6a and
6b (Table 2) showed little difference when compared to the results
obtained from the BMM method. Further screening was not carried
out, although it is known that MICs of drugs differ slightly
depending on the assay method used [25]. Due to the activity of 8a
and 8b we decided to investigate the effect of using long alkane
chains in place of the alkene derivatives to determine whether this
functionality is essential for efficacy. This led to the synthesis of
compounds 9a (ten carbon alkane), 9b (fifteen carbon alkane) and
9c (twenty carbon alkane) (Scheme 4). Anti-mycobacterial
screening of compounds 9b and 9c was not possible however; due
to their insolubility at biological pH presumably due to their highly
lipophilic nature (Clog P values are 14.43 ꢂ 0.52 and 19.74 ꢂ 0.52
respectively). Compounds 8a, 8b and 9a were tested against an XDR
strain of M. tuberculosis [strain X194, resistant to first line drugs
(Isoniazid and Rifampicin) and second line drugs (Kanamycin,
Ofloxacin and Amikacin)] (BACTEC 460 TB system). These
compounds were further analyzed for cytotoxicity on mammalian
cells from the MDBK (Madin Darby bovine kidney epithelium) line
(Table 2).
8b had MIC values of 6.09 and 5.04
mM respectively. The activity of
these lipophilic alkene bearing compounds (8a and 8b) were
similar (expressed in
mg/mL) to EMB, when screened against H37Rv.
b
a
O
O
O
OH
HO
Compounds 8a and 8b show potent activity against the H37Rv
strain with 8b being approximately two-fold more active than 8a
against both H37Rv and XDR 194 strains respectively. The MIC for
14
15
16
c
EMB on H37Rv is 0.94
lower (0.5–0.25 g/mL, Table 2). The XDR strain used was resistant
to EMB at the breakpoint concentration of 2.5 g/mL in BACTEC
[24]. Therefore, the MIC for the XDR strain lies beyond 2.5 g/mL.
mg/mL [25] the MIC for 8b against H37Rv was
m
m
m
This means that the efficacy of compound 8b is between 2.5 and 5
times more effective in inhibiting the XDR strain than EMB (8b
killed the strain between 1 mg/mL and 0.5 mg/mL, Table 2). Also, the
O
O
OTs
TsO
OH
HO
sensitivity for 8b between H37Rv and X194 was approximately two
fold (Table 2). The results indicate that our compounds probably
have a different cellular target to that of EMB.
9a proved to be twice as active as compound 8a against the XDR
strain of M. tuberculosis. It was not considered necessary to test 9a
18
17
Scheme 1. Reagents and conditions: (a) H2C]CHCH2MgBr, dry THF; (b) Dean–Stark
apparatus, H2SO4, benzene, reflux; (c) O3, dry CH3OH, NaBH4; (d) p-TsCl, powdered
KOH, THF.

4300
O.K. Onajole et al. / European Journal of Medicinal Chemistry 44 (2009) 4297–4305
O
HN
N
O
O2N
Cl
20
c
19
b
a
O
O
O
O
N
N
N
N
N
N
OTs
N
N
TsO
N
N
18
n
n
n
n
N
H
N
H
O
O
O
O
5a: n = 1
5b: n = 2
NO2 O2N
6a: n = 1
7
6b: n = 2
Scheme 2. Reagents and conditions: (a) piperazine/homopiperazine, DCM, ꢀ78 ꢁC, rt, 24 hrs; (b) DCM, Et3N, reflux 16 h; (c) CH3CN, Et3N, reflux, N2 atm.
using the BMM method since the BACTEC system is much more
accurate. Compound 9a has a higher Clog P value compared to 8a
(Table 2), suggesting that lipophilicity could play an important role
in the activity of this class of compounds. However, this assumption
could not be further evaluated due to insolubility problems
encountered with compounds 9b and 9c. It should be noted that
the MICs for 9a and 8b (in ug/ml) fall in the same range
4. Conclusion
A series of novel PCU tetra-amines (fourteen compounds) were
synthesized and screened for their anti-TB activity of which five
compounds (6a, 6b, 8a, 8b and 9a) displayed the highest anti-TB
activity against M. tuberculosis H37Rv. Compounds 8a, 8b and 9a
were the most potent with excellent activities giving MICs of 3.04,
(1 > MIC > 0.5), however conversion to
m
M makes quite a signifi-
1.26 and 1.50 mM respectively against an XDR-TB strain (X194).
cant difference as compound 8b has a higher molar activity than 9a
(8b has a larger molecular weight).
These results indicate that lipophilicity is an important component
of their efficacy (having larger Clog P values than SQ109). This does
however limit their potential application due to solubility problems
(as seen for molecules 9b and 9c). We are examining the design of
new derivatives in terms of their structure–activity relationship. It
is hoped that new more soluble active species can be obtained. The
first generation of these active compounds however, is currently
undergoing further in vitro and in vivo analysis.
Compounds 8a and 8b are structurally related to SQ109, they all
have polycyclic ‘‘cage’’ moieties, diamines (linear or cyclic) and long
alkene chains, which form the basis of this comparison. As illus-
trated in Table 2, 8a is only half as active as SQ109 whereas 8b is
twice as active as SQ109 [8]. This suggests that the length of the
alkene chain (lipophilicity again) may be playing an important role
in the activity of these compounds. The cytotoxicities (IC50, Table 2)
of these compounds (8a, 8b and 9a) and that of SQ109 fall in the
5. Experimental
same range between 20 and 30
least toxic.
mM with compound 8a being the
The NMR data were recorded on Bruker AVANCE III 400 MHz
and 600 MHz instruments using CDCl3 as a solvent. All chemical
shifts (d) were quoted in parts per million downfield from TMS and
A potential anti-TB candidate needs to be evaluated as a drug
combination, i.e. the possibility of incorporating such an anti-TB
candidate with other existing anti-TB drugs. Based on this, further
studies were carried out to investigate the interaction of compound
8b (the most active compound in this series) with two known anti-
TB drugs (Isoniazid and Rifampicin) by means of in vitro testing
against the H37Rv strain of M. tuberculosis. The interaction of
compound 8b with these two anti-TB drugs could be antagonist,
additive, synergistic or have no effect at all. Results obtained using
once again the BACTEC 460 TB system showed that compound 8b
the coupling constants (J) recorded in Hertz. Splitting pattern
abbreviations are as follows: s ¼ singlet, d ¼ doublet, t ¼ triplet,
m ¼ multiplet, br ¼ broad. Infrared spectra were obtained on
a Perkin Elmer Spectrum 100 instrument with an Attenuated Total
Reflectance attachment recorded in cmꢀ1. All reactions were
monitored using Thin Layer Chromatography (TLC, Merck Kieselgel
60, F254). All purifications were carried by Column Chromatog-
raphy using Fluka Kieselgel 60 (70–230 mesh) and
CH3Cl:CH3OH:NH4OH (88:10:2) as the eluent (solvent mixture).
Level of purity for all compounds was judged to be >95% based
upon 1H NMR and LC-MS analysis. Mass Spectra were obtained
using a Waters LCT Premier Time of Flight mass spectrometer.
Tetrahydrofuran was freshly distilled before use from a sodium
benzophenone under N2 atmosphere while dichloromethane was
dried using phosphorus pentoxide prior to use. The syntheses of
the precursors are described in their corresponding references.
Clog P gives an indication of the lipophilicity of the drug with
reference to its pharmacological importance (pharmacokinetics
and pharmacodynamics) [15,26]. The values were calculated using
ACD/Labs LogP software v11.0.
(0.5 mg/mL) had no antagonist or synergistic effect; it did however
have an additive effect.
HN
NH
n'
Br
n
Br
n
24a: n = 2
24b: n = 3
24c: n = 4
a
a
21a: n = 2
21b: n = 3
HN
N
HN
N
n
n
n'
25a: n = 2
25b: n = 3
25c: n = 4
22a: n' = 1, n = 2
22b: n' = 1, n = 3
23a: n' = 2, n = 2
23b: n' = 2, n = 3
5.1. Synthesis of PCU tetra-amine 5a & 5b
To a vigorously stirred solution of the cyclic diamines (piperazine/
homopiperazine) (22 mmol) in DCM (400 mL) at ꢀ78 ꢁC (dry ice, 2-
propanol) under N2 atmosphere was added dropwise PCU ditosylate
Scheme 3. Reagents and conditions: (a) piperazine/homopiperazine, DCM, ꢀ78 ꢁC, rt,
24 h.

O.K. Onajole et al. / European Journal of Medicinal Chemistry 44 (2009) 4297–4305
4301
O
O
OTs
TsO
HN
N
18
a
a
a
26
O
O
O
O
N
N
N
N
N
N
N
N
N
N
N
N
N
b
N
R
N
N
R
O
O
n
n
11
9a: R = C10H21
9b: R = C15H31
9c: R = C20H41
10
8a: n = 2
8b: n = 3
Scheme 4. Reagents and conditions: (a) 22 or 25 or 26, K2CO3, CH3CN, reflux under N2 atm; (b) LiAlH4, dry THF and reflux under N2 atm, 36 h.
(18, 1.2 g, 2.2 mmol) in DCM (100 mL) over 45 min. The reaction
mixture was left to attain room temperature with stirring for 24 h.
The solution was washed with water to remove excess diamines, the
obtained organic extract was dried over Na2SO4 and concentrated in
vacuo. The crude residue was purified via column chromatography
using CH3Cl:MeOH:NH4OH (88:10:2) to obtain a pure product.
reaction mixture was cooled and diluted with 60 mL of ethyl
acetate and washed sequentially with (2 ꢃ 50 mL) 10% NaHCO3,
(2 ꢃ 50 mL) water and (2 ꢃ 50 mL) brine. The organic layer was
dried over anhydrous Na2SO4 and concentrated in vacuo. The crude
residue was purified via column chromatography using
CH3Cl:MeOH:NH4OH (88:10:2).
5.1.1. PCU dipiperazine (5a)
5.2.1. 5-Nitrofuran-2-carbonyl piperazine PCU (6a)
A yellow oil (Rf ¼ 0.2, 0.56 g, 68% yield). IR vmax: broad absorp-
tion (N–H) 3367, 2949, 1655, 1463, 1125 and 746 cmꢀ1. HRMS
calculated for C23H36N4O (M þ Hþ) 385.2955, found 385.2967. 1H
NMR [CDCl3, 400 MHz]; dH 1.52 (AB, JAB ¼ 10.1 Hz, 1H), 1.87 (AB,
JAB ¼ 10.1 Hz, 1H), 1.99 (t, J ¼ 7.92 Hz, 2H), 2.38–2.60 (m, 8H), 2.95
(2H). 13C NMR [CDCl3, 100 MHz]: dC 29.7 (t), 41.8 (d), 43.4 (t), 44.5
(d), 45.6 (t), 48.0 (d), 53.7 (t), 55.1 (t), 58.8 (d), 94.87 (s).
A brown oil (Rf ¼ 0.7, 0.56 g, 65%). IR nmax: 2927, 1630, 1352,1021
and 752 cmꢀ1. HRMS calculated for C33H38N6O9 (M þ Hþ) 663.2779,
found 663.2758. 1H NMR [CDCl3, 400 MHz]; dH 1.52 (JAB ¼ 10.1 Hz,
1H), 1.87 (JAB ¼ 10.1 Hz, 1H), 1.99 (t, J ¼ 7.36 Hz, 2H), 2.38–2.61 (m,
10H), 3.76 (s, 2H), 3.85 (s, 2H), 7.15 (1H), 7.32 (1H). 13C NMR [CDCl3,
100 MHz]: dC 29.9 (t), 41.8 (d), 43.1 (t), 43.1 (t), 43.5 (t), 44.5 (d), 46.6
(t), 48.1 (d), 53.4 (t), 54.4 (t), 58.8 (d), 94.8 (s), 111.7 (d), 118.1 (d),
148.7 (s), 151.2 (s), 156.6 (s).
5.1.2. PCU dihomopiperazine (5b)
A light yellow oil, (Rf ¼ 0.2, 0.55 g, 62%). IR vmax: broad absorp-
tion (N–H) 3386, 2818, 1465, 1108, 927, 918 and 748 cmꢀ1. HRMS
calculated for C25H40N4O (M þ Hþ) 413.3280 found 413.3287. 1H
NMR [CDCl3, 400 MHz] dH 1.48 (AB, JAB ¼ 10.2 Hz, 1H), 1.84 (AB,
JAB ¼ 10.2 Hz, 1H), 1.76 (2H), 1.98 (2H), 2.35–2.69 (m, 10H), 2.89–
2.94 (m, 4H). 13C NMR [CDCl3,100 MHz]: dC 30.0 (t), 30.4 (t), 41.8 (d),
43.4 (t), 44.5 (d), 47.2 (t), 48.0 (d), 48.6 (d), 54.3–55.0 (t), 57.8 (t),
58.8 (d), 94.8 (s).
5.2.2. 5-Nitrofuran-2-carbonyl homopiperazine PCU (6b)
A brownoil(Rf ¼ 0.51 g, 61%). IRvmax: 3479, 2951, 2859,1627,1530,
1352, 810 and 729 cmꢀ1. HRMS calculated for C35H42N6O9 (M þ Hþ)
691.3092, found 691.3105. 1H NMR [CDCl3, 400 MHz]; dH 1.52
(JAB ¼ 10.1 Hz, 1H), 1.87 (JAB ¼ 10.1 Hz, 1H), 1.96–2.03 (m, 4H) 2.37–
2.88 (m, 10H), 3.72–3.77 (m, 2H), 3.84 (2H), 7.19 (1H), 7.34 (1H). 13C
NMR [CDCl3, 100 MHz]: dC 27.0 (t), 28.7 (t), 30.4 (t), 41.8 (d), 43.4 (t),
44.4 (d), 46.2 (t), 47.3 (t), 47.9 (t), 48.0 (d), 48.8 (t), 54.2–54.9 (t), 54.7
(t), 58.8 (d), 94.8 (s), 111.7 (d), 118.0 (d), 149.2 (s), 151.2 (s), 157.9 (s).
5.2. Synthesis of 5-nitrofuran-2-carbonyl diamine PCU 6a & 6b
To a stirred mixture of PCU tetra-amines (5a and 5b, 1.2 mmol)
5.3. Synthesis of N-benzyl homopiperazine PCU (7)
in DCM (2 mL) and Et3N (670
was added freshly prepared 5-nitrofuran-2-carbonyl chloride (19
m
L, 4.8 mmol) under N2 atmosphere
A mixture of N-benzyl homopiperazine (20, 0.7 g, 3.68 mmol)
and PCU ditosylate (18, 0.93 g, 1.67 mmol) and triethylamine
[15] 3.6 mmol) in DCM (3 mL) and stirred for 18 h at reflux. The
(350 mL, 2.5 mmol) in CH3CN (20 mL) was refluxed for four days
N
NH2
H2N
OH
N
28
b
N
H
27
O
N
H
OH
a
30
29
31
Scheme 5. Reagents and conditions: (a) CH3CH2OH, rt, l h, NaBH4, reflux, stir overnight, N2 atm; (b) CH3OH, rt, 2 h, NaBH4, stir overnight, N2 atm.

4302
O.K. Onajole et al. / European Journal of Medicinal Chemistry 44 (2009) 4297–4305
b
a
a
O
O
O
O
N
N
OTs
TsO
NH
HN
N
N
N
N
HO
OH
N
N
12
32
18
13
Scheme 6. Reagents and conditions: (a) 30 or 31, CH3CN, Et3N, reflux, N2 atm; (b) 10% Pd/C, ammonium formate, dry MeOH, reflux, 16 h, N2 atm.
under N2 atmosphere. The reaction mixture was cooled, filtered and
concentrated in vacuo. The crude residue was purified via column
chromatography on silica gel using CH3Cl:MeOH:NH4OH (88:10:2,
Rf ¼ 0.65) as eluent to give the product as a yellowoil (0.84 g, 85%). IR
vmax: 2935, 2811, 1452, 1351, 1111, 729 and 695 cmꢀ1. HRMS calcu-
lated for C39H52N4O (M þ Hþ) 593.4219 found 593.4230. 1H NMR
[CDCl3, 400 MHz] dH 1.48 (JAB ¼ 10.2 Hz,1H),1.84 (JAB ¼ 10.2 Hz,1H),
1.77 (2H), 1.96 (t, J ¼ 8.0 Hz, 2H), 2.35–2.73 (m, 14H), 3.60 (s, 2H),
7.26–7.31 (m, 5H).13C NMR [CDCl3,100 MHz]: dC 27.5 (t), 30.4 (t), 41.8
(d), 43.4 (t), 44.5 (d), 48.0 (d), 53.4–55.1 (t), 58.8 (d) 62.8 (t), 94.9 (s),
126.8 (d), 128.2 (d), 128.9 (d), 139.5 (s).
5.4.1. N-Geranyl piperazine (22a)
A yellow oil (Rf ¼ 0.5, 0.52 g, 68%). 1H NMR [CDCl3, 400 MHz]; dH
1.56 (s, 3H), 1.60 (s, 3H), 1.64 (s, 3H), 1.98–2.08 (m, 4H), 2.52 (NH
proton), 2.98 (d, 2H), 5.04 (t, J ¼ 7.12 Hz, 1H), 5.22 (m, 1H). 13C NMR
[CDCl3, 100 MHz]: dC 16.4 (q), 17.7 (q), 25.7 (q), 26.4 (t), 39.8 (t), 45.1
(t), 52.8 (t), 56.2 (t), 120.1 (d), 124.0 (d), 131.6 (s), 139.7 (s).
5.4.2. N-Trans–trans farnesyl piperazine (22b)
A yellow oil (Rf ¼ 0.6, 0.5 g, 65%). 1H NMR [CDCl3, 400 MHz]; dH
1.51 (s, 3H), 1.52 (s, 3H), 1.56 (s, 3H), 1.59 (s, 3H), 1.86–2.01 (m, 8H),
2.40 (NH), 2.90 (m, 2H), 4.98–5.02 (m, 2H), 5.17 (t, 1H, J ¼ 6.6 Hz).
13C NMR [CDCl3, 100 MHz]: dC 16.0 (q), 16.4 (q), 17.6 (q), 25.7 (q),
26.3 (t), 26.7 (t), 39.7 (t), 39.7 (t), 45.2 (t), 53.1 (t), 56.0 (t), 56.3 (t),
120.2 (d), 123.8 (d), 124.3 (d), 131.1 (s), 135.0 (s), 139.2 (s).
5.4. Synthesis of N-isoprenyl/linear alkane diamines
To a vigorously stirred solution of diamine (piperazine/homo-
piperazine) (10 mmol) in DCM (750 mL) at ꢀ78 ꢁC (dry ice, 2-
propanol) under N2 atmosphere was added dropwise a solution of
isoprenyl bromide (21a or 21b, 2 mmol)/linear alkane bromide
(24a, 24b or 24c, 2 mmol) in DCM (250 mL) over 45 min. The
reaction mixture was left to attain room temperature with stirring
for 24 h. The solution was washed with water to remove excess
piperazine, the organic extract was dried over Na2SO4 and
concentrated in vacuo. The crude residue was purified via column
chromatography using CH3Cl:MeOH:NH4OH (88:10:2).
5.4.3. N-Geranyl homopiperazine (23a)
A yellow oil (Rf ¼ 0.5, 0.48 g, 64%). 1H NMR [CDCl3, 400 MHz]; dH
1.47 (s, 3H), 1.50 (s, 3H),1.55 (s, 3H),1.69 (sxt, 2H),1.88–1.98 (m, 4H),
2.52–2.55 (m, 4H), 2.82–2.88 (m, 4H), 2.98 (d, 2H), 3.51 (NH), 4.93–
4.97 (m, 1H), 5.11–5.13 (m, 1H). [CDCl3, 100 MHz]: dC 16.1 (q), 17.6
(q), 25.5 (q), 26.1 (t), 29.4 (t), 39.5 (t), 46.5 (t), 48.0 (t), 54.3 (t), 55.9
(t), 56.9 (t), 121.4 (d), 123.9 (d), 131.2 (s), 138.2 (s).
5.4.4. N-Trans–trans farnesyl homopiperazine (23b)
A yellow oil (Rf ¼ 0.6, 0.52 g, 66%). 1H NMR [CDCl3, 400 MHz]; dH
1.58 (s, 3H), 1.59 (s, 3H), 1.63 (s, 3H), 1.68 (s, 3H), 1.80 (sxt, 2H), 1.95–
1.98 (m, 2H), 2.04–2.12 (m, 6H), 2.59 (NH), 2.63–2.68 (m, 4H), 2.92–
2.98 (m, 4H), 3.11 (d, 2H), 5.07–5.12 (m, 2H), 5.27 (t, 1H). 13C NMR
[CDCl3, 100 MHz]: dC 16.0 (q), 16.4 (q), 17.6 (q), 25.7 (q), 26.3 (t), 26.3
(t), 26.7 (t), 29.9 (t), 39.7 (t), 39.7 (t), 46.9 (t), 48.3 (t), 54.5 (t), 56.2
(t), 57.7 (t), 121.5 (d), 123.8 (d), 124.3 (d), 131.1 (s), 135.1 (s), 138.5 (s).
Table 1
Clog P values and in vitro anti-mycobacterial BMM activity on day 21 of novel PCU
amine compounds against M. tuberculosis H37Rv.
5.4.5. N-C10 piperazine (25a)
O
A white solid (Rf ¼ 0.4, 0.46 g, 58%). 1H NMR [CDCl3, 600 MHz];
dH 0.83 (t, 3H), 1.22 (m, 14H), 1.44 (s, 2H), 1.93 (m, 1H), 2.30 (m, 2H),
2.45 (s, 4H), 2.92 (s, 4H). 13C NMR [CDCl3, 150 MHz]: dC 14.1 (q), 22.6
(t), 26.5 (t), 27.5 (t), 29.3 (t), 29.5 (t), 31.9 (t), 45.1 (t), 53.2 (t), 59.1 (t).
R
R
Compound
R
MW
Clog Pa
H37Rv (MIC)
m
g/mL
mM
1
Ethambutol
–N0(CH2)4NH
–N0(CH2)5NH
–N0(CH2)4NCO(C4H2O)NO2
–N0(CH2)5NCO(C4H2O)NO2
–N0(CH2)5NCH2C6H5
–N0(CH2)4NC10H17
–N0(CH2)4NC15H25
–N0(CH2)4NCOC6H5
–N0(CH2)4NCH2C6H5
2-Pyridylmethylaminob
–N[(CH2)2OH]CH2C6H5
204
385
413
663
691
592
657
793
592
565
429
515
NA
4
19.6
ND
155
5.4.6. N-C15 piperazine (25b)
5a
5b
6a
6b
7
8a
8b
10
11
12
13
0.41 ꢂ 0.55
1.18 ꢂ 0.54
0.57 ꢂ 0.83
1.32 ꢂ 0.81
4.11 ꢂ 0.73
7.35 ꢂ 0.73
11.42 ꢂ 0.77
1.47 ꢂ 0.67
2.61 ꢂ 0.83
0.82 ꢂ 0.40
4.22 ꢂ 0.54
>128
64
16
16
64
4
4
128
64
128
128
A white solid (Rf ¼ 0.4, 0.45 g, 58%). 1H NMR [CDCl3, 600 MHz];
dH 0.89 (t, 3H), 1.26–1.28 (m, 24H), 1.49 (s, 2H), 1.81 (s, 1H), 2.30 (m,
2H), 2.41 (s, 3H), 2.91 (s, 4H). 13C NMR [CDCl3, 150 MHz]: dC 14.1 (q),
22.7 (t), 26.7 (t), 27.6 (t), 29.3 (t), 29.6 (t), 29.6 (t), 29.6 (t), 29.7 (t),
31.9 (t), 46.1 (t), 54.7 (t), 59.5 (t).
24.1
23.2
108.1
6.09
5.04
216.2
113.3
298
5.4.7. N-C20 piperazine (25c)
A white solid (Rf ¼ 0.4, 0.48 g, 53%). 1H NMR [CDCl3, 600 MHz];
dH 0.85 (t, 3H), 1.22–1.24 (m, 30H), 1.45 (m, 2H), 1.99 (s, NH), 2.28
(m, 2H), 2.37 (s, 3H), 2.86 (m, 4H). 13C NMR [CDCl3, 150 MHz]: dC
14.0 (t), 22.7 (t), 26.6 (t), 27.6 (t), 29.3 (t), 29.6 (t), 29.6 (t), 29.6 (t),
29.7 (t), 31.9 (t), 46.6 (t), 54.6 (t), 59.5 (t).
248
MIC: Minimal Inhibitory Concentration; ND: Not determined. MW: Molecular
weight. NA: Not applicable.
a
Clog P was calculated using ACD/labs software v11.0.
Connected through the amino group.
b

O.K. Onajole et al. / European Journal of Medicinal Chemistry 44 (2009) 4297–4305
4303
Table 2
BACTEC and IC50 results using the MDBK cell line for selected compounds.
Compound
Clog P
H37Rv (MIC,
mM)
H37Rv (MIC,
m
g/mL)
XDR 194 (MIC,
mM)
XDR 194 (MIC,
m
g/mL)
IC50 (mM)
6a
6b
8a
8b
0.57 ꢂ 0.83
1.32 ꢂ 0.81
7.35 ꢂ 0.73
11.42 ꢂ 0.77
9.12 ꢂ 0.52
6.04 ꢂ 0.45
MIC > 12.07 (SD ꢂ 7.47)
MIC > 11.58 (SD ꢂ 7.16)
1.52 > MIC > 0.76 (SD ꢂ 0.54)
0.63 > MIC > 0.32 (SD ꢂ 0.22)
NT
MIC > 8 (SD ꢂ 4.24)
MIC > 8 (SD ꢂ 4.24)
1 > MIC > 0.5 (SD ꢂ 0.35)
0.5 > MIC > 0.25 (SD ꢂ 0.18)
NT
NT
NT
NT
NT
NT
NT
30
24
25
26
3.04 > MIC > 1.52 (SD ꢂ 0.54)
1.26 > MIC > 0.63 (SD ꢂ 0.45)
1.5 > MIC > 0.75 (SD ꢂ 0.53)
–
2.0 > MIC > 1.0 (SD ꢂ 0.71)
1 > MIC > 0.5 (SD ꢂ 0.35)
1 > MIC > 0.5 (SD ꢂ 0.35)
–
9a
SQ109a
0.63
–
a
Literature value [8]; NT: Not tested.
5.5. Synthesis of PCU piperazine isoprenyl/linear alkane
(d), 43.4 (t), 44.2 (d), 47.9 (d), 53.2 (t), 54.6 (t), 58.8 (d), 58.8 (t),
94.7 (s).
A mixture of N-isoprenyl piperazine/linear alkane piperazine
(1.9 mmol) was reacted with (0.47 g, 0.84 mmol) PCU ditosylate
(18) and K2CO3 (0.175 g, 1.27 mmol) in CH3CN (10 mL) with reflux
under nitrogen atmosphere for four days. The reaction was cooled,
filtered and concentrated in vacuo to obtain a crude product. The
residue was purified via column chromatography on silica gel using
CHCl3/MeOH/NH4OH:88:10:2 as eluent to give the product as
yellow oil.
5.5.5. C20- piperazine PCU (9c)
A yellow solid (Rf ¼ 0.8, 0.58 g, 73%). IR vmax: 2915, 2849, 2809,
1469, 1375, 1163, 1119, 827 and 718 cmꢀ1. MS (TOF) calculated for
C63H116N4O (M þ Hþ) 945.9222, found 945.9210; 1H NMR [CDCl3,
400 MHz] dH 0.86 (m, 3H), 1.22 (s, br, 30H), 1.45 (m, 3H), 1.84
(JAB ¼ 10.2 Hz, 1H), 1.96 (t, 2H), 2.27–2.56 (m, 15H). 13C NMR [CDCl3,
100 MHz]: dC 14.1 (t), 22.7 (t), 26.8 (t), 27.6 (t), 29.4 (t), 29.4 (t), 29.5
(t), 29.6 (t), 29.7 (t), 30.1 (t), 31.9 (t), 41.8 (d), 43.4 (t), 44.5 (d), 48.0
(d), 53.1 (t), 53.2 (t), 54.6 (t), 58.8 (d), 58.8 (t), 94.8 (s).
5.5.1. N-Geranyl piperazine PCU (8a)
A yellow oil (Rf ¼ 0.7, 0.40 g, 71%). IR vmax: 2929, 1672, 1448,
1375, 1294, 1006 and 821 cmꢀ1. HRMS calculated for C43H68N4O
(M þ Hþ) 657.5471, found 657.5446; 1H NMR [CDCl3, 400 MHz] dH
1.51 (JAB ¼ 10.2 Hz, 1H), 1.84 (JAB ¼ 10.2 Hz, 1H), 1.95–2.06 (m, 8H),
2.35–2.57 (m, 12H), 2.95 (d, 2H), 5.04 (t, J ¼ 6.72 Hz, 1H), 5.22 (t,
J ¼ 6.4 Hz, 1H). 13C NMR [CDCl3, 100 MHz]: dC 16.4 (q), 17.7 (q), 25.7
(q), 26.4 (t), 30.1 (t), 39.8 (t), 41.8 (d), 43.4 (t), 44.5 (d), 48.0 (d), 53.0
(t), 53.2 (t), 54.6 (t), 55.8 (t), 56.0 (t), 58.8 (d), 94.9 (s), 120.7 (d),
124.1 (d), 131.5 (s), 138.9 (s).
5.6. N-Benzoyl piperazine PCU (10)
A mixture of N-benzoyl piperazine (26, 0.7 g, 3.68 mmol) and
PCU ditosylate (18, 0.93 g, 1.67 mmol) and K2CO3 (0.28 g, 2.0 mmol)
in CH3CN (20 mL) was refluxed for four days under N2 atmosphere.
The reaction mixture was cooled, filtered and concentrated in
vacuo. The crude residue was purified via column chromatography
on silica gel using CH3Cl:MeOH:NH4OH (88:10:2, Rf ¼ 0.75) as
eluent to give the product as a yellow oil (0.84 g, 85%). IR vmax
:
5.5.2. N-Trans–trans farnesyl piperazine PCU (8b)
3463, 2952, 1623, 1430, 1292 and 723 cmꢀ1. HRMS calculated for
C37H44N4O3 (M þ Hþ) 593.3492, found 593.3505. 1H NMR [CDCl3,
400 MHz]; dH 1.50 (JAB ¼ 10.3 Hz, 1H), 1.85 (JAB ¼ 10.3 Hz, 1H), 2.36–
2.59 (m, 10H), 3.42 (s, 2H), 3.78 (s, 2H) and 7.37 (s, 5H). 13C NMR
[CDCl3, 100 MHz]: dC 29.8 (t), 41.8 (d), 42.0 (t), 43.5 (t), 44.4 (d), 47.6
(t), 48.0 (d), 52.9 (t), 94.7 (s), 127.0 (d), 128.1 (d), 128.5 (d), 135.7 (s),
170.3 (s).
A yellow oil (Rf ¼ 0.8, 0.42 g, 63%). IR vmax: 2929, 1670, 1448,
1375, 1294, 1153, 1006 and 822 cmꢀ1
. HRMS calculated for
C53H84N4O (M þ Hþ) 793.6723, found 793.6714. 1H NMR [CDCl3,
400 MHz] dH 1.48 (JAB ¼ 10.2 Hz, 1H), 1.84 (JAB ¼ 10.2 Hz, 1H), 1.56 (s,
3H), 1.57 (s, 3H), 1.60 (s, 3H), 1.65 (s, 3H), 1.91–2.08 (m, 10H), 2.35–
2.60 (m, 12H), 2.95 (d, 2H), 5.07 (m, 2H), 5.23 (m, 1H). 13C NMR
[CDCl3, 100 MHz]: dC 16.0 (q),16.5 (q),17.7 (q), 25.7 (q), 26.4 (t), 26.7
(t), 30.1 (t), 39.7 (t), 39.8 (t), 41.8 (d), 43.4 (t), 44.5 (d), 48.0 (d), 53.1
(t), 53.2 (t), 56.0 (t), 58.8 (d), 94.9 (s), 120.6 (d), 124.0 (d), 124.2 (d),
131.3 (s), 135.2 (s), 138.9 (s).
5.7. N-Benzyl piperazine PCU (11)
N-Benzoyl piperazine PCU (10, 1 g, 1.7 mmol) was added to
a
stirred suspension of lithium aluminium hydride (0.25 g,
5.5.3. C10-piperazine PCU (9a)
6.8 mmol) in dry THF under nitrogen. The solution was refluxed for
36 h under N2. The solution was diluted with diethyl ether and the
excess LAH was quenched by the dropwise addition of saturated
aqueous Na2SO4. The solution was filtered, dried over anhydrous
Na2SO4 and the solvent was removed in vacuo to afford a residue
which was purified via column chromatography on silica gel using
CH3Cl:MeOH:NH4OH (88:10:2, Rf ¼ 0.62) as eluent to give the
product as a yellow oil (55%). IR vmax: 2948, 1656, 1149 and
734 cmꢀ1. HRMS calculated for C37H48N4O (M þ Hþ) 565.3906,
found 565.3911. 1H NMR [CDCl3, 400 MHz]; dH 1.48 (JAB ¼ 10.3 Hz,
1H), 1.84 (JAB ¼ 10.3 Hz, 1H), 1.96 (t, J ¼ 7.92 Hz, 2H) 2.34–2.56 (m,
10H), 3.47 (s, 2H) and 7.2–7.28 (m, 5H). 13C NMR [CDCl3, 100 MHz]:
dC 29.9 (t), 41.8 (d), 43.4 (t), 44.5 (d), 48.0 (d), 53.1 (t), 54.6 (t), 58.8
(d), 63.0 (t), 94.8 (s), 127.0 (d), 128.2 (d), 129.3 (d), 138.0 (s).
A dark brown oil (Rf ¼ 0.8, 0.41 g, 73%). IR vmax: 2852, 2807,
1464, 1.295, 1161, 826 and 721 cmꢀ1
. HRMS calculated for
C43H76N4O (M þ Hþ) 665.6092, found 665.6114; 1H NMR [CDCl3,
400 MHz] dH 0.79 (t, J ¼ 6.44 Hz, 3H), 1.17–1.19 (m, 14H), 1.40
(JAB ¼ 9.2 Hz, 1H), 1.42 (s, 2H), 1.77 (JAB ¼ 10.2 Hz, 1H), 1.91 (m, 2H),
2.28–2.32 (m, 3H), 2.37–2.52 (m, 11H). 13C NMR [CDCl3, 100 MHz]:
dC 14.1 (q), 22.6 (t), 26.4 (t), 27.5 (t), 29.3 (t), 29.5 (t), 29.5 (t), 31.8
(t), 41.7 (d), 43.4 (t), 44.4 (d), 47.9 (d), 52.6 (t), 54.5 (t), 58.6 (t),
58.7 (d), 94.6 (s).
5.5.4. C15-piperazine PCU (9b)
A light brown oil (Rf ¼ 0.8, 0.48 g, 71%). IR vmax: 2915, 2850,
2808, 1466, 1375, 1163, 1120, 826 and 720 cmꢀ1. MS (TOF) calcu-
lated for C53H96N4O (M þ Hþ) 805.7657, found 805.7698; 1H NMR
[CDCl3, 400 MHz] dH 0.77 (m, 3H), 1.15 (s, br, 24H), 1.40 (m, 3H),
1.76 (JAB ¼ 10.0 Hz, 1H), 1.89 (3H), 2.20 (m, 2H), 2.27–2.40 (m, 11H),
2.49 (s, 2H). 13C NMR [CDCl3, 100 MHz]: dC 14.1 (q), 22.6 (t), 26.9
(t), 27.6 (t), 29.3 (t), 29.5 (t), 29.6 (t), 29.6 (t), 30.1 (t), 31.9 (t), 41.8
5.8. Synthesis of 2-(aminomethyl) pyridine PCU (12)
A mixture of 2-(aminomethyl) pyridine (27, 1 g, 9.2 mmol) and
of benzaldehyde (29, 0.98 g, 9.2 mmol) in ethanol (15 mL) was

4304
O.K. Onajole et al. / European Journal of Medicinal Chemistry 44 (2009) 4297–4305
stirred for 1 h at room temperature under nitrogen atmosphere, the
corresponding imine was reduced with solid NaBH4 (0.7 g,
18 mmol) which was added slowly over 30 min, the mixture was
further stirred for an additional 30 min, the mixture was then
refluxed overnight. The mixture was allowed to cool to RT, an
additional ethanol (15 mL) was added to the reaction vessel after
which 10% HCl was added to quench excess NaBH4. The acidic
mixture was basified with 25% aqueous ammonia solution. The
desired product was extracted with dichloromethane (2 ꢃ 50 mL)
and the solution was dried over Na2SO4 and concentrated in vacuo.
The crude product was distilled under vacuum to afford (1.2 g, 65%)
N-benzyl-N-{(pyridin-2-yl)methyl}amine (30) [21,22]. 1H NMR
[CDCl3, 400 MHz]: dH: 2.25 (NH), 3.84 (s, 2H), 3.92 (s, 2H), 7.14 (t,
1H), 7.26 (d, 1H), 7.30–7.37 (m, 5H), 7.62 (t, 1H), 8.56 (d, 1H). 13C
NMR [CDCl3,100 MHz]: dC 53.4 (t), 54.4 (t),121.8 (d), 122.2 (d),126.9
(d), 128.1 (d), 128.3 (d), 136.3 (d), 140.0 (d), 149.2 (s), 156.7 (s).
In the next step, a mixture of N-benzyl-N-{(pyridin-2-yl)me-
thyl}amine (30,1.5 g, 7.6 mmol), PCU ditosylate (18,1.9 g, 3.4 mmol)
CH3CN (50 mL) was refluxed under nitrogen for four days. The
reaction was monitored using TLC. After completion the reaction
was filtered and concentrated in vacuo. The crude product was
purified via column chromatography on silica gel using
CH3Cl:MeOH:NH4OH (88:10:2, Rf ¼ 0.7) as eluent to give 2-(ben-
zylamino)ethanol PCU (13, 1.3 g, 70%). IR vmax: 3371, 2951, 1601,
1452, 1043, 732 and 697 cmꢀ1. HRMS calculated for C33H42N2O3
(M þ Hþ) 515.3274, found 515.3288. 1H NMR [CDCl3, 400 MHz]: dH
1.44 (JAB ¼ 10.2 Hz, 1H), 1.77 (JAB ¼ 10.2 Hz, 1H), 1.96 (t, J ¼ 7.36 Hz,
2H), 2.25–2.64 (m, 8H) 3.54 (t, 2H), 3.61 (s, 2H), 7.27–7.28 (m, 5H).
13C NMR [CDCl3,100 MHz]: dC 29.3 (t), 41.5 (d), 43.4 (t), 44.1 (d), 47.7
(d), 49.6 (t), 55.0 (t), 58.4 (t), 58.4 (d), 58.7 (t), 95.1 (s), 127.1 (d),
128.3 (s), 129.0 (s).
5.10. Biological testing
5.10.1. Broth macrodilution method
All mycobacterial work was carried out in a Level III Biosafety
laboratory. All synthesized novel ‘‘cage’’ compounds were evalu-
ated in triplicate against M. tuberculosis H37Rv. Each PCU amine
compound was dissolved in methanol to prepare a final concen-
tration of 12.8 mg/mL and dilution a 100 fold with 7H9 broth
and Et3N (710 mL, 5.1 mmol) in CH3CN (20 mL) was refluxed under
nitrogen for four days. The reaction was monitored on TLC, after
completion the reaction was filtered and concentrated in vacuo. The
crude product was purified via column chromatography on silica
gel using CH3Cl:MeOH: NH4OH (88:10:2, Rf ¼ 0.8) as eluent to give
N-benzyl-N-(2-methylpyridine) PCU (32) (1.55 g, 75%). 1H NMR
[CDCl3, 400 MHz]: dH 1.39 (JAB ¼ 10.2 Hz, 1H), 1.73 (JAB ¼ 10.2 Hz,
1H), 2.01 (t, J ¼ 7.8 Hz, 2H), 2.18–2.58 (m, 8H), 3.61 (s, 2H), 3.72
(s, 2H), 7.10 (t, J ¼ 5.2 Hz, 1H), 7.19–7.33 (m, 5H), 7.52 (d, J ¼ 7.7 Hz,
1H), 7.60 (t, J ¼ 7.5 Hz, 1H), 8.47 (d, J ¼ 4.2 Hz, 1H). 13C NMR [CDCl3,
100 MHz]: dC 30.0 (t), 41.6 (d), 43.4 (t), 44.3 (d), 47.8 (d), 50.1 (t),
58.4 (t), 58.7 (d), 60.1 (t), 94.8 (s),121.8 (d),122.8 (d),126.9 (d),128.2
(d), 128.8 (d), 136.4 (d), 139.4 (s), 148.8 (d), 160.4 (s).
medium to give a stock concentration of 128 mg/mL. EMB was dis-
solved in DMSO and water similarly to serve as a standard. These
was diluted two fold in sterile 30 mL universal tubes containing
Middlebrook 7H9 broth supplemented with casitone, glycerol and
10% OADC enrichment to give a concentration range of 128
mg/mL–
0.125 g/mL. Tubes containing Middlebrook 7H9 broth without
m
compounds served as the compound free controls while Mid-
dlebrook 7H9 broth containing solvents alone served as controls to
monitor their inhibitory effect.
A mixture of (1.8 g, 1.7 mmol) N-benzyl-N-{(pyridin-2-yl)me-
thyl}amine PCU (32), ammonium formate (0.54 g, 8.5 mmol) and
150 mg of 10% Pd/C in methanol was refluxed under nitrogen
atmosphere for 15 h [27]. The mixture was cooled to RT, filtered and
solution concentrated. The residue obtained was made alkaline
with NaHCO3 and extracted with CHCl3. The was mixture dried over
Na2SO4, concentrated in vacuo to afford pure 2 (aminomethyl)
pyridine PCU (12) (CHCl3:MeOH:NH4OH, 88:10:2, Rf ¼ 0.65, 0.85 g,
67%). IR vmax: 3314, 2955, 1665, 1590, 1433 and 753 cmꢀ1. HRMS
calculated for C27H32N4O (M þ Hþ) 429.2654, found 429.2667. 1H
NMR [CDCl3, 400 MHz]: dH 1.48 (JAB ¼ 10.2 Hz, 1H), 1.83
(JAB ¼ 10.2 Hz, 1H), 2.01 (t, J ¼ 7.28 Hz, 2H), 2.3–2.56 (m, 4H), 2.76
(m, 2H), 3.89 (s, 2H), 7.12 (t, 1H), 7.30 (d, 1H), 7.60 (t, 1H), 8.51 (d,
1H). 13C NMR [CDCl3, 100 MHz]: dC 32.5 (t), 41.5 (d), 43.5 (t), 44.2
(d), 46.1 (t), 47.9 (d), 55.1 (t), 58.5 (d), 95.4 (s), 122.0 (d), 122.2 (d),
136.5 (d), 149.2 (d), 159.3 (s).
A standardized inoculum was prepared by vortexing a log phase
culture of M. tuberculosis H37Rv in a sterile tube containing 4.5 mL
phosphate buffer, 0.05% Tween 80 and 4–6 glass beads (5 mm
diameter). After allowing the clumps to settle, the supernatant was
aspirated and adjusted to a MacFarland number 1 standard,
equivalent to 1 ꢃ107 colony forming units (CFU)/mL. This was
diluted in 7H9 broth to obtain a concentration of 1 ꢃ105 CFU/mL. Of
this, 500 mL was added into each tube of broth containing the
diluted test compound concentrations, the compound free controls
and the solvent controls. Colony counts of the test inoculum were
prepared by plating out 20 mL onto Middlebrook 7H11 agar plates.
Plates and tubes were incubated aerobically at 37 ꢁC for twenty one
days. Macroscopic readings of the tubes were documented every
seven days until twenty one days.
5.10.2. BACTEC 460 TB analysis
M. tuberculosis reference strain H37Rv (ATCC 25618) and XDR
5.9. Synthesis of N-benzyl ethanolamine PCU (13)
strain ꢃ 194 (drug sensitivity: Isoniazid > 0.2
cin > 1.0 g/mL; EMB > 2.5 g/mL; Kanamycin > 5.0
acin > 2.0 g/mL; Amikacin > 5.0 g/mL) were cultured in
m
g/mL; Rifampi-
m
m
mg/mL; Oflox-
A mixture of benzaldehyde (29, 2.1 g, 20 mmol) and N-etha-
nolamine (28, 1.2 g, 20 mmol) in methanol (20 mL) was stirred at
25 ꢁC under N2 atmosphere for 2 h. The mixture was cooled to 0 ꢁC
using an external ice–salt bath after which NaBH4 (1.5 g, 40 mmol)
was added slowly. The mixture was stirred overnight at RT. Excess
NaBH4 was quenched by adding 10% HCl (20 mL), the mixture was
basified with 25% NH4OH and the product was extracted from the
mixture with dichloromethane (2 ꢃ 30 mL). The solvent was dried
over Na2SO4 and concentrated in vacuo to afford pure 2-(Benzyla-
mino) ethanol (31) (2.2 g, 72%). 1H NMR [CDCl3, 400 MHz]: dH 2.71
(t, 2H), 3.04 (NH), 3.60 (t, 2H), 3.75 (s, 2H), 7.23–7.33 (m, 5H). 13C
NMR [CDCl3, 100 MHz]: dC: 50.6 (t), 53.4 (t), 60.6 (t), 127.0 (d), 128.1
(d), 128.3 (d), 139.6 (s).
m
m
Middelbrook 7H9 medium [28], enriched with OADC (0.005%, v/v,
oleic acid; 0.5%,171 w/v, BSA; 0.2%, w/v, glucose; 0.02%, v/v, catalase
and 0.085%, w/v, NaCl) [24]. Incubation was done with continuous
stirring at 37 ꢁC. Reproducible growth (<1.0% difference) was
recorded under standardized conditions. Cultures with an optical
density between 0.1 and 0.6 (at 600 nm) were in exponential
growth. At an optical density of approximately 0.16, 0.1 mL of
culture was inoculated into a BACTEC vial (Becton Dickinson,
Franklin Lakes, NJ, USA) and incubated at 37 ꢁC to a growth index
(GI) of 500 (þ/ꢀ 50). This culture was used as the primary culture
for drug testing. The growth index is a quantitative determination
of radioactive CO2 on a scale from 0 to 999. Drug compounds were
diluted with methanol (filter sterilized through Millex LG syringe
driven filters, Millipore). Compound concentrations in BACTEC vials
A mixture of 2-(Benzylamino) ethanol (31, 1.2 g, 7.9 mmol), PCU
ditosylate (18, 2 g, 3.6 mmol) and K2CO3 (0.745 g, 5.4 mmol) in

O.K. Onajole et al. / European Journal of Medicinal Chemistry 44 (2009) 4297–4305
4305
ranged between 8
Growth index results were recorded every 24 h. Growth rates of the
mycobacterial strains were calculated as daily growth index
m
g/mL and 0.03125
m
g/mL (final concentration).
Medical Microbiology; University of KwaZulu-Natal for providing
technical assistance during the BMM biological testing of the
compounds.
difference (DGI) where a DGI >10 was considered positive growth.
Purity of M. tuberculosis cultures was monitored by plating onto
7H11 mycobacterium agar plates [24].
References
[1] Global tuberculosis control-surveillance, planning, financing, WHO
report.http://www.who.int/tb/publications/global_report/2007/pdf/full.pdf
(2007) (accessed on 26 June 2008).
5.11. Cytotoxicity analysis
5.11.1. Materials
[2] ‘‘Tuberculosis’’ WHO Fact Sheet No 104, Health Communications, WHO,
Geneva,
2006.http://www.who.int/mediacentre/factsheets/fs104/en/index.
The RPMI 1640 (with 25 mM HEPES and L-glutamine), penicillin/
html (accessed on 14 Nov. 2008).
streptomycin solution, and trypsin-versene mixture were
purchased from Lonza and the heat-inactivated foetal bovine
serum (FBS) was obtained from Invitrogen. Cytotoxicity was
assessed using the Promega CellTiter 96 non-radioactive cell
proliferation assay and an EL ꢃ 800 automated microplate reader
from Bio-Tek Instruments, Inc.
[3] V.L.C. White, J. Moore-Gillon, Thorax 55 (2000) 962–963.
[4] M.A. Espinal, S.J. Kim, P.G. Suarez, K.M. Kam, A.G. Khomenko, G.B. Migliori,
J. Baez, A. Kochi, C. Dye, M.C. Raviglione, J. Am. Med. Assoc. 283 (2000)
2537–2545.
[5] N.R. Gandhi, A. Moll, A.W. Sturm, R. Pawinski, T. Govender, U. Lalloo, K. Zeller,
J. Andrews, G. Friedland, Lancet 368 (2006) 1575–1580.
[6] Medecins Sans Fronteres, Development of new drugs for TB chemo-
therapy.http://www.doctorswithoutborders.org/news/tuberculosis/tb_xdr_
report_full_10-2006.pdf (accessed on 26 Sept. 2008).
5.11.2. Methods
[7] R.E. Lee, M. Protopopova, E. Crooks, R.A. Slayden, M. Terrot, C.E. Barry, J. Comb.
Chem. 5 (2003) 172–187.
The toxicities of the compounds were determined on the MDBK
(Miadin Darby bovine kidney epithelium) cell line (supplied by the
Department of Biochemistry, University of Kwa-Zulu Natal, South
Africa). The cells were grown in a 37 ꢁC incubator as a monolayer in
RPMI 1640 supplemented with 10% (v/v) heat-inactivated FBS and
penicillin/streptomycin at a final concentration of 0.1 mg/mL. Cells
were then trypsinised and plated at a density of 5 ꢃ104 cells per
well into a 96-well plate and incubated for 6 h at 37 ꢁC. Media was
removed from all wells and replaced with fresh media. The test
compounds were dissolved in 1% methanol with the aid of soni-
cation. Dilutions of the test compounds were then prepared using
fresh media before addition to the cells. The control untreated cell
culture was incubated in fresh media containing an appropriate
volume of the 1% methanol solution. This was included as a 100%
viability internal control. All test samples were done in quintupli-
cate and repeated twice on different days. Cells were incubated
with the test samples for 42 h at 37 ꢁC. Cell viability was assessed
on the basis of the conversion of the tetrazolium salt {3-[4,5-
dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide} (MTT)
into purple formazan crystals which can be dissolved using acidi-
fied isopropanol and then spectrophotometrically analyzed. The
Promega CellTiter 96 non-radioactive cell proliferation assay was
performed as per manufacturer’s instructions. Briefly, MTT was
added to each well and the plate incubated at 37 ꢁC for 3 h before
the formazan crystals were dissolved and the absorbance read at
570 nm. Percent cytotoxicity was calculated as follows: percent
cytotoxicity ¼ 100 ꢀ [(OD of sample/OD of control) ꢃ 100%]. All
standard deviations were below 10%.
[8] M. Protopopova, C. Hanrahan, B. Nikonenko, R. Samala, P. Chen, J. Gearhart,
L. Einck, C.A. Nacy, J. Antimicrob. Chemother. 56 (2005) 968–974.
[9] E. Bogatcheva, C. Hanrahan, B. Nikonenko, R. Samala, P. Chen, J. Gearhart,
F. Barbosa, L. Einck, C.A. Nacy, M. Protopopova, J. Med. Chem. 49 (2006)
3045–3048.
[10] J. Zah, G. Terre’Blanche, E. Erasmus, S.F. Malan, Bioorg. Med. Chem. 11 (2003)
3569–3578.
[11] H.T. Nagasawa, J.A. Elberling, F.N. Shirota, J. Med. Chem. 16 (1973)
823–826.
[12] W.J. Geldenhuys, S.F. Malan, J.R. Bloomquist, A.P. Marchand, C. Van der Schyf, J.
Med. Res. Rev. 25 (2005) 21–48.
[13] H.T. Nagasawa, J.A. Elberling, F.N. Shirota, J. Med. Chem. 18 (1975) 826–830.
[14] F.M. Ito, J.M. Petroni, D.P. de Lima, A. Beatriz, M.R. Marques, M.O. de Moraes,
L.V. Costa-Lotufo, R.C. Montnegro, H.I.F. Magalhaes, C.D.O. Pessoa, Molecules
12 (2007) 271–282.
[15] R.P. Tangallapally, R. Yendapally, R.E. Lee, K. Hevener, V.C. Jones,
A.J.M. Lenaerts, M.R. McNeil, Y.H. Wang, S. Franzblau, J. Med. Chem. 47 (2004)
5276–5283.
[16] R.P. Tangallapally, R. Yendapally, R.E. Lee, A.J.M. Lenaerts, J. Med. Chem. 48
(2005) 8261–8269.
[17] R.P. Tangallapally, R.E.B. Lee, A.J.M. Lenaerts, R.E. Lee, Bioorg. Med. Chem. Lett.
16 (2006) 2584–2589.
[18] G.A. Boyle, T. Govender, H.G. Kruger, G.E.M. Maguire, Tetrahedron Asym. 15
(2004) 2661.
[19] R.C. Cookson, E. Crundwell, R.R. Hill, J. Hudec, J. Chem. Soc. (1964)
3062–3067.
[20] A.P. Marchand, K.A. Kumar, A.S. McKim, K. Mlinaric-Majerski, G. Kragol,
Tetrahedron 53 (1997) 3467–3474.
[21] D.J.E. Spencer, B.J. Johnson, W.B. Tolman, Org. Lett. 4 (2002) 1391–1393.
[22] H. Mishra, R. Mukherjee, J. Org. Chem. 692 (2007) 3248–3260.
[23] R.P. Tripathi, N. Saxena, V.K. Tiwari, S.S. Verma, V. Chaturvedi, Y.K. Manju,
A.K. Srivastva, A. Gaikwad, S. Sinha, Bioorg. Med. Chem. 14 (2006) 8186–8196.
[24] S.H. Siddiqi, BACTEC 460 TB System. Product and Procedure Manual Revision,
D. Becton Dickinson Microbiology System, Sparks, MD, 1995.
[25] L.A. Collins, S.G. Franzblau, Antimicrob. Agents Chemother. 41 (1997)
1004–1009.
[26] G. Navarrete-Vazquez, G.M. Molina-Salinas, Z.V. Duarte-Fajardo, J. Vargas-
Villarreal, S. Estrada-Soto, F. Gonzalez-Salazar, E. Hernandez-Nunez, S. Said-
Fernandez, Bioorg. Med. Chem. 15 (2007) 5502–5508.
Acknowledgements
[27] S. Scapecchi, E. Martini, D. Manetti, C. Ghelardini, C. Martelli, S. Dei, N. Galeotti,
L. Guandalini, N.M. Romanelli, E. Teodori, Bioorg. Med. Chem. 12 (2004) 71–85.
[28] G. Middlebrook, Z. Reggiardo, W.D. Tigertt, Am. Rev. Respir. Dis. 115 (1977)
1066–1069.
We thank the National Research Foundation, Gun 2073251,
Aspen Pharmacare and the University of KwaZulu-Natal for finan-
cial support. We also thank Mr. M. Mthethwa, Department of