Tetraazamacrocyclic derivatives and their metal complexes as antileishmanial leads

Article abstract of DOI:10.1016/j.poly.2019.02.027

A total of 44 bis-aryl-monocyclic polyamines, monoaryl-monocyclic polyamines and their transition metal complexes were prepared, chemically characterized, and screened in vitro against the Leishmania donovani promastigotes, axenic amastigotes and intracellular amastigotes in THP1 cells. The IC₅₀ and/or IC₉₀ values showed that 10 compounds were similarly active at about 2-fold less potent than known drug pentamidine against promastigotes. The most potent compound had an IC₅₀ of 2.82 μM (compared to 2.93 μM for pentamidine). Nine compounds were 1.1–13.6-fold more potent than pentamidine against axenic amastigotes, the most potent one being about 2-fold less potent than amphotericin B. Fourteen compounds were about 2–10 fold more potent than pentamidine, the most potent one is about 2-fold less potent than amphotericin B against intracellular amastigotes in THP1 cells. The 2 most promising compounds (FeL7Cl₂ and MnL7Cl₂), with strong activity against both promastigotes and amastigotes and no observable toxicity against the THP1 cells are the Fe²⁺- and Mn²⁺-complexes of a dibenzyl cyclen derivative. Only 2 of the 44 compounds showed observable cytotoxicity against THP1 cells. Tetraazamacrocyclic monocyclic polyamines represent a new class of antileishmanial lead structures that warrant follow up studies.

Full text of DOI:10.1016/j.poly.2019.02.027

Accepted Manuscript
Tetraazamacrocyclic derivatives and their metal complexes as antileishmanial
leads
Timothy J. Hubin, Ashlie N. Walker, Dustin J. Davilla, TaRynn N. Carder
Freeman, Brittany M. Epley, Travis R. Hasley, Prince N.A. Amoyaw, Surendra
Jain, Stephen J. Archibald, Timothy J. Prior, Jeanette A. Krause, Allen G. Oliver,
Babu L. Tekwani, M. Omar F. Khan
PII:
S0277-5387(19)30128-7
https://doi.org/10.1016/j.poly.2019.02.027
POLY 13771
DOI:
Reference:
Polyhedron
To appear in:
Received Date:
Revised Date:
Accepted Date:
25 October 2018
2 February 2019
14 February 2019
Please cite this article as: T.J. Hubin, A.N. Walker, D.J. Davilla, T.N. Carder Freeman, B.M. Epley, T.R. Hasley,
P.N.A. Amoyaw, S. Jain, S.J. Archibald, T.J. Prior, J.A. Krause, A.G. Oliver, B.L. Tekwani, M.O.F. Khan,
Tetraazamacrocyclic derivatives and their metal complexes as antileishmanial leads, Polyhedron (2019), doi: https://
doi.org/10.1016/j.poly.2019.02.027
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Tetraazamacrocyclic derivatives and their metal complexes as antileishmanial leads
Timothy J. Hubin,1* Ashlie N. Walker,1 Dustin J. Davilla,1 TaRynn N. Carder Freeman,1 Brittany
M. Epley,1 Travis R. Hasley,1 Prince N. A. Amoyaw,2 Surendra Jain,3 Stephen J. Archibald,4
Timothy J. Prior,4 Jeanette A. Krause,5 Allen G. Oliver,6 Babu L. Tekwani,3,7 M. Omar F.
Khan2,8
1Department of Chemistry and Physics, Southwestern Oklahoma State University. 100 Campus
Drive, Weatherford, OK 73096
2Department of Pharmaceutical Sciences, College of Pharmacy, Southwestern Oklahoma State
University. 100 Campus Drive, Weatherford, Ok 73096
3National Center for Natural Products Research and Department of BioMolecular Sciences,
School of Pharmacy, University of Mississippi, University, MS 38677
4Department of Chemistry, University of Hull, Cottingham Road, Hull, HU6 7RX
5Department of Chemistry, University of Cincinnati, 301 Clifton Ct., Cincinnati, OH 45221-0172
6Department of Chemistry and Biochemistry, University of Notre Dame, Notre Dame, IN 46556
7(Present address) Southern Research, Division of Drug Discovery, 2000 9th Avenue South
Birmingham, AL 35205
8(Present address) University of Charleston School of Pharmacy, 2300 MacCorkle Ave SE
Charleston, WV 25304
* Corresponding author. TJH: Tel.: +1 580 774 3026; fax: +1 580 774 3115; E-mail address:
tim.hubin@swosu.edu
1

ABSTRACT
A total of 44 bis-aryl-monocyclic polyamines, monoaryl-monocyclic polyamines and
their transition metal complexes were prepared, chemically characterized, and screened in vitro
against the Leishmania donovani promastigotes, axenic amastigotes and intracellular amastigotes
in THP1 cells. The IC50 and/or IC90 values showed that 10 compounds were similarly active at
about 2-fold less potent than known drug pentamidine against promastigotes. The most potent
compound had an IC50 of 2.82 µM (compared to 2.93 µM for pentamidine). Nine compounds
were 1.1-13.6-fold more potent than pentamidine against axenic amastigotes, the most potent one
being about 2-fold less potent than amphotericin B. Fourteen compounds were about 2-10 fold
more potent than pentamidine, the most potent one is about 2-fold less potent than amphotericin
B against intracellular amastigotes in THP1 cells. The 2 most promising compounds (FeL7Cl2
and MnL7Cl2), with strong activity against both promastigotes and amastigotes and no
observable toxicity against the THP1 cells are the Fe2+- and Mn2+- complexes of a dibenzyl
cyclen derivative. Only 2 of the 44 compounds showed observable cytotoxicity against THP1
cells. Tetraazamacrocyclic monocyclic polyamines represent a new class of antileishmanial lead
structures that warrant follow up studies.
Key Words: Cyclen, cyclam, polyamine, antileishmanial drug lead, metal complexes
2

1.
Introduction
Leishmaniasis, a vector born parasitic disease, affects 350 million people worldwide.
Visceral leishmaniasis/kala azar (VL) is the most severe form resulting in 20,000 deaths every
year. The other forms of leishmaniasis are cutaneous leishmaniasis (CL), mucocutaneous
leishmaniasis (MCL), and post kala azar dermal leishmaniasis (PKDL).1 One to two million new
cases of the cutaneous form and half a million new cases of the visceral form occur each year. It
is endemic in 98 countries, closely associated with poverty, with new cases mostly appearing in
India, Bangladesh, Nepal, Brazil, Kenya, Sudan, and Ethiopia.1,2 Multiple different
morphological forms arise in the life cycles of Leishmania, differing mainly by the position,
length and the cell body attachment of the flagellum.3 Amastigotes are a typical morphology
during an intracellular lifecycle stage in a human host. The promastigote form is a common
morphology in the insect host. The extracellular form of Leishmania promastigotes is transmitted
to humans through phlebotomine sandflies. The intracellular form of promastigotes replicates in
the human host macrophages through a complicated life cycle.3
The currently available drugs for treatment of leishmaniasis are severely toxic, costly, or
not effective due to increased resistance. The current first-line drug for treatment of visceral
leishmaniasis is pentavalent antimonial injections. The injections are inconvenient for therapies,
lasting up to 21 days, since no oral forms of pentavalent antimonials are currently available. The
use of pentavalent antimonials as first-line drugs for the past 70 years has led to resistance in
many areas worldwide.4-6 The second-line drug for therapy is amphotericin B, which requires a
slow IV infusion and has severe side effects that can potentially be lethal.5,6 Recent clinical trials
have shown successful treatment of VL with a single dose of liposomal amphotericin B and
3

combination therapies namely, amphotericB+ miltefosine, amphotericine B+paromomycin,
miltefosine+ paromomycin and pentavalent antimonials+ paromomycin.7
The ability to metabolize polyamines is critical for L. donovani replication and survival.
The enzymes associated with the pathway of polyamine biosynthesis and transport are promising
targets for new antileishmanial drug discovery.8-11 The critical enzymes in the metabolic pathway
are ornithine decarboxylase (ODC), S-adenosylmethionine decarboxylase (AdoMetDC),
spermidine synthase (SpdS), trypanothione synthetase (TryS), trypanothione reductase (TR)
tryparedoxin peroxidase (TXNPx), deoxyhypusine synthase (DHS) and deoxyhypusine
hydroxylase (DOHH).9 Targeting these enzymes can potentially obstruct L. donovani replication.
The inhibition of growth of L. donovani and L. infantum promastigotes in culture by
difluoromethyl ornithine (DFMO),12,13 an ODC inhibitor, was the first indication that polyamine
biosynthesis is a potential drug target for leishmaniasis. Sing et al14 has shown that the
polyamine depletion due to the inhibition of ODC by 3-aminooxy-1-aminopropane (Fig. 1) is
leishmanicidal. Leishmania parasites synthesize the polyamines putrescine and spermidine
through the catalysis by ODC, AdoMetDC, and SpdS. Spermidine is utilized as a substrate to
synthesize trypanothione for redox control. Trypanothione synthetase (TryS) and trypanothione
reductase (TR) are two enzymes involved. Ullman and colleagues15-17 have genetically validated
all these enzymes as drug targets for L. donovani.
Several polyamine and oligoamine derivatives (Fig. 1) have been reported to have potent
antileishmanial activity through the interference of polyamine biosynthetic enzymes or transport
system. 18,19 CGP40215A and MDL73811 are potent inhibitors of AdoMetDC, which have been
shown to inhibit growth of L. donovani promastigotes at micromolar concentration.16,20
CGP40215A also inhibits growth of L. donovani amastigotes in mouse macrophages at
4

concentrations up to 90 μM [59].21 The polyamine analogue MDL27695 was shown to eliminate
77-100% of L. donovani amastigotes from mouse macrophages at a 1 μM concentration, possibly
by interfering with DNA and RNA synthesis or inhibiting polyamine transport.18,22 The
oligoamines CGC-11211 and CGC-11226, and the macrocyclic polyamine analogue CGC-11235
showed potent antileismanial activity in vitro (IC50 < 1μM), predominantly by interfering with
the polyamine biosynthetic enzymes.19 The polyamine-dependent redox metabolism including
trypanothione as well as the polyamine transport mechanism have also been shown to be
promising drug targets.9,10,23-26
Figure 1. Representative structures of polyamine and oligoamine derivatives reported to have
potent antileishmanial activity.18,19
In recent years there has been an increasing interest in the application of polyamines
analogs and their transition metal complexes in medicine, including against malaria and
leishmaniasis.27-33 Several reports have shown that incorporation of transition metal ions into
organic pharmacophores offer molecular diversity in drug discovery in addition to enhancement
of the biological activity.34-36 The synthetic tetraazamacrocyclic compounds, particularly, cyclen
5

and cyclam, and their analogs or their metal complexes have been extensively utilized in
applications of a variety of diagnostic and magnetic resonance imaging (MRI) contrast agents.37
Our group has extensively investigated tetraazamacrocycles and their metal complexes in drug
discovery against variety of disease states including malaria, HIV, cancer, and schistosomiasis.38-
51
The facts that polyamine metabolism and related enzymes are established targets, and several
oligoamine and polyamine derivatives have shown potent antileishmanial activity, triggered us to
screen a set of tetraazamacrocyclic derivatives (Fig. 2) and their metal complexes against L.
donovani, the causative agent for visceral leishmaniasis as an extension of our antiparasitic drug
discovery initiatives.
Figure 2. Structures of the tetraazamacrocyclic ligands used in this study.
6

2.
2.1. Chemistry
All the ligands L1–L10 in Figure 2 contain only one tetraazamacrocycle and different
Results and Discussion
combinations of aromatic units for screening to determine which combination of groups
maximizes antileishmanial efficacy. Ligands L1 and L2 contain one aromatic group, whereas L3
– L10 contain two aromatic, or heteroaromatic groups. The tetraazamacrocyclic parent of L1,
L3-L7 is cyclen, whereas for L2, and L8-L10 it is cyclam.
2.1.1. Design Principles
The tetraazamacrocyclic ligands forming the foundation of this study have all been
published over the past 20+ years (see experimental section for specific references). We selected
a few structural types involving varying substitution patterns, including one, two, or three
aromatic groups as a way to probe the importance of hydrophobicity to antileishmanial activity.
We also chose to vary the bridging superstructure included on the macrocycle, from more
flexible un-bridged, to short and rigid ethylene cross-bridged, to longer and less constrained
propylene cross-bridged. The presence of the cross-bridge rigidifies the resulting transition
metal complex, more fully engulfs the hydrophilic metal ion, and increases the kinetic stability of
the metal complex compared to unbridged analogues.52 If redox processes involving the metal
ion of these complexes is involved in the antileishmanial activity, such stabilization would be
important so that the potential drug molecule does not decompose upon oxidation/reduction.
Finally, we synthesized and screened complexes of the selected ligands with six different
transition metal ions namely, Mn2+, Fe2+, Co2+, Ni2+, Cu2+ and Zn2+. The biological significance
7

of all of these metal ions are well-known. They are typically found in a variety of organisms,
including humans, and are not expected to be toxic, particularly if bound strongly to a
macrocyclic ligand. By producing metal complexes, we hoped to change the three-dimensional
shape, hydrophobicity, and redox reactivity of the resulting potential drug leads. Choosing a
range of metal ions allowed us to identify the metal complexes with the most potent
antileishmanial activity. Establishing the properties of the most active metal ion complexes
might help indicate the potential mechanism of action of the lead compounds against leishmania
For example, Fe/Mn are known to have rich redox chemistries, while Zn is not.
2.1.2. Synthesis of ligands
A representative synthetic scheme for the benzyl substituted cyclam ligands is shown in
Scheme 1. All of these ligands are previously published. Synthesis starting from cyclam
proceeds through bis-aminals to give regioselective benzylation at non-adjacent nitrogens.
Removal of the methylene aminal carbons yields L10,53 which can be bridged by ditosyl propane
to give L9.54 Reductive ring cleavage by sodium borohydride of the dialkylated cyclam-glyoxal
results in L249 and L8.55 Analogous chemistry starting from cyclen results in L1,51 L3,55 L4,56
and L7.57
8

Scheme 1. Synthetic scheme for the literature syntheses of cyclam ligands L2,49 L8,55 L9,54 and
L10.53 Cyclen analogues L1,51 L3,55 L4,56 and L757 are made by analogous reactions.
The cyclen bisquinoline ligands L5 and L6 were prepared according to the synthetic
route shown in Scheme 2.39,47 The initial synthesis by direct derivatization using inorganic
K2CO3 in DMF 39 was inefficient and the desired product (L5) was isolated in low yield (~35%).
By choosing N-methylpyrrolidinone as the reaction solvent, with triethylamine as base, and
elevating the reaction temperature, L5 and L6 were accessed in fair yields of up to 45%.47
Scheme 2. Synthetic scheme for the literature syntheses of cyclen bisquinoline ligands L5 39,47
and L6.47
9

2.1.3. Synthesis of transition metal complexes
Scheme 3. Generic complexation reaction illustrated using L2 and showing an acetate complex.
Syntheses of Mn2+, Fe2+, Co2+, Ni2+, Cu2+ and Zn2+ complexes were attempted with L1-
L4 and L7-L10 using anhydrous MCl2 and/or M(OAc)2 salts in anhydrous solvents such as
MeOH, MeCN, and/or DMF (see Scheme 3). The products of complexations resulting in pure
complexes are given in Table 1 and specific reaction conditions and basic characterization data
are given in the Experimental Section. In general, unbridged tetraazamacrocycles (like L7 and
L10) can be complexed to metal ions in protic solvents like methanol without the ligand
preferentially protonating from the solvent. However, in most cases, complexation to ethylene-
and propylene-cross-bridged ligands (L1-L4, L8-L9) is most successful in aprotic solvents like
acetonitrile or dimethylformamide where the highly basic proton sponge ligands are not
protonated by solvent, which impedes metal ion complexation.52 Complexations were generally
carried out in an inert atmosphere glovebox to protect the metal ions from air oxidation and the
ligands from protonation by water.53 Not all complexations were successful, but a significant
variety of complexes with different metal ion and ligand type were obtained to survey for
antileishmanial activity among these tetraazamacroyclic ligands and complexes. Future work
will include completing the full range of metal complexes for each ligand by finding successful
reaction conditions for each metal-ligand pair. The reactions of MnCl2 or FeCl2 with ligand L5 in
10

a 1:1 ratio at 80oC for 1 week using DMF as a solvent in a glove box gave the complexes
Mn(L5)Cl2 and Fe(L5)(Cl2, respectively in 45% and 29% yields, respectively.50
2.1.4. X-ray Crystallographic Structural Studies
In order to better understand the three dimensional structures of these potential drug
molecules, and to correlate those structures with the biological activity observed, we have
obtained a number of X-ray crystal structures. Not all ligands or metal complexes gave X-ray
quality crystals, but we have obtained crystal structures of the following ligands, either alone or
in a metal complex: L1, L3, L7, L8, and L10. We have also previously published additional X-
ray crystal structures of metal complexes of L2,49 L3,58 and L848,51 which will be discussed.
Tables containing data collection and refinement and metrical parameters for each structure are
found in the electronic supplementary information.
[Cu(L1)Cl]PF6 was prepared and crystallized (Fig. 3) as the acetate complex did not
crystallize. In [Cu(L1)Cl]PF6, the copper(II) center adopts a distorted square based pyramidal
geometry. (Fig. 3a). The tetraazamacrocycle is forced into a folded (cis) arrangement by the
ethyl bridge, and the ring nitrogens occupy three of the equatorial sites and the axial site, the
typical coordination geometry for Cu2+ complexes of cross-bridged tetraazamacrocycles.59,60,49,51-
52
The single benzyl group is folded into the open coordination site of the five-coordinate Cu2+,
which is likely due to packing forces in the solid state, since no close interaction of the benzyl
group with the metal ion is observed.60 The previously published49 structure of [Cu(L2)(OAc)]+
(Fig. 3b) has an acetate bound to the Cu2+ center, and this bulky group causes the benzyl arm to
be rotated away from the metal ion. This demonstrated flexibility may play a role in the
antileishmanial activity discussed below.
11

a)
b)
Figure 3. The X-ray crystal structures of a) [Cu(L1)Cl]+ (top) and [Cu(L2)(OAc)]+49
demonstrate the flexibility of the benzyl group, which can fold toward an empty coordination site
as in [Cu(L1)Cl]+ or away from a more sterically crowded metal ion as in [Cu(L2)(OAc)]+.
[Mn(L3)Cl2] (Fig. 4a) has previously been published,58 but none of the other L3
complexes screened here crystallized. However, the acetate precursor to [Fe(L3)(OAc)]PF6,
[Fe(L3)(OAc)2], did form X-ray quality crystals (Fig. 4b).
12

a)
b)
Figure 4. The X-ray crystal structures of a) [Mn(L3)Cl2] (left)58 and b) [Fe(L3)(OAc)2] (right).
13

Both structures have six-coordinate distorted octahedral coordination geometries around
M2+ ions with folded, or cis, configurations of L3 bonded by all four nitrogen atoms.
[Mn(L3)Cl2] has both benzyl groups rotated away from the Mn2+ ion and nearly coplanar (24.36o
of their planes). However, in [Fe(L3)(OAc)2], the benzyl group to rotate approximately
perpendicular to the planes of the acetate carboxyl groups in order to fit between them in the
solid state (65.26o of their planes). It remains to be determined if this flexibility plays a role in
antileishmanial activity, or if the simple presence of two hydrophobic benzyl groups, regardless
of flexibility, is all that is required for increased activity (vide infra).
No crystal structures for L4-L6 or L9 are known, according to the Cambridge Database
(version 5.37) or grown by us. We include Cu(CB3-DO2A)+ in figure (Fig. 5) 61 to demonstrate
the similarity to ethylene cross-bridged complexes (L1, L3 above) except for the 3-carbon cross-
bridge, which makes the ligand cavity slightly larger and better able to encapsulate the metal ion
(Cu2+ in this case). Obviously L4’s benzyl groups will not be coordinating, and we have
demonstrated above how they would likely be arranged in complexes of L4. As dibenzyl
compounds, some of the L4 compounds demonstrate similar antileishmanial properties as the L3
compounds (vide infra).
a)
14

b)
Figure 5. a) The X-ray crystal structure61 of Cu(CB3-DO2A)+ and b) the CB3-DO2A ligand
Neutral ligand L8 itself, rather than a metal complex, crystallized from organic solvent.
In aqueous solution, cross-bridged cyclams are proton sponges and are typically diprotonated.62
In L8, the cross-bridge extends in front of the page (Fig. 6), showing the cavity for metal
binding. Similarly to the [Fe(L3)(OAc)2] complex, the benzyl rings are significantly non-planar
(plane-plane angle of 54.00o). As will be discussed below, certain metal complexes of these
ligands are typically more active against Leishmania than the free ligands themselves.
Figure 6. The X-ray crystal structure of L8.
The final two X-ray crystal structures to be discussed are Ni(L7)(OAc)+, which
crystallized when the screened complex Ni(L7)Cl2 did not, and Cu(L10)(DMF)22+, which
crystallized when the screened complex [Cu(L10)(OAc)]PF6 did not (see Fig. 7).
15

a)
b)
2+
Figure 7. The X-ray crystal structures of a) Ni(L7)(OAc)+ (left) and b) Cu(L10)(DMF)2
These two crystal structure include, unlike any of those above, unbridged
tetraazamacrocycles, and demonstrate the configurational flexibility of these ligands by
.
producing both a folded (or cis) geometry (Ni(L7)(OAc)+ (Fig. 7a) and a flat (or trans) geometry
(Cu(L10)(DMF)22+ (Fig. 7b)). A design principle (vide supra) used in producing this overall set
of compounds for antileishmanial screening was to determine if the rigid, cross-bridged ligand
16

complexes would be more or less active than the more flexible unbridged analogues. However,
these two crystal structures clearly demonstrate the flexibility of the unbridged ligands. Not only
are both cis and trans structures achieved, but the benzyl groups show even more rotational
flexibility than seen in the cross-bridged structures discussed above. The plane-plane angle in
Ni(L7)(OAc)+ is 80.41o, the closest to perpendicular we have yet seen. While this plane-plane
angle in Cu(L10)(DMF)22+ is 0.00o, making them coplanar (this is also a required outcome of the
crystallographic space group, P 21/n, where one half of the molecule is the asymmetric unit and
generates the other. Both ligand L7 and L10 complexes are, in general, more active against
leishmania than the cross-bridged ligands presented above. In fact, the Mn2+ and Fe2+ complexes
of L7 are identified as the most likely potential drug molecules identified in this study. Perhaps
this is a result of the greater flexibility discussed above—a possibility that warrants further
investigation.
2.2. Biological Evaluations of In vitro antileishmanial activity
2.2.1. Antileishmanial activity against promastigotes
Ten compounds resulting from L1-10 were significantly active to about 2-fold less potent
than pentamidine against promastigotes. The IC50 values ranged from 2.82 to >20 µM compared
to 2.93 µM for pentamidine (Table 1). The diaryl-monocyclic polyamine derivatives were more
active than the monoaryl-monocyclic derivatives. For example, all of the monoaryl compounds’
IC50’s were >20 µM and were not further tested, while most of the diaryl compounds and/or their
metal complexes showed IC50 values less than 20 µM. The most potent is a Mn2+-complex of
bisbenzyl cyclen derivative (MnL7Cl2), which has an IC50 of 2.82 µM compared to 2.93 µM for
pentamidine (Table 1). An Fe2+-complex of cyclen bisquinoline (FeL5Cl2) was found to be most
potent by IC90 values, with IC90 of 7.07 µM compared to 6.52 µM for pentamidine. However,
17

this analog also showed IC50 and IC90 values of 3.95 and 8.79 µM, respectively against THP1
cells suggesting potential cytotoxicity. No simple conclusion regarding which metal ion (Mn2+,
Fe2+, Co2+, Ni2+, Zn2+ and Cu2+) best enhanced the antileishmanial activity over the parent
ligands. However, it was generally observed that Mn2+, Fe2+, and Zn2+-complexes were more
potent than the corresponding free ligands, with Mn2+-complexes typically being the most potent.
The three most potent ligands are L3, L9 and L10 with IC50 values 12.68, 12.98 and 9.40
µM; and IC90 values of 11.44, 18.97, and 17.84 µM, respectively (Table 1). Ligands L3 is a
dibenzyl ethyl cross-bridged cyclen, L9 and L10 are dibenzyl derivatives of cyclam (Fig. 2).
The three most potent compounds against promastigotes are Mn(L7)Cl2, Mn(L9)MnCl4 and
Fe(L5)Cl2, among which only Mn(L7)Cl2 has no toxicity against the THP1 cells, suggesting
Mn(L7)Cl2 as a lead for further study.
Table 1
The in vitro Inhibitory Activity of the Monocyclic Compounds against L. donovani.
Sample Code
L. donavani
L. donavani
L. donavani
THP1 cells
Promastigotes Amastigotes Amastigotes
+ THP1
(Cytotoxicity)
IC50
IC90
IC50
IC90
IC50
IC90
IC50
IC90
(µM) (µM) (µM) (µM) (µM) (µM) (µM) (µM)
Pentamidine
2.93
6.52
20.42 >25
29.38 9.14
>20
>10
>20
>25
>10
>20
Amphotericine B
0.134 0.262 0.790 0.959 0.48
1.94
>20
Bn1Me1Bcyclen (L1)
>20
>20
>20
>20
>20
18

[Cu(L1)(OAc)]PF6
[Zn(L1)(OAc)]PF6
Bn1Me1Bcyclam (L2)
[Ni(L2)(OAc)]PF6 . 0.35NH4PF6
[Cu(L2)(OAc)]PF6
[Zn(L2)OAc)]PF6
Bn2Bcyclen (L3)
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>25
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>25
>20
>20
>20
>20
>20
>25
>10
>10
>10
>10
>20
>20
>20
>20
>20
>20
>20
>25
>20
>20
>20
>20
>20
>25
>10
>10
>10
>10
>20
12.68 11.44 15.14 22.29 >25
Mn(L3)Cl2
18.93 >20
11.50 17.09 >20
[Fe(L3)(OAc)]PF6
[Ni(L3)(OAc)]PF6
[Cu(L3)(OAc)]PF6
[Zn(L3)(OAc)]PF6
Bn2PBCyclen (L4)
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
>20
10.95 >20
>20
>20
>20
4.73
13.64 >20
15.61 >20
>25
>10
>10
>10
>10
>20
>25
>10
>10
>10
>10
>20
18.24 >25
[Mn(L4)(C2H3O2)]PF6 . 1.9NH4PF6 >10
[Fe(L4)(C2H3O2)]PF6 . 1.2NH4PF6 >20
>10
>20
>10
>10
>10
>10
>20
>10
>10
>10
>10
>20
[Co(L4)(C2H3O2)]PF6 . DMF
[Zn(L4)(C2H3O2)]PF6 . 0.2DMF
CNBQ (L5)
20.66 >20
22.92 >20
>20
>20
19

Mn(L5)Cl2
7.40
4.13
>20
11.46 >15
>15
>10
>20
>25
>20
>15
>20
>20
>15
>15
>15
>15
>15
8.59
3.52
>20
-
10.99 13.20 >15
Fe(L5)Cl2
7.07
>20
>10
>20
>25
>20
5.24
>20
2.8
3.95
>20
>25
>20
>15
>20
8.79
>20
>25
>20
>15
>20
>20
Bis-BQ-Cyclen (L6)
Bn2Cyclen (L7)
10.47 >20
Mn(L7)Cl2
2.82
5.50
9.12
1.02
1.89
>20
2.40
6.56
>20
Fe(L7)Cl2
13.09 >15
Ni(L7)Cl2 . 3H2O
Bn2Bcyclam (L8)
Mn(L8)Cl2
18.63 >20
20.81 >20
21.99 >20
>20
>20
>15
10.06 10.43 >20
>15
>15
>15
>15
>15
>15
>15
>15
>15
>15
14.27 >15
Fe(L8)Cl2
7.80
20.09 >15
>15
>15
>15
>15
>15
>15
>15
>15
[Co(L8)(OAc)]PF6 . H2O
[Ni(L8)(OAc)]PF6
[Zn(L8)(OAc)]PF6
Bn2PBCyclam (L9)
[Mn(L9)][MnCl4] . CH3CN
[Fe(L9)][FeCl4] . H2O
[Co(L9)(OAc)1.1](PF6)0.9 . DMF
[Ni(L9)(OAc)1.1](PF6)0.9 . DMF
35.29 >20
>15
>15
>15
6.51
4.73
>20
>20
12.98 18.97 18.54 23.11 11.10 11.43 20.63 22.80
3.48
9.11
1.50
2.68
1.89
2.17
3.49
>14
>13
>13
5.90
>14
>13
>13
11.69 >20
10.14 >14
11.93 >14
7.98
9.33
>13
>13
10.45 13.09 >13
10.12 12.96 >13
>13
>13
20

[Zn(L9)(OAc)]PF6.0.3NH4PF6
Bn2Cyclam (L10)
8.17
9.40
6.62
5.64
>15
>20
5.65
10.50 9.91
17.84 >25
12.36 >19
13.91 >19
13.48 7.81
8.53
>14
>14
>25
>19
>19
>15
>15
>15
>25
>19
>15
>15
>15
>15
12.93 13.19 >25
12.36 12.68 >19
Mn(L10)Cl2 . H2O
Fe(L10)Cl2 . 0.5 H2O
9.61
>15
>15
6.80
9.68
>15
>15
7.04
>19
>15
>15
>15
[Co(L10)(OAc)](PF6) . H2O
Cu(L10)(OAc)](PF6) . H2O
[Zn(L10)(OAc)]PF6 . 0.1H2O
>15
>20
>15
>15
11.28 >15
2.2.2. Antileishmanial activity against axenic amastigotes
L. donovani promastigotes in culture have been transformed into potential amastigotes
forms, which normally grow intracellularly in the host macrophages. These culture-adapted
forms are referred as axenic amastigotes and have been suggested as a useful model for in vitro
antileishmanial screening and also for investigating mechanisms regulating the differentiation,
survival and pathogenicity of the leishmania parasite.63,64 Against axenic amastigotes, 9
compounds were 1.1-13.6-fold more potent than pentamidine, the most potent (MnL9MnCl4)
being about 2-fold less potent than amphotericin B. This compound is also highly potent against
promastigotes as well as intracellular amastigotes in THP1 cells. However, it is cytotoxic to the
THP1 cells as mentioned above. Other compounds with reasonable activity against the axenic
amastigotes are L9 and all of its metal complexes, among which L9 is the only other compound
with comparable IC50 and IC90 values against THP1 cells as MnL9MnCl4, again suggesting
21

potential toxicity. Against intracellular amastigotes in THP1 cells, 14 compounds were about 2
to 10-fold more potent than pentamidine, the most potent (MnL7Cl2) is about 2-fold less potent
than amphotericin B. This same compound was also found to be the most potent against
promastigotes without sign of toxicity. The second most potent compound against the
intracellular amastigotes in THP1 cells is Fe(L7)Cl2, without sign of toxicity against the THP1
cells. This compound was also found to have good activity against promastigotes suggesting
another potential lead for further study.
2.2.3. Summary of Biological Activity
In general, the diaryl compounds and their metal complexes (specifically Zn2+, Fe2+, or
Mn2+ complexes) were found to be more potent compared to the monoaryl system possibly due
to the higher lipophilicity and thus cell penetration. Two compounds with strong activity against
both promastigotes and amastigotes and no observable toxicity against the THP1 cells are the
Fe2+- and Mn2+- complexes of L7, a dibenzyl cyclen derivative (Fig. 2 and Table 1), which
warrant further investigation.
The present study revealed several important findings. First, the diaryl monocyclic
systems are more potent than monoaryl derivatives, perhaps due to more lipophilicity as well.
Second, Fe2+, and Mn2+ complexes increase the activity of both nonpolar and polar macrocyclic
derivatives against both promastigotes and amastigotes of leishmanial parasites, although other
metals also have inconsistently shown improvement in activity. Third, the polyamine ring
systems are likely to be key structural targets for antileishmanial activity, as all the compounds
showed reasonable growth inhibition of the leishmania parasites.
22

3. Conclusions
The results indicate that excellent antileishmanial potency can be obtained by transition
metal complexation to tetraazamacrocycles. Interestingly, all of the most active compounds were
transition metal complexes, not traditional “organic” compounds. The analogs with hydrophobic
aryl rings showed higher antileishmanial potency, possibly due to improved cell permeability of
these analogs. The improvement in the inhibitory potency beyond that of pentamidine required
the complexation of Fe2+, or Mn2+ and may indicate the role of polyamine biosynthesis and/or
oxygen radicals formed by the redox cycling of ionic iron and manganese. It has previously been
shown that several polyamines and oligoamines are potent growth inhibitors of leishmanial
parasites through inhibition of ODC, AdoMetDC, or polyamine transport system.15,16 This study
suggests the possible involvement of these compounds with interference of any of these enzymes
in the polyamine biosynthetic pathway of leishmania parasites. Future detailed mechanistic and
stability studies involving these ligands, and their metal complexes will be designed to unveil the
role of complexed metal ions, especially iron, and manganese in antileishmanial activity. This
study has laid a solid foundation towards the development of a new antileishmanial
chemotherapy based on tetraazamacrocyclic derivatives and their metal complexes.
4. Experimental Section
4.1. Chemistry
4.1.1. General
All the materials were reagent grade, and used as supplied. The reactions were performed in
anhydrous solvents under inert atmosphere unless otherwise indicated. Anhydrous solvents
(acetonitrile and DMF) as well as all other reagents were used as received from a commercial
23

source. Elemental analysis was carried out by Quantitative Technologies, Inc. in Whitehouse,
NJ. Electrospray MS was obtained using a Shimadzu LCMS-2020 in 1:1 methanol/ water
mixture. 1H and 13C NMR spectra were recorded at 300 MHz and 75 MHz, respectively on
Bruker 300 spectrometer with TMS as internal standard.
4.1.2. Synthesis of ligands and their Metal Complexes
Several ligands and their metal complexes were synthesized according to literature
procedures: L1, [Cu(L1)(OAc)]PF6, [Zn(L3)(OAc)]PF6, [Zn(L8)(OAc)]PF6, [Zn(L2)(OAc)]PF6,
[Cu(L3)(OAc)]PF6, and [Zn(L1)(OAc)]PF6;51 L2 and [Cu(L2)(OAc)]PF6;49 L3, L8,55 and L4;56
L539,47 Fe(L5)Cl2 and Mn(L5)Cl2;50 L6 39,47, L9;54 L10;53 L7 and;57 [Fe(L3)Cl2], [Mn(L3)Cl2],
[Mn(L8)Cl2], and [Fe(L8)Cl2]48; [Ni(8)(OAc)]PF6.48 Others were synthesized according to the
following methods.
[Ni(L2)(OAc)]PF6 • 0.35NH4PF6: (1.00 mmol, 0.331 g) of L2 and 0.177 g (1.00 mmol)
of anhydrous nickel(II) acetate were added to 25 ml of dry DMF in an inert atmosphere
glovebox. The reaction was stirred at room temperature for 18 h. The crude
[Ni(L2)(OAc)][(OAc)] solution was removed from the glovebox, filtered to remove any trace
solids, and evaporated to dryness. The crude product was dissolved in 10 ml of dry methanol, to
which was added dropwise a 5 ml dry methanol solution of 5 equivalents (0.815 g, 5.00 mmol)
of NH4PF6. A pale purple powder precipitated, was collected, washed with cold methanol and
ether, and dried under vacuum. Yield: 0.356 g (60%). Elemental analysis(%) calcd.
[Ni(C20H34N4)(C2H3O2)]PF6 • 0.35NH4PF6 (650.266 g/mol): C 40.64, H 5.95, N 9.37; Found C
40.46, H 5.70, N 9.59. MS (ES) m/z 447 [Ni(L2)(OAc)]+.
24

[Ni(L3)(OAc)]PF6: (1.00 mmol, 0.379 g) of L3 and 0.177 g (1.00 mmol) of anhydrous
nickel(II) acetate were added to 25 ml of dry DMF in an inert atmosphere glovebox. The
reaction was stirred at room temperature for 18 h. The crude [Ni(L3)(OAc)][(OAc)] solution
was removed from the glovebox, filtered to remove any trace solids, and evaporated to dryness.
The crude product was dissolved in 10 ml of dry methanol, to which was added dropwise a 5 ml
dry methanol solution of 5 equivalents (0.815 g, 5.00 mmols) of NH4PF6. A pale purple powder
precipitated, was collected, washed with cold methanol and ether, and dried under vacuum.
Yield: 0.414 g (67%). Elemental analysis (%) calcd. [Ni(C24H34N4)(C2H3O2)]PF6 (641.260
g/mol): C 48.70, H 5.82, N 8.74; Found C 48.58, H 6.00, N 8.79. MS (ES) m/z 495
[NiL(OAc)]+.
[Mn(L4)(OAc)]PF6 • 1.9NH4PF6: (1.00 mmol, 0.392 g) of L4 and 0.173 g (1.00 mmol)
of anhydrous manganese(II) acetate were added to 10 ml of dry DMF in an inert atmosphere
glovebox. The reaction was stirred at room temperature for 7 days, during which a white powder
precipitated. This powder was filtered from solution and washed with minimal DMF and then
diethyl ether before drying overnight open to the glovebox atmosphere. The white solid was
then dissolved in 5 ml MeOH in the glovebox, and 5 equivalents (5.00 mmol, 0.815 g) of
NH4PF6 dissolved in 3 ml MeOH was added, resulting in precipitation of the white powder
product (0.194 g, 20%), which was filtered from solution and dried open to the glovebox
atmosphere. Elemental analysis(%) calcd. [Mn(C25H36N4)(C2H3O2)]PF6 • 1.9NH4PF6 (961.230
g/mol): C 33.74, H 4.89, N 8.60; Found C 33.58, H 5.03, N 8.39. MS (ES) m/z 445 [Mn(L4)]+.
[Fe(L4)(OAc)]PF6 • 1.2NH4PF6: (1.00 mmol, 0.392 g) of L4 and 0.174 g (1.00 mmol) of
anhydrous iron(II) acetate were added to 10 ml of dry DMF in an inert atmosphere glovebox.
The reaction was stirred at room temperature for 7 days, during which a light brown powder
25

precipitated. This powder was filtered from solution and washed with minimal DMF and then
diethyl ether before drying overnight open to the glovebox atmosphere. The light brown solid
was then dissolved in 5 ml MeOH in the glovebox, and 5 equivalents (5.00 mmol, 0.815 g) of
NH4PF6 dissolved in 3 ml MeOH was added, resulting in precipitation of the light brown product
(0.264 g, 31%), which was filtered from solution and dried open to the glovebox atmosphere.
Elemental analysis(%) calcd. [Fe(C25H36N4)(C2H3O2)]PF6 • 1.2NH4PF6 (848.040 g/mol): C
38.24, H 5.21, N 8.59; Found C 38.32, H 5.14, N 8.59. MS (ES) m/z 446 [Fe(L4)]+.
[Co(L4)(OAc)]PF6 • DMF: (1.00 mmol, 0.392 g) of L4 and 0.177 g (1.00 mmol) of
anhydrous cobalt(II) acetate were added to 10 ml of dry DMF in an inert atmosphere glovebox.
The reaction was stirred at room temperature for 7 days, during which the solution became deep
blue. The solution was removed from the glovebox, filtered to remove trace solids which were
discarded, and the DMF was removed under vacuum. The deep purple oily solid remaining was
then dissolved in 5 ml MeOH to give a pink solution, and 5 equivalents (5.00 mmol, 0.815 g) of
NH4PF6 dissolved in 3 ml MeOH was added. This resulted in precipitation of the pink product
(0.324 g, 44%), which was filtered from solution, washed with methanol and diethyl ether, and
dried under vacuum. Elemental analysis(%) calcd. [Co(C25H36N4)(C2H3O2)]PF6 • DMF (728.62
g/mol): C 49.45, H 6.36, N 9.61; Found C 49.77, H 6.12, N 9.25. MS (ES) m/z 452 [Co(L4)]+.
[Zn(L4)(OAc)]PF6 • 0.2DMF • 0.2H2O: (1.00 mmol, 0.392 g) of L4 and 0.183 g (1.00
mmol) of anhydrous zinc(II) acetate were added to 10 ml of dry DMF in an inert atmosphere
glovebox. The reaction was stirred at room temperature for 7 days, during which the solution
became orange. The solution was removed from the glovebox, filtered to remove trace solids
which were discarded, and the DMF was removed under vacuum. The deep yellow solid
remaining was then dissolved in 5 ml MeOH to give a yellow solution, and 5 equivalents (5.00
26

mmol, 0.815 g) of NH4PF6 dissolved in 3 ml MeOH was added. This resulted in precipitation of
the tan product (0.337 g, 49%), which was filtered from solution, washed with methanol and
diethyl ether, and dried under vacuum. Elemental analysis (%) calcd.
[Zn(C25H36N4)(C2H3O2)]PF6 • 0.2DMF • 0.2H2O (683.82 g/mol): C 48.48, H 6.07, N 8.60;
Found C 48.08, H 5.78, N 8.95. MS (ES) m/z 480 [Zn(L4)Na]+.
[Mn(L7)Cl2]: (1.00 mmol, 0.353 g) of L7 and 0.126 g (1.00 mmol) of anhydrous
manganese(II) chloride were added to 10 ml of dry DMF in an inert atmosphere glovebox. The
reaction was stirred at room temperature for 7 days, during which a white powder formed. This
solid was filtered off inside the glovebox, rinsed with diethyl ether, and allowed to dry open to
the glovebox atmosphere to give the white powder product. Yield: 0.211 g (44%). Elemental
analysis (%) calcd. [Mn(C22H32N4)Cl2] (478.36 g/mol): C 55.24, H 6.74, N 11.71; Found C
55.10, H 6.51, N 11.58. MS (ES) m/z 455 [Mn(L7)(MeOH)(OH)]+.
[Fe(L7)Cl2] • 0.4DMF • 0.5 Et2O: (1.00 mmol, 0.353 g) of L7 and 0.127 g (1.00 mmol)
of anhydrous iron(II) chloride were added to 10 ml of dry DMF in an inert atmosphere glovebox.
The reaction was stirred at room temperature for 7 days, during which a tan powder formed.
This solid was filtered off inside the glovebox, rinsed with diethyl ether, and allowed to dry open
to the glovebox atmosphere to give the white powder product. Yield: 0.228 g (42%). Elemental
analysis (%) calcd. [Fe(C22H32N4)Cl2] • 0.4DMF • 0.5 Et2O (545.57 g/mol): C 55.48, H 7.35, N
11.30; Found C 55.96, H 7.23, N 11.76. MS (ES) m/z 437 [Fe(L7)(MeOH)]+; m/z 481
[Fe(L7)Cl2]+.
[Ni(L7)Cl2] • 3H2O: (1.50 mmol, 0.529 g) of L7 and 0.194 g (1.00 mmol) of anhydrous
nickel(II) chloride were added to 15 ml of MeOH and 5 ml DMF. The reaction was heated to
reflux with stirring under nitrogen for 3 days, during which a clear emerald green solution
27

formed. The solution was cooled to room temperature, filtered to remove trace solids which
were discarded, and evaporated to dryness forming a green oil. The oil was dissolved in 10 ml
MeOH, to which 100 ml diethyl ether was added to precipitate out a green powder product. The
green powder product was filtered, washed with diethyl ether, and dried under vacuum. Yield:
0.353 g (44%). Elemental analysis (%) calcd. [Ni(C22H32N4)Cl2] • 3H2O (482.12 g/mol): C
49.28, H 7.14, N 10.45; Found C 49.23, H 7.08, N 10.30. MS (ES) m/z 447 [Ni(L7)Cl]+.
[Co(L8)(OAc)]PF6 • H2O: (1.00 mmol, 0.407 g) of L8 and 0.177 g (1.00 mmol) of
anhydrous cobalt(II) acetate were added to 25 ml of dry DMF in an inert atmosphere glovebox.
The reaction was stirred at room temperature for 18 h. The crude [Co(L7)(OAc)][(OAc)]
solution was removed from the glovebox, filtered to remove any trace solids, and evaporated to
dryness. The crude products was dissolved in 10 ml of dry methanol, to which was added
dropwise a 5 ml dry methanol solution of 5 equivalents (0.815 g, 5.00 mmols) of NH4PF6. A
pale pink powder precipitated, was collected, washed with cold methanol and ether, and dried
under vacuum. Yield: 0.416 g (60%). Elemental analysis (%)
calcd. [Co(C26H38N4)(C2H3O2)]PF6 • H2O (687.572 g/mol): C 48.91, H 6.30, N 8.15; Found C
49.19, H 6.50, N 8.29. MS (ES) m/z 524 [Co(L8)(OAc)]+.
[Mn(L9)][MnCl4] • CH3CN: (0.452 mmol, 0.190 g) of L9 and 0.063 g (0.500 mmol) of
anhydrous manganese(II) chloride were added to 5 ml of dry MeCN in an inert atmosphere
glovebox. The reaction was stirred at room temperature for 7 days, during which a pale tan
solution formed. This solution was filtered to remove trace solids, and evaporated to dryness
under vacuum to give the pale tan powder product. Yield: 0.152 g (85% based on MnCl2).
Elemental analysis (%) calcd. [Mn(C27H40N4)][MnCl4] • CH3CN (713.37 g/mol): C 48.83, H
6.08, N 9.82; Found C 49.65, H 6.47, N 9.47. MS (ES) m/z 511 [Mn(L9)]+.
28

[Fe(L9)][FeCl4] • H2O: (0.452 mmol, 0.190 g) of L9 and 0.064 g (0.500 mmol) of
anhydrous iron(II) chloride were added to 5 ml of dry MeCN in an inert atmosphere glovebox.
The reaction was stirred at room temperature for 7 days, during which a brown solution formed.
This solution was filtered to remove trace solids, and evaporated to dryness under vacuum to
give the brown powder product. Yield: 0.137 g (79% based on FeCl2). Elemental analysis (%)
calcd. [Fe(C27H40N4)][FeCl4] • H2O (692.15 g/mol): C 46.85, H 6.11, N 8.10; Found C 46.97, H
5.71, N 7.85. MS (ES) m/z 512 [Fe(L9)]+.
[Co(L9)(OAc)1.1](PF6)0.9 • DMF: (0.500 mmol, 0.211 g) of L9 and 0.089 g (0.500 mmol)
of anhydrous cobalt(II) acetate were added to 10 ml of dry DMF in an inert atmosphere glovebox
and stirred overnight, forming a bright blue solution. The solution was removed from the
glovebox, filtered to remove trace solids which were discarded, and evaporated to dryness
forming a bright blue oil. The oil was dissolved in 7 ml dry MeOH, to which a 3 ml MeOH
solution of 5 equivalents (2.50 mmol, 0.408 g) of NH4PF6 was added with stirring, and then
stored overnight at -10 oC. The pink powder product was filtered, washed with methanol and
diethyl ether, and dried under vacuum. Yield: 0.183 g (49%).
[Co(C27H40N4)(C2H3O2)1.1](PF6)0.9 • DMF (748.08 g/mol): C 51.70, H 6.78, N 9.36; Found C
51.61, H 7.09, N 9.37. MS (ES) m/z 557 [Co(L9)(OAc)(H2O)]+.
[Ni(L9)(OAc)1.1](PF6)0.9 • DMF: (0.500 mmol, 0.211 g) of L9 and 0.089 g (0.500 mmol)
of anhydrous nickel(II) acetate were added to 10 ml of dry DMF in an inert atmosphere glovebox
and stirred overnight, forming a brown solution. The solution was removed from the glovebox,
filtered to remove trace solids which were discarded, and evaporated to dryness forming a brown
oil. The oil was dissolved in 5 ml dry MeOH, to which a 2 ml MeOH solution of 5 equivalents
(2.50 mmol, 0.408 g) of NH4PF6 was added with stirring, and then stored overnight at -10 oC.
29