**3. Case study: Evaluation of the chemotherapeutic potential of metal complexes containing nucleobase-analogues against** *Leishmania infantum* **and** *Leishmania braziliensis*

On the other hand, use of metal-based nanosystems as drug carriers, e.g. noble metal nano‐ particles, might provide additional advantages such as the possibility for diagnosis by imaging techniques and the combined effect of producing ROS as it is the case for silver. ROS can induce oxidative stress, DNA damage, alkylation of target proteins and eventually apoptosis of the

In order to inhibit *Leishmania* parasites with a ROS-based treatment, these oxygen derivatives must be produced in a physical way rather than in an enzymatic way that can be blocked by parasites. Metal nanoparticles are able to produce high amounts of ROS, as they are more

Nanocarriers also offer the possibility to specifically target the parasites by attaching appro‐ piate targeting molecules onto their surface. This way side effects to the host would be minimised. In addition, drug delivery vehicles such as nanoparticles allow prompt interactions with biomolecules present within as well as on the surface of the cell and may be tuned into

Recent advancements in molecular biology have identified a few parasite targets that are likely to be very sensitive to metal-based compounds. These targets usually are enzymes, some of them bearing free thiols at their active sites that manifest a high propensity to react with soft Lewis acids, i.e. metal ions such as Ag(I), Au(III) or Zn(II). Therefore these parasite targets will be susceptible to strong and selective inhibition by this kind of metals. This is the case of dithiol reductases like trypanothione reductase (T(SH)2), which have been shown to play a key role in the *Leishmania* metabolism (see Section 2.1.1) and therefore constitute primary targets for metal compounds. Cysteine proteases, such as cathpesin L-like or cathepsin B-like, are another example of proteins with thiol-containing active sites and thus responsive to inhibition by metal compounds. Inhibitors that would effectively target both types of cysteine proteases in *Leishmania*, while maintaining some selectivity versus homologous host enzymes, would be

Regarding DNA interaction, previous studies have shown that DNA-binding metal com‐ pounds such as cisplatin display antiparasitic activity. These findings along with the obser‐ vation that many antiparasitic drugs bind to DNA, have led to propose that in general every

Therefore DNA-intercalating molecules have been used as ligands to form metal complexes showing antiparasitic activity. Intercalating ligands are usually polyaromatic systems with two or more donor atoms in close disposition to "chelate" metal ions. These ligands would not only be responsible for interaction of the metal compound with DNA but also they could act as carriers of the metal, increasing interaction of complexes with DNA by minimizing exposure

DNA interacting compound is potentially active against parasites.

of metal to inactivating cellular nucleophiles such as thiols.

reactive than the corresponding bulk metal (see example of AgNPs in section 2.1.9).

different sizes to get the optimal uptake rate and blood circulation times of the drug.

*2.2.4. Specifically targeted drugs: Metal inhibitors of parasite enzymes and DNA-binders*

parasite.

480 Leishmaniasis - Trends in Epidemiology, Diagnosis and Treatment

ideal drug leads.

In this section we will describe some of our latest findings, which have been published recently. [40] Through this case study, we seek to provide the reader with an useful insight on our research, which is aimed at the rational design of new biomimetic metal-based systems as potential antiparasitic agents. Our research activity can be summarized in the following tasks:


#### **3.1. Transition metal complexes with 1,2,4-triazolo[1,5-a]pyrimidines**

1,2,4-triazolopyrimidines are bicyclic heterocycles that are formed from the condensation of a ring of 1,2,4-triazole and another one of pyrimidine. Depending on the relative orientation of both rings, four different isomeric families can arise: 1,2,4-triazolo[1,5-a]pyrimidines, 1,2,4 triazolo[1,5-c]pyrimidines, 1,2,4-triazolo[4,3-a]pyrimidines, and 1,2,4-triazolo[4,3-c]pyrimi‐ dines. Among them, 1,2,4-triazolo[1,5-a]pyrimidine derivatives are the most stable thermodynamically and, because of this, the object of our present studies.

In previous works, 1,2,4-triazolo[1,5-a]pyrimidines have proved to be excellent ligands for a wide range of transition metal ions. [43] This fact is due to their, at least, three coordination positions, N1, N3 and N4, which can lead to several coordination modes. The coordination capability of these derivatives can be increased by the presence of heteroatoms as ringsubstituents. However, a systematic revision on the existing results indicates a clear trend of these ligands to coordinate monodentately by N3, followed by N3,N4-bidentate and N1,N3 bidentate bridging modes (Figure 8).

The rich coordination chemistry of these derivatives has led in the last years to a great variety of multidimensional coordination compounds showing interesting properties, especially from the magnetic and biological viewpoints.

In addition, the biomimetic character of 1,2,4-triazolo[1,5-a]pyrimidines with purine nucleo‐ bases confers a potential biological activity to these derivatives and to their metal complexes, which can be used for therapeutic aims. Our studies have revealed the high potential of this kind of compounds for acting as leishmanicidal agents.

**Figure 8.** Basic structure of 5,7-substituted 1,2,4-triazolo[1,5-a]pyrimidines (a) and purines (b). Numbering scheme and possible binding sites to metal ions are also depicted for triazolopyrimidines. X=donor atom (N, O, S, etc.)

Herein we report the results obtained with three of the most promising metal compounds we have obtained so far: [Cu(HmtpO)2(H2O)3](ClO4)2 H2O (1), {[Cu(HmtpO)2(H2O)2](ClO4)2 2HmtpO}n (2) and {Co(HmtpO)(H2O)3](ClO4)2 2H2O}n (3), Figure 9. All of them contain the neutral form of 5-methyl-1,2,4-triazolo[1,5-a]pyrimidin-7(4*H*)-one (HmtpO) and perchlorate as counteranion. The three compounds show different topology and dimensionality. Com‐ pound 1 is a monomeric complex in which HmtpO shows both N3 monodentate and N1,O71 bidentate modes; compound 2 is a two-dimensional framework in which HmtpO ligand shows an N3,O71 bidentate bridging mode; and the structure of compound 3 consists of onedimensional chains in which HmtpO displays an N1,N3,O71 tridentate bridging mode. The structural diversity of these compounds is mainly due to the mode of the triazolopyrimidine ligand.

As depicted in Figure 9, the compounds 1-3 were synthesized by mixing their corresponding metal perchlorate salts with HmtpO derivative in aqueous media and bringing to reflux for 30 min before acidification with HCl. In all cases, compounds were isolated as crystals from their respective solution after several days standing at room temperature. Obtention of single crystals allowed to determine their crystal structure by X-ray analysis and their characteriza‐ tion was completed by elemental and thermal analysis (thermogravimetry and differential scanning calorimetry), and spectroscopic techniques such as FTIR and UV-Vis. Magnetic studies indicate that compound 1 exhibits simple paramagnetism in 2-300 K while the overall behaviour of 2 and 3 corresponds to weak ferromagnetically and antiferromagnetically coupled systems, respectively. [44]

#### **3.2.** *In vitro* **antiproliferative activity against promastigote forms (extracellular forms) and toxicity against a mammalian host cell model**

Firstly we evaluated the toxic activity of the free triazolopyrimidine compound HmtpO and its Cu(II) and Co(II) complexes 1-3 against promastigotes of two species of *Leishmania* (*L.*

**Figure 9.** Synthetic scheme and structures of triazolopyrimidine derivative HmtpO and its metal complexes 1-3. Please note that the graphs of 1-3 correspond only to the cationic part of the metal compounds.

Herein we report the results obtained with three of the most promising metal compounds we have obtained so far: [Cu(HmtpO)2(H2O)3](ClO4)2 H2O (1), {[Cu(HmtpO)2(H2O)2](ClO4)2 2HmtpO}n (2) and {Co(HmtpO)(H2O)3](ClO4)2 2H2O}n (3), Figure 9. All of them contain the neutral form of 5-methyl-1,2,4-triazolo[1,5-a]pyrimidin-7(4*H*)-one (HmtpO) and perchlorate as counteranion. The three compounds show different topology and dimensionality. Com‐ pound 1 is a monomeric complex in which HmtpO shows both N3 monodentate and N1,O71 bidentate modes; compound 2 is a two-dimensional framework in which HmtpO ligand shows an N3,O71 bidentate bridging mode; and the structure of compound 3 consists of onedimensional chains in which HmtpO displays an N1,N3,O71 tridentate bridging mode. The structural diversity of these compounds is mainly due to the mode of the triazolopyrimidine

**Figure 8.** Basic structure of 5,7-substituted 1,2,4-triazolo[1,5-a]pyrimidines (a) and purines (b). Numbering scheme and possible binding sites to metal ions are also depicted for triazolopyrimidines. X=donor atom (N, O, S, etc.)

As depicted in Figure 9, the compounds 1-3 were synthesized by mixing their corresponding metal perchlorate salts with HmtpO derivative in aqueous media and bringing to reflux for 30 min before acidification with HCl. In all cases, compounds were isolated as crystals from their respective solution after several days standing at room temperature. Obtention of single crystals allowed to determine their crystal structure by X-ray analysis and their characteriza‐ tion was completed by elemental and thermal analysis (thermogravimetry and differential scanning calorimetry), and spectroscopic techniques such as FTIR and UV-Vis. Magnetic studies indicate that compound 1 exhibits simple paramagnetism in 2-300 K while the overall behaviour of 2 and 3 corresponds to weak ferromagnetically and antiferromagnetically

**3.2.** *In vitro* **antiproliferative activity against promastigote forms (extracellular forms) and**

Firstly we evaluated the toxic activity of the free triazolopyrimidine compound HmtpO and its Cu(II) and Co(II) complexes 1-3 against promastigotes of two species of *Leishmania* (*L.*

ligand.

coupled systems, respectively. [44]

**toxicity against a mammalian host cell model**

482 Leishmaniasis - Trends in Epidemiology, Diagnosis and Treatment

*infantum* and *L. braziliensis*). IC50 values registered after 72 h of exposure are shown in Table 2, including Glucantime as reference drug. Antileishmanial activity of metal complexes 1-3, expressed as IC50, was similar to that found for Glucantime for both *L. infantum* and *L. braziliensis*. In contrast, the free derivative HmtpO is significantly less active than its metal compounds.

To evaluate toxicity on the host, J774.2 macrophages (mammalian cells) were used as cell model. Cytotoxic studies showed that metal complexes 1-3 are much less toxic than Glucantime and the free HmtpO derivative (Table 2).

On the other hand, selectivity and thus efficacy of assayed compounds towards parasite cells was evaluated and quantified by using the selectivity index (SI). This parameter is defined as the cocient between IC50 for cells and IC50 for parasites. A value greater than 1 is considered more selective for activity against parasites, and a value less than 1 is considered more selective for activity against cells.[45] SI of these derivatives was 30-fold or more higher than SI of Glucantime and HmtpO. These results are indicative of the higher potential of metal com‐ pounds 1-3 as antiparasitic agents compared with the current treatments, in this case Glucan‐ time. Moreover it is evident that the presence of the metal ion in the scaffold enhances significantly triazolopyrimidine derivative activity and selectivity. This example constitutes another proof of the validity of the metal-drug synergism approach.

#### **3.3. Effects on the infection rate and the intracellular replication of the amastigote forms**

Most studies on *in vitro* biological activity of new compounds against *Leishmania* spp. are performed on promastigote forms because it is much easier to work with these forms *in vitro*.


a Towards J774.2 macrophages after 72 h of culture. IC50 is the concentration required to give 50% inhibition, calculated by linear regression analysis from the Kc values at concentrations employed (1, 10, 25, 50 and 100 µM).

b Selectivity index (SI) =IC50 macrophages/IC50 parasite

**Table 2.** *In vitro* activity of reference drugs, free HmtpO derivative and metal compounds 1, 2 and 3 against promastigote forms of *Leishmania* spp.

However, in our studies we also include the effects of these compounds on the forms that develop in the host (amastigotes). This study is of great importance to determine effects in the definitive host and thus it gives a better idea of the potential application as antiparasitic drugs.

To predict the effect of metal complexes 1-3 on the capacity for infection and growth inhibition of intracellular forms of *L. infantum* and *L. braziliensis*, adherent J774.2 macrophages (1×105 macrophages) were incubated for two days and then infected with 1×106 promastigote forms of *L. infantum* and *L. braziliensis* for 12 h. Non-phagocytosed parasites were afterwards removed and culture was kept in fresh medium for 10 days. Parasites invaded cells and then converted into amastigotes within one day after infection. On the 10th day, the rate of host-cell infection reached the maximum. When drugs 1-3 were added at their respective IC25 concentration to macrophages infected with *Leishmania* spp. promastigote forms in exponential growth phase, infection rate decreased significantly after 12 h with respect to control measurements, follow‐ ing the trend 1>3>2 for *L. infantum* and 3>1>2 for *L. braziliensis*, with percentages of infestationinhibition capacity of 84%, 79% and 67%, respectively, in the case of *L. infantum* and 86%, 79% and 75%, respectively, in the case of *L. braziliensis*. These values are remarkably higher than those for inhibition by Glucantime (56% and 36% for *L. infantum* and *L. braziliensis*, respec‐ tively). The three complexes inhibited *Leishmania* spp. amastigote replication in macrophage cells *in vitro*, following a similar pattern to that for infection rate inhibition and again being more effective than reference drug. Although not always it is possible to establish a direct relationship between drug action on extracelular promastigote and intracellular amastigote forms, in case of compound 3, it was effective against both forms.

#### **3.4. Studies on the mechanism of action**

In order to investigate the possible mechanism of action of metal compounds 1-3 on the parasite, their effect on metabolite excretion is analyzed, and microscopy studies on the treated parasites are carried out to visualize any ultrastructural alteration that may be provocked by the compounds.

#### *3.4.1. Metabolite excretion effect*

However, in our studies we also include the effects of these compounds on the forms that develop in the host (amastigotes). This study is of great importance to determine effects in the definitive host and thus it gives a better idea of the potential application as antiparasitic drugs.

Towards J774.2 macrophages after 72 h of culture. IC50 is the concentration required to give 50% inhibition, calculated

by linear regression analysis from the Kc values at concentrations employed (1, 10, 25, 50 and 100 µM).

**Table 2.** *In vitro* activity of reference drugs, free HmtpO derivative and metal compounds 1, 2 and 3 against

To predict the effect of metal complexes 1-3 on the capacity for infection and growth inhibition of intracellular forms of *L. infantum* and *L. braziliensis*, adherent J774.2 macrophages (1×105

of *L. infantum* and *L. braziliensis* for 12 h. Non-phagocytosed parasites were afterwards removed and culture was kept in fresh medium for 10 days. Parasites invaded cells and then converted into amastigotes within one day after infection. On the 10th day, the rate of host-cell infection reached the maximum. When drugs 1-3 were added at their respective IC25 concentration to macrophages infected with *Leishmania* spp. promastigote forms in exponential growth phase, infection rate decreased significantly after 12 h with respect to control measurements, follow‐ ing the trend 1>3>2 for *L. infantum* and 3>1>2 for *L. braziliensis*, with percentages of infestationinhibition capacity of 84%, 79% and 67%, respectively, in the case of *L. infantum* and 86%, 79% and 75%, respectively, in the case of *L. braziliensis*. These values are remarkably higher than those for inhibition by Glucantime (56% and 36% for *L. infantum* and *L. braziliensis*, respec‐ tively). The three complexes inhibited *Leishmania* spp. amastigote replication in macrophage cells *in vitro*, following a similar pattern to that for infection rate inhibition and again being more effective than reference drug. Although not always it is possible to establish a direct relationship between drug action on extracelular promastigote and intracellular amastigote

In order to investigate the possible mechanism of action of metal compounds 1-3 on the parasite, their effect on metabolite excretion is analyzed, and microscopy studies on the treated

promastigote forms

macrophages) were incubated for two days and then infected with 1×106

forms, in case of compound 3, it was effective against both forms.

**3.4. Studies on the mechanism of action**

b Selectivity index (SI) =IC50 macrophages/IC50 parasite

484 Leishmaniasis - Trends in Epidemiology, Diagnosis and Treatment

promastigote forms of *Leishmania* spp.

a

To the best of our knowledge, none of the trypanosomatids studied is capable of completely degrading glucose to CO2 under aerobic conditions, so they excret a great part of the carbon skeleton into medium as fermented metabolites, which can differ according to the employed species.[46] *Leishmania* spp. have a high rate of glucose consumption, thereby acidifying culture medium due to incomplete oxidation to acids. 1 H-NMR spectra enable us to determine fermented metabolites that are excreted by the parasites during their *in vitro* culture. One of the major metabolites excreted by *Leishmania* spp. is succinate, the main role of which is probably to maintain the glycosomal redox balance by providing two glycosomal oxidore‐ ductase enzymes. These enzymes allow reoxidation of NADH that is produced by glyceral‐ dehyde-3-phosphate dehydrogenase in the glycolytic pathway. Succinic fermentation offers one significant advantage, since it requires only half of the produced phosphoenolpyruvate (PEP) to maintain the NAD+ /NADH balance. The remaining PEP is converted into acetate, depending on the species being considered. Figure 10 (on the left) shows 1 H-NMR spectrum of cell-free medium four days after inoculation with *L. infantum*. Additional peaks, corre‐ sponding to the major metabolites that were produced and excreted during growth, could be detected when this spectrum was compared with the one made in fresh medium. Taking into account that *L. infantum* excretes mainly succinate and acetate, 1 H-NMR spectra show that only compound 2 significantly altered excreted metabolites by *L. infantum*. When promastigote forms of *L. infantum* were treated with compound 2 at IC25 doses, the excretion of catabolites (succinate and acetate) was clearly disturbed and a new peak, identified as pyruvate, appeared (Figure 10). These results mean that compound 2 inhibits glycosomal enzymes, causing pyruvate to be excreted as a final metabolite. On the other hand, compounds 1 and 3 inhibite excreted metabolites only slightly. In the case of *L. braziliensis*, compounds 1-3 showed a similar behavior as for *L. infantum*, being again compound 2 the most inhibitory.

#### *3.4.2. Ultrastructural alterations*

Transmission electron microscopy images showed that compounds 1-3 induced morphologi‐ cal alterations in *L. infantum* and *L. braziliensis* promastigotes when parasites were treated with the respective IC25. Compound 2 was the most effective against both parasite species.

**Figure 10.** NMR spectra of promastigote forms of *L. infantum*, which show the characteristic peaks of the major ex‐ creted metabolites of non-treated parasites (left) and parasites that have been treated with IC25 of compound 2 (right) for four days.

**Figure 10.** NMR spectra of promastigote forms of *L. infantum*, which show the characteristic peaks of the major ex‐ creted metabolites of non-treated parasites (left) and parasites that have been treated with IC25 of compound 2 (right)

for four days.

486 Leishmaniasis - Trends in Epidemiology, Diagnosis and Treatment

**Figure 11.** TEM images showing ultrastructural alterations in *L. infantum* and *L. braziliensis* after being treated with compounds 1, 2 and 3 (at IC25 concentrations) for 72h. (a) Control parasite of *L. infantum* showing organelles with their characteristic aspect, such as nucleus (N), kinetoplast (K), flagellum (F), glycosomes (G) and mitochondrion (M). Bar=1.00 µm. (b) Control parasite of *L. braziliensis* with structures such as nucleus (N), vacuoles (V) and mitochondrion (M). Bar=1.00 µm. (c) *L. infantum* treated with compound 2, showing cellular rest (CR), intense vacuolization (V) and reservosomes (R). Bar=1.59 µm. (d) *L. infantum* treated with compound 3, showing electrodense cytoplasm, vacuoles (V), glycosomes (G) and kinetoplast (K). Bar=1.00 µm. (e) *L. braziliensis* treated with compound 1, showing intense va‐ cuolization (V), giant reservosomes (R) and kinetoplast (K) and swelling mitochondrion (M). Bar=1.00 µm. (f) Promasti‐ gotes of *L. braziliensis* treated with compound 2, with vacuoles (V) and electrodense organelles (arrows). Bar=1.00 µm.

After treating *L. braziliensis* promastigotes with compound 2, many of the parasites appeared dead and others adopted distorted shapes, while in others a uniformly electrodense cytoplasm was formed, in which no cytoplasmic organelles were visible. Parasites vacuolization was pronounced and many of these vacuoles contained strongly electrodense inclusions. In case of *L. infantum*, compound 2 led mostly to cell destruction (Figure 11c), which was evident from the presence of a great quantity of cell remains in supernatant. Likewise parasites had strongly electrodense cytoplasm with intense vacuolization, with both empty vacuoles and membranes, and reservosomes, which appeared in greater numbers than in non-treated promastigotes (Figure 11a).

On the other hand, compound 1 was again very effective against *L. braziliensis* as some parasites appeared dead and others completely altered (Figure 11e), replete with reservosomes and enormous vacuoles. Some promastigotes appeared to be distorted and strongly electrodense, and showed condensed kinetoplast and very swollen mitochondria. In contrast, compound 3 was effective against *L. infantum* (Figure 11d), whose alterations were similar to those already described, with unrecognizable parasites, filled with vacuoles, which distorted their morphol‐ ogy, as well as a great quantity of reservosomes that occupied practically the entire cytoplasm. In these parasites kinetoplast and mitochondria also appeared swollen, resulting in a strongly electrodense cytoplasm. Dead parasites were also visible.

#### **3.5. Final remarks**

In addition to these studies, it should be noted that compounds 1-3 have displayed a high *in vitro* activity against both extra and intracellular forms of *T. cruzi* and are effective at concen‐ trations similar to those of benznidazole. At the same time, they are much less toxic for host cells than the latter. Moreover antileishmanial activity of metal compounds is much higher than that of isolated HmtpO ligand, which is an evidence of the critical role of metal ions in antiparasitic activity. Furthermore, promising *in vivo* activity was observed for all of them, with results consistent with those observed *in vitro*.
