4-chloro-cinnamic acid (4-ClcinnH), 4-methoxy-cinnamic acid (4-OMecinnH), 4-chloro-phenylacetic acid (4-ClphacH), 4-methoxy-phenylacetic acid (4-OMephacH) and LnCl₃·xH₂O (Ln = Y, Ce, Nd, Sm, Yb) were purchased from Sigma- Aldrich, and other chemicals were obtained comercially and used without further purification. Sodium cinnamates were synthesized by neutralization of its respective acid with a NaOH solution in water. Microanalyses were performed with a Flash EA 1112 Series CHN Analyzer. Metal analyses were performed in duplicate by ethylene-diamine tetraacetic acid (EDTA) titration (direct complex-metric titration)⁹ and using xylenol orange as indicator. Molar conductance of 10^-3 M solutions of the complexes in DMF was recorded on a Thermo Orion 150 conductivity meter by using 0.01 M KCl water solution as calibrant. A Shimadzu UV-1700 spectrophotometer was used to record the electronic spectra. Infrared spectra were recorded with a NICOLET 6700 Spectrometer (4000-400 cm^-1 with a recording accuracy of 1 cm^-1). Thermogravimetric analyses were performed with a NETZSCH TG 209 F1 thermal analyzer. Heating was conducted under static conditions in nitrogen with a range of 20-700 °C at 20 °C.min_-1 in an Al₂O₃ crucible.

The Ln(III) complexes were prepared using similar methods to those described in the literature (Fig. 2)¹⁰.

Fig. 2
General reaction scheme for the synthesis of rare earth complexes.

Y(4-Clcinn)₃] (1). A 10 ml aqueous solution of Na (4- Clcinn) (3 eq: 1012 mg, 4.95 mmol) was slowly added to a 15 ml aqueous solution of YCl₃·6H2O (500 mg, 1.65 mmol); a precipitate formed instantly upon the addition of the sodium salt. The solution was adjusted to pH 5 with drops of a 0.1 M HCl or NaOH solution controlled by a pH-meter and then stirred for 3 h and filtered. The white precipitate was then washed with distilled water and dried in a desiccator for 2 days. Yield: 82 %. C₂₇H₁₈Cl₃O₆Y, elemental analysis: C 51.21 (calc. 51.17), H 2.91 (2.86), Y 13.85 (14.03) %. IR (KBr pellet cm^-1): 3038 w, 1640 vs, 1592 m, 1525 s, 1492 s, 1417 vs, 1395 s, 1285 w, 1248 m, 1202 w, 1176 w, 1091 s, 1011 m, 982 s, 880 m, 855 w, 825 vs, 759 m, 731 m, 693 w, 664 w, 557 m, 496 s, 458 s. Conductivity (μS·cm^-1, 0.001 M, DMF): 11.7. UV–Vis (ethanol): λmax = 270 nm.

[Y(4-OMecinn)₃] (2). The synthesis of 2 was performed in an identical manner to the synthesis of 1. White powder, yield: 80%. C₃₀H₂₇O₉Y, elemental analysis: C 58.16 (calc. 58.08), H 4.40 (4.39), Y1 4.32 (14.33) %. IR (KBr pellet cm^-1): 3006 w, 2953 w, 2837 w, 1635 s, 1605 s, 1574 m, 1511 s, 1459 m, 1426 s, 1405 s, 1306 m, 1246 vs, 1172 s, 1109 w, 1027 m, 988 m, 876 w, 830 s, 782 w, 728 w, 699 w, 560 m, 516 w, 479 w, 456 w. Conductivity (μS·cm^-1, 0.001 M, DMF): 7.3. UV–Vis (ethanol): λmax= 280 nm.

[Ce(4-Clcinn)₃] (3). The synthesis of 3 was performed in an identical manner to the synthesis of 1. White powder, yield: 88 %. C₂₇H₁₈Cl₃O₆Ce, elemental analysis: C 47.59 (calc. 47.35), H 2.66 (2.65), Ce 20.02 (20.46) %. IR (KBr pellet cm^-1): 3034 w, 2926 w, 1639 vs, 1590 m, 1516 s, 1492 s, 1416 vs, 1393 vs, 1284 w, 1247 m, 1175 w, 1091 s, 1012 m, 982 s, 880 m, 854 w, 825 s, 748 m, 693 w, 665 w, 545 m, 495 m, 456 m. Conductivity (μS·cm^-1, 0.001 M, DMF): 4.4. UV–Vis (ethanol): λmax = 270 nm.

[Ce(4-OMecinn)₃]·H₂O (4). The synthesis of 4 was performed in an identical manner to the synthesis of 1. White powder, yield: 90 %. C₃₀H₂₇O9Ce(OH2), elemental analysis: C 51.89 (calc. 52.25), H 4.21 (4.24), Ce 20.31 (20.32) %. IR (KBr pellet cm^-1): 3420 w, 3003 w, 2934 w, 2836 w, 1638 s, 1606 s, 1573 m, 1512 s, 1426 s, 1401 s, 1304 m, 1245 vs, 1173 s, 1108 w, 1031 m, 983 m, 877 w, 830 vs, 779 m, 720 m, 637 w, 558 s, 518 m, 476 w, 458 w. TGA mass loss 2.80 % (70 - 150 °C, 1 step, calc. 1 × H2O = 2.61 %), 4.70 % (180 - 270 °C, 1 step, calc. 1 × OCH3 [C29H25O8Ce formation] = 4.62 %), 52.0% (310 - 600 °C, 1 step, calc. Ce2O3 formation = 52.4 %). Conductivity (μS·cm^-1, 0.001 M, DMF): 5.9. UV–Vis (ethanol): λmax = 280 nm.

[Ce(4-Clphac)₃]·H₂O (5). The synthesis of 5 was performed in an identical manner to the synthesis of 1. Pale yellow powder, yield: 74 %. C₂₄H₁₈Cl₃O₆Ce(OH₂), elemental analysis: C 43.04 (calc. 43.22), H 3.01 (3.02), Ce 21.13 (21.01) %. IR (KBr pellet cm^-1): 3440 s, 3044 w, 1631 s, 1544 vs, 1493 m, 1422 s, 1398 s, 1286 m, 1200 w, 1092 m, 1017 w, 935 w, 857 w, 809 m, 743 m, 683 m, 594 w, 507 w, 476 m, 424 w. TGA mass loss 2.40% (75- 50 °C, 1 step, calc. 1 × H2O = 2.70 %), 64.6 % (300-600 °C, 1 step, calc. CeOCO3 formation = 67.6 %). Conductivity (μS·cm^-1, 0.001 M, DMF): 5.5.

[Ce(4-OMephac)³]·H₂O (6). The synthesis of 6 was performed in an identical manner to the synthesis of 1. Pale yellow powder, yield: 81 %. C₂₇H₂₇O₉Ce(OH₂), elemental analysis: C 49.96 (calc. 49.61), H 4.32 (4.47), Ce 21.47 (21.44) %. IR (KBr pellet cm^-1): 3460 s, 3058 w, 2905 w, 2836 w, 1636 vs, 1574 m, 1543 s, 1517 vs, 1430 s, 1400 s, 1284 m, 1251 vs, 1178 m, 1106 w, 1034 m, 949 w, 859 w, 821 w, 730 m, 697 w, 588 w, 538 w, 481 w, 446 w. TGA mass loss 3.10 % (95-160 °C, 1 step, calc. 1 × H2O = 2.76 %), 70.5 % (280-600 °C, 1 step, calc. CeO2 formation = 73.7%). Conductivity (μS·cm^-1, 0.001 M, DMF): 11.6.

[Nd(4-Clcinn)³]·2H₂O (7). The synthesis of 7 was performed in an identical manner to the synthesis of 1. White powder, yield: 85 %. C₂₇H₁₈Cl₃O₆Nd(2OH₂), elemental analysis: C 45.02 (calc. 44.73), H 2.99 (3.06), Nd 19.97 (19.89) %. IR (KBr pellet cm^-1): 3344 m, 3046 w, 1640 vs, 1591 w, 1507 vs, 1420 vs, 1393 s, 1249 m, 1093 s, 1014 w, 984 s, 826 s, 756 m, 542 w, 496 m, 456 m. TGA mass loss 6.00 % (80- 150 °C, 1 step, calc. 2 × H₂O = 4.97 %), 48.0 % (340-580 °C, 1 step, calc. Nd₂O₂CO₃ formation = 47.5 %). Conductivity (μS·cm^-1, 0.001 M, DMF): 3.9. UV–Vis (ethanol): λmax = 268 nm.

[Nd(4-OMecinn)₃]·H₂O (8). The synthesis of 8 was performed in an identical manner to the synthesis of 1. White powder, yield: 89 %. C₃₀H₂₇O₉Nd(OH₂), elemental analysis: C 52.31 (calc. 51.94), H 4.23 (4.21), Nd 20.15 (20.79) %. IR (KBr pellet cm^-1): 3340 w, 3004 w, 2932 w, 2837 w, 1682 m, 1637 s, 1606 s, 1573 m, 1512 vs, 1427 s, 1401 vs, 1304 m, 1245 vs, 1173 s, 1109 w, 1029 m, 984 m, 879 w, 831 s, 779 m, 722 m, 558 s, 520 m, 459 w. Conductivity (μS·cm^-1, 0.001 M, DMF): 6.1. UV–Vis (ethanol): λmax = 279 nm.

[Sm(4-Clcinn)₃]·H₂O (9). The synthesis of 9 was performed in an identical manner to the synthesis of 1. White powder, yield: 82 %. C27H18Cl3O6Sm(OH2), elemental analysis: C 45.66 (calc. 45.47), H 2.90 (2.83), Sm 21.48 (21.08) %. IR (KBr pellet cm^-1): 3353 m, 3041 w, 2925 w, 1639 s, 1590 m, 1510 s, 1494 s, 1417 vs, 1394 s, 1285 w, 1248 m, 1203 w, 1091 s, 1013 m, 983 s, 880 m, 854 w, 825 s, 755 m, 689 w, 667 w, 544 m, 496 m, 457 m. Conductivity (μS·cm^-1, 0.001 M, DMF): 2.7. UV–Vis (ethanol): λmax = 267 nm.

[Sm(4-OMecinn)₃]·H₂O (10). The synthesis of 10 was performed in an identical manner to the synthesis of 1. White powder, yield: 96 %. C₃₀H₂₇O₉Sm(OH₂), elemental analysis: C 51.08 (calc. 51.48), H 4.34 (4.18), Sm 21.18 (21.48) %. IR (KBr pellet cm^-1): 3346 w, 2962 w, 2931 w, 2838 w, 1683 m, 1637 s, 1605 s, 1575 s, 1511 vs, 1427 s, 1403 vs, 1305 m, 1245 vs, 1172 s, 1110 w, 1027 m, 985 m, 881 w, 832 s, 780 m, 723 m, 557 m, 520 m, 459 w. Conductivity (μS·cm^-1, 0.001 M, DMF): 10.5. UV–Vis (ethanol): λmax = 278 nm.

[Yb(4-Clcinn)₃]·H₂O (11). The synthesis of 11 was performed in an identical manner to the synthesis of 1. White powder, yield: 84 %. C₂₇H₁₈Cl₃O₆Yb(OH₂), elemental analysis: C 43.98 (calc. 44.07), H 2.68 (2.74), Yb 23.24 (23.52) %. IR (KBr pellet cm^-1): 3355 m, 3026 w, 1640 s, 1590 m, 1523 s, 1493 s, 1417 vs, 1396 s, 1290 w, 1248 m, 1202 w, 1091 s, 1011 m, 984 s, 880 m, 825 s, 764 m, 748 m, 726 m, 666 w, 557 w, 497 m, 458 m. Conductivity (μS·cm^-1, 0.001 M, DMF): 4.9. UV–Vis (ethanol): λmax = 269 nm.

[Yb(4-OMecinn)₃] (12). The synthesis of 12 was performed in an identical manner to the synthesis of 1. White powder, yield: 80 %. C₃₀H₂₇O₉Yb, elemental analysis: C 51.36 (calc. 51.14), H 3.82 (3.86), Yb 24.72 (24.56) %. IR (KBr pellet cm^{- 1}): 2932 w, 2837 w, 1634 s, 1603 s, 1573 m, 1512 vs, 1427 vs, 1405 vs, 1307 m, 1248 vs, 1173 s, 1110 w, 1027 m, 991 m, 875 w, 830 s, 783 w, 730 w, 697 w, 558 m, 517 m, 459 w. Conductivity (μS·cm^-1, 0.001 M, DMF): 3.5. UV–Vis (ethanol): λmax = 280 nm.

Cytotoxic activity of the compounds was evaluated based on the viability of the human promonocytic cell line U-937 (ATCC CRL-1593.2^TM) determined using the MTT enzymatic micro-method (3-(4,5-dimethylthiazol-2-yl)-2,5- diphenyl tetrazolium bromide). U-937 cells were grown in 96-wells plates at 1x10⁵ cells/ml in RPMI-1640 enriched with 10 % FBS and incubated at 37 °C with 5 % CO2 in the presence of lanthanide complexes. After 72 h of incubation, the viability of cells was determined by assaying the reduction of MTT to formazan as described previously by adding 10 μl/well of an MTT solution (5 mg/ml) and incubating the solution for 3 h. The enzymatic reaction was terminated by the addition of 100 μl of a solution of 50% isopropanol with 10% sodium dodecyl-sulfate. Formazan production was measured at 570 nm on a 96-well microplate reader (BioRad Benchmark). Cells cultivated in the absence of lanthanide complexes and maintained under the same conditions were used as a control. All of the compounds were tested in triplicate in two independent experiments. The results are expressed as 50 % lethal concentration (LC₅₀), calculated by the probit probabilistic method¹¹.

The in vitro leishmanicidal activity of the lanthanide complexes against intracellular amastigotes was evaluated using flow cytometry based on methodology described by Pulido et al.12. Promastigotes of L.(V.) panamensis UA140-epirGFP, which are responsible for the mucocutaneous form of the disease, were grown in a modified biphasic NNN environment with phosphate-buffered saline (PBS) and glucose in the liquid phase. Infection of U-937 cells involved adjusting the number of cells to 1x10⁵ cells/ml and adding 0.1 μg/ml of phorbol myristate acetate (PMA) to induce the transformation of cells to macrophages. The experiment was conducted in 24-well plates where cells were infected with promastigotes in the growth stationary phase. After 4 h of incubation at 34 °C (5 % CO₂), free parasites were removed by two successive washes with PBS. Adapted dilutions of lanthanide complexes were added, and cultures were incubated for 72 h at 37 °C (5 % CO₂). Infected cells cultivated in the absence of lanthanide complexes served as a control. Amphotericin B was used as a reference drug. The anti-parasitic effect was measured with an argon laser in a flow cytometer at 488 nm (excitation) and 525 nm (emission). Each lanthanide complex concentration, as well as each control, was tested in two independent experiments. The results are expressed as the percentages of inhibition¹¹.

In vitro anti-trypanosomal activity of lanthanide complexes were evaluated on Trypanosoma cruzi cells (Tulahuen strain) for intracellular amastigotes that express the Escherichia coli beta-galactosidase gene. After U-937 cells were infected by epimastigotes and incubated for 24 h, lanthanide compounds were added at concentrations of 20 μg/ml. The experiments were performed in duplicate two times, and the results are expressed as the percentages of inhibition. The LC₅₀ of benznidazole, a compound conventionally used for the treatment of Chagas disease, was used as a control¹²,¹³,¹⁴.

The P. falciparum 3D7 strain was grown asynchronously in standard conditions described by Trager and Jensen¹⁵. Compound activity affecting parasite growth as determined in 24- well suspension cultures using fluorometry and quantifying the DNA parasite with ethidium bromide (BrEt) staining¹⁶. The intensity of fluorescence is directly proportional to the amount DNA, which is proportional to the amount of viable parasites. The compounds were dissolved in RPMI 1640 at a concentration of 20 μg/ml. The parasites were incubated with the compounds for 72 h at 37 °C in an atmosphere of 5 % O2, 5 % CO2, and 90 % N₂. Chloroquine was used as a positive control to measure activity. The stained DNA was quantified with a Varioskan Flash spectrofluorometer at an excitation wavelength of 510 nm and an emission wavelength of 595 nm. For each concentration, the experiment was performed in duplicate two times. The results are expressed as EC50 (effective concentration 50), estimated by the probit method¹⁷.

Lanthanide complexes 1-12 have been prepared using methods that involve a metathesis or double displacement reaction, similar to those described in the literature. Complexes 1-12 were isolated as solids in powder, soluble in DMSO, DMF, and acetone, and partially soluble in water and hexane. The results obtained by elemental analysis (carbon and hydrogen), by EDTA complexometry and by TG-DTG analysis, agree with all structures proposed.

The experimental values of the percentage of lanthanide, carbon and hydrogen obtain showed sufficient corresponddence with the calculated values for the suggested general formula of the complexes such as [Ln(L)₃]·xH₂O.

Molar conductance (Λ_M) values of complexes 1-12 (3.5– 11.7 S·cm².mol^-1) were measured in dimethylformamide (DMF) at room temperature. These conductance values are much lower than the values reported for 1:1 electrolytes in these solvents¹⁸. The results indicated a non-ionic nature of the compounds, and hence, they are neutral complexes. The overall charge of the trivalent metal ions was balanced by three negatively charged coordinated ligands (bidentate coordination of the carboxylates ligands).

The UV-Visible spectra of the complexes with cinnamic ligands were recorded by UV-Visible spectroscopy in the wavelength range of 200-800 nm. The band observed at 265 nm for (4-ClcinnH) and at 275 nm for (4-OMecinnH) is due to π-π* transition of conjugated system in the ligand and it was shifted to higher wavelength (red shift) upon complexation. The amounts of these shifts however are small, i.e., 2–5 nm. The occurrence of red shifts for lanthanide chelates indicates the decrease in the corresponding energy gaps in the electronic energy levels of the ligands as a result of chelation.

Infrared spectra of all of the complexes obtained show characteristic absorption bands of the cinnamate and phenylacetate ligands. The absence of the absorption band (COOH) at approximately 1700 cm^-1 of the ligands in all of the spectra indicates coordination of the ligands to the metal. The experimental data obtained by IR spectroscopy of the Ln (III) complexes, as well as of the respective ligands, are presented in Table 1.

Table 1

Select IR bands of Na ligands and their Ln (III) complexes.

For complexes 4-11, the bands in the 3460-3340 cm^-1 region are assigned to the stretching of the OH group, (OH)_w, with the hydration water molecules. For the precursors and complexes derived from cinnamic acid 1-4 and 7-12, bands assigned to the υ(C=C) vibrations are observed in the 1600 cm^-1 region, confirming that the π electron system of the olefinic double bond does not participate in the coordination¹⁰,¹⁹,²⁰,²¹. Stretching of the Ln- O bond (metal-carboxylate) has been attributed to the v(Ln-O) bands that have been reported at 780 cm^-1 for the lanthanum complex²², and differences have been found among all of the compounds due to the presence of the substituents (-Cl, -OMe) and the (C=C)propenyl bond. A significant change to lower frequencies (bathochromic effect) is observed for the stretching υ(CO₂) compared to the sodium salts. Frequency analysis of the bands corresponding to the COO- group demonstrates that ΔυCOO- = υas(COO-) - υs(COO-). The calculated values of ΔυCOOare in the range of 100-122 cm-1, indicating that the ligand is in a chelating environment²³,²⁴. This finding confirms that the ligands and the lanthanide ions are coordinated through the oxygen atoms of the carboxylate groups in a bidentate fashion²⁴.

Thermal decomposition results of compounds 4‒7 in nitrogen are presented in Table 2 and Figure 3. For all compounds, the TG curves show endothermic decomposition between 70 °C and 160 °C, corresponding to the loss of hydration water molecules in one step. For the neodymium complex (7), the loss of two water molecules has been calculated; other compounds studied using TG are monohydrate. In the case of 4, the possible loss of the methoxy substituent on one of the ligands indicates the formation of C₂₉H₂₅O₈Ce at 180-270 °C.

Table 2
Thermal decomposition data on complexes 4-7 in nitrogen

* Measured and calculated values of the percentage of mass loss from the initial sample (initial temperature at 20 °C) to the final product.

Fig. 3
TG curves of compounds: (a) [Ce(4-OMecinn)3].H2O; (b) [Ce(4-Clphac)3].H2O; (c) [Ce(4-OMephac)3].H2O; (d) [Nd(4-Clcinn)3].2H2O

Thermal analysis of Ce(III) anhydrous compounds shows three decomposition oxides between 280 °C and 600 °C: Ce₂O₃ to 4, CeOCO₃²⁵ to 5 and CeO2 to 6 (common for Ce(IV) oxides). Between 340-580 °C, complex 7 shows the decomposition of anhydrous cinnamate to neodymium(III) carbonate oxide (Nd₂O₂CO₃), a phase which has been reported in other studies of lanthanide complexes²⁶,²⁷.

To evaluate in vitro cytotoxicity of lanthanide complexes, experiments were performed using the human promonocytic cellular line (ATCC CRL-1593.2^TM) and the methyl thiazol tetrazolium (MTT) enzymatic micro-method. Results show that lanthanide complexes are potentially non-toxic, with values of LC₅₀ > 200 μg/ml. The in vitro anti-parasitic activity of lanthanide complexes against L. panamensis, P. falciparum and T. cruzi are shown in Table 3. In vitro anti-plasmodial activity has been evaluated on an asynchronous culture of the P. falciparum strain 3D7 (sensitive to chloroquine), according to a previously described procedure¹⁵. Percentages of inhibition vary between 0 and 17 %. Ce(III) complexes with phenylacetate ligands (5 and 6) display a higher percentage of inhibition against P. falciparum. All of the synthesized compounds have been evaluated in vitro against intracellular amastigotes of Leishmania (viannia) paramensis, according to a previously described procedure¹². Results displayed in Table 3 indicate that Ce(III) complexes 3-6 significantly inhibit the growth of Leishmania parasites. Compound 4, which was able to reduce the infection by 22 %, could have leishmanicidal potential if the effective concentration 50 (EC50), i.e. the concentration of compound that achieves decrease infection in at least 50 %, less than 50 μg/ml, a concentration classified as moderately active compound. Factors that most likely contribute to this anti-parasitic activity are the higher solubility of these complexes in culture media and the possibility that this metal is involved in biological oxidation-reduction reactions (Ce³⁺/Ce⁴⁺), has shown that the cerium cinnamate presents corrosion resistance, and according XPS curves, 3+ and 4+ oxidation states are involved²⁸. The activity of these cerium compounds could increase intracellular oxidative stress which could affect parasite redox metabolism²⁹. Lanthanides in complex with methoxy groups display higher anti-leishmanial activity than complexes with chorine substituents, however, no significant correlation has been found between modifications of the C=C double bond of an alkenyl substituent (compounds 3 and 4) to an alkyl fragment (compounds 5 and 6) and the leishmanicidal activity of the Ce(III) complexes. Lanthanide complexes 1-12 have been evaluated for in vitro trypanocidal activity against T. cruzi (tulahuen strain). All complexes show antiparasitic activity against T. cruzi. Compounds 2, 4, 6, 8, 10 and 12 with methoxy groups in the aromatic ring show the highest inhibition percentages (> 40 %), based on their different abilities to interact with vital biomolecules (DNA and specific proteins) as shown in other studies³⁰, even its mechanism of action could be related to the ability to produce toxic free radical species in the parasite³¹. The results show that although the structures of the complexes are similar, the biological activity depends on the ligands and the metal center.

Table 3

In vitro anti-parasitic and cytotoxic activities of lanthanide complexes 1-12.

a Growth inhibition of U-937 cells or parasites determined by flow cytometry for intracellular amastigotes of L. panamensis, by fluorometry for all forms of P. falciparum and by the colorimetric β-galactosidase method for intracellular amastigotes of T. cruzi. Data are expressed as the average of at least two independent experiments, each performed in triplicate. b B amphotericin was applied at 0.05 μg/ml, the concentration at which infection is reduced by 50 %. c EC50 = 0.1 μg/ml d Benznidazole was applied at 10 μg/ml, the concentration at which infection is reduced by 50 %.

dorian.polo@correounivalle.edu.co