Received Date: October 02, 2015, Accepted Date: October 28, 2015, Published Date:October 31, 2015
Objective: To prepare a new series of [Cu(L-dipeptide)(5-NO2-phen)] complexes and study the influence of the nitro group on the structure, DNA interaction and cytotoxic activity, aimed at finding new copper compounds with antitumor activity. Methodology and results: A series of mononuclear mixed ligand copper(II) complexes of the type [Cu(L-dipeptide)(5-NO2-phen)]·nH2O (where L-dipeptide: Ala-Phe, Phe-Ala, Phe-Val and Phe-Phe) were synthesized and characterized. The crystal structure of [Cu(Phe-Ala)(5-NO2-phen)]?4H2O was solved by X-Ray diffraction. The complexes present square-based pyramidal coordination geometry. UV-Vis spectroscopy results suggest that the coordination observed in solid state is maintained in solution. The complexes bind to isolated DNA, as studied by Circular Dichroism. Biological experiments showed that all the complexes induce cell death in HeLa (human cervical adenocarcinoma) and MDA-MB-231 (human metastatic breast adenocarcinoma) cell lines. Their cytotoxic activity is higher than that of the Cisplatin. Conclusions: Four new [Cu(L-dipeptide)(5-NO2-phen)]·nH2O were prepared, which present good cytotoxic activity. The introduction of the nitro group on the phen impaired DNA binding and cytotoxic activity.
Copper complexes; Dipeptide; 5-NO2-phen; DRX; DNA interaction; Cytotoxic activity
Coordination chemistry has provided medicine with drugs for clinical anticancer chemotherapy as well as a wider group of candidate compounds in different stages of research with promising activities. With the aim to obtain new antitumor metallo-drugs, complexes of several metals other than Pt are being studied . In particular, under the hypothesis that endogenous metals may be less toxic than Pt, several copper complexes were studied looking for cytotoxic activity . As a consequence, a variety of Cu-complexes were found to present high antitumor activity, as tested in vitro on several cancer cell lines, and a few of them on in vivo experiments [2,3]. One of the most relevant examples of active Cu-complexes are the coordination compounds Casiopeinas®, heteroleptic complexes containing an aromatic diimine and a coligand, two of which are already approved for clinical trials as antitumor drugs . In this series of hetereolopetic compounds it was found that the aromatic diiminic ligand was the most influent ligand in relation to the biological activity, with subtle differences in its structure influencing strongly its activity [5,6].
Two heteroleptic Cu-5-NO2-1,10-phenanthroline complexes were reported to present potent antitumor activity in vitro, bringing the attention to this phen derivative as a ligand for further development of heteroleptic complexes with antitumor activity . The 5-NO2-1,10-phenanthroline (5-NO2-phen) ligand presents different hydrophobicity and hydrogen bonding properties than the 1,10-phenanthroline (phen). It acts as an electron-attractor substituent, modifying the electronic properties of the molecule. These properties may influence their complexes reactivity, possibly showing different ability to bind to DNA than Cu-phen complexes [8-11]. As a result, their cytotoxic activity may be different. This difference possibly extends to heteroleptic 5-NO2- phen complexes.
Our research group has previously studied homoleptic Cu-Ldipeptide complexes [12-15] and heteroleptic Cu-L-dipeptidephen complexes . While both present cytotoxic activity (IC50 Cu-L-dipeptide 20-50 μM, IC50 Cu-L-dipeptide-phen 1-20 μM on HeLa cells) phen containing complexes present higher activity showing the relevance of the phen ligand to the activity .
To move on with these studies, heteroleptic Cu(II) complexes with L-dipeptide and 5-NO2-1,10-phenanthroline were studied with the aim to evaluate the influence of the 5-NO2-phen moiety in their structure, DNA binding and cytotoxic activity. As coligands, four L-dipeptides were used, selecting those which presented good cytotoxic activity in the Cu-L-dipeptide-phen series (L-Ala-Phe, L-Phe-Ala, L-Phe-Val and L-Phe-Phe) . In this context, this work reports the synthesis, characterization and DNA binding of four new [Cu(L-dipeptide)(5-NO2-1,10- phenanthroline)] compounds. In addition, their cytotoxic activity was evaluated in two cancer cell lines of human origin. To the best of our knowledge, no previous report on cytotoxic activity of copper complexes with dipeptides and 5-nitro-phen as ligands has been published to the date.
Reagents for synthesis and biochemical studies were used as commercially available: Copper salts (Fluka), L-dipeptides, 5-NO2- 1,10-phenanthroline (5-NO2-phen, SIGMA) and Calf thymus DNA (CT-DNA, SIGMA). Distilled water was used for all the experiments.
Synthesis of the complexes and analytical characterization
Ternary Cu-dipeptide-5-NO2-phen complexes (where L-dipetide: Ala-Phe, Phe-Ala, Phe-Val or Phe-Phe) were obtained as follows: 0.1 mmol of dipeptide were dissolved in 100 mL of warm water and 0.1 mmol of CuSO4·5H2O was added. The pH was adjusted to 7 with a 0.1 M NaOH solution. A solution of 0.1 mmol of 5- NO2-phen in 10 mL of ethanol was added. A blue solution was obtained. The compounds were isolated by evaporation at 50- 60° C, until blue crystals were formed. Yield 60-70%. To obtain single crystals, a small amount of the solution was allowed to slowly evaporate at room temperature. Blue crystals, suitable for X-ray diffraction, were only obtained for [Cu(Phe-Ala)(5- NO2- phen)]·4H2O.
Elemental analysis: [Cu(Ala-Phe)(5- NO2-phen)]·11H2O (C1) Calc. for C24H43CuN5O16: C, 39.97, N, 9.71, H, 6.01 Found: C, 40.32, N, 9.64, H, 5.61; [Cu(Phe-Ala)(5- NO2-phen)]·4H2O (C2) Calc. for C24H29CuN5O9: C, 48.44, N, 11.77, H, 4.91 Found: C, 48.09, N, 11.80, H, 4.90; [Cu(Phe-Val)(5- NO2-phen)]·4H2O (C3) Calc. for C26H33CuN5O9: C, 50.12, N, 11.24, H, 5.34 Found: C, 50.08, N, 11.27, H, 5.32; [Cu(Phe-Phe)(5- NO2-phen)]·3,5H2O (C4) Calc. for C30H32CuN5O8.5: C, 54.42, N, 10.58, H, 4.87 Found: C, 54.41, N, 10.64, H, 4.74.
Chemical analyses for carbon, nitrogen, hydrogen and sulfur were performed with a Carlo Erba analyzer. Infrared spectra were recorded with a Bomen FT-IR spectrophotometer from 4000 to 400 cm−1 using KBr disks. UV–visible (UV–vis) spectra of the complexes in solution were carried out with a Milton Roy Spectronic 3000 spectrophotometer, using 1 cm path length quartz cells. Circular Dichroism spectra were recorded on a Jasco 720 instrument using 1 cm path length quartz cells.
Crystal structure determination of [Cu(Phe-Ala) (5-NO2-phen)]·4H2O, C2
Data for a suitable single crystal for complex C2 was collected at 293  K on an Enraf-Nonius FR590 Kappa-CCD diffractometer using graphite monochromated MoKα radiation (0,71073 Å). Bruker AXS Collect software was used for data collection and the HKL Denzo-Scalepack program suite for data reduction. The structure was solved by direct methods using SIR-92  and refined with SHELXL-2014 . Multi-scan absorption correction was applied . SHELXL-2014 and ORTEP-3 programs were used within the WinGX . Molecular structure graphics were prepared using MERCURY program . All non-hydrogen atoms were refined using anisotropic displacement parameters. Carbon bonded hydrogen atoms were stereochemically position and refined isotropically using the riding-model, thermal parameters set at 1.2 times the Uiso of the C atom they are bonded to Amino and water hydrogen atoms were found in the difference Fourier map, positionally fixed and their thermal parameters set to 1.2 times the Uiso of the nitrogen atom for the amine and 1.5 times the Uiso of the oxygen atom for the water molecules. Summary of the crystallographic data, experimental details and refinement results are listed in Table 1.
|Wave length||0.71073 Å|
|Unit cell dimensions|
|Density (calculated)||1.486 Mg/m3|
|Absorption coefficient||0.882 mm-1|
|Crystal size||0.292 × 0.206 × 0.124 mm3|
|Theta range for data collection||3.122 to 26.716°|
|Index ranges||-26<=h<=28 -14<=k<=14 -10<=l<=13|
|Independent reflections||5407 [R(int)=0.0274]|
|Completeness to theta=25.242°||99.7%|
|Maximum and minimum transmission||0.896 and 0.804|
|Refinement method||Full-matrix least-squares on F2|
|Data / restraints / parameters||5407 / 1 / 353|
|Goodness-of-fit on F2||1.037|
|Final R indices [I>2sigma(I)]||R1=0.0367, wR2=0.0777|
|R indices (all data)||R1=0.0502, wR2=0.0833|
|Absolute structure parameter||-0.021(7)|
|Largest different peak and hole||0.347 and -0.343 e.Å-3|
Table 1 Crystal data and structure refinement for C2.
Binding to DNA: CD studies
Absorption titration measurements were carried out on Calf Thymus-DNA in phosphate buffer solution (pH=7.4) as follows. To a 2.5 mL sample of 1.7 × 10-4 M CT-DNA 10 μL aliquots of the complexes 2.0 × 10-4 M in buffer were added. The CD spectra were registered after 15 m of incubation at room temperature.
Cell lines were obtained from the American Type Culture Collection (ATCC): HeLa (CCL-2™, human cervical adenocarcinoma), MDAMB- 231 (HTB-22™, human metastatic breast adenocarcinoma). HeLa cells were grown in high glucose (4.5 g/L) Dulbecco's Modified Eagle Medium (DMEM) and MDA-MB-231 cells were grown in RPMI Medium both containing stable L-Glutamine and supplemented with 10% Fetal Bovine Serum (FBS, GE Healthcare). Cells were removed enzymatically from flasks using 0.01% Trypsin-EDTA solution. Cultured cells were incubated at 37°C on 5% CO2 atmosphere for 48 hours, containing solutions of the Cu complexes (final concentrations ranging from 1 to 50 μM). Cell viability in response to the complexes was determined by a colorimetric assay using Cell Counting Kit – 8 (Fulka) on 15.000 cells grown in 96-well plates. The kit utilizes a water-soluble tetrazolium salt that is reduced by dehydrogenases in cells to give a yellow colored product (formazan). The amount of the formazan dye generated by the activity of dehydrogenases in cells is directly proportional to the number of living cells. Absorbance of converted dye was measured at 450 nm using a micro plate reader. Cu-complexes solutions were prepared in 5% DMSO in water (due to poor solubility in water). The higher concentration of DMSO in culture medium was 0.5% (for the 100 μM test) and proved not to induce cytotoxicity. The IC50 was estimated from the dose-response curves, deviation between duplicates did not exceed 10%.
5-NO2-phen copper complexes have not been widely studied yet. There are only six X-ray crystal structures of copper complexes containing 5-NO2-phen reported to the date. None of them contains dipeptides as co-ligand [7,22-26]. The crystal structure of a new complex: [Cu(L-Phe-Ala)(5-NO2-phen)]·4H2O (C2) was obtained. The ortep type view of the compound with anisotropic displacement ellipsoids (50% probability) and numbering scheme is shown in Figure 1. The complex presents a coordination sphere where the copper ion is situated in a distorted squared pyramidal environment, equatorially coordinated to two N and one O atom from the dipeptide, and one N atom from the phen. The coordination sphere is completed by one N atom of the phen in the apical position. Table 2 shows bond distances and angles around the copper atom. The 5-NO2-phen ligand presents a usual bite distance (2.687 Å vs 2.64 mean distance for Cu(II)- 5-NO2- phen complexes according to the Cambridge Structural Data Base) and angle (77.20° vs 81.08° mean angle) and it is in a nearly orthogonal position with the Cu-peptide moiety. The distortion of the pyramidal environment was evaluated by means of τ factor which varies from 0 for perfect square pyramidal to 1 for trigonalbipyramid geometry . τ=0.16, therefore the contribution of the trigonal-bipyramid geometry is 16%.
|Selected Bond Length (Å)||Cu1 N5||2.011(3)|
|Cu1 N4||1.891(4)||Cu1 N2||2.025(4)|
|Cu1 O3||1.993(3)||Cu1 N1||2.271(4)|
|Selected angles (°)|
|N4 Cu1 O3||83.39(13)||N5 Cu1 N2||94.72(13)|
|N4 Cu1N5||83.40(13)||N4 Cu1 N1||109.10(14)|
|O3 Cu1 N5||164.08(13)||O3 Cu1 N1||96.02(13)|
|N4 Cu1 N2||173.55(18)||N5Cu1 N1||96.68(14)|
|O3 Cu1 N2||97.43(14)||N2 Cu1 N1||77.21(15)|
Table 2 Selected bond lengths and angles.
In order to determine the structural similarity between this 5-NO2- phen complex with its previously reported analogue containing phen  the crystal structures of [Cu(Phe-Ala)(5-NO2-phen) and [Cu(Phe-Ala)(phen)] were compared by means of the solid form module in the Mercury program. The restrictions applied to the search of similarity by the module were 20% distance and 10 degree angle tolerance. The comparison led up to a 97% similarity of the complexes’ powder diffraction patterns. When comparing one molecule of each there is a root mean square (rms) of 0.19 Å. Comparison of cell parameters are shown in Table 3 and the diagrams resulting from the comparison are shown in Figure 2. The compounds are isostructural. Moreover, the spatial disposition of atoms around the coordination center and throughout the crystal lattice was practically identical.
|[Cu(phe-ala)(phen)]·4H2O||C2||22.5305(4) Å||11.5646(2) Å||10.7493(2) Å||90°||116.108(1)°||90°|
|[Cu(phe-ala)(5-NO2-phen)]·4H2O||C2||22.9400(8) Å||11.7710(5) Å||10.6910(5) Å||90°||112.860(2)°||90°|
Table 3: Cell parameters comparison.
Infrared spectra of the complexes
The infrared spectra of the complexes present several commonfeatures. They are also similar to those of the Cu-dipeptide-phen complexes . For instance, a broad, very strong peak around 1600 cm-1 attributed to ν(C=O)+ν(C-N)+νas(COO), indicative of the coordinated dipeptide moiety [12,13,28]. Absorption peaks corresponding to ring stretching frequencies of the 5-NO2- phen are slightly modified in relation to the free ligand and appear around 1515 cm-1 and 1420 cm-1, in agreement with the coordination of the 5-NO2-phen . Other distinctive bands, like the ones due to C-H rocking in the 5-NO2-phen, appear from 1200 cm-1 to 720 cm-1, as shown in Table 4. Taking into account the similarity found in the infrared spectra of the complexes it can be assumed that the four studied complexes present a coordination similar to that observed for the [Cu(Phe-Ala)(5-NO2-phen)]·4H2O complex (C2).
|Assignment||n(C=O)+n(C-N)+ n(COO)as||n(C=C)||n(C=N)||r(C-H)||r(C-H)||r (C-H)||r(C-H)|
|[Cu(phe-phe)( 5-NO2-phen)]||1591, 1576||1514||1417||1150||1079||836||734|
Table 4 Wavenumber (cm-1) of representative absorption bands in the complexes and in free 5-NO2-phen, as well as their tentative assignment.
Table 5 presents the wavelength of the maxima of the visible spectra of the complexes. They present a broad peak around 625 nm with a shoulder at about 800 nm, characteristic of copper in penta coordinated environments . The λmax values are similar to those of their [Cu(dipeptide)(phen)] analogs . In an attempt to evaluate if the coordination observed in solid state is maintained in aqueous solution, UV-visible spectra of the complexes were analyzed using the work of Prenesti et al. who have previously developed visible spectra/structure correlations for Cu(II) complexes [31-33].
Table 4: Wavelength of the maxima of the visible spectra of the complexes (λmax) and molar absorbance (εM).
The expected λmax for the observed coordination is 610-630 nm composed by: 580 nm (octahedral arrangement with an N3O equatorial coordination scheme)  plus 30-40 nm due to the presence of one N axially coordinated . This value is in accordance with the observed values for this series of [Cu(dipeptide)(5-NO2-phen)] complexes. This analysis suggests that the same coordination observed in solid state is maintained for the major species in aqueous solution at the studied concentration (around 5 × 10-3 M). Similar results were previously reported for [Cu(Gly-Gly)(5-NO2-phen)] .
The CD spectrum of CT-DNA was monitored alone, in the presence of 5-NO2-phen ligand and the complexes. Spectra for the ligand 5-NO2-phen is presented in Figure 3 (spectra for the other compounds can be found as supplementary material). Upon addition of the ligand 5-NO2-phen to CT-DNA, an intensity increase in the positive band of the spectrum is observed. Similar results are observed for all the complexes. These observation shows that 5-NO2-phen (and its complexes) binds to isolated DNA . It also suggests an intercalative mode of binding of the 5-NO2-phen and its complexes to the DNA, as observed for other Cu-5-NO2-phen heteroleptic complexes . The stacking of the complex molecules, possible through the 5-NO2-phen moiety, between the base pairs of DNA led to an enhancement in the positive band. These results are also similar to that obtained in the same conditions for the phen ligand. The extent of hyperchromism is qualitatively minor in the case of 5-NO2-phen and its complexes (around 20% for 5-NO2-phen and its complexes vs 50% for phen in the same experimental conditions) suggesting that the 5-nitro group impairs DNA binding. A similar behavior was also observed for a series of Ru mixed ligand complexes .
Table 6 presents the IC50 values for the studied complexes. All of them present higher cytotoxic activity (lower IC50) on HeLa and MDA-MB-231 cells than Cisplatin. [Cu(dipeptide)(5-NO2- phen)] complexes show similar activity to [Cu(5-NO2-phen)2Cl2. Compared with their phen analogs, the 5-NO2-phen complexes present lower activity (for instance, [Cu(ala-phe)(phen)] presents an IC50 2 μM on HeLa cells . This tendency was previously observed in the Casiopeinas series, where complexes containing 5-NO2-phen had larger IC50 than their counterparts with phen , showing the relevance of the phen moiety on the activity of the complexes. This reduction of the cytotoxic activity is possibly related to the fact that the nitro group impairs DNA binding. The electron-attractor effect of the nitro substituent on the phen can also influence the reactivity of the Cu center in the complexes and therefore their activity. The activity of the complexes makes them good candidates to perform in vivo antitumor tests comparing they tissue distribution, which may be different from that of the related phen complexes.
|Compound||IC50 (mM) Hela||IC50 (mM) MDA MB 231|
Table 6: Cytotoxic activity (expressed by IC50) of the studied complexes against HeLa and MDA-MB-23 cell lines.
Four new complexes were synthesized and characterized. They present square-based pyramidal coordination geometry with a dipeptide and a 5-NO2-phen coordinated to the Cu(II) center, perpendicular to each other. UV-vis spectra suggest no change in coordination upon dissolution. The complexes bind to isolated DNA, possibly by intercalation. The extent of binding is lower than that observed for Cu-dipeptide-phen complexes. The complexes present cytotoxic activity against the studied cell lines, higher than Cisplatin. The introduction of the nitro group in the phen reduces the cytotoxic activity of the complexes.
The authors thank CSIC and PEDECIBA Química and Biología (Uruguay); CNPq and FAPESP (Brazil) and CAPES/UdelaR (Brazil- Ururguay bilateral research grant) for financial support.
CCDC 1055554 contains the supplementary crystallographic data for C2. This data can be obtained free of charge via https:// summary.ccdc.cam.ac.uk/structure-summary-form, or from the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK, Fax: (+44) 1223-336-033. CD spectra of DNA alone and in the presence of the studied compounds (Figure S1 – S5).
All Published work is licensed under a Creative Commons Attribution 4.0 International License
Copyright © 2018 All rights reserved. iMedPub LTD Last revised : September 22, 2018