In Vitro and in Vivo Anticancer Activity of Copper(I) Complexes with

Article pubs.acs.org/jmc

In Vitro and in Vivo Anticancer Activity of Copper(I) Complexes with Homoscorpionate Tridentate Tris(pyrazolyl)borate and Auxiliary Monodentate Phosphine Ligands Valentina Gandin,† Francesco Tisato,‡ Alessandro Dolmella,† Maura Pellei,§ Carlo Santini,§ Marco Giorgetti,∥ Cristina Marzano,*,† and Marina Porchia‡ †

Dipartimento di Scienze del Farmaco, Università di Padova, via Marzolo 5, 35131 Padova, Italy IENI-CNR, Corso Stati Uniti 4, 35127 Padova, Italy § Scuola di Scienze e Tecnologie − Sez. Chimica, Università di Camerino, via S. Agostino 1, 62032 Camerino, Macerata, Italy ∥ Dipartimento di Chimica Industriale, Università di Bologna, Viale Risorgimento 4, 40136 Bologna, Italy ‡

S Supporting Information *

ABSTRACT: Tetrahedral copper(I) TpCuP complexes 1−15, where Tp is a N,N,N-tris(azolyl)borate and P is a tertiary phosphine, have been synthesized and characterized by means of NMR, ESI-MS, and XAS-EXAFS, and X-ray diffraction analyses on the representative complexes 1 and 10, respectively. All copper(I) complexes were evaluated for their antiproliferative activity against a panel of human cancer cell lines (including cisplatin and multidrug-resistant sublines). The two most effective complexes [HB(pz)3]Cu(PCN), 1, and [HB(pz)3]Cu(PTA), 2, showed selectivity toward tumor vs normal cells, inhibition of 26S proteasome activity associated with endoplasmic reticulum (ER) stress, and unfolded protein response (UPR) activation. No biochemical hallmarks of apoptosis were detected, and morphology studies revealed an extensive cytoplasmic vacuolization coherently with a paraptosis-like cell death mechanism. Finally, the antitumor efficacy of complex 1 was validated in the murine Lewis Lung Carcinoma (LLC) model.

■

INTRODUCTION Copper is an essential trace element that plays a pivotal role in a wide range of physiological cellular processes. However, due to the high redox reactivity, free copper ions are extremely cytotoxic, and intracellular copper levels must be therefore tightly regulated.1 Recently, it has been found that copper metabolism is severely altered in neoplastic diseases. In particular, elevated serum copper concentrations correlate well with tumor burden, progression, and recurrence in a variety of human cancers such as Hodgkin’s lymphoma, sarcoma, cervix, prostate, liver, lung, brain, and breast cancers.2−4 Although the molecular mechanism underlying copper rise in malignant cells remains poorly understood, it looks partially explainable taking into consideration the role that copper plays in tumor angiogenesis, especially at the early stages.5 Notably, copper seems to take part in angiogenesis by stimulating proliferation and migration of human endothelial cells and acting as a cofactor of several angiogenic factors, such as VEGF, bFGF, TNF-a, and IL1.5−8 Furthermore, it has been demonstrated that the human copper transporter hCTR1 could modulate the activation of cell signaling pathways in embryogenic cells, thus affording the development and progression of cancers.9 The altered metabolism of cancer cells and the different response between normal and tumor cell to copper have laid © XXXX American Chemical Society

the basis for the development of copper complexes as anticancer agents. A huge number of Cu(I/II) complexes with different sets of donor ligands have been synthesized and characterized for this purpose, and many of these derivatives display a prominent in vitro cytotoxic activity.10,11 Despite these great efforts, for very few copper complexes the in vitro studies have been translated into preclinical in vivo models.11 For the development of copper-based complexes endowed with antitumor activity, the ligand (framework and donor atom set) plays a crucial role in modulating the hard/soft properties of the metal and the lipophilic/hydrophilic balance of the resulting complexes, thus modifying their solubility as well as their ability to cross cell membranes. Furthermore, the nature of the ligand(s) determines the stability of the complex toward transchelation reactions with biological molecules and toward disproportionation reactions in the case of copper(I) species. In previous studies, we found that stable tetrahedral, heteroleptic copper(I) complexes were obtained by combining the property of scorpionates (either mono-, bi-, or tridentate poly(azolyl)borate and bi- or tridentate poly(pyrazolyl)methane) with the reducing ability of tertiary phosphines.12−21 Mixed-ligand complexes (BpCuP2-type including Received: February 20, 2014

A

dx.doi.org/10.1021/jm500279x | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Scheme 1. Structures of Phosphinesa Used in This Work

a

PCN = tris(cyanoethyl)phosphine; PTA = 1,3,5-triaza-7-phosphaadamantane; thp = tris(hydroxymethyl)phosphine; DAPTA = 3,7-diacetyl-1,3,7triazaphosphabicyclo[3.3.1]nonane; PTA-SO2 = 2-thia-1,3,5-triaza-phosphaadamantane-2,2-dioxide; P(C6H5)3 = triphenylphosphine; and P(pC6H4F)3 = tris(4-fluorophenyl)phosphine.

N,N-bidentate chelate (Bp) and two monodentate phosphines (P), and TpCuP-type including N,N,O- or N,N,N-tridentate scorpionate (Tp) and one monodentate phosphine (P)) have shown promising in vitro antitumor activity. Among them, the neutral TpCuP-type species was revealed to be the most effective compounds.22 On the basis of these, we have further investigated the structure−activity relationship of TpCuP-type species by systematically modifying the nature of the phosphine or the substituents within the tridentate backbone (Schemes 1 and 2). In this study, we report on the synthesis of three series of copper(I) complexes containing tridentate tris(pyrazolyl)borate and tertiary phosphine ligands. The first series includes [HB(pz)3]Cu(PR3)-type complexes 1−7 (Scheme 3) where [HB(pz)3] is the hydrotris(pyrazol-1-yl)borate, [(HB(pz)3)]−,

(Scheme 2) and PR3 are PCN (tris(cyanoethyl)phosphine), PTA (1,3,5-triaza-7-phosphaadamantane), DAPTA (3,7-diacetyl-1,3,7-triazaphosphabicyclo[3.3.1]nonane), PTA-SO2 (2thia-1,3,5-triaza-phosphaadamantane-2,2-dioxide), thp (tris(hydroxymethyl)phosphine), PPh3 (tryphenylphosphine), and P(C 6 H 4F) 3 (tris(p-fluorophenyl)phosphine), respectively (Scheme 1). In the second and third series, we kept unaltered the phosphine (PCN or PTA) and modified the tris(pyrazolyl)borate framework (complexes 8−15) (Scheme 3). All of these complexes have been characterized both in the solid state and in the solution state by means of conventional physicochemical techniques, including multinuclear NMR spectroscopy and Xray crystal structure determination for [HB(3-(NO2)pz)3]Cu(PCN), 10. Relevant structural information concerning the active Cu site of complex [HB(pz)3]Cu(PCN) 1, was collected by X-ray absorption spectroscopy (XAS). The cytotoxicity profile of all derivatives was evaluated in vitro on a panel of human cancer cell lines, some of which were selected for their resistance to cisplatin or for retaining the multidrug-resistant phenotype. On the basis of the in vitro screening, [HB(pz)3]Cu(PCN), 1, and [HB(pz)3]Cu(PTA), 2, were identified as valuable candidates for further biological investigations in order to obtain insight into the mechanism by which they promote the antiproliferative effects. As stated before, despite the huge interest in identifying copper-based compounds that are poorly toxic and highly active as antitumor agents, nowadays there is still a paucity of studies investigating their in vivo antitumor activity. Therefore, we thought to perform in vivo experiments as a step following the in vitro assays. Accordingly, the antitumor activity of the lead compound [(HB(pz)3]Cu(PCN), 1, was evaluated in a murine solid tumor model, the Lewis Lung Carcinoma (LLC) model.

Scheme 2. Structures of Poly(azolyl)boratea Ligands Used in This Work

■

RESULTS Synthesis and Characterization of Copper(I) Complexes. All copper(I) complexes 1−15 were prepared starting from the labile Cu(I) precursor, [Cu(CH3CN)4]BF4, by ligand exchange reactions according to the following general synthetic procedure. To an acetonitrile solution of [Cu(CH3CN)4]BF4, a stoichiometric amount of phosphine was added at room temperature followed by the addition of the salt of the relevant scorpionate after 15−30 min. The order of the addition of the

a [HB(pz)3]− = tris(pyrazol-1-yl)borate; [HB(3-(NO2)pz)3]− = tris(3nitro-pyrazol-1-yl)borate; [(HB(Btz)3]− = tris(benzotriazol-1-yl)borate; [HB(3-(CF3)pz)3]− = tris(3-trifluoromethyl-pyrazol-1-yl)borate; and [HB(3,5-Me2pz)3]− = tris(3,5-dimethylpyrazol-1-yl)borate.

B



Article

Scheme 3. Structures of the Complexes Studied in This Work

of the free ligands and in most cases evidenced a loss of multiplicity. Also phosphines protons were affected by copper coordination which determined signal broadening and/or loss of multiplicity. 31P NMR spectra were much more influenced by copper coordination (31P spectra of representative compounds 1 and 2 are displayed in Figures S3−S4, Supporting Information). Complexes displayed broadened singlet signals (half-line-width in the range 600−2000 Hz) significantly downfield shifted when compared to those observed in uncoordinated phosphines. In detail, the chemical shift variation of the 31P signal (Δδ is defined as the chemical shift difference between the value of coordinated vs uncoordinated phosphine) in complexes 1−7 was detected in the range from 12.4 to 23.2 ppm.26 In complexes 8−15, in PTA-containing and PCN-containing complexes, Δδ depended on the nature of the tridentate coligand. In the case of PCN derivatives, Δδ varied from 10.9 ppm in 10 to 27.3 ppm in 9, reflecting the electron-withdrawing and electron donor properties of the pyrazolyl rings substituents, i.e., NO2 vs CH3 groups, respectively. For PTA derivatives, the range of Δδ values was smaller and varied from 10.0 ppm in 13 to 17.0 ppm in 2. The signal broadening observed in both 1H and 31P NMR spectra of these TpCuP complexes is likely due to the quadrupolar relaxation induced by the 63Cu and 65Cu nuclei (both having I = 3/2) rather than a dynamic behavior in solution. Ligand exchange at room temperature at the NMR time scale was ruled out by the appearance of the free phosphine signal when free PCN was added to the solution of 1 (see Figure S5, Supporting Information). ESI(+)-MS spectra of complexes 1−15 were recorded in acetonitrile or methanol solutions. Representative ESI(+)-MS spectra of complexes 11 and 15 are illustrated in Figures S6−S7 (Supporting Information). Generally, they showed the presence of the parent ion (except for compounds 4, 5, and 6), the occurrence of clusters, preferential loss of the chelate ligand with respect to the phosphines, and, in some cases, formation of oxidized copper species. The presence of the molecular ion peak in most of these complexes suggested that the mixedligand TpCuP configuration conferred a superior stability to these arrangements compared to those exhibited, for instance, in BpCuP2 species.22 Instead, compounds 4 and 5 showed a loss of phosphine with the formation of the oxidized adduct [CuII[HB(pz)3]2 + H]+ at m/z 489 as the most abundant peak.

ligands was crucial, as tridentate boron chelates Tp, if added first, favored the formation of the dimer adduct [Cu(Tp)]2, which in turn disproportionated giving a mixture of metallic copper and the blue-violet bis-substituted [Cu II(Tp) 2] species.23 In the case of PCN ligand, the addition of phosphine first favored the formation of byproducts such as [Cu(PCN)(CH3CN)]+ and [Cu(PCN)2]+,24 and the subsequent addition of the tridentate chelate afforded TpCuP-type compounds. Also, the use of an excess of phosphine ligand invariably led to TpCuP-type compounds.25 Yields varied from 30 to 80%. Generally, mixed-ligand complexes were stable both in the solid and in the solution states with the exception of complex 5 that degraded in air over time and needed to be stored in an inert atmosphere. Compounds 1−15 were soluble in chloroform, acetonitrile, and dimethyl sulfoxide and insoluble in water and diethyl ether. The use of water-soluble ligands such as PTA, DAPTA, or thp did not confer water solubility to the resulting copper complexes. All compounds were white except for nitro-containing derivatives 10 and 14 which were orangeyellow. The identity of copper complexes has been assessed by elemental analysis (C, H, N), multinuclear NMR (1H, 31P) and IR spectroscopies, and high resolution ESI(+) mass spectrometry. Moreover, the crystal structure of the representative complex 10 was confirmed by X-ray analysis, and XAS investigation on the most active complex 1 (vide infra) was performed. The infrared spectra of all compounds showed the expected bands for the tridentate and phosphine ligands. In particular, B−H stretching vibrations slightly shifted to a higher frequency with respect to the free ligand, in the region between 2440 and 2520 cm−1, and CN stretching vibrations of terminal cyano groups of PCN were at 2245−2255 cm−1 (1, 8−11), N− O stretching at 1535−1545 cm−1 and 1373−1382 cm−1 (10 and 14), SO2 stretching at 1374 and 1170 cm−1 (4), and CO stretching at 1636 cm−1 (3). It is worth noting that the largest batochromic shift of B−H stretching frequency with respect to an uncoordinated ligand is observed for compound 1 (38 cm−1). 1 H NMR spectra of complexes 1−15, recorded in DMSO or CDCl3 solution (proton spectra of representative compounds 1 and 2 are displayed in Figures S1−S2, Supporting Information), evidenced a small downfield shift of the signals relative to the pyrazolyl (or benzotriazolyl) protons with respect to the signals C



Article

Compounds 3, 6, 7, 11, and 14 showed adducts obtained via the addition of a phosphine (P) to the parent ion peak ([M + P + Cu]+ and [M + P + Na]+ fragments). Moreover, fragment ions due to the loss of the chelate were detected in complexes 1, 2, 6, 7, 8, 10, and 11 with the formation of [Cu(P)]+ and rearrangements to [Cu(P)2]+ and [Cu(P)(MeCN)]+ adducts. Clusters of the type [2 M + Cu]+ containing three copper(I) ions were detected in complexes 3, 4, and 9. In the latter, such a cluster represented the base peak of the spectrum at m/z 1200. The detection of oxidized species and cluster ions in the ESI(+)-MS spectra depends on the perturbation effects on the analyte solutions due to the application of ESI methods (e.g., oxidative environment generated in ESI(+) conditions and increasing local concentration at the droplet surface during spray phenomena). Instead, as previously reported,27 fragment ions derived from dissociation reactions could be indicative of species present in solution at micromolar concentrations (e.g., coordinatively unsaturated [Cu(P)]+ and [Cu(P)2]+). Only these species might have an impact on the cytotoxic profile of the pertinent complex. The crystal structure of the representative complex [HB(3(NO2)pz)3]Cu(PCN) 10 was determined by single-crystal Xray diffraction analysis. The main details of the crystal structure determination are listed in Table 1, and a selection of bond

Table 2. Selected Bond Lengths (Å) and Angles (Deg) for Complex 10 Cu1−P1 N1−N2 N2−C1 C1−C2 P1−C4 N1−Cu1−P1 Cu1−N1−N2 N1−N2−B1 N2−C1−C2 N1−C3−C2 C3−C2−C1 C4−C5−C6 a

2.2022 1.355 1.354 1.363 1.845 127.1 114.0 121.0 108.9 113.7 102.8 112.8

(7) (2) (2) (2) (2) (1) (1) (1) (1) (1) (1) (2)

Cu1−N1 N1−C3 N2−B1 C2−C3

2.1769 1.329 1.541 1.391

(12) (2) (2) (2)

Cu1−P1−C4 Cu1−N1−C3 N1−N2−C1 N3−C3−N1 C3−N1−N2 P1−C4−C5 C5−C6−N4

118.8 139.8 110.8 120.2 103.7 116.6 178.0

(1) (1) (1) (1) (1) (1) (2)

a

The precision of the data listed in the above table was as follows: four decimal digits for distances referring to Cu; three decimal digits for other distances involving atoms other than H; one decimal digit for bond angles.

Table 1. Crystallographic Data for Complex 10 complex 10 empirical formula formula weight temperature (K) crystal system space group a (Å) b (Å) c (Å) α (deg) β (deg) γ (deg) volume (Å3) Z ρcalc (Mg/m3) μ (mm−1) F(000) crystal size (mm) reflections collected independent reflections (R(int)) data/restraints/parameters goodness-of-fita on F2 final R indexes [I > 2σ (I)] largest diff. peak/hole/e (Å)−3

C18H19BN12O6PCu 604.77 298(2) trigonal P̅3c1 12.4337(2) 12.4337(2) 21.5249(3) 90.00 90.00 120.00 2881.86(7) 4 1.394 0.867 1232 0.32 × 0.24 × 0.18 44 486 2611/0.0301 2611/135/137 1.045 R1b = 0.0362, wR2c = 0.1098 0.972/−0.394

Figure 1. ORTEP full view of the neutral complex 10. Thermal ellipsoids have been drawn at the 30% probability level. The alternate positions of the oxygen atoms and all of the hydrogen atoms have been omitted for clarity.

the tris(3-nitro-pyrazolyl)borate ligand are rare. In the CCDC repository,29 there is only one other report describing a copper compound in which PCN is replaced by PPh3.30 Moreover, also crystallographic reports containing PCN are not common. To the best of our knowledge, the structure of complex 10 is the first example of a tetrahedrally coordinated copper compound exhibiting the combination of PCN with a tris(azolyl)borate ligand. When bound to copper, the scorpionate ligand formed three six-membered heterocycles. All of them had a twisted boat (Cs) pucker, with the N1 and N2 atoms of two adjacent pyrazole rings coplanar within 0.09 Å, while the Cu1 and B1 atoms lay out of the four-nitrogen atom plane, by 1.01 and 0.66 Å, respectively. Because of chelation, each one of the aforementioned four-nitrogen mean planes was almost perpendicular to the opposed pyrazole ring (85.1°), while the mean planes passing through the rings made dihedral angles of 63.0° with each other. All the N−Cu−P bond angles were alike by symmetry, and the constraints imposed by the chelating ligand made the angles wider (127.1(1)°) than expected for an ideal

Goodness-of-fit = [Σ (w (F02 − Fc2)2]/(Nobsevns − Nparams)]1/2, based on all data. bR1 = Σ (|F0| − |Fc|)/Σ |F0|. cwR2 = [Σ[w (F02 − Fc2)2]/ Σ[w (F02)2]]1/2. a

distances and angles is reported in Table 2. Figure 1 reports the ORTEP28 diagram of the full view of the complex. The compound crystallized in a trigonal space group, with the atoms of copper, phosphorus, boron, and the hydrogen bound to boron lying on the ternary axis of symmetry. The environment of the Cu(I) atom, made by the P1 phosphorus and by the N1 nitrogen atoms of three pyrazole rings, was distorted tetrahedral. It is worth noting that crystal structures containing D



Article

tetrahedron. By looking down through the Cu−P axis, the three cyanoethyl arms of the PCN ligand and the three Cu−N bonds appeared almost perfectly staggered, the torsional angle N1− Cu1−P1−C4 being 53.1°. The dihedrals mainly responsible for the orientation of the cyanoethyl arms Cu1−P1−C4−C5 and P1−C4−C5−C6 were −51.1° and 69.9°, respectively. Thus, they both showed a synclinal arrangement instead of the expected anti-periplanar one. This orientation put the cyanoethyl arms rather close to the nitro groups of the pyrazole rings. The Cu1−P1 and Cu1−N1 bond lengths were 2.202(1) and 2.177(1) Å, respectively. A search in the CCDC database for tetra-coordinated copper complexes showing the P1N3 donor set returned slightly more than 100 structures, in which the calculated mean values for the Cu−P and Cu−N bond lengths were 2.190 and 2.092 Å, respectively. Both parameters in complex 10 were longer than the pertinent mean values, and the departure was more appreciable for the Cu−N bond (+0.085 Å). By limiting the comparison to tetra-coordinated copper tris(azolyl)borato complexes (five entries),30−33 we observed that the deviation from the mean bond lengths (Cu− P, 2.164 Å; Cu−N, 2.077 Å) increased in both cases (+0.038 Å for Cu−P and +0.100 Å for Cu−N). The closest similarity with our compound was displayed by [HB(3-(NO2)pz)3]Cu(PPh3).30 Bond distances and angles in the pyrazolyl rings were nearly identical to those of 10, while the Cu−P and Cu− N distances were 2.193 and 2.142 Å (mean), respectively. The elongation of the Cu−N bond in 10 was due to the particular “umbrella” orientation assumed by the cyanoethyl arms of PCN. Comparably long Cu−N bonds were found in complexes featuring at the same time bulky phosphine ligands and 3substituted tris-azolyl ligands.34−36 In all these compounds, the Cu−P bond lengths were about 2.2 Å, as in 10. The inspection of the packing diagram allowed one to appreciate the position of the large void mentioned in the Experimental Section that runs along the crystallographic c axis. As for the intermolecular contacts, the packing diagram highlights a couple of rather loose approaches (H-acceptor distance of about 2.64 and 2.72 Å). The first one involved the O2a and the H1 atom of a molecule at y, 1− x + y, and 1− z, with an angle O2a−H1−C1 of 118.4°; the second one engaged the O1 oxygen and the H4a atom of a molecule at 1− x, 1− x + y, and 1/2 − z, with an angle O1−H4a−C4 of 136.9°. The tetrahedral N3P-coordination sphere outlined for complex 10 was confirmed by the local atomic environment of the metal found in complex 1 by means of X-ray absorption spectroscopy (XAS).37−40 The normalized XANES spectra of complex 1 are reported in Figure 2a. XANES spectra of Cu complexes displayed a typical edge feature which was assigned to the 1s-4p electronic transition.41 The observed position at 8982.7 eV fell among the values observed for Cu(I) complexes with a similar environment.38,41 In addition, the intensity of such a transition was reported to depend on the local coordination geometry around the Cu(I) site.41 Here, the observed value for complex 1 was not indicative for a digonal or trigonal geometry, but it matched well within the values relative to 4-coordinated Cu(I) complexes.41,42 More precise structural information for 1 can be obtained by the analysis of the extended X-ray absorption spectrum. Therefore, an EXAFS analysis has been conducted by using the atomic coordinates available for complex 10 as the starting structural model. The local atomic environment of Cu was postulated to be 3 + 1; i.e., three N atoms from the Tp chelate

Figure 2. XAS data for complex 1. (a) Normalized XANES spectrum which displays the 1s-4p edge transition, peak P. (b) Best fit of the Cu K-edge EXAFS signal. The figure shows the comparison of the experimental (−) and theoretical (·) k3-extracted EXAFS signals and (c) comparison of the theoretical (·) and experimental (−) Fourier transform signals relative to b.

plus 1 P of PCN (details are available in the Supporting Informations). The outcome of the fitting procedure is displayed in Figure 2b and c. Specifically, Figure 2b indicates the comparison of the k3-extracted EXAFS signal (solid line) with the theoretical one (dotted line), whereas the corresponding Fourier transform (FTs) signal is displayed in Figure 2, panel c. Theoretical curves matched well with the experimental ones in both panels confirming the 4-fold coordination of the Cu site in 1. The best fit of the interatomic distances for the three equivalent Cu−N interactions and the Cu−P one were 2.069(4) Å and 2.135(5) Å, respectively. Their corresponding EXAFS bond variance was 0.009(3) and 0.005(2) Å2 for the Cu−N and Cu−P, respectively. These values were in very good agreement with those reported for complex [HB(pz)3]Cu(PPh3) that contains the same [HB(pz)3]− ligand (Cu−N = 2.076(6))32 but lower than those observed in complex 10 (see Table 2). The elongation of the Cu−N bond distances in 10 is likely due to the electron withdrawing effect provided by the nitro groups in the chelate [HB(3-(NO2)pz)3]− ligand.

■

CYTOTOXICITY AGAINST CULTURED CANCER CELLS The cytotoxic properties of all copper compounds and of the corresponding uncoordinated ligands were evaluated against a panel of human tumor cell lines containing examples of breast (MCF-7), cervical (A431), colon (HCT-15), pancreatic (BxPC3), and lung (A549) cancers, neuroblastoma (SHSY5Y), and melanoma (A375). For comparison purposes, the cytotoxicity of cisplatin, the most widely used metal-based E



Article

Table 3. In Vitro Antitumor Activitya IC50 (μM) ± SD compd

MCF-7

A431

HCT-15

A375

BxPC3

SH-SY5Y

A549

[HB(pz)3]Cu(PCN) (1) [HB(pz)3]Cu(PTA) (2) [HB(pz)3]Cu(DAPTA) (3) [(HB(pz)3)Cu(PTA-SO2)] (4) [HB(pz)3]Cu(thp) (5) [HB(pz)3]Cu(PPh3) (6) [HB(pz)3]Cu[P(p-C6H4F)3] (7) [HB(3-(CF3)pz)3]Cu(PCN) (8) [HB(3,5-Me2pz)3]Cu(PCN) (9) [HB(3-(NO2)pz)3]Cu(PCN) (10) [HB(Btz)3]Cu(PCN) (11) [HB(3-(CF3)pz)3]Cu(PTA) (12) [HB(3,5-Me2pz)3]Cu(PTA) (13) [HB(3-(NO2)pz)3]Cu(PTA) (14) [HB(Btz)3]Cu(PTA) (15) PPh3 PTA PCN PTA-SO2 DAPTA [P(p-C6H4F)3] Na[HB(pz)3] Na[HB(3-(CF3)pz)3] K[HB(3,5-Me2pz)3] Na[HB(Btz)3] Na[HB(3-(NO2)pz)3] cisplatin

0.73 ± 0.17 0.79 ± 0.38 9.59 ± 2.23 2.12 ± 1.06 ND 3.01 ± 0.23 3.23 ± 1.32 3.61 ± 1.41 8.02 ± 2.36 13.23 ± 2.52 19.25 ± 1.96 4.45 ± 1.02 6.98 ± 1.52 17.33 ± 2.36 11.12 ± 2.65 34.18 ± 3.35 >100 >100 69.46 ± 3.35 85.32 ± 3.95 63.32 ± 3.99 >100 67.39 ± 2.73 63.32 ± 2.39 >100 49.31 ± 3.24 7.60 ± 0.21

0.41 ± 0.18 1.88 ± 1.7 6.65 ± 1.7 1.00 ± 1.16 ND 1.98 ± 1.11 2.34 ± 1.02 7.92 ± 1.06 9.27 ± 1.35 5.43 ± 1.75 11.12 ± 2.34 15.02 ± 1.98 9.83 ± 2.09 8.28 ± 2.24 10.25 ± 3.14 47.53 ± 1.14 >100 >100 61.12 ± 4.36 >100 56.04 ± 3.46 40.44 ± 1.44 46.47 ± 3.27 42.26 ± 1.98 >100 66.09 ± 2.98 1.65 ± 0.51

0.99 ± 0.54 2.12 ± 1.13 11.83 ± 2.26 2.01 ± 1.05 ND 2.99 ± 1.54 0.99 ± 0.36 4.25 ± 1.61 10.15 ± 1.57 11.24 ± 1.64 12.56 ± 2.45 6.48 ± 2.98 8.64 ± 1.74 23.78 ± 2.11 4.65 ± 1.76 54.13 ± 4.25 >100 >100 >100 67.12 ± 3.14 >100 77.58 ± 1.97 58.72 ± 2.25 58.79 ± 2.19 >100 >100 16.65 ± 2.63

0.65 ± 0.23 1.85 ± 1.82 7.57 ± 2.11 2.25 ± 0.47 ND 3.17 ± 1.07 2.97 ± 0.54 8.14 ± 1.32 5.63 ± 1.25 8.21 ± 1.79 9.64 ± 238 8.47 ± 2.22 7.02 ± 1.51 17.45 ± 3.11 12.72 ± 2.96 52.31 ± 2.46 >100 >100 >100 >100 86.11 ± 4.86 66.87 ± 2.34 >100 68.27 ± 3.84 >100 69.63 ± 2.12 3.11 ± 0.98

0.62 ± 0.46 2.24 ± 1.1 7.64 ± 1.41 1.43 ± 0.97 ND 1.45 ± 1.73 0.76 ± 0.37 7.96 ± 1.87 5.74 ± 2.15 6.47 ± 2.38 16.27 ± 4.09 7.46 ± 2.17 9.53 ± 2.35 7.53 ± 2.81 8.36 ± 2.11 62.24 ± 2.17 >100 >100 >100 89.53 ± 4.13 >100 75.37 ± 3.74 >100 73.35 ± 5.01 >100 44.48 ± 2.84 10.17 ± 1.65

0.77 ± 0.83 1.45 ± 1.71 8.46 ± 2.28 2.26 ± 0.75 ND 3.13 ± 1.96 1.12 ± 0.42 11.34 ± 1.46 5.36 ± 1.19 7.43 ± 1.65 14.72 ± 3.25 11.14 ± 2.13 12.66 ± 1.99 8.41 ± 3.07 14.27 ± 3.93 41.24 ± 2.94 >100 >100 >100 >100 >100 97.43 ± 2.16 >100 98.48 ± 3.33 >100 59.16 ± 4.19 5.36 ± 1.74

0.84 ± 0.25 1.49 ± 1.91 13.13 ± 1.25 3.65 ± 1.02 ND 3.17 ± 0.48 2.22 ± 1.31 6.54 ± 2.23 10.08 ± 2.54 8.19 ± 2.11 15.56 ± 2.28 7.77 ± 2.12 8.32 ± 1.22 22.51 ± 2.25 19.76 ± 1.71 30.12 ± 2.13 >100 >100 >100 >100 56.45 ± 4.42 >100 >100 >100 >100 >100 12.64 ± 0.81

Cells (3−8 × 104·mL−1) were treated for 72 h with increasing concentrations of tested compounds. Cytotoxicity was assessed by the MTT test. IC50 values were calculated by a four parameter logistic model (P < 0.05). SD = standard deviation. ND: not determined.

a

Table 4. Cross-Resistance Profilesa IC50 (μM) ± SD compd [HB(pz)3]Cu(PCN) (1) [HB(pz)3]Cu(PTA) (2) [HB(pz)3]Cu(DAPTA) (3) [HB(pz)3]Cu(PTA-SO2) (4) [HB(pz)3]Cu(PPh3) (6) [HB(pz)3]Cu[P(p-C6H4F)3] (7) [HB(3-(CF3)pz)3]Cu(PCN) (8) [HB(3,5-Me2pz)3]Cu(PCN) (9) [HB(3-(NO2)pz)3]Cu(PCN) (10) [HB(Btz)3]Cu(PCN) (11) [HB(3-(CF3)pz)3]Cu(PTA) (12) [HB(3,5-Me2pz)3]Cu(PTA) (13) [HB(3-(NO2)pz)3]Cu(PTA) (14) [HB(Btz)3]Cu(PTA) (15) cisplatin doxorubicin

2008

C13*

R.F.

LoVo

LoVo MDR

R.F.

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.39 ± 0.11 1.45 ± 0.6 8.86 ± 1.3 1.95 ± 1.63 3.92 ± 1.22 2.25 ± 0.95 2.21 ± 0.62 3.23 ± 1.01 8.46 ± 2.26 7.84 ± 1.32 2.89 ± 0.92 3.52 ± 1.26 6.88 ± 0.92 3.77 ± 0.94 22.77 ± 2.01

1.3 1.4 1.0 0.7 1.2 1.0 1.0 0.9 1.1 0.9 0.9 1.1 1.3 1.1 10.2

1.41 ± 0.56 2.46 ± 1.01 7.35 ± 2.14 2.25 ± 0.74 3.28 ± 0.85 3.54 ± 1.11 6.14 ± 1.73 4.88 ± 1.98 11.26 ± 3.16 11.73 ± 2.43 5.46 ± 1.31 5.32 ± 1.54 7.58 ± 3.14 4.43 ± 0.93

0.84 ± 0.31 2.11 ± 0.93 8.65 ± 1.25 1.62 ± 1.03 2.14 ± 1.15 2.65 ± 0.87 6.03 ± 1.33 5.05 ± 1.12 12.53 ± 2.98 12.63 ± 1.64 5.44 ± 1.65 4.43 ± 1.75 8.85 ± 2.95 5.15 ± 1.09

1.3 1.4 1.2 0.7 0.6 0.7 1.0 1.0 1.1 1.1 0.9 0.8 1.1 1.2

1.11 ± 0.86

19.21 ± 2.37

17.3

0.29 1.05 9.11 2.65 3.21 2.16 2.11 3.37 7.36 8.65 3.26 3.04 5.47 3.54 2.22

0.09 0.76 1.15 1.36 2.11 0.98 0.96 0.97 3.14 2.23 1.05 1.23 2.13 1.51 1.03

Cells (3 × 104·mL−1) were treated for 72 h with increasing concentrations of tested compounds. Cytotoxicity was assessed by the MTT test. IC50 values were calculated by a four parameter logistic model (P < 0.05). SD = standard deviation. RF = IC50 resistant/IC50 sensitive. a

over all cell lines (average IC50 values over 80 μM). PTA, PTASO2, DAPTA, PCN, and [P(p-C6H4F)3]phosphine ligands elicited a very low cytotoxic activity (mean IC50 values over 80 μM), whereas the lipophilic phosphine PPh3 showed average IC50 values of 46 μM. Results obtained after treatment with

anticancer drug, was evaluated under the same experimental conditions. IC50 values, calculated from the dose−survival curves (for examples, see Figure S9, Supporting Information) obtained after 72 h of drug treatment from the MTT test, are shown in Table 3. Uncoordinated tris(pyrazolyl)borate ligands proved to be scarcely effective in decreasing cancer cell viability F



Article

complex [HB(pz)3]Cu(thp) 5 were not reproducible, thus precluding statistically significant IC50 calculations. In the series of [HB(pz)3]Cu(PR3) complexes 1−7 holding the unsubstituted tris(pyrazolyl)borate ligand, the cytotoxic activity was modulated by the nature of the pertinent phosphine. [HB(pz)3]Cu(PCN), 1, emerged as the most promising derivative, eliciting mean IC50 values about 1 order of magnitude lower than those of cisplatin (IC50 of 0.7 μM and 8.2 μM for 1 and cisplatin, respectively). In A549 lung and HCT-15 colon cancer cells, 1 afforded antiproliferative effects up to 15 and 17 times superior than those of cisplatin, respectively. Complexes 2, 4, 6, and 7 yielded mean IC50 values from 3 to 5 times lower than those of the reference metallodrug. As a unique exception, [HB(pz)3]Cu(DAPTA), 3, showed a cytotoxic activity with IC50 ranging from 6.65 μM, against human cervix A431 cells, to 13.13 μM, against human lung A549 cells, slightly higher than those elicited by cisplatin. The introduction of both electron-donor (Me) or electronacceptor (NO2 and CF3) groups in the tris(pyrazolyl)borate moiety significantly weakened the cytotoxic profile of the resulting copper(I) species. In detail, in PCN-containing derivatives 8−10, we observed an 11-fold reduction of the in vitro antitumor activity (average IC50 values of 7.1, 7.7, and 8.5 μM compared to 0.7 μM of 1). Analogously, in PTA-containing complexes 12−14, we detected IC50 values up to 9 times higher than those exhibited by 2. Similarly, substitution of the pyrazolyl rings with the bulkier benzotriazolyl ones (complexes 11 and 15) diminished about 20 and 7 times the cytotoxic activity compared to that of 1 and 2, respectively. All copper derivatives have been additionally tested for their in vitro antitumor activity on two human cell line pairs which have been selected for sensitivity/resistance to cisplatin (ovarian cancer cells 2008/C13*) or belonging to the multidrug-resistant (MDR) phenotype (colon cancer cells LoVo/LoVo MDR). Cytotoxicity on sensitive and resistant cells was assessed after 72 h of drug treatment by an MTT test. Table 4 shows the cytotoxicity parameters, in terms of IC50 and the resistance factor (RF), the latter defined as the ratio between IC50 values calculated for resistant cells and those obtained with sensitive ones. All derivatives demonstrated a similar cytotoxic profile both in cisplatin-sensitive and -resistant cell lines. For all copper(I) derivatives, the RFs calculated on 2008/C13* cells were from 6- to 10-fold lower than that of cisplatin, indicating the absence of cross-resistance phenomena. Analogously, on colon cancer LoVo/LoVo MDR cells, all derivatives yielded RFs on average 30-fold lower than that obtained with doxorubicin, a drug belonging to the MDR spectrum, thus suggesting that these agents are not potential MDR substrates. Once again, complex 1 emerged as the most effective derivative, with IC50 values ranging from 0.29 to 1.41 μM against both sensitive and resistant cancer cells. On the basis of the overall in vitro cytotoxicity data, complexes 1 and 2 were selected for further biological evaluations. Cytotoxicity against Cultured Nontumor Cells. The cytotoxicity of complexes 1 and 2 was also evaluated against nontumor cells in rapid proliferation: the human embryonic kidney HEK293 cells (Figure 3). Both copper(I) complexes elicited a cytotoxic activity roughly 4 times higher than that recorded after cisplatin treatment, with IC50 values in the lowmicromolar range. The selectivity index values (SI = quotient of the average IC50 toward normal cells divided by the average IC50 for the malignant cells) calculated for complexes 1 and 2

Figure 3. Cytotoxicity against nontumor HEK293 cells. Cells (5 × 104) were treated for 72 h with increasing concentrations of 1 (○), 2 (▼) or cisplatin (●). The cytotoxicity was assessed by the MTT test. IC50 values were calculated by 4-PL (P < 0.05). The inset shows SI values (the quotient of the average IC50 toward normal cells divided by the average IC50 for the malignant cells). Error bars represent standard deviation.

were approximately 3 times higher than that calculated with cisplatin, attesting to a preferential cytotoxicity of copper(I) complexes versus neoplastic cells (inset of Figure 3). Inhibition of 26S Proteasome. It has been previously demonstrated that some copper(I) complexes exerted their antiproliferative activity by inhibiting proteolytic proteasome activities, inducing endoplasmic reticulum (ER) stress and triggering a nonapoptotic mechanism of programmed cell death (PCD), termed paraptosis.43,44 On the basis of these, the ability of 1 and 2 to hamper the functioning of each individual proteasome active sites, chymotrypsin-like (CT-L), trypsin-like (T-L), and caspase-like (C-L) activities were assessed in human ovarian 2008 cancer cells treated with increasing concentrations of copper drugs for 24 h. Lactacystin, an irreversible nonpeptidomimetic proteasome inhibitor, was used as a positive control. Copper(I) complexes inhibited all the proteolytic (CT-L, T-L and C-L) activities with IC50 values (μM) of 23.2, 58.5, and 27.7 for 1 and of 34.6, 61.3, and 33.8 for 2, respectively (Figure 4, panel A). Although less effective than lactacystin in inhibiting CT-L activity (IC50 of 23.2 μM, 34.6 μM, and 1.8 μM for 1, 2, and lactacystin, respectively), both copper(I) species exerted a superior hindering ability on T-L and C-L activities, the latter being roughly 2 times more inhibited. Coherently with the inhibition of the 26S proteasome activities, Western blot analyses demonstrated a time-dependent accumulation of poly ubiquitinated proteins in 2008 cells treated for 24 and 48 h with IC50 doses of 1 and 2 (Figure 4, panel B). Endoplasmic Reticulum (ER) Stress and Unfolded Protein Response (UPR) Induction. Since the proteasome inhibition and the accumulation of ubiquitinated proteins in the ER lumen provoke an ER stress response that ultimately leads to UPR activation, the phosphorylation level of two well-known protein markers (PERK and IRE1) involved in UPR were also assayed by Western blot analyses. As depicted in Figure 4, panel C, high levels of phosphorylated ER stress-related proteins were detected in 2008 cells treated with both copper(I) complexes at the scheduled treatments. G



Article

Figure 4. Proteasome inhibition and UPR induction. (A) Inhibition of the human 26S proteasome. 2008 cells were treated for 24 h with increasing concentrations of 1, 2, or lactacystin. Proteasome activity was measured by means of specific fluorogenic peptide substrates. IC50 values were calculated by a four-parameter logistic model (P < 0.05). Error bars represent standard deviation. (B) Detection of poly ubiquitinated proteins. 2008 cells were treated for 24 and 48 h with IC50 of 1 (1.5 and 0.5 μM) or 2 (4.5 and 1.5 μM) and ubiquitinated proteins estimated by Western blotting. (C) Detection of p-PERK and p-IRE1. 2008 cells were treated with IC50 of 1 (3 and 1.5 μM) or 2 (10 and 4.5 μM) for 12 and 24 h. p-PERK and pIRE1 levels were determined by Western blot analysis.

Figure 5. Morphological changes. (A) Phase-contrast microphotographs (100×) of 2008 cells. (a) Control cells. (b and c) Cells treated for 24 h with IC50 of 1 (1.5 μM) or 2 (4.5 μM), respectively. (B) Hoechst staining. 2008 cells were incubated for 48 h with IC50 of 1 (0.5 μM, panel b), 2 (1.5 μM, panel c), and cisplatin (panel d). Panel a represents the control untreated 2008 cells.

Cell Death Mechanism. Morphological analysis (cell vacuolization and nuclear condensation) and molecular cell death determinant analysis (caspase-3/7 and −9 activity, DNA fragmentation and AIP-1/Alix protein levels) were assessed in complexes 1- and 2-treated human ovarian 2008 cells. In these cells, 1 and 2 determined a massive cytoplasmic vacuolization (Figure 5, panel A). Interestingly, autophagy inhibitors, 3-MA (3-methyladenine) and monensin, did not abrogate cellular cytoplasmatic vacuolization induced by 1 and 2, and vacuoles

were not positive to monodansylcadaverine staining (data not shown), thus confirming the nonautophagic nature of copperinduced vacuoles. Furthermore, 1- and 2-treated cells stained with Hoechst 33258 did not show nuclear DNA condensation or apoptotic body formation, classical hallmarks of apoptosis (Figure 5, panel B, b and c). On the contrary, cisplatin determined a net increase of cells with condensed DNA (Figure 5, panel B, d). The lack of DNA fragmentation was also detected either by ELISA (Figure 6, panel A) and H



Article

Figure 6. (A) Nuclear DNA fragmentation. 2008 cells were treated for 12 or 24 h with IC50 of 1 (3 and 1.5 μM), 2 (10 and 4.5 μM), and cisplatin (27 and 12 μM). Quantitative estimation of DNA fragmentation was obtained with an ELISA test. Error bars represent standard deviation. (B) FACS analysis. Forward scattering vs side scattering of 2008 cells untreated and treated for 24 h with IC50 of 1 (1.5 μM). (C) Caspase activity. 2008 cells were incubated for 48 h with 1 (0.5 μM), 2 (1.5 μM), or staurosporine (3 μM) and processed for caspase-3/-7 and -9 activity. Error bars indicate standard deviation. (D) AIP-1/Alix levels. 2008 cells were treated with IC50 concentrations of 1 (1.5 and 0.5 μM) or 2 (4.5 and 1.5 μM) for 24 and 48 h. The amount of AIP-1/Alix protein was detected by Western blot analysis.

In Vivo Antitumor Activity. The most promising complex [HB(pz)3]Cu(PCN), 1, was evaluated for its potential antitumor effects in a model of solid tumor, the murine LLC. Tumor growth inhibition induced by the novel copper(I) complex was compared with that promoted by the reference metallodrug cisplatin. Once the tumor was visible (7 days after tumor inoculation), tumor-bearing mice were randomized into vehicle control and treatment groups (7 mice per group). Control mice received the vehicle (50% Cremophor EL (v/v) and 50% PEG400 (v/v)), whereas treated groups received daily doses of 1 (15 and 25 mg·kg−1) or cisplatin (1.5 mg·kg−1). Tumor growth was estimated at day 15, and the results are summarized in Table 6. For the assessment of the adverse side

cytofluorimetric techniques (Table 5 and Figure 6, panel B). The absence of nucleosome formation and of sub-G1 cell Table 5. Cell Cycle Analysesa sub-G1 G1 S/G2/M

control

1

2

2.10 60.87 33.22

2.31 48.25 47.77

3.75 47.34 46.37

a

Percentages of 2008 cells in different cell cycle phases after treatment for 24 h with IC50 doses of complex 1 or 2 vs control cells.

population following exposure to copper derivatives confirmed the induction of a cell death process different from apoptosis. Cytofluorimetric analyses (Figure 6, panel B) highlighted the presence of an increase of the ultrastructural cellular complexity and size (shift of cell population on the upper right quadrant of the cytogram), thus confirming the ability of complex 1 to cause extensive cytosolic vacuolization. The activity of the initiator caspase-9 and of the effector caspase-3/-7 was determined by monitoring fluorimetrically the release of free AMC from AMC-conjugated caspase-specific peptide substrates. Staurosporine, an alkaloid inducing apoptosis through caspase-3/7 and -9 activation, and z-VADfmk, a pan caspase inhibitor, were used as positive and negative controls, respectively. As shown in Figure 6, panel C, staurosporine determined a substantial activation of caspases with respect to control cells, and cotreatment with z-VAD-fmk significantly decreased cleavage of both caspases. Conversely, no increase in AMC fluorescence has been detected after treatment for 48 h with IC50 doses of 1 and 2, and cotreatment with the pan caspase inhibitor did not change caspase activity profiles. In addition, treatment with 1 and 2 determined a timedependent decrease of AIP-1/Alix protein, a known inhibitor protein of paraptosis45,46 (Figure 6, panel D).

Table 6. In Vivo Anticancer Activity of Complex 1 toward LLCa compd

daily dose (mg·kg−1)

c

control (1) (1) cisplatin

15 25 1.5

average tumor weight (g) (mean ± SD) 0.491 0.254 0.106 0.107

± ± ± ±

0.07 0.06**b 0.09**b 0.07**b

inhibition of tumor growth (%) 48.26 78.20 76.60

a

Starting from day 7 after tumor implantation, the tested compounds were daily administered intraperitoneally (ip). At day 15, mice were sacrificed, and tumor growth was detected as described in the Experimental Section. The Tukey−Kramer test was performed. b**p < 0.01. cVehicle (50% Cremophor EL (v/v) and 50% PEG400 (v/v)).

effects, changes in the body weights of tumor-bearing mice were daily monitored (Figure 7). Chemotherapy with 1 at the lowest concentration induced a 48% reduction of tumor mass compared to that in the control group. Following the administration of 1 at 25 mg·kg−1, tumor mass was reduced over 78%. Remarkably, the time course of body weight changes I



Article

cytotoxic potential of this class of compounds. Neutral TpCuPtype copper(I) complexes were straightforwardly synthesized by the addition of a stoichiometric amount of both ligands onto the labile [Cu(MeCN)4]BF4 precursor in MeCN solutions. As outlined in Figure 1 for the representative complex 10, the molecular structure adopted by this class of TpCuP-type compounds is distorted tetrahedral, the scorpionate ligand acting as tripodal chelate via three Npz donors with the phosphine P completing the coordination sphere of the metal. Such a structure was retained in all of the complexes in the solution state at elevated concentration (ca. 5.10−2 M detected via 1H NMR) and, in most cases (complexes 1−3 and 7−15), also at micromolar concentration (ca. 5.10−5 M detected via ESI(+)-MS). The cytotoxic activity of complexes 1−15 exceeded those of the corresponding ligands by at least 1 order of magnitude, indicating that the coordination to copper ion led to an increase of antitumor efficacy. Considering complexes 1−7, in which the structural modification concerns the phosphine ligand, they exhibited significant cytotoxic activities (in the low to submicromolar range) when holding phosphines of moderate hydrophilicity (PCN, PTA, and PTA-SO2, with log P in the −0.18/−1.11 window; see Table S1, Supporting Information). The coordination of more hydrophilic phosphines (DAPTA, log P = −1.44) or lipophilic phosphines (PPh3 and p-C6H4F, log P = 5.34 and 5.64, respectively) resulted in moderate cytotoxic activity. In this series, complexes [HB(pz)3]Cu(PCN), 1, and [HB(pz)3]Cu(PTA), 2, elicited the most promising antiproliferative activity with mean IC50 values of 0.7 and 1.7 μM, respectively. In order to shed light on the structure−activity relationships among copper(I) analogues, PCN- and PTA-containing complexes 1 and 2 were additionally subjected to modifications of the tridentate ligand. The introduction of electron-donating methyl groups (complexes 9 and 13) or electron-withdrawing nitro and trifluoromethyl groups (complexes 8, 10, 12, and 14) in the pyrazolyl backbones or the replacement of pyrazolyl with benzotriazolyl condensed rings (complexes 11 and 15) caused a substantial decrease of the in vitro antitumor activity that was however comparable in five out of eight cases with that shown by cisplatin. All synthesized TpCuP-type complexes were able to circumvent clinically relevant models of drug resistance, namely, cisplatin and multidrug resistances, and elicited a preferential antiproliferative activity against neoplastic cells. These features confirm a peculiar behavior of phosphinecontaining copper(I) complexes and are consistent with the hypothesis of a mechanism of action different from Pt(II) metallodrugs. Therefore, a major part of the experimental studies was directed to the elucidation of the mechanism by which phosphine copper(I) complexes promoted cancer cell killing effects. Attention has been mainly focused on complexes 1 and 2 due to their promising in vitro antitumor efficacies. Experimental data collected with different morphological and biochemical analyses in human ovarian cancer cells indicated the inhibition of the 26S proteasome and ER stress/UPR activation as the major mechanistic features of cancer cell death. These results, besides confirming the 26S proteasome as a putative molecular target for copper complexes,11,47 pointed out the capability of these copper(I) species to interact with all three catalytic peptidase activities of 26S. Although a primary role for the CT-L site has been recognized, the relative importance of the different catalytically active sites of proteasome in cytosolic protein degradation has not been

Figure 7. Body weight changes. The body weight changes of LLCbearing C57BL mice treated with vehicle or tested compounds. Each drug was administered daily after 7 days from the tumor cell inoculum. Weights were measured at day 1 and daily from day 7. Error bars indicate the standard deviation.

indicated that treatment with 1 did not induce significant body weight loss (

In Vitro and in Vivo Anticancer Activity of Copper(I) Complexes with

Recommend Documents