Antiproliferative effects of ruthenium-based nucleolipidic nanoaggregates in human models of breast cancer in vitro: insights into their mode of action

Looking for new metal-based anticancer treatments, in recent years many ruthenium complexes have been proposed as effective and safe potential drugs. In this context we have recently developed a novel approach for the in vivo delivery of Ru(III) complexes, preparing stable ruthenium-based nucleolipidic nanoaggregates endowed with significant antiproliferative activity. Herein we describe the cellular response to our ruthenium-containing formulations in selected models of human breast cancer. By in vitro bioscreens in the context of preclinical studies, we have focused on their ability to inhibit breast cancer cell proliferation by the activation of the intrinsic apoptotic pathway, possibly via mitochondrial perturbations involving Bcl-2 family members and predisposing to programmed cell death. In addition, the most efficient ruthenium-containing cationic nanoaggregates we have hitherto developed are able to elicit both extrinsic and intrinsic apoptosis, as well as autophagy. To limit chemoresistance and counteract uncontrolled proliferation, multiple cell death pathways activation by metal-based chemotherapeutics is a challenging, yet very promising strategy for targeted therapy development in aggressive cancer diseases, such as triple-negative breast cancer with limited treatment options. These outcomes provide valuable, original knowledge on ruthenium-based candidate drugs and new insights for future optimized cancer treatment protocols.

low toxicity profiles, have been recently identified and intensively studied 9,10 . Many ruthenium complexes have in fact shown selective bioactivity, as well as the ability to overcome the resistance encountered with platinum-based drugs, ranking them as strong antitumoral candidates in a rational drug discovery approach 11 . Among these, ruthenium(III) compounds behave as valid prodrugs with somehow limited side-effects 11 , being likely activated to the more reactive and cytotoxic ruthenium(II) derivatives within the reducing microenvironment of solid tumours 12 . Along with ligands release and/or substitution -which occurs rapidly under physiological conditions in vitro and in vivo -the biological reduction of ruthenium(III) complexes is a possible process, especially in high proliferating cells, thereby promoting a unique activation process of this kind of metal-based drug in tumour tissues 12 . Nevertheless, in the general uncertainty concerning their mechanism of action, it cannot be excluded that Ru(III) complexes, properly transported into tumour cells, can interact in their original redox status with potential molecular targets. Furthermore, the possibility to be transported by the transferrin/transferrin receptor (Tf/TfR) network in place of iron, might allow for a natural ruthenium accumulation within cancer cells, typically requiring high iron amounts to accommodate for their rapid proliferation 13,14 .
Despite the positive outcomes throughout advanced preclinical and clinical evaluations of the anticancer ruthenium(III)-based compounds NAMI-A 15 -the most advanced candidate drug having completed Phase II clinical trials -and KP1019 16 (NKP1339, the sodium salt of KP1019, is described to be ready for clinical appliance 17 ), various drawbacks have been observed, mainly related to their limited stability in physiological conditions, impairing both their general pharmacokinetic and pharmacodynamic profiles 18 . In addition to interfering with the internalization process into targeted cancer cells, degradation events can also limit the therapeutic effectiveness, thereby becoming a central issue that has claimed a reconsideration of the efficacy of low molecular weight ruthenium complexes currently in preclinical or clinical studies 19 . Additionally, the lack of convincing evidences on their biological mode of action, as well as on targets identification, has significantly limited their further development 20,21 . Aiming at improving the stability of the ruthenium(III)-based drugs in biological environments, as well as their suitability for biomedical applications, we have recently proposed an innovative strategy for their transport in vivo incorporating them within liposomial aggregates. Starting from a core molecular scaffold based on a ruthenium complex inspired to NAMI-A, named AziRu showing per se higher in vitro cytotoxicity than NAMI-A 22,23 , we have developed a mini-library of highly functionalized amphiphilic ruthenium(III) complexes, including differently decorated nucleolipids (see Fig. 1). The resulting nucleolipidic Ru(III) complexes (named ToThyRu, HoThyRu and DoHuRu) are endowed with a marked propensity for self-aggregation in aqueous solutions and high in vitro antiproliferative activity against human cancer cells, thus proving to be promising lead compounds for future in vivo studies. Moreover, on the lookout for biomimetic nanosystems, stable nucleolipidic Ru(III) complexes were produced by co-aggregation with either the zwitterionic phospholipid POPC (1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine) or the cationic lipid DOTAP (1,2-dioleoyl-3-tr imethylammoniumpropane chloride). The final nanoaggregates -which exhibit liposomial structure and have been subjected to an in-depth microstructural characterization -are ad hoc designed to present high stability under physiological conditions as well as to transport high ruthenium amounts in cells, thereby ensuring more effective metal-based treatments [24][25][26][27] . Indeed, recent in vitro bioactivity investigations have shown the great potential of our biocompatible ruthenium-containing liposomes as active antineoplastic agents against human carcinoma cells of different histological origin 24 . Of special interest have been the results obtained on estrogen and progesterone receptor positive MCF-7 adenocarcinoma cells, an in vitro model reflecting to a large extent the features of luminal breast cancer cells in vivo, worldwide the most common invasive cancer in women. Most remarkably, liposomes loaded with the Ru(III)-containing nucleolipids have shown a promising selectivity towards cancer cells in bioscreens in vitro 19,22,[25][26][27] . To expand knowledge on these compounds so to optimally develop ruthenium-based candidate drugs, mechanisms including interaction with DNA and induction of apoptosis have been investigated, but little has been hitherto explored on their overall modes of action as well as on biomolecular targeting 14 . Thus, in continuation with our previous studies, we have herein investigated in-depth the anticancer profile and mode of action of Ru(III)-containing liposomes. In particular, by focusing on well-established breast cancer cell lines as in vitro models for human mammary tumours, we here describe via preclinical bioscreens how nucleolipidic ruthenium complexes are able to activate specific molecular cell death pathways, thereby interfering with cancer cells growth and proliferation. We also provide evidence that could strengthen novel drug-discovery strategies and ultimately lead to more selective ruthenium-based anticancer agents.

Results
In vitro bioscreens for anticancer activity. The antiproliferative properties of the liposomes containing the Ru(III)-nucleolipids were screened on human breast cancer models by means of selected mammary malignant cells endowed with different phenotypic and/or genotypic features and different replicative and/or invasive potential (MCF-7, CG-5, MDA-MB-231, MDA-MB-468 and MDA-MB-436 cell lines). The concentration-effect curves (Fig. 2a) -here reported in terms of a "cell survival index" which combines the measurements of cell number and viability -show a typical concentration-dependent sigmoid trend yielding IC 50 values in the low micromolar range. According to the IC 50 values (see Table 1), the ToThyRu/DOTAP and DoHuRu/DOTAP formulations are the most effective in reducing the proliferation of all the used cell lines. This result, consistent with our previous reports, is likely associated to the positive charge of DOTAP formulations which can promote the interaction with the plasma membranes allowing a faster and quantitative drug cellular uptake 19 . Moreover, in relation to their effectiveness in vitro, there are no significant differences between the ToThyRu and DoHuRu nucleolipidic Ru(III) complexes used in our liposomial formulations, consistently with the rationale that the ruthenium center is the bioactive species, and the nucleolipids are simply carrier molecules 24 . IC 50 data normalization in favour of the actual ruthenium content enclosed in ToThyRu/DOTAP and DoHuRu/DOTAP liposomes (15% mol/mol), leads up to values of IC 50 of about 3-4 μ M in CG-5 cells and of about 10 μ M in MCF-7 cells, suggestive of a marked antiproliferative bioactivity in vitro. Overall, the calculated ruthenium IC 50 values for DOTAP liposomes are in the low micromolar range (≤ 20 μ M), similar or lower than those measured for cisplatin used in the same experimental conditions as a reference drug. All the breast cancer cells used in this screening are therefore sensitive to the action of metal-based drugs. The ruthenium IC 50 values relative to cell growth inhibition in the presence of ToThyRu/POPC and DoHuRu/POPC formulations are also relevant, reaching values below 20 μ M as in the case of MDA-MB-468, MCF-7 and CG-5 cancer cells. Looking further at IC 50 values, the "naked" low molecular weight Ru(III) complex AziRu exhibits a milder antiproliferative activity on the same breast cancer cells, showing IC 50 values constantly greater than 250 μ M. These values are generally in agreement with those reported in the literature for the antimetastatic NAMI-A 28 , and once more highlight the critical importance of the delivery strategy to ensure the drug stability in the extra-cellular environment, a quantitative transport across the membranes and the bioavailability at the biological targets. Noteworthy, the bioactivity potentiating factors calculated as the ratio of IC 50 values of the Ru(III)-containing liposomes with respect to the IC 50 of AziRu, reach values of more than 20 in terms of antiproliferative ability 27 .

Cellular morphological changes.
Monitoring the cellular morphological changes over time can provide an additional support to the antiproliferative effects produced by ruthenium-based nanoaggregates. Throughout the in vitro trials, imaging end-points were collected by light microscopy from cells to assess cellular morphological changes induced by drug administration. Microphotographs in Fig. 2b of MCF-7 and MDA-MB-231 breast adenocarcinoma cells, nowadays among the most reliable in vitro models of breast cancer, in the presence of DoHuRu/POPC and DoHuRu/DOTAP (at their IC 50 concentrations for 48 and 72 h) are suggestive of the characteristic cell shrinkage as well as loss of cell-cell contact that accompanies apoptosis occurrence. Moreover, after in vitro treatments cells were morphologically examined at a higher magnification by a phase-contrast microscope for autophagic vacuoles formation. As shown in Fig. 2c, in addition to apoptotic hallmarks, autophagic vacuoles clearly appear when MCF-7 cells were treated with the DoHuRu/DOTAP liposomes for 48 h at IC 50 concentration, providing a morphological support of autophagy activation. These observations were further confirmed by FACS analysis, discussed below. Thus, the AziRu in vitro treatment by the DoHuRu/POPC liposome seems to trigger exclusively apoptosis, whereas the one with the cationic DoHuRu/DOTAP seems to simultaneously activate apoptosis and autophagy.

Pro-apoptotic effects in breast cancer cells.
Changes both in the cell morphology and in the monolayers appearance suggest that liposomes containing the nucleolipidic Ru(III)-complexes might have chemotherapeutic effects in human breast cancer initiating cells towards specific cell death pathways. The evaluation of apoptosis induction via FACS analysis has revealed that both DoHuRu/POPC and DoHuRu/DOTAP formulations are able to induce remarkable pro-apoptotic effects on MCF-7 cells and MDA-MB-231. In fact, as depicted in Fig. 3a,c with reference to MCF-7 cells, the 46% of total cell population is in advanced stage of apoptosis following 48 h of exposure to IC 50 concentrations of DoHuRu/DOTAP; further 24 h of treatment results in the 82% of cell population in late apoptosis phase. Milder pro-apoptotic effects, but nonetheless of significance, are detected in the case of DoHuRu/POPC treatment, leading after 48 h to about 36% of total MCF-7 cell population in early apoptosis and to a 37% in late apoptosis after 72 h. As a further validation of these data, no significant signals of apoptosis perturbation, such as an increased necrosis, were detectable by FACS analysis. A similar distribution of cell population among the different apoptotic stages has been observed by investigating the effects of liposomes containing nucleolipidic Ru(III)-complexes on MDA-MB-231 cells (see Fig. 3b,d), confirming the increased in vitro efficacy -especially as a trend over time -of the cationic nanoaggregates (i.e., the DoHuRu/DOTAP ones) with respect to the zwitterionic POPC ones. This scenario is fully consistent with our previous studies 19,27 . As a result of different cellular uptake kinetics, ruthenium-containing cationic liposomes (DoHuRu/DOTAP) induce faster biological effects by way of cellular apoptosis activation than those dependent by the neutral liposomes (DoHuRu/POPC). DNA fragmentation in MCF-7 and MDA-MB-231. It is generally accepted that DNA damage and subsequent induction of apoptosis is the primary cytotoxic mode of action of cisplatin and other metal-based antiproliferative drugs. In addition to shrinkage and fragmentation of the cells and nuclei, the apoptotic process is accompanied by degradation of the chromosomal DNA into nucleosomal units. Indeed, late events of apoptosis typically lead to DNA cleavage, resulting in a "ladder" formation detectable by agarose gel electrophoresis. To verify apoptosis induction in breast cancer cells in a direct manner, we analyzed DNA degradation on genomic DNA samples obtained from treated cells. As depicted in Fig. 3e, though with a non-canonical nuclear fragmentation pattern in MCF-7 due to inherent caspase-3 expression deficiency, the DNA damage markedly increased in cells exposed to both DoHuRu/POPC and DoHuRu/DOTAP for 48 h at IC 50 doses. Moreover, the typical internucleosomal DNA laddering of cell undergoing apoptosis clearly appeared in MDA-MB-231 cells, with a fragmentation pattern similar to that induced by IC 50 doses of cisplatin in vitro, herein used as positive control 29 .

Apoptotic-related proteins in MCF-7 and MDA-MB-231 breast cancer cells.
Given that the expression profile of pro-apoptotic proteins is a central determinant for the death mode induced by ruthenium-based drugs, we analyzed the expression of proteins linked to the two main pathways that control mammalian apoptosis, i.e. the extrinsic (or death receptor) and intrinsic (or mitochondrial) apoptotic pathways. As determined by Western blot analysis reported in Fig. 4a,c,e, protein extracts from MCF-7 exposed for 48 and 72 h to IC 50 concentrations of DoHuRu/POPC and DoHuRu/DOTAP showed a remarkable increase in caspase-9 activity compared with control, whereas in the same experimental conditions no caspase-8 activation was detectable. Indeed, the activation of pro-caspase-9 involves intrinsic proteolytic processing, resulting in the production of an active p35 subunit. Moreover, as clearly detectable by immunoblot, an additional cleavage occurred producing a large p37 subunit which is known to amplify the apoptotic response. On the other hand, full length pro-caspase-8 expression was not affected by in vitro treatments. Taken together, these results suggest a ruthenium-dependent activation of the apoptotic mitochondrial pathway, wherein Bax and Bid are key cell death factors increasing mitochondrial permeability and release of cytochrome c. Conversely, cell survival factor Bcl-2 inhibits actions of Bax and Bid on mitochondria 30 . In MCF-7 cells Bax was significantly increased and Bcl-2 decreased upon DoHuRu/POPC and DoHuRu/DOTAP treatment -a cellular response predisposing to apoptotis activation. These biological effects were time-dependent and more evident following in vitro treatment with the cationic DoHuRu/ DOTAP liposomes.
In the same way, ruthenium treatment in MDA-MB-231 breast cancer cells via DoHuRu/POPC and DoHuRu/ DOTAP in vitro administration at IC 50 concentrations elicited caspase-9 activation after 48 h, and this anew was coupled to the simultaneous Bax and Bcl-2 up-regulation and down-regulation, respectively (Fig. 4b,d,f). Furthermore, in this breast cancer cell line DoHuRu/DOTAP seemed able to promote full length pro-caspase-8 cleavage, as evidenced by the formation of the active p10 and p18 fragments. In turn, activated initiator caspases further process other caspase members, including caspase-3 and caspase-7, to initiate a caspase cascade, which generally leads to complete the programmed cell death process. In fact, the immunoblotting analysis performed on MDA-MB-231 cells exposed to Ru(III)-containing liposomes further revealed a marked proteolytic cleavage of the inactive proenzyme to activate caspase-3. Hence, the evaluation of apoptosis regulatory proteins in breast cancer models for the in vitro preclinical evaluation of ruthenium biological effects suggests the invariable induction of the mitochondrial apoptotic cell death pathway, but also the cell-specific activation of the extrinsic death pathway particularly in the case of the cationic nanoaggregate.
Interestingly, the treatment with either DoHuRu/POPC or DoHuRu/DOTAP alters the expression profile of Bax and Bcl-2 proteins with respect to basal amounts, radically changing the Bax/Bcl-2 ratio. The Bax up-regulation and Bcl-2 down-regulation observed in concert following ruthenium action may predispose cells to apoptosis 30 .

Autophagy activation in MCF-7 and MDA-MB-231 breast cancer cells.
In addition to apoptosis, cellular suicide can also be executed via non-apoptotic forms of programmed death such as autophagy. The simultaneous evaluation of apoptosis and autophagy by means of the herein used methods is not mutually exclusive, i.e. it is possible to ascertain the degree of autophagy independently from that of apoptosis.

Discussion
Innovative anticancer drugs with new molecular mechanisms of action are essential in chemotherapeutic treatment to kill specific cancer types, and to overcome toxic side effects as well as chemoresistance 31 . Current research efforts are focused on a deeper understanding of the cellular response and/or resistance to anticancer treatments, including the role of cell death pathways activation, such as apoptosis and autophagy, by chemotherapeutics 32 .
Recently, we have developed new biocompatible ruthenium-based nanosystems, proved to be particularly effective against specific cell lines derived from human solid tumours 19,22,[24][25][26][27] . Starting from these encouraging data, in the present work we have first confirmed their efficacy focusing on a panel of human tumour cells arising from breast cancer. At the moment, the endocrine-responsive (ER) breast adenocarcinoma MCF-7 and the triple-negative breast adenocarcinoma (TNBC) MDA-MB-231 cell models account for the great majority of investigations on breast cancer cells and are considered the most reliable in vitro models of breast cancer together with their variants CG5, and MDA-MB-436 and MDA-MB-468, respectively 33,34 . All these cells are sensitive to cisplatin in vitro, so that cisplatin is used in standard first line treatment protocols for many breast cancers, often in combination with other drugs, such as taxanes, vinca alkaloids, and 5-fluorouracil, resulting in synergistic or additive effects. Therefore, classical chemotherapy is currently the only drug-based option in the therapeutic armamentarium, and metal-based chemotherapy is also emerging as an upcoming treatment modality especially in TNBC 35 . Although new types of ruthenium complexes with superior anticancer activity in vitro have been meanwhile reported 36 , the induction of comparable or even greater cytotoxic effects than cisplatin have been herein confirmed for nucleolipidic Ru(III) formulations. Then, aiming at an in-depth investigation of their mode of action, we have shown that amphiphilic ruthenium complexes, properly delivered by suitable nano-systems, are able to kill cancer cells by activating specific apoptotic processes, coupled in some cases to cellular autophagy. Indeed, in the last decade many different ruthenium compounds have been tested for their anticancer properties. However, all the reported studies have not clearly evidenced a unique mode of action at cellular level, nor unambiguously defined a specific mechanism of action at molecular level 10,21,24 . In-depth bioanalytical studies have been so far performed especially for cisplatin, which have validated nuclear DNA as the main final drug target causing adducts formation and DNA damage, leading to cell division inhibition and cytotoxicity 5 . In analogy with cisplatin, some of us have recently demonstrated that also AziRu and NAMI-A can interact with DNA model systems, with Ru(III) ions being incorporated into oligonucleotide structures via stable linkages 37 . However, since an increasing number of evidences demonstrate that both ruthenium(II) and (III)-based drugs are able to interact with both intra-and extra-cellular protein targets, currently other molecular mechanisms of action cannot be excluded 9,38 . Beyond molecular targeting, we have observed an invariable activation of programmed cell death pathways which confirms that the primary mode of action in breast cancer models of AziRu via DoHuRu/POPC and DoHuRu/DOTAP formulations is the induction of apoptosis. These data are largely in accordance with several reports highlighting the occurrence of distinct hallmarks of apoptosis after ruthenium administration in vitro 16,39 . Cellular morphological changes and DNA fragmentation provided additional evidence of an apoptosis-inducing activity at the basis of the anticancer properties of AziRu. The regulation of apoptotic cell death orchestrated by intracellular caspases plays a fundamental role in the response to chemotherapeutics, as evasion of apoptosis is one of the central features of malignant progression as well as of drug resistance. Indeed, survival of malignant mammary cells is a key event in disease occurrence and progression, but also in therapy failure and chemoresistance development 40 . Physiological mammary cells growth is controlled by a balance between cell proliferation and apoptosis. A large body of evidence has clarified that tumour growth is not just a result of uncontrolled proliferation but also of reduced apoptosis, so that the balance between proliferation and apoptosis is crucial in determining the overall growth or regression of the tumour in response to chemotherapy 41 . Two main pathways of caspase activation have been described in mammalian cells, which result in final control of apoptosis. In the intrinsic pathway, typically activated by intracellular stress signals, pro-apoptotic cell death factors belonging to the Bcl-2 family increase mitochondrial permeability and release of cytochrome c, as well as of other proteins from the intermembrane space of mitochondria. Apaf-1, a downstream mediator of apoptosis, along with cytochrome c, associates with caspase-9 in cytoplasm and leads to its activation 42 . The resulting apoptosome initiates a cascade of effector caspases, which include caspases-3, -6, and -7. In turn, the active caspase-3 triggers DNA fragmentation factor (Caspase-Activated DNase, CAD) and promotes DNA internucleosomal cleavage. All the formulations containing nucleolipidic Ru(III)-complexes we have tested activate the mitochondrial apoptotic cell death pathway in breast cancer cells, as highlighted by a remarkable activation of caspase-9. Interestingly, this occurs independently of the cell ability to complete apoptosis process via the executioner caspase-3, as demonstrated by Ru-dependent activation of apoptosis in MCF-7. This adenocarcinoma model is known to be resistant to some chemotherapeutics due to a deletion in the CASP-3 gene that leads to an inherited deficiency of caspase-3 43 . Caspase-3 -commonly turned on by numerous death signals -cleaves a variety of important cellular proteins and is ultimately responsible for apoptotic DNA fragmentation. Despite the lack of caspase-3 expression, liposomes containing nucleolipidic Ru(III)-complexes have hitherto shown to be particularly effective on this in vitro model. It has been also reported that MCF-7 undergoes cell death by apoptotic stimuli in the absence of the typical DNA fragmentation, and recent observations further suggest that large and small DNA fragments coupled to even single-strand cleavage events occur during apoptotic death 44 . Consistently with our results, these observations have raised relevant questions on the degradation pattern of nuclear DNA in agarose gel electrophoresis detection, which remains controversial. However, morphological changes and MCF-7 cell death were independent of caspase-3 and may correlate with the activation of different apoptotic pathways and other effector caspases, such as caspase-6 or -7. As far as the activation of programmed cell death pathways is concerned, DoHuRu/DOTAP formulation seems capable to concurrently activate the two major pathways of apoptosis in a cell-specific mode. In fact, MDA-MB-231 cells, undergoing the apoptotic mitochondrial pathway after exposure to DoHuRu/DOTAP, show an apparent proteolytic processing of pro-caspase-8 to form various fragments, including the active p18 and p10. The extrinsic pathway is activated by extracellular ligands able to bind to death receptors on the cell surface, which leads to the formation of the death-inducing signaling complex (DISC). This death receptor pathway is triggered by members of the death receptor superfamily such as CD95 and tumour necrosis factor receptor. Formation of a death-inducing signalling complex induces caspase-8 activation and thereby the downstream caspase cascade 45 . We hypothesize that the cationic DOTAP nanoaggregate, by means of its inherent surface charge, can interact in peculiar manner with the external surface of cell membranes, as suggested by the faster cellular uptake kinetics compared to POPC formulations observed in a previous study 19 . Although a hypothesis, exclusive molecular interactions coupled to local drug release could stimulate specific surface receptors involved in the activation of the extrinsic pathway. Indeed, not infrequently the killing of tumor cells by anticancer chemotherapeutics has been linked to activation of extrinsic apoptosis pathways 46 . These outcomes are consistent with former investigations and demonstrate that some ruthenium(II) and (III) complexes can simultaneously trigger intrinsic and extrinsic apoptosis pathways 47 .
As far as the Bcl-2 family proteins are concerned, we believe that some central aspects closely linked to ruthenotherapy have emerged. Members of this family are regulatory proteins involved in the control of cell death, by either inducing (pro-apoptotic factors) or inhibiting (anti-apoptotic factors) apoptosis 48 . Bcl-2 is a crucial anti-apoptotic protein to be regarded as an oncogene, playing an important role in promoting cellular survival by inhibition of pro-apoptotic proteins. On the other hand, the pro-apoptotic effectors of the Bcl-2 family, including Bax, normally act on the mitochondrial membrane to promote permeabilization coupled to the release of both cytochrome c and ROS as important signals in the apoptosis cascade. Several clinical studies have provided support for an overexpression of the antiapoptotic Bcl-2 protein as a negative prognostic marker in various tumours 49 . Alternatively, decreased Bax levels have been found in correlation with shorter survival in patients with breast cancer and colorectal cancer 50 . In MCF-7 and MDA-MB-231 we found an important effect induced by both DoHuRu/POPC and DoHuRu/DOTAP treatment on the cellular content of Bcl-2 and Bax, which could be correlated to the induction of the mitochondrial cell death pathway. The significant increase in Bax/Bcl-2 ratio detected in treated cells with respect to untreated cells could have an important impact in the regulation of cell fate by interfering with breast cancer cell survival. Indeed, in accordance with experimental and clinical investigations, tumours dependent on Bcl-2 family members are likely sensitive to Bcl-2 modulation in order to survive; in turn high Bax expression has been associated with a better response to chemotherapy in many cancers forms 51 . Consistently with our results, it should be possible to circumvent the inherent apoptosis deficiency of malignant cells by directly affecting the mitochondrial function. In this way, following mechanisms yet to be clarified, the ruthenium complexes may interact with mitochondrial targets, possibly via selective accumulation and ROS generation 52 . At the same time, dysfunction of mitochondria might be responsible for autophagic cell death. More and more studies underline the important interplay between apoptosis and autophagy, and suggest that apoptosis activation is often related to increased autophagy processes 53 . As in the case of apoptosis, autophagy plays a vital role in cellular proliferation and survival, and dysregulated autophagy activation has been described in many pathologies, including cancer. Although the role of the autophagic programmed cell death in neoplastic development remains to be clarified in vivo, cancer cells with up-regulated autophagy exhibit a less aggressive behaviour as well as an increased susceptibility to chemotherapy 54 . Since the effectiveness of many apoptosis-inducing chemotherapies can be affected by specific mutations in genes orchestrating apoptotic regulation, the activation of alternative cell death pathways, in tandem with or in the absence of an efficient apoptotic machinery, may represent an attractive goal for novel metal-based chemotherapeutics. In this perspective, recent evidence supports the occurrence of the simultaneous induction of apoptosis and autophagy in cancer cells as a result of specific signaling 55 . In addition, autophagy can be also activated upon exposure to genotoxic compounds, including several metal-based drugs able to target DNA. The induction of autophagy in the case of cationic nucleolipidic formulations could be linked to the ruthenium-induced down-regulation of the prosurvival protein Bcl-2, evident in both MCF-7 and MDA-MD-231 cells. Indeed, disturbances in the interaction between Beclin-1 and Bcl-2 family proteins, by which Beclin-1 is inhibited in normal conditions, has been established to stimulate cellular autophagy 56 . Thus Ru(III) complexes activity in inhibiting breast cancer cells proliferation via DoHuRu/ DOTAP administration would lie in the crosstalk connecting the main molecular mechanisms involved in the regulation of apoptosis and autophagy processes. Nevertheless, the existence of other non-apoptotic nor necrotic cell death pathways, triggered upon exposure to ruthenium-containing nanoaggregates we have documented in breast cancer models, needs further and more targeted studies. According to our previous findings 24 and to recent literature reports 36,52 , it cannot be excluded that the simultaneous activation of different mechanisms of cell death can be caused by multiple potential interactions at the subcellular/molecular level, both nuclear and cytosolic. Nevertheless, the activation of multiple death pathways by metal-based chemotherapeutics in aggressive cancer diseases with limited treatment options is a largely desired goal, in order to possibly restrict the onset of chemoresistance as well as to efficiently counteract uncontrolled proliferation. Upcoming investigations on the critical proteins and pathways involved in autophagy control exerted by Ru(III)-containing liposomes are underway to strengthen knowledge in favour of future in vivo applications for these high potential candidate drugs.  Fig. 1 along with AziRu) were prepared by reacting in stoichiometric amounts the starting nucleolipids, named ToThy or DoHu 57 , with the Ru complex [trans-RuCl 4 (DMSO) 2 ] − Na + following a previously described procedure 25 . In all cases, the desired nucleolipidic Ru(III) complexes were obtained in a pure form, as confirmed by TLC and ESI-MS analysis, and almost quantitative yields. The lipid formulations of ToThyRu and DoHuRu in POPC and DOTAP were prepared as previously reported 24 and characterized by DLS analysis.

Methods
In vitro bioscreens. The anticancer activity of ruthenium-containing nucleolipidic nanoparticles and of AziRu was investigated through the estimation of a "cell survival index", arising from the combination of cell viability evaluation with cell counting 25 . Cells were inoculated in 96-microwell culture plates at a density of 10 4 cells/well, and allowed to grow for 24 h. The medium was then replaced with fresh medium and cells were treated for further 48 h with a range of concentrations (1 → 1000 μ M) of AziRu, and of ToThyRu and DoHuRu complexes lodged either in POPC or DOTAP liposomes (ToThyRu/POPC and ToThyRu/DOTAP, DoHuRu/POPC and DoHuRu/DOTAP, respectively). Using the same experimental procedure, cell cultures were also incubated with ruthenium-free ToThy/POPC, ToThy/DOTAP, DoHu/POPC and DoHu/DOTAP liposomes as negative controls, as well as with cisplatin (cDDP) -a positive control for cytotoxic effects 27 . Cell viability was evaluated using the MTT assay procedure, which measures the level of mitochondrial dehydrogenase activity using the yellow 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT, Sigma) as substrate. The assay is based on the redox ability of living mitochondria to convert dissolved MTT into insoluble purple formazan. Briefly, after the treatments, the medium was removed and the cells were incubated with 20 μ l/well of a MTT solution (5 mg/mL) for 1 h in a humidified 5% CO 2 incubator at 37 °C. The incubation was stopped by removing the MTT solution and by adding 100 μ l/well of DMSO to solubilize the obtained formazan. Finally, the absorbance was monitored at 550 nm using a microplate reader (iMark microplate reader, Bio-Rad, Milan, Italy). Cell number was determined by TC10 automated cell counter (Bio-Rad, Milan, Italy), providing an accurate and reproducible total count of cells and a live/dead ratio in one step by a specific dye (trypan blue) exclusion assay. Bio-Rad's TC10 automated cell counter uses disposable slides, TC10 trypan blue dye (0.4% trypan blue dye w/v in 0.81% sodium chloride and 0.06% potassium phosphate dibasic solution) and a CCD camera to count cells based on the analyses of captured images. Once the loaded slide is inserted into the slide port, the TC10 automatically focuses on the cells, detects the presence of trypan blue dye and provides the count. When cells are damaged or dead, trypan blue can enter the cell allowing living cells to be counted. Operationally, after treatments in 96-microwell culture plates, the medium was removed and the cells were collected. Ten microliters of cell suspension, mixed with 0.4% trypan blue solution at 1:1 ratio, were loaded into the chambers of disposable slides. The results are expressed in terms of total cell count (number of cells per ml). If trypan blue is detected, the instrument also accounts for the dilution and shows live cell count and percent viability. Total counts and live/dead ratio from random samples for each cell line were subjected to comparisons with manual hemocytometers in control experiments.
The calculation of the concentration required to inhibit the net increase in the cell number and viability by 50% (IC 50 ) is based on plots of data (n = 6 for each experiment) and repeated five times (total n = 30). IC 50 values were obtained by means of a concentration response curve by nonlinear regression using a curve fitting program, GraphPad Prism 5.0, and are expressed as mean values ± SEM (n = 30) of five independent experiments. Cell morphology. Human breast cancer cell lines were grown on 60 mm culture dishes by plating 5 × 10 5 cells. After reaching the subconfluence, cells were incubated for 48 h with IC 50 concentrations of the ruthenium-containing liposomes (DoHuRu/POPC and DoHuRu/DOTAP) under the same in vitro experimental conditions described above and were then morphologically examined by a phase-contrast microscope (Labovert microscope, Leizt) for autophagic vacuoles and apoptotic markers. Microphotographs at a 200 × total magnification (20 × objective and 10 × eyepiece) were taken with a standard VCR camera (Nikon).
DNA fragmentation assay. MCF-7 and MDA-MB-231 cells were grown on standard plastic 60 mm culture dishes by plating 5 × 10 5 cells. After 24 h of growth the cells were treated for 48 h with IC 50 doses of DoHuRu/ POPC and DoHuRu/DOTAP liposomes under the same experimental conditions described for bioscreens, as well as with IC 50 doses of cisplatin (cDDP) -a positive control for in vitro apoptosis 29 . The DNA fragmentation assay was carried out as previously reported 27 . After treatments, cells were collected and the pellets were suspended in lysis buffer (50 mM Tris-HCl, pH 8.0, 5 mM EDTA, 100 mM NaCl, 1% SDS, 0.5 mg/mL Proteinase K) and incubated at 50 °C. After 1 h incubation, 10 mg/mL RNase was added to the lysates and incubated for 1 h at 50 °C. DNA was precipitated with NaOAc pH 5.2 and ice cold 100% EtOH and centrifuged at 14000 × g for 10 min. Pellets were dissolved in TE buffer (10 mM Tris-HCl, pH 8.0, 1 mM EDTA). A 20 μ L aliquot of each DNA sample was analyzed on a 1.5% agarose gel stained with ethidium bromide and visualized under UV light.  61 . The supernatant fraction was obtained by centrifugation at 15,000 × g for 10 min at 4 °C and then stored at − 80 °C. Protein concentration was determined by the Bio-Rad protein assay (Bio-Rad, Milan, Italy).
Statistical Analysis. All data were presented as mean values ± SEM. The statistical analysis was performed using Graph-Pad Prism (Graph-Pad software Inc., San Diego, CA) and ANOVA test for multiple comparisons was performed followed by Bonferroni's test.