Lipid-conjugated Ru(iii) complexes as inhibitors of pathological peptide aggregation in neurodegenerative disorders Authors: Claudia Riccardi, Ettore Napolitano, Chiara Platella, Andrea Conte, Simona Paladino, Domenica Musumeci, Daniela Montesarchio Published: 2026 ABSTRACT Neurodegenerative diseases (NDs), characterized by the progressive loss of neuronal homeostasis and function, represent an increasing global health burden. Their severity and profound impact on quality of life highlight the urgent need for effective and targeted therapeutic strategies. Recent studies have drawn attention to the potential of certain Ru(iii) complexes, including NAMI-A and KP1019, as promising therapeutic candidates for NDs. Originally developed as anticancer agents, these compounds have demonstrated remarkable activity in inhibiting the aggregation of model misfolded proteins, which represent a common pathological feature of various neurological disorders. Herein, searching for Ru(iii) complexes which could be effective as anti-neurodegenerative agents, we selected AziRu – a NAMI-A-like compound – and some of its lipid-conjugated derivatives, previously explored as anticancer drug candidates, and tested them for their ability to interfere with the self-aggregation of selected model peptides involved in the pathogenesis of NDs, such as Alzheimer's disease (AD), Parkinson's disease (PD) and prion diseases. Among them, the Ru(iii) complexes we named MyriPyRu and PalmiPyRu, bearing myristic and palmitic acid-based tails, respectively, emerged as promising candidates. These complexes effectively modulated and inhibited the aggregation of amyloid peptides linked to AD pathogenesis with IC50 values in the low micromolar range (ca. 5 μM for both compounds). More detailed studies on their interaction with AD peptides revealed their capacity to be coordinated by histidine residues in the N-terminal region. Biological investigations on human neuroblastoma-derived cells indicated no relevant cytotoxicity at concentrations comparable to those that caused anti-aggregation effects. Overall, these findings are highly promising, as MyriPyRu or PalmiPyRu could be the starting compounds to evolve further modified derivatives functionalized with specific targeting moieties for amyloid peptides, potentially paving the way to a new generation of Ru(iii)-based anti-amyloid agents for the treatment of Alzheimer's disease. Introduction Neurodegenerative diseases (NDs) are a heterogeneous group of aging-related disorders characterized by the progressive loss of structure and/or function of neurons in specific brain regions.1,2 A common underlying mechanism in many of these disorders is the misfolding of specific proteins involved in disease onset and progression. Accordingly, NDs are often referred to as ‘protein misfolding’ diseases or proteinopathies.3,4 Proteins are indeed essential for maintaining cellular structure and function but, to correctly perform their roles, they must fold into specific three-dimensional shapes. When proteins misfold, they lose functionality and may aggregate, forming toxic species that contribute to cellular damage.5,6 The accumulation of misfolded proteins is a hallmark of several NDs, including Alzheimer's (AD), Parkinson's (PD) and Huntington's diseases (HD) – where the associated proteins are amyloid-beta (Aβ) peptides, α-synuclein and mutant huntingtin,4,6–9 respectively – as well as prion diseases, a group of fatal infectious or genetic neurodegenerative disorders caused by the conformational conversion of a cellular prion protein into its pathological isoform.10,11 Therefore, misfolded proteins represent crucial therapeutic targets for the development of novel and effective treatments aimed at slowing or halting the progression of these severe conditions.7,8,12 In this context, some metal-based complexes have been explored as potential drug candidates in neurodegenerative disorders due to their ability to interact with proteins implicated in disease pathogenesis.13–19 Particular attention has been devoted to platinum20–26 and ruthenium derivatives,27–34 which can target various misfolded proteins involved in different neurodegenerative conditions. Specifically, within the class of Ru(iii)-based complexes, NAMI-A, imidazolium trans-[tetrachloridobis(imidazole)ruthenate(iii)] (Fig. S1, left), and KP1019, indazolium trans-[tetrachloridobis(indazole)ruthenate(iii)] (Fig. S1, right), are two well-known Ru(iii) complexes extensively studied over the past 30 years for their selective anticancer and/or antimetastatic properties.35–38 More recently, the potential therapeutic use of these Ru(iii) complexes has broadened beyond oncology. For instance, NAMI-A has been shown to inhibit α-synuclein aggregation and fibril formation, thereby reducing neurotoxicity and motor impairments in animal models.39 In another study, KP1019 has been reported to bind Aβ peptides, modulate their aggregation, and reduce associated neurotoxicity.40 Similarly, the aggregation of the prion protein fragment 106–126 has been suppressed by both NAMI-A and KP1019.41 Collectively, these studies demonstrated that NAMI-A-type Ru(iii) complexes can interfere with Aβ aggregation, highlighting the N-terminal metal-binding region of the peptide as the main interaction site. These observations suggest that Ru(iii)-based compounds may represent promising modulators of pathological protein aggregation in neurodegenerative disorders and prompted us to investigate whether structurally related Ru(iii) derivatives could display similar anti-aggregation properties. Thus, we investigated a set of Ru(iii)-based compounds, previously explored in our group as antitumor agents, for their potential as anti-neurodegenerative candidates. In detail, we focused on AziRu (Fig. 1), a compound structurally related to NAMI-A carrying a pyridine ligand replacing the imidazole one,42,43 and on a series of lipid-functionalized analogues we named MyriPyRu, PalmiPyRu, StePyRu, OlePyRu and CholPyRu according to the nature of their lipophilic tail (Fig. 1).44 This set of Ru(iii) complexes was recently prepared by some of us44 through conjugation of the AziRu pyridine ligand with selected biocompatible lipids (Fig. 1)44 in order to overcome the limited cell membrane permeability45 and early deactivation in the extracellular environment of AziRu likely due to ligand exchange reaction in aqueous media.46–51 Indeed, monitoring the hydrolysis in pseudo-physiological solutions over time by UV-vis measurements, significantly prolonged half-lives values were obtained for the lipid-conjugated Ru(iii) complexes (ranging from 90 to 240 min) compared to AziRu (approximately 45 min), as previously determined (Fig. S2 and S3).44 Chemical structures of AziRu and its lipophilic Ru(iii) derivatives analyzed in this study. More importantly, within this series, several compounds, especially PalmiPyRu and StePyRu, exhibited high antiproliferative activity against human breast cancer cells (BCC) while showing no toxicity toward control healthy cells.44 Notably, PalmiPyRu demonstrated significant binding to bovine serum albumin (BSA) – used as a model serum protein – and to model DNA G-quadruplex (G4) structures. In addition, PalmiPyRu showed a 20-fold increase in cellular uptake compared to AziRu in two histologically distinct BCC lines44 and the ability to induce ferroptosis through disruption of iron homeostasis and oxidative stress pathways.52 Building upon our previous studies on lipid-conjugated Ru(iii) complexes originally developed as anticancer agents,42–44,50,52 the present work investigates, for the first time, their potential as modulators of aggregation of key misfolded proteins. To this aim, four different peptide sequences (Table 1) were studied: Table 1Peptide sequences analyzed in this studyPeptideSequenceAβ1–40DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGGVVAβ1–15DAEFRHDSGYEVHHQSup35p7–13GNNQQNYα-SynMDVFMKGLSKAKEGVVAAAEKTKQGVAEAAGKTKEGVLYVGSKTKEGVVHGVATVAEKTKEQVTNVGGAVVTGVTAVAQKTVEGAGSIAAATGFVKKDQLGKNEEGAPQEGILEDMPVDPDNEAYEMPSEEGYQDYEPEA Peptide sequences analyzed in this study 1) the full-length Aβ1–40 peptide, a proteolytic fragment of the amyloid precursor protein (APP) encoded by a gene on chromosome 21;53,54 2) the Aβ1–15 fragment, spanning the first 15 residues of the beta-amyloid peptide;55 3) the amyloid-forming peptide, i.e. the heptapeptide GNNQQNY, corresponding to residues 7–13 of the N-terminal region of the yeast prion protein Sup35p (Sup35p7–13);24,56 4) α-synuclein, a 140-residue polypeptide genetically and neuropathologically linked to Parkinson's disease (PD).57 The selected peptide fragments were chosen for their relevance in mediating the aggregation processes of their corresponding full-length proteins. Once prepared following our previously published protocols,44 all the Ru(iii) complexes were investigated in their interaction with selected peptide sequences in PBS through various biophysical techniques. Specifically, fluorescence and dynamic light scattering (DLS) measurements were employed to evaluate the ability of the studied Ru(iii) derivatives to modulate peptide aggregation. Subsequently, 1H NMR and UV-vis spectroscopies were used to gain deeper insight into their specific interaction with the selected peptides. Finally, cytotoxicity of Ru(iii) complexes on human neuroblastoma-derived SH-SY5Y cells, a widely used human neuronal model, was assessed using viability assays. In particular, the present study is aimed at evaluating whether the nature of the appended lipid moiety influences the interaction of these Ru(iii) complexes with model self-assembled peptides and their anti-aggregation properties, thereby providing preliminary structure–activity relationship information for this class of compounds. Results and discussion AziRu and its lipid-conjugated derivatives were prepared according to previously reported protocols.44 In order to here evaluate their interaction with the selected peptide sequences, stock solutions of each Ru(iii) complex were prepared at mM concentration in acetonitrile, from which appropriate aliquots were withdrawn and diluted with PBS to reach the desired concentration for each experiment, always maintaining the final acetonitrile content in the working solutions below 1%. UV-vis stability studies of the Ru(iii) complexes in acetonitrile The stability of 100 μM solutions of the studied complexes in acetonitrile was assessed by UV-vis analysis over a 24 h period (Fig. S4). Under these conditions, all Ru(iii) complexes exhibited spectra with the characteristic absorption band centered around 400 nm, attributed to ligand-to-metal charge transfer (LMCT) transitions involving the Ru–Cl bonds.43,58,59 For all the tested compounds, time-dependent UV-vis spectra, commonly used to monitor hydrolysis and ligand substitution events in Ru(iii) complexes, did not show significant changes of the characteristic LMCT band around 400 nm. Very minor variations observed in the UV region were not systematic across the analyzed lipid-conjugated Ru(iii) complexes, with a very small absorbance decrease for AziRu and a moderate increase for MyriPyRu and StePyRu (Fig. S4), and were therefore not considered indicative of solvent coordination. Overall, these results demonstrate that the investigated compounds remain sufficiently stable in solution under the adopted experimental conditions, supporting their subsequent evaluation in peptide aggregation studies. Thioflavin T (ThT) fluorescence emission experiments ThT is a small molecule consisting of a dimethylated benzothiazole ring linked to a dimethylaminobenzyl ring via a single C–C bond. It represents a widely explored dye to study the kinetics of the self-assembling process associated with amyloid fibril formation as it exhibits a characteristic fluorescence response upon binding to Aβ fibrils.60–65 In detail, in the absence of amyloid-like aggregates, ThT displays only basal fluorescence. Upon amyloid fibril formation, ThT fluorescence markedly increases, making ThT-based assays a highly sensitive and reliable method for assessing the effect of potential therapeutic agents on Aβ aggregate formation.62–65 Upon addition of an inhibitor of protein aggregation, the fluorescence intensity should be reduced to an extent depending on the effective potency of the inhibitor. For these experiments, each Ru(iii) complex was incubated with equimolar concentrations of the selected peptides and ThT, following previously described protocols.66,67 In the case of Aβ1–40, 24 h incubation with MyriPyRu and PalmiPyRu led to reduce the fluorescence signals to approximately 20% of their initial values (Fig. 2a). Intermediate fluorescence decrease was observed for StePyRu, OlePyRu, and CholPyRu, which retained around 50% of the initial emission intensity, while the reference compound AziRu proved to be the least effective Aβ1–40 aggregation inhibitor, with fluorescence emission reaching ca. 80% of the starting ThT fluorescence value (Fig. 2a). ThT-based fluorescence emission experiments (λexc = 440 nm and λem = 495 nm) on Aβ1–40 (10 μM concentration, a), Aβ1–15 (20 μM, b), Sup35p7–13 (50 μM, c) and α-syn (2 μM, d) in the absence or presence of the selected Ru(iii) complexes using a peptide/ThT/Ru system in 1 : 1 : 0 or 1 : 1 : 1 ratio. The percentages of aggregate formation at 24 h were calculated using the fluorescence emission of ThT in peptide/Ru samples in comparison with the fluorescence values observed for each peptide alone in solution. Data are expressed as mean values ± SD (error bars) for multiple determinations. Statistical significance was assessed by using Student's t-test: **** p < 0.0001, *** p < 0.001, ** p < 0.01, * p < 0.05 for Ru complexes relative to each peptide alone. Notably, the overall anti-aggregation properties observed for the present lipid-conjugated Ru(iii) complexes were comparable to those previously reported for the most active imidazole-containing derivatives67 and higher than those observed for the thiazole66 analogues, analyzed under the same experimental conditions (peptide concentration, incubation conditions and metal-to-peptide ratio) with Aβ1–40. As control experiments, all Ru(iii) complexes were also tested alone and in the presence of ThT at 10 μM concentration and 1 : 1 molar ratio in PBS for 24 h (Fig. S5). The results indicated that all the Ru(iii) complexes exhibited negligible fluorescence intensities, even in the presence of ThT, thereby excluding any significant interaction with the fluorescent dye. In all cases, the observed fluorescence intensity was comparable to that of ThT alone, under identical experimental conditions, and however ca. 12.5 times lower than signals observed in the presence of Aβ1–40 treated with the fluorescent probe for the same time (Fig. S5). Although Aβ1–15 has considerably lower propensity to form mature amyloid fibrils than full-length Aβ, previous studies have shown that short Aβ-derived fragments may still exhibit measurable aggregation-related behaviour68 and therefore this fragment was also included in ThT-based assays to provide complementary information on the effect of the Ru(iii) complexes on the self-assembly behaviour of the N-terminal region of the peptide. When Aβ1–15 was tested (Fig. 2b), the Ru(iii) complexes induced changes in ThT fluorescence that qualitatively followed the same order of activity observed for Aβ1–40. Indeed, the treatment with AziRu resulted in only a slight reduction of the fluorescence emission value. Similarly, addition of CholPyRu led to a modest reduction of ca. 20% of the initial fluorescence signal. A more significant decrease in the fluorescence emission intensity was observed in the presence of StePyRu and OlePyRu, but the most effective inhibitors of Aβ1–15 aggregation were found to be MyriPyRu and PalmiPyRu, which produced an overall reduction of about 70% in fluorescence compared to the untreated peptide (Fig. 2b). To assess whether these Ru(iii) complexes exert a similar inhibitory effect on other relevant misfolded proteins, ThT assays were also performed using Sup35p7–13 and α-syn model peptides (Fig. 2c and d). A comparable inhibitory trend was also observed with Sup35p7–13 (Fig. 2c), where all Ru(iii) complexes affected the peptide fluorescence emission intensity even if the fluorescence intensity values observed in this case were always lower than those recorded with amyloid AD peptides. Notably, a reduction of about 50% was recorded only upon treatment with MyriPyRu and PalmiPyRu (Fig. 2c). In the case of α-syn, only a moderate decrease in ThT fluorescence was observed in all cases, even upon treatment with MyriPyRu and PalmiPyRu, resulting in a slight reduction of around 30% compared to the control (Fig. 2d). Since only limited inhibitory effects were observed for Sup35p7–13 and α-syn, subsequent investigations were focused on Aβ-derived peptides associated with Alzheimer's disease. Among the Ru-complexes tested, MyriPyRu and PalmiPyRu, identified as the most potent inhibitors of peptide self-aggregation, were selected for further analysis. Next, we evaluated if the aggregation inhibition produced by MyriPyRu and PalmiPyRu on Aβ1–40 was concentration-dependent. To this aim, the same ThT-based experiment was performed using fixed concentrations of the peptide (10 μM) and increasing amounts of each Ru(iii) complex in the 1 → 30 μM concentration range (Fig. 3), in analogy to previous reports.24 Inhibition of Aβ1–40 aggregation, analyzed at 10 μM concentration, upon addition of MyriPyRu and PalmiPyRu (with increasing concentration in the 1 → 30 μM range), evaluated as a ThT-based fluorescence variation. In the table, IC50 values (μM) of MyriPyRu and PalmiPyRu were reported. Results are representative of two independent experiments. Data are reported as mean values ± SD. Both compounds inhibited Aβ1–40 aggregation in a concentration-dependent manner, displaying comparable IC50 values, in both cases below 5 μM (Fig. 3). DLS analysis DLS is a highly sensitive and powerful technique for assessing particle size and aggregation in solution,69,70 and is particularly useful for evaluating the impact of metal complexes on Aβ fibril formation.71 In the absence of any Ru(iii) complex, Aβ1–40 exhibited a single, broad peak centered at approximately 370 nm (Fig. 4a, black line), consistent with previously reported data.66,67,72 DLS analysis of Aβ1–40 (100 μM concentration in PBS) in the absence or presence of equimolar amounts of the indicated lipid-conjugated Ru(iii) complexes upon 24 h incubation time. a) Representative size intensity distribution profiles (diameter, nm); b) average particle size (diameter, nm) of the investigated systems. Data are reported as mean values ± SD for multiple determinations. Upon addition of equimolar amounts of the selected lipid-conjugated Ru(iii) complexes (Fig. 4, blue and red lines), a remarkable shift in particle size distribution was observed, indicating the formation of smaller species with narrower size distributions. Specifically, the peptide/Ru complexes systems showed particle populations of ca. 240 nm for MyriPyRu and 160 nm for PalmiPyRu, suggesting a stronger inhibitory effect on Aβ1–40 aggregation after incubation with PalmiPyRu (Fig. 4), corroborating the previous ThT fluorescence data. Overall, these data support the interaction between the lipid-conjugated Ru(iii) complexes and the peptide, which influences aggregate formation by significantly reducing particle size in solution. These results are also consistent with previous observations obtained testing Ru(iii) complexes containing thiazole-,66 imidazole-67 and pyridine-based72 ligands. NMR analysis with Aβ1–15 To determine the binding ability of these Ru(iii) complexes to Aβ peptide, two complementary spectroscopic methods were utilized: 1H-NMR and UV-vis spectroscopy. The full-length Aβ peptide contains three histidine residues (His-6, His-13, and His-14), located within its hydrophilic N-terminal domain, which have been shown to exhibit strong affinities for various metal ions, such as Cu, Zn, and Fe.73–75 Ruthenium complexes are also known to strongly interact with imidazole groups of histidine residues, both in serum proteins such as albumin76–79 and within the Aβ peptide itself.29,66,67,72,80–82 To investigate whether MyriPyRu and PalmiPyRu effectively coordinate to histidine residues, mixtures of each Ru(iii) complex with the peptide were monitored over time by 1H-NMR spectroscopy analysis. Typically, during standard NMR analysis, signals from the ligands coordinated to the paramagnetic Ru(iii) metal center tend to broaden and/or eventually disappear into the baseline.58,83 This phenomenon has been previously observed for both free paramagnetic metal ions in solution84,85 and metallotherapeutics interacting with β-amyloid peptides.29,66,67,72,80,82 As such, signal broadening in NMR spectra is a well-established indicator of histidine coordination to metal centers in Aβ peptides.29,66,67,72 For these experiments, each ruthenium complex was mixed with an equimolar amount of Aβ1–15 at a concentration of 100 μM in a buffered (PBS) D2O solution and the 1H-NMR spectra of the resulting peptide/Ru mixtures were recorded at different time points over 24 h. As previously reported,29,66,67,72,82,86 the truncated Aβ1–15 variant of the full-length Aβ peptide is an optimal model for NMR studies, as it retains the three N-terminal histidine residues critical for metal binding while being less prone to aggregation in solution,86 thus allowing a clear and direct evaluation of Ru binding. In the spectrum of free Aβ1–15 peptide, the protonic signals of the histidine residues, His-6, His-13, and His-14, appear as sharp, well-resolved peaks in the aromatic region (6.7–8.0 ppm), along with the proton signals from Phe-4 and Tyr-10, consistently with previously published data (Fig. 5, top).29,66,67,72 Expansion of the 8.3–6.5 ppm region of the 1H-NMR spectra of Aβ1–15 in the absence (top) and presence of MyriPyRu and PalmiPyRu, investigated at a 1 : 1 peptide/Ru molar ratio at 24 h in a buffered D2O solution (PBS). The peaks corresponding to the histidine residues His-6, His-13 and His-14 are labelled with red triangles and blue circles, respectively for distal (Hε) and proximal (Hδ) histidine signals. The peaks corresponding to the Tyr-10 and Phe-4 protons are marked as * and #, respectively. Following incubation with MyriPyRu or PalmiPyRu (Fig. 5 and S6), all the imidazole signals of the three histidine residues underwent signal attenuation and a noticeable loss of resolution was detected. Additionally, a subtle upfield shift was detected in the distal histidine signals (Hε) in the presence of both compounds, compared to the free peptide. These findings strongly support the coordination between the paramagnetic Ru(iii) centers and the histidine nitrogen atoms in both tested complexes, in agreement with previous studies on similar compounds.29,66,67,72,82 Interestingly, unlike other Ru(iii) complexes described in the literature,29,66,67,72,82 the treatment with PalmiPyRu led to a loss of resolution not only for proximal and distal histidine signals but also for the Phe-4 peaks (7.15–7.30 ppm), suggesting a possible interaction with this residue as well (Fig. 5 and S6). In contrast, no effect was observed in samples treated with MyriPyRu, even after extended incubation times (Fig. 5). The appearance of the first chemical shift perturbations after approximately 1 h is consistent with the hydrolysis behaviour of these Ru(iii) complexes in a phosphate buffered solution (Fig. S4), in which MyriPyRu and PalmiPyRu display half-life times of ca. 90 and 120 min,44 respectively. These relatively small perturbations could therefore be consistent with early coordination events involving the N-terminal region of Aβ. While MyriPyRu produced detectable perturbations already after 1 h and then the peptide signals remained essentially unchanged during the following incubation period, PalmiPyRu induced a more gradual spectral evolution over time (Fig. S6). This behaviour likely reflects not only the hydrolysis events but also the subsequent peptide recognition and coordination steps, which may proceed with different kinetics. To verify if the intensities of the histidine proton signals decreased upon Ru treatment and to quantify the effect of each Ru(iii) complex on Hδ and Hε peaks, the integrated areas of imidazole proton signals were calculated for both the free peptide and after incubation with each complex. As previously reported, these values were then normalized to the two methyl doublets of the Val-12 residue66,67,72,87 located at approximately 0.8 and 0.9 ppm, which remain unaffected upon Ru addition.66,72,88 This normalization enables a direct comparison of the impact of each Ru compound on the reduction of histidine NMR signal intensities. Taken together, both MyriPyRu and PalmiPyRu induced a decrease in the intensity of both proximal and distal histidine signals (Fig. S7), with PalmiPyRu producing a more pronounced overall effect. UV-vis analysis with Aβ1–15 To gain further insight into the interaction of MyriPyRu and PalmiPyRu with Aβ1–15, UV-vis spectroscopy analysis was performed. Indeed, this technique offers a useful comparison between electronic absorption spectra of the free Ru(iii) complexes in solution and after peptide addition, so that spectral changes can be a useful tool for monitoring coordination with Aβ. Due to the presence of a single tyrosine residue (Tyr-10), the absorbance spectrum of Aβ1–15 shows only a weak signal at 280 nm (Fig. S8), as previously reported.72 Consequently, the main bands observed in the UV-vis spectra of MyriPyRu and PalmiPyRu originate from the Ru complexes centres. Therefore, in parallel experiments, to a 100 μM solution of each Ru complex equimolar amounts of the Aβ peptide were added. Spectra were then recorded after 1 and 24 h incubation. After 1 h incubation, significant changes were observed in the UV-vis spectra of both Ru(iii) complexes (Fig. 6). In the UV region, the absorbance peaks between 230–270 nm and 280–310 nm decreased, resulting in two clear isosbestic points at 234 and 284 nm. In the visible region, a marked decrease in the LMCT band at 400 nm was also detected. All these effects became more pronounced after 24 h incubation (Fig. 6, blue lines). The presence of isosbestic points suggests the conversion of the initial species into distinct Ru-containing species upon interaction with the peptide. Representative overlapped UV-vis absorption spectra of MyriPyRu (a) and PalmiPyRu (b), both at 100 μM concentration in PBS, in the absence and presence of Aβ1–15 at a 1 : 1 peptide/Ru molar ratio at different times, as indicated. Although a similar reduction in the signal over time was observed for MyriPyRu and PalmiPyRu analyzed alone after dissolution in similar saline conditions (Fig. S2 and S3),44 the slower kinetics observed in the presence of Aβ1–15 suggests an interaction of the lipid-conjugated Ru(iii) complexes with the amyloid peptide. Cytotoxicity evaluation of Ru(iii) complexes on human neuroblastoma SH-SY5Y cells The above-described results revealed a promising capability of the tested Ru(iii) complexes, especially MyriPyRu and PalmiPyRu, to effectively inhibit the aggregation of amyloid peptides in vitro. To evaluate their potential biological applicability, we tested the cytotoxicity of AziRu, MyriPyRu, PalmiPyRu, StePyRu and OlePyRu, on human neuroblastoma SH-SY5Y cells, a widely used in vitro neuronal model. CholPyRu was excluded from this analysis because of its limited solubility in cell culture medium at the concentrations required for biological testing. The other Ru(iii) complexes were used at a concentration of 10, 20, or 100 μM to treat SH-SY5Y cells for 48 h, and then a standard metabolic assay was performed to measure the mitochondrial conversion of WST-8 [2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium, monosodium salt] by viable cells. As indicated in Fig. 7, all the tested complexes showed no significant toxicity at the concentrations of 10 and 20 μM, values that are consistent with those displaying marked anti-aggregation activity in the ThT assays. Conversely, all the tested complexes, except for AziRu, displayed significant toxicity at 100 μM concentration, probably due to the effect of their lipidic chains on the delicate membrane structures typical of neuronal cells. Evaluation of cell viability of SH-SY5Y cells after treatment with the various Ru(iii) complexes for 48 h compared to untreated cells. Results are expressed as mean values ± SD of treated compared with untreated cells (set equal to 1); biological triplicate of two independent experiments are shown,* p < 0.01 Student t- test. Conclusions In this study, a series of lipid-functionalized Ru(iii) complexes derived from AziRu were investigated for their ability to modulate the aggregation of amyloidogenic peptides associated with neurodegenerative disorders. Among the tested compounds, MyriPyRu and PalmiPyRu emerged as the most effective inhibitors of Aβ peptide aggregation, displaying marked activity with IC50 values in the low micromolar range, as determined by ThT fluorescence assays. Complementary biophysical and spectroscopic analyses further supported the interaction of these Ru(iii) complexes with Aβ peptides. DLS experiments revealed a significant reduction in aggregate size upon treatment with MyriPyRu and PalmiPyRu, while 1H NMR and UV-vis studies suggested the involvement of histidine residues located in the N-terminal region of Aβ peptides in the interaction with the Ru centers. A particularly interesting outcome of this study is the clear role of the lipid moiety on the anti-aggregation properties of the complexes. While both MyriPyRu and PalmiPyRu exhibited the highest activity, the longer-chain derivative StePyRu, together with OlePyRu and CholPyRu, which also present less flexible appendages, proved to be less effective, indicating that the anti-amyloid activity cannot be simply correlated with the increasing chain length. Rather, both the length and structural features of the lipid chain in terms of overall flexibility appear to play a crucial role in modulating peptide recognition and aggregation inhibition, providing valuable structure–activity relationship information for the rational design of new Ru(iii)-based anti-amyloid agents. Notably, the investigated compounds showed limited cytotoxicity toward SH-SY5Y neuroblastoma cells at concentrations relevant for anti-aggregation activity, supporting their potential biological applicability. Overall, these findings highlight lipid-conjugated Ru(iii) complexes as promising scaffolds for the development of novel anti-amyloid agents targeting pathological peptide aggregation in Alzheimer's disease. More importantly, they demonstrate that lipid functionalization represents an effective strategy to tune the anti-aggregation properties of NAMI-A-type Ru(iii) complexes through modulation of the lipid chain properties. Future studies will be directed toward elucidating the molecular basis underlying the anti-aggregation activity of these Ru(iii) complexes, identifying the biologically relevant Ru species formed under physiological conditions, evaluating the possible contribution of redox-related processes, and assessing their ability to protect neuronal cells from Aβ-induced toxicity in more advanced biological models, while further optimizing the structural features responsible for their biological activity. Experimental section Materials and general methods Tablets of phosphate-buffered saline (PBS), Thioflavin T and 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP) were purchased from Merck Life Science s.r.l. PBS solutions (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4, pH = 7.3) were prepared by dissolving a PBS tablet in Milli-Q water and then filtered through 0.22 μm filters. Peptides preparation β-Amyloid (1–40), β-amyloid (1–15), amyloid-forming peptide Sup35p and α-synuclein were purchased from Tebubio Italia. Peptide powders were dissolved in HFIP (to have them as monomers), fractionated, lyophilized and stored at −20 °C until use, following literature procedures.89 Then, each aliquot was dissolved in the proper PBS volume to reach the desired concentration for each experiment. Ru(iii) complexes preparation The lipid-conjugated Ru(iii) complexes44 and AziRu43,90 were prepared according to previously reported protocols. For all the compounds, the obtained spectroscopic (1H-NMR) and spectrometric (ESI-MS) data were in perfect agreement with those reported in literature. • MyriPyRu 1 H NMR [400 MHz, (CD 3 ) 2 CO]: significant signals at δ = −1.61 (very broad signal, Py-Ru); −12.78 (very broad signal, CH3 DMSO). ESI-MS (negative ions): calculated for C22H39Cl4NO3RuS: 640.50; found m/z: 640.8 (M−). • PalmiPyRu 1 H NMR [400 MHz, (CD 3 ) 2 CO]: significant signals at δ = −1.62 (very broad signal, Py-Ru); −12.76 (very broad signal, CH3 DMSO). ESI-MS (negative ions): calculated for C24H43Cl4NO3RuS: 668.55; found m/z: 668.9 (M−). • StePyRu 1 H NMR [400 MHz, (CD 3 ) 2 CO]: significant signals at δ = −1.61 (very broad signal, Py-Ru); −12.71 (very broad signal, CH3 DMSO). ESI-MS (negative ions): calculated for C26H47Cl4NO3RuS: 696.60; found m/z: 696.9 (M−). • OlePyRu 1 H NMR [400 MHz, (CD 3 ) 2 CO]: significant signals at δ = −1.62 (very broad signal, Py-Ru); −12.84 (very broad signal, CH3 DMSO). ESI-MS (negative ions): calculated for C26H45Cl4NO3RuS: 694.59; found m/z: 694.8 (M−). • CholPyRu 1 H NMR [400 MHz, (CD 3 ) 2 CO]: significant signals at δ = −1.60 (very broad signal, Py-Ru); −12.82 (very broad signal, CH3 DMSO). ESI-MS (negative ions): calculated for C39H61Cl4NO5RuS: 898.85; found m/z: 899.9 (M−). Concentrated stock solutions of all Ru compounds were prepared at 1–3 mM concentration in CH3CN (HPLC grade, Merck Life Science S.r.l.) to minimize premature hydrolysis of the labile Ru–Cl bonds during sample preparation and storage. Apart from UV-visible studies over time reported in Fig. S4, immediately before each experiment, suitable aliquots were withdrawn from these stock solutions and diluted with PBS buffer at the desired concentration, always maintaining in the final working solutions an acetonitrile concentration well lower than 1%. UV-vis experiments UV-vis measurements were performed on a JASCO V-750 UV-vis spectrophotometer using 1 cm path length cuvettes (1 mL internal volume, Hellma). Spectra were recorded at r.t. in the 250–600 nm range with a response of 0.24 s, a scanning speed of 200 nm min−1 and a 2.0 nm bandwidth, with appropriate baseline subtraction. All the UV-vis spectra were averaged over 3 scans and each experiment performed in triplicate. The target compounds were analyzed from freshly diluted samples at 100 μM concentration in CH3CN. UV-vis spectra were recorded over time up to 24 h. ThT-based fluorescence assays A 1.4 mM stock solution of thioflavin T was prepared in double distilled water using the molar extinction coefficient of 36 000 M−1 cm−1 at 412 nm and then filtered with 0.2 μm Millipore filters according to a previously reported procedure.91,92 Fluorescent ThT-based measurements were performed with a Varioskan LUX microplate reader. Samples were plated on black polystyrene 96-well plates (Life Technologies Italia), where the final volume of each well was 150 μL. Aβ1–40, Aβ1–15, Sup35p7–13 and α-syn were tested at 10, 20, 50 and 2 μM concentration in PBS, respectively. All samples were prepared with a peptide/ThT/Ru complex in 1 : 1 : 1 molar ratio.66,67 Mixtures were prepared in triplicate, with appropriate control samples for each Ru complex in the absence of Aβ. Fluorescence of the samples was measured at 37 °C from the top of the plate (without lid) using λexc and λem of 440 nm and 495 nm,66,67 respectively, and using an excitation bandwidth of 12 nm. ThT fluorescence of the samples containing both peptides and Ru(iii) complexes was compared to the fluorescence emission of samples with only the peptides to determine the relative percentage of aggregation. For all the tested peptides, end-point fluorescence values were recorded after 24 h of incubation. The plate was shaken in continuous mode at 600 rpm for 5 s before each measurement. Each experiment was repeated in triplicate, and data are reported as mean value ± SD. Statistical significance was assessed by using Student's t-test: **** p < 0.0001, *** p < 0.001, ** p < 0.01, * p < 0.05 for Ru complexes relative to the sole peptide. Similar settings were also used to determine the IC50 values of MyriPyRu and PalmiPyRu on Aβ1–40 aggregation. The amyloid peptide was tested at 10 μM concentration in 1 : 1 molar ratio with ThT, and Ru(iii) complexes were analyzed in the 1 → 30 μM concentration range. Then, the assay was performed as described above. IC50 values were calculated, as previously reported,93 by fitting the data with a decay equation: y = a × e−x/t + b, where a is the initial amplitude of the decay, t is the decay constant and b is the asymptotic value. Each experiment was repeated in duplicate, and data are reported as mean value ± SD. DLS analysis on Aβ1–40 To analyze the size of the aggregates formed by Aβ1–40 in the absence and presence of the selected Ru(iii) complexes, DLS measurements were recorded on a Zetasizer Nano ZS (Malvern Instruments) using 12 mm square polystyrene cuvettes (DTS0012, Malvern Instruments). Samples were prepared by adding equimolar amounts of each lipid-conjugated Ru(iii) complex to the Aβ1–40 peptide, prepared at 100 μM concentration in PBS, and then filtered through a 0.2 μm membrane filter (Merck Life Science S.r.l.), following described protocols.66,67,72 All the measurements were carried out after 24 h of incubation at r.t. using a scattering angle of 173° and an equilibration time of 60 s, as previously reported.44,94,95 Size values (expressed in diameter) were directly obtained from the measurements fitting the correlation functions with the cumulant analysis algorithm as implemented by the Malvern software (ISO 13321). Each experiment was repeated in duplicate. Size values are reported as mean values ± SD for multiple determinations. 1H-NMR spectroscopy experiments with Aβ1–15 Sample preparation for 1H-NMR spectroscopy followed previously described protocols29,66,67,72 with minor modifications. Briefly, equimolar amounts of the Aβ peptide and the desired Ru complex were mixed in PBS to have a 100 μM concentration for both components. The samples were then rapidly taken to dryness using an Eppendorf Concentrator Plus (Eppendorf srl, Milan, Italy) and re-dissolved in 600 μL of D2O (Merck LifeScience s.r.l.) immediately before NMR acquisition, thereby removing the residual acetonitrile from the stock solutions while maintaining identical buffer composition and the overall concentration of compounds. Thereafter, the spectra of the peptide/Ru mixtures were recorded at different time points up to 24 h on a Bruker Avance III HD 400 MHz spectrometer. UV-vis spectroscopy analysis with Aβ1–15 For this analysis, the peptide was added to each Ru complex in 1 : 1 molar ratio. The resulting mixtures were immediately diluted with PBS to reach a final concentration of 100 μM for both Aβ15 and each Ru complex, overall containing less than 1% acetonitrile. As a reference, the spectrum of Aβ1–15 alone in PBS was also recorded. All UV-vis spectra were acquired after 1 and 24 h of Ru/peptide incubation on a JASCO V-750 UV-vis spectrophotometer. Measurements were carried out in 1 cm path length cuvettes (1 mL internal volume, Hellma) at r.t., in the 225–600 nm range, with a scanning speed of 100 nm min−1 and a bandwidth of 1.0 nm. The appropriate baseline was subtracted. Each spectrum represents the average of three scans, and all three experiments were performed in duplicate. Cell cultures and cytotoxicity assay SH-SY5Y cells, kindly provided by Prof. Claudia Crosio (University of Sassari, Italy), were cultured in RPMI-1640 (Euroclone) with 10% fetal bovine serum (FBS; Hyclone, Fisher Scientific) and 2 mM l-glutamine (Euroclone), and maintained at 37 °C in a saturated humidity atmosphere containing 95% air and 5% CO2. Cell viability was evaluated by measuring cell metabolic activity using the Cell Counting Kit-8 assay (Dojindo Laboratories Co, Ltd, Japan). Specifically, cells were seeded in 24 well plates at 5 × 104 cells per well and, 24 h after seeding, they were incubated with the different Ru(iii) complexes. After 48 h treatment with the metal-based compounds, cells were incubated with WST-8 [2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium, monosodium salt] (10 μL + 200 μL RPMI per well) for 2–3 h at 37 °C under 5% CO2. Subsequently, the resulting coloured solution was quantified by measuring absorbance at 450 nm with a Microplate Reader according to the manufacturer's protocol. Background absorbance of wells containing culture medium and WST-8 reagent without any cells or wells containing cells and culture medium without reagent were measured and were subtracted from sample readings. Conflicts of interest There is no conflict of interest to declare. Supplementary Material