Curcumin: A multi-Target disease-modifying agent for late-stage transthyretin amyloidosis

1 Scientific RepoRts | 6:26623 | DOI: 10.1038/srep26623 www.nature.com/scientificreports Curcumin: A multi-target diseasemodifying agent for late-stage transthyretin amyloidosis Nelson Ferreira1,2, Nádia P. Gonçalves1,2,3, Maria J. Saraiva1,2,3 & Maria R. Almeida1,2,3 Transthyretin amyloidoses encompass a variety of acquired and hereditary diseases triggered by systemic extracellular accumulation of toxic transthyretin aggregates and fibrils, particularly in the peripheral nervous system. Since transthyretin amyloidoses are typically complex progressive disorders, therapeutic approaches aiming multiple molecular targets simultaneously, might improve therapy efficacy and treatment outcome. In this study, we evaluate the protective effect of physiologically achievable doses of curcumin on the cytotoxicity induced by transthyretin oligomers in vitro by showing reduction of caspase-3 activity and the levels of endoplasmic reticulumresident chaperone binding immunoglobulin protein. When given to an aged Familial Amyloidotic Polyneuropathy mouse model, curcumin not only reduced transthyretin aggregates deposition and toxicity in both gastrointestinal tract and dorsal root ganglia but also remodeled congophilic amyloid material in tissues. In addition, curcumin enhanced internalization, intracellular transport and degradation of transthyretin oligomers by primary macrophages from aged Familial Amyloidotic Polyneuropathy transgenic mice, suggesting an impaired activation of naïve phagocytic cells exposed to transthyretin toxic intermediate species. Overall, our results clearly support curcumin or optimized derivatives as promising multi-target disease-modifying agent for late-stage transthyretin amyloidosis. The accumulation of misfolded proteins as insoluble amyloid fibrils is a key pathognomonic feature of many neurodegenerative diseases, including Familial Amyloidotic Polyneuropathy (FAP), type II diabetes, Alzheimer’s, Parkinson’s and Huntington’s diseases. In most of these pathologies, the end stage aggregated species that accumulate in tissues are composed mostly by 6–10 nm twisted β -pleated-fibrils, in which the polypeptide chain is arranged in sheets perpendicular to the fibril axis being hydrogen bonding parallel1. Although amyloid fibrils have been reported to be toxic2, accumulating evidence suggests that the non-fibrillar oligomer intermediaries are the primary culprits for neurodegeneration and exert their toxicity through distinct mechanisms3. Transthyretin (TTR) is an abundant plasma protein that is mainly synthesized by the liver and choroid plexuses of the brain. TTR in its homotetrameric structure acts as a carrier protein for thyroid hormones and retinol-binding protein in plasma and cerebrospinal fluid4. Extracellular TTR aggregation and deposition triggers inflammation and oxidative damage5, disruption of calcium homeostasis6, extracellular matrix remodeling7, activation of heat shock response and endoplasmic reticulum (ER) stress pathways6,8, including common molecular actors and scenarios, that resemble in many features those associated to local amyloidoses affecting the central nervous system such as Alzheimer’s and Parkinson’s diseases. FAP is a fatal neurodegenerative disorder characterized by the extracellular deposition of aggregates and fibrils of mutant forms of TTR, particularly in peripheral nerves and ganglia of the peripheral nervous system. Although the most common TTR mutation leading to FAP gives origin to a substitution of methionine for valine at position 30 (TTR V30M), more than one hundred amyloidogenic TTR variants have been described. Different mutations in the TTR gene are associated with distinct age of onset, disease penetrance, clinical phenotype, prognosis and clinical outcome9. In addition to familial TTRassociated forms of amyloid, senile systemic amyloidosis (SSA) is a nonhereditary, late onset sporadic form of TTR amyloidosis that is typically related with wild-type TTR amyloid deposition in the heart, leading to cardiac dysfunction and ultimately death10. 1IBMC - Instituto de Biologia Molecular e Celular, Universidade do Porto, Rua Alfredo Allen, 208, 4200 – 135 Porto, Portugal. 2i3S – Instituto de Investigação e Inovação em Saúde da Universidade do Porto, Rua Alfredo Allen, 208, 4200 – 135 Porto, Portugal. 3ICBAS, Instituto de Ciências Biomédicas Abel Salazar, Universidade do Porto, Rua Jorge Viterbo Ferreira 228, 4050 – 313 Porto, Portugal. Correspondence and requests for materials should be addressed to M.R.A. (email: [email protected]) received: 01 December 2015 accepted: 04 May 2016 Published: 20 May 2016 OPEN www.nature.com/scientificreports/ 2 Scientific RepoRts | 6:26623 | DOI: 10.1038/srep26623 Through a combination of compound library screens and epidemiological data, many low molecular weight organic compounds have been suggested as effective therapeutic and prophylactic agents against amyloidosis. The most striking aspect that has emerged from recent studies is that some of these small molecules, in particular natural occurring polyphenols, do not act on a single target in the multi-step amyloidogenic cascade; rather, they seem to inhibit/redirect different steps of the amyloid molecular dynamics and also modulate the biology of cells and tissues where amyloid deposition takes place11,12. Among those polyphenols, curcumin, a natural occurring diarylheptanoid isolated from Curcuma longa, has been found to modulate oligomerization and fibril formation of several amyloid proteins, including TTR12,13, α -synuclein14 and amyloid-β 15. Curcumin has anti-oxidant and anti-inflammatory properties that have also been suggested to play a protective role in a variety of chronic pathologies such as cancer, atherosclerosis, and neurodegenerative diseases16. In addition, curcumin can efficiently chelate amyloid associated metal ions, such as Fe2+, Zn2+ and Cu2+ which may reduce amyloid aggregation or oxidative neurotoxicity in vivo17. Furthermore, curcumin is widely used as therapeutic agent in Chinese and Indian Ayurvedic medicine and is “generally recognized as safe” by the U.S. Food and Drug Administration. Epidemiological studies conducted in India, where curcumin is highly consumed, showed that Alzheimer’s disease incidence is significantly lower than that reported from the western world18. These observations together with the findings that curcumin disaggregates TTR amyloid fibrils in vitro11 and activates innate immune system promoting amyloid clearance19, prompted us to investigate the effect of curcumin supplementation in a late-stage mouse model for FAP. Results Curcumin reduces ER stress and protects Schwann cells from entering into the apoptotic signaling pathway upon exposure to TTR aggregates. We investigated the impact of curcumin on TTR aggregate-induced ER stress response and apoptosis by exposing RN22 cells to TTR oligomers formed in the absence or presence of different curcumin concentrations. As expected from previous work6, TTR oligomers, but not soluble TTR, triggered a remarkable increase in the intracellular amount of BiP, the central regulator for the ER-stress response20, indicating cellular toxicity (Fig.1). In strong contrast, we observed no such effect when identical amounts of oligomers formed in the presence of curcumin were added to the RN22 cells, indicating that these species are less toxic than untreated oligomeric intermediaries (Fig.1). Moreover, exposure of cells to Figure 1. Curcumin inhibits TTR oligomers-dependent activation of ER-chaperone BiP. (A) Anti-BiP Western blot of protein extracts from Schwannoma cells exposed to soluble TTR Y78F (TTR) or to TTR Y78F oligomers (Oligomers) or TTR Y78F oligomers pre-treated with curcumin for 24 h (final concentration of protein in cell culture medium 2 μ M and curcumin 0.001–11 μ M). In parallel, cells were incubated with curcumin alone, at highest concentration used (Curc. 11 μ M). (B) Bar graph represents quantification of data and is presented as mean ± SEM (*p < 0.05; **p < 0.01; ***p < 0.001). www.nature.com/scientificreports/ 3 Scientific RepoRts | 6:26623 | DOI: 10.1038/srep26623 curcumin alone (up to 11 μ M) did not interfere with BiP expression along the range of concentrations used in the assay. In addition, toxicity of TTR aggregate intermediaries was also studied through a standardized caspase-3 assay in this Schwann cell line. Unlike soluble TTR, TTR oligomeric species induced substantial caspase-3 activity in RN22 cells, indicative of cell damage and entry into the apoptotic signaling pathway. On the other hand, cell exposure to oligomers formed in the presence of curcumin resulted in a significant dose-dependent reduction of caspase-3 activity (Fig.2). Moreover, no significant alteration in this enzyme activity was noted when cells were incubated with curcumin alone (11 μ M) (Fig.2). Evaluation of curcumin effect in a late-stage FAP mouse model. In view of the promising data described above and available in the literature concerning in vitro and in vivo studies12,13,21, we investigated the effect of dietary supplementation of curcumin on TTR amyloid formation using a late-stage FAP mouse model expressing the amyloidogenic human TTR V30M variant on a Hsf-1 heterozygous background (hTTR V30M/Hsf)8. The impairment of Hsf-1 expression leads to an extensive and early deposition of non-fibrillar TTR in several tissues, including the gastrointestinal tract and the peripheral nervous system. Non-fibrillar TTR species start to deposit between 3–6 months of age and gradually evolve to congophilic material, typically after 12 months of age. Therefore, this mouse model is highly relevant for testing new therapeutic strategies targeting different stages of the pathology. In the current study, we aimed at evaluating the effect of curcumin in a late-stage disease in which deposition of non-fibrillar TTR coexists with birefringent congophilic material in tissues. Protocol design, drug dosage and selection of endpoints were based on information from a previous study13. As expected, curcumin supplementation was very well-tolerated in aged mice and did not produce any observable adverse side effects. No alteration was observed in body weight or mortality between animals treated with curcumin and age-matched controls. In addition, no histological abnormalities were observed in liver sections stained with hematoxylin and eosin for morphologic assessment (data not shown). Curcumin binds to plasma TTR at the thyroxine binding sites and increases its resistance to dissociation. To confirm curcumin binding to TTR in vivo, plasma from mice fed with curcumin-enriched diet or standard diet (control) was incubated with radiolabeled thyroxine (T4) ([125I]-T4) and subjected to gel electrophoresis under native conditions. T4-binding proteins were visualized by phosphor imaging analysis, as shown in Supp. Fig. 1. The results showed that samples from curcumin treated mice presented considerably less intense TTR bands, indicating that after gut absorption and systemic distribution, curcumin selectively competes with T4 for its binding to TTR (41.74% competition) (Supp. Fig. 1). Densitometry analysis of the IEF gels demonstrated Figure 2. Curcumin reduces caspase-3 activity associated with extracellular TTR oligomers. Activity of caspase-3 was measured in Schwannoma cells exposed to soluble TTR Y78F (TTR) or to TTR Y78F oligomers (Oligomers) or TTR Y78F oligomers pre-treated with curcumin for 24 h (same conditions as indicated in Fig.1) (**p < 0.01; ***p < 0.001; N.S. = non-significant). www.nature.com/scientificreports/ 4 Scientific RepoRts | 6:26623 | DOI: 10.1038/srep26623 that curcumin chronic supplementation significantly increases TTR resistance to dissociation when compared with plasma TTR from untreated animals under the tested conditions (Supp. Fig. 2). Curcumin decreases TTR deposition and rescues tissue toxicity. TTR levels in plasma were quantified by radial immunodiffusion and showed no statistical difference between mice fed with curcumin-enriched diet and control mice (556.8 ± 5.1 μ g TTR/ml and 556.2 ± 5.5 μ g TTR/ml, respectively), indicating that curcumin administration did not affect TTR liver synthesis and turnover in vivo. The effect of curcumin supplementation on TTR deposition and tissue damage was analyzed by SQ-IHC and western blotting in different target organs. At the end of treatment, mice were 15.5 months of age thus, as expected, control animals showed widespread TTR deposition throughout the interstitial connective tissue of DRG, surrounding the perikaryon and in close contact with satellite glial cells8. In strong contrast, sub-chronic administration of curcumin inhibited TTR deposition in ganglia associated with spinal nerves, as denoted in Figs3 and 4. Similar results were found in stomach (Figs5 and 6). Since extracellular accumulation of TTR aggregates closely correlates with cellular damage and toxicity5,22, we next investigated the effect of curcumin treatment on several molecular markers associated with the disease in the peripheral nervous system and gastrointestinal tract. In both DRG (Figs3 and 4) and stomach (Figs5 and 6), we found reduced levels of ER-chaperone BiP. In addition, reduced levels of Fas death receptor were observed in curcumin treated mice compared to controls (Fig.3). Supporting these observations, several key pro-inflammatory mediators associated with disease pathogenesis, namely NF-κ B, IL-1β and TNF-α 8, were found reduced in DRG of curcumin treated mice (Fig.3). Overall, these results sustained that curcumin modulates TTR deposition in both gastrointestinal tract and peripheral nervous system. Curcumin remodels amyloid deposits and enhances its clearance. Previous data has shown that curcumin disaggregates TTR fibrils in vitro12. This prompted us to investigate the effect of curcumin in amyloid deposition and extracellular matrix remodeling. The presence of amyloid was detected by Congo red staining in stomach, the primary amyloid target-organ in this particular mouse model. The results showed that 7 out of 14 (50%) control animals exhibited Congo Red-positive material, whereas only 1 out of 12 (8%) curcumin treated animals presented congophilic amyloid deposits (Fig.5). These results were corroborated by the reduced levels of MMP-9 (biomarker associated with amyloid extracellular deposition) in stomach (Fig.5), thus clearly indicating that curcumin remodels TTR amyloid deposits and extracellular matrix in vivo. Curcumin increases internalization and intracellular transport of TTR aggregates in macrophages. Curcuminoids are known to improve the function of the innate immune system by increasing clearance of aggregates in several amyloid diseases23,24. Therefore, we further investigated the effect of curcumin in bone marrow derived macrophages exposed to TTR aggregates. Macrophage culture purity was confirmed for all time points analyzed, by staining with F4/80, a well-characterized and extensively referenced mature macrophage marker25,26. In control macrophages (treated with vehicle) uptake of extracellular TTR aggregates was scarce (Fig.7). In addition, intracellular transport of TTR aggregates progressed very slowly as shown by limited co-localization between those and EEA1 (t = 2 h, Fig.7) or LAMP-1 (t = 24 h, Fig.7). Treated macrophages were exposed to curcumin before addition of TTR aggregates to the cell culture medium. Using confocal microscopy, we found that curcumin stimulation resulted in rapid internalization of TTR aggregates and lysosomal degradation as shown by co-localization with EEA1 (t = 2 h, Fig.7) and LAMP-1 (t = 24 h, Fig.7), respectively. The optimal dose of curcumin (1 μ M) was determined using a wide range of concentrations (0.01–100 μ M). Concentrations higher than 1 μ M did not increase aggregates internalization by macrophages. Discussion The working hypothesis for the TTR amyloid cascade, supported by extensive work and literature, led to the notion that stabilizing TTR native structure could be a promising therapeutic approach to stop the production of toxic TTR amyloidogenic species. Since FAP and other TTR-related amyloidoses are progressive and complex multi-system disorders, therapeutic strategies aiming multiple molecular targets simultaneously, either through combination therapies or multi-target-directed compounds, could improve therapy efficacy and treatment outcome. Curcumin has a symmetric and extended conjugated molecular structure that resembles the structure of Congo red, the classic gold standard for detecting amyloid deposits21. In addition, curcumin is able to efficiently cross the blood-brain barrier, which is an essential feature for therapies aiming amyloid deposition in central nervous system27,28. Concerning TTR amyloidosis, previous work from our group has shown that, in vitro, curcumin binding to the largely unoccupied thyroxin binding pockets in the native TTR structure increases its resistance to dissociation in non-native monomers that polymerize into toxic aggregates12. Very recently, the crystal structure of the complex formed between TTR and curcumin has been determined and it corroborates our initial findings by showing that intact curcumin binds at the two T4 binding sites located in the TTR tetramer central channel29. In the present study, we investigated the effect of curcumin supplementation in a late-stage mouse model for FAP. After gut absorption and systemic distribution, curcumin selectively interacted with TTR in plasma, in the largely unoccupied T4 binding pockets, increasing TTR stability. In addition, curcumin administration significantly lowered TTR non-fibrillar deposition and toxicity in tissues, in particular along the gastrointestinal tract and DRG. These results confirm the previous study in early stage FAP mice13 and are in accordance with the www.nature.com/scientificreports/ 5 Scientific RepoRts | 6:26623 | DOI: 10.1038/srep26623 Figure 3. Curcumin decreases TTR deposition and associated toxicity in DRG of aged hTTR V30M/ Hsf mice. Representative images of immunohistochemistry analysis of TTR, BiP, Fas death receptor, cleaved caspase-3, NF-κ B, IL-1β and TNF-α in DRG of mice treated with curcumin (right panels; n = 12) and agematched controls (left panels; n = 14); 40× magnification. Bar graphs: quantification of immunohistochemistry images is represented as percentage of area occupied ± SEM (***p < 0.001). These observations were supported by Western blot analysis of TTR levels after normalization to GAPDH signal intensities (Fig.4). www.nature.com/scientificreports/ 6 Scientific RepoRts | 6:26623 | DOI: 10.1038/srep26623 Figure 4. (A) Representative anti-BiP and anti-TTR Western blots of DRG from curcumintreated (n = 6) and control mice (n = 6). (B) Bar graphs illustrate normalized BiP/GAPDH and TTR/GAPDH density quantifications ± SEM (***p < 0.001). Similar results were found throughout the GI tract, in particular in stomach (Figs5 and 6). Figure 5. Curcumin decreases TTR aggregates deposition and associated toxicity in the stomach of aged hTTR V30M/Hsf mice. Representative images of immunohistochemistry analysis of TTR, BiP, MMP-9 and also Congo red staining of stomach sections of mice treated with curcumin (right panels; n = 12) and age-matched controls (left panels; n = 14); 20× magnification. Bar graphs: quantification of immunohistochemistry images is represented as percentage of area occupied ± SEM (***p < 0.001). www.nature.com/scientificreports/ 7 Scientific RepoRts | 6:26623 | DOI: 10.1038/srep26623 working hypothesis for TTR amyloidosis by which stabilizing TTR native fold by small molecules would block protein aggregation and toxicity30. Moreover, after Congo red birefringence analysis of stomach sections, we found that curcumin treated mice presented fewer and noticeably smaller congophilic deposits as compared to control mice. These results were supported by a substantial reduction on the levels of MMP-9 in treated mice and clearly indicate that curcumin promotes amyloid remodeling, reabsorption and extracellular matrix recovery. Our observations corroborate the results obtained for other amyloidoses since, for instance, administration of dietary curcumin to aged mouse models of AD resulted in labeling Aβ plaques and reducing amyloid levels and plaque burden28,31. Though our data indicates that curcumin modulates the TTR cascade by direct interaction with TTR aggregate intermediates, we speculate whether the diverse array of molecular targets for curcumin might potentiate its neuroprotective effect and improve treatment outcome in vivo. For instance, since high concentrations of Zn2+ and Cu2+ can trigger TTR amyloid formation32, it seems reasonable to speculate whether curcumin known ability to form complexes with these ions17 might contribute to inhibit abnormal TTR aggregation and toxicity in vivo. Moreover, compelling evidence indicates that curcumin is able to suppress inflammation through multiple signaling pathways33,34. Thus, reduction of the pro-inflammatory response to TTR deposition in DRG, observed in curcumin treated mice, might be attributed not only to curcumin anti-amyloidogenic effects but also to its well-known capacity to block the production of key mediators of inflammation such as NF-kB, TNF-α, or IL-1β 33. However, curcumin has been recognized mainly as an anti-oxidant compound35–37. In accordance, a recent study by Moustapha et al.38 revealed that this is the predominant effect at very low doses of curcumin (lower than 1 μ M) while at higher concentrations, 10 μ M, it starts inducing early events of autophagy and at even higher dosages (> 25 μ M) it induces apoptosis and endoplasmic reticulum stress causing calcium release, with destabilization of the mitochondrial compartment and promoting tumor cell death38,39. This hormetic characteristic of curcumin has been detailed using a tumor cell line namely, hepatocellular carcinoma Huh-7 cells38. Nonetheless, molecular effects of curcumin on complex cell systems might most likely be cell-type specific in a concentration and time-dependent manner. Defects in the autophagic machinery or mitochondrial dynamics might play a pathological role in neurodegenerative disorders40 and might also be impaired in FAP. In fact, it was recently described that curcumin is able to decrease monomeric TTR, by promoting LC3 cleavage, thus enhancing autophagy, in a cell culture system41. Importantly, studies of Hsf-1 knock-out mice suggest that Hsf-1 might play a role in the regulation of autophagy activity42. Therefore, future studies disclosing the role of autophagy in FAP should also encompass the use of other animal models. Transthyretin, either wild-type or mutant, can be internalized by diverse types of cells, including hepatocytes, mouse embryonic fibroblasts, yolk sac cells, sensory neurons and glial cells43–46. Recent work from Misumi and colleagues has shown that fibroblasts endocyte and degrade TTR aggregates in vitro and in vivo47. All together these studies provide evidence for the complex cellular dynamics underlying TTR catabolism and clearance and its impact in the FAP pathology. We speculate that cellular impairment of one or more of these mechanisms might shift the amyloid formation/degradation equilibrium towards pathogenic TTR aggregate accumulation in the extracellular matrix. Moreover, TTR aggregates internalization might also be deleterious to cells leading to death and organ dysfunction, a topic that will need further investigation. During the last years, considerable focus has been given to the role of the immune system on amyloid pathologies. Macrophages and microglia are the innate immune cells responsible for clearance of pathogens and waste products. Curcuminoids have been reported to enhance phagocytosis of Aβ by upregulating the transcription of 1,4-mannosyl-glycoprotein 4-β -Nacetyl glucosaminyl transferase and other genes, including Toll-like receptors, typically down-regulated in mononuclear cells of AD patients19. Similarly, we found that pre-exposition of Figure 6. (A) Representative anti-BiP and anti-TTR Western blots of stomachs from curcumin-treated (n = 6) and control mice (n = 6). (B) Bar graphs: normalized BiP/GAPDH and TTR/GAPDH density quantifications ± SEM (***p < 0.001). www.nature.com/scientificreports/ 8 Scientific RepoRts | 6:26623 | DOI: 10.1038/srep26623 macrophages isolated from aged hTTR V30M/Hsf mice, to physiologically achievable doses of curcumin (1–10 μ M), improved their phagocytic uptake and degradation of extracellular TTR amyloid aggregates compared to control macrophages (exposed to vehicle). Thus, our observations indicate that aged FAP macrophages present deficient phagocytic ability, suggesting a possible impairment in their functions. In fact, it has been previously Figure 7. Curcumin treated macrophages from FAP mice (hTTR V30M/Hsf mice) present increased internalization and degradation of TTR aggregates. Representative pictures of double immunofluorescence labeling for TTR (green) and EEA1 (red), TTR (green) and F4/80 (red) or TTR (green) and Lamp-1 (red), denoting TTR intracellular aggregates colocalizing with endosomes and lysosomes in curcumin treated mice (arrows). Superposition of the labels, with DAPI (blue), is shown. 63× magnification. Data presented is representative of three independent experiments. Treated macrophages were pre-exposed to curcumin before addition of TTR aggregates to cell culture medium and, by double immunofluorescence confocal microscopy, we found that preexposition of macrophages with curcumin resulted in rapid internalization of TTR extracellular aggregates and lysosomal degradation as shown by co-localization of TTR oligomers and EEA1 (t = 2 h, Fig. 7) or LAMP-1 (t = 24 h, Fig. 7), respectively. www.nature.com/scientificreports/ 9 Scientific RepoRts | 6:26623 | DOI: 10.1038/srep26623 demonstrated that nerve biopsies from FAP patients did not display innate cellular infiltrate surrounding amyloid deposits, as should be expected5 and that macrophage density is downregulated in a FAP mouse model after an injured stimulus26. Despite the molecular mechanisms underlying the clearance impairment of macrophage are mostly unknown, the current study shows that chronic administration of curcumin may modulate the innate immune system response to extracellular accumulation of TTR and provide a previously uncharacterized approach for FAP therapy. Future studies will further address the role of macrophages in FAP pathology using different FAP mouse lines at different stages of TTR amyloid formation. Currently, a diverse array of emerging disease-modifying agents is under analysis in human clinical trials38. From TTR stabilizers (diflunisal, tafamidis), gene therapies to suppress TTR expression (siRNAs) and amyloid fibrils disruptors (doxycycline/TUDCA), the perspectives for an effective therapy for TTR amyloidosis seem more encouraging nowadays than ever before. Nevertheless, much is still unknown regarding long-term safety and efficacy of such approaches and, with the exception of doxycycline/TUDCA therapy, most of them target only early stages of the disease. Our results suggest curcumin as an interesting compound for late-stage TTR amyloidosis. Taking into account its pluripotency, high tolerability, long history of use, and inexpensive cost, we speculate that its combined administration with available therapies might boost treatment efficacy and outcome for advanced amyloidosis. In conclusion, the present work demonstrates that curcumin inhibits TTR aggregation in a dose dependent manner and enhance TTR aggregates clearance by macrophages in vitro. When administrated to aged FAP mice, curcumin not only reduced TTR burden and toxicity but also remodeled congophilic material in tissues. Overall, our results clearly support curcumin or optimized derivatives as promising multi-target disease-modifying agent for late-stage amyloidosis. Materials and Methods Curcumin. Curcumin was purchased from Sigma-Aldrich, St. Louis, MO, USA. Curcumin had ≥ 80% purity (HPLC), with total curcuminoids ≥ 94% (HPLC). Production of recombinant TTR variants. Recombinant TTR V30M and TTR Y78F production, isolation and purification were performed as previously described48. Preparation of fluorescent TTR oligomers. Soluble TTR V30M in PBS was filtered through 0.2 μ m Anotop syringe filters (Whatman, England) and labeled with the fluorescent dye Alexa Fluor 488 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. TTR aggregates were generated by incubating the protein (2 mg/ml), with stirring, at room temperature for 7 days6. Preparations were analyzed by dynamic light scattering (DLS) at 25 °C in a Malvern Zetasizer Nano ZS (Malvern, Worcestershire, UK) as previously reported, to confirm TTR pathogenic aggregation11. Cell culture assays to assess ER stress and apoptosis. Soluble TTR was dissolved at 2 mg/ml in PBS and curcumin stocks were prepared in dimethyl sulfoxide (DMSO). Soluble TTR was mixed with different concentrations of curcumin (0.02–200 μ M) or solvent alone (DMSO). A sample containing only curcumin 200 μ M was also prepared. All samples were kept in Eppendorf amber tubes (for light protection) and incubated for 6 days at 37 °C to allow aggregation. These samples were then diluted (18× ) in cell culture medium to attain a concentration of 2 μ M TTR and increasing concentrations of curcumin ranging 0.001–11 μ M. Following, these samples were added to the cells as following described. Rat Schwannoma cells (RN22) (American type Cell Collection) were propagated and maintained as described previously11. Briefly, 80% confluent cells in Dulbecco’s minimal essential medium (DMEM) supplemented with 1% fetal bovine serum were exposed, for 24 h, with: i) vehicle (DMSO); ii) 2 μ M of freshly prepared soluble TTR; and the diluted samples prepared above, namely iii) TTR Y78F oligomers or iv) curcumin-induced TTR oligomers or v) 11 μ M of curcumin alone. After treatment, cells were trypsinized and lysed using ice-cold lysis buffer containing 5 mM ethylenediamine tetraacetic acid, 2 mM ethylene glycol tetra acetic acid, 20 mM 3-(N-morpholino) propane sulfonic acid, 1% Triton X-100, 1 mM phenyl methane sulfonyl fluoride (PMSF) and a protease inhibitor mix (GE Healthcare). Cell lysates were used for determination of BiP intracellular levels and caspase-3 activity. Protein concentration in lysates was determined using a Bio-Rad protein assay kit. Caspase-3 activity was accessed using the CaspACE fluorimetric 96-well plate assay system (SigmaAldrich, St. Louis, MO, USA) according to the manufacturer’s instructions. For determination of BiP levels by western blot analysis equal amounts of protein from lysates were separated in 15% SDS-PAGE and transferred onto a nitrocellulose Hybond-C membrane (Amersham Biosciences) using a semi-dry system. The primary antibodies and the respective dilutions used were: rabbit polyclonal anti-BiP (1:1000) and mouse monoclonal anti-GAPDH (1:3000) (Abcam, Cambridge, UK). Detection was performed with LuminataTM Crescendo (enhanced chemiluminescence, Millipore, Billerica, MA). Quantification of blots was performed with a Bio-Rad ChemiDoc XRS system using the IMAGELAB software and immunosignals were normalized with GAPDH expression. Results are presented as normalized density ± Standard Error of the Mean (SEM). Ethics statement. All the experiments described herein were approved by the Portuguese General Veterinarian Board (authorization number 024976 from DGV-Portugal) and are in compliance with national rules and the European Communities Council Directive (2010/63/EU), for the care and handling of laboratory animals. FAP transgenic mouse model. Aged transgenic mice for human TTR V30M in a TTR null background, heterozygous for the heat shock transcription factor 1 (Hsf-1), labeled hTTR V30M/Hsf mice, were used for the