The inhibition of PrP^c-to-PrP^Sc conversion by curcumin has been demonstrated in hamster PrP [4] and in truncated human PrP [5]. However, the effect of curcumin on the structural conversion of mPrP is not known. This prompted us to initially investigate the effect of curcumin on cell-free conversion of mPrP.

The effect of curcumin on the amyloid formation of mPrP was studied based on the kinetics of amyloid formation with and without curcumin. Incubation of recombinant mPrP in 2 M guanidine hydrochloride (GdnHCl) at 37 °C resulted in amyloid fibrils [14,15]. The formation of amyloid fibrils was determined by the fluorescence intensity of thioflavin T (ThT) at 480 nm arising from specific binding of ThT to cross β-sheet structures in amyloid fibrils. Formation of amyloid fibrils includes nucleation in the lag phase and subsequent elongation in the growth phase. As shown in Figure 1, in the absence of curcumin, the ThT-fluorescence reading of mPrP fibrils rose at the fifth hour and then increased significantly in the following three hours. Finally, the ThT-fluorescence intensity remained constant. This sigmoidal kinetic curve indicates that the duration of the lag phase and of the growth phase of mPrP fibril formation can be estimated as 5 h and 3 h, respectively. In contrast, the sigmoidal curve observed in the presence of curcumin (20 μM or 50 μM) exhibited a weak increase such that the ThT-fluorescence reading remained low. In the presence of 20 μM curcumin, the ThT-fluorescence intensity increased slightly after 5–6 h of reaction but did not increase significantly in the following incubation, indicating partial inhibition of the growth phase. This kinetic study confirms that curcumin reduces the formation of amyloid fibrils and that partial inhibition likely occurs at the growth phase. In the presence of 50 μM curcumin, the ThT-reading increased slightly after 3–4 h of incubation but remained low after the forth hour. This shortened lag phase implies the existence of an alternative pathway for nucleation, while the fibril elongation may have also been inhibited in the alternative pathway. The amyloid fibrils obtained in this experiment were further analyzed in the following experiments.

An important feature of PrP^Sc is the resistance to proteinase K (PK) activity. Therefore, we treated the mPrP fibrils with PK, and then analyzed the PK-resistant fragments with high-resolutiontricine-SDS-PAGE [16]. In tricine-SDS-PAGE, small protein fragments can be clearly resolved at ~12 and 10 kDa, as shown in Figure 2. At PK: PrP = 1: 500, the fragments with 12 and 10 kDa clearly existed in all the samples. A weak band at ~16 kDa was shown in curcumin-free samples but this band disappeared in curcumin-treated samples. When PK was increased such that PK: PrP =1:50, in the presence of 20 μM curcumin, the 12-kDa fragment was strongly weakened while a smaller fragment (~8 kDa) was observed. This pattern implies that the amyloid forms of PrP generated under curcumin treatment can be further digested. When 50 μM of curcumin was added, the 12-kDa fragment was not observed while the 10- and 8- kDa fragments were more pronounced than the same fragment in the absence of curcumin. This pattern indicates the presence of an alternative amyloid form generated under curcumin treatment. Further analysis of this alternative amyloid will be the subject of future study.

As determined by X-ray fiber diffraction, amyloid fibrils associated with various diseases all appear as unbranched straight fibrils [17]. Thus, it is essential to observe the morphology of the fibrils converted in the presence of curcumin. TEM images of the fibrils under various conditions were analyzed. As illustrated in Figure 3a, the mPrP fibrils grown in the absence of curcumin are long and straight as observed in TEM. The straight mPrP fibrils were also observed in the presence of 20 μM curcumin as shown in Figure 3b, but the length of the fibrils were generally shorter than those grown in the absence of curcumin. The analysis of the fibril-length distribution plotted in Figure 3d indicates that the most populated fibrils were 201–250 nm long in the absence of curcumin, whereas the major population was shifted to 101–150 nm in the presence of 20 μM curcumin. The average length of mPrP fibrils grown in the absence of curcumin and in the presence of 20 μM curcumin is 238 and 174 nm, respectively. The diameter of each of these mPrP fibrils is about 25 nm. In the presence of 50 μM curcumin, the majority of the mPrP aggregated amorphously as shown in Figure 3c. Amyloid fibrils were rarely formed and the observed fibrils were curvy and short (~60 nm). These TEM images clearly suggest that curcumin inhibited the elongation of the mPrP amyloid fibrils along the long-axis of the fibrils. This is consistent with previous study indicating curcumin binds to the intermediates of PrP resulting inhibition of amyloid formation [5]. The grown mPrP fibrils were further treated with 20 μM and 50 μM curcumin for 12 h as shown in Figure 3e,f, respectively. Clearly, 20 μM curcumin impaired the morphology of fibrils such that the residual fibrils observed exhibited both straight morphology and amorphous aggregation. When the concentration of curcumin was increased to 50 μM, the amyloid-forms were mostly abolished. This phenomenon is consistent with a study of quercetin where quercetin induced the amorphous aggregation of mature insulin fibrils [18].

To investigate the effect of curcumin in cells, we first confirmed the disruption of cell membranes in erythrocytes from mouse blood. Subsequently, cell viability, apoptosis and ROS level were all studied with various concentrations of curcumin.

Plasmid pET101 encoding mPrP 23–231 was transformed into competent E. coli BL21 (DE3), and expressed in inclusion bodies by induction with isopropyl β_-D-thiogalactopyranoside. The protein was purified on a Ni-Sepharose column according to a previously described procedure [14]. The purity of the isolated protein was confirmed by SDS–PAGE.

The fibril conversion from 10 μM mouse prion protein (mPrP) was incubated on the shaker rotating at 1000 rpm at 37 °C in a buffer solution containing 50 mM MES (pH 6.0) and 2 M GdnHCl. In addition, tested compounds were added into the sample mixture. The kinetics of fibril formation was monitored by thioflavin T (ThT), a fluorescence dye recognizing amyloids. The aliquots of fibril samples during the time course of incubation were diluted with sodium acetate buffer (pH 5.0) containing 10 μM ThT to the final fibril concentration at 0.5 μM. The fluorescence spectra were collected from 470 to 550 nm with the excitation wavelength at 450 nm by using a Hitachi F-4500 fluorimeter. The maximum fluorescence emission at 482 nm was recorded.

The fibril samples were stained with 2.6% tungsten phosphoric acid on carbon-coated 200-mesh copper grids. The samples were adsorbed onto the copper grids for 1 min and subsequently washed with PBS and H₂O. The samples were air-dried before imaging. The TEM images were collected using a Hitachi H-7100 TEM. The analysis of fibril length was performed with ImageJ software.

The 10 μM of fibril samples were treated with proteinase K (PK) at 37 °C in 100 mM Tris (pH 7.5). After one hour incubation, the PK-treated samples were added with 2× sample buffer for SDS-PAGE followed by heating at 95 °C for 10 min and then analyzed by tricine-SDS-PAGE. The densitometric quantification of proteins was analyzed with ImageJ software.

The mouse blood was primarily centrifuged at 1,000 g for 10 min. After removing the supernatant, the erythrocytes were washed three times with phosphate buffered saline (PBS, pH7.4). Subsequently, the mPrP fibrils were added into the cell suspensions (1% hematocrit), and incubated at 37 °C for 40 min. The mixtures were centrifuged at 1,000 g for 10 min. Finally, the aliquots of cells were distinguished by microscopy and the aliquots of supernatant were collected for absorption measurement.

Mouse neuroblastoma (N2a) cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with penicillin/streptomycin and 10% (v/v) fetal bovine serum in a humidified atmosphere with 5% CO₂ at 37 °C. N2a cells (1 × 10⁴) were seeded to 96-well plates for 24 h, and then treated with curcumin and/or mPrP fibrils. After 72 h of incubation, the medium was removed and the cell plates were washed by PBS. Subsequently, 10% (v/v) WST-1 was added into the plates and incubated with cells for 3 h. The cell viability was determined by the absorbance of formazan was recorded at 450 nm using an ELISA reader

Cells were fixed by adding 100% methanol drop by drop then stored at 4 °C for over 30 min. The fixed cells were spin down followed by re-suspension in 500 μL of PBS. Cells were then incubated with RNase to final concentration at 200 μg/mL and with propidium iodide (PI) to final concentration at 40 μg/mL at room temperature in the dark for 30 min. The cell cycle profiles were analyzed by Cytomics FC500 Flow Cytometry (Beckman Coulter).

N2a cells (1 × 10⁵) were seeded to 24-well plates for 24 h, and then treated with 20 μM of mPrP fibrils, or 2.5 μM curcumin prior to co-incubation of mPrP fibrils. Two positive controls including 5 μg/μL of tunicamycin and 100 μM H₂O₂ were compared. After 2 h of incubation, 2, 7-dichrofluorescein diacetate (DCFH-DA) was added to the cell plates to final concentration of 50 μM, and then incubated for 30 min. Subsequently, the medium was removed and the cell plates were washed by PBS, followed by the addition of 1 mM N-acetylcysteine dissolved in DMSO to quench the oxidation of DCFH-DA. The fluorescence emission of reduced 2, 7-dichrofluorescein (DCF) at 530 nm was recorded with the excitation wavelength at 485 nm by using a fluorescence ELISA reader.