br Materials and methods br Results br Discussion In
Materials and methods
Discussion In the search for drugs for human prion diseases, we conducted docking simulation using a DEGIMA supercomputer. Results from the in vitro and ex vivo drug screening revealed that NPR-053 and -056 significantly reduced PrPSc levels (Figs. 2 and 5). Although the affinity of NPR-056 for PrPC was approximately 2-fold greater than NPR-053 (Supplementary Fig. S1), thermal stabilisation of the NPR-053–PrPC complex was significantly stronger than the GN8–PrPC complex (Fig. 4 and Table 2), indicating that bonding varies among NPRs. Intriguingly, although NPR-015 and -065 did not bind with PrPC in the SPR assay, all NPRs (-015, -050, -053, -056 and -065) exhibited anti-prion effects in FK-1 and 22L strains (Supplementary Fig. S3B). These results suggested that NPRs might exhibit suppressant effects against PrPSc in universal prion strains depending on the binding of NPRs to PrPC, although these results differed from drugs with a strain-dependent function such as Cpd B (Kawasaki et al., 2007). Conversely, NPRs had no effect on PrPC levels and cellular toxicity (Supplementary Fig. S5), suggesting that NPRs do not promote PrPC degradation, compared with activation of the ubiquitin–proteasome system by IU-1 (Homma et al., 2015) and macroautophagy by lithium (Heiseke et al., 2009) and FK506 (Nakagaki et al., 2013). Results from the bioassay showed that NPR-053 and -056 treatments markedly suppressed PrPSc and gliosis levels in the brains of mice at onset, but had no effect on survival (Fig. 5 and Supplementary Fig. S6), indicating that the effective drugs which is left in gingerol after NPR-treatment might be very low levels. The pharmacokinetics, such as transportation to the brain and drug stabilisation, remains poorly understood, so future studies should focus on the routes, periods, frequency and dosages of NPRs. The intracellular aggregations of dysfunctional misfolded proteins are removed by a variety of homeostatic mechanisms, including endoplasmic reticulum-associated degradation, the ubiquitin–proteasome system and the autophagy system. PrPSc has been shown to co-localise with p62 and form aggresomes in prion-infected cells (Homma et al., 2014b). Intriguingly, results from this study shown that some NPRs significantly reduced the development of aggresomes (Fig. 6). Previous reports have shown that aggresomes are degraded by a selective autophagy system that recognises LC3-tagged protein complexes (Bjorkoy et al., 2005; Ichimura et al., 2008; Komatsu et al., 2007; Pankiv et al., 2007). The location of senile plaques in the Alzheimer\'s disease brain, which is the result of Aβ and neurofibrillary tangles caused by hyper-phosphorylated tau and inclusion bodies, which are formed by alpha-synuclein in Parkinson\'s disease cell culture models, is identical to the location of aggresomes (Kothawala et al., 2012; Shen et al., 2011). These results suggest that NPRs could protect against prion disease as well as other conformational diseases that accumulate aggresomes. Molecular simulation and analysis of atomic level interactions between NPRs and PrPC showed that the binding sites of NPR-053 and -056 were located around the four amino acid residues — N159, Q160, K194 and E196 (Fig. 1A). High-pressure NMR shows that this position is less stable (Kuwata et al., 2002) and is considered to be a “hot spot” for the pathogenic conversion of prion diseases (Kuwata et al., 2007; Yamamoto, 2014). Indeed, GN8 can stabilise PrPC and inhibit conversion by binding to this position (Ishikawa et al., 2009; Kuwata et al., 2007). Furthermore, several different structure compounds were reported, which have anti-prion activity by binding to the similar position (Ferreira et al., 2014). Our simulation results show that the overall binding characters of our compounds are consistent with these previous studies, suggesting that the hydrophobic pocket might be important for binding of other anti-prion compounds with heterogeneous PrP. However, the FMO calculations clearly demonstrate that the detailed interaction mechanisms of NPR-053 and -056 are different than GN8. In the case of GN8, polar interactions, including hydrogen-bonding interactions with N159, Q160, K194 and E196 were critical (Ishikawa et al., 2009). Conversely, HF level calculations revealed no large negative interaction energy (Fig. 3A), suggesting that polar interactions are not important for the binding of NPR-053 and -056. Instead, MP2 energies of several residues were negatively large, indicating that van der Waals interactions play an important role in complex stability. In particular, G124, L125, Y162, Q186 and H187 exhibit key interactions with the compounds. The docking structures of NPR-053 and -056 are shown in Fig. 3B. We note that the aromatic rings of these compounds are in contact with the amino acid residues, which are capable of generating a large attractive interaction with PrPC. Thus, our analysis of interaction energies using the FMO method was consistent with binding conformations from the docking simulation.