Cytotoxicity evaluation
To determine the cytotoxicity of DB772 and the potential role of cell death as the mechanism of anti-prion action, microglial and Rov9 cells were exposed to half-log dilutions of DB772. The CC50 was calculated to be 10.662.3 mM for microglial cells and 10.561.4 mM for Rov9 cells (Table 2). The tissue culture selectivity index (CC50/EC50) for microglia and Rov9 cells is 4.6 and 5.5, respectively. The percent cytotoxicity was determined at relevant anti-prion concentrations: the anti-PrPSc TCEC50 (Microglia: 2.4 mM; Rov9: 1.9 mM), and at 4 mM (concentration used in Fig. 2). Since the planned ANOVA-based analysis failed the normality assumption, the Kruskal-Wallis one-way ANOVA based on ranks, with a Tukey method of multiple comparisons wasFigure 4. DB772 inhibits prion infectivity in Rov9 cells. Rov9Sc/ DB772 and Rov9Sc/UnTx cells were mechanically lysed and used as prion inoculum. Rov9 cells inoculated with these new inocula were then tested for PrPSc via ELISA. Data columns represent the corrected optical density (OD450 ?OD620, per manufacturer’s instructions to correct for nonspecific optical density) means 6 one standard deviation of three independent experiments, each run in triplicate. The y-axis reference line indicates the minimum detection limit of the ELISA. *, P,0.001.used. No change in cell viability was detected at the anti-PrPSc TCEC50 (Microglial % viability2.4mM: 98.262.1%; Rov9 % viability1.9mM: 99.560.5%). At 4 mM (the concentration initially used to treat the cells) no cytotoxicity was definitively detected in microglial cells; however, early cytotoxic effects are likely at this concentration (Microglial % viability4mM: 94.0611%; P = 0.079). Significant cell death, however, was detected in Rov9 cells at 4 mM (Rov9 % viability4mM: 94.462.5%; P,0.001). To further reduce the possibility of low levels of cytotoxicity being a non-specific cause of PrPSc inhibition, we tested the compound curcumin, which does not inhibit sheep scrapie in Rov9 cells [47]. Rov9 cells cultured in 100 and 56.4 mM of curcumin exhibited 100% cell death; thus, these samples could not be used to assay for PrPSc. The calculated curcumin CC50 was 68.6620 mM in Rov9 cultures. Rov9 cells cultured at 31.7 mM subjectively exhibited significant cell death and the total protein levels in the 31.7 mM-treated samples (42.7639.7 mg/ml) were significantly (P,0.001) lower than all sample groups treated with less curcumin (17.8 mM: 1032 mg/ml, 10 mM: 1116 mg/ml, 1 mM: 1105 mg/ml, and 0 mM: 1097 mg/ml). No anti-PrPSc TCEC50 can be calculated for curcumin as there was no inhibition of PrPSc accumulation, even at the clearly cytotoxic concentration of 31.7 mM. In fact, the relative levels of PrPSc in the treated groups were actually slighter higher than the untreated groups (Fig. 6).
Discussion
Despite previous research investigating compounds with antiPrPSc activity [43,44,45], no effective chemotherapeutics exist for the treatment or prevention of prion diseases. Identification of new classes of anti-prion compounds is therefore vital, not only for the practical application of in vivo chemotherapeutics, but also for investigations studying the mechanisms of PrPSc conversion and accumulation. The data herein describe the discovery of in vitro anti-prion activity of a novel aromatic monocation. The anti-prion activity was demonstrated in two different cell culture models, including a cell type that is relevant to natural prion disease (sheep microglial cells). While the anti-prion effects described herein werediscovered while using DB772 to eliminate BVDV from primary sheep microglial cells and Rov9 cells, to the authors’ knowledge there are no published reports of phenyl-furan-benzimidazole cations with anti-prion activity. The anti-PrPSc activity resulted in complete loss of PrPSc signal in sheep microglial cells and Rov9 cells by the end of the treatment (P-4), which paralleled a loss in prion infectivity. All replicates were not completely cured of PrPSc, however, as after four passages without DB772 (P-8), one group from each of the microglialSc/ DB772 and Rov9Sc/DB772 samples had low levels of detectable Sc PrP . Regardless, these results do demonstrate significant inhibition of sheep-derived PrPSc accumulation in two cell types. In addition to the PrPSc inhibition, DB772 treatment also inhibited BVDV in both cell lines; however, it did not cure most of the cell replicates as BVDV antigen and BVDV RNA returned to detectable levels in one microglial replicate and in all of the Rov9 cell replicates. This incomplete pestivirus inhibition is different from what was demonstrated in primary bovine fibroblasts [56]. The differing results may be due to differences in the strains of BVDV that were tested, as well as the different cell types. The cytotoxicity of DB772 was evaluated in sheep microglial cells and Rov9 cells.
The 50% cytotoxicity point was similar between sheep microglial cells and Rov9 cells (10.5 mM and 10.0 mM, respectively) and is also similar to the previously demonstrated CC50 of 8.6 mM in B16 melanoma cells [58]. These values are in contrast to previous cytotoxicity studies using DB772 in Madin-Darby bovine kidney (MDBK) cells, in which the CC50 was substantially higher at 215 mM [57]. The discrepancy between these CC50 values is possibly a reflection of the different cell types, but may also be a result of the different culture conditions used. Initial investigations into the mechanism of action were conducted and while no mechanism was identified, some potential mechanisms have been ruled out. Expression of PRNP is required for PrPSc permissiveness [62] and the level of expression correlates with PrPSc permissiveness [63,64]; thus, one obvious mechanism of PrPSc inhibition would be the partial to complete inhibition of PRNP expression. There was no evidence that DB772 inhibited PRNP expression, as PRNP transcript levels and total PrP protein levels were not decreased. In fact at passage four (the end of the DB772 treatment), microgliaSc/DB772 and microgliaC/DB772 cells have significant increases in PRNP transcript and total PrP protein levels as compared to the untreated controls. This confirms that DB772 does not inhibit PrPC expression in microglial cells and suggests that PrPC expression may increase in response to DB772 exposure. While the levels of PrPC were increased, it is unclear if the increase is significant enough to be biologically relevant as the magnitude of change was small (less than two fold). Similarly, there was no evidence of PRNP expression inhibition in Rov9 cells as the direction of changes in PRNP transcript levels in Rov9 cells was towards an increase in PRNP transcript levels with DB772 treatment, and no change was identified in total PrP levels. The difference between the sheep microglial cells and Rov9 cells regarding PRNP transcript levels and total PrP levels is possibly attributed to the artificial PRNP expression system used in Rov9 cells (i.e., tetracycline-inducible promoter), which is unlikely to respond to the same stimuli as the natural PRNP promoter.
