Highly infectious prions are not directly neurotoxic

Abstract

Prions are infectious agents which cause rapidly lethal neurodegenerative diseases in humans and animals following long, clinically silent incubation periods. They are composed of multichain assemblies of misfolded cellular prion protein. While it has long been assumed that prions are themselves neurotoxic, recent development of methods to obtain exceptionally pure prions from mouse brain with maintained strain characteristics, and in which defined structures-paired rod-like double helical fibers-can be definitively correlated with infectivity, allowed a direct test of this assertion. Here we report that while brain homogenates from symptomatic prion-infected mice are highly toxic to cultured neurons, exceptionally pure intact high-titer infectious prions are not directly neurotoxic. We further show that treatment of brain homogenates from prion-infected mice with sodium lauroylsarcosine destroys toxicity without diminishing infectivity. This is consistent with models in which prion propagation and toxicity can be mechanistically uncoupled.

Keywords: neurodegeneration; neurotoxicity; prions.

Fig. 1.

Development of the IncuCyte-based neurite retraction assay. (A and B) Neurite outgrowth in primary neurons plated at 15,000 cells/well plateaued after 10 d in vitro (DIV); n = 6. Dark blue traces: neurites; yellow: cell features excluded from neurite analysis. (Scale bar, 200 µm.) (C) Neurite retraction in response to the treatment with RML-infected and uninfected Prnp^+/+ brain diluted to a concentration of 10⁻³ (RML, 10^5.2 IU/mL) for 12 h occurred in a cell-density–dependent manner (n = 3, *P < 0.03, unpaired two-tailed t test). (D) Dose–response curve for the neurite retraction assay; n = 3 to 12 independent cell cultures were analyzed in triplicate after 12 h of treatment; mean ± SD. Data were fitted using a nonlinear regression model [Y = 100/(1+10^((LogIC50-X)×HillSlope))]. ****P < 0.0001, one-way ANOVA with Tukey’s multiple comparison test.

Fig. 2.

Purified RML prions are not neurotoxic in primary cell culture. (A and B) Neurite length after treatment (designated by arrow) with pRML diluted in tissue culture medium at 10^4.2 (n = 3), 10^5.2 (n = 3), 10^5.7 (n = 3), and 10^6.7 (n = 3) IU/mL or RML prion-infected brain (RML BH) at a concentration of 3 × 10⁻³ in medium (10^5.7 IU/mL, n = 3) or uninfected brain at a concentration of 3 × 10⁻³ in medium (n = 3). (A) Normalized response over time, mean ± SEM, (B) at 72 h post treatment, mean ± SD; *P < 0.05 (unpaired two-tailed t test). (C and D) Analysis of neurite fragmentation (C) and dendritic spine density (D) after treatment with pRML in medium at 10^4.2 (n = 7), 10^5.2 (n = 7), 10^5.7 (n = 7), 10^6.7 IU/mL (n = 6), RML brain at a concentration of 10⁻³ in medium (10^5.2 IU/mL, n = 8), uninfected brain at a concentration of 10⁻³ in medium (n = 7) or medium alone (n = 8), mean ± SEM, *P < 0.05, ***P < 0.001, ****P < 0.0001. Gray asterisks: comparison of RML BH to uninfected control BH, unpaired two-tailed t test. Black asterisks: comparison to medium only, one-way ANOVA with Sidak’s correction for multiple comparisons; ns: not significant. (E) Representative images of dendritic spine density; green and red: raw imaging data; blue and gray: image analysis. (Scale bar, 10 µm.)

Fig. 3.

Purified RML prions in Prnp^0/0 mouse brain homogenate are not neurotoxic in primary cell culture. (A and B) Neurite length after treatment (designated by arrow) with pRML in Prnp^0/0 BH at a concentration of 10⁻⁴ at 10^4.2 (n = 3), 10^5.2 (n = 3), 10^5.7 (n = 3), and 10^6.7 IU/mL (n = 3) or RML BH at a concentration of 3 × 10⁻³ (10^5.7 IU/mL, n = 3), 10⁻⁴ (10^4.2 IU/mL, n = 3) or Prnp^0/0 BH at 10⁻⁴ (n = 3). The same dataset for RML BH at concentrations of 3 × 10⁻³ and 10⁻⁴ is shown across the figures for comparison with pRML. (A) Normalized response over time, mean ± SEM (B) at 72 h post treatment and mean ± SD pRML in Prnp^0/0 BH (10⁻⁴) at 10^6.7 IU/mL is significantly less toxic than RML BH at 10^4.2 IU/mL (10⁻⁴). *P < 0.05, unpaired two-tailed t test. (C and D) Analysis of neurite fragmentation (C) and dendritic spine density (D) after treatment with pRML in Prnp^0/0 BH at a concentration of 10⁻⁴ at 10^4.2 (n = 3), 10^5.2 (n = 7), 10^5.7 (n = 3), and 10^6.7 IU/mL (n = 3) and Prnp^0/0 BH at a concentration of 10⁻⁴ (n = 8), RML BH at concentration of 10⁻⁴ (10^4.2 IU/mL, n = 5) and 10⁻³ (10^5.2 IU/mL, n = 8), or medium alone (n = 8), mean ± SEM; comparison to Prnp^0/0 BH at 10⁻⁴, one-way ANOVA with Sidak’s correction for multiple comparisons. *P < 0.05 and ***P < 0.001 ns: not significant. (E) Representative images of dendritic spine density; green and red: raw imaging data; blue and gray: image analysis. (Scale bar, 10 µm.)

Fig. 4.

Purified RML prions in Prnp^+/+ mouse brain homogenate are not neurotoxic in primary cell culture. (A and B) Neurite length after treatment (designated by arrow) with pRML in Prnp^+/+ BH at a concentration of 10⁻⁴ at 10^4.2 (n = 3), 10^5.2 (n = 3), 10^5.7 (n = 3), and 10^6.7 IU/mL (n = 3), uninfected Prnp^+/+ BH at a concentration of 10⁻⁴ (n = 3), or RML BH at concentrations of 10⁻⁴ (10^4.2 IU/mL, n = 3), 10⁻³ (10^5.2 IU/mL, n = 3), and 3 × 10⁻³ (10^5.7 IU/mL, n = 3). The same dataset for RML BH at concentrations of 3 × 10⁻³ and 10⁻⁴ is shown across the figures for comparison with pRML. (A) Normalized response over time, mean ± SEM, (B) at 72 h post treatment; mean ± SD pRML in Prnp^+/+ BH (10⁻⁴) at 10^6.7 IU/mL is significantly less toxic than RML BH at 10^4.2 IU/mL (10⁻⁴); *P < 0.05, unpaired two-tailed t test. (C and D) Analysis of neurite fragmentation (C) and dendritic spine density (D) after treatment with pRML in Prnp^+/+ BH at a concentration of 10⁻⁴ at 10^4.2 (n = 3), 10^5.2 (n = 7), 10^5.7 (n = 7), and 10^6.7 IU/mL (n = 3) and Prnp^+/+ BH at a concentration of 10⁻⁴ (n = 8), RML BH at a concentration of 10⁻⁴ (10^4.2 IU/mL, n = 5) and 10⁻³ (10^5.2 IU/mL, n = 8) or medium alone (n = 8); mean ± SEM, comparison to Prnp^+/+ BH at 10⁻⁴, one-way ANOVA with Sidak’s correction for multiple comparisons; ***P < 0.001; ns: not significant. (E) Representative images of dendritic spine density. Green and red: raw imaging data; blue and gray: image analysis (Scale bar: 10 µm.)

Fig. 5.

Toxicity of RML brain homogenate is abolished by increasing concentrations of sarkosyl while prion infectivity titer is fully preserved. RML BH (10% [wt/vol]) was adjusted to final concentrations of 0.1 to 4% (wt/vol) sarkosyl (srk) in the sample, incubated at 37 °C for 30 min, and analyzed for acute toxicity to primary neurons at a final brain concentration of 10^−3.1 (A and B) and 10^−4.1 (C). Infectivity of the sarkosyl-treated RML BH samples (C) was assessed simultaneously by using the ASCA (D). (A and B) Neurite length after treatment (designated by arrow) with RML BH at a concentration of 10^−3.1 (n = 4). (A) Normalized response over time (B) at 56 h post treatment; mean ± SD, **P < 0.01, ***P < 0.001, and ****P < 0.0001, comparison to medium, one-way ANOVA with Sidak’s correction for multiple comparisons. ns: not significant. (C) Normalized response at 72 h post treatment (n = 3); mean ± SD, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, comparison to medium, one-way ANOVA with Dunnett’s multiple comparisons test. ns: not significant. (D) Infectivity titer of RML BH treated with sarkosyl (n = 3) was determined by ASCA in PK1/2 cells after the third split. *P < 0.05, comparison to RML BH, one-way ANOVA with Dunnett’s multiple comparisons test.