Partial Deletion of the Carboxyl-Terminal Signal Sequence of the Cellular Prion Protein Alters Protein Expression via Endoplasmic Reticulum-Associated Degradation

Abstract

Cellular prion protein (PrP^C) is a glycoprotein tethered to the plasma membrane via a GPI-anchor, and it plays a crucial role in prion diseases by undergoing conformational change to PrP^Sc. To generate a knock-in (KI) mouse model expressing bank vole PrP^C (BVPrP^C), a KI targeting construct was designed. However, a Prnp gene sequence that encodes PrP^C lacking seven C-terminal amino acid residues of the GPI-anchoring signal sequence (GPI-SS) was unintentionally introduced into the construct. The resulting KIBVPrP248 mice exhibited very low PrP^C expression and resistance to prion infection. To investigate the underlying mechanism of reduced PrP^C expression, RK13 cells expressing either full-length GPI-SS (BVPrP255) or truncated GPI-SS (BVPrP248) and KIBVPrP248 mice were analyzed. In RK13-BVPrP248 cells, PrP^C protein levels were nearly ten-fold lower than in RK13-BVPrP255 cells, mimicking the extremely low PrP^C expression of the KIBVPrP248 mice. The abundance, stability, and translational efficiency of the Prnp mRNA were not the primary causes for the low PrP^C expression in RK13-BVPrP248 cells. A pharmacological analysis revealed that BVPrP248 underwent enhanced degradation via the ER-associated degradation pathway, with increased PrP ubiquitination detected in both the cell and animal models. An immunofluorescence analysis showed that BVPrP248 was mislocalized to the ER, co-localizing with Grp78, an ER chaperone. Although mislocalization of BVPrP248 under the transient overexpression condition led to mild activation of the unfolded protein response in RK13-BVPrP248 cells, low-level chronic expression of BVPrP248 in stable transfectants and KIBVPrP248 mice did not facilitate such events. These findings suggested that the C-terminal GPI-SS of PrP^C plays a critical role in PrP^C biogenesis.

Keywords: GPI‐linked proteins; endoplasmic reticulum–associated degradation; gene expression regulation; prion proteins; unfolded protein response.

FIGURE 1

Western blot analysis of PrP^C and PrP^Sc in KIBVPrP248 mice. (A) Western blot analysis of PrP^C expression in the brains of littermate WT (+/+), heterozygous (+/Rc), and homozygous (Rc/Rc) KIBVPrP248 mice. The blot for homozygous KIBVPrP248 mouse samples was overexposed. The arrow indicates PrP^C, likely in the unglycosylated form, that was barely expressed. (B) Western blot analysis of PK‐resistant PrP^Sc levels in the brains of RML prion–infected WT (+/+), heterozygous (+/Rc) and homozygous (Rc/Rc) KIBVPrP248 mice at the terminal stage of the disease. Western blots for β‐Actin represent the level of internal control for the experiment.

FIGURE 2

Characterization of RK13‐BVPrP248 cells. (A) Western blot analysis of BVPrP^C expression in RK13‐BVPrP255 and RK13‐BVPrP248 cells. RK13, mock‐transfected cells; vec, cells transfected with an empty vector. Densitometry of BVPrP^C levels normalized to β‐Actin (right panel, n = 4). (B) RT‐qPCR analysis of Prnp mRNA levels in RK13‐BVPrP255 and RK13‐BVPrP248 cells (n = 5). Relative copy number analysis via qPCR using gDNA (right panel, n = 5). (C) Western blot analysis of glycosylation status following PNGase F treatment. (D) Western blot analysis following PI‐PLC treatment to assess GPI‐anchoring in cell lysates and media. The media were collected from cultures of 1.5 to 1.9 × 10⁶ cells. (E, F) Double immunofluorescence staining of BVPrP and ganglioside GM1 (lipid raft marker) using cholera toxin B or Grp78 (ER marker) in RK13‐BVPrP255 and RK13‐BVPrP248 cells. Nuclei were stained with DAPI. Scale bar = 50 μm. (G) GFP detection in RK13‐GFP254 and RK13‐GFP247 cells. Scale bar = 50 μm. ***p < 0.001. **p < 0.01.

FIGURE 3

Mechanisms underlying BVPrP248 downregulation. (A) mRNA decay within RK13‐BVPrP255 and RK13‐BVPrP248 cells. Cells were treated with actinomycin D, and intracellular mRNA decay was measured at 2, 4, 8, and 14 h by RT‐qPCR. (B) In vitro mRNA stability. Total RNA isolated from RK13‐BVPrP255 and RK13‐BVPrP248 cells was left at 37°C for up to 24 h, followed by RT‐qPCR. (C) Western blot analysis of reaction products from a wheat germ extract–coupled in vitro transcription/translation system. EGFP was used as a positive control, and an empty vector (vec) and double distilled water (ddH₂O) were used as negative controls. The densitometry of the Western blots is shown in the bottom panel (n = 5). (D) Proteasomal degradation. The cell culture was incubated with a proteasome inhibitor (MG132). Total BVPrP levels were analyzed by Western blotting after PNGase F digestion. The densitometry of BVPrP levels was normalized to β‐Actin (right panel, n = 3). (E) Ubiquitination analysis. The empty vector–transfected RK13, RK13‐BVPrP255, and RK13‐BVPrP248 cells were incubated with or without MG132. Cell lysates were subjected to immunoprecipitation with an anti‐PrP SAF32 antibody, followed by Western blotting with an anti‐ubiquitin antibody. The densitometry of ubiquitinated PrP levels was analyzed with or without normalization to total PrP (right panels, n = 4). (F) Western blot analysis of ER stress and UPR‐related protein levels in RK13 cells transiently transfected with empty vector, BVPrP255, or BVPrP248, 48 h post‐transfection. The densitometry of marker proteins is shown in the right panel (n = 3). ***p < 0.001; **p < 0.01; *p < 0.05.

FIGURE 4

ER‐associated degradation of BVPrP248 and UPR responses in KIBVPrP248 mice. (A) Immunohistochemistry of brain tissue from littermate WT, KIBVPrP248 (Rc/Rc), and Prnp null mice. Sections were counterstained with Nuclear fast red. Scale bar = 500 μm (upper panel) and 20 μm (lower panel). The CA1 region of the hippocampus (box with discontinuous line) was magnified to compare the pattern of expressed BVPrP^C (lower panels). so, stratum oriens; sp., stratum pyramidale; sr, stratum radiatum. (B) Prnp mRNA expression levels in the brains of KIBVPrP248 (Rc/Rc) mice, as analyzed by RT‐qPCR. (C) Double immunofluorescence staining of Grp78 and PrP in the cortex of mouse brain. The co‐localization was visualized in merged images. Magnified images were obtained from the area marked with discontinuous boxes. Scale bar = 20 μm. (D) Ubiquitination of BVPrP in KIBVPrP248 (Rc/Rc) mice. Mouse brain homogenate was immunoprecipitated with anti‐PrP SAF32 antibody or anti‐ubiquitin antibody, and detected with anti‐ubiquitin antibody or anti‐PrP (SAF32 or 8H4) antibody, respectively. Arrows, tetra‐ubiquitinated PrP. Asterisks, mono‐ or di‐ubiquitinated PrP. The predicted molecular size of these bands was calculated based on addition of approximately 8.5–34 kDa from mono‐ to tetra‐ubiquitins to approximately 22.9–24.7 kDa from WT PrP and BVPrP. The expression of BVPrP as an unglycosylated form in KIBVPrP248 was shown in the overexposed input blots. The densitometry of tetra‐ubiquitinated PrP levels normalized to total PrP (bottom panel). (E) Western blot analysis of UPR‐related protein levels in WT and KIBVPrP248 (Rc/Rc) mouse brains. The densitometry of marker proteins (right panel, n = 3).