In Vitro and In Vivo Evidence towards Fibronectin's Protective Effects against Prion Infection

Abstract

A distinctive signature of the prion diseases is the accumulation of the pathogenic isoform of the prion protein, PrP^Sc, in the central nervous system of prion-affected humans and animals. PrP^Sc is also found in peripheral tissues, raising concerns about the potential transmission of pathogenic prions through human food supplies and posing a significant risk to public health. Although muscle tissues are considered to contain levels of low prion infectivity, it has been shown that myotubes in culture efficiently propagate PrP^Sc. Given the high consumption of muscle tissue, it is important to understand what factors could influence the establishment of a prion infection in muscle tissue. Here we used in vitro myotube cultures, differentiated from the C2C12 myoblast cell line (dC2C12), to identify factors affecting prion replication. A range of experimental conditions revealed that PrP^Sc is tightly associated with proteins found in the systemic extracellular matrix, mostly fibronectin (FN). The interaction of PrP^Sc with FN decreased prion infectivity, as determined by standard scrapie cell assay. Interestingly, the prion-resistant reserve cells in dC2C12 cultures displayed a FN-rich extracellular matrix while the prion-susceptible myotubes expressed FN at a low level. In agreement with the in vitro results, immunohistopathological analyses of tissues from sheep infected with natural scrapie demonstrated a prion susceptibility phenotype linked to an extracellular matrix with undetectable levels of FN. Conversely, PrP^Sc deposits were not observed in tissues expressing FN. These data indicate that extracellular FN may act as a natural barrier against prion replication and that the extracellular matrix composition may be a crucial feature determining prion tropism in different tissues.

Keywords: extracellular matrix (ECM); k fibronectin; muscle; myotube; prion; reserve cells; scrapie; tropism.

Figure 1

Prion deposition is associated with myotubes in RML-infected dC2C12 cultures at 20 days post-inoculation. Cultures were immunohistochemically stained with SAF83 anti-PrP Ab (red), anti-desmin Ab (green) and DAPI (blue). Four serial micrographs taken along the Z-axis show the multilayer culture composed of 2 morphologically different cell types, the long multi-nucleated myotubes (MT) abundant in the upper layers (Z3, Z4) and the mono-nucleated spindle-shape reserve cells (RC), mostly in the bottom layers (Z1, Z2). PrP^Sc immunostaining appears as fine intracytoplasmic punctae (yellow arrowheads) solely in myotubes with no PrP^Sc deposits found in reserve cells. Interval of the confocal Z-stack images is 0.5 µm. Scale bar, 20 µm. PrP^Sc signals in reserve cells-enriched (Z1) and myotubes-enriched layers (Z4) were quantified by measuring mean pixel intensity and coefficient of variation of punctate red fluorescent signals (right panels). Statistical significance was set at p < 0.01 (**) and p < 0.001 (***) in an unpaired t test.

Figure 2

Silver stain and immunoblot analysis of PTA-purified PrP^Sc from RML-infected mouse brain and dC2C12 cultures. NBH and RML -infected PTA precipitated samples were lysed with sarkosyl (S), triton deoxycholate (TD) or RIPA buffer, digested with different concentrations of PK and analyzed by SDS-PAGE followed by silver staining (gel on top) or immunoblotting (membrane at the bottom). High molecular weight (HMW) proteins > 100 kDa were found under all purification conditions for dC2C12 samples as shown by silver staining. PrP^Sc from brain and dC2C12 cultures was detected by western blotting, whereas only PrP^Sc from brain was visualized with silver staining. SAF83 dil. 1:10,000 was used to detect PrP. Arrowheads indicate the migrating position of PrP glycoforms.

Figure 3

Silver staining and immunoblotting of fractions from velocity gradients. RML-infected mouse brain homogenate (A) and dC2C12 culture (B) were lysed in sarkosyl (left) or TD (right) buffer, PK digested, and the proteins were fractionated by velocity gradient centrifugation in an iodixanol gradient. The presence of HMW proteins (arrowhead, >100 kDa) was evaluated by silver staining and PrP^Sc by immunoblotting (SAF83, dil. 1:10,000) in the 14 fractions collected from top (fraction in lane 1) to bottom (fraction in lane 14).

Figure 4

Infectivity of PrP^Sc from dC2C12 before and after purification evaluated by SSCA. (A) SSCA of whole lysate (light grey bar) and PTA-purified PrP^Sc (dark grey bar) from the same RML-infected dC2C12 culture. An uninfected dC2C12 lysate (Mock, white bar) was also analyzed. (B) SSCA of the RML-infected dC2C12 fractions collected after velocity gradient centrifugation (compare Figure 3B). In all cases, dC2C12 cells were lysed in TD buffer. Dotted line represents the cut off threshold for prion infection. Average ± SE was plotted. *, p < 0.05; **, p < 0.01;***, p < 0.001. Mock, uninfected dC2C12 lysate.

Figure 5

Confirmation of extracellular matrix proteins by immunoblotting and presence of prion-like rods confirmed by electron microscopy. (A) Immunoblotting of collagen, fibrillin, fibronectin, PrP^res and Coomassie blue staining for the fractions of RML-infected dC2C12 cells after velocity gradient centrifugation. (B,C) Fractions from velocity gradients with strong PrP^res signal but no HMW bands (following Coomassie staining), were examined by electron microscopy. Prion rods were detected in cells lysed with sarkosyl (B) whereas, following triton X-100 extraction, amorphous aggregates like blobs and oligomers were observed (C). (D) Electron microscopic examination of a PTA-precipitated sample from RML-infected dC2C12 cultures contains a mesh of fibrillar structures, such as collagen (asterisk), in which a prion rod-like structure (arrow) can be distinguished. Scale bar 50 nm.

Figure 6

Detection of PrP and FN proteins in muscle-derived cells. Immunocytochemical detection of PrP^Sc (red) and FN (green) in NBH-treated (A) and RML-infected dC2C12 (B) cultures at day 20 post-inoculation. No co-localization of the two proteins was observed in either the NBH-treated (A) or RML-infected (B) cultures. Nuclei in blue. MT, myotube (delimited by dashed line); RC, reserve cells. Scale bar, 20 µm for the top row and 10 µm in bottom row (corresponding to the boxed section in the top row).

Figure 7

Time course of PrP^res and FN in dC2C12 cultures. Immunoblot analysis of PrP^res and FN at 7 and 14 days post-inoculation show the negative correlation of the presence of these proteins with time. (A) FN decreases from day 7 to 14 in both NBH-treated and RML-infected cultures whereas PrP^res increases with time. (B) No differences in FN levels between NBH and RML-infected cultures at day 20 are detected. The detergents used in the homogenization of the lysates yield different amount of FN by immunoblotting, sarkosyl yielding the larger FN signal.

Figure 8

Dexamethasone effect on dC2C12 cultures confirms the inverse correlation between prion accumulation and FN. Differentiated C2C12 cultures were treated with dexamethasone (10 µM) at day 7 post-exposure to NBH (A) or RML prions (B). (A,B) Immunocytochemical detection in cultures at 21 days post-inoculation of PrP^Sc (in red) and desmin (in green) reveal some atrophic myotubes remaining (indicated by arrowheads) caused by dexamethasone treatment and an abundance of mono-nucleated cells (delimited by dashed line). (B) PrP^Sc was observed, for the first time, associated with the mono-nucleated cells as gross particulates. Scale bar, 20 μm for the top row and 10 µm in bottom row (corresponding to the boxed section in the top row).

Figure 9

Protein expression of PrP and extracellular matrix components in dC2C12 at 20 days post-inoculation. treated with dexamethasone. dC2C12 culture was exposed to RML prions or NBH, and then, at 7 days post-infection, treated with dexamethasone (10 μM in ethanol) or mock control (0.25% ethanol). The cells were harvested at 20 days post-exposure and analyzed by immunoblotting. Dexamethasone treatment of NBH and RML infected dC2C12 reduces (A) total PrP levels and (B) FN with a concomitant increase in (A) PK-resistant PrP compared with non-dexamethasone-exposed cultures. (A) The expression of MyoD and Px7, myogenic regulatory factors, was reduced while (B) other extracellular matrix (ECM) proteins such as collagen I and vitronectin did not change. Conversely, LRP1, reported to mediate endocytosis of PrP^C in neuronal cells, was increased proportionally to the accumulation of PrP^res. The immunoblot intensity was normalized by GAPDH loading control. Error bars represent SD. *, p < 0.05 and **, p < 0.01 in comparison with mock treatment control. FN, fibronectin; LRP1, low-density lipoprotein receptor-related protein 1; COL-I, collagen I; VN, vitronectin; GAPDH, Glyceraldehyde 3-phosphate dehydrogenase.

Figure 10

PrP^Sc is inversely correlated with FN in tissues from naturally scrapie-infected sheep. (A,C,E,G,I,K) Immunohistochemical detection of FN and (B,D,F,H,J,L) PrP^Sc in serial sections of tissues from naturally scrapie-infected sheep. (A,B) In striated skeletal muscle, fine punctae of FN immunolabelling are observed at the endomysium sheath of the myofibers. (C–F) In smooth muscle layers of the digestive tract, FN is found at basal lamina of each myocyte with PrP^Sc restricted to the myenteric plexus, surrounded by a FN-positive perineurium (asterisk). Boxes in (C,D) correspond to areas shown in higher magnification in (E,F). (G,H) The central nervous system shows intense PrP^Sc accumulation. (I,J) Blood vessel displayed FN at the basal lamina of endothelium and at the adventitia. (K) The adrenal gland shows FN staining with a diffuse pattern in the entire adrenal cortex at intercellular space (inset: detailed FN intercellular staining); (L) PrP^Sc accumulation is associated with chromaffin cells at the adrenal medulla. Scale bar, 50 μm for A, B, E, F; 100 μm for G, H, I, J, K, L and 200 μm for C,D.