Follicular dendritic cell-specific prion protein (PrP) expression alone is sufficient to sustain prion infection in the spleen

Abstract

Prion diseases are characterised by the accumulation of PrP(Sc), an abnormally folded isoform of the cellular prion protein (PrP(C)), in affected tissues. Following peripheral exposure high levels of prion-specific PrP(Sc) accumulate first upon follicular dendritic cells (FDC) in lymphoid tissues before spreading to the CNS. Expression of PrP(C) is mandatory for cells to sustain prion infection and FDC appear to express high levels. However, whether FDC actively replicate prions or simply acquire them from other infected cells is uncertain. In the attempts to-date to establish the role of FDC in prion pathogenesis it was not possible to dissociate the Prnp expression of FDC from that of the nervous system and all other non-haematopoietic lineages. This is important as FDC may simply acquire prions after synthesis by other infected cells. To establish the role of FDC in prion pathogenesis transgenic mice were created in which PrP(C) expression was specifically "switched on" or "off" only on FDC. We show that PrP(C)-expression only on FDC is sufficient to sustain prion replication in the spleen. Furthermore, prion replication is blocked in the spleen when PrP(C)-expression is specifically ablated only on FDC. These data definitively demonstrate that FDC are the essential sites of prion replication in lymphoid tissues. The demonstration that Prnp-ablation only on FDC blocked splenic prion accumulation without apparent consequences for FDC status represents a novel opportunity to prevent neuroinvasion by modulation of PrP(C) expression on FDC.

Figure 1. Cre-mediated gene recombination in FDC in the spleens, lymph nodes and Peyer's patches of CD21-Cre ROSA26^flox/Flox mice.

A) Analysis of the cellular sites of LacZ expression (blue) in the spleens, inguinal lymph nodes, Peyer's patches and mesenteric lymph nodes of CD21-Cre ROSA26^flox/flox mice shows Cre-mediated recombination in a focus of cells within the B cell follicles. Sections were counterstained with nuclear fast red (red). B) IHC analysis of FDC (CD35⁺ cells, upper row, red) and B cells (CD45R⁺ cells, lower row, red) confirmed that Cre-mediated LacZ expression (blue) was associated with FDC in the spleens of WT→CD21-Cre ROSA26^flox/flox mice. No LacZ expression was associated with FDC in spleens from ROSA26^flox/flox mice that lacked Cre. C) IHC analysis of the status of FDC (CD35⁺ and C4-binding cells; red) and B cells expressing CD45R, CD19, and CD1d (red) in spleens from WT, CD21-Cre ROSA26^flox/flox, WT→CD21-Cre ROSA26^flox/flox and ROSA26^flox/flox mice. Scale bars 100 µm. n = 6 mice/group.

Figure 2. FDC-restricted PrP^c expression in the spleens of Prnp ^stop/-→CD21-Cre Prnp ^stop/- mice.

A) The anticipated distribution of PrP^C expression on FDC and B cells in tissues from each mouse group. B) The detection of Cre in the tail and spleen but not blood of the Prnp ^stop/-→CD21-Cre Prnp ^stop/- mice confirmed the restriction of the Cre-expression to the stromal but not haematopoietic compartments of these mice (upper panel). Efficient Cre-mediated recombination of Prnp^stop (Prnp^{stop (R)}) was restricted to the FDC-containing stromal compartment of the spleens of Prnp ^stop/-→CD21-Cre Prnp ^stop/- mice when compared to control mice. Cre-mediated recombination by CD21-expressing lymphocytes was efficiently prevented in these mice by the irradiation and transfer of Prnp ^stop/- bone marrow as demonstrated by the lack of a Prnp ^stop(R) band in DNA extracted from blood (lower panel). B, blood; S, spleen; T, tail; M, DNA size markers; a, b, c, control DNA samples for each transgene combination tested which were (a) Prnp ^-/-, (b) Prnp ^stop/- and (c) complete recombination of the stop cassette within the Prnp ^stop/- allele. C) IHC analysis of PrP^C expression (blue) by FDC (CD35⁺ cells; red) and sympathetic nerves (TH⁺ cells, green) confirmed PrP^C expression was restricted to FDC in spleens of Prnp ^stop/-→CD21-Cre Prnp ^stop/- mice. Scale bar, 100 µm. D) Morphometric analysis confirmed that the magnitude of the PrP^C expression co-localized upon the surfaces of FDC in the spleens of Prnp ^stop/-→CD21-Cre Prnp ^stop/- mice was similar to that observed upon FDC in spleens from Prnp ^+/-→Prnp ^+/- control mice (p<0.690, n = 48 FDC networks/group). In contrast, in the absence of Cre-recombinase expression by FDC in CD21-Cre Prnp ^stop/-→Prnp ^stop/- mice, PrP^C expression was substantially lower than that observed upon FDC in spleens from Prnp ^+/-→Prnp ^+/- control mice (p<1×10^-25, n = 48 FDC networks/group). E) Morphometric analysis confirmed that PrP^C expression upon the surfaces of sympathetic nerves in the spleens of Prnp ^stop/-→CD21-Cre Prnp ^stop/-, CD21-Cre Prnp ^stop/-→ CD21-Cre Prnp ^stop/- and CD21-Cre Prnp ^stop/-→Prnp ^stop/- mice was significantly ablated when compared to that observed upon sympathetic nerves in spleens from Prnp ^+/-→Prnp ^+/- control mice (p<1×10^-25, n = 48 sympathetic nerves/group). For all panels n = 6 mice/group.

Figure 3. Effect of FDC-restricted PrP^c expression on the spleens of Prnp ^stop/-→CD21-Cre Prnp ^stop/- mice.

A) IHC analysis of the status of FDC networks (C4-binding cells and CD21/CD35⁺ cells; red), B cells expressing CD45R (red), and CD3⁺ T cells (green). Morphometric analysis confirmed that there were no significant difference in the size (B) and number (C) of the CD35⁺ FDC networks in spleens from Prnp ^stop/-→CD21-Cre Prnp ^stop/-, CD21-Cre Prnp ^stop/-→CD21-Cre Prnp ^stop/-, CD21-Cre Prnp ^stop/-→Prnp ^stop/- mice and Prnp ^+/-→Prnp ^+/- control mice (n = 32 FDC networks/group). D and E), Comparison of the sympathetic innervation in spleens from each mouse group. D) IHC detection of TH-positive sympathetic nerves (green) and FDCs (CD35⁺ cells; red). Scale bar, 50 µm. E) Quantitative analysis of the relative positioning of the FDC and sympathetic nerves showed there were no significant differences in average distance between these cell populations in spleens from each mouse group (P = 0.932, n = 48 FDC networks/group). For all panels n = 6 mice/group.

Figure 4. Effect of FDC-restricted PrP^c expression on PrP^Sc accumulation in the spleen.

Mice were injected i.p. with the ME7 scrapie agent and tissues collected 35, 70 days and 105 days after exposure. A and B) High levels of PrP^d were detected in association with FDC (CD21/35 positive cells) in the B cell follicles (CD45R positive cells) of spleens of mice with PrP^C-expressing FDC: Prnp ^stop/-→CD21-Cre Prnp ^stop/- mice, CD21-Cre Prnp ^stop/-→CD21-Cre Prnp ^stop/- mice and Prnp ^+/-→Prnp ^+/- control mice. B) High magnification images of the sites of PrP^d accumulation (red) at 70 days post-injection with scrapie. Arrowheads show PrP-accumulation within tingible body macrophages. C) Analysis of adjacent sections by PET-immunoblot analysis confirmed the presence of PK-resistant PrP^Sc (blue/black). In contrast, no PrP^d or PrP^Sc was detected in spleens of CD21-Cre→Prnp ^stop/-mice that lacked PrP^C-expressing FDC. Arrows indicate PrP^Sc accumulation upon FDC. A, scale bar  = 100 µm. B, scale bar  = 20 µm. C, scale bar  = 500 µm. For all panels n = 4 mice/group.

Figure 5. FDC-restricted PrP^c-ablation in the spleens of Prnp ^flox/-→CD21-Cre Prnp ^flox/- mice.

A) The anticipated distribution of PrP^C expression on FDC and B cells in tissues from each mouse group. B) PCR analysis of DNA isolated from the spleens, blood and tails of Prnp ^flox/-→CD21-Cre Prnp ^flox/- mice confirmed that efficient Cre-mediated DNA recombination and Prnp-ablation (Prnp ^deflox) was restricted to the FDC-containing stromal compartment of the spleen. Cre-mediated recombination of CD21-expressing lymphocytes was efficiently prevented in these mice by the irradiation and transfer of Prnp ^flox/- bone marrow as demonstrated by the lack of a Prnp ^deflox band in DNA extracted from blood (lower panel). B, blood; S, spleen; T, tail; M, DNA size markers; a, b, c, d control DNA samples for each transgene combination tested which were (a) Prnp ^flox/flox,(b) Prnp ^flox/-, (c) Prnp ^-/- and (d) Prnp ^flox/- with complete recombination of the floxed exon 3. C) IHC analysis of PrP^C expression (blue) by FDC (CD35⁺ cells; red) and sympathetic nerves (TH⁺ cells, green) confirmed the PrP^C-ablation was restricted to FDC in spleens of Prnp ^flox/-→CD21-Cre Prnp ^flox/- mice. Scale bar, 100 µm. D) Morphometric analysis confirmed that the magnitude of the PrP^C expression co-localized upon the surfaces of FDC in the spleens of Prnp ^flox/-→CD21-Cre Prnp ^flox/- mice was significantly lower than that observed upon FDC from Prnp ^+/-→Prnp ^+/- control mice (P<1.0×10^-24, n = 48 FDC networks/group). In contrast, in the absence of Cre-recombinase expression by FDC in CD21-Cre Prnp ^flox/-→Prnp ^flox/-mice, PrP^C expression was similar to that observed upon FDC in spleens from Prnp ^+/-→Prnp ^+/- control mice (P = 0.106, n = 48 FDC networks/group). E) Morphometric analysis confirmed that the magnitude of the PrP^C expression co-localized upon the surfaces sympathetic nerves in the spleens of Prnp ^flox/-→CD21-Cre Prnp ^flox/-, CD21-Cre Prnp ^flox/-→ CD21-Cre Prnp ^flox/- and CD21-Cre Prnp ^flox/-→Prnp ^flox/- mice was not significantly different when compared to that observed upon sympathetic nerves in spleens from Prnp ^+/-→Prnp ^+/- control mice (p = 0.400, n = 48 sympathetic nerves/group). For all panels n = 6 mice/group.

Figure 6. Effect of FDC-restricted PrP^C-ablation on FDC status.

A) IHC analysis of the status of FDC (C4-binding cells and CD21/CD35⁺ cells; red), B cells expressing CD45R (red), and CD3⁺ T cells (green). Morphometric analysis confirmed that there were was no significant difference in the size (B) and number (C) of the CD35⁺ FDC networks in spleens of mice from each mouse group (n = 32 FDC networks/group). D and E), Comparison of the sympathetic innervation in spleens from Prnp ^flox/-→CD21-Cre Prnp ^flox/-, CD21-Cre Prnp ^flox/-→CD21-Cre Prnp ^flox/-, CD21-Cre Prnp ^flox/-→Prnp ^flox/-mice and Prnp ^+/-→Prnp ^+/- control mice. D) IHC detection of TH-positive sympathetic nerves (green) and FDCs (CD35⁺ cells; red). Scale bar, 50 µm. E) Quantitative analysis of the relative positioning of the FDC networks and sympathetic nerves showed there was no significant difference in the average distance between these cell populations in spleens from each mouse group (P = 0.765, n = 48 FDC networks/group). For all panels n = 6 mice/group.

Figure 7. Effect of FDC-restricted PrP^C-ablation on immune complex trapping.

A) Mice were passively immunized with preformed PAP immune complexes, and 24 h later, the presence of immune complexes (red) upon FDC (CD35⁺ cells, green) assessed by IHC. Scale bar, 100 µm. B) Morphometric analysis confirmed that the magnitude of the immune complex-trapping co-localized upon the surfaces of FDC from Prnp-ablated Prnp ^flox/-→CD21-Cre Prnp ^flox/- mice was not significantly different from that observed in spleens from control mice. This analysis also confirmed that the immune complexes were preferentially associated with FDC in these tissues and significantly greater than the null hypothesis that the pixels were randomly distributed. *, P<1×10⁻²¹; **, P<1×10⁻³²; *** P<9×10⁻²⁸; n = 40 FDC networks/group. For all panels n = 6 mice/group.

Figure 8. Effect of FDC-restricted PrP^C-ablation on PrP^Sc accumulation in the spleen.

Mice were injected i.p. with the ME7 scrapie agent and tissues collected 70 days after exposure. A) High levels of PrP^d (red, left-hand column) were detected in association with FDC (red, middle column) in the B cell follicles (red, right-hand column) of spleens from CD21-Cre Prnp ^flox/-→Prnp ^flox/-mice and Prnp ^+/-→Prnp ^+/- control mice that contained PrP^C-expressing FDC. B) High magnification images of the sites of PrP^d accumulation (red) at 70 days post-injection with scrapie. C) PET blot analysis of analysis of adjacent sections by PET-immunoblot analysis confirmed presence of PK-resistant PrP^Sc (blue/black). In contrast, no PrP^Sc was detected in spleens of Prnp ^flox/-→CD21-Cre Prnp ^flox/- and CD21-Cre Prnp ^flox/-→CD21-Cre Prnp ^flox/- mice that lacked PrP^C-expressing FDC. In the spleens of some of these mice, low levels of PrP^d were occasionally localised within tingible body macrophages (B, arrowheads). A, scale bar  = 100 µm. B, scale bar  = 20 µm. C, scale bar  = 500 µm. Arrows indicate PrP^Sc accumulation upon FDC. For all panels n = 4 mice/group.

Figure 9. Effect of FDC-restricted PrP^C-ablation on PrP^Sc accumulation in the brains and spleens of scrapie-affected mice.

Control mice (Prnp ^+/- mice) and Prnp ^flox/-→CD21-Cre Prnp ^flox/- mice that lacked PrP^C-expressing FDC were injected i.c. with the scrapie agent directly into the CNS. Brains and spleens were collected from clinically scrapie-affected mice to compare the neuropathology and cellular sites of PrP^Sc accumulation. A) High levels of spongiform pathology (H&E, upper row), heavy accumulations of PrP^d (brown, second row), reactive astrocytes expressing GFAP (brown, third row) and active microglia expressing Iba-1 (brown, bottom row) were detected in the hippocampi of the brains of all clinically scrapie-affected mice. Scale bars, 500 µm. B) High levels of PrP^d (red) were detected in association with FDC in spleens from clinically scrapie-affected control mice that contained PrP^C-expressing FDC. C) PET blot analysis of analysis of adjacent sections by PET-immunoblot analysis confirmed the presence of PK-resistant PrP^Sc (blue/black). In contrast, no PrP^d or PrP^Sc was detected in spleens of Prnp ^flox/-→CD21-Cre Prnp ^flox/- that lacked PrP^C-expressing FDC. Scale bars  = 500 µm.