Production of cattle lacking prion protein

Abstract

Prion diseases are caused by propagation of misfolded forms of the normal cellular prion protein PrP(C), such as PrP(BSE) in bovine spongiform encephalopathy (BSE) in cattle and PrP(CJD) in Creutzfeldt-Jakob disease (CJD) in humans. Disruption of PrP(C) expression in mice, a species that does not naturally contract prion diseases, results in no apparent developmental abnormalities. However, the impact of ablating PrP(C) function in natural host species of prion diseases is unknown. Here we report the generation and characterization of PrP(C)-deficient cattle produced by a sequential gene-targeting system. At over 20 months of age, the cattle are clinically, physiologically, histopathologically, immunologically and reproductively normal. Brain tissue homogenates are resistant to prion propagation in vitro as assessed by protein misfolding cyclic amplification. PrP(C)-deficient cattle may be a useful model for prion research and could provide industrial bovine products free of prion proteins.

Figure 1

Generation of PRNP^−/− cattle. (a) PRNP^−/− cattle at 13 months of age. (b) Verification of PRNP^−/− genotype in the ear biopsy fibroblasts by genomic PCR. P, positive control; N, negative control. The PRNP^−/− calf is positive with puroF14 × puroR14 (middle) and neoF7 × neoR7 (top) primers, which are specific to the targeting events at both alleles of PRNP. The calf is negative with wild-type alleles of PRNP amplified with BPrPex3F × BPrPex3R (bottom) primers. (c) RT-PCR analysis on the PRNP^−/− calves shows disruption of mRNA expression in fibroblasts of PRNP^−/− calves. Primers:PrPmF3 × PrPmR3 (ref. 6). (d) Western blot analysis on the PRNP^−/− calf shows absence of PrP^C protein in fibroblasts. As a positive control, a wild-type (WT) calf was analyzed. As a negative control, protein extracts from mouse fibroblasts were used because the monoclonal antibody used is claimed to be specific to bovine PrP^C protein. Protein extracts from wild-type calf show the presence of 33–35 kDa of bovine PrP^C protein in size, but no positive band from the PRNP^−/− calf. Its replica blot was probed with anti-β actin antibody and served as an internal positive control. (e) Absence of PrP^C protein in peripheral blood lymphocytes (PBLs) of PRNP^−/− calf by western blot analysis. (f) Absence of PrP^C protein in brain stem of PRNP^−/− calves by western blot analysis.

Figure 2

Histopathological analysis of obex and cerebellum of 14-month-old cattle. (a,b) PRNP^+/+ dorsal motor nucleus of vagus (a) and molecular layer, granular layer and white matter (b). (c,d) PRNP^−/− dorsal motor nucleus of vagus (c) and molecular layer, granular layer and white matter (d). There are neither plaques of spongiform tissues nor apparent neurodegeneration in the tissues. H & E stain. Scale bars, 100 µm.

Figure 3

Comparative analysis of immune system of PRNP^−/− and wild-type (WT) control cattle at 12–13 months old. (a) Flow cytometry in peripheral blood lymphocyte (PBL), stained with anti-IgM and anti-CD21 antibodies. (b) PBLs stained with anti-IgM and anti-lambda light-chain antibodies. (c) PBLs stained with anti-CD4 and anti-CD8 antibodies. (d) PBLs stained with anti-CD3 and anti-γδ T cell–receptor antibodies. (e) Secondary antibody isotype control staining. (f) In vitro mitogenic response of T cells in PRNP^−/− and WT cattle. PBLs from four PRNP^−/− and four WT cattle were cultured with medium only (Med) or stimulated with immobilized anti-CD3 monoclonal antibody (CD3), Con A (concanavalin A) or PHA (phytohemagglutinin) mitogens for 48 h and proliferation was measured by ³H thymidine incorporation. Mean of T-cell response of PRNP^−/− group and WT group cattle and their s.e.m. are shown. No significant differences were found. (g) Intracellular cytokine analysis of IFNγ expression in PRNP^−/− and WT control cattle by dual-color flow cytometry. PBLs were stimulated by immobilized anti-CD3 monoclonal antibody for 72 h and intracellular IFNγ production was analyzed by surface CD3 and intracellular IFNγ (positive, green; negative, red) dual-color immunofluorescent staining. Percentage of IFNγ⁺ T cells are shown in the upper right quadrant. (h) In vitro IFNγ production by PBLs in PRNP^−/− and WT cattle. PBLs isolated from four PRNP^−/− and four WT cattle were stimulated by (i) immobilized anti-CD3 monoclonal antibody or (ii) Con A for 72 h and secreted IFNγ in the culture supernatant was analyzed by calibrated bovine IFNγ ELISA. Mean of the IFNγ production (ng/ml) in PRNP^−/− group and WT control cattle and their s.e.m. are shown. Statistical analysis using Student’s t-test showed no significant difference between PRNP^−/− and WT cattle (P = 0.5). (i) Humoral immune response to ovalbumin protein antigen in PRNP^−/− and WT cattle. Four PRNP^−/− and four WT cattle were immunized with ovalbumin twice at day 0 (V1) and day 21 (V2) and ovalbumin-specific IgG antibody titers at 7 d after V2 were determined. Mean antibody titers of PRNP^−/− group and WT group cattle and their s.e.m. are shown. Statistical analysis using Student’s t-test showed no significant difference between PRNP^−/− and WT cattle (P = 0.9).

Figure 4

In vitro propagation of PrP^BSE and PrP^TME in PRNP^−/− and PRNP^+/+ wild-type (WT) cattle brain homogenates. (a,b) In vitro propagation of PrP^BSE in 10% homogenates from cortex (a) or hypothalamus (b). (c) in vitro propagation of PrP^TME in 10% homogenates from cortex (c). The pathological form of the prion protein, PrP^BSE or PrP^TME, in the inoculum, was derived from BSE- or TME-affected cattle, respectively. We used 1:50, 1:100 and 1:200 dilutions of the infectious material. Samples were either frozen immediately after mixture (F) or subjected to 48 PMCA amplification cycles (S). The appearance of PrP^BSE or PrP^TME was assessed by western blot analysis after proteinase K (PK) digestion. Samples from PRNP^+/+ wild-type (WT-Bo), PRNP^−/− (KO-Bo) cattle (substrates) and the BSE-positive control brain homogenate (inoculum) are shown for comparison with and without PK treatment.