Skip Navigation

Bovine Spongiform Encephalopathy Agent in a Prion Protein (Prp)ARR/ARR Genotype Sheep after Peripheral Challenge: Complete Immunohistochemical Analysis of Disease-Associated Prp and Transmission Studies to Ovine-Transgenic Mice

  1. Anna Bencsik and
  2. Thierry Baron
  1. Agence Française de Sécurité Sanitaire des Aliments, Unité Agents Transmissibles Non Conventionnels, Lyon, France
  1. Reprints or correspondence: Dr. Anna Bencsik, AFSSA, Unité ATNC, 31 ave. Tony Garnier, 69364 Lyon, cedex 07, France (a.bencsik{at}lyon.afssa.fr).

Abstract

Possible transmission of the bovine spongiform encephalopathy (BSE) agent to ovine species has been considered for several years. It has been recently demonstrated that the BSE agent, after intracerebral challenge, can infect sheep believed to be the most resistant genetically to prion diseases (prion protein [PrP]ARR/ARR genotype). We report here the results of a detailed immunohistochemical analysis of the disease-associated PrP (PrPd) in all organs from a PrPARR/ARR sheep infected with the BSE agent by a peripheral route. Because PrPd was detected in the brain in the absence of any clinical symptoms, transmission studies were also performed using a sensitive ovine-transgenic mouse model—Tg(OvPrP4)—that can identify the BSE agent on the basis of the occurrence of florid plaques in the mouse brain. The data indicated that these PrPd deposits were linked to the BSE agent and were associated with infectivity. This suggests that PrPARR/ARRM sheep may be silent carriers of the BSE agent.

Transmissible spongiform encephalopathies (TSEs), or prion diseases, are fatal human and animal neurodegenerative diseases that include Creutzfeldt-Jakob disease (CJD) in humans, bovine spongiform encephalopathy (BSE) in cattle, and scrapie in sheep and goats. A common characteristic of these diseases is the accumulation of an abnormal protease-resistant conformer of the host-encoded prion protein (PrP), sometimes termed PrPSc, to indicate its association with scrapie; PrPres, to indicate resistance to proteinase K digestion; and PrPBSE or PrPCJD, in the case of BSE and CJD [1]. We use here the abbreviation PrPd, already adopted by our group and others, because it embraces the biochemical and species properties of the abnormal form of disease-associated PrP in a more universal way [2, 3].

BSE is the most probable source of contamination of human beings, after the occurrence of the CJD variant [4]. Scrapie, in contrast, is a contagious form of TSE in sheep [5] that has never before been associated with human disease. However, infection with the BSE agent may have occurred when BSE-contaminated meat and bone meal were given to sheep and goats during the 1980s and may have been subsequently maintained within the sheep and goat population [6]. The consumption of meat from sheep infected with the BSE agent might present a risk to humans. Numerous experimental transmissions of the BSE agent to sheep and goats were, therefore, performed by oral, intravenous, or intracerebral inoculation. They all demonstrated that the BSE agent could easily cross the species barrier and showed that both the clinical expression and distribution of the abnormal form of PrP was indistinguishable from that of natural sheep scrapie [7].

Programs have been in progress throughout Europe to search for possible “ovine BSE,” and, in France in 2005, a BSE signature was first reported in a natural case of TSE in a goat [8]. To reduce the scrapie susceptibility of the sheep population and, if present, the prevalence of ovine BSE cases, selective breeding is now under way in many countries, based on the selection of TSE-resistant genotypes [9]. In fact, polymorphisms in the gene for PrP at residues 136, 154, and 171 are associated with susceptibility to TSEs [10]. The A136R154R171/A136R154R171 genotype, used for selective breeding, is considered to impart the strongest resistance to TSE in sheep, whereas the A136R154Q171/A136R154Q171 genotype induces susceptibility to these diseases, especially the BSE agent [5, 7]. Unexpectedly, infection of ARR homozygous sheep with the BSE agent is possible by the intracerebral route [11], but the significance of this susceptibility needs to be determined under more natural conditions of exposure. It was shown more recently that early oral exposure to the BSE agent led to the detection of abnormal PrP in the spleen of a single ARR homozygous sheep 10 months after exposure [12].

Here, we report the transmission of the BSE agent to an ARR homozygous sheep by a peripheral route (intrasplenic) and the detection of PrPd in several areas of the central nervous system (CNS) but not in other peripheral organs. Our main purpose was to see whether PrPd was linked to infectivity, because of its detection in the brain in the absence of any clinical symptoms, and our second purpose was to check its signature in terms of BSE origin. Transmission studies were performed using a sensitive ovine-transgenic mouse model—Tg(OvPrP4)—that is known to distinguish the BSE agent [1315]. BSE identification in inbred wild-type mice essentially relies on the demonstration of a specific distribution of spongiform lesions throughout the brain [1618], whereas, in this ovine-transgenic mouse model that uses inoculation with experimental ovine BSE, the BSE signature is considered to be the identification of typical florid plaques, as previously described in humans with variant CJD or in macaques infected with the BSE agent [14, 19]. Brain extracts from the midbrain, brain stem, and thoracic level of the medulla of the BSE-challenged ARR homozygous sheep were therefore injected into 3 groups of Tg(OvPrP4) mice, and the development of a murine TSE disease, PrPd expression, and florid plaques were examined.

Materials and Methods

Ovine experimental scheme. The experimental transmission of the BSE agent to sheep was started in 1997 and has been described elsewhere [2, 20, 21]. Two Lacaune sheep of PrPARQ/ARQ genotype were inoculated by peripheral routes (either intraperitoneal or intrasplenic) with 5 mL of a 10% brain homogenate from a French BSE-affected cow. Both sheep showed clinical signs of neurological disorders, died 672 and 1444 days after inoculation, and exhibited PrPd accumulation in the brain and peripheral organs [21].

The same peripheral routes and inoculum were used to inoculate 2 other Lacaune sheep of the ARR/ARR genotype. The intraperitoneally and intrasplenically challenged sheep were killed 673 and 2191 days after inoculation, respectively, without any clinical signs of prion disease. Samples from healthy (noninoculated) 2-year-old sheep either presenting ARR/ARR and ARR/ARQ genotypes or of unknown genotype but <1 year old were used as negative controls. The negative clinical status of these control animals was confirmed by checking the absence of PrPd from the brain and tonsils by immunohistochemistry.

Studies of the PrPARR/ARR sheep inoculated with the BSE agent. Complete PrPd immunohistochemical analysis of the final Lacaune sheep of ARR/ARR genotype killed 2191 days after inoculation is described here. This animal did not present any clinical symptoms and was killed by intravenous injection of pentobarbital (10 mg/kg) followed by exsanguination. Samples were rapidly removed from each organ and fixed for several days in buffered 10% formalin (4% formaldehyde), giving a total of 160 tissue samples (table 1).

Transmission studies in Tg(OvPrP4) mice. Groups of 10 female Tg(OvPrP4) mice, 4–6 weeks old, were injected by the intracerebral route with 20 μL of 10% brain homogenates from the midbrain, brain stem, and thoracic level (T7–T8) of the medulla of a PrPARR/ARR sheep inoculated with the BSE agent. The mice were coded and scored for clinical signs of TSE according to criteria described elsewhere [14, 15] and were killed at clinical end point. The incubation period for each transgenic mouse was calculated as the interval between injection and death. Brains were removed and either frozen for PrPd biochemical studies [22] or fixed in 10% formol saline for histopathological assessment.

issue processing. Coronal slices (5 mm thick) of the different fixed sheep tissue samples (n = 160) were placed in a formic acid bath (98%–100%; VWR International) for 1 h at room temperature, to reduce infectivity [23]. The mouse brains were trimmed to give coronal sections at levels described in TSE strain-typing studies [24]. After classical embedding in paraffin (Excelsior; Thermo Electron), 5-μm brain sections were prepared and placed on pretreated slides (Starfrost; Medite Histotechnic). Once dewaxed (Automate; TechInter), slides were stained for either histopathological or immunohistochemical examination. Hematoxylin-eosin staining was used for the histopathological investigations. Sections were also stained with Congo red to reveal any amyloid-type deposits characteristic of PrPd florid plaques [14].

PrPd immunohistochemistry. Tissue slices were immunostained for PrPd by use of different anti-PrP monoclonal antibodies (MAbs) (table 2). The SAF84 MAb was used on sheep and mouse tissue sections, as described elsewhere [26]. Briefly, the tissue sections were first treated with 98%–100% formic acid for 10 min at room temperature, followed by hydrated autoclaving for 20 min at 121°C in distilled water (Prestige Medical). Proteinase K digestion was performed at 20 μg/mL (Roche Diagnostics) for 15 min at 37°C.

The P4 MAb was used in the ARR homozygous sheep, as described elsewhere [2]. The 2G11 MAb was used with a newly described sensitive amplifying method [26] that includes an additional step with streptomycin sulphate before incubation with the primary antibody, combined with the conventional use of a peroxidase-labeled avidin-biotin complex system (Vectastain Elite ABC; Vector Laboratories); this method remarkably enhances PrPd detection [26]. Final revelation was achieved with diaminobenzidine intensified with nickel chloride (DABNi) to give black deposits. The sections were then counterstained with aqueous hematoxylin, dehydrated, mounted using Eukitt (Sigma-Aldrich), and observed under a light microscope (BX51; Olympus) coupled to an image-analysis workstation (MorphoExpert software; Explora Nova).

Results

Detection of PrPd in the brain of a PrPARR/ARR sheep inoculated with the BSE agent but not in the peripheral organs. Although this ARR homozygous sheep did not show any clinical symptoms, the animal was killed 2191 days after intrasplenic challenge with the BSE agent. The potential presence of PrPd was first determined by immunohistochemistry using different anti-prion antibodies on the brain stem sample, where PrPd was clearly detected in the obex (figure 1G). In contrast, no abnormal PrP was detected in the same region of the first ARR homozygous sheep that was challenged with the BSE agent by the intraperitoneal route with the same inoculum and that died 673 days after inoculation or in another ARR homozygous sheep from a flock affected with natural scrapie [21, 30]. Remarkably, the 2G11 MAb gave the strongest PrPd signals, whereas SAF84 only faintly labeled the obex region, underscoring the necessity of choosing the best antibody for this genotype. Very few neurons accumulated PrPd in this PrPARR/ARR sheep—even within a particular nucleus, such as the dorsal motor nucleus of the vagus (figure 1G)—but, paradoxically, in each neuron the PrPd deposits were heavily covering the cytoplasm. No neuropile-associated PrPd deposits or stellate or perivascular accumulations were observed. The P4 MAb was used on the brain-stem sample to characterize the probable BSE origin of this PrPd, leading to loss of the strong intraneuronal pattern. All other brain areas were then systematically analyzed using the 2G11 MAb, and the results are summarized in figure 1. Rostrocaudal study of the neuroanatomical distribution of PrPd indicated that all subdivisions (preoptic, anterior, tuberal, and mammillary) of the hypothalamic complex were targeted. The cell bodies of the median preoptic, supraoptic, paraventricular, and ventromedial nuclei, as well as the medial mammillary, supramammillary, and ventral tuberomammillary nuclei, had all accumulated PrPd (figure 1A and 1B). In the hippocampus, PrPd was clearly associated with neural cell processes subjacent to the granular cell layer of the CA1 field, as illustrated in figure 1C. Only a few cells were clearly labeled intraneuronally within the periaqueducal gray area (figure 1D). In themidbrain, several PrPd-accumulating cells were identified in the oculomotor nucleus, the central linear nucleus, the substantia nigra, the interpedoncular nucleus, and the raphe nuclei, in which a few neurons presented vacuolated cytoplasm (figure 1E). No PrPd deposition was apparent in the cortical parts of the cerebellum. Significant amounts of PrPd were detected in the deep, medial, interposed, and lateral cerebellar nuclei (figure 1F). Small but intensively labeled amounts of PrPd-accumulating cells were also observed in the medulla oblongata, the dorsal motor nucleus of the vagus, the spinal tract of the trigeminal nerve, the reticular formation, and the raphe nuclei (figure 1E and 1G). No PrPd was detected in the cortex, striatum, or thalamus. All spinal cord sections from the cervical to lumbar regions presented PrPd-accumulating cells in the central gray matter, especially in the intermediolateral nucleus and in some neurons of the ventral, but never the dorsal, horn (figure 2AD).

Figure 1.

Disease-associated prion protein (PrPd) distribution in the brain of a PrPARR/ARR sheep challenged with the bovine spongiform encephalopathy agent by the intrasplenic route. On the left, the blue areas on the schematic coronal brain sections indicate regions where PrPd was observed after amplified immunohistochemistry using the 2G11 monoclonal antibody; on the right are illustrations of representative types of PrPd deposition. A, Hypothalamic median preoptic nucleus. PrPd was seen in the hypothalamic median preoptic nucleus predominantly as granular deposits within the cytoplasm. The very limited granular deposits in the neuropile could be seen in this nucleus. PrPd was apparently not associated with the vascular elements. B, Paraventricular nucleus of the hypothalamus. In the paraventricular nucleus of the hypothalamus, a few neurons were heavily labeled with a granular intracytoplasmic deposit (inset shows higher magnification). C, Hippocampus. In the hippocampus, PrPd deposition was observed at the CA1 level as fine deposits suggestive of the labeling of neural processes (inset shows higher magnification). D, Periaqueducal gray matter. In the periaqueducal gray matter, only a few neurons strongly accumulated PrPd within their cytoplasm (inset shows higher magnification). E, Raphe nuclei. In the raphe nuclei, PrPd was detected within the neuron cytoplasm, which was also sometimes vacuolated; vacuolation was not detected elsewhere. F, Deep cerebellar nuclei. Granular PrPd deposits were also observed in the deep cerebellar nuclei, as illustrated here in the medial cerebellar nucleus. G, Obex. In the obex, the dorsal nucleus of the vagus nerve showed few neurons strongly accumulating PrPd within their cytoplasm. Bars are as follows: 160 μm for panel E; 95 μm for panel B; 60 μm for panels A, C, F, and G; and 30 μm for panels D and E.

Figure 2.

A, Disease-associated prion protein (PrPd) distribution in the T7 spinal cord section of a PrPARR/ARR sheep challenged with the bovine spongiform encephalopathy agent. PrPd was not detectable in the dorsal horn. At higher magnification, a limited number of PrPd-positive neurons were observed in the ventral horn (B). Panels C and D illustrate PrPd detection in the left and right intermediolateral nuclei, which show granular intracytoplasmic and neuropile deposition. Bars are as follows: 630 μm for panel A, and 60 μm for panels B–D.

Complete immunohistochemical analyses of the other organs are summarized in table 1. No PrPd accumulation could be detected in any of these organs. No PrPd could be detected in the lymphoid system of this sheep.

Figure 3.

Neuropathological analysis of Tg(OvPrP4) mice inoculated with spinal cord (A), brain stem (B), and midbrain (C and D) samples from a prion protein (PrP)ARR/ARR sheep challenged with the bovine spongiform encephalopathy agent revealed the presence of typical florid plaques. Amyloid florid plaques formed clusters in the cortex (PrPd immunohistochemistry; A) as well as in the hippocampus (stained with red Congo dye; C). In these structures, the diameters of the florid plaques ranged from 30 to 100 μm (D and B). Bars are as follows: 60 μm for panel A, 30 μm for panel C, and 16 μm for panels B and D.

Table 1.

Organs analyzed for disease-associated prion protein (PrPd) by immunohistochemistry in PrPARR/ARR sheep inoculated with the bovine spongiform encephalopathy agent.

Detection of PrPd and infectivity in the midbrain, brain stem, and medulla of the asymptomatic PrPARR/ARR sheep inoculated with the BSE agent, and observation of the BSE signature, florid plaque, in Tg(OvPrP4) mouse transmission studies. Three groups of Tg(OvPrP4) mice were inoculated with midbrain, brain stem, and spinal cord extracts from this asymptomatic ARR homozygous sheep challenged with the BSE agent and were clinically monitored to see whether the PrPd detected in the CNS of this sheep was associated with infectivity. A typical mouse TSE occurred in each group (the panel of clinical symptoms has been described elsewhere [14, 15]). The mean incubation times are given in table 3. Immunohistochemistry analyses with the SAF84 MAb revealed the presence of PrPd in each mouse. The mapping of PrPd within a group and between groups was similar and is summarized in table 4. In all groups, PrPd was detected in the cortex (figure 3B and 3D), hippocampus (figure 3A and 3C), thalamus, substantia nigra, and raphe nuclei as florid plaques. These plaques also bound to Congo red dye and displayed the green-gold birefringence in polarized light that is characteristic of amyloid. Some florid plaques were particularly huge, attaining 100 μm in diameter (figure 3B). PrPd deposition in the hypothalamus and brain stem was granular. No PrPd was detected in the subcallosal region, habenula, or cerebellum. Florid plaques were detected in the colliculus, but not when the midbrain was used as inoculum. Similarly, they were not detected in the septum of brain stem—inoculated mice, in contrast to the 2 other groups. Florid plaques were only detected in the caudate putamen of mice inoculated with spinal cord extract.

Table 2.

Prion protein (PrP) primary antibodies used.

Table 3.

Transmission of disease-associated prion protein (PrPd) from an asymptomatic PrPARR/ARR sheep inoculated with the bovine spongiform encephalopathy agent to Tg(OvPrP4) mice.

Table 4.

Disease-associated prion protein (PrPd) distribution within Tg(OvPrP4) mice (n = 5/group) inoculated with midbrain, brain stem, and spinal cord extracts from an asymptomatic PrPARR/ARR sheep inoculated with the bovine spongiform encephalopathy agent.

Discussion

A very sensitive immunohistochemistry procedure detecting PrPd in the brain of a PrPARR/ARR sheep inoculated with the BSE agent by the intrasplenic route was used to demonstrate possible peripheral neuroinvasion in such an animal and confirmed earlier results observed after the administration of the BSE agent by the intracerebral route [11, 31, 32]. The previously described P4 MAb discriminator property [3, 11, 31] was used to check whether PrPd deposition was truly linked to transmission of the BSE agent, and the expected loss of intraneuronal labeling was observed. Only 1 animal was studied, which makes the extrapolation of results difficult, but these data confirm the possibility that some TSE agents, such as that for BSE, may affect PrPARR/ARR sheep by a peripheral route of infection [33].

The systematic immunohistochemical assessment of PrPd accumulation revealed differences in the distribution of PrPd in the brain and the type of deposition. In contrast to the reported infection of PrPARR/ARR sheep with the BSE agent by the intracerebral route, the presence of PrPd in our intrasplenically inoculated PrPARR/ARR sheep was detected in the hippocampus but not in the cerebral and cerebellar cortexes, striatum, and thalamus 31]. Most sites exhibited few strongly labeled cells. Deposition was mainly granular, and aggregates were detected in the neuronal perikarya and neuronal processes. In general, the PrPd profile in this PrPARR/ARR sheep inoculated intrasplenically with the BSE agent was quite different from that reported in a PrPARR/ARR sheep inoculated intracerebrally with the BSE agent [31]. Despite the difficulty of extrapolating from a single case, these discrepancies may simply reflect the different route of inoculation used or different stages in the development of the neuropathological process, because the intrasplenically inoculated PrPARR/ARR sheep was asymptomatic, unlike the 2 intracerebrally inoculated sheep.

In light of the route of inoculation, the absence of detectable PrPd in the lymphoid compartment was particularly unexpected but may be linked to the ARR/ARR genotype, because in PrPARR/ARR sheep intracerebrally inoculated with the BSE agent the lymphoid system seems to be similarly and systematically (n = 5) devoid of any detectable PrPd accumulation [32], unlike the other genotypes studied. Indeed, it has recently been reported that, after early oral exposure to the BSE agent, a single PrPARR/ARR sheep had accumulated PrPd in the spleen 10 months after exposure, so it may be that the lymphoreticular system is only transiently affected in sheep of this particular genotype, as in the transient lymphoreticular targeting described in cattle experimentally infected with the BSE agent [12, 34].

In this context, that the presence of PrPd could not be detected in any other tissue samples analyzed despite the use of a highly enhanced immunohistochemistry procedure must not be thought to indicate its total absence but rather its absence during a certain time of the incubation process. Transmission studies in mice, particularly in ovine PrP—transgenic mouse lines, should help to clarify the real status of these PrPd-free tissues in the presence of the BSE infectious agent.

The successful transmission from 3 different CNS regions of this asymptomatic sheep to ovine-transgenic Tg(OvPrP4) mice provides clues to the significance of the PrPd detected in this case. However, it mainly demonstrates that this transgenic mouse model, highly susceptible to ovine BSE from ARQ/ARQ genotype sheep [14], is also susceptible to ovine BSE from an asymptomatic ARR/ARR genotype sheep—despite the sole expression of the ARQ PrP allele. A typical mouse TSE was induced in each group, with the mean incubation periods ranging from 423 to 518 days after inoculation, and PrPd accumulation was detected in each brain. The apparent variability in the mean incubation periods between the 3 groups probably reveals the different amounts of infectious agent actually present in the various CNS segments and suggests that the midbrain contained less infectious agent than the spinal cord, which is in accordance with the different estimated PrPd levels in these regions. These incubation periods were longer than those described for ovine BSE from ARQ/ARQ genotype sheep and could also be the result of the different ovine prion allele [14, 19].

In any case, the PrPd detected in this ARR homozygous sheep challenged with the BSE agent was associated with an infectious property. In addition, the presence of typical amyloid florid plaques in each group indicated that the infectious agent present in all 3 brain samples was linked to the presence of the BSE agent. PrPd mapping revealed remarkable differences in PrPd distribution relative to that described for Tg(OvPrP4) mice inoculated with the ovine BSE agent from ARQ/ARQ genotype sheep [14]. No florid plaque was observed in the subcallosal region, nor was granular PrPd deposition observed in the habenula. This again may be an “allele effect.” Some subtle differences in PrPd florid plaque mapping were also apparent between the 3 groups, and second-passage experiments from each group should help to evaluate the significance of this observation.

In conclusion, the present study reports essential data on transmission from ARR homozygous sheep challenged with the BSE agent to transgenic mice and helps clarify the issue of genetic resistance to TSE in sheep. Use of the Tg(OvPrP4) mouse line demonstrated the existence of infectivity associated with an asymptomatic PrPARR/ARR sheep and indicated that the PrPd detected in the mouse brain was of BSE origin. These data suggest the possible existence of silent carriers of the BSE agent in ARR homozygous sheep (in the case of BSE contamination of the sheep population) and the need to investigate the kinetics of experimental TSE in sheep of the ARR/ARR genotype, to complete the few published experimental data available for BSE in sheep.

Acknowledgments

We are grateful for the excellent histotechnical assistance of Céline Raynaud and Mikaël Leboidre. We also thank Latefa Chouaf-Lakhdar, Emilie Antier, Clément Lavigne, and Romain Godoye, for their contribution to the mouse experimental studies. We further thank all those who took part in the “ovine BSE” experiment at any level, who are too numerous to be cited here in full.

Footnotes

  • Potential conflicts of interest: none reported.

  • Financial support: “Programme National de Recherches sur les ESST et les Prions” Groupement d'Intérêt Scientifique “Infections à Prions.”

  • Received August 29, 2006.
  • Accepted October 2, 2006.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
| Table of Contents