Assessing the Susceptibility of Transgenic Mice Overexpressing Deer Prion Protein to Bovine Spongiform Encephalopathy

Abstract

Several transgenic mouse models have been developed which facilitate the transmission of chronic wasting disease (CWD) of cervids and allow prion strain discrimination. The present study was designed to assess the susceptibility of the prototypic mouse line, Tg(CerPrP)1536^+/−, to bovine spongiform encephalopathy (BSE) prions, which have the ability to overcome species barriers. Tg(CerPrP)1536^+/− mice challenged with red deer-adapted BSE resulted in 90% to 100% attack rates, and BSE from cattle failed to transmit, indicating agent adaptation in the deer.

TEXT

Chronic wasting disease (CWD) is a burgeoning transmissible spongiform encephalopathy (TSE) epidemic of captive and free-ranging cervids in North America and South Korea (1, 2). CWD is the only known TSE of wild animals and appears to be transmitted with unprecedented efficiency. Consequently, since its initial identification in northern Colorado and southeastern Wyoming, the disease has spread inexorably among North American states and provinces (3). To minimize the incidence of and ultimately to eliminate CWD, extensive surveillance and research projects have been developed (4, 5). The ability to study the biology of CWD prions has been enhanced by the development of transgenic mice which overexpress cervid PrP on a mouse null PrP background (6–8).

While TSEs usually show species specificity, bovine spongiform encephalopathy (BSE) has the ability to overcome several species barriers with relative ease. It has been shown that BSE can either naturally or experimentally infect a wide range of species, including mice, sheep, goats, cats, macaques, mink, red deer, and humans (9–15). Until BSE is eliminated, it will remain a threat to susceptible species, including cervids, which could potentially act as a reservoir of the agent. It is therefore necessary to develop and validate sensitive experimental models which offer the ability to reliably detect, isolate, and characterize BSE after it has been passaged through deer. In this respect, transgenic mouse lines which overexpress cervid PrP transgenes on a mouse PrP null background are the models of choice since PrP primary structure identity between inoculated PrP^Sc and transgene-expressed PrP^C in recipient mice contributes in reducing transmission barriers (16). Here we tested the susceptibility of Tg(CerPrP)1536^+/− mice (16) to BSE and European red deer (Cervus elaphus elaphus)-adapted BSE.

Red deer-adapted BSE was produced after intracerebral (IC) challenge of European red deer with a bovine BSE source (BBP 12/92) (15). Two of these deer, each with a heterozygous E/Q polymorphism at codon 226, were included in the current study. Tg(CerPrP)1536^+/− mice were challenged either intracerebrally (IC; 0.02 ml) or IC and intraperitoneally (IC & IP; 0.02 and 0.1 ml, respectively) in groups of 10 mice with a 10% (wt/vol) concentration of either (i) one of two deer-adapted BSE sources (D1 and D2) or (ii) the original bovine BSE inoculum that was used to inoculate the deer (BBP12/92). As an experimental control, tg110 mice, a transgenic line which overexpresses bovine PrP on a murine PrP null background and is very sensitive to bovine BSE, was also challenged with the BBP12/92 inoculum. All mice were monitored from 30 days postinoculation (dpi) and were euthanized using carbon dioxide when they showed clinical signs associated with TSE disease or due to other welfare considerations; the mice in the second group were categorized as representing intercurrent deaths and included TSE-negative and preclinical TSE-positive mice. At necropsy, the brain of each mouse was removed and cut parasagitally into two parts. The larger part was fixed in buffered formalin and processed for histology to detect and semiquantitatively assess spongiosis to produce lesion profiles (17). Fixed material was also assessed by immunohistochemistry (IHC) using Rb486 antibody for detection of the PrP^Sc distribution in the brain (18, 19). Western blot analysis was performed on brain tissue (the smaller part) from challenged mice and also on the inoculum using a Bio-Rad TeSeE Western blot kit. Murine brain tissues were extracted as described previously (20). Inoculum samples were centrifuged at 348,000 × g for 30 min and the pellets rehomogenized in proprietary kit (Bio-Rad Western blot) homogenization buffer prior to continuation of the extraction process according to kit instructions. Samples were diluted in Laemmli buffer to achieve good band definition, and 15 μl was loaded per lane. Proteins were separated on 12% Bis-Tris (Criterion XT; Bio-Rad) gels before electrotransfer to an activated membrane was performed. Unbound membrane was blocked using 5% (wt/vol) bovine serum albumin (BSA) in PBS supplemented with 0.1% (vol/vol) Tween 20 (PBST). Bound PrP^Sc was labeled with biotinylated Sha31 anti-PrP monoclonal antibody, detected with streptavidin peroxidase (Sigma), and visualized by enhanced chemiluminescence (ECL; Amersham) (20, 21). All animal experiments were performed in accordance with the Animals (Scientific Procedures) Act of 1986 using Home Office project licence number (PPL70/7167) and approved by a local ethics committee.

Our results show that Tg(CerPrP)1536^+/− mice challenged with either deer-adapted BSE source (D1 or D2) succumbed to disease with a 90% to 100% attack rate (Table 1), regardless of the inoculation route. The incubation periods of mice inoculated i.c. were 238 ± 27 and 252 ± 34 (mean ± standard deviation) dpi, respectively. The incubation periods produced by the same sources inoculated IC & IP were 242 ± 17 and 261 ± 18 dpi, respectively. These data suggest that a combined IC & IP route of inoculation does not offer significant advantages with respect to the sensitivity of this mouse line to detect red deer-adapted BSE compared to an IC-only challenge. Incubation period, attack rate, lesion profile, IHC, and Western blot analyses all indicated that a single agent was isolated from D1 and D2 (Fig. 1, 2, and 3). In contrast, all Tg(CerPrP)1536^+/− mice that were inoculated with the bovine BSE source (BBP12/92) were TSE negative. The attack rate of BBP12/92 in tg110 mice, however, was 90%, with incubation periods ranging from 231 to 270 dpi (Table 1). These values are comparable to the results of previous BSE transmissions in tg110 mice (22, 23), suggesting that the lack of transmission of BBP12/92 in Tg(CerPrP)1536^+/− mice was not due to instability of the agent during storage. Western blot analysis of BBP12/92, D1, and D2 inocula also suggests that they contained similar quantities of PrP^Sc (Fig. 3).

TABLE 1.

Transmission of deer-adapted BSE and bovine BSE to transgenic mice

BSE source	Mouse line	Challenge	np/nta	Incubation periodc
D1	Tg(CerPrP)1536+/−	IC	9/10b	238 ± 27
		IC & IP	9/10b	242 ± 17
D2		IC	9/10b	252 ± 34
		IC & IP	10/10	261 ± 18
BBP12/92		IC	0/10	868d
		IC & IP	0/10	892d
	Tg110	IC	9/10b	255 ± 14

^an_p, numbers of TSE-positive mice irrespective of clinical status; n_t, numbers of mice tested. Differences between n_p and n_t values are due to intercurrent deaths attributed to TSE-negative mice.

^bIn these groups, intercurrent deaths attributed to TSE-negative mice were detected prior to the diagnosis of the first TSE-positive mouse.

^cIncubation period was recorded as days postinoculation (dpi) ± standard deviation. Only clinically positive and TSE-confirmed mice were included in the assessment of incubation periods.

^ddpi of last TSE-negative mouse.

FIG 1.

Susceptibility of Tg(CerPrP)1536^+/− mice to BSE and cervid-adapted BSE. (A) Survival curves of Tg(CerPrP)1536^+/− mice inoculated with bovine BBP12/92 and cervid D1 and D2 isolates. Mice with confirmed clinical TSE status affect the profile of the curve, and they are indicated as downward-oriented steps. TSE-negative mice are indicated by upward-oriented marks. Downward-oriented marks indicate TSE-confirmed mice which died prior to the development of clinical signs. The last two categories of subjects were treated as intercurrent deaths and do not alter the profile of the curve. (B) Lesion profiles of cervid-adapted BSE isolates D1 and D2 in Tg(CerPrP)1536^+/− mice. IC, intracerebral challenge; IC&IP, combined intracerebral and intraperitoneal challenge. Lesion profiles depicting means and standard errors of the means (SEMs) of the results from experiments performed with at least five clinically positive mice per brain area are shown. Gray matter areas were scored as follows: G1, dorsal medulla nuclei, including cochlear nuclei; G2, granular layer of the cerebellar cortex adjacent to the fourth ventricle; G3, superior colliculus; G4, hypothalamus; G5, thalamus; G6, hippocampus; G7, septal nuclei of the paraterminal body; G8, cerebral cortex (at the level of G4 and G5); and G9, cerebral cortex (at the level of G7). White matter areas were scored as follows: W1, cerebellar white matter; W2, mesencephalic tegmentum; and W3, the cerebral peduncles.

FIG 2.

Immunohistochemical staining of PrP^Sc in the central nervous system (CNS) of a diseased Tg(CerPrP)1536^+/− mouse. (A) PrP^Sc-specific deposits in the hippocampal and dorsal thalamic areas of a mouse challenged with cervid-adapted BSE. (B) Lack of PrP^Sc-specific deposits in Tg(CerPrP)1536^+/− mice challenged with the BBP12/92 BSE inoculum. Bars, 500 μm.

FIG 3.

Western blot of Tg(CerPrP)1536^+/− mice challenged with cervid-passaged BSE and cattle-derived BSE. BBP 12/92, bovine BSE brain pool; D1 inoculum and D2 inoculum, 10% brain homogenates of diseased European red deer challenged with BBP12/92; D1 and D2, mice challenged with D1 inoculum and D2 inoculum, respectively; IC, intracerebral challenge; IC&IP, combined intracerebral and intraperitoneal challenge; BBP12/92 inoculum, a 10% homogenate of BBP12/92 which was used to inoculate European red deer that were used as deer BSE sources in the current study; BBP12/92 IC or IC&IP, mice were inoculated with BBP12/92 inoculum IC or IC&IP, respectively. Samples were diluted to give optimal band definition. The equivalent values for wet weight tissue loaded per lane were as follows: BBP12/92, 0.05 mg; D1 inoculum, 0.19 mg; D1 IC, 0.06 mg; D1 IC/IP, 0.19 mg; D2 inoculum, 0.19 mg; D2 IC, 0.06 mg; D2 IC/IP, 0.075 mg; BBP12/92 inoculum, 0.19 mg; BBP12/92 IC and BPP12/92 IC/IP, 3.75 mg. Note that equal amounts of tissue (0.19 mg per lane) were loaded for all inocula.

The bioassay results could be attributed to differences in the cervid PRNP between European red deer (Cervus elaphus elaphus), which are susceptible to bovine BSE, and mule deer (Odocoileus hemionus hemionus), whose PrP sequence was used to generate Tg(CerPrP)1536^+/− mice. Sequence analysis revealed only one variation between the two genera, a substitution of glutamic acid (E) with glutamine (Q) at codon 226. In European red deer, this position is dimorphic, showing either E or Q. Indeed, the red deer originally inoculated were heterozygous (E/Q) at residue 226. Other cervid species in which CWD has been detected are not polymorphic at this position. Thus, mule deer, white-tailed deer, and moose show Q at 226, while Rocky Mountain elk show E at 226. Nevertheless, this polymorphism does not seem to alter the susceptibility of red deer to bovine BSE (15).

Similar data have been reported elsewhere where another mouse line overexpressing a cervid transgene, Tg(ElkPrP), was resistant to bovine BSE but was rendered susceptible when the BSE had been adapted to sheep (24). Collectively, these observations suggest that the biological characteristics of bovine BSE can be altered after it is passaged in deer or sheep to allow the agent to circumvent the species barrier in relation to transgenically modified mice. This phenomenon could be attributed, at least partially, to genetic factors other than PRNP which could modulate, presumably in conjunction with PRNP, the transmission of specific TSE strains to certain hosts as reported previously (25–28).

It could theoretically be argued that the data presented here represent titer differences between bovine- and deer-adapted BSE inocula. In our view, this explanation is improbable as the infectivity data for the deer-adapted BSE sources are associated with high attack rates and short mean incubation periods with tight variances which are compatible with high titers. In contrast, the bovine BSE source failed to propagate in the cervid mice although it transmitted readily to the bovine-adapted tg110 mice with attack rates and incubation period data associated with high titers in this mouse line (23). Also, the deer-adapted and bovine BSE inocula had similar levels of PrP^Sc (equal amounts of tissue were loaded in each case, and the intensities of the Western blot signals were similar), despite the disparity in the bioassay data. Taken together, these observations suggest that adaptation of BSE into the deer was necessary for the agent to propagate into the cervid mice, indicating that other, yet unknown species-specific factors are involved in the conversion of PrP^C to PrP^Sc.

In conclusion, our data show that the Tg(CerPrP)1536^+/− line is susceptible not only to CWD (19) and to natural scrapie prions (29) but also to cervid-adapted BSE, suggesting that this line can provide a successful experimental model for surveillance and research of cervid TSEs to investigate infectivity and characterize cervid-derived TSE strains. They also support the view that caution must be exercised when using transgenic models to predict the susceptibility of the respective natural host species to infection across a species barrier and to highlight the importance of other genetic factors that may be involved in TSE pathogenesis.