Experimental infection of meadow voles (Microtus pennsylvanicus) with sheep scrapie

Abstract

Meadow voles (Microtus pennsylvanicus) are permissive to chronic wasting disease (CWD) infection, but their susceptibility to other transmissible spongiform encephalopathies (TSEs) is poorly characterized. In this initial study, we intracerebrally challenged 6 meadow voles with 2 isolates of sheep scrapie. Three meadow voles acquired a TSE after the scrapie challenge and an extended incubation period. The glycoform profile of proteinase K-resistant prion protein (PrP^res) in scrapie-sick voles remained similar to the sheep inocula, but differed from that of voles clinically affected by CWD. Vacuolization patterns and disease-associated prion protein (PrP^Sc) deposition were generally similar in all scrapie-affected voles, except in the hippocampus, where PrP^Sc staining varied markedly among the animals. Our results demonstrate that meadow voles can acquire a TSE after intracerebral scrapie challenge and that this species could therefore prove useful for characterizing scrapie isolates.

Sheep scrapie belongs to a group of diseases referred to as transmissible spongiform encephalopathies (TSEs) or prion diseases. Other TSEs include bovine spongiform encephalopathy (BSE), chronic wasting disease (CWD), and Creutzfeldt-Jakob disease (CJD). Neuropathological hallmarks of TSEs are spongiosis and neuronal loss in the central nervous system (CNS), which lead to dysfunction and death (1). These processes are thought to result from conformational conversion of the normal host-encoded prion protein (PrP^C) to an abnormal, disease-associated isoform (PrP^Sc) that resists degradation and accumulates in insoluble protein aggregates in infected tissues in the nervous system (2).

Studies of scrapie, as the archetypal prion disease, have provided fundamental knowledge on the role of host genetics in susceptibility to prion disease, the distribution of infectivity in host tissues, and the occurrence of different prion strains (2). Research on these topics has relied heavily on bioassays in rodent species, especially mice, due to the impracticality and cost of using large animals in protracted challenge studies. Other advantages of using mice in scrapie research are the ability to generate genetically defined rodent hosts and the finding that the strain diversity of scrapie can be maintained and propagated after serial subpassage in inbred mice (3,4).

Rodent models of scrapie are hindered, however, by some critical limitations including lengthy incubation periods (1 to 2 y) and inefficient or unsuccessful transmission of a number of natural scrapie isolates to mice (5,6). Transgenic mice that overexpress ovine PrP^C have been developed to overcome some of these limitations and serve as competent hosts for scrapie isolates from sheep expressing different genotypes of the prion protein gene (Prnp) (7).

Voles have recently been identified as rodent hosts that are uniquely susceptible to a wide range of different TSEs, including scrapie, CWD, CJD, and rodent-adapted TSEs (8–13). Early challenge studies showed that field voles (Microtus agrestis) were more susceptible to mouse- and rat-passaged scrapie than were gerbils, guinea pigs, and Chinese or golden hamsters (14,15). More recently, fast (145 to 200 d) and efficient transmission of sheep and goat scrapie isolates to European bank voles (Myodes glareolus), including isolates that failed to transmit to laboratory mice or hamsters, has been demonstrated (6,8,9). Our group has identified 2 North American vole species, meadow voles (Microtus pennsylvanicus) and red-backed voles (Myodes gapperi), that are highly permissive to CWD in experimental challenges (12). In the present study, we investigate the susceptibility of meadow voles to sheep scrapie as a potential rodent model for use in scrapie research and to begin to characterize the spectrum of prion disease phenotypes in meadow voles.

Outbred meadow voles were raised from wild-caught progenitors in a colony maintained at the US Geological Survey (USGS) National Wildlife Health Center (NWHC) in Madison, Wisconsin, USA. Animals were handled in accordance with NWHC Animal Care and Use Committee protocol #EP080715-A2. Six 4-to-8-week-old meadow voles were inoculated intracerebrally with 20 μL of a 10% weight/volume (w/v) brain homogenate in phosphate-buffered saline (PBS) made from 1 of 2 scrapie-infected brains of clinically positive, homozygous Prnp V¹³⁶R¹⁵⁴Q¹⁷¹ genotype (GenBank accession number CAA04276.1) sheep maintained at the University of Idaho Caine Veterinary Teaching Center (Caldwell, Idaho, USA). Meadow voles were monitored daily for disease progression and euthanized when clinical disease signs consistent with a TSE developed. The prion protein coding sequences from all meadow voles used in this study were identical and contained glycine at residue 64 (GenBank accession number ACV85681.1).

Meadow vole brains were harvested immediately postmortem and divided sagittally, with half prepared for evaluation by immunoblot and the other half fixed in 10% neutral-buffered formalin for histopathology and immunohistochemistry (IHC). All challenged animals were tested for the presence of PrP^Sc in brain tissue by immunoblotting and IHC. We have previously described our NuPAGE and immunoblotting procedures using prion protein monoclonal antibodies (mAbs) SAF 83 and BAR 224 (Cayman Chemical, Ann Arbor, Michigan, USA) (12). Densitometry was carried out using VisionWorks LS software version 6.6a on an EC3 imaging system (UVP Bioimaging Systems, Upland, California, USA). Compositional analysis of glycoform ratios was done using CoDaPack 2.01 software (16). For IHC, formalin-fixed vole brain hemispheres were embedded in paraffin, sectioned to 5 μm, and mounted onto glass slides. Hematoxylin and eosin (H&E)-stained tissues were prepared using Harris hematoxylin (Thermo Fisher Scientific, Rockford, Illinois, USA) and eosin Y stain (Ricca Chemical, Arlington, Texas, USA). Immunostaining for PrP^Sc followed a previously described decloaking and staining procedure (17), using mAb SAF 83 (1:5000) and NovaRed horseradish peroxidase detection kit (Vector Laboratories, Burlingame, California, USA).

Spongiosis was assessed in 9 neuroanatomic regions in H&E-stained medial sagittal brain sections in TSE-affected meadow voles, as well as a 411-day-old control vole. The vacuolization severity scale ranged from 0 to 5 and tissues were blindly scored by 2 investigators (18). Characteristics of PrP^Sc deposition were recorded by identifying the localization and pattern of PrP^Sc staining within each of the same 9 brain regions and scoring intensity of PrP^Sc immunostaining using a 5-point scale: 0 — no staining; 1 — weak staining; 2 — moderate staining; 3 — strong staining; and 4 — intense staining.

Meadow voles were intracerebrally challenged with either sheep scrapie isolate 1 (voles A, B, and C) or isolate 2 (voles D, E, and F). Only 1 of the 3 voles challenged with scrapie isolate 1 (vole A) developed clinical signs of a TSE and was euthanized 593 d post-inoculation (dpi). Voles B and C lived for 571 and 653 dpi, respectively, and did not display clinical signs. Of the animals challenged with scrapie isolate 2, voles D and E showed clinical signs of disease and were euthanized at 354 and 483 dpi, respectively. Vole F suffered an intercurrent death at 304 dpi, a timepoint that may not have been sufficient to observe clinical signs of TSE onset. Clinical disease manifestations were unremarkable for a rodent TSE and included initial ataxia, followed by a state of excessive lethargy when animals were nonresponsive to physical stimuli and lacked motivation to burrow and acquire food or water. No notable differences in clinical disease signs were observed for sick animals challenged with either scrapie isolate.

Proteinase-K-resistant prion protein (PrP^res) is a marker for PrP^Sc and TSE infection (2). We found that only animals that displayed clinical signs were PrP^res-positive (Figure 1A). Two asparagine-linked sites in the prion protein (PrP) molecule are variably glycosylated. Proteinase-K-resistant prion protein (PrP^res) can be found in un-, mono-, or diglycosylated forms (19). The ratio among these 3 forms of PrP^res corresponds to factors controlled by both host and TSE agent. We observed that meadow vole-passaged scrapie maintained the general glycosylation site occupancy ratio (“glycoform profile”) of the ovine inocula: diglycosylated > monoglycosylated > unglycosylated (Figure 1). On average, the proportion of diglycosylated PrP was reduced and unglycosylated PrP appeared to be increased in the vole-passaged scrapie samples, relative to the ovine inocula. The monoglycosylated form remained similar across all samples. None of these differences between the glycoform profiles of vole-passaged scrapie and the ovine inocula, however, was found to be statistically significant at a confidence level of 95% (Figure 1B). We did find that the glycoform profile for meadow vole-passaged scrapie differed significantly from those we previously reported for clinically ill meadow voles that had been challenged with chronic wasting disease (CWD) from white-tailed deer (Figure 1B) (12). Considering the limited data set with which we are working, our results suggest that passage of scrapie in meadow voles generally maintains the glycoform characteristics of the original sheep inocula and PrP^res glycoform ratios in meadow vole-passaged scrapie differ from those in meadow vole-passaged CWD.

Figure 1.

Meadow vole-passaged scrapie retains biochemical profile of ovine inocula. A — Immunoblotting confirms the presence of proteinase K-resistant prion protein (PrP^res) in the brains of sheep used for vole challenges and 3 voles (A, D, and E) of 6 challenged meadow voles. B — Triplots of glyocoform ratios for sheep scrapie, cervid chronic wasting disease (CWD), and their meadow vole-passaged counterparts. Colored dots represent geometric centers for the groups identified in the key. Colored outlines represent 95% confidence intervals for each group. Data on CWD are from Heisey et al (12).

Quantitation of spongiosis in specific neuroanatomic areas of scrapie-challenged meadow voles had regions of both similarity and differences (Figure 2). In all 3 animals, pronounced vacuolization was consistently found in the medulla (region 1), cerebellum (region 2), and thalamus (region 5). The degree of spongiosis was somewhat variable in the superior colliculus (region 3) and hippocampus (region 6).

Figure 2.

Severity of vacuolization of brains from scrapie-challenged meadow voles. Neuroanatomic brain regions were scored for severity of vacuolation in clinically diseased meadow voles. Regions were: 1. dorsal medulla; 2. cerebellar cortex; 3. superior colliculus; 4. hypothalamus; 5. thalamus; 6. hippocampus; 7. septal nuclei of the paraterminal body; 8. cerebral cortex anterior to 7; and 9. cerebral cortex dorsal to the corpus callosum. Each line depicts the vacuolization pattern for a single animal ± standard deviations of lesion severity scores reported by 2 independent scorers.

To further characterize scrapie-induced neuropathology in meadow voles, we investigated PrP^Sc deposition in the brains of infected animals. The distribution and amount of PrP^Sc staining are summarized in Table I. A consistent granular pattern of staining that sometimes coalesced into focal aggregates was observed throughout multiple gray matter regions of brains of all 3 TSE-positive voles. Affected areas include the dorsal medulla, cerebellar cortex, superior colliculus, thalamus, septal nuclei, and cerebral cortex.

Table I.

Average deposition patterns of prion protein (PrP) for scrapiechallenged meadow voles

Brain region	PrP deposition score	PrP deposition pattern
Brainstem
Dorsal medulla	+	Granular
Cerebellum
Molecular layer	−	−
Purkinje layer	−	−
Granular tracts	+	Focal granular deposits
White matter tracts	+	Granular, diffuse
Superior colliculus	+	Granular
Hypothalamus	−	−
Thalamus	+++	Dense, granular
Hippocampus	$\frac{\text{vole\hspace{0.17em}A}/\text{vole\hspace{0.17em}D}/\text{vole\hspace{0.17em}E}}{+/-/++}$	Variable (Figures 3M, N, O)
Septal nuclei	++	Granular
Cerebral cortex	++	Granular; small aggregates

− no staining; + weakly positive; ++ moderately positive; +++ strongly positive; ++++ intensely positive.

Representative images of each brain region are displayed in Figure 3. The thalamus consistently displayed the heaviest staining of all brain regions observed (Figure 3H), while staining was conspicuously absent from the hypothalamus (Figure 3G). Although the intensity of PrP^Sc staining in these regions appeared to roughly correspond with the density of spongiosis, that relationship did not hold true for all brain regions examined, e.g., cerebellum and dorsal medulla for vole E. Unlike the other brain regions examined, the hippocampus displayed a unique PrP^Sc immunostaining pattern for each of the 3 TSE-positive voles, with a multifocal stellate staining throughout multiple layers of the hippocampus of vole A (Figure 3M), no hippocampal staining in vole D (Figure 3N), and dense, granular staining largely restricted to the strata oriens, pyrimidale, and radiatum of the CA1, 2, and 3 regions of the hippocampus and the molecular and granular layers of the dentate gyrus of vole E (Figure 3O).

Figure 3.

Immunohistochemical detection of disease-associated prion protein (PrP^Sc) in meadow vole brains. Sagittal sections of brains from clinically diseased and uninfected meadow voles were immunostained for PrP^Sc using mAb SAF 83. Representative images of the following brain regions for infected and control voles are shown: (A) infected and (D) uninfected dorsal medulla; (B) infected and (E) uninfected cerebellum (arrow indicates location of inset image from the granular layer); (C) infected and (F) uninfected superior colliculus; (G) infected and (J) uninfected hypothalamus; (H) infected and (K) uninfected thalamus; and (I) infected and (L) uninfected cerebral cortex. Marked differences in PrP^Sc deposition were observed in the hippocampus of individual scrapie-challenged voles: vole A (M), vole D (N), and vole E (O). Uninfected voles showed no PrP^Sc staining in the hippocampus (not shown). Scale bars = 500 μm and are equally applicable to both images in each infected/uninfected set.

Our data show that after the challenge with scrapie from homozygous Prnp V¹³⁶R¹⁵⁴Q¹⁷¹ genotype sheep, TSE developed in a subset of meadow voles after lengthy survival times, which indicates the presence of an interspecies transmission barrier. Glycoform analysis revealed that passage of scrapie in meadow voles retained the overall biochemical profile of the ovine inocula (Figure 1). The PrP^res glycosylation site occupancy ratios for white-tailed deer CWD passaged in meadow voles differs significantly from those for scrapie in meadow voles. (Figure 1B) (12). The glycoform profile data, in combination with differences in attack rate and incubation period between CWD and scrapie passaged in meadow voles, suggest that each of these TSEs produces a different strain in meadow voles. The meadow voles used in this study possessed identical PrP amino acid sequences, which indicates that observed differences among scrapie-challenged animals were not due to host-encoded PrP polymorphisms. Meadow voles can possess a naturally occurring glycine to serine polymorphism at amino acid position 64 (12). This polymorphism does not appear to influence survival in meadow voles challenged with white-tailed deer CWD (12) and it will be interesting to investigate its role in the susceptibility of meadow voles to scrapie in the future.

Immunostaining of PrP^Sc in scrapie-positive meadow voles indicated a consistent granular deposition pattern across nearly all brain regions examined, with some animal-to-animal variation (Table I, Figure 3). The hippocampus, however, was an exception in that it displayed completely different PrP^Sc staining patterns for all 3 TSE-positive meadow voles. In contrast to these results, identical brain regions were reported to contain similar amounts of PrP^Sc immunolabeling in scrapie-infected bank voles (9). Differences in the staining in our study may reflect variation in the strains affecting each vole or the outbred nature of both meadow voles and bank voles used in these studies.

Overall, the distribution of PrP^Sc staining in scrapie-positive meadow voles was remarkably similar to that of published results for scrapie-challenged bank voles despite the fact that our 2 studies used scrapie isolates of different sheep genotypes. The similarity in our results could suggest that both species of voles select for similarly compatible strains or conformers of PrP^Sc that may be commonly found in scrapie-infected sheep. Also interesting in light of our results in meadow voles are completely dissimilar PrP^Sc distribution and staining patterns for scrapie-challenged ovinized Tg338 mice, in which labeling is characterized by plaques, intraneuronal PrP^Sc deposition, and involvement of the hypothalamus (20).

In conclusion, meadow voles displayed much longer survival times and lower attack rates than bank voles when challenged with sheep scrapie. Some of this may be due to differences in host genetics or scrapie isolates. Our results suggest that meadow voles may not be as susceptible to sheep scrapie as bank voles or transgenic ovinized mice, but that meadow voles can have value for characterizing scrapie isolates.