Previous Article | Next Article ![]()
Journal of Virology, July 2004, p. 6792-6798, Vol. 78, No. 13
0022-538X/04/$08.00+0 DOI: 10.1128/JVI.78.13.6792-6798.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Jason C. Bartz,1 Anthony E. Kincaid,2 and Richard A. Bessen3*
Department of Medical Microbiology and Immunology,1 Department of Physical Therapy, Creighton University, Omaha, Nebraska 68178,2 Department of Veterinary Molecular Biology, Montana State University Bozeman, Montana 597173
Received 29 September 2003/ Accepted 3 March 2004
|
|
|---|
|
|
|---|
Modest levels of prion agent replication in skeletal muscle have been reported in a few studies following intracerebral or extraneural inoculation of the prion agent. Prion infectivity in skeletal muscle was first demonstrated in mink with transmissible mink encephalopathy (TME); the amount of infectious agent in skeletal muscle was 10,000-fold less than the amount found in brain (26). Prion infectivity or accumulation of the abnormal form of the prion protein, PrPSc, in skeletal muscle has more recently been described in experimental prion infection of rodents and in sporadic Creutzfeldt-Jakob disease (CJD) of humans (14), but the cellular location of agent deposition and/or replication in muscle has not been identified (3, 6, 40). In natural prion diseases, PrPSc is found in the peripheral nervous system of scrapie-infected sheep and goats, and it has been proposed that it is the peripheral nerves which transverse skeletal muscle that are the source of prion infectivity in muscle (16, 17, 19, 20). Attempts to measure the prion agent in muscle of goats infected with the scrapie agent (18, 20) and cattle infected with the bovine spongiform encephalopathy (BSE) agent (10, 42) have not detected prion infectivity. However, a murine bioassay was used to measure prion infectivity in these studies, and this assay cannot detect below 104.1 lethal median doses (LD50) of the BSE agent (10, 42). Therefore, low levels of prion agent that could be present in muscle tissue of domestic livestock would not be found using this bioassay. Routine testing for PrPSc in brains from cattle in Europe and in lymph nodes from deer and elk in the United States has been implemented to reduce human exposure to BSE and chronic wasting disease by removing infected animals from the food supply. However, human consumption of skeletal muscle from a prion-infected animal is currently considered a low risk for contracting a prion disease.
In this study, we report that in hamsters infected with the TME agent, by either an intralingual or intracerebral route, PrPSc can accumulate in skeletal muscle cells of the tongue. Localization of PrPSc to the neuromuscular junction (NMJ) indicates that this synapse can be a site for prion agent replication and may provide a path by which the prion agent spreads between skeletal muscle cells and axon terminals. PrPSc was also found in the lingual papillae of the tongue, specifically in nerve fibers of the lamina propria and in the stratified squamous epithelium of the tongue surface. These findings have public health implications and suggest that (i) ingestion of prion-infected tongue could be a source of prion exposure for humans, since livestock tongue is not included in the specified risk material; and (ii) normal shedding of the outer layer of the tongue epithelium can release the prion agent into saliva, which may be a source for prion agent transmission in livestock and cervids.
|
|
|---|
Tissue preparation for PrPSc Western blotting. PrPSc was enriched from tongue prior to Western blotting by tissue extraction in detergent and differential ultracentrifugation as previously described (3, 4). PrPSc-enriched samples were digested with proteinase K and resuspended in sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) sample loading buffer. Five-percent brain homogenates were also digested with proteinase K and prepared for Western blotting as previously described (4). SDS-PAGE and Western blot analysis were performed using monoclonal antibody 3F4 hybridoma supernatant (22) (a gift of Victoria Lawson, NIH Rocky Mountain Labs, Hamilton, Mont.) as previously described. Quantification of PrPSc signal in Western blot assays was performed using ImageQuant software as previously described (4).
PrPSc immunohistochemistry, immunofluorescence, and confocal microscopy. Immunostaining for PrPSc in tongue was performed as previously described (3). Briefly, animals were intracardially perfused with paraformaldehye-lysine-periodate fixative followed by postfixation in paraformaldehye-lysine-periodate for 5 h. Paraffin-embedded tissue sections (4 µm) were subjected to antigen retrieval by pretreatment with formic acid for 20 min. A minimum of 10 tissue sections per animal was examined for each antibody staining procedure. PrPSc was detected by incubation with monoclonal 3F4 hybridoma supernatant (1:600 dilution) and the ABC-horseradish peroxidase Elite staining method (Vector Laboratories, Burlingame, Calif.). For PrPSc staining in the double immunofluorescence procedure, the ABC-horseradish peroxidase incubation step in the ABC method was replaced by incubation with an Alexa Fluor 488 streptavidin conjugate (Molecular Probes, Portland, Oreg.) at a 1:400 dilution. PrPSc immunofluorescence staining was combined with rabbit polyclonal antibody staining for either desmin (1:50 dilution; DAKO Corp., Carpinteria, Calif.), synaptophysin (1:50 dilution; DAKO Corp.), S100 (1:800 dilution; DAKO Corp.), cytokeratin (1:800 dilution; DAKO Corp.), or calcitonin growth-related peptide (CGRP; 1:125 dilution; Peninsula Labs, Inc., San Carlos, Calif.). This latter group of primary antibodies was visualized by incubation with anti-rabbit antibody conjugated to Alexa Fluor 568 (Molecular Probes). The nuclear counterstain TO-PRO3 (Molecular Probes) was applied to some tissue sections at a concentration of 0.25 µM for 10 min. Tissue sections were mounted with Vectashield Antifade (Vector Laboratories). Images were visualized with epifluorescence using a Bio-Rad Radiance 2000 confocal system attached to a Nikon Eclipse 800 microscope.
|
|
|---|
![]() View larger version (63K): [in a new window] |
FIG. 1. PrPSc deposition in tongue and brain following inoculation of the HY TME agent. Hamsters were inoculated with the TME agent or a normal brain homogenate (mock) by either the intratongue or intracerebral route. Tongue and brain tissues were collected from animals that were sacrificed after the onset of clinical disease as described in the text. PrPSc in brain homogenates or PrPSc-enriched preparations from tongue digested with proteinase K were prepared as described in Materials and Methods and analyzed by SDS-PAGE and PrP Western blotting. The amount of tissue analyzed in each lane is indicated in milligram tissue equivalents (mg eq), and the molecular masses of polypeptides in kilodaltons are indicated to the left of the panel.
|
![]() View larger version (66K): [in a new window] |
FIG. 2. PrPSc deposition in tongue following inoculation of the HY TME agent. Hamsters were inoculated with a normal brain homogenate (A) or the TME agent (B and C) by either the intratongue (A and B) or intracerebral (C) route. Tongues were collected from animals that were sacrificed after the onset of clinical disease and prepared for immunohistochemistry as described in Materials and Methods. PrPSc immunohistochemistry illustrated PrPSc (red punctate staining) in skeletal muscle cells (B) and the lamina propria (C) but an absence of immunostaining in mock-infected tongue (A). The tissue was counterstained with hematoxylin. The lamina propria (LP) and stratified squamous epithelium (SSE) of the tongue are indicated. Bar, 10 µm.
|
![]() View larger version (61K): [in a new window] |
FIG. 3. Confocal microscopy of PrPSc in skeletal muscle cells and axons. Hamsters were inoculated in the lingual muscles with the HY TME agent. At the onset of clinical symptoms, the tongue was prepared for immunofluorescence staining and confocal microscopy as described in Materials and Methods. Following immunofluorescence staining for identification of PrPSc with Alexa Fluor 488 (A, D, and G), tissue sections were also stained for either desmin (B), S100 (E), or synaptophysin (H) with goat anti-rabbit immunoglobulin conjugated to Alexa Fluor 568. To investigate the colocalization of PrPSc with desmin (C), S100 (F), and synaptophysin (I), green and red fluorescent images were merged. Bar, 10 µm. The insets at the lower left corner (C, F, and I) are two- to threefold enlargements of smaller boxes in each of the corresponding panels.
|
The spatial relationship between PrPSc and the NMJ was investigated following intratongue inoculation of the HY TME agent, since PrPSc was found in both skeletal muscle cells and axons in the tongue. An antibody to synpatophysin, a marker for synaptic vesicles in the axon terminal, was used for identification of NMJs in hamster tongue from clinically ill hamsters at 95 days postinfection. Sixty-four percent of synaptophysin-positive skeletal muscle cells were also PrPSc positive; colocalization of synaptophysin and PrPSc was found in 25% of these cells (Fig. 3G to I). In the example illustrated (Fig. 3G to I), PrPSc staining was localized to the NMJ, indicating that this is one potential site for PrPSc accumulation in skeletal muscle cells and/or axon terminals.
The majority of PrPSc staining in the lingual mucosa was located in the lamina propria following intracerebral inoculation of HY TME, especially within the connective tissue core of the lingual papillae (Fig. 2C). To investigate the spatial relationship between PrPSc and nerve fibers in the lamina propria, we examined PrPSc and CGRP staining in the tongue at 75 to 85 days postinfection following intracerebral inoculation of the HY TME agent. CGRP staining in the lamina propria was used to identify a subset of sensory nerves, since not all sensory fibers are CGRP positive. PrPSc immunostaining was identified in 35% of CGRP-positive papillae, but colocalization of CGRP and PrPSc was found in only 5% of CGRP-positive papillae (Fig. 4A to C). In 4% of the lingual papillae, PrPSc staining was also found in the stratified squamous epithelium of the tongue mucosa following intracerebral inoculation (Fig. 4D). In these examples, the PrPSc and cytokeratin staining were in close proximity in the papillary epithelium (Fig. 4D to F); cytokeratin immunohistochemistry was used as a marker for the stratified squamous epithelial cells of the tongue. However, there was no direct overlap of the two staining patterns (Fig. 4F, inset). A greater amount of PrPSc staining was found associated with tongue papillae following intracerebral inoculation compared to inoculation of the HY TME agent into the lingual muscles. No PrPSc-specific staining was observed in the tongue of mock-infected hamsters (Fig. 4G to I).
![]() View larger version (74K): [in a new window] |
FIG. 4. Confocal microscopy of PrPSc in sensory nerve fibers and the stratified squamous epithelium. Hamsters were inoculated in the cerebrum with the HY TME agent (A to F) or a normal brain homogenate (G to I) (mock infected). At the onset of clinical symptoms, the tongue was prepared for confocal microscopy as described in Materials and Methods. Following immunofluorescence staining for PrPSc with Alexa Fluor 488 (A, D, and G), tissue sections were stained for either CGRP (B) or cytokeratin (E and H) with goat anti-rabbit immunoglobulin conjugated to Alexa Fluor 568. The stain TO-PRO3 was used to visualize nuclei under blue fluorescence. To investigate the colocalization of PrPSc with CGRP (C) and cytokeratin (F and I), green and red fluorescent images were merged. Bar, 10 µm. The insets at the bottom (C and F) are two- to threefold enlargements of smaller boxes in each of the corresponding panels.
|
|
|
|---|
Prior studies have demonstrated PrPSc accumulation in the tongue (3, 40) or prion infectivity in skeletal muscle (6, 26, 40), but PrPSc staining by immunomorphological methods has only been reported in nerve fibers that transverse skeletal muscle (3). In the present study, PrPSc deposition was found in nerve fibers and skeletal muscle cells of the tongue. The normal isoform of the prion protein, PrPC, has been demonstrated to spread within nerve fibers by axonal transport (5) and is located in synapses, including subsynaptic areas in both the presynaptic and postsynaptic cells of the NMJ (2, 15). Since PrPC is required for PrPSc formation (7), we propose that the NMJ is a site for PrPSc formation and/or PrPSc spread between skeletal muscle cells and axon terminals. In this case, PrPSc spread to, and accumulation in, skeletal muscle cells in the tongue following intracerebral inoculation could be due to TME agent replication in the brain and subsequent anterograde transport within the hypoglossal nerve, which provides motor innervation to the lingual muscles. PrPSc may spread across the NMJ and enter skeletal muscle cells, where additional TME agent replication can occur, since PrPC is also expressed in skeletal muscle cells (15, 29). This hypothesis is supported by studies that describe (i) the spread of the prion agent in both the anterograde and retrograde directions between synaptically linked nerve cell groups in the central nervous system (4, 12, 13, 27), (ii) the retrograde transport of the HY TME agent from the tongue to the brain in the hypoglossal nerve following intratongue inoculation (3), and (iii) prion infectivity in skeletal muscle of transgenic mice that express PrPC in myocytes (6). Although the NMJ is morphologically distinct from synapses in the central nervous system, our findings are consistent with this peripheral synapse serving as a pathway for intercellular spread of the prion agent between motor axon terminals and muscle cells.
PrPSc deposition in the lamina propria below the mucosal epithelium was associated with sensory nerve fibers; however, we cannot exclude the possibility that other types of cells or axons are infected with the TME agent, since only a portion of PrPSc staining colocalized with CGRP staining. Given that CGRP staining only labels a subset of sensory fibers, it is possible that other types of sensory fibers located in the lamina propria also contain PrPSc. Based on the location of PrPSc staining and morphology of the adjacent tissue, there was no evidence that PrPSc was present in the lingual tonsils, which are also a potential target for TME agent replication. However, from these studies we cannot exclude that the lingual tonsils are a site of TME agent replication. PrPSc staining in the lamina propria was often found in the connective tissue in the core of papillae; this area has a high concentration of sensory fibers that innervate mechanoreceptors and taste buds in the tongue (30, 35). PrPSc staining in the lamina propria and tongue epithelium was more pronounced following intracerebral inoculation compared to intralingual inoculation. This finding could be due to TME agent replication in the brain and subsequent transganglionic transport within the tongue-associated sensory cranial nerves. PrPSc that was localized to cytokeratin-positive cells could also be explained by TME infection of sensory nerve fibers that project into the epithelial cell layers of the tongue (11, 25, 32, 38). Alternatively, prion agent could spread to basal keratinocytes from sensory nerve endings and lead to PrPSc formation, since keratinocytes also express PrPC (33, 39). This proposed pathway for prion agent spread to the mucosa along sensory fibers is analogous to the previous reports of spread of the CJD agent from the brain to the cornea and olfactory mucosa in sporadic CJD (24, 45; P. Duffy, J. Wolf, G. Collins, A. G. DeVoe, B. Streeten, and D. Cowen, Letter, N. Engl. J. Med. 290:692-693, 1974). The low level or absence of the prion agent in the lymphoreticular system in sporadic CJD (14, 21) suggests that the optic and olfactory cranial nerves, respectively, transport the CJD agent from the brain to these mucosal or exposed surfaces. Ultrastructural studies will be necessary to further define the location of PrPSc in the lamina propria and sensory nervous system in the tongue and its spatial relationship to epithelial cells of the tongue.
In natural prion diseases, prion agent transport from the brain to the tongue could have implications for human food safety and animal prion transmission. In sheep with scrapie, cattle with BSE, and deer with chronic wasting disease, the brain stem regions containing the tongue-associated cranial nerve nuclei have been reported to be targets for prion infection. In one or more of these diseases, PrPSc has been found in the hypoglossal nucleus (9, 37, 43), the nucleus of the solitary tract (9, 34, 37, 43), and several of the trigeminal nuclei (9, 34, 37, 43, 44). The ability of the prion agent to establish infection in brain nuclei that contribute to, or synapse with, the four cranial nerves that innervate the tongue suggests that this could be a pathway for centrifugal prion agent spread. The tongue is one of the most densely innervated extraneural tissues in the human body, with a very high concentration of both motor and sensory axons (28, 30, 35, 36, 41). Anterograde prion transport within the hypoglossal nerve would be expected to result in prion infection of the lingual muscles, while transganglionic transport within the trigeminal, facial, and glossopharyngeal nerves could result in prion spread to the sensory fibers in the tongue. It is noteworthy that prion infectivity has been found in the trigeminal ganglia of sheep with scrapie (19) and cattle with BSE (42), indicating that centrifugal spread of these prion agents within the trigeminal nerve can occur and could result in spread to peripheral sites including the tongue. Since livestock and cervid tongues are not banned for human consumption, it is possible that humans could be exposed to animal prion diseases by ingestion of prion-infected tongue. Attempts to identify prion infectivity in bovine tongues have not been successful. However, these studies have used an infectivity bioassay that cannot detect below 104.1 LD50 of BSE prions, which may not detect low levels of the prion agent (10, 42). Our findings also have implications for intraspecies transmission of animal prion diseases. Establishment of prion infection in the stratified squamous epithelial cells of the tongue, as suggested by our findings, could result in sloughing of the prion agent into the saliva, since this tissue is undergoing continual cell turnover. In this case, animal behaviors resulting in the exchange of saliva between hosts may play a role in prion transmission.
Special thanks are given to Maria Christensen and Emily Hansen for excellent technical assistance.
Present address: Department of Anatomy and Cell Biology, University of Kansas Medical Center, Kansas City, KS 66160. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»