SEARCH

SEARCH BY CITATION

Keywords:

  • copper;
  • enteric glia;
  • enteric nervous system;
  • myenteric plexus;
  • nitric oxide

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Funding
  8. Author contributions
  9. Disclosures
  10. References
  11. Supporting Information

Background  Cellular prion protein (PrPC) is expressed in the enteric nervous system (ENS), however, its physiological role has not been identified. Studies suggest that PrPC can function as a metal-binding protein, as absence of the protein has been linked to altered copper metabolism and atypical synaptic activity. Because copper is known to modulate smooth muscle relaxation, we tested the hypothesis that PrPC deficiency would alter intestinal contractility.

Methods  We examined electrically evoked ileal contractility in Prnp−/− or wild type littermate mice and the effects of copper or copper chelation. PrPC expression was studied in whole mount ileal preparations of mice and guinea pigs by immunohistochemistry.

Key Results  Relative to wild type mice, ileal tissues of Prnp−/− mice exhibited reduced electrical field stimulation (EFS)-evoked contractility. Furthermore, EFS-induced relaxation, as a percentage of that induced by a nitric oxide donor, was enhanced. Addition of a copper donor to the organ bath increased, whereas the addition of a copper chelator inhibited, nitric oxide donor-induced ileal relaxation in Prnp−/− mice. PrPC was expressed on nerve fibers or terminals, and some cell bodies in the myenteric and submucosal plexuses of wild type mice. PrPC colocalized with a neuron-specific ectonucleotidase, nucleoside triphosphate diphosphohydrolase 3 (NTPDase3), but to only a limited extent with GFAP, a marker of enteric glia. Guinea pigs expressed PrPC in nerve fibers or terminals and enteric glia in the myenteric and submucosal plexuses.

Conclusions & Inferences  Our findings suggest that PrPC, which is abundant in the ENS, has a role in the regulation of ileal contractility.

The misfolding of cellular prion protein (PrPC) into its pathological isoform leads to several transmissible neurodegenerative disorders that in humans include Creutzfeldt–Jakob disease, Gerstmann–Straussler–Scheinker syndrome, kuru, and fatal familial insomnia.1 PrPC, a glycoprotein that is attached to the membrane surface of cells via a glycosylphosphatidyl anchor,2 is highly conserved within vertebrate species3 and widely expressed at high levels in the central nervous system (CNS).4 There is evidence that PrPC modulates intracellular signaling pathways,5–7 ion channels,8 glutamate-mediated excitoxicity,9 and that it has neuroprotective activity.8,10 PrPC has been localized to synapses where it is important for synaptic development and function.11,12 Outside of the brain, recent functional studies have shown PrPC to be an essential modulator of redox homeostasis and contractile function in skeletal muscle.7,13 In the gut, there is strong evidence that PrPC is expressed by components of the enteric nervous system (ENS),14–16 gut-associated lymphoid tissue,16,17 and on the epithelium.18 The localization of PrPC remains somewhat controversial with some articles showing it is expressed on enteric glia19 and others enteric neurons.15,20

The prion gene (Prnp) encodes a 253 amino acid protein consisting of two distinct domains; a C-terminus, that contains three alpha helices, two of which are linked by a disulfide bridge, and an unstructured N-terminus, that contains an octamer repeat region capable of binding up to five copper atoms.21,22 Copper binding to PrPC has been shown to cause endocytosis of, PrPC either by clathrin-coated pits,23 or by caveolae-like membranous domains.24 Using synchrotron-based x-ray fluorescence imaging, we recently showed that Prnp−/− mice have reduced brain copper levels relative to their normal littermates, whereas mice that overexpressed PrPC, had elevated copper levels.25 This is likely of importance as there is evidence that PrPC protects voltage-gated calcium channels from copper interference26 and electrophysiological studies have shown that PrPC modulates NMDA receptor activity in a copper-dependent manner.27,28 Moreover, PrPC handling of synaptic copper has been shown to affect the function and transmission through purinergic P2X4 receptors, by preventing copper-induced inhibition.29 The PrPC-copper relationship has not been examined in the gut, but it is possible that as a result of a loss or deficiency of PrPC, there could be a physiological consequence as a result of either increased or decreased amounts of free copper.

The binding of copper to PrPC (in vivo and in vitro), and the resultant endocytosis of the copper–PrPC complex, strongly suggests that PrPC is involved in copper homeostasis.26 As there is a relationship between PrPC and copper, and as it has been observed that copper can enhance nitric oxide (NO) donor-evoked aortic relaxation,30 we hypothesized that mice that lacked PrPC would have altered intestinal contractility and relaxation. In this study we investigated a functional role of PrPC in ileal contractility using Prnp−/− mice. As the localization of PrPC in the ENS is controversial,14,15,19,31 we also attempted to further clarify this issue using immunohistochemistry.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Funding
  8. Author contributions
  9. Disclosures
  10. References
  11. Supporting Information

Ethical approval

The animal studies were conducted in accordance with the guidelines established by the Canadian Council of Animal Care. The procedures and animal care protocol (#M09048) were approved by the Health Sciences Animal Care Committee at the University of Calgary.

Animals

Mice with a targeted disruption of the prion gene (Prnp−/−) of the Zurich I strain32 (backcrossed for >10 generations onto the C57BL/6 background) and their Prnp+/+ littermates were used in these experiments. Hemizygous Tga20 mice, which have ∼seven-fold overexpression of PrPC (kindly provided by Dr. S. Lindquist, MIT) were also used.33 The mice were matched for age, sex (male), body weight (22–25 g), and allowed ad libitum access to standard laboratory chow and water. All mice were maintained in a double barrier unit in a room that had controlled temperature (22 ± 1 °C), humidity (65%–70%), and light cycle (12 h light/12 h dark). Male albino guinea pigs (350–600 g, Charles River Laboratories, Montreal, QC, Canada) housed under identical conditions were also used for these studies. Guinea pigs were euthanized by isoflurane overdose and exsanguination; mice were euthanized by CO2 and cervical dislocation. Animal numbers for each study are provided in the Figure legends.

Ileal contractility and relaxation studies

Muscle contractility and relaxation studies were performed as described previously.34 In brief, mice were killed by cervical dislocation and then segments of distal ileum were removed, submerged in ice-cold oxygenated Krebs’ solution and gently flushed to remove luminal contents. Four 1 cm segments from each mouse were then ligated at each end with 5–0 silk and suspended longitudinally in organ baths (Krebs’ solution, 37 °C, aerated with 95% O2, 5% CO2). These preparations were positioned between a pair of parallel electrodes that were attached to an isometric force transducer (Harvard Apparatus, Kent, UK). The mechanical activity of the muscle was enhanced using a transducer amplifier relayed to a bioelectric amplifier and then recorded (model 8811A; Hewlett-Packard, Mississauga, ON, Canada). The samples were allowed to equilibrate for 20 min under a basal tension of 0.5 g. Bethanechol (100 μ mol L−1; Sigma, St. Louis, MO, USA) was added and after 2 min, the organ baths were rinsed and tissues allowed to return to a steady-state resting level. To examine muscle contractility, ileal preparations were subjected to electrical field stimulation (EFS; 2, 4, 8, 16 Hz for 10 s, 60 V, 0.5 ms pulse duration, 2 min intervals) repeated three times for each experiment. The values presented are the average of the three repetitions of the EFS-generated contractility response shown as a percentage of the maximal bethanechol response, to normalize the results for the volume of tissue.

Atropine (1 μ mol L−1; Sigma) and guanethidine (5 μ mol L−1; Sigma) were then added to the organ bath to produce nonadrenergic noncholinergic (NANC) conditions. After 20 min, prostaglandin F2a (PGF2a, 1 μ mol L−1, Sigma) was added which provided a plateau of contraction after 5 min that was stable over the following 15 min. To examine muscle relaxations, ileal sections were subjected to EFS (2, 4, 8, 16 Hz for 10 s, 60 V, 0.5 ms pulse duration, 2 min intervals), repeated three times, then the NO donor sodium nitroprusside (SNP, 100 μ mol L−1; Sigma) was added. The values presented are the average of the three repetitions of the EFS-generated relaxation responses shown as a percentage of the maximal SNP response, to normalize the results.

Copper and copper chelation studies

Ileal tissue samples were harvested and prepared as described for the ileal contractility and relaxation studies. After equilibration, tissues were incubated with a copper donor (CuCl2, Sigma) or the copper chelator, diethylthiocarbamate (DEDCA, Sigma), both at 100 μ mol L−1 for 20 min followed by bethanechol (100 μ mol L−1). Once bethanechol-induced steady-state resting levels were obtained, SNP (1, 10, 100 μ mol L−1) was added sequentially. The amplitudes of SNP-induced relaxations were calculated based on the nadir relaxations and bethanechol induced steady-state levels and presented in milligrams.

Immunohistochemistry

Segments of the ileum from guinea pigs and Prnp+/+, Prnp−/− and Tga20 mice were removed and placed in phosphate-buffered saline plus azide (PBS) containing 1 μ mol L−1 nifedipine. Intestinal tissues were then opened along the mesenteric border, rinsed with PBS and processed for whole mount preparations. These tissues were stretched and pinned flat with the mucosal side up, fixed overnight in Zamboni’s fixative, and subsequently rinsed and stored at 4 °C in PBS. The longitudinal muscle-myenteric plexus and submucosa, containing the submucosal plexus were dissected and processed for immunohistochemistry. In brief, whole mount preparations were washed in PBS plus azide containing 0.1% Triton X–100 and incubated with diluted antiserum for 48 h at 4 °C. For a list of primary and secondary antibodies used in this study, see Table 1. For PrPC, two mouse monoclonal antibodies were used: SAF-83, a monoclonal antibody that recognizes the conformational epitope located within residues 126–164 and is specific for the mouse (and hamster) prion protein, and SAF-32, which recognizes a seven amino acid motif (QPHGGGW repeated four times; residues 59–89) in the octapeptide repeat region of the mouse prion protein. This antibody has a broader specificity as it also recognizes human, bovine, and ovine prion protein.35

Table 1.   Antibodies used for immunofluorescence
AntigenSourceDilutionDetectionDilution
  1. The source of all secondary antibodies was Jackson ImmunoResearch (West Grove, PA, USA).

SAF-83Cayman Chemical1 : 500Donkey anti-mouse CY31 : 100
SAF-83Cayman Chemical1 : 500Goat anti-mouse Alexa4881 : 200
SAF-32Cayman Chemical1 : 500Donkey anti-mouse CY31 : 100
PGP9.5Ultraclone1 : 2000Goat anti-rabbit FITC1 : 50
GFAPBiomedical Technologies1 : 250Goat anti-rabbit FITC1 : 50
SynpAbcam1 : 100Donkey anti-rabbit CY31 : 100
nNOSBD Biosciences1 : 250Goat anti-rabbit FITC1 : 50
NTPDase3Jean Sevigny (Laval)1 : 100Donkey anti-guinea pigCY31 : 100

We also colocalized PrPC with other cell types including enteric neurons, using PGP9.5, and enteric glia, using glial fibrillary acidic protein (GFAP). In mice, whole mount preparations of the myenteric plexus were also probed with an nNOS specific antibody.36 A neuron-specific ectonucleotidase, which is localized to the membrane of enteric neurons, nucleoside triphosphate diphosphohydrolases 3 (NTPDase3) was used to determine if PrPC localized to the cell membrane.37

To study the potential of synaptic localization, PrPC was colocalized with synaptophysin, a marker of synaptic vesicles.38 The source of all secondary antibodies was Jackson ImmunoResearch (West Grove, PA, USA). The tissues were mounted in bicarbonate-buffered glycerol at pH 8.6, and immunofluorescence was visualized with an Olympus FV1000 confocal microscope. An observer blinded to the genotype confirmed all tissues examined.

Specificity of the two PrPC antibodies used in these studies was confirmed by immunoblotting gut and brain homogenates from mice and guinea pigs (Fig. S1).

Statistical analysis

Results are expressed as mean ± SE. Comparisons between groups were made using either a one-way analysis of variance (anova) followed by a post-hoc Tukey’s test or a repeated-measures two-way anova followed by a Bonferroni post-test.

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Funding
  8. Author contributions
  9. Disclosures
  10. References
  11. Supporting Information

Ileal contractility and relaxation in Prnp−/− mice

Electrical field stimulation (EFS) caused frequency-dependent contractions in the ileum of Prnp+/+ and Prnp−/− mice (Fig. 1A). Contractions were abolished by tetrodotoxin (data not shown). The maximal responses to bethanechol (100 μ mol L−1) were not different between Prnp+/+ and Prnp−/− mice (data not shown). EFS-induced contractions, as a percentage of bethanechol-induced responses, were significantly reduced at 2 and 4 Hz in Prnp−/− relative to Prnp+/+ mice (Fig. 1A). We also found that EFS-induced relaxations, as a percentage of NO donor-induced relaxation responses, were larger at lower frequencies (2 and 4 Hz) in Prnp−/− mice (Fig. 1B).

image

Figure 1.  Ileal contractility and relaxation responses in Prnp+/+ and Prnp−/− mice. (A) Electrically evoked contractility responses were significantly reduced at 2 and 4 Hz in ileal tissues of the Prnp−/− mice compared to Prnp+/+ (*P < 0.05). (B) Under nonadrenergic noncholinergic conditions, EFS-evoked relaxation responses were larger at lower frequencies in the ileal tissues of Prnp−/− mice compared to Prnp+/+ (P < 0.05). n ≥ 15 for each group.

Download figure to PowerPoint

PrPC did not colocalize with nNOS neurons of the myenteric plexus

As we found that EFS-induced ileal relaxation responses were greater in Prnp−/− mice, we next examined whether PrPC was localized to nNOS immunoreactive neurons or with nNOS terminals in the myenteric plexus. We also wanted to see if there were differences in the number of nNOS-positive neurons between Prnp+/+ and Prnp−/− mice as this could have been responsible for the enhanced relaxation response we observed. PrPC was not detected on myenteric neurons of Prnp−/− mice, and in the myenteric plexus of Prnp+/+ mice, did not appear to colocalize with nNOS neurons; some minor degree of colocalization on nerve fibers or terminals was observed (Fig. 2). In addition, there were no significant differences in the number of nNOS-positive neurons when Prnp+/+ and Prnp−/− mice were compared (Prnp+/+, 5.6 ± 0.4 vs Prnp−/−, 5.3 ± 0.5 nNOS immunoreactive neurons/ganglia; 537 neurons counted in Prnp+/+, 508 in Prnp−/−, n = 12/group).

image

Figure 2.  Expression of PrPC and nNOS in the ileal myenteric plexus. PrPC did not appear to colocalize with nNOS-positive neurons (white asterisks, D–F) and was not detected in the myenteric plexus of Prnp−/− mice (A–C). Some colocalization on nerve fibers or terminals was observed (arrows). PrPC antibody: SAF-83. n ≥ 8; Scale bar: 25 μm.

Download figure to PowerPoint

Copper chelation abolished NO donor-induced ileal relaxations in Prnp−/− mice

We also examined relaxations induced by the NO donor SNP. These experiments showed that SNP induced concentration-dependent relaxation responses (1–100 μ mol L−1) in Prnp−/− and Prnp+/+ mice. NO donor-induced ileal relaxation was enhanced in the presence of 10 μ mol L−1 SNP in the tissues from Prnp−/− mice (P < 0.05 vs Prnp+/+, Fig. 3A,B). As copper reportedly has a role in NO-donor-induced relaxation responses,30,39 we next examined SNP-induced relaxation in response to the addition of a copper donor or a copper chelator in Prnp−/− and Prnp+/+ mice. These experiments demonstrated that exogenous copper, in the presence of a NO donor, increased ileal relaxations in Prnp−/− mice, whereas copper did not alter SNP-induced ileal relaxations in Prnp+/+ mice (Fig. 3A). In agreement, the addition of a copper chelator decreased NO donor-induced relaxation responses in Prnp−/− mice, but not in Prnp+/+ mice (Fig. 3B).

image

Figure 3.  The effect of copper and copper chelation on ileal relaxation. (A) The addition of a copper donor significantly enhanced NO donor-induced ileal relaxation responses in Prnp−/− mice when compared to those produced by Prnp+/+ mice. In addition, ileal relaxation in the presence of 10 μ mol L−1 SNP alone was greater in the Prnp−/− mice. (B) In the presence of the copper chelator DEDCA, NO donor-induced relaxation responses were reduced in the ileal tissues of Prnp−/− mice relative to Prnp+/+. Relaxation response units = mg force/mg dry tissue weight. [n ≥ 4; Cu, copper; DEDCA, copper chelator; ***P < 0.001 or **P < 0.01 vs all other groups; Φ< 0.05 for (WT + Cu) vs (Prnp−/− + Cu)].

Download figure to PowerPoint

PrPC distribution in neurons and enteric glia of the myenteric plexus

As PrPC is strongly expressed within the ENS, whole mount preparations of the myenteric plexus of Prnp+/+ and Prnp−/− mice were probed with SAF-83 and SAF-32 (PrPC), PGP9.5 (neurons), or GFAP (enteric glia). The pattern of labeling of the two prion protein antibodies was virtually identical, although SAF-83 had lower background. As expected, PrPC immunoreactivity was absent in the myenteric plexus of Prnp−/− mice (Fig. 4A,B). In Prnp+/+ mice, PrPC immunoreactivity was colocalized on neural elements (Fig. 4A) and to a minor degree, with enteric glial cells (Fig. 4B). Although the pattern of PrPC immunoreactivity was similar to the distribution of enteric glia, colocalization was very limited or absent (Fig. 4B). The prion protein colocalized with the neuronal marker PGP9.5 and appeared to be either on the cell surface or within neuronal processes (Fig. 4A). In some preparations, PrPC staining appeared smooth, whereas in others, punctate staining was evident.

image

Figure 4.  Ileal PrPC expression in the myenteric plexus. (A) Immunohistochemical localization of PrPC and enteric neurons. No PrPC was detected in the myenteric plexus of Prnp−/− mice (a–c). However, PrPC colocalized with the neuronal marker PGP9.5 in neural elements of the Prnp+/+ mice (arrows, d–f). (B) Immunohistochemical localization of PrPC with enteric glia. No PrPC was detected in the myenteric plexus of Prnp−/− mice (a–c). Colocalization of PrPC [using both SAF-83 (d–f) and SAF-32 (g–i) antibodies] with GFAP was sparse to absent in Prnp+/+ mice (arrows). n ≥ 12; Scale Bar: 25 μm.

Download figure to PowerPoint

To further clarify the distribution of PrPC, we labeled enteric neurons with the ectonucleotidase, NTPDase3, a neuron-specific cell surface marker, and we also performed colocalization studies in mice that overexpressed PrPC. There were no differences in the distribution of NTPDase3 when Prnp+/+, Prnp−/− and Tga20 tissues were compared (Fig. 5A). Interestingly, PrPC immunoreactivity was strongly colocalized with NTPDase3 in both wild type and Tga20 mice (Fig. 5A). We observed that the distribution of PrPC, and the colocalizations with PGP9.5, GFAP and synaptophysin in the Tga20 mice was similar to that in the Prnp+/+ mice (Fig. 5B).

image

Figure 5.  Colocalization of PrPC with NTPDase3. (A) Distribution and colocalization of NTPDase3 with PrPC in the myenteric plexus of Prnp−/− (a–c), Prnp+/+ (d–f) and in mice that overexpress PrPC (g–i). PrPC immunoreactivity strongly colocalized with NTPDase3 in both Prnp+/+ and Tga20 mice (arrows). (B) PrPC expression in the myenteric plexus of mice that overexpressed PrPC. PrPC distributed within these tissues colocalized with PGP9.5 (a–c), GFAP (d–f) and Synp (g–i) in a similar manner as observed for Prnp+/+ mice. PrPC antibody: SAF-83. n = 4; Scale Bar: 25 μm.

Download figure to PowerPoint

PrPC distribution in neurons and enteric glia of the submucosal plexus

We also examined PrPC expression in the submucosal plexus. As noted for the myenteric plexus, PrPC immunoreactivity was absent in the submucosal plexus of Prnp−/− mice (Fig. S2A,B), while in Prnp+/+ mice, we observed that PrPC colocalized with the neuronal marker PGP9.5 (Fig. S2A). As in the myenteric plexus, PrPC immunoreactivity was found to be on either the cell surface or within the neuronal processes. We again observed that the colocalization of PrPC with GFAP was relatively sparse. (Fig. S2B).

PrPC colocalized with synaptophysin in the myenteric and submucosal plexuses

Because PrPC is reportedly expressed at synapses in the CNS,35 we set out to determine whether or not this was also true of the ENS. Ileal tissues of Prnp+/+ mice exhibited extensive colocalization of PrPC and punctate synaptophysin protein in both the myenteric (Fig. 6A) and submucosal plexuses (Fig. S3A). There were no apparent differences in the degree of synaptophysin expression in preparations from Prnp+/+ and Prnp−/− mice.

image

Figure 6.  Ileal expression of PrPC and synaptophysin in the myenteric plexus of mice and guinea pigs. (A) PrPC and synaptophysin (Synp) distribution in the myenteric plexus of Prnp−/− (a–c) and Prnp+/+ (d–f) mice. Synp colocalized extensively with PrPC in the ileal tissues of Prnp+/+ mice (arrows), but was undetectable in Prnp−/− mice (n = 6 for each group). (B) PrPC expression in the myenteric plexus of the guinea pig. PrPC expressed within myenteric ganglia colocalized with PGP9.5 (a–c), GFAP (d–f) and Synp (g–i). The majority of the PrPC that colocalized with PGP9.5 or GFAP appeared to be on the cell surface of neurons and neuronal processes or distributed on enteric glia (arrows, n = 3). Antibodies used: for mice, PrPC SAF-83; for guinea pigs, SAF-32. Scale Bar: 25 μm.

Download figure to PowerPoint

PrPC in the ENS of the guinea pig

To determine whether the expression pattern of PrPC was similar among different species, we repeated the colocalization studies in whole mount preparations from the guinea pig. SAF-83 immunoreactivity in the guinea pig ENS was not observed (data not shown), however, the pattern of staining using SAF-32 was similar to that observed in the mice. In these studies, PrPC was found to colocalize with PGP9.5, GFAP and synaptophysin in the myenteric (Fig. 6B) and submucosal (Fig. S3B) plexus. The colocalization of PrPC with GFAP appeared to be more robust in the enteric plexuses of the guinea pig than that observed in mice.

Cytosolic staining of PrPC-positive PGP9.5-negative cells

Interestingly, in some ganglia, in preparations from the Prnp+/+ mice, we observed that there were PrPC-positive cells that appeared to have a neuronal phenotype, yet were PGP9.5-negative (Fig. 7). Moreover, the staining pattern in these cells was markedly different as PrPC immunoreactivity was evenly distributed within the cytosol, rather than on the cell surface or on neuronal processes as noted for the majority of the preparations examined. We did not observe similar staining in Prnp-deficient mice. These relatively rare PrPC-positive PGP9.5-negative cells appeared in approximately 1 of 10–15 of the ganglia examined.

image

Figure 7.  Cytosolic immunoreactivity of PrPC. A few of the myenteric and submucosal ganglion preparations had PrPC-positive cells that appeared to have a neuronal phenotype, yet were PGP9.5-negative. Note that the cytosolic region of the cell appears to express PrPC (white arrows) rather than on the cell surface or on neuronal/glial processes (illustrated by open arrows) as noted for the majority of the preparations. These PrPC-positive PGP9.5-negative cells appeared in approximately 1 out of 10–15 ganglia examined. PrPC antibody: SAF-83. n = 26 mice in total were examined. Scale Bar: 25 μm.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Funding
  8. Author contributions
  9. Disclosures
  10. References
  11. Supporting Information

An understanding of PrPC function, and its distribution, is important to gain a better appreciation of the physiological role of this protein in the gut. In the present study, we have extended our previous observations40 describing the localization of PrPC in the colon. Here we show that there were functional differences associated with Prnp deficiency, since loss of the protein was associated with both a reduction in EFS-induced ileal contractility and an enhanced relaxation response. Moreover, the addition of copper increased, whereas treatment with a copper chelator inhibited, NO donor-induced ileal relaxation in Prnp−/− mice, while having no effect in Prnp+/+ mice. We also confirmed that PrPC is constitutively expressed throughout the myenteric and submucosal plexuses of mice and guinea pigs. PrPC immunoreactivity was mainly distributed on the surface of neurons or on nerve fibres or terminals in the ENS, though some degree of colocalization with enteric glia was also observed.

The results from our organ bath studies showed that Prnp−/− mice have reduced EFS-induced ileal contractility when compared to the responses of Prnp+/+ controls. In addition, EFS-induced relaxation (as a percentage of that induced by SNP) was enhanced in Prnp−/− mice. We hypothesized that this might be due to dysregulated copper handling in Prnp−/− mice, since it has been shown that elevated copper concentrations can augment s-nitrosothiol formation and degradation41 thus resulting in an increase in NO levels. This mechanism is consistent with our findings, since a reduction in the clearance of copper, stemming from the absence of PrPC, may have enhanced copper-induced s-nitrosothiol degradation, resulting in increased NO levels, and hence, a reduction in ileal contractility.

Neuronally expressed PrPC has been shown to effectively chelate copper42 and is thus likely to be involved in copper cycling at the synapse.26,43 As copper concentrations are increased in the synaptic cleft following depolarization,43 this metal must be cleared quickly, presumably to prevent the development of damaging reactive oxygen species. In this study, we observed that in the ENS of mice and guinea pigs, PrPC colocalized strongly with synaptophysin, a synaptic vesicle glycoprotein found in neuroendocrine cells and in nearly all neurons of the central nervous system.44 The distribution, and the strong colocalization of PrPC with synaptophysin, further suggests that there may be a role for PrPC in maintaining homeostasis at enteric synapses.

Previous studies have shown that electrophysiological abnormalities, such as impaired long-term potentiation,45 reduced late after-hyperpolarization currents46 and low resting intracellular Ca2 + levels,47 are likely the result of changes in the modulation of L-type voltage-gated calcium channels in cells lacking PrPC.26 Because copper can reduce the influx of calcium through L-type calcium channels, it is plausible that PrPC assists in the maintenance of intracellular calcium homeostasis by providing protection from inhibitory copper concentrations.26 It has also been shown that enzymatic removal of cell surface PrPC results in reduced copper influx,48 and in mice lacking the copper-binding octarepeats region, there was a decreased resistance to oxidative stress and an increased sensitivity to copper toxicity.49

Studies have shown that the vasorelaxant actions of NO are enhanced by copper supplementation, and inhibited when copper is restricted.30,39,50 Furthermore, intravenously administered copper has been shown to induce NOS activity in a number of tissues, which secondarily led to an increase in serum nitrite/nitrate levels.51 Moreover, chronic copper overload in fibroblast cell lines obtained from two mouse mutants, C57BL/6-Atp7aMobr and C57BL/6-Atp7aModap, cell lines that accumulate copper to high levels in normal media as result of a defect in copper export, was shown to upregulate Prnp expression.52 Copper has also been shown to induce Prnp and PrPC expression in primary hippocampal and cortical neurons.53 In our study, the ileal tissues from Prnp−/− mice that were pre-incubated with a copper chelator had marked reductions in NO donor-induced relaxation relative to Prnp+/+ controls. This is interesting as it has been reported that rats fed a copper-free diet developed a defect in NO-mediated vasodilation54 and reduced colonic smooth muscle contractility in response to EFS.55 Collectively, these results suggest that copper deficiency inhibits the NO-mediated mechanism of vascular smooth muscle relaxation, without altering the capacity of the smooth muscle to relax. Thus, in the presence of a copper chelator, augmented copper deficiencies in the Prnp−/− mice may have decreased NO radical availability, disrupted NO-guanylate cyclase interactions, or inhibited s-nitrosothiol release of NO.

There is evidence that PrPC is expressed within components of the ENS.14,15,19 In keeping with this, we observed that in the ileum, nerves of the myenteric and submucosal plexus of mice and guinea pigs expressed PrPC, and more specifically, that the protein colocalized with synaptophysin. However, in these same tissues, PrPC that colocalized with GFAP was limited. In the gut, Ford et al. (2002) observed that enteric neurons expressed high levels of PrPC, yet they were unable to determine whether or not enteric glia expressed the protein.14 Our findings are similar to the observations of this group in which they found PrPC expression in cells from the CNS to be predominantly neuronal, whereas glial cells expressed very low levels of message and undetectable levels of the protein.20 In contrast, our results are not in complete accord with the findings of Albanese et al. (2008) who reported that PrPC in both wild type and overexpressing mice had greater expression in enteric glia relative to neurons.19 These discrepancies could be the result of differences in the antibodies used. For example, Albanese et al. used 03R22, an anti-PrPC polyclonal that recognizes residues 218–232 within the C-terminus, whereas we used either SAF-32, a monoclonal antibody that recognizes residues 59–89 of the octapeptide repeat region, or SAF-83, a conformational monoclonal antibody that recognizes residues 126–164.56 When looking at the distribution of PrPC (a protein known to localize to the cell surface) and GFAP on enteric glial cells, they do appear to be distributed similarly (as also noted by Albanese et al. 2008). However, when we merged these images, there was little or no apparent overlap in our study. In contrast, while GFAP sparsely colocalized with PrPC in the mouse, in the guinea pig, there appeared to be a far greater colocalization of PrPC with GFAP. In part, this might be explained as a result of having to use SAF-32 (a linear antibody) to localize PrPC, rather than SAF-83.

We next looked at the distribution of PrPC in mice that overexpress PrPC to see if the increased expression of PrPC would be helpful in determining colocalization. We again observed that the expression pattern in these Tga20 mice was identical to that as we observed in wild type mice. We also colocalized PrPC with the ectonucleotidase NTPDase3, a neuron-specific cell surface marker,37 and observed that the distributions of PrPC in the myenteric plexus of both Prnp+/+ and Tga20 mice clearly showed these to be colocalized.

We observed that some cells that had a neuronal phenotype, stained positive for PrPC, but were negative for the neuronal marker PGP9.5. Although it has been shown that PGP9.5 does not label all myenteric neurons,57 the finding that PrPC labeled only those devoid of PGP9.5 and showed dramatic intracellular staining, is very intriguing. One possibility is that the antibodies recognized a PrPC-like protein,19 although this would be somewhat surprising for SAF-83 as it recognizes the conformational epitope of the protein. As this did not occur in all preparations, another possibility is that these cells are some type of enteric progenitor cells. Although speculative, this would be an interesting prospect as the level of PrPC expression has been shown to be related to pluripotency.58 For example, it has been shown that Prnp has a relationship with several pluripotency genes (i.e., Nanog) and is a marker of hematopoietic stem cells.59 Moreover, Prnp has shown to play a crucial role as a participant in the regulation of self-renewal/differentiation status of embryonic stem cells,58 and in human mesenchymal stem cells, there is evidence that an increase in the level of PrPC expression enhances proliferation and differentiation.60 Further study will be required to examine this intriguing possibility.

In conclusion, PrPC, which is expressed in both the myenteric and submucosal plexuses of the ENS, was found to colocalize with the synaptic marker, synaptophysin. We also observed that in the myenteric plexus, PrPC colocalized with NTPDase3, a neuron-specific cell surface protein. These findings are likely of importance as the loss of PrPC had functional consequences; ileal tissues from mice lacking this protein had reduced EFS-induced contractility and augmented smooth muscle relaxation responses in the presence of an NO donor. A reduction in the capacity to bind copper in the Prnp−/− mice was likely involved as the addition of copper enhanced, while the addition of a chelator reduced, NO-donor-induced ileal relaxation. These studies suggest that PrPC has a previously unrecognized role in modulating ileal contractility and relaxation. These findings add to the body of literature that illustrates the importance of the prion protein in the ENS in both health and inflammatory and infectious disease.40,61

Acknowledgments

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Funding
  8. Author contributions
  9. Disclosures
  10. References
  11. Supporting Information

The authors would like to acknowledge Garnet Walker for her assistance in genotyping and mouse colony maintenance.

Funding

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Funding
  8. Author contributions
  9. Disclosures
  10. References
  11. Supporting Information

This study was supported by a grant from the Alberta Prion Research Institute (APRI), Alberta Innovates – Bio Solutions (to K.A.S. and F.R.J.). F.R.J. was supported by a Canada Research Chair award. K.A.S. is an Alberta Innovates-Health Solutions (AI-HS) Scientist and holds the CCFC Chair in IBD Research at the University of Calgary. A.L.A. is an AI-HS Graduate Student.

Author contributions

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Funding
  8. Author contributions
  9. Disclosures
  10. References
  11. Supporting Information

The experiments were performed in the labs of KAS and FRJ; GRM, KAS, and MB designed and interpreted the data for the organ bath studies; MB performed the experiments; CMK and ALA performed the immunohistochemical experiments and all authors interpreted the data from these experiments; GRM drafted the paper. All authors contributed to revising the article for critical intellectual content and have approved the final version.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Funding
  8. Author contributions
  9. Disclosures
  10. References
  11. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Funding
  8. Author contributions
  9. Disclosures
  10. References
  11. Supporting Information

Figure S1. Immunoblot of PrPC antibodies in mouse and guinea pig tissues. PrPC expression in brain and ileal homogenates (10% w/v) were prepared and visualized as previously described.40 (A) Immunoblotting, using SAF-32 or SAF-83 antibodies, was used to determine PrPC expression in whole brain homogenates of Prnp+/+ mice (M) or guinea pigs (GP). PrPC expression levels, as determined by SAF-83 or SAF-32, were comparable in mice. In contrast, PrPC expression as determined by SAF-32 was slightly reduced, whereas SAF-83 was absent, in the guinea pig. (B) PrPC expression in Prnp+/+ mice or guinea pigs was determined in 3 cm full-thickness segment homogenates of the terminal ileum. The expression of PrPC using these two antibodies mirrored that as described for the brain. Note the variability at potential N-glycosylation sites between the brain and the ileum. β-actin was used as the lane loading control for all immunoblots.

Figure S2. Ileal PrPC expression in the submucosal plexus. (A) PrPC colocalized with PGP9.5 in ileal whole mount preparations from the Prnp+/+ mice (d-f, arrows). The most abundant colocalization was found to be on the cell surface (arrows) or on neuronal processes. PrPC immunoreactivity was absent in tissues of Prnp-/- mice (a-c). (B) Colocalization of PrPC and GFAP was limited in Prnp+/+ mice (d-f, arrows); PrPC was not detected in submucosal plexus preparations obtained from the Prnp-/- mice (a-c). PrPC antibody: SAF-83. n ≥ 12; Scale bar: 25 μm.

Figure S3. Ileal expression of PrPC and synaptophysin in the submucosal plexus of mice and guinea pigs. (A) PrPC colocalized with Synp in the ileal submucosal plexus of Prnp+/+ mice (d-f, arrows), but was absent in Prnp-/- mice (a-c, n = 6). (B) PrPC expression in the submucosal plexus of the guinea pig (n =3). PrPC expressed within the submucosal ganglia colocalized with PGP9.5 (a-c), GFAP (d-f) and Synp (g-i). PrPC antibody: SAF-83. Scale bar: 25 μm.

FilenameFormatSizeDescription
NMO_1970_sm_FigureS1.tif1921KSupporting info item
NMO_1970_sm_FigureS2.tif3534KSupporting info item
NMO_1970_sm_FigureS3.tif4215KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.