. Author manuscript; available in PMC: 2017 Jul 6.

Published in final edited form as: Nature. 2016 Aug 8;536(7617):464–468. doi: 10.1038/nature19312

The prion protein is an agonistic ligand of the G protein-coupled receptor Gpr126/Adgrg6

Alexander Küffer ^1,^*, Asvin KK Lakkaraju ^1,^*, Amit Mogha ², Sarah C Petersen ², Kristina Airich ¹, Cédric Doucerain ¹, Rajlakshmi Marpakwar ¹, Pamela Bakirci ¹, Assunta Senatore ¹, Arnaud Monnard ¹, Carmen Schiavi ¹, Mario Nuvolone ¹, Bianka Grosshans ³, Simone Hornemann ¹, Frederic Bassilana ³, Kelly R Monk ², Adriano Aguzzi ^1,^†

PMCID: PMC5499706 NIHMSID: NIHMS867483 PMID: 27501152

Abstract

Ablation of the cellular prion protein PrP^C leads to a chronic demyelinating polyneuropathy (CDP) affecting Schwann cells. Neuron-restricted PrP^C expression prevents the disease¹, suggesting that it acts in trans through an unidentified Schwann cell receptor. We found that the cAMP concentration in PrP^C-deficient sciatic nerves is reduced, suggesting the involvement of a G protein-coupled receptor (GPCR). The amino-terminal “flexible tail” (FT, residues 23-120) of PrP^C triggered a concentration-dependent cAMP increase in primary Schwann cells, in the Schwann-cell line SW10, and in Hek293T cells overexpressing the GPCR Gpr126/Adgrg6. In contrast, naïve HEK293T cells and HEK293T cells expressing several other GPCRs did not react to the FT, and ablation of Gpr126 from SW10 cells abolished the FT-induced cAMP response. The FT contains a polycationic cluster (KKRPKPG) similar to the GPRGKPG motif of the Gpr126 agonist, type-IV collagen² (Col4). A KKRPKPG-containing PrP^C-derived peptide (FT_23-50) sufficed to induce a Gpr126-dependent cAMP response in cells and mice, and improved myelination in hypomorphic Gpr126 zebrafish mutants. Substitution of the cationic residues with alanines abolished the biological activity of both FT_23-50 and the respective Col4 peptide. We conclude that PrP^C promotes myelin homeostasis through FT-mediated Gpr126 agonism. Besides clarifying the physiological role of PrP^C, these observations are relevant to the pathogenesis of demyelinating polyneuropathies, common debilitating diseases with limited therapeutic options.

Neuronal Prnp ablation triggers CDP¹, suggesting the existence of a PrP^C receptor on Schwann cells. We therefore assessed the binding of full-length PrP^C (recPrP, residues 23-231), FT (residues 23-110), or its refolded globular domain (GD, residues 121-231), to primary Schwann cell cultures (PSC) from Prnp^ZH1/ZH1 sciatic nerves³ using POM1 and POM2 antibodies⁴. Both recPrP and FT, but not GD, were found to stain PSC (Extended Data Fig. 1A).

Using transcription activator-like effector nucleases, we generated a Prnp-ablated subclone (termed SW10_ΔPrP) of the SW10 Schwann-cell line (Extended Data Fig. 1B–C). Again recPrP and FT, but not GD, adhered to SW10_ΔPrP cells (Fig. 1A, S1D–E). Neither recPrP nor FT adhered to the Prnp⁻^/⁻ hippocampal cell line HpL⁵ (Extended Data Fig. 1F), suggesting that binding was specific to Schwann cells. We next measured the binding of synthetic PrP-derived peptides (2 μM, 20 min) to SW10_ΔPrP cells (Fig. 1B). FT_23-50, but neither FT_39-66 nor any of the carboxy-proximal peptides (Extended Data Table 1), showed binding (Fig. 1C) suggesting that residues 23-38 are essential for the interaction.

A: *Prnp*-ablated SW10 cells (SW10_ΔPrP) were exposed to recombinant PrP^C, FT, or GD (“ligand”). PrP^C and FT, but not GD, adhered to Schwann cells (red: POM1 and POM2 antibodies; grey: DAPI). p75^NGFR antibodies identified Schwann cells. Scale bar: 26 μM. B: FT-derived peptides and PrP^C domains. CC1 and CC2: charge clusters 1 and 2. OR: octapeptide repeats. Peptides are color-coded as in panel C. C: SW10_ΔPrP cells were exposed to FT-derived peptides (2 μM; 20 min) carrying a carboxy-terminal hemagglutinin tag (HA). Flow cytometry showed strong binding by peptide FT_23-50. D: *Prnp*^ZH1/ZH1 sciatic nerves displayed lower cAMP than wild-type BL6 nerves. Dots: individual mice (11–15 mice/group). N.S.: non-significant. Error bars: standard error of the mean. Unpaired Student’s t test was used for statistical analysis. Data (A, C) are representative of 3 biological replicates.

We treated SW10 cells with trypsin (2.5% w/v, 10 min) in order to degrade membrane-resident proteins. Following trypsin inactivation, we added non-trypsinized SW10 cells labeled with the cell-tracking dye Deep Red. HA-tagged FT_23-50 (2μM) was added, and binding was monitored with anti-HA antibodies. 51% of non-trypsinized cells and 5% of trypsinized cells showed FT_23-50 binding (Extended Data Fig. 1G), suggesting the existence of a surface FT receptor on Schwann cells. Cytofluorimetry revealed no altered binding of HA-tagged FT_23-50 to SW10 cells treated with phosphoinositol phospholipase C (PI-PLC, 30′ at 37°C), indicating that binding did not require GPI-anchored surface proteins, whereas binding of POM2 was greatly reduced indicating that PrP^C had been stripped from cell surfaces (Extended Data Fig. 1H).

Might the FT alter cAMP signaling in Schwann cells? At 4 days of age, Prnp^ZH1/ZH1 and WT sciatic nerves showed similar cAMP levels (Extended Data Fig. 2A). At 4 weeks of age, Prnp^ZH1/ZH1sciatic nerves showed a trend towards decreased cAMP levels (Extended Data Fig. 2B). At 12–16 weeks, when CDP is incipient in Prnp^ZH1/ZH1mice¹, Prnp^ZH1/ZH1 nerves exhibited significantly lower cAMP levels (Fig. 1D, p=0.0115). Strictly isogenic C57BL/6J Prnp^ZH3/ZH3 (ref. 6); sciatic nerve lysates from these mice (10–16 week old) also displayed significantly lower levels of cAMP than wild-type mice (Extended Data Fig. 2C).

Sciatic nerves from 12–16 week old tgNSE-PrP mice, expressing PrP^C in neurons, showed cAMP levels similar to wild-type mice, whereas tgMBP-PrP sciatic nerves, which express PrP^C in Schwann cells and suffer from CDP¹, showed lower cAMP levels (Extended Data Fig. 2D). Moreover, when treated with FT (0.1–5 μM; 20 min), Prnp^ZH1/ZH1 PSC and SW10 cells displayed a concentration-dependent cAMP increase (Figs. 2A, Extended Data Fig. 2E) with an EC₅₀ of 860 nM (Extended Data Fig. 2F).

A: Primary *Prnp*^ZH1/ZH1 Schwann cells were treated (20’) with increasing concentrations of recombinant FT or with 10 μM GD. ∅: untreated cells. cAMP levels were determined in cell lysates (5x10⁵ cells/assay). Addition of FT, but not of GD, induced a concentration-dependent cAMP response in Schwann cells. Here and henceforth: *: p<0.05; **: p<0.01; ***: p<0.001. B: cAMP concentrations in primary *Prnp*^ZH1/ZH1 Schwann cell cultures exposed to medium conditioned by HEK293T cells overexpressing wild-type murine PrP^C (HEK^PrP, right) or a non-coding vector (HEK^empty, left) as control. FT-containing medium resulted in cAMP induction. C: Synthetic peptides (27-44 residues) were added to SW10_ΔPrP cells (2 μM each, 20 min). Only FT_23-50 induced cAMP. D: FT_23-50 was preincubated with a two-fold molar excess of miniantibodies Fab3 or Fab71, and added to SW10_ΔPrP cells (20 min). Preincubation with Fab3, but not with Fab71, significantly quenched the FT-dependent cAMP spike. Panels depict independent triplicates; unpaired Student’s t-test was used for statistical analysis.

We transfected HEK293T cells, which express little endogenous PrP^C, with a plasmid expressing murine PrP^C or with an empty plasmid (HEK^PrP and HEK^empty, respectively). Immunoprecipitations showed that HEK^PrP cells released soluble FT (ca. 37 ng FT/ml, Extended Data Fig. 3A–B). Exposure to conditioned medium from HEK^PrP cells, but not from HEK^empty cells, raised cAMP levels in PSC cultures generated from Prnp^ZH1/ZH1 mice (Fig. 2B). Spent medium from PSC cultures and sciatic nerves from 10-week old C57BL/6 mice did not contain any FT, suggesting that little FT is released by Schwann cells (Extended Data Fig. 3C–D).

To assess if the FT contains motifs responsible for inducing the cAMP response, we treated SW10_ΔPrP cells with the same synthetic peptides previously used to assess binding to Schwann cells. Peptide FT_23-50 induced cAMP, whereas FT_34-69 and all carboxy proximal peptides were inactive (Fig. 2C). These results suggest that residues 23-33, which contain the lysine-rich “charge cluster 1” (CC1), represent the biologically active region of the FT. To test the latter prediction, we incubated FT_23-50 with monovalent recombinant phage-derived miniantibodies recognizing the CC1 (Fab3) or the octapeptide repeats (Fab71) of PrP^C. Preincubation with Fab3, but not with Fab71, significantly quenched the ability of FT_23-50 to induce cAMP in SW10_ΔPrP cells (Fig. 2D).

Peripheral nervous system myelination is controlled by Gpr126, an adhesion GPCR expressed by Schwann cells^7,8. We therefore generated HEK293T cells stably overexpressing human Gpr126 or, for control, Gpr124 and Gpr176 (denoted HEK^Gpr126, HEK^Gpr124, and HEK^Gpr176) with a C-terminal V5 epitope tag⁹ (Extended Data Fig. 3E). Flow cytometry revealed FT_23-50 binding to HEK^Gpr126 but neither to HEK^Gpr124 nor to HEK^Gpr176 cells (Extended Data Fig. 3F–G). Moreover, FT_23-50 exposure (20’; ≥500 nM) raised cAMP in HEK^Gpr126 cells but not in HEK^Gpr124 and HEK^Gpr176 cells (Fig. 3A). We then treated HEK293T, HEK^Gpr124, and HEK^Gpr126 cells with recombinant FT or GD (2μM, 20’). FT was detected only in HEK^Gpr126 lysates immunoprecipitated with anti-V5 antibodies, and GD was never detected (Extended Data Fig. 4A).

A: Intracellular cAMP in wild-type HEK293T cells and in Gpr176, Gpr124 and Gpr126 overexpressors exposed to FT_23-50 (0.5 μM, 20 min). Only HEK^Gpr126 cells showed a cAMP increase. B: Wild-type (left) and Gpr126-ablated (right) SW10 cells were exposed to FT_23-50 (2μM, 20 min). SW10 cells, but not SW10_ΔGpr126 cells, respond to FT_23-50 with a cAMP spike. Moreover, SW10 cells did not respond to alanine-substituted ${FT}_{23 - 50} ({FT}_{23 - 50}^{K \to A})$ C: Protein was isolated from wild-type or *Prnp*^ZH3/ZH3 sciatic nerves (13-week old female mice) and Western blots were probed for EGR2 and actin. Densitometry (below) showed reduced Egr2 in *Prnp*^ZH3/ZH3 nerves (p=0.028). For uncropped gels see Supplementary Information File 1. Panels depict independent triplicates; unpaired Student’s t-test was used for analysis.

Next, we generated clonal SW10 cells devoid of endogenous Gpr126 (designated SW10_ΔGpr126). When treated with FT_23-50, SW10 (but not SW10_ΔGpr126) cells reacted with increased cAMP levels. A modified ${FT}_{23 - 50}^{K \to A}$ peptide (with lysine residues replaced with alanines) was ineffective in binding cells and inducing cAMP (Fig. 3B). We then treated SW10_ΔGpr126 cells transfected with human Gpr126, Gpr124, Gpr176 or Gpr56 with FT_23-50. Only Gpr126-transfected cells showed a cAMP response (Extended Data Fig. 4B) similar to that of naïve SW10 cells, indicating that the tag did not affect the function of Gpr126 (Extended Data Fig. 4C). When treated with conditioned media from HEK^PrP or HEK^empty cells, SW10 but not SW10_ΔGpr126 cells responded with a cAMP spike (Extended Data Fig. 4D). Moreover, FT adsorption was reduced in SW10_ΔGpr126 cells (Extended Data Fig. 4E).

We then administered FT_23-50 (2μM, 20’) to HEK^Gpr126 cells and HEK293(H) cells transfected with plasmids encoding human Gpr56, Gpr64, Gpr133, or Gpr97. Only Gpr126-expressing cells showed a cAMP response (Extended Data Fig. 4F). The magnitude of cAMP response was not enhanced by increasing the transfected plasmid, suggesting that other signaling components became limiting (Extended Data Fig. 5A). There was no cAMP induction in Prnp^ZH1/ZH1 cerebellar granule neuronal cultures treated with FT_23-50 or ${FT}_{23 - 50}^{K \to A}$ (Extended Data Fig. 5B), as expected from the minimal Gpr126 expression in the brain¹⁰. The FT is released from PrP^C by metalloproteases¹¹; after treatment with the metalloprotease inhibitor TAPI-2, HEK^PrP-conditioned medium contained significantly less FT (Extended Data Fig. 5C–D) and displayed reduced cAMP-inducing activity (Extended Data Fig. 5C).

Egr2/Krox-20 controls the expression of myelin genes and is implicated in myelin maintenance¹². Egr2 expression was decreased in 13-week-old Prnp^ZH3/ZH3 sciatic nerves (p<0.05; Fig. 3C), and recombinant FT (2 μM, 24h) activated Egr2-dependent luciferase expression in SW10 cells (Extended Data Fig. 5E). Similarly, Egr2 transcription was upregulated in primary Schwann cells treated with recombinant FT (2 μM; 1h) (Extended Data Fig. 5F). Also, Akt phosphorylation increased 5 min after treatment with recombinant FT (2μM) and peaked at 10 min in SW10_ΔPrP but not in SW10_ΔGpr126 cells (Extended Data Fig. 5G). The integrity of SW10 cells and their subclones was confirmed by the expression of myelin genes (Extended Data Fig. 6A).

We identified two regions of similarity between FT (KKRPKPG and QGSPG) and the Gpr126 ligand, Type-IV collagen (Col4)² (GPRGKPG and QGSPG, Fig. 4A). Replacement of the conserved cationic residues with alanines (KKRPKPG ➭ AAAPAPG), but not other substitutions, abrogated cAMP induction in SW10_ΔPrP cells (Fig. 4B); treatment with FT_23-34 (2μM, 20’), which contains KKRPKPG, sufficed to induce cAMP in SW10_ΔPrP but not in SW10_ΔGpr126 cells (Fig. 4C). We next generated murine PrP^C mutants containing alanine substitutions in either of the two conserved motifs. After transient transfection, both mutants were highly expressed by HEK293T cells (Extended Data Fig. 6B), and cleaved FT was recovered in the medium (Extended Data Fig. 6C). When applied to SW10_ΔPrP cells, conditioned media from HEK293T cells expressing wild-type or QGSPG-mutated ${PrP}^{C} ({HEK}_{\sum QGSPG}^{PrP})$ induced cAMP, whereas medium from cells expressing KKRPK-mutated ${PrP}^{C} ({HEK}_{\sum KKRPK}^{PrP})$ did not (Extended Data Fig. 6D). We then generated 21-mer peptides bearing the corresponding Col4 sequence (GPRGKPG) or an alanine-substituted variant (AAAGAAG). The native Col4 peptide (8 μM), but not the mutated peptide, induced cAMP in SW10_ΔPrP cells (Extended Data Fig. 6E).

A: Sequence alignment revealed two regions of similarity between the FT and Col4 (red boxes). Yellow and green shades represent high and moderate similarity, respectively. Dotted line: non-homologous residues. B: SW10_ΔPrP cells were treated with synthetic FT_23-50 or modified version of FT_23-50 in which the KKRPK or QGSPG motifs were replaced with alanines (2μM, 20’). Alanine substitution of KKRPK (peptide ${FT}_{23 - 50}^{K \to A}$ ), but not of QGSPG, abrogated cAMP induction. C: SW10_ΔPrP and SW10_ΔGpr126 cells were exposed (2μM, 20’) to the synthetic peptides FT_23-50, FT_23-34 or SFT_23-34 (peptide containing scrambled amino acid sequence of FT2_3-34). FT_23-50 and FT_23-34 induced cAMP in SW10_ΔPrP cells but not SW10_ΔGpr126 cells. Panels depict independent triplicates; unpaired Student’s t-test was used for analysis.

In zebrafish, Gpr126 controls myelination, yet Prnp⁻^/⁻ mice have no obvious myelination defects but develop a late-onset peripheral neuropathy with demyelination, onion bulb formation (Fig. 5A, Extended Data Fig. 7A), decreased Remak bundles containing ≥10 axons and increased bundles containing <10 axons (Extended Data Fig. 7B), indicative of impaired axon-Schwann cell interactions. To test if Gpr126 dysfunction may cause late-onset phenotypes in mice, we examined sciatic nerves from Dhh^Cre::Gpr126^fl/fl mice experiencing Schwann-cell specific Gpr126 ablation from ~E12.5 onwards^13,14. At one year of age, we noted neuropathic traits similar to those of Prnp⁻^/⁻ mice including reduced bundles containing >20 unmyelinated axons, increased bundles containing <10 axons, abnormal cytoplasmic Schwann cell protrusions¹ and onion bulbs (Fig. 5B and Extended Data Fig. 7C–D).

A: Transmission electron micrographs from 14 month old wild-type BL6 and *Prnp*^ZH3/ZH3 sciatic nerves (N=3/4 animals). Thinly myelinated axons (arrowhead), loss of axon-Schwann cell interaction (boxes), abnormal cytoplasmic Schwann cell protrusions (white arrowhead) and initial onion bulb formation (asterisk) were observed in *Prnp*^ZH3/ZH3 mice. Scale bars: 500 nm. B: Neuropathic phenotype of one-year-old *Dhh^Cre;Gpr126^fl/fl* mutant mouse nerves. Left panels: toluidine blue-stained sections of sciatic nerves from control *Gpr126^fl/fl* (phenotypically wild-type) and *Dhh^Cre::Gpr126^fl/fl* (*Gpr126*^ΔSchwann) mice. *Gpr126^fl/fl* nerves were well myelinated (N = 3/3 animals), whereas *Gpr126*^ΔSchwann nerves exhibited myelin loss with readily apparent onion bulb-like structures (arrowheads) (N = 3/3 animals). Right panels: Transmission electron micrographs from *Gpr126^fl/fl* and *Gpr126*^ΔSchwann sciatic nerves. Myelinated axons (“M”) and Remak bundles (“R”) were found in *Gpr126^fl/fl* sciatic nerves (N = 3/3 animals). Numerous defects were observed in *Gpr126*^ΔSchwann sciatic nerves (N = 3/3 animals) including onion bulbs (black arrowheads), abnormal cytoplasmic protrusions (white arrows), and loss of axon-Schwann cell interactions (boxes) similar to *Prnp*^ZH3/ZH3 mice. Scale bars = 20 μm (a–b), 2 μm (c–d). C: *gpr126^st63* hypomorphic mutant larvae were treated with vehicle (DMSO) or FT_23-50 (20 μM) at 50–55 hours post fertilization (hpf) and posterior lateral line nerve was immunostained at 5 days post fertilization (dpf) for myelin basic protein (Mbp, green). AcTub: acetylated tubulin (red) labeling axons. Scale bar = 20 μm. The intensity of immunofluorescence was assessed by morphometry (right graphs). FT treatment enhanced Mbp immunofluorescence without affecting AcTub. D: Mbp expression was scored in larvae treated with FT_23-50 or vehicle (DMSO). FT_23-50 treatment resulted in a higher proportion of rescued (“some” and “strong”) MBP expression in *gpr126^st63* (53% vs. 34%), but not in *gpr126^st49* larvae (p < 0.05, Fisher’s two-tailed exact test). N.S. = not significant. N ≥ 25 larvae per replicate treatment. E: *Prnp*^ZH3/ZH3 and BL6 mice were intravenously exposed to either FT_23-50 or its non-charged analogue ${FT}_{23 - 50}^{K \to A}$ (600 μg/animal, 20 min). *Prnp*^ZH3/ZH3 sciatic nerves showed a significant cAMP increase after injection of FT_23-50, but not of ${FT}_{23 - 50}^{K \to A}$ . FT_23-50-treated *Prnp*^ZH3/ZH3 sciatic nerves reached cAMP levels similar to those of BL/6 mice injected with ${FT}_{23 - 50}^{K \to A}$ . Each dot represents an individual animal. F: Heart cAMP levels were also increased in FT_23-50-injected mice.. G: Control *Gpr126^fl/fl* (WT) and *Dhh^Cre::Gpr126^fl/fl* mutant (*Gpr126*^ΔSchwann) mice were intravenously injected with either FT_23-50 or uncharged ${FT}_{23 - 50} ({FT}_{23 - 50}^{K \to A})$ (600 μg/animal). Sciatic nerves were isolated 20 min post injection. FT_23-50 elicited a significant cAMP increase in WT mice, but not in *Gpr126*^ΔSchwann mice. ${FT}_{23 - 50}^{K \to A}$ injection did not alter cAMP levels (*N=3*). Unpaired Student’s t-test was used for analysis.

The gpr126^st63 zebrafish mutant has a point mutation reducing Gpr126 signaling and shows reduced myelin basic protein (Mbp) expression by Schwann cells of the posterior lateral line nerve (PLLn) (Fig. 5C, Extended Data Fig. 8A) which can be rescued by Gpr126 activators^8,15,16. When applied to gpr126^st63 zebrafish larvae at 50–55h post-fertilization (hpf), FT_23-50 (20 μM) increased Mbp expression in the PLLn at 5 days post-fertilization (dpf) compared to DMSO-treated larvae (Fig. 5C–5D, p<0.05). We also treated gpr126^st49 larvae, which encode a truncated Gpr126 incapable of G_s signaling^8,16. Mbp was barely detectable in both FT_23-50 and DMSO-treated mutant gpr126^st49 larvae (Fig. 5C, Extended Data Fig. 8B).

We next administered intravenous FT_23-50 or ${FT}_{23 - 50}^{K \to A}$ peptide (600 μg) to 10–16 week old Prnp^ZH3/ZH3 and wild-type mice (N=8 per group, all littermates). FT_23-50, but not ${FT}_{23 - 50}^{K \to A}$ , induced cAMP elevation in both Prnp^ZH3/ZH3 and Prnp^ZH1/ZH1 sciatic nerves 20 minutes post injection; cAMP reached levels similar to those of BL6 mice injected with ${FT}_{23 - 50}^{K \to A}$ (Fig. 5E, Extended Data Fig. 8C). FT_23-50 induced a cAMP spike also in the heart, which expresses Gpr126 (Fig. 5F, Extended Data Fig. 8D) but not in kidney and brain (Extended Data Fig. 8E–F) which do not express Gpr126¹⁷. Finally, FT_23-50 but not ${FT}_{23 - 50}^{K \to A}$ (600 μg) induced a robust cAMP spike in Gpr126^fl/fl but not in Dhh^Cre::Gpr126^fl/fl sciatic nerves (Fig. 5G).

Here we report the first molecular elucidation of a phenotype caused by PrP^C deficiency. While Gpr126 is crucial for peripheral nerve myelination during development, the late-onset phenotype of Prnp-ablated mice suggests additional roles in myelin maintenance. While Gpr126⁻^/⁻ mice exhibit drastic hypomyelination, the Prnp^−/− phenotype is relatively mild and late-onset akin to peripheral neuropathies, whose prevalence is 2.4% generally and 8% among the aged¹⁸. Perhaps Gpr126 provides basal signaling in Prnp^−/− mice through its interaction with type-IV collagen and/or laminin-211 (ref. 2,16), which may not suffice for long-term myelin maintenance. The Gpr126-agonistic properties of systemically administered FT peptides suggests that soluble ligands may be beneficial in diseases caused by hypomorphic Gpr126 mutations¹⁹, and perhaps also in other hereditary motor-sensory neuropathies.

A moderately conserved homology region between PrP^C and Col4 proved essential to the biological activity of both proteins, whereas laminin-211 may activate Gpr126 through different mechanisms. Although Gpr126 is not expressed in the central nervous system, certain mutants of PrP^C cause myelin pathology in vivo²⁰. This observation raises the question whether inappropriate GPCR activation may play a role in prion diseases within the CNS.

Methods

Mice

Mice were bred in specified-pathogen free facilities at the University Hospital Zurich and Washington University, and housed in groups of 3–5, under a 12 h light/12 h dark cycle (from 7 am to 7 pm) at 21±1°C, with sterilized chow (Kliba No. 3431, Provimi Kliba, Kaiseraugst, Switzerland) and water ad libitum. Animal care and experimental protocols were in accordance with the Swiss Animal Protection Law, and approved by the Veterinary office of the Canton of Zurich (permits 123, 130/2008, 41/2012 and 90/2013). The following mice were used in the present study: C57BL/6J, Prnp^ZH1/ZH1 (ref. 3), co-isogenic C57BL/6J Prnp^ZH3/ZH3 and Prnp^WT/WT control mice⁶ and Schwann cell-specifc Dhh^Cre;Gpr126^fl/fl mutants^3,4. Mice of both genders were used for experiments unless specified. Archival tissues from previous studies^1,6 were also analyzed in the current study.

Primary Schwann cell culture

Sciatic nerves from postnatal day 2–5 were dissected using microsurgical techniques. Nerves were dissociated in serum-free DMEM supplemented with 0.05% collagenase IV (Worthington) for 1h in the incubator. Sciatic nerves were mechanically dissociated using fire-polished Pasteur pipettes. Cells were filtered in a 40 μM cell strainer and washed in Schwann cell culture medium (DMEM, Pen-Strep, Glutamax, FBS 10%) by centrifugation at 300g for 10 min. Resuspended cells were plated on 3.5 cm petri dishes previously coated with poly-L-lysine 0.01% (wt/vol) and laminin (1 mg/ml). Laminin (Cat. No: L2020; from Engelbreth-Holm-Swarm murine sarcoma basement membrane) and poly-L-lysine were obtained from Sigma-Aldrich.

Purification of recombinant proteins

Full-length recombinant PrP (recPrP, residues 23-231) and globular domain (GD, residues 121-231) were purified as previously described^21–23. The generation of the GST fusion FT-PrP expression vector (pGEX-KG FT-PrP) was described previously; a modified purification protocol was used²⁴. The FT-PrP expression vector was transformed into BL21 (DE3) strain of Escherichia coli (Invitrogen). Bacteria were grown in Luria-Bertani medium to an OD of 0.6, and the expression of the fusion protein was induced with 0,5mM isopropyl-1-thio-β-D-galactopyranoside (AppliChem). Cells were then grown for another 4h at 37 °C and 100 rpm shaking. Cells were pelleted at 5000g for 20 min at 4 °C (Sorvall centrifuge, DuPont). The pellet was resuspended on ice in lysis buffer (phosphate-buffered saline supplemented with complete protease inhibitors (EDTA-free, Roche), phenylmethyl sulfonyl fluoride (Sigma) and 150 μM lysozyme (Sigma) and incubated on ice for 30 min. Triton-X 100 (1%), MgCl₂ (10mM) and DNase I (5 μg/ml, Roche) were added, and the lysate was incubated on ice for 30 min. The lysate was than centrifuged for 20 min at 10’000g at 4°C. Glutathione sepharose beads were washed with PBS and incubated with the cell lysate for 1h at 4°C on a rotating device. Beads were packed into a column and washed with PBS until a stable baseline was reached as monitored by absorbance at A₂₈₀ using an ÄKTAprime (GE healthcare). The fusion protein was cleaved on the beads with 5U/ml Thrombin (GE Healthcare) for 1h at room temperature under agitation. For thrombin removal, benzamidine sepharose beads were added and incubated for 1h at 4°C on a rotating wheel. Protein preparations were analyzed by 12% NuPAGE gels followed by Coomassie- or silver-staining. To achieve a higher purity of the protein, we next applied the protein to a sulphopropyl (SP) sepharose column equilibrated with 50mM Tris-HCl buffer, pH=8.5. Elution was performed with a linear NaCl gradient ranging from 0–1000 mM. Fractions containing the protein were collected and concentrated (AMICON; MWCO 3500). The protein was then injected in 500 μl portions into a size-exclusion chromatography system (TSK-GEL G2000SW_XL column (Tosoh Bioscience)) and eluted with a linear gradient using PBS. Pure fractions were combined, concentrated and stored at −20°C. The purity of FT-PrP was > 95–98 % as judged by a silver-stained 12% NuPAGE gel.

Cell Culture

SW10 cells and clones derived from it were all grown in DMEM medium supplemented with 10% fetal bovine serum (FBS), penicillin-streptomycin and Glutamax (all obtained from Invitrogen). HEK293T cells and clones derived therefrom were grown in DMEM-F12 medium supplemented with 10%FCS, penicillin-streptomycin and Glutamax (all obtained from Invitrogen). Human Gpr126 (NM_020455), Gpr124, Gpr64, Gpr56, Gpr133, Gpr56 and Gpr176 expression plasmids (pCGpr126-V5, pCGpr124-V5, pCGpr65-V5, pCGpr56-V5, pCGpr133-V5, pCGpr56-V5 and pCGpr176-V5) were generated by PCR amplification of the respective cDNA followed by TOPO cloning into the pCDNA3.1/V5-His-TOPO vector. The cDNA was in frame with the V5 tag (sequence: GKPIPNPLLGLDST) at the C terminus. HEK^GPR126, HEK^GPR176 cells were generated by transfecting 1μg of plasmid in one well of a subconfluent 6-well plate using 3μl Fugene (Roche) according to the manufacturer’s protocol. 24h after transfection, cells were transferred to a 10cm dish and grown in selective medium containing 0.4 mg/ml G418 (Invitrogen) until emergence of resistant colonies. A limiting dilution was carried out to obtain clonal lines. Membrane expression of the transgene was assessed in the selected clones by confocal microscopy using 1:100 diluted anti-V5 antibody (Invitrogen) and the Cytofix/Cytoperm kit (Pharmingen Cat. Nr. 554714), according to manufacturer’s protocol.

Cerebellar granule neuronal culture

Cerebellar granule neurons were generated from 7–8 day old Prnp^ZH1/ZH1 mice with methods described previously²⁵. Cultures were plated at 350,000 cells/cm² in Basal Medium Eagle (BME) (Invitrogen) with 10% (v/v) FCS and maintained at 37°C in 5% CO₂.

Plasmids and Transfections

pCDNA-PrP^C was generated by cloning murine PrP^C into pCDNA3.1 vector as described previously²⁶. Site-specific mutagenesis kit (Stratagene) was used to induce alanine substitutions of QPSPG and KKRPK domains in PrP^C. Primers used for generating Ala-QPSPG plasmid are: forward- GTG GAA GCC GGT ATC CCG GGG CGG CAG CCG CTG CAG GCA ACC GTT ACC C reverse: GGG TAA CGG TTG CCT GCA GCG GCT GCC GCC CCG GGA TAC CGG CTT CCA C. Primers for Ala-KKRPK are: forward- CTA TGT GGA CTG ATG TCG GCC TCT GCG CAG CGG CGC CAG CGC CTG GAG GGT GGA ACA CCG, reverse: CGG TGT TCC ACC CTC CAG GCG CTG GCG CCG CTG CGC AGA GGC CGA CAT CAG TCC ACA TAG. Transfections were performed with Lipofectamine 2000 (Invitrogen) according to manufacturer’s protocol. 3μg of DNA was used per well of a 6-well plate. 24h post transfections, cells were washed using PBS, and fresh medium was added to the cells.

Immunoprecipitation

HEK293T and HEK^GPR126 cells growing in T75 flasks at 50% density were treated with recombinant FT or GD (2μM, 20 min). Cells were washed twice in PBS and lysed in IP buffer: 1% Triton X-100 in PBS, 1x protease inhibitors (Roche) and Phospho stop (Roche) for 20 min on ice followed by centrifugation at 5000 rpm for 5 min at 4°C. BCA assays were performed to quantify the amount of protein, and 500 μg of protein was used for Immunoprecipitations. 2 μg of anti-V5 antibody were added to the cell lysate and incubated on a wheel rotator overnight at 4°C. On the following day, Protein G dynabeads (Invitrogen) were added to the samples and incubated for further 3h on the wheel at 4°C. Beads were washed three times, 5 min each using the IP buffer followed by addition of 2x sample buffer containing DTT (1mM final). Samples were heated at 95°C for 5 min, loaded on 4–12% Novex Bis-tris gels (Invitrogen), and migrated for 1.5h at 150V followed by western blotting. Immunoprecipitations were performed by adding 2 μg of POM2 antibody to 500μl of cell medium and incubating overnight on a wheel rotator at 4°C. Protein G beads were then added, and incubation on a wheel rotator at 4°C was performed again.

RNA Isolation and quantitative PCR

RNA extraction and quantitative PCR was performed as described previously¹. The following primers were used: EGR2 fw: 5′-AATGGCTTGGGACTGACTTG-3′; EGR2 rev: 5′-GCCAGAGAAACCTCCATT-3′ CA; GAPDH fw: 5′-CCACCCCAGCAAGGAGAC-3′; GAPDH rev: 5′-GAAATTGTGAGGGAGATGCT-3′.

Zebrafish mutant strains and analysis

Adult zebrafish were maintained in the Washington University Zebrafish Consortium facility (http://zebrafishfacility.wustl.edu/) and all experiments were performed in compliance with institutional protocols. Embryos were collected from harem matings or in vitro fertilization, raised at 28.5°C, and staged according to standard protocols²⁷. The gpr126^st49 and gpr126^st63 mutants were described previously^7,8.

Zebrafish peptide treatment, immunostaining, and quantification

gpr126^st63 or gpr126^st49 mutants were collected from homozygous mutant crosses and WT larvae were collected from AB* strain crosses and raised to 50 hours post fertilization (hpf). FT_23-50 treatment of gpr126 mutants was performed as previously described¹⁵. Briefly, egg water was then replaced with either 20 μM FT_23-50 in egg water or egg water containing an equivalent volume of DMSO. At 55 hpf, larvae were washed twice and raised in egg water to 5 days post fertilization (dpf). WT and gpr126 larvae were fixed in 2% paraformaldehyde + 1% tricholoroacetic acid in phosphate buffered saline, and Mbp and acetylated tubulin immunostaining was performed as described previously^8,28. Expression scoring was performed with observer blind to treatment according to the following rubric: “strong” = strong and consistent expression throughout PLLn, “some” = weak but consistent expression in PLLn, “weak” = weak and patchy expression in PLLn, “none” = no expression in PLLn. N = three independent replicate gpr126^st63 assays and one gpr126^st49 assay. n = 87 DMSO-treated gpr126^st63 larvae, 81 Prp-FT-treated gpr126^st63 larvae, n = 27 DMSO-treated gpr126^st49 larvae, n = 25 Prp-FT-treated gpr126^st49 larvae.

Fluorescent nerve images were analyzed using the Fiji software²⁹. A rectangle region-of-interest (ROI) was drawn longitudinally over the fluorescent nerve. The longitudinal grey-scale histogram of the myelin basic protein (MBP) was normalized pixel-by-pixel to the corresponding intensity of the acetylated tubulin (AcTub). The size of the measured ROIs was kept constant across different treatment modalities.

Flow cytometry

SW10 cells were grown in P75 flasks at 50% density, rinsed with PBS, and detached from culture flasks with dissociation buffer containing EDTA (GIBCO). After detaching, cells were washed to remove residual EDTA and counted using a Neubauer chamber. Batches of 10⁵ SW10 cells were transferred to FACS tubes, treated with HA-tagged recombinant peptides for 20 minutes, washed, and incubated with Alexa-488 conjugated anti-HA antibody for 30 minutes. After further washes and centrifugations, cells were resuspended in 200μl of FACS buffer (PBS +10% FBS) and analyzed with a FACS Canto II cytofluorimeter (BD Biosciences). Data were analyzed using FloJo software.

Western blot analysis

Schwann cells were lysed in cell-lysis buffer (Tris-HCl 20mM, NaCl 137mM, Triton-X-100 1%) supplemented with protease inhibitor cocktail (Roche complete mini). The lysate was homogenized by passing several times through a 26G syringe, and cleared by centrifugation at 8000g, 4°C for 2 min. in a tabletop centrifuge (Eppendorf 5415R). Protein concentration was measured with the BCA assay (Thermo Scientific). 10 μg total protein was boiled in 4 x LDS (Invitrogen) at 95°C for 5 min. After a short centrifugation, samples were loaded on a gradient 4–12% Novex Bis-Tris Gel (Invitrogen) for electrophoresis at constant voltage of 200V. Gels were transferred to PVDF membranes with the iBlot system (Life technologies). Membranes were blocked with 5% Top-Block (Sigma) in PBS-T for 1h at room temperature. Primary antibody was incubated overnight in PBS-T with 5% Top-Block. Membranes were washed three times with PBS-T for 10 min. and incubated for 1 hour with secondary antibodies coupled to horseradish peroxidase at room temperature. After three washes with PBS-T, the membranes were developed with Crescendo chemiluminescence substrate system (Millipore). Signals were detected using a Stella 3200 imaging system (Raytest, Straubenhardt, Germany).

Antibodies

Monoclonal antibodies against PrP^C were obtained and used as described previously⁴. Fab3 and Fab71 antibodies were generated using the phage display technology and their epitopes were mapped with overlapping peptides. Anti AKT, p-AKT were obtained from Cell signaling and used at 1:2000 dilutions for western blotting. Anti-p75NGF receptor antibody was obtained from Abcam and used at 1:200 dilutions for immunofluorescence. Anti V5 antibody was from Invitrogen and used at a dilution of 1:500 for western blot and 2μg of the antibody was used for immunoprecipitation on 500μg of cell lysate.

Cyclic AMP measurements

In the Direct cAMP ELISA assay, cAMP levels were assessed with a colorimetric competitive immunoassay (Enzo Life Sciences). Quantitative determination of intracellular cAMP was performed in cells or tissues lysed in 0.1 M HCl to stop endogenous phosphodiesterase activity and to stabilize the released cAMP. SW10 or HEK293T cells (100’000 cells/well) were plated in 6-well plates to a ca. 50% density. Cells were treated with conditioned media or recombinant peptides (2μM, unless specified) for 20 minutes unless otherwise mentioned. Cells were lysed with 0.1M HCl lysis buffer (Direct cAMP ELISA kit, Enzo). To ensure complete detachment of cells cell scrapers were used. Lysates were homogenized with 26G needle and syringe before clearing by centrifugation at 600g for 10 min. The subsequent steps were performed according to the manufacturer’s protocol based on competition of sample cAMP with a cAMP-alkaline phosphatase conjugate. Cyclic AMP levels were calculated using a cAMP standard curve in the case of ELISA based assay. Finally, cAMP concentrations were normalized to total protein content in each sample. cAMP changes are represented as fold changes to the respective controls. For each experiment, at least three independent biological replicates were used. For normalization purposes, the median value of the respective control sample was defined as “1”. All measurements within each panel were normalized to this control value.

Generation of Schwann cell lines devoid of Gpr126

We designed two CRISPR short-guide RNA (sgRNAs) against exon 2 of Gpr126 (upper Guide CCTGTGTTCCTCTCTCAGGT and lower Guide AACAGGAACAGCAGGGCGCT). The DNA sequences corresponding to the sgRNAs were cloned into expression plasmids and transfected with EGFP-expressing Cas9-nickase plasmids. Single EGFP-expressing Schwann cells were isolated with a FACS sorter (Aria III). To determine the exact sequence of indels induced by genome editing we amplified the sgRNA-targeted locus by PCR and subcloned the fragments into blunt-TOPO vectors. Ten colonies per cell line were sequenced and showed distinct indels on each allele. A clonal subline devoid of Gpr126 was used for further studies. This cell line possessed insertions on both the alleles; a 49bp insertion at position 118 and a 5bp insertion at position 84 on each allele. Both insertions led to a frameshift and to the generation of premature stop codons leading to early translation termination.

Promoter luciferase assay

Luciferase reporter constructs were generated containing a 1.3kB sequence upstream of the transcription-starting site of Egr2. SW10 Schwann cells were transfected with Egr2 reporter construct and a renilla plasmid using lipofectamine 2000. After one day in vitro, Schwann cells were treated with recombinant full-length PrP (23-231), the globular domain of PrP (121-231) or PBS control. Luciferase activity was measured 24 hours after stimulation with Dual-Luciferase Reporter Assay System (Promega) according to the manufacturer’s recommendations. Results were normalized to renilla transfection controls.

Immunocytochemistry

Glass coverslips were placed in 12-well plates (Thermo scientific) and coated with 0.01% wt/vol Poly-l-lysine solution (Sigma) overnight at room temperature. Coverslips were washed three times with ddH₂O and dried for 2h in a laminar-flow hood. Schwann cells were seeded and cultured at 50% density. Cells were treated with recombinant FT-PrP, full length recPrP or C1-PrP for 20 min., and washed with serum-free DMEM. Cells were further washed with PBS followed by fixation with 4% paraformaldehyde. Fixed cells were incubated in blocking buffer (PBS+10%FBS) for 1h. Cells were treated with various primary antibodies followed by washes and incubation with Alexa 488 and Alexa 647 tagged rabbit or mouse secondary antibodies (Life technologies). Imaging was performed by Leica SP2 confocal microscope using a 20x objective; images were processed by Image J software.

Transmission electron microscopy

Transmission electron microscopy was performed as previously described⁶. Briefly, mice under deep anesthesia were subjected to transcardial perfusion with PBS heparin and sciatic nerves were fixed in situ with 2.5% glutaraldehyde + 2% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4 and embedded in Epon. Ultrathin sections were mounted on copper grids coated with Formvar membrane and contrasted with uranyl acetate/lead citrate. Micrographs were acquired using a Hitachi H-7650 electron microscope (Hitachi High-Tech, Japan) operating at 80 kV. Brightness and contrast were adjusted using Photoshop. For quantification of Remak bundles and onion bulb like structures, images were captured at 1500X magnification and axon numbers and abnormal onion bulb like structures were counted manually. Quantification was performed in a blinded fashion by assigning numbers to the images and upon completion of quantification genotypes were revealed.

Recombinant peptides

HA-tagged and untagged synthetic peptides were produced by EZ Biosciences. A stock solution of 2mM was prepared by dissolving the peptides in PBS and they were used at a final concentration of 2μM unless specified. The sequences all the peptides used in this study can be found in Extended Data Table 1.

Extended Data

Extended Data Figure 3 — A: HEK293T cells were transfected with an empty vector (HEK^empty) or a plasmid expressing murine PrP^C (HEK^PrP). Cell medium was collected 48h after transfection and subjected to immunoprecipitation with monoclonal antibody POM2 (against PrP^C), followed by western blotting using biotinylated POM2 and streptavidin-HRP. FT was observed only in the medium from HEK^PrP cells. B: FT released into the conditioned medium of HEK^PrP was immunoprecipitated using POM2 and visualized by Western blotting with biotinylated POM2. Dilutions of recombinant FT (3.125 – 100 ng) were used for calibration, and the concentration of FT released into the medium was estimated at 37 ng/ml. C: Conditioned medium from primary BL6 Schwann cells cultures (PSC^BL6) was subjected to immunoprecipitation with antibody POM2 (against PrP^C) followed by Western blotting with POM2. For control, we used conditioned medium from HEK cells transfected with a non-coding plasmid (HEK^empty) or a with a plasmid encoding murine PrP^C (HEK^PrP). FT was only detected in conditioned medium from HEK^PrP cells (lane 2) but not in conditioned medium from two independent PSC^BL6 cultures (lanes 3–4). Asterisks denote immunoglobulins detected by the secondary antibody. D: Sciatic nerves lysates obtained from 10 week old *Prnp*^ZH1/ZH1 and BL6 mice and subjected to immunoprecipitation with POM2 antibody followed by Western blotting with POM2. Full length PrP^C, but no FT, was detectable in the immunoprecipitates from BL6 mice. E: Wild-type HEK293T cells (HEK^WT) or HEK293T cells overexpressing various GPCRs bearing V5 epitope tags (HEK^Gpr126, HEK^Gpr124, and HEK^Gpr176) were grown on coverslips and stained with anti V5 antibody (detecting tagged GPCRs; magenta). Nuclei were stained with DAPI (blue). Staining revealed cell surface expression of all transfected GPCRs. Scale bar: 8μm. F: HA-tagged FT_23-50 peptide (2μM) was added to wild-type HEK293T (HEK^WT) cells or to HEK^GPR126 cells, labeled with anti-HA antibody, and subjected to cytofluorimetry. Overexpression of Gpr126 increased the binding of FT_23-50. G: Binding of HA-tagged FT_23-50 to HEK^GPR126 cells (right panel, monitored by cytofluorimetry) was conspicuously increased over that of wild-type, Gpr176, and Gpr124-overexpressing HEK293T cells. Data are representative of three biologically independent experiments; statistical significance was evaluated by unpaired Student’s t-test.

Extended Data Figure 4 — A: HEK293T, HEK^GPR124 and HEK^GPR126 cells were exposed (20 min) to recombinant FT, GD (2μM), or PBS, and subjected to immunoprecipitation using the anti-V5 antibody, followed by Western blotting using POM2, anti-V5 or POM1. Anti-V5 detected both full-length Gpr126, Gpr124 (denoted as GprV5 for both proteins) and the respective C-terminal fragments (Gpr126V5-CTF, Gpr124V5-CTF). POM2 revealed a band corresponding to the FT (lane 3) that co-precipitated with GPR126. POM1 indicated that GD did not bind. Lanes 1, 2 and 3: HEK^GPR126 cells treated with PBS, GD and FT. Lanes 4, 5 and 6: HEK^GPR124 cells treated with PBS, GD and FT. Lanes 7, 8 and 9: HEK293T cells treated with PBS, GD and FT Asterisks: immunoglobulin heavy and light chains. For uncropped gels see Supplementary Information File 1. B: SW10_ΔGpr126 cells plated at a density of 100’000 cells/well in 6-well plates were transfected with control plasmid (pCDNA3) or plasmids encoding various GPCRs (Gpr126, 124, 176, and 56) bearing C-terminal V5 tags. Only cells transfected with pCGpr126-V5 showed a cAMP response to FT_23-50 48h post transfection. PBS treatment was used for control. C: Intracellular cAMP responses to FT treatment (2 μM, 20’) in SW10 and SW10_ΔGpr126 cells, as well as $SW 10_{Δ Gpr 126}^{huGpr 126}$ cells expressing V5-tagged human Gpr126 (pCGpr126-V5). A significant increase in cAMP was observed in SW10 cells, whereas SW10_ΔGpr126 showed no change. In contrast, $SW 10_{Δ Gpr 126}^{huGpr 126}$ cells showed a significant cAMP increase, indicative of successful complementation. D: SW10 and SW10_ΔGpr126 cells were incubated (20 min) with conditioned medium from HEK^empty or HEK^PrP cells.. HEK^PrP-conditioned medium induced a robust cAMP spike in SW10, but not SW10_ΔGpr126 cells. E: SW10 and SW10_ΔGpr126 cells were grown on coverslips for 24h and exposed to recombinant FT (2 μM, 20 min). Cells were stained with POM2 (detecting FT; red, DAPI-stained nuclei: grey) and antibodies to p75^NGFR (yellow). Deletion of Gpr126 largely suppressed FT binding. Scale bar: 26 μm. F: HEK293(H) cell lines were transfected with plasmids expressing different adhesion GPCRs (Gpr: 97, 133, 64, 56), followed by selection of cells expressing the receptor in presence of geneticin. GPCR expressing cells and HEK^Gpr126 cells were then treated with either FT_23-50 or ${FT}_{23 - 50}^{K \to A}$ for control (FT and C, respectively). Only cells expressing Gpr126 responded to FT_23-50 with a cAMP spike. Interestingly, cells expressing Gpr133 reacted with a decrease in cAMP levels. Data are representative of three biologically independent experiments; statistical significance was evaluated by unpaired Student’s t-test.

Extended Data Figure 5 — A: HEK293T cells were transfected with increasing amounts of human Gpr126 plasmid (2–5μg/well of a 6 well plate). 48h post transfection cells were treated with FT_23-50 or PBS as a control. Increasing amount of Gpr126 cDNA did not result in amplification of the cAMP signal. B: Primary *Prnp*^ZH1/ZH1 cerebellar granule neuron cultures were seeded in 6-well plates at a density of 5x10⁵ cells and treated with FT_23-50, ${FT}_{23 - 50}^{K \to A}$ , or PBS. No alterations in the levels of cAMP were noticed. C: SW10 cells were exposed to conditioned medium from HEK cells that had been transfected with empty vector (HEK^empty) or a PrP^C expression vector (HEK^PrP). HEK^PrP were optionally treated with 100 μM of the TAPI-2 protease inhibitor for 24h before harvesting the medium. TAPI-2 treatment resulted in reduced cAMP induction, suggesting that impaired proteolytic cleavage of the FT from PrP^C resulted in decreased signaling. D: Quantification of FT released into the medium relative to the total amount of PrP^C in lysates by Western blotting. The spent medium of HEK^PrP cells treated with TAPI-2 contained less FT. E: SW10 and SW10_ΔGpr126 cells were transfected with an Egr2-controlled firefly luciferase reporter and treated with recombinant FT (2μM) or PBS (24 hrs). Ordinate: luciferase expression normalized to a renilla luciferase control (n=3; *: p<0.05; t-test). Luciferase activity was observed only in SW10 cells stimulated with FT but not in SW10_ΔGpr126 cells. F: Primary Schwann cells were exposed to recombinant FT (2 μM, 1h) or PBS. Egr2 mRNA expression was measured by quantitative RT-PCR and normalized against a panel of housekeeping genes. For uncropped gels see Supplementary Information File 1. G: SW10_ΔPrP and SW10_ΔGpr126 cells were grown in 6-well plates, exposed to recombinant FT (≤30 min), and analyzed by Western blotting (left). Densitometry (right) showed increased phospho-AKT/AKT ratio in SW10_ΔPrP cells, but not in SW10_ΔGpr126 cells. Data are representative of three biologically independent experiments; statistical significance was evaluated by unpaired Student’s t-test.

Extended Data Figure 6 — A: SW10, SW10_ΔPrP and SW10_ΔGpr126 cells were grown on coverslips and stained with antibodies against Myelin associated glycoprotein (MAG), Myelin oligodendrocyte glycoprotein (MOG), glial fibrillary acidic protein (GFAP) and p75 nerve growth factor receptor (p75NGFR) (left panel, all green; DAPI-stained nuclei: blue). Cells labeled with secondary antibody alone (2° Ab) were used as control to determine unspecific staining. Scale bars: 10μm. Expression in all cell lines was confirmed by western blotting (right panel). Lysate from HEK293T wild-type cells (HEK^WT) was used as control. All proteins except Myelin basic protein (MBP) were expressed in SW10 cells and its derivatives. For uncropped gels see Supplementary Information File 2. B: Western blot (developed with POM2) of HEK 293T cells transfected with expression plasmids for wild-type murine PrP^C or for PrP^C bearing lysine-to-alanine substitutions in the KKRPK and QGSPG motifs (lanes 3 and 4, respectively). The mutations did not affect the biogenesis and processing of PrP^C C: Western blot of the medium collected from the cells shown in panel B. FT fragments bearing the mutations were released into the medium similarly to wild-type FT. For uncropped gels see Supplementary Information File 1. D: SW10_ΔPrP cells were treated with conditioned medium from HEK293T cells transfected with an empty vector (HEK^empty), with PrP^C (HEK^PrP), or with full-length PrP^C versions in which the QGSPG (HEK^PrPΣ*^QGSPG*) or KKRPK (HEK^PrPΣ*^KKRPK*) motifs were substituted (Σ) with alanines. The charge neutralization within the KKRPK motif abrogated the cAMP induction. E: SW10_ΔPrP cells were treated with FT_23-50 (2 μM) or a Col4-derived 21-meric synthetic peptide containing either the GPRGKPG domain or its alanine-substituted variant (AAAGAAG; both 8 μM). Both FT_23-50 and the native Col4 peptide, but not the alanine-substituted peptide (Ala-Col4), induced cAMP. Data are representative of three biologically independent experiments; statistical significance was evaluated by unpaired Student’s t-test.

Extended Data Figure 7 — A: Transmission electron micrographs of 14 month-old *Prnp*^ZH1/ZH1 sciatic nerves (ZH1). Black arrowhead: thinly myelinated axons; white arrowhead: abnormal cytoplasmic Schwann cell protrusions; boxes: loss of axon-Schwann cell interactions; asterisk: initial onion bulb formation. Scale bar: 2 μm in upper left panel; 500 nm in all other panels. **B–C**: Quantification of unmyelinated axons in Remak bundles was performed manually by counting number of axons in the bundles from electron microscopy images (1500x magnification, 10 images per mouse were analyzed and three mice per genotype were used in total). The bundles were further sorted into three categories; <10 axons, 10–20 axons and >20axons/bundle. Comparisons were performed between either BL6 and *Prnp*^ZH3/ZH3 (B) or *Gpr126^fl/fl* (WT) and *Dhh^Cre::Gpr126^fl/fl* (*Gpr126*^ΔSchwann) mice (all mice were 13 month old in age). Both *Prnp*^ZH3/ZH3 and *Gpr126*^ΔSchwann mice showed a similar inclination towards decrease in number of axons per bundle. Statistical significance was established by performing a two-way annova with Bonferroni correction. D: Onion bulb like structures were quantified from electron microscopy images (1500x magnification, 10 images per mouse were analyzed and three mice per genotype were used in total) of either BL6 and *Prnp*^ZH3/ZH3 or *Gpr126^fl/fl* (WT) and *Dhh^Cre::Gpr126^fl/fl* (*Gpr126*^ΔSchwann) mice. These onion bulb like structures were prevalent only in *Prnp*^ZH3/ZH3 and *Gpr126*^ΔSchwann mice, with *Gpr126*^ΔSchwann exhibiting more.

Extended Data Figure 8 — A: Immunofluorescence for myelin basic protein (Mbp; green) in the posterior lateral line nerve of wild-type zebrafish larvae. AcTub: acetylated tubulin (red) labels the axons. Scale bar = 20 μm. B: *gpr126^st49* hypomorphic mutant larvae were treated with vehicle (DMSO) or FT_23-50 (20 μM) at 50–55 hours post fertilization (hpf) and immunostained at 5 days post fertilization (dpf) for myelin basic protein (Mbp, green). AcTub: acetylated tubulin (red) labeling axons. Scale bar = 20 μm. FT_23-50 treatment did not alter Mbp immunofluorescence. C: FT_23-50 or ${FT}_{23 - 50}^{K \to A}$ was intravenously administered to *Prnp*^ZH1/ZH1 and BL6 mice (600 μg/mouse, 20 min). After FT_23-50 injection, cAMP levels in *Prnp*^ZH1/ZH1 mice increased to levels approaching those of BL6 mice. Each dot represents an individual animal. D: cAMP spiked also in hearts of FT_23-50 injected but not in ${FT}_{23 - 50}^{K \to A}$ injected mice. *: p<0.05; **: p<0.01. **C–D:** FT_23-50 or ${FT}_{23 - 50}^{K \to A}$ was injected intravenously into 10–16-week old *Prnp*^ZH1/ZH1 or BL6 mice (600 μg/animal, 20 minutes). cAMP levels in kidneys (E) and brain (F) showed no significant changes. Each dot represented an individual animal; statistical significance was evaluated by unpaired Student’s t-test.

Extended Data Table 1.

Sequences of synthetic peptides used in the present study. The collagen-4 homology domain necessary for cAMP induction is highlighted in yellow.

Name	Sequence
FT_23-50	GWNTGGSRYPGQGSPGGNRYP
FT_23-50-HA	GWNTGGSRYPGQGSPGGNRYPYPYDVPDYA
FT_34-69-HA	PGQGSPGGNRYPPQGGTWGQPKGGGWGQYPYDVPDYA
FT_51-94-HA	PQGGTWGQPHGGGWGQPHGGSWGQPHGGSWGQPHGGGW
FT_83-109-HA	PHGGGWGQGGGTHNQWNKPSKPKTNLKYPYDVPDYA
FT_94-110-HA	GGGTHNQWNKPSKPKTNLKHYPYDVPDYA
FT_94-110	GGGTHNQWNKPSKPKTNLKH
AlaQGSPG	KKRPKPGGWNTGGSRYPGAAAAAGNRYP
AlaKKRPK	AAAPAPGGWNTGGSRYPGQGSPGGNRYP
CoIIV	GPRGKPGVDGYNGSRGDPGYP
CoIIV-Mut	AAAGAAGVDGYNGSRGDPGYP
FT_23-34	GWNTG
SFT_23-34	AAAGAAGGWNTG

Open in a new tab

Supplementary Material

Supp Data Fig 1

Supplementary Information Files 1–2: Uncropped blots from the figures of the manuscript. To generate the final figure panels, images were cropped using Adobe Photoshop and realigned with Adobe Illustrator.

NIHMS867483-supplement-Supp_Data_Fig_1.pdf^{(216.7KB, pdf)}

Supp Data Fig 2

NIHMS867483-supplement-Supp_Data_Fig_2.pdf^{(169.9KB, pdf)}

Acknowledgments

We thank Rita Moos, Petra Schwarz, Cinzia Tiberi, Christine Sturzenegger, Victor Escalante, Mirzet Delic, Charleen Johnson and Zachary Spence for technical help, Juliane Bremer for discussions, Valerie Beck, Marie Angela-Wulf, Karl Frontzek, Tory Johnson, Tracy O’Connor, Agnes Lau, and Claudia Rehwald for help in performing experiments, and Xianhua Piao for providing antibodies to Gpr126. AA is the recipient of an Advanced Grant of the European Research Council (Prion2020), a European Union Framework 7 Grant (NEURINOX), the Swiss National Foundation, the Clinical Research Priority Programs “Small RNAs” and “Human Hemato-Lymphatic Diseases”, SystemsX.ch, and the Novartis Research Foundation. AK was supported by MD/PhD fellowship of the Swiss National Science Foundation. AL is supported by a grant of the Synapsis Foundation. SH is a recipient of SystemsX.ch grant. This work was supported by NIH grants F32 NS087786 to S.C.P. and NS079445 to K.R.M.

Footnotes

Author contributions:

AA, AK, KRM and AL designed experiments, analyzed data and wrote the manuscript. AK generated recombinant proteins, SW10ΔPrP and SW10_GPR126 cell lines, performed cAMP, q-PCR, FACS and luciferase assay. AL performed cAMP assays, IP, western blots, immunofluorescence, and FACS. CD contributed to immunofluorescence, extracted sciatic nerves from mice, performed intravenous injections and performed repetitions of cAMP and FACS assays. AM performed all Schwann-cell Gpr126 mutant mouse experiments and SCP performed all zebrafish experiments, which were analyzed by AM, SCP, KRM, and AAG. RM performed cAMP assays, western blots, and FACS analyses. KA contributed to cAMP assays and western blots. PB performed the experiments in Fig. 1C and repeated cAMP assay for Fig. 5B–D. AS generated Fab3 and Fab71 antibodies. AM, CS, and KA contributed to Figures 2B, 2G, and 2A, respectively. MN generated ZH3 mice and performed electron microscopy. BG and FB generated GPR-overexpressing HEK cell lines. SH supervised generation of recombinant proteins, designed synthetic peptides, and contributed to writing the manuscript. All authors approved of the final version.

Competing Financial Interests: None

References

1.Bremer J, et al. Axonal prion protein is required for peripheral myelin maintenance. Nat Neurosci. 2010;13:310–318. doi: 10.1038/nn.2483. [DOI] [PubMed] [Google Scholar]
2.Paavola KJ, Sidik H, Zuchero JB, Eckart M, Talbot WS. Type IV collagen is an activating ligand for the adhesion G protein-coupled receptor GPR126. Sci Signal. 2014;7:ra76. doi: 10.1126/scisignal.2005347. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Bueler H, et al. Normal development and behaviour of mice lacking the neuronal cell-surface PrP protein. Nature. 1992;356:577–582. doi: 10.1038/356577a0. [DOI] [PubMed] [Google Scholar]
4.Polymenidou M, et al. The POM monoclonals: a comprehensive set of antibodies to non-overlapping prion protein epitopes. PLoS One. 2008;3:e3872. doi: 10.1371/journal.pone.0003872. [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Kuwahara C, et al. Prions prevent neuronal cell-line death. Nature. 1999;400:225–226. doi: 10.1038/22241. [DOI] [PubMed] [Google Scholar]
6.Nuvolone M, et al. Strictly co-isogenic C57BL/6J-Prnp−/− mice: A rigorous resource for prion science. J Exp Med. 2016;213:313–327. doi: 10.1084/jem.20151610. [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Pogoda HM, et al. A genetic screen identifies genes essential for development of myelinated axons in zebrafish. Dev Biol. 2006;298:118–131. doi: 10.1016/j.ydbio.2006.06.021. [DOI] [PubMed] [Google Scholar]
8.Monk KR, et al. A G protein-coupled receptor is essential for Schwann cells to initiate myelination. Science. 2009;325:1402–1405. doi: 10.1126/science.1173474. [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Kasof GM, Gomes BC. Livin, a novel inhibitor of apoptosis protein family member. J Biol Chem. 2001;276:3238–3246. doi: 10.1074/jbc.M003670200. [DOI] [PubMed] [Google Scholar]
10.Zhang Y, et al. An RNA-sequencing transcriptome and splicing database of glia, neurons, and vascular cells of the cerebral cortex. J Neurosci. 2014;34:11929–11947. doi: 10.1523/JNEUROSCI.1860-14.2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Altmeppen HC, et al. The sheddase ADAM10 is a potent modulator of prion disease. Elife. 2015;4 doi: 10.7554/eLife.04260. [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Decker L, et al. Peripheral myelin maintenance is a dynamic process requiring constant Krox20 expression. J Neurosci. 2006;26:9771–9779. doi: 10.1523/JNEUROSCI.0716-06.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Jaegle M, et al. The POU proteins Brn-2 and Oct-6 share important functions in Schwann cell development. Genes Dev. 2003;17:1380–1391. doi: 10.1101/gad.258203. [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Mogha A, et al. Gpr126 functions in Schwann cells to control differentiation and myelination via G-protein activation. J Neurosci. 2013;33:17976–17985. doi: 10.1523/JNEUROSCI.1809-13.2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Liebscher I, et al. A tethered agonist within the ectodomain activates the adhesion G protein-coupled receptors GPR126 and GPR133. Cell Rep. 2014;9:2018–2026. doi: 10.1016/j.celrep.2014.11.036. [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Petersen SC, et al. The adhesion GPCR GPR126 has distinct, domain-dependent functions in Schwann cell development mediated by interaction with laminin-211. Neuron. 2015;85:755–769. doi: 10.1016/j.neuron.2014.12.057. [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Uhlen M, et al. Proteomics. Tissue-based map of the human proteome. Science. 2015;347:1260419. doi: 10.1126/science.1260419. [DOI] [PubMed] [Google Scholar]
18.Ishida C, Okino S, Kitamoto T, Yamada M. Involvement of the peripheral nervous system in human prion diseases including dural graft associated Creutzfeldt-Jakob disease. J Neurol Neurosurg Psychiatry. 2005;76:325–329. doi: 10.1136/jnnp.2003.035154. [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Kou I, et al. Genetic variants in GPR126 are associated with adolescent idiopathic scoliosis. Nat Genet. 2013;45:676–679. doi: 10.1038/ng.2639. [DOI] [PubMed] [Google Scholar]
20.Radovanovic I, et al. Truncated prion protein and Doppel are myelinotoxic in the absence of oligodendrocytic PrPC. J Neurosci. 2005;25:4879–4888. doi: 10.1523/JNEUROSCI.0328-05.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Zahn R, von Schroetter C, Wuthrich K. Human prion proteins expressed in Escherichia coli and purified by high-affinity column refolding. FEBS Lett. 1997;417:400–404. doi: 10.1016/s0014-5793(97)01330-6. [DOI] [PubMed] [Google Scholar]
22.Lysek DA, Wuthrich K. Prion protein interaction with the C-terminal SH3 domain of Grb2 studied using NMR and optical spectroscopy. Biochemistry. 2004;43:10393–10399. doi: 10.1021/bi0494828. [DOI] [PubMed] [Google Scholar]
23.Hornemann S, Christen B, von Schroetter C, Perez DR, Wuthrich K. Prion protein library of recombinant constructs for structural biology. FEBS J. 2009;276:2359–2367. doi: 10.1111/j.1742-4658.2009.06968.x. [DOI] [PubMed] [Google Scholar]
24.Guillot-Sestier MV, Sunyach C, Druon C, Scarzello S, Checler F. The alpha-secretase-derived N-terminal product of cellular prion, N1, displays neuroprotective function in vitro and in vivo. J Biol Chem. 2009;284:35973–35986. doi: 10.1074/jbc.M109.051086. [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Schulz JB, Weller M, Klockgether T. Potassium deprivation-induced apoptosis of cerebellar granule neurons: a sequential requirement for new mRNA and protein synthesis, ICE-like protease activity, and reactive oxygen species. J Neurosci. 1996;16:4696–4706. doi: 10.1523/JNEUROSCI.16-15-04696.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Dametto P, et al. Neurodegeneration and unfolded-protein response in mice expressing a membrane-tethered flexible tail of PrP. PLoS One. 2015;10:e0117412. doi: 10.1371/journal.pone.0117412. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Kimmel CB, Ballard WW, Kimmel SR, Ullmann B, Schilling TF. Stages of embryonic development of the zebrafish. Dev Dyn. 1995;203:253–310. doi: 10.1002/aja.1002030302. [DOI] [PubMed] [Google Scholar]
28.Lyons DA, et al. erbb3 and erbb2 are essential for schwann cell migration and myelination in zebrafish. Curr Biol. 2005;15:513–524. doi: 10.1016/j.cub.2005.02.030. [DOI] [PubMed] [Google Scholar]
29.Schindelin J, et al. Fiji: an open-source platform for biological-image analysis. Nat Methods. 2012;9:676–682. doi: 10.1038/nmeth.2019. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp Data Fig 1

NIHMS867483-supplement-Supp_Data_Fig_1.pdf^{(216.7KB, pdf)}

Supp Data Fig 2

NIHMS867483-supplement-Supp_Data_Fig_2.pdf^{(169.9KB, pdf)}

PERMALINK

The prion protein is an agonistic ligand of the G protein-coupled receptor Gpr126/Adgrg6

Alexander Küffer

Asvin KK Lakkaraju

Amit Mogha

Sarah C Petersen

Kristina Airich

Cédric Doucerain

Rajlakshmi Marpakwar

Pamela Bakirci

Assunta Senatore

Arnaud Monnard

Carmen Schiavi

Mario Nuvolone

Bianka Grosshans

Simone Hornemann

Frederic Bassilana

Kelly R Monk

Adriano Aguzzi

Abstract

Figure 1. Schwann cells selectively bind the flexible tail (FT) of PrPC.

Figure 2. The FT fragment elicits a concentration-dependent cAMP response.

Figure 3. FT-dependent cAMP signaling in Gpr126-ablated Schwann cells.

Figure 4. FT and collagen-IV share a cAMP-inducing domain.

Figure 5. Myelinotrophic effect of FT in zebrafish and mice.

Methods

Mice

Primary Schwann cell culture

Purification of recombinant proteins

Cell Culture

Cerebellar granule neuronal culture

Plasmids and Transfections

Immunoprecipitation

RNA Isolation and quantitative PCR

Zebrafish mutant strains and analysis

Zebrafish peptide treatment, immunostaining, and quantification

Flow cytometry

Western blot analysis

Antibodies

Cyclic AMP measurements

Generation of Schwann cell lines devoid of Gpr126

Promoter luciferase assay

Immunocytochemistry

Transmission electron microscopy

Recombinant peptides

Extended Data

Extended Data Figure 1.

Extended Data Figure 2.

Extended Data Figure 3.

Extended Data Figure 4.

Extended Data Figure 5.

Extended Data Figure 6.

Extended Data Figure 7.

Extended Data Figure 8.

Extended Data Table 1.

Supplementary Material

Acknowledgments

Footnotes

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases

Figure 1. Schwann cells selectively bind the flexible tail (FT) of PrP^C.