Notch Signal Mediates the Cross-Interaction between M2 Muscarinic Acetylcholine Receptor and Neuregulin/ErbB Pathway: Effects on Schwann Cell Proliferation

The cross-talk between axon and glial cells during development and in adulthood is mediated by several molecules. Among them are neurotransmitters and their receptors, which are involved in the control of myelinating and non-myelinating glial cell development and physiology. Our previous studies largely demonstrate the functional expression of cholinergic muscarinic receptors in Schwann cells. In particular, the M2 muscarinic receptor subtype, the most abundant cholinergic receptor expressed in Schwann cells, inhibits cell proliferation downregulating proteins expressed in the immature phenotype and triggers promyelinating differentiation genes. In this study, we analysed the in vitro modulation of the Neuregulin-1 (NRG1)/erbB pathway, mediated by the M2 receptor activation, through the selective agonist arecaidine propargyl ester (APE). M2 agonist treatment significantly downregulates NRG1 and erbB receptors expression, both at transcriptional and protein level, and causes the internalization and intracellular accumulation of the erbB2 receptor. Additionally, starting from our previous results concerning the negative modulation of Notch-active fragment NICD by M2 receptor activation, in this work, we clearly demonstrate that the M2 receptor subtype inhibits erbB2 receptors by Notch-1/NICD downregulation. Our data, together with our previous results, demonstrate the existence of a cross-interaction between the M2 receptor and NRG1/erbB pathway-Notch1 mediated, and that it is responsible for the modulation of Schwann cell proliferation/differentiation.


Drug Treatments
Arecaidine Propargyl Ester hydrobromide (APE, Sigma-Aldrich, Milan, Italy) is a preferred agonist of the M2 muscarinic receptor subtype. Its selectivity has been previously determined by pharmacological binding experiments and M2 knockdown in different cell models [20,[25][26][27]. As previously used in other works, APE was used at the final concentration of 100 µM [20,25,28,29]. Pituitary extract (PE, Sigma-Aldrich, Milan, Italy) is derived from the pituitary gland and contains several hormones including glial growth factor (GGF); it was used at a final dilution of 6 µL/mL. NRG1 (Immunological Sciences, Rome, Italy) was used at the final concentration of 50 ng/mL. NRG1 or PE were added 1 h before APE treatment.
All experiments were performed in technical and experimental triplicate.

Cell Viability
SCs were seeded on a 24-well plate at a density of 50 × 10 3 cells/well. The day after, cells were treated as described in "drug treatments"; in the co-treatment, NRG was added 1 h before APE. Cell growth was assessed by colorimetric assay based on 3-(4,5-dimethylthiazol 2-y1)-2,5-diphenyltetrazolium bromide (MTT; Sigma-Aldrich, Milan, Italy). For each well, the optical density (OD) at 570 nm was measured by the GloMax Multi Detection System (Promega, Milan, Italy).

Flow Cytometry Analysis
SCs were treated as described in "Drug treatments", for 16, 24 and 48 h. At the end of the treatment, cells were incubated for 90 min with bromodeoxyuridine (BrdU, Sigma-Aldrich, Milan, Italy) at a final concentration of 45 µM, collected by trypsinization, centrifuged for 10 min at 1,000 rpm and washed three times with PBS. Cells were then fixed in methanol/PBS (1:1; v/v). To identify cells in S phase, DNA content and BrdU incorporation were determined in simultaneous analysis by staining with propidium iodide (PI) and anti-BrdU, respectively. Partial DNA denaturation was performed by incubating the cells in 3 N HCl for 45 min, followed by neutralization with 0.1 M sodium tetraborate.
Flow cytometry analysis was performed with a flow cytometer Coulter Epics XL with 488 nm wavelength excitation and 10 4 events were collected for each sample. Biparametric (DNA content versus BrdU content) analysis was performed using WinMDI 2.7 software.

RT-PCR and qRT-PCR Analysis
Total RNA was extracted using Tri-Reagent (Sigma-Aldrich, Milan, Italy) and digested with DNAseI (Ambion-Life technologies Italia, Monza, Italy). Total RNA was reversetranscribed into cDNA with 1 µg of random Primers (Promega, Milan, Italy) and 200 U of Moloney Murine Leukaemia Virus (M-MLV reverse) transcriptase (Promega, Milan, Italy). Glyceraldehyde-3-phosphate dehydrogenase (Gapdh) was used as the housekeeping gene. qRT-PCR was performed with SYBR Green Mastermix (Promega, Milan, Italy) and specific primers at a final concentration of 200 nM were added at the respective wells and analysed by Thermofisher Quantstudio3 (Waltham, MA, USA). Data were normalized for the housekeeping gene gapdh and the ∆∆Ct method was used to determine the fold changes in the gene expression compared with the control.
The sequences of the primers used were: Reverse 5 -TGATGGCAACAATGTCCACT-3

Protein Extraction and Western Blot
Protein samples were extracted in a Lysis buffer (10 nM Tris, 0.5% NP40, 150 mM NaCl). A sample buffer (4×) was added to the protein samples, and they were heated for 5 min at 100 • C, loaded onto 10% SDS (Sodium dodecyl sulphate) polyacrylamide gel and run at 30 mA using a running buffer (25 mM Tris, 190  To detect β-actin, anti-mouse IgG alkaline phosphatase-conjugated secondary antibody was used. Bands were stained with nitro blue tetrazolium in the presence of 5-bromo-4chloro-3-indolyl-phosphate (NBT-BCIP).
The optical density (OD) of each protein band was analysed with ImageJ software (National Institutes of Health, NIH, 469 Bethesda, MD, USA) and normalized against the OD of the protein reference band.

Cell Infection with Adenovirus Expressing Notch-NICD
SCs were infected with recombinant adenoviruses expressing the constitutively active form of Notch-1 (NICD) and green fluorescent protein (GFP), using previously established protocol [30]. Viruses were used at a multiplicity of infection of 50 MOI. SCs were infected with adeno-GFP or adeno-GFP-NICD for 1 h in a serum free medium. After 1 h, complete media were replaced. The day after, cells were treated with APE for 24 h and then SCs were collected for Western blot analysis.

Data Analysis
Data analyses were performed with GraphPad Prism 8 (Graphpad Software, La Jolla, CA, USA). Data were presented as the average ± standard error of the mean (SEM). Student's t-test or one-way ANOVA analyses with Bonferroni's post-tests were used. A value of p < 0.05 was considered statistically significant: * p < 0.05, ** p < 0.01, *** p < 0.001 and **** p < 0.0001. The densitometric analyses of Western blot and PCR bands were measured by ImageJ software (National Institutes of Health, NIH, 469 Bethesda, MD, USA).

M2 Receptor Stimulation Counteracts SC Proliferation Mediated by PE/NRG1
Previous data have demonstrated that M2 agonist APE was able to arrest rat and human SC proliferation [20,23]. In order to explain the mechanism responsible for this effect, we firstly evaluated the ability of M2 agonist APE to counteract SC proliferation induced by PE/NRG1. PE contains several factors including glial growth factors (GGF/NRG), for this reason it has been largely used in SC cultures to promote cell proliferation [20].
Cytofluorimetric analysis confirmed that PE treatment increased the percentage of the SCs in S phase already after 16 h of treatment ( Figure 1A,B). In fact, SCs maintained in FBS and forskolin, showed the typical cell proliferating profile, with a percentage of S phase of 16.41% ± 3.36%, whereas, after PE exposure, the percentages were 21.15 ± 2.4% after 16 h, 24.95 ± 0.85% after 24 h and 17.13 ± 0.46% after 48 h, supporting the idea that PE has a positive effect on SC proliferation. This increase was counteracted by APE treatment already after 16 h of treatment. After 24 h and 48 h, the M2-mediated decrease of the cells in S phase was more evident (Region R3, Figure 1A). In fact, after 16 h of co-treatment, the percentage of cells in S phase was 15.1 ± 2.45% but, after 48 h, it was strongly reduced to 1.38 ± 0.41%. In order to understand if the effect of PE may be explained by the presence of GGF/NRG1, MTT assays were performed upon NRG1 and APE treatments ( Figure 1C). As already demonstrated, NRG1 treatment increased cell growth, whereas APE treatment reduced significantly the cell growth; interestingly, APE plus NRG1 co-treatment showed the same cell number observed after APE treatment, confirming the inhibitory effect of APE on cell proliferation-NRG1 induced. It was also interesting that SCs treated with PE showed a significant downregulation of the transcript levels of nrg1 type I ( Figure 1D), suggesting that NRG1 present in PE negatively counteracted its autocrine production by SCs.
the same cell number observed after APE treatment, confirming the inhibitory effect of APE on cell proliferation-NRG1 induced. It was also interesting that SCs treated with PE showed a significant downregulation of the transcript levels of nrg1 type I ( Figure 1D), suggesting that NRG1 present in PE negatively counteracted its autocrine production by SCs.

M2 Receptor Activation Downregulates NRG1 Expression
M2 activation, APE-mediated, significantly downregulated nrg1 transcript levels already after 16 h of APE exposure (Figure 2A). Using different primers able to recognize the nrg1 type I isoform, RT-PCR analysis showed a significant downregulation of the nrg1/I transcripts after 24 h and 48 h of treatment ( Figure 2B). The immunocytochemistry analysis demonstrated a progressive reduction in immunopositivity for NRG1 protein in SCs after 24 h and 48 h from APE treatment ( Figure 2C). SC growth (Ctrl vs. APE, *** p < 0.001; n = 3), whereas NRG1 exposure increases cell number vs. NRG1, **** p < 0.0001; n = 3). APE plus NRG1 co-treatment shows a cell growth compara APE treatment (NRG1 vs. APE + NRG1, *** p < 0.001; Ctrl vs. APE+NRG1, ** p < 0.01; n = 3 Representative RT-PCR shows a significant decrease in the transcript levels of nrg1 type I i after 24 h of PE treatment. Gapdh was used as housekeeping gene. The graph reports the av of the OD ± SEM of the bands normalized against the gapdh of three independent experiment vs. PE, * p < 0.05; n = 3).

M2 Receptor Activation Downregulates NRG1 Expression
M2 activation, APE-mediated, significantly downregulated nrg1 transcript leve ready after 16 h of APE exposure (Figure 2A). Using different primers able to reco the nrg1 type I isoform, RT-PCR analysis showed a significant downregulation o nrg1/I transcripts after 24 h and 48 h of treatment ( Figure 2B). The immunocytochem analysis demonstrated a progressive reduction in immunopositivity for NRG1 prot SCs after 24 h and 48 h from APE treatment ( Figure 2C).

M2 Stimulation Alters erbB2 Receptor Expression and Distribution
To respond to NRG signals, Schwann cells express erbB2 and erb3 receptors [6,31]. In order to better understand the ability of the M2 receptor to impair the NRG pathway, the expression of erbB receptors was also evaluated. As shown in Figure 3A, erbB2 receptor transcripts were significantly downregulated already after 16 h of APE treatment; after 24 h, the transcript level was comparable to untreated cells ( Figure 3A). On the other hand, Western blotting analyses confirmed that erbB2 receptors were significantly downregulated after 16 and 24 h of APE treatment ( Figure 3B).
In order to better understand the ability of the M2 receptor to impair the NRG pathway the expression of erbB receptors was also evaluated. As shown in Figure 3A, erbB2 recep tor transcripts were significantly downregulated already after 16 h of APE treatment; afte 24 h, the transcript level was comparable to untreated cells ( Figure 3A). On the other hand Western blotting analyses confirmed that erbB2 receptors were significantly downregu lated after 16 and 24 h of APE treatment ( Figure 3B). The immunocytochemistry analysis had also shown that the erbB2 receptor expres sion was mainly localized on the cell membrane surface in untreated cells ( Figure 4A) whereas after APE treatment, erbB2 appeared mainly accumulated in the perinuclear area ( Figure 4B). The immunolocalization of BIP protein, a typical endoplasmic reticulum (ER protein, suggested that the perinuclear region where erbB2 receptors were localized afte APE treatment may be ER region ( Figure 4C). Moreover, immunostaining had even shown that erbB2 after M2 agonist treatment, co-localized with Lamp-1, marker of lyso some ( Figure 4E), but not in the Golgi, as indicated by the GM-130 marker ( Figure 4F-G) The immunocytochemistry analysis had also shown that the erbB2 receptor expression was mainly localized on the cell membrane surface in untreated cells ( Figure 4A), whereas after APE treatment, erbB2 appeared mainly accumulated in the perinuclear area ( Figure 4B). The immunolocalization of BIP protein, a typical endoplasmic reticulum (ER) protein, suggested that the perinuclear region where erbB2 receptors were localized after APE treatment may be ER region ( Figure 4C). Moreover, immunostaining had even shown that erbB2 after M2 agonist treatment, co-localized with Lamp-1, marker of lysosome ( Figure 4E), but not in the Golgi, as indicated by the GM-130 marker ( Figure 4F-G).

M2 Receptor Modulates erbB2 Expression via Notch-1 Pathway
As mentioned above, Notch-1 signalling is one of the master pathways involved in the regulation of transition from precursor to immature SCs [32]. In our previous paper we demonstrated that the M2 agonist APE did not modulate the expression of full-length Notch-1 but progressively decreased the expression of the active form of Notch (Notch Intracellular Domain; NICD) [29].
In order to evaluate whether the decreased expression of erbB2 was controlled by M2 receptors directly or via the Notch-1 pathway, infection with recombinant adenovirus expressing the construct GFP-NICD was performed. The expression of the GFP signal indicated the percentage of cells infected ( Figure 5A).

M2 Receptor Modulates erbB2 Expression via Notch-1 Pathway
As mentioned above, Notch-1 signalling is one of the master pathways involved the regulation of transition from precursor to immature SCs [32]. In our previous pa we demonstrated that the M2 agonist APE did not modulate the expression of full-len Notch-1 but progressively decreased the expression of the active form of Notch (No Intracellular Domain; NICD) [29].
In order to evaluate whether the decreased expression of erbB2 was controlled by receptors directly or via the Notch-1 pathway, infection with recombinant adenovirus pressing the construct GFP-NICD was performed. The expression of the GFP signal in cated the percentage of cells infected ( Figure 5A).
In the cells infected with the control vector adeno-GFP, APE treatment induce reduction in erbB2 protein expression, similar to that observed in not infected cells (Fig  3B and 5B). On the contrary, when SCs were infected with the adeno-NICD-GFP, any v iation in erbB2 protein expression was observed after 24 h of APE treatment ( Figure 5 suggesting that the erbB2 expression was directly controlled by the NICD fragment. In the cells infected with the control vector adeno-GFP, APE treatment induced a reduction in erbB2 protein expression, similar to that observed in not infected cells (Figures 3B and 5B). On the contrary, when SCs were infected with the adeno-NICD-GFP, any variation in erbB2 protein expression was observed after 24 h of APE treatment ( Figure 5B), suggesting that the erbB2 expression was directly controlled by the NICD fragment.

M2 Receptor Activation Negatively Controls the Expression of erbB3 Receptor
The effect of M2 receptor stimulation on erbB3 receptor expression was also evaluated. As shown in Figure 6A, erbB3 receptor transcript levels were significantly upregulated already after 16 h and up to 24 h of APE exposure, but its protein level was significantly downregulated after 16 and 24 h of APE treatment ( Figure 6B). FACS analysis for erbB3 receptor, reported in Figure 6C,D, confirmed a significant increase in the erbB3 positive cells after PE treatment ( Figure 6D, 85.5 ± 1.7%), whereas the number of positive cells significantly decreased when APE was added to the SC cultures ( Figure 6D, 28 ± 1.32%).

M2 Receptor Activation Negatively Controls the Expression of erbB3 Receptor
The effect of M2 receptor stimulation on erbB3 receptor expression was also evaluated. As shown in Figure 6A, erbB3 receptor transcript levels were significantly upregulated already after 16 h and up to 24 h of APE exposure, but its protein level was significantly downregulated after 16 and 24 h of APE treatment ( Figure 6B). FACS analysis for erbB3 receptor, reported in Figure 6C and D, confirmed a significant increase in the erbB3 positive cells after PE treatment ( Figure 6D, 85.5 ± 1.7%), whereas the number of positive

Discussion
The role of neurotransmitters in the control of neuron-glia interactions and in the modulation of glial cell development and physiology is increasingly emerging [16,20,29,33,34]. ACh, released along cholinergic axons [35], contributes to the regulation of SC proliferation to address rat and human SC differentiation towards a myelinating phenotype, increasing the expression of the transcription factor Egr2/Krox20 and myelin proteins (i.e., MBP and P0) [20,23,29].
In this work, we explored whether the effects produced by M2 receptor may be correlated with the NRG1/erbB2-3 pathway, considering the role of this growth factor in the control of SC proliferation and differentiation [6,10,31,36,37].

Discussion
The role of neurotransmitters in the control of neuron-glia interactions and in the modulation of glial cell development and physiology is increasingly emerging [16,20,29,33,34]. ACh, released along cholinergic axons [35], contributes to the regulation of SC proliferation to address rat and human SC differentiation towards a myelinating phenotype, increasing the expression of the transcription factor Egr2/Krox20 and myelin proteins (i.e., MBP and P0) [20,23,29].
In this work, we explored whether the effects produced by M2 receptor may be correlated with the NRG1/erbB2-3 pathway, considering the role of this growth factor in the control of SC proliferation and differentiation [6,10,31,36,37].
The FACS analysis and the MTT assay performed in the presence of PE or NRG1/I have confirmed the role of GGF/NRG1 in the positive control of SC proliferation. However, when the M2 agonist APE was added to the SC culture medium, the SC proliferative rate showed a significant reduction, indicating the ability of the M2 agonist to counteract the GGF/NRG1 effects also when the exogenous NRG or PE were provided. Moreover, when the PE was supplied to the cells, the SC ability to synthesize NRG was dramatically reduced. It is relevant to note that GGF in the pituitary extract is NRG1 type II; it may play the same regulatory role as NRG1 type III via its EGF domain in the downregulating autocrine secretion of NRG1 type I in SCs. Similarly, it appears relevant that M2 agonist APE is also able to reduce the NRG autocrine production, downregulating the NRG1 expression in SCs, both as transcript and protein.
The MTT results, however, suggest that SCs, after APE treatment, are not able to respond to NRG1 exposure, implying a possible effect of M2 agonist also on NRG receptors. The data obtained clearly demonstrate the ability of APE to negatively modulate the expression of both erbB2 and erbB3 protein expression. In particular, the erbB2 receptor transcript levels were drastically reduced after 16 h of treatment. Conversely, APE stimulation produced a progressive increase in the erbB3 and erbB2 receptor transcripts after 24 h of expression. This may suggest that the decreased expression of erbB proteins may be also due to post-transcriptional control-M2 receptor mediated, possibly regulated by some miRNA activation. However, it is not possible to exclude that the increased levels of erbB3 or erbB2 transcripts may be a compensatory effect due to the decreased expression of the respective proteins.
Immunocytochemistry analysis has shown an increased expression of erbB2 at the level of the perinuclear region after M2 stimulation, compared to untreated cells. BIP staining, a common ER marker, has shown a positivity for this marker on the same area, where erbB2 localization has been observed after M2 agonist treatment. Moreover, an increased erbB2 localization at the level of the lysosomes, as detectable by the Lamp1 co-immunostaining. Although further experiments are needed to fully address the ErbB2 localization after M2 receptor stimulation, we could assume that M2 receptor activation causes a significant decreased expression of the receptor and alters its localization on the plasma-membrane. In all cases, according to the previous observations, it is clear that after M2 receptor stimulation, SCs lose the ability to respond to NRG, with a consequent reduction in SC proliferation.
Several papers described how, in the PNS, Notch-Neuregulin1 pathways collaborate to SC survival and proliferation. Notch-1 promotes SC transition from precursors to an immature SC phenotype [32], but it also works as a negative regulator of myelination [36]. As a matter of fact, overexpression of the active intracellular domain of Notch1 (NICD) delays myelination, while Notch-1 inactivation accelerates it [32,38]. In our previous work, the analysis of Notch-1 and the active form NICD expression in cultured SCs after APE treatment showed that, although the expression of full-length Notch-1 was not modulated, NICD was progressively reduced in APE-treated cells in a dose dependent manner [29]. Considering that the M2 agonist is able to downregulate the expression of the erbB2 receptor at a transcriptional and protein level, we tried to understand if these two effects were related to each other. The infection of SCs with adenovirus expressing NICD-GFP has allowed us to demonstrate that M2 agonist treatment was not able to downregulate erbB2 expression in NICD-GFP infected cells, whereas a significant downregulation was observed in not infected SCs and in SCs infected with adenovirus containing an GFP-empty construct. This result suggests that the levels of erB2 are indirectly controlled by M2 selective activation, through NICD downregulation.

Conclusions
M2 muscarinic receptor activation, through the selective agonist APE, is able to decrease the expression of erbB receptors on the plasma-membrane of the SCs, preventing NRG1 binding. This result, supported by immunocytochemistry and FACS analysis studies, suggests that the M2 receptor is able to alter the formation of the erbB2/3 complex. Moreover, a cross-interaction between M2 receptor-Notch-1 (NICD) and NRG/erbB pathways was also defined; in fact, M2 receptor activation, downregulating Notch active form NICD, indirectly inhibits the erbB2 receptor expression.
These data, in agreement with the results obtained from our previous studies [16,20,23,29], suggest that the cholinergic stimulus mediated by the M2 muscarinic receptor, may be used by the SCs as a signal to counteract cell proliferation by switching off the Notch-NRG pathways.