Merlin status regulates p75(NTR) expression and apoptotic signaling in Schwann cells following nerve injury

Abstract

After nerve injury, Schwann cells (SCs) dedifferentiate, proliferate, and support axon regrowth. If axons fail to regenerate, denervated SCs eventually undergo apoptosis due, in part, to increased expression of the low-affinity neurotrophin receptor, p75(NTR). Merlin is the protein product of the NF2 tumor suppressor gene implicated in SC tumorigenesis. Here we explore the contribution of merlin to SC responses to nerve injury. We find that merlin becomes phosphorylated (growth permissive) in SCs following acute axotomy and following gradual neural degeneration in a deafness model, temporally correlated with increased p75(NTR) expression. p75(NTR) levels are elevated in P0SchΔ39-121 transgenic mice that harbor an Nf2 mutation in SCs relative to wild-type mice before axotomy and remain elevated for a longer period of time following injury. Replacement of wild-type, but not phospho-mimetic (S518D), merlin isoforms suppresses p75(NTR) expression in primary human schwannoma cultures which otherwise lack functional merlin. Despite elevated levels of p75(NTR), SC apoptosis following axotomy is blunted in P0SchΔ39-121 mice relative to wild-type mice suggesting that loss of functional merlin contributes to SC resistance to apoptosis. Further, cultured SCs from mice with a tamoxifen-inducible knock-out of Nf2 confirm that SCs lacking functional merlin are less sensitive to p75(NTR)-mediated cell death. Taken together these results point to a model whereby loss of axonal contact following nerve injury results in merlin phosphorylation leading to increased p75(NTR) expression. Further, they demonstrate that merlin facilitates p75(NTR)-mediated apoptosis in SCs helping to explain how neoplastic SCs that lack functional merlin survive long-term in the absence of axonal contact.

Keywords: Axotomy; Cell death; Nerve growth factor; Neurofibromatosis type 2; Receptor; Schwann cells.

Figure 1

Merlin is phosphorylated in Schwann cells following nerve injury. A. Immunoblots of protein lysate from cut and uncut rat sciatic nerves probed with anti-phospho-merlin (p-merlin) and merlin antibodies. B. Average p-merlin/merlin levels based on densitometry based on samples from 3 nerves pooled for each condition and averaged from 4 separate repetitions. Error bars present SEM. *p=0.0165, Student’s unpaired t-test. C. Frozen sections of cut and uncut sciatic nerves were immunolabeled with anti-neurofilament 200 (NF200, red) and anti-p-merlin (@green) antibodies. Nuclei were labeled with DAPI (blue). Right column demonstrates combined image with superimposed staining showing presence of p-merlin confined to the axons in in uncut nerve and increased diffuse p-merlin labeling of denervated SCs following axotomy. Scale bar=10 µm.

Figure 2

Merlin is phosphorylated spiral ganglion Schwann cells after kanamycin-induced hair cell loss. Cochlear frozen sections were labeled with anti-myosin VII, (green), anti-NF200 (red), and phospho-merlin (p-merlin, green) antibodies.

Figure 3

Merlin status regulates p75^NTR expression. A. Protein lysates from cut and uncut sciatic nerves were collected 7,21, and 180 days following unilateral axotomy in wild type and P0SchΔ39-121 mice. Blots were probed with anti-p75^NTRantibody and then stripped and re-probed with anti-β-actin antibodies. B. Average p75^NTR/β-actin levels based on densitometry. Blots are from 3 nerve samples pooled for each condition and averaged from 4 separate repetitions for each time point. Error bars present SEM. *p<0.05 by two-tailed Student t-test.

Figure 4

Merlin regulates p75^NTR expression in cultured Schwann and schwannoma cells. A. SC cultures from RosaCre:Nf2^f/f mice were treated with (Tx +) or without (Tx −) tamoxifen. Protein lysates were probed with anti-p75^NTR and merlin antibodies. The blots were stripped and reprobed with anti-β-actin antibodies. B. Average p75^NTR/β-actin levels based on densitometry based on blots from 3 separate repetitions. Error bars present SEM. *p=0.012 by two-tailed Student t-test. C. Primary VS cultures were transduced with Ad-empty vector, Ad-merlin (wild-type), Ad-merlin S518A (unphosphorylatable), or Ad-merlin S518D (phospho-mimetic). Protein lysates were probed with anti-p75^NTR and merlin antibodies. The blots were stripped and reprobed with anti-β-actin antibodies.

Figure. 5

Lack of functional merlin results in decreased Schwann cell proliferation following axotomy. A. Cut and uncut sciatic nerves from wild type (WT) and P0SchΔ39-121 mice treated with EdU were collected at 7 and 180 days following unilateral axotomy and frozen sections were labeled for EdU using the Click-IT reaction (red). Nuclei are labeled with DAPI (blue). Scale bar=100 µm. B. The number of EdU-positive SC nuclei was scored. Counts represent the mean from 4 animals per group. Error bars present SEM. One way ANOVA with post hoc Holm-Sidak was used to test for significance of differences *p< 0.05, ** p<0.001.

Figure 6

Lack of functional merlin blunts Schwann cell apoptosis following axotomy. A. Cut and uncut sciatic nerves from wild type (WT) and P0SchΔ39-121 mice were collected at 7, 21, and 180 days following unilateral axotomy and frozen sections were labeled with TUNEL (red). Nuclei are labeled with DAPI (blue). Scale bar=100 µm. B. The average number of TUNEL-positive SC nuclei was scored. Counts represent the mean from 4 animals per group. Error bars present SEM. One way ANOVA with post hoc Holm Sidak was used to test significance of differences *p< 0.05. The greatest difference between wild-type and mutant mice occurred at post-axotomy day 7 (p< 0.001).

Figure 7

Merlin is necessary for p75^NTR-mediated Schwann cell apoptosis. A. SC cultures from RosaCre:Nf2^f/f mice were treated with tamoxifen (Tx+) or carrier (Tx−) and subsequently maintained in the presence or absence of proNGF (3nM). Cultures were labeled with TUNEL and the number of TUNEL-positive SC nuclei was scored. B. Average number of TUNEL-positive SC nuclei from 3 separate cultures for each condition. Error bars present SEM. One way ANOVA with post hoc Holm Sidak was used to test for significance of differences. Scale bar=100 µm. C. Protein lysates from cultures were immunoblotted wth anti-cleaved caspase3 antibody. Blots were stripped and reprobed with anti-β-actin antibody.