Interactions among Merlin, Arkadia, and SKOR2 mediate NF2-associated human Schwann cell proliferation

Abstract

Background: NF2-related Schwannomatosis (previously referred to as Neurofibromatosis Type 2, or NF2) is a genetic-associated disease resulting from mutations in the gene, NF2. NF2 encodes the Merlin protein, which acts as a tumor suppressor. Bilateral vestibular schwannoma (VS) is a hallmark of NF2. Although the exactly molecular mechanism mediating NF2-driven schwannomatosis is not fully understood, it is known that defective Merlin protein functionality leads to abnormal cell proliferation.

Methods: Herein, we utilized a human induced pluripotent stem cell (hiPSC)-based Schwann cell (SC) model to investigate the role of Merlin in human SCs. SCs were derived from hiPSCs carrying a NF2 mutation (c.191 T > C; p. L64P), its isogenic wild-type control cell line, and a NF2 patient-derived hiPSC line. Phenotypes were determined via immunocytochemistry and various bioassays. Different proteins interacting with Merlin in wild-type and NF2 mutation SCs were identified using co-immunoprecipitation followed by mass spectrometry.

Results: SC derived from NF2^L64P hiPSCs showed significantly higher proliferation and abnormal morphology compared to NF2^WT SCs. Phenotypes that could be restored by wildtype NF2 overexpression. Interactome profiling of Merlin (NF2) in SCs derived from NF2^WT- and NF2^L64P- hiPCSs identified differential protein binding levels. Among identified proteins, we validated the interaction among Merlin, an E3 ubiquitin ligase (Arkadia), and a SKI family co-repressor (SKOR2). This complex plays a significant role for this interaction in SC proliferation. Our findings were further validated by SCs derived from the patient-derived hiPSCs carrying a deletion in the chromosome 22 which spans the NF2 gene.

Conclusions: Our results presented a hiPSC-derived SC system for SC-related disease modeling and established a new model in which Merlin interacts with Arkadia and SKOR2. This interaction is required for the proper cell proliferation in human SCs.

Keywords: Human induced pluripotent stem cells; Merlin; NF2; Proteomic analysis; Schwann cells; Schwannoma.

Fig. 1

Differentiation of Schwann cells (SCs) from NF2^WT and NF2^L64P hiPSC lines. A. Schematic of the differentiation protocol. B, C, and D. Quantification of Schwann cell progenitors (B), SCs (C) and neurons (D) induction based on the ratio of marker + cells over DAPI + cells. B’, B”, C’, C”, D’ and D”. Representative immunohistochemistry (IHC) images of markers for SCPs, SCs, and neurons, respectively. Scale bar = 100 μm

Fig. 2

Abnormal phenotypes in NF2^L64P-derived Schwann cells (SCs). A. Higher cell proliferation level in NF2^L64P-derived SCs based on the BrdU ELISA assay. B-C’. Representative images of the cell proliferation marker, Ki67. Scale bar = 100 μm D-E. Representative bright-field images of SCs derived from NF2^WT and NF2^L64P hiPSC lines. Scale bar = 10 μm. F. Significantly larger cell surface area in NF2^L64P-derived SCs. G and G’. Representative IHC images of SCs derived from NF2^WT and NF2^L64P hiPSC lines staining with phalloidin for F-actin. Scale bar = 100 μm

Fig. 3

Restoration of abnormal phenotypes in NF2^L64P-derived Schwann cells (SCs). A-C. Representative images of SCs staining with phalloidin in NF2^WT, NF2^L64P, and NF2^L64P + NF2 conditions. D. The size of cell surface area reduced in NF2^L64P-derived SCs with the overexpression of wildtype NF2 (NF2^L64P + NF2). E. Cell proliferation level in NF2^L64P-derived SCs decreased with the transfection of NF2-expressing plasmid (NF2^L64P + NF2) based on the BrdU assay. Scale bar = 50 μm

Fig. 4

Proteomic analyses of proteins after co-immunoprecipitation (Co-IP) with the Merlin antibody from NF2^WT- and NF2^L64P-derived SCs. (A) Different protein pattern between NF2^WT and NF2^L64P after IP in SDS-PAGE with the Merlin antibody that recognizes both WT and mutant Merlin. (B) Six differentially concentrated proteins (DCPs) were identified via analyzing presence-absence data. (C) Six DCPs were identified in the TMT-based quantification data. (D) TMT-based data showed distinctions in proteomes between samples from NF2^WT and NF2^L64P in the principal component analysis (PCA). (E) Canonical analysis suggested strong effects of the NF2 p.L64P mutation on SC proteome

Fig. 5

Merlin interacts with Arkadia and SKOR2 in the SMAD-dependent pathways in the TGFβ signaling. (A) Western blots of Arkadia and SKOR2 following the Co-IP with the Merlin antibody. (B) Western blots of Merlin and SKOR2 following the Co-IP with the Arkadia antibody. (C) Similar Western blot pattern of Ubiquitin following the Co-IP with the Merlin antibody with the Western blots of SKOR2 in panel A and B. (D) Western blots in the cytoplasmic fraction. (E) Western blots in the nuclear fraction. F-H. Quantification of Arkadia, SKOR2, and pSMAD2/3 in the nuclear fraction with TGFβ activation. I. The SBE assay indicated significantly higher response to the TGFβ activation in NF2^WT-SCs comparing to its in NF2^L64P- derived SCs

Fig. 6

Abnormal cell proliferation in SCs derived from the patient-derived hiPSCs (NF2^+/−). Similar Western blot results among Merlin, Arkadia, SKOR2, pSMAD2/3 in nuclear fractions in responding to the TGFβ activation indicated that heterozygous loss of NF2 results in the higher SKOR2 and the significantly lower SBE activity level. A-A’. Representative IHC images of SOX10 + SCPs derived from NF2^+/− hiPSCs. B-B’ Representative IHC images of S100β + SCs derived from NF2^+/− hiPSCs. C-C’ Representative images of NF2^WT hiPSCs- and NF2^+/−hiPCSs-derived SCs staining with phalloidin. D. Significantly higher proliferation activity in NF2^+/−-derived SCs. E. Western blots in whole lysates. F. Higher level of SKOR2 in NF2^+/−-derived SCs with the TGFβ activation. G. Significantly lower SBE activity in NF2^+/−-derived SCs comparing NF2^WT-derived SCs after the TGFβ activation

Fig. 7

Reduction of SKOR2 protein level decreases the cell proliferation in SCs. (A) The proposed model of interaction among Merlin, Arkadia, and SKOR2 in responding to the SMAD-dependent TGF-β signaling pathway. (B) Representative images of NF2^L64P SCs infected with lentivirus for 72 h. Control: empty vector. SKOR2_KO: vector carrying a shRNA fragment. (C) Representative Western blots. (D) Quantitative ratio of SKOR2/β-actin. (E) Significantly lower cell proliferation level in SKOR2_KO SCs than control SCs based on the BrdU ELISA assay. Scale bad = 100 μm. t-test: p < 0.05