The NF2 tumor suppressor merlin interacts with Ras and RasGAP, which may modulate Ras signaling

Abstract

Inactivation of the tumor suppressor NF2/merlin underlies neurofibromatosis type 2 (NF2) and some sporadic tumors. Previous studies have established that merlin mediates contact inhibition of proliferation; however, the exact mechanisms remain obscure and multiple pathways have been implicated. We have previously reported that merlin inhibits Ras and Rac activity during contact inhibition, but how merlin regulates Ras activity has remained elusive. Here we demonstrate that merlin can directly interact with both Ras and p120RasGAP (also named RasGAP). While merlin does not increase the catalytic activity of RasGAP, the interactions with Ras and RasGAP may fine-tune Ras signaling. In vivo, loss of RasGAP in Schwann cells, unlike the loss of merlin, failed to promote tumorigenic growth in an orthotopic model. Therefore, modulation of Ras signaling through RasGAP likely contributes to, but is not sufficient to account for, merlin's tumor suppressor activity. Our study provides new insight into the mechanisms of merlin-dependent Ras regulation and may have additional implications for merlin-dependent regulation of other small GTPases.

Fig. 1

Nf2^−/− mSCs exhibit elevated Ras activity and depend on Ras activity for focus formation. a, b Ras activity in Nf2^+/+ and Nf2^−/− mSCs was analyzed by GST-Raf1 RBD pulldown followed by western blotting. Spliced lanes are from the same exposure of the same blot. a Cells were cultured at low (L) or high (H) cell density before lysis. b Cells were growth factors starved overnight and lysed following stimulation with 10 ng/ml PDGF-BB for 2 min, or without stimulation. c, d Nf2^−/− mSCs were transduced with lentiviral vectors encoding the indicated transgenes and selected with blasticidin to eliminate nontransduced cells. Cell density (monolayer) and foci were photographed 5 and 6 days after transduction, respectively (c); transgene expression was verified by western blotting (d). PD: pulldown; VC: vector control. See also Fig. S1. Representative data from one out of two independent experiments are shown for all panels

Fig. 2

Merlin can directly interact with Ras. a Domain organization of merlin isoform 1 and 2 (iso1/2) and GST-fusion fragments used for pulldown. Note that aa 1–579 are identical in iso1 and 2. b Western blot analysis of GST-merlin fragments pulldown of HRas 1–166 preloaded with GDP (D) or GTPγS (T). GST-Raf1 RBD was included for comparison. All proteins were purified from E. coli. GST-fusion protein inputs were stained with Ponceau S. c Western blot analysis of GST-merlin fragments pulldown of KRas4B overexpressed in 293 cells by transient transfection. TurboGFP (tGFP) transfected sample shows endogenous total Ras, much lower than KRas4B. Representative data from one out of two independent experiments are shown. d Varying amounts of constructs for proximity biotinylation (with an N-terminal OLLAS tag) were transfected into 293 cells; ~24 h later, media was replenished with 0.1 µM biotin; ~48 h after transfection, cells were lysed for streptavidin (SA) pulldown, analyzed by western blotting. Note that Ras and OLLAS were sequentially probed without stripping, taking advantage of different host species for the antibodies. Prior signals were inactivated by H₂O₂ treatment before the second round of probing. See also Fig. S3

Fig. 3

Merlin interacts with RasGAP. a Western blot analysis of Co-IP of RasGAP with merlin. b Western blot analysis of GST-merlin pulldown of RasGAP. Note that GST-merlin S518A and lysate from RT4/Tet-NF2 S518D cells were used here. Spliced lanes are from the same exposure of the same blot. GST-fusion protein inputs were stained with Ponceau S. Asterisks indicate the correct band. c Bimolecular fluorescence complementation (BIFC) analysis of merlin and RasGAP interaction. Constructs fused with a split cerulean protein (Cerf or Cerl) were cotransfected with pEYFP-C1 (transfection control) into 293T cells. Approximately 48 h later, cells were fixed and stained with wheat germ agglutinin (WGA)-Texas Red-X to outline outside membranes. CFP/YFP: cyan/yellow fluorescent protein. Representative data from one out of at least two independent experiments are shown for all panels

Fig. 4

Merlin FERM and tail interact with the PH-C2 domains of RasGAP. a Domain organization of merlin and GST-fusion fragments used for pulldown. b, c Western blot analysis of GST-merlin fragments pulldown of RasGAP from Sc4 cell lysate (b) or full-length (FL) RasGAP purified from E. coli (c). d Domain organization of RasGAP and GST-fusion fragments used for pulldown. The locations of two caspase cleavage sites [57] and a patient-derived mutation C540Y are depicted. e, f Western blot analysis of GST-RasGAP fragments pulldown of merlin from indicated cell lysates. GST-fusion protein inputs were stained with Ponceau S. Asterisks indicate correct bands. Representative data from one out of at least two independent experiments are shown for b and f

Fig. 5

NF2 patient-derived mutations impair merlin:RasGAP interaction in vivo. a BiFC analysis of merlin and RasGAP interaction (WT and mutants). RasGAP ∆PH-C2 served as a binding-deficient control. See also legend for Fig. 3c. Note that the BiFC signals from merlin mutants were not only weak, but also relocated from the membrane to the cytoplasm. b Expression of BiFC constructs in a parallel transfection experiment validated by western blotting. The polyclonal GFP antibody detects both fragments of the split cerulean (Cerf/Cerl) and EYFP. Representative data from one out of three independent experiments are shown for a and b

Fig. 6

Merlin cannot increase RasGAP activity in vitro, but suppresses PDGF-BB activated Ras signaling in vivo. a RasGAP fragments used for in vitro GAP activity experiments. b, c Equal molar amounts of RasGAP fragments were preincubated with or without equal molar amounts of merlin (FL in b, aa 1–313 in c) for 1 h at room temperature and then used to catalyze the hydrolysis of GTP-loaded HRas 1–166. BSA was used as a noncatalyzing control. Remaining Ras-GTP was detected by GST-Raf1 RBD pulldown followed by western blotting. All proteins were purified from E. coli. See also Fig. S8a, b. d Cells were seeded and cultured for ~20 h (close to confluence), then growth factor starved overnight, then stimulated with 10 ng/ml PDGF-BB and lysed at indicated time points for Ras-GTP pulldown by GST-Braf RBD, analyzed by western blotting. Note that p-MEK, p-Erk, MEK, and Erk were sequentially probed without stripping, taking advantage of different host species for phosphorylation-specific and pan antibodies. Phosphorylation-specific signals were inactivated by H₂O₂ treatment before probing MEK. Note that uneven Erk signals were due to steric interference from previously bound antibodies (p-Erk signals were very strong before H₂O₂ treatment). Representative data from one out of two (b, c) or three (d) independent experiments are shown

Fig. 7

Comparison of RasGAP loss with merlin loss in tumorigenicity and detachment-induced signaling. a, b Bioluminescence imaging of orthotopic tumors formed by mSCs/Luc derivatives. Cells were injected into sciatic nerves of NOD/SCID mice (at one side) and tumor growth was monitored by measuring luciferase signals weekly for a total of 3 weeks. Endpoint data (images of representative animals, luciferase total flux from each animal, and the mean) are shown. **P ≤ 0.01; ns, nonsignificant; ordinary one-way ANOVA followed by Tukey’s multiple comparisons test. See also Fig. S9c. c Western blot analysis of signaling profiles of mSC derivatives under a detaching condition. Cells were treated with PBS/5 mM EDTA, harvested at indicated time points. Stripping was performed between sequential probing of phosphorylated and total protein level of the same target. Representative data from one out of two independent experiments are shown. d The model wherein the presence of merlin modulates Ras signaling. The GAP domain consists of two parts: GAP_extra and GAP_core [58]; the C2-GAP domains constitute the catalytic unit promoting Ras-GTP hydrolysis into Ras-GDP. The SH2-SH3-SH2 domains mediate RasGAP translocation from the cytoplasma to the plasma membrane through binding the consensus phosphotyrosine in various activated receptor tyrosine kinases (RTKs). Merlin FERM and tail interact with the PH-C2 domains of RasGAP, whereas the FERM also interacts with Ras (GTP-independent). Merlin may recruit RasGAP in concert with activated RTKs, or independently, to fine-tune Ras-GTP inactivation. Note that merlin might not simultaneously bind to both Ras and RasGAP. GF: growth factor