Erbin Inhibits RAF Activation by Disrupting the Sur-8-Ras-Raf Complex*

Erbin is a member of the LAP (leucine-rich repeat (LRR) and PDZ domain) family. It inhibits Ras-mediated activation of ERK in response to growth factors. In this study, we investigated the mechanisms by which Erbin regulates the Ras-Raf-MEK pathway. The N-terminal LRR domain was necessary and sufficient to inhibit neuregulin-activated expression of ϵ416-Luc, a reporter of ERK activation. On the other hand, Erbin had no effect on Ras activation, but it attenuated neuregulin-induced Raf activation, suggesting that Erbin may regulate Raf activation by Ras. Via the LRR domain, Erbin interacts with Sur-8, a scaffold protein necessary for the Ras-Raf complex. Expression of Erbin attenuated the interaction of Sur-8 with active Ras and Raf. Moreover, Erbin-shRNA, which suppressed Erbin expression at mRNA and protein levels, increased the interaction of Sur-8 with Ras and Raf, ERK activation, and neuregulin-induced expression of endogenous acetylcholine receptor ϵ-subunit mRNA. These results demonstrate a regulatory role of Erbin in the Ras-Raf-MEK pathway, suggesting that Erbin may inhibit ERK activation by disrupting the Sur-8-Ras/Raf interaction.

Neuregulin is a family of epidermal growth factor-containing polypeptides implicated in regulating neuron migration, neurite outgrowth, and expression of neurotransmitter receptors (1,2). Neuregulins act by activating their receptors, the 180-kDa ErbB receptor tyrosine kinases ErbB2, ErbB3, and ErbB4 (3,4). Upon activation, tyrosine residues in the carboxyl termini become phosphorylated and serve as docking sites for cytoplasmic signaling molecules such as Shc and Grb2 to activate various downstream signaling pathways. Grb2 brings guanyl nucleotide exchange factor (SOS) to the plasma membrane in proximity with Ras and expedites the exchange of GDP for GTP on Ras (5). Activated Ras (GTP-bound) then directly binds to Raf and allows the latter to be activated (6,7). Active Raf triggers sequential activation of MEK, 2 a MAPK kinase, and ERK (8).
In this study, we investigated mechanisms of Erbin in inhibiting ERK activation. Erbin bound to Sur-8 and inhibited the interaction of Sur-8 with Ras and Raf. Suppression of Erbin expression in cells increased the interaction of Sur-8 with active Ras and Raf, ERK activation, and neuregulin-induced expression of endogenous AChR ⑀-subunit mRNA. Taken together, these results demonstrate a regulatory role of Erbin in the Ras-Raf-MEK pathway, suggesting that Erbin may inhibit ERK activation by disrupting the Sur-8-Ras-Raf complex.
Immunoprecipitation and Immunoblotting-Cell lysates (ϳ400 g of protein) were incubated without or with indicated antibodies 1 h at 4°C and subsequently with protein A-or protein G-agarose beads overnight at 4°C on a rotating platform. After centrifugation, beads were washed five times with the modified radioimmune precipitation assay buffer. Bound proteins were eluted with the SDS sample buffer, resolved by SDS-PAGE, and transferred onto nitrocellulose membranes (Schleicher and Schuell). Nitrocellulose membranes were incubated at room temperature for 1 h in blocking buffer containing Tris-buffered saline with 0.1% Tween (TBS-T) and 5% milk or 5% bovine serum albumin, followed by incubation with indicated antibodies in the blocking buffer. After being washed three times for 5 min each with TBS-T, the membrane was incubated with horseradish peroxidase-conjugated donkey anti-mouse or anti-rabbit IgG (Amersham Biosciences) followed by washing. Immunoreactive bands were visualized with enhanced chemi-luminescence substrate (Pierce). In some experiments, the nitrocellulose filter was incubated in a buffer containing 62.5 mM Tris-HCl, pH 6.7, 100 mM ␤-mercaptoethanol, and 2% SDS at 50°C for 30 min and then washed with 0.1% Tween 20 in 50 mM Tris-buffered saline at room temperature for 1 h and reblotted with different antibodies. The following antibodies were used: FLAG (M2, Sigma), Myc (9E10, Santa Cruz Biotechnology), phospho-Raf-1 (Ser-338, Upstate), phospho-MAPK (Promega), anti-Ras (clone RAS10, Upstate), GST (Biocompare), and Erbin (28). For quantitative analysis, autoradiographic films were scanned with Epson Expression 1680, and the captured image was analyzed with NIH Image software.
Luciferase Assay-C2C12 myoblasts were cotransfected with or without Myc-Erbin, plus the ⑀-subunit promoter-luciferase transgene that contains 416 nucleotides of the 5Ј-untranslated region of the ⑀-subunit gene (30). 24 h after transfection, myoblasts were incubated in the differentiation medium to induce myotube formation. Myotube formation was completed 48 h after the switch to differentiation medium. C2C12 myotubes were stimulated with neuregulin at a final concentration of 10 nM at 37°C for 24 h. pRL-SV40, which expresses Renilla luciferase under the control of the SV40 promoter (Promega), was cotransfected as a control to monitor the transfection efficiency. 48 h after transfection, cells were lysed, and the activities of the two different luciferases were assayed with their respective substrates with a dual luciferase assay kit (Promega).
Reverse Transcription-PCR-Total RNA was extracted by TRIzol (Invitrogen) and first-strand cDNA prepared using the SuperScript III First-strand Synthesis Kit (Invitrogen) according to the manufacturer's instruction. PCR parameters were 25 cycles (94°C for 30 s, 55°C for 30 s, 72°C for 30 s) followed by a final extension cycle at 72°C for 7 min. PCR product was resolved on 6% acrylamide gels and visualized by ethidium bromide. Primers were as follows: 5Ј-TGA TGC TGA AAG TGG CCC ACC AGC C-3Ј and 5Ј-TGA AGA AAC TTC TCG TAC AAT GAT G-3Ј for Erbin; 5Ј-CCA TGT CCC CGC GGC TGC GC-3Ј and 5Ј-GAG CCC ACG CTG AAG AGC AC-3Ј for the AChR ⑀-subunit; and Inhibition of Erbin Expression by Small Hairpin RNA (shRNA)-shRNA virus was generated by the BLOCK-iT Lentiviral RNAi Expression System (Invitrogen) according to the manufacturer's instruction. Briefly, analysis of the Erbin sequence by a program provided by Invitrogen, three sequences were picked and cloned into the pENTR/U6 entry construct to yield pENTR/U6-Erbin. They were transfected in HEK 293 cells for their ability to suppress endogenous Erbin expression assayed by Western blot. Two clones, 2583 and 4049, were found effective in suppressing Erbin expression and were used for viral production. 2583 encodes 5Ј-CAC CGC TGA TGA CAC TCA CAA ATT ACG AAT AAT TTG TGA GTG TCA TCA GC-3Ј, whereas 4049 encodes 5Ј-CAC CGC ATC CCT CTA GAG AAC AAC TCG AAA GTT GTT CTC TAG AGG GAT GC-3Ј. pLenti6-Erbin was generated by left-right recombination between the pENTR/U6 entry construct and pLenti6/ BLOCK-iT-DEST. 293FT producer cells were cotransfected with pLenti6-Erbin and a packaging mix to produce lentivirus.

RESULTS AND DISCUSSION
Erbin Inhibits Activation of Raf but Not Ras-In an earlier report, we showed that Erbin inhibits ERK activation (27). The site of action was mapped to be between Raf and Ras using phospho-ERK as the readout, because Erbin inhibits ERK activation by active Ras but not by active Raf.
Raf is a serine-threonine kinase that phosphorylates and activates the dual specificity kinase, MEK (MAPKK) (33) which, in turn, phosphorylates and activates ERK. Although these observations suggest that Erbin may inhibit Raf activation, it is possible that Ras activation may also be a target of regulation. To address these questions, we studied the effects of Erbin on Ras and Raf activation in HEK 293 cells. Tagged Ras and Raf were cotransfected with Erbin or LRR. After stimulation with neuregulin, active Ras was purified by the Ras-binding domain of Raf immobilized on beads. The RBD binds to active Ras with an affinity 3 orders of magnitude higher than inactive GDP-bound Ras and has been used widely to purify active Ras (31,32,34). Purified active Ras was revealed by immunoblotting with anti-Ras antibody. As shown in Fig. 1, A and B, active Ras was increased by neuregulin. However, the increase was not altered by Erbin or LRR, suggesting that Erbin may have little effect on Ras activation.
The LRR Domain Is Necessary and Sufficient to Inhibit ⑀416-Luc Expression-Erbin has three domains: the N-terminal LRR domain, an LAP-specific domain that is C-terminal to the LRR, and the C-terminal PDZ domain (13). The LRR domain has been shown to be sufficient to inhibit ERK activation (27). To exclude the possible involvement of other domains, it was important to determine whether the LRR domain is required for the inhibitory effect. To this end, we characterized the effects of ⌬LRR, a deletion mutant without the LRR domain, on neuregulin-induced expression of ⑀416-Luc (Fig. 2B). ⑀416-Luc is a reporter transgene containing a 416-bp 5Ј-flanking region of the AChR ⑀-subunit gene and the luciferase gene, which responds to neuregulin stimulation in skeletal muscle cells (30,32,35). Expression of constitutively active Ras (V12), Raf (BXB), or Mek1 (ddMek1) increased the promoter activity of ⑀416-Luc. Expression of constitutively active Ras and Raf increases expression of the ⑀416 transgenes (30). As shown in Fig. 2A, co-expression of Erbin suppressed RasV12-induced activation of ⑀416-Luc. In contrast, however, Erbin appeared to be unable to inhibit RafBXB-or ddMek1-induced ⑀416-Luc expression. These results confirm our early report that Erbin may function downstream or at the level of Ras and, at the same time, validate the reporter for a reliable readout. Erbin, LRR, or ⌬LRR was cotransfected with ⑀416-Luc in C2C12 myoblasts, and the resulting myotubes were stimulated with neuregulin and assayed for luciferase activity as described under "Experimental Procedures." Although Erbin and LRR attenuated neuregulin-induced expression of ⑀416-Luc, the deletion of LRR (⌬LRR) prevented Erbin from inhibiting neuregulin-induced expression. These results suggest that the LRR domain is necessary and sufficient for inhibition of ⑀416-Luc expression.
Erbin Interaction with Sur-8-Sur-8 is believed to be a positive regulator of the Ras-Raf-MEK pathway, functioning as a scaffold that enhances ERK activation by facilitating the interaction between Ras and Raf (33,36). Sur-8 is composed almost entirely of 18 LRRs that show homology to the LRR domain of Erbin. Interestingly, like Erbin, Sur-8 interacts with active RasV12 but not with inactive RasN17 (36). We reasoned that Erbin may interact with Sur-8 and thus disrupt its interaction with Ras to inhibit ERK activation. First we tested whether Erbin

. The LRR domain is required for Erbin inhibition of neuregulin induction of ⑀416-Luc.
A, Erbin inhibition of ⑀416-Luc expression by active Ras but not active Raf or MEK1. C2C12 myoblasts were transfected with ⑀416-Luc with various amounts of expression constructs of active Ras, Raf, or MEK1. A Renilla luciferase plasmid, pRL-TK, was cotransfected as a control to monitor the transfection efficiency and sample handling. After differentiation, cells were lysed, and firefly and Renilla luciferase activities were assayed. Cells transfected with an empty vector were taken as 100%. Shown is a representative experiment in duplicates, which was repeated three times with similar results. B, the LRR domain is necessary and sufficient to inhibit neuregulin-induced ⑀416-Luc expression. C2C12 cells were transfected with ⑀416-Luc, pRL-TK, and the indicated Erbin constructs (structures are shown below the histogram). Luciferase activities were assayed as in A. *, p Ͻ 0.01.  and Sur-8 interact in cells. GST-Sur-8 and Myc-Erbin were cotransfected into HEK 293 cells. Cell lysates were incubated with glutathioneconjugated agarose beads to pull down Sur-8, and the resulting precipitates were blotted with anti-Myc antibody. As shown in the top panels of Fig. 3A, Myc-Erbin was detected in the precipitates of GST-Sur-8. In a reciprocal experiment, GST-Sur-8 was present in immunoprecipitates of Myc-Erbin (Fig. 3A, bottom panels). These results suggest that Erbin interacts with Sur-8 in cells. Further analyses showed that Erbin interaction with Sur-8 requires the LRR domain. Deletion of this domain prevented Erbin from interacting with Sur-8 (Fig. 3B). To determine whether the interaction occurs between the two proteins at endogenous levels, HEK 293 cell lysates were incubated with anti-Sur-8 serum, and the resulting precipitates were probed with anti-Erbin antibodies (Fig.   3C). Erbin was detected in Sur-8 precipitates, suggesting that endogenous Erbin and Sur-8 may interact in cells.
Disruption of the Sur-8-Ras and -Raf Interaction by Erbin-To determine whether Erbin regulates the Sur-8-Ras interaction, HEK 293 cells were transfected with GST-Sur-8, FLAG-RasV12, and Myc-Erbin. Cell lysates were incubated with glutathione-conjugated agarose beads to pull down Sur-8, and the resulting precipitates were blotted for RasV12 with anti-FLAG antibody. Sur-8 and RasV12 co-precipitated (Fig. 4A), in agreement with earlier studies (36). However, co-expression blocked the interaction of Sur-8 with RasV12. This effect was dose-dependent (Fig. 4B). The amount of RasV12 interacting with Sur-8 decreased as the concentration of Erbin increased. These results demonstrated that Erbin may be able to disrupt the Sur-8-Ras complex. In addition, Erbin expression appeared to FIGURE 5. Enhanced interaction of Sur-8 with Ras and Raf when Erbin expression was suppressed. A, inhibition of Erbin expression by Erbin-shRNA constructs. HEK 293 cells were transfected with LacZ-shRNA or Erbin-shRNA 2583 and 4049. The resulting lysates were analyzed for Erbin expression by Western blot with ␤-actin as a control. IB, immunoblot. B, Erbin-shRNA specifically suppressed expression of Erbin but not Densin-180. HEK 293 cells were transfected Erbin-shRNA 4049 alone or with pRK5-Erbin or pEF6-Densin-180. Expression of Erbin was analyzed by antibodies against Erbin (lanes 1 and 2) or Myc (lanes 3-6) with ␤-actin as a control. C, Erbin-shRNA increased Sur-8 interaction with Ras and Raf. HEK 293 cells were transfected with LacZ-shRNA or Erbin-shRNA 4049. Cell lysates were subjected to immunoprecipitation with antibodies against Ras and Raf, and the resulting complexes were probed with anti-Sur-8 antibody. Nontransfected cell lysates were also incubated with anti-Myc antibodies as control. Expression of Erbin and actin is shown in the bottom panels. inhibit the interaction between Sur-8 and Raf (Fig. 4C). These observations are in agreement with the notion that Erbin inhibits ERK activation by disrupting Sur-8 interaction with Ras and Raf.
Increased Sur-8-Ras and -Raf Interaction in Cells in Which Expression of Erbin Was Suppressed-To further study the role of Erbin in regulating ERK activation, we explored the consequences of suppression of Erbin expression by using a small interfering RNA approach, which diminishes the expression of a specific gene in cells. RNA interference has been shown recently to specifically suppress the expression of endogenous and heterologous genes in mammalian cell lines (37)(38)(39). The Erbin DNA sequence was analyzed by a Web-based program (BLOCK-iT TM RNAi Designer) for putative siRNA sequences. Three distinct sequences were chosen that are conserved in human, mouse, and rat and have no homology to any DNA sequence in the NCBI sequence bank. They were subcloned in pENTR/U6 and tested for the ability to suppress endogenous Erbin expression in HEK 293 cells. Two clones, designated 2583 and 4049, suppressed expression of Erbin but not Densin-180, a protein that is homologous to Erbin (Fig. 5, A and B). Because of apparent higher efficiency, Erbin 4049 was used in the following experiments except when indicated otherwise.
If Erbin inhibits the interaction of Sur-8 with Ras and Raf, suppression of Erbin expression is expected to increase the interaction. To test this hypothesis, HEK 293 cells were transfected with Erbin 4049 to knock down Erbin expression; cell lysates were incubated with antibodies against Ras or Raf, and the resulting precipitates were probed for Sur-8. As shown in Fig. 5C, the amount of Ras and Raf precipitated with Sur-8 was increased in cells in which Erbin expression was suppressed (in comparison with cells transfected with the control vector pENTR that encodes LacZ shRNA), in line with the notion that Erbin disrupts Sur-8 interaction with Ras and Raf. Notice that the interaction of endogenous Raf with Sur-8 was barely detectable in LacZ shRNA-transfected cells but increased dramatically when Erbin was depleted, suggesting that Erbin may act by preventing Raf recruitment to the Ras-Sur-8 complex.
Enhanced Activation of Raf and ERK When Erbin Expression Was Suppressed-We next examined the effect of Erbin suppression on Raf and ERK activation by neuregulin. As shown in Fig. 6, transfection of HEK 293 cells with Erbin 2583 and 4049 decreased Erbin protein by 60 and 80%, respectively. However, Raf or ERK expression in cells transfected with LacZ shRNA or the Erbin shRNA constructs remained similar (Fig. 6), demonstrating specific suppression of Erbin expression. The basal levels of active Raf and ERK, as indicated by respective phospho-counterparts, appeared to be increased in the absence of neuregulin. Nevertheless, stimulation with neuregulin increased levels of both phospho-Raf and -ERK. In comparison with control cells (transfected transducing units/ml) 12 h later. The resulting myotubes were lysed and assayed for firefly and Renilla luciferase activities. The relative firefly/Renilla activity in cells infected with the control viruses was taken as 100%. Shown was a representative experiment in duplicates, which was repeated three times with similar results. B and C, reciprocal relationship between Erbin protein level and neuregulin-induced AChR e-subunit mRNA. C2C12 myoblasts were infected with Erbin-shRNA 4049 as in A. Resulting myotubes were subjected to analysis of AChR, Erbin, and actin mRNA by reverse transcription-PCR. Representative gels are shown in B, and quantitative analyses are shown in C.
with LacZ shRNA), neuregulin-induced increase in phospho-Raf and -ERK was enhanced in cells transfected with Erbin shRNA (Fig. 6). Moreover, the level of increase in Erbin 4049-transfected cells was higher than in cells transfected with Erbin 2583 (Fig. 6B), which correlated reciprocally with the levels of Erbin suppression. These results demonstrate that neuregulin-induced Raf and ERK activation was increased when Erbin expression was suppressed.
Erbin Regulation of Neuregulin-induced Expression of the AChR-Neuregulin has been implicated in local AChR synthesis at the synapse (40). Neuregulin activates ERK, which is required for induced AChR expression (30). To study the role of Erbin in regulating ERK activation in vivo, we generated lentiviral Erbin 4049, the more potent form of two shRNAs (Fig. 5). Infection of C2C12 myotubes with the Erbin-shRNA virus inhibited expression of endogenous Erbin (Fig. 7B) and appeared to have little, in any, effect on muscle differentiation (data not shown). Neuregulin-induced expression of the ⑀416-Luc transgene, however, was increased in muscle cells infected with the Erbin-shRNA virus (Fig.  7A), in line with the notion that Erbin functions as a negative regulator of neuregulin signaling. To determine whether inhibition of Erbin expression affects the expression of endogenous AChR mRNA level, virus-infected C2C12 myotubes were stimulated with neuregulin, and expression of the AChR ⑀-subunit mRNA was analyzed by reverse transcription-PCR. As shown in Fig. 7, B and C, viral infection decreased the mRNA levels of Erbin but not ␤-actin. Remarkably, the AChR ⑀-subunit mRNA increased in a concentration-dependent manner. The reciprocal relationship between the levels of Erbin and ⑀-subunit mRNA indicates a role of Erbin in regulating neuregulin-induced AChR expression.
These observations demonstrate that Erbin inhibits Raf activation probably by disrupting the interaction of Sur-8 with Ras and Raf. Convincing evidence is provided by studies using small interference RNA techniques. First, Erbin 4049 reduced Erbin expression at the mRNA and protein level. The decrease in Erbin expression correlated negatively with readouts of the proposed function of Erbin: increase in Sur-8 interaction with Ras and Raf (Fig. 5), enhanced activation of Raf and ERK by neuregulin (Fig. 6), and elevated expression of AChR ⑀-subunit mRNA and promoter activity (Fig. 7). Second, the effect of Erbin shRNA was concentration-dependent (Fig. 7). A reciprocal relationship of Erbin expression and AChR ⑀-subunit mRNA was demonstrated. Third, Erbin shRNA was gene-specific and had no effect on expression of Densin-180, a protein that is homologous to Erbin (Fig. 5). Finally, a similar effect was observed with two different Erbin shRNA constructs targeted to different sites in the Erbin mRNA. Both Erbin 2583 and 4049 suppressed Erbin expression and increased Raf and ERK activation by neuregulin in transfected cells (Fig. 6).
The results of this study and an earlier report (27) suggest that Erbin may serve as a negative feedback molecule to down-regulate the Ras-Raf-MEK pathway. Because Erbin interacts with various proteins, some of which appears to be scaffold components, it may play a role in finetuning the local signaling strength in a subcellular domain.