BRAF Splice Variant Resistance to RAF Inhibitor Requires Enhanced MEK Association

BRAF Splice Variant Resistance to RAF Inhibitor Requires Enhanced MEK Association

Article BRAF Splice Variant Resistance to RAF Inhibitor Requires Enhanced MEK Association Graphical Abstract Authors Michael J. Vido, Kaitlyn Le, Ed...

NAN Sizes 0 Downloads 0 Views

Article

BRAF Splice Variant Resistance to RAF Inhibitor Requires Enhanced MEK Association Graphical Abstract

Authors Michael J. Vido, Kaitlyn Le, Edward J. Hartsough, Andrew E. Aplin

Correspondence [email protected]

In Brief BRAF splice variants represent a common resistance mechanism to FDAapproved RAF inhibitors in melanoma. Through co-IP and functional studies, Vido et al. demonstrate that the phosphobinding site serine 729 mediates enhanced association between splice variants and their substrate, MEK, that is required for resistance to RAF inhibitors.

Highlights d

BRAF splice variants mediate resistance to FDA-approved RAF inhibitors

d

Low-dose RAF inhibitor treatment disrupts BRAF homodimers

d

Enhanced association with MEK1/2 is required for splicevariant-mediated resistance

d

BRAF S729 affects splice variant-MEK association and sensitivity to RAF inhibitor

Vido et al., 2018, Cell Reports 25, 1501–1510 November 6, 2018 ª 2018 Elsevier Inc. https://doi.org/10.1016/j.celrep.2018.10.049

Cell Reports

Article BRAF Splice Variant Resistance to RAF Inhibitor Requires Enhanced MEK Association Michael J. Vido,1,2 Kaitlyn Le,1 Edward J. Hartsough,1 and Andrew E. Aplin1,3,4,* 1Department

of Cancer Biology, Thomas Jefferson University, Philadelphia, PA 19107, USA College of Biomedical Sciences, Thomas Jefferson University, Philadelphia, PA 19107, USA 3Sidney Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA 19107, USA 4Lead Contact *Correspondence: [email protected] https://doi.org/10.1016/j.celrep.2018.10.049 2Jefferson

SUMMARY

Expression of aberrantly spliced BRAF V600E isoforms (BRAF V600E DEx) mediates resistance in 13%–30% of melanoma patients progressing on RAF inhibitors. BRAF V600E DEx confers resistance, in part, through enhanced dimerization. Here, we uncoupled BRAF V600E DEx dimerization from maintenance of MEK-ERK1/2 signaling. Furthermore, we show BRAF V600E DEx association with its substrate, MEK, is enhanced and required for RAF inhibitor resistance. RAF inhibitor treatment increased phosphorylation at serine 729 (S729) in BRAF V600E DEx. Mutation of S729 to a non-phosphorylatable residue reduced BRAF V600E DEx-MEK interaction, reduced dimerization or oligomerization, and increased RAF inhibitor sensitivity. Conversely, mutation of the BRAF dimerization domain elicited partial effects on MEK association and RAF inhibitor sensitivity. Our data implicate BRAF S729 in resistance to RAF inhibitor and underscore the importance of substrate association with BRAF V600E DEx. These findings may provide opportunities to target resistance driven by aberrantly spliced forms of BRAF V600E. INTRODUCTION The BRAF gene is mutated frequently in human cancers, including cutaneous melanoma and thyroid carcinoma (Davies et al., 2002); the most common mutation is a valine to glutamic acid substitution at codon 600 (V600E). BRAF V600E is constitutively active and signals downstream via MEK-ERK1/2 (Conner et al., 2003; Wan et al., 2004) to promote cellular transformation independent of RAS binding and RAF dimerization (Poulikakos et al., 2011; Ritt et al., 2010; Ro¨ring et al., 2012). Inhibiting BRAF V600E with US Food and Drug Administration (FDA)-approved RAF inhibitors, vemurafenib or dabrafenib, with or without MEK inhibitor, causes objective responses in 50%–70% of BRAF V600E melanoma patients and improves progression-free survival; however, resistance invariably arises (Chapman et al., 2011; Flaherty et al., 2010; Hartsough et al., 2014a; Sosman et al., 2012). Acquired resistance to RAF inhibitors and/or MEK inhibitors is often characterized by ERK1/2

pathway reactivation; common mechanisms include the expression of mutant RAS (Nazarian et al., 2010), amplification of BRAF V600E (Shi et al., 2012), and expression of alternatively spliced BRAF V600E isoforms (BRAF V600E DEx) (Basile et al., 2013; Hartsough et al., 2014b; Moriceau et al., 2015; Poulikakos et al., 2011; Shi et al., 2014; Wagle et al., 2014). Targeting resistance to RAF inhibitor monotherapy and RAF-MEK inhibitor combination therapy represents an unmet clinical need. Aberrantly spliced BRAF V600E (BRAF V600E DEx) isoforms have been identified in patients progressing on RAF inhibitors alone and in RAF-MEK inhibitor combinations, as well as in preclinical resistance assays (Basile et al., 2013; Moriceau et al., 2015; Poulikakos et al., 2011; Wagle et al., 2014). Additional alterations, including double kinase fusions (Kemper et al., 2016) and deletions of the BRAF N terminus (Johnson et al., 2018), have been identified in targeted inhibitor resistance. Chromosomal rearrangements of the BRAF gene that act as oncologic drivers are also found in multiple cancer types (Jones et al., 2008; Kulkarni et al., 2017; Lin et al., 2012). BRAF V600E DEx activates the MEK-ERK1/2 pathway during vemurafenib treatment and displays enhanced dimerization compared to full-length BRAF V600E (Poulikakos et al., 2011). A mutation in the BRAF dimerization domain (R509H) partially impairs maintenance of ERK1/2 phosphorylation levels in the presence of vemurafenib (Poulikakos et al., 2011), but effects on cell growth and viability have not been demonstrated. Crystal structures with vemurafenib bound depict BRAF as an asymmetrical dimer (Karoulia et al., 2016). This has led to a proposed model whereby vemurafenib binds one BRAF protomer, resulting in a conformational change that prevents vemurafenib binding to the second protomer. By contrast, others observe in bioluminescence resonance energy transfer (BRET) assays that vemurafenib binding disrupts BRAF homodimerization (Thevakumaran et al., 2015). These data are supported by immunoprecipitation data that show the disruption of BRAF V600E DEx oligomers by PLX4720 (Hartsough et al., 2018; Hatzivassiliou et al., 2010; Thevakumaran et al., 2015). It is possible that contrary effects seen on wild-type BRAF-CRAF heterodimerization may be dependent on background cellular and mutational contexts (Karoulia et al., 2016; Poulikakos et al., 2010). Whereas enhanced BRAF dimerization has been proposed as a common attribute underlying vemurafenib resistance (Karoulia et al., 2016; Yao et al., 2015), increased association between BRAF and its substrate MEK has also been observed in the setting of resistance to concurrent RAF-MEK inhibition

Cell Reports 25, 1501–1510, November 6, 2018 ª 2018 Elsevier Inc. 1501

Figure 1. aC-OUT Inhibitors Reduce BRAF V600E DEx Homodimerization (A) 1205LuTR melanoma cells were induced with 100 ng/mL doxycycline (DOX) for 24 hr to co-express myc-tagged and FLAG-tagged BRAF V600E or BRAF V600E DEx2-8 with the expression of myc-tagged BRAF V600E DEx2-8 as a control. Cells were then treated with or without 1 mM PLX4720 for 24 hr. Cell lysates were immunoprecipitated with anti-FLAG Affinity Gel, and homodimerization was assessed by western blot. Arrows indicate BRAF isoforms. FLAG immunoprecipitate is shown on the top; whole-cell lysate (WCL) is shown on the bottom. (B) Quantitation of myc and FLAG from the western blot of FLAG immunoprecipitate was performed using Quantity One software to measure homodimerization. Displayed values are normalized to BRAF V600E DEx2-8 and represent the mean from six independent experiments. (C) As in (A) except cells were induced with 5 ng/mL DOX and treated with 0.001, 0.005, 0.01, 0.05, 0.1, 0.5, 1, 5, and 10 mM PLX4720 for 4 hr. FLAG immunoprecipitate is shown on top; WCL is shown on the bottom. (D) Quantitation of myc and FLAG from the western blot of FLAG immunoprecipitate and phosphoMEK from the western blot of WCL was performed using Quantity One software. Displayed values are normalized to vehicle treatment and represent the mean from three experimental replicates. (E) As in (C) except cells were treated with vemurafenib (left) or GDC-0879 (right); displayed blots are representative of three experimental replicates. Error bars represent SEM; significance determined by Student’s t test assuming unequal variance was indicated as follows: *p < 0.05 and ****p % 0.0001. See also Figures S1 and S2.

(Moriceau et al., 2015). BRAF mutational status and RAF inhibitor binding can alter the degree of BRAF-MEK interaction in a dimerization-independent manner (Haling et al., 2014). The degree of MEK association with BRAF V600E DEx has not been studied. Here, we sought to better define the mechanisms underlying BRAF V600E DEx resistance. We demonstrate that BRAF V600E DEx exhibits enhanced association with substrate MEK compared to full-length BRAF V600E and that this association is required for BRAF V600E DEx-mediated resistance. These findings support a view that disrupting BRAF-MEK association could represent a potential pharmacologic goal during RAF inhibitor therapy. To this end, we show that the phospho-binding site, serine 729 (S729), is required for BRAF V600E DEx-MEK association as well as dimerization or oligomerization and, importantly, resistance to RAF inhibitor. RESULTS BRAF V600E DEx Homodimerization Is Reduced by LowConcentration Vemurafenib Treatment BRAF V600E DEx displays enhanced homodimerization; however, assays measuring BRAF V600E DEx homodimerization in the presence of vemurafenib and other aC-OUT inhibitors are limited. 1502 Cell Reports 25, 1501–1510, November 6, 2018

Thus, we co-expressed myc-tagged and FLAG-tagged full-length and splice variant forms of BRAF in 1205LuTR melanoma cells and performed co-immunoprecipitation (co-IP). This expression model recapitulates the patient resistance setting in which BRAF V600E DEx is expressed concurrently with full-length BRAF V600E (Basile et al., 2013; Poulikakos et al., 2011). As previously reported, BRAF V600E DEx2-8 homodimerized to a greater extent than full-length BRAF V600E and association was disrupted by the R509H mutation to the dimerization domain (Poulikakos et al., 2011; Figure S1A). For simplicity of notation, we will refer to BRAF homodimers, although our co-IP experiments may also capture larger BRAF oligomers. Treatment with PLX4720, the non-clinical analog of vemurafenib, significantly disrupted BRAF V600E DEx2-8 co-IP (Figure 1A; quantitated in Figure 1B). Despite the disruption of BRAF association, cellular phospho-MEK levels remained largely unaffected at this dose. To more precisely define the relationship between PLX4720 disruption of BRAF V600E DEx oligomerization and cellular phospho-MEK levels, we measured association across a range of PLX4720 doses (1 nM–10 mM; Figure 1C; quantitated in Figure 1D). BRAF V600E DEx association decreased starting at 5 nM of PLX4720 and remained unaltered at concentrations above 10 nM. By contrast, cellular phospho-MEK levels were largely unaffected at PLX4720 doses (1 nM–5 mM), where

reductions in BRAF V600E DEx association were noted. Consistent results were measured with the clinical compound, vemurafenib (Figure 1E). The cellular phospho-MEK dose response matched that published for vemurafenib-resistant cells expressing endogenous BRAF V600E DEx (Karoulia et al., 2016; Poulikakos et al., 2011). Importantly, our data suggest that, even at relatively low concentrations, vemurafenib can bind to BRAF protomers and disrupt BRAF V600E DEx homodimerization. PLX4720 and vemurafenib represent aC-OUT/DFG-IN RAF inhibitors (Karoulia et al., 2016). We also tested the effect of GDC0879 (aC-IN/DFG-IN) on BRAF V600E DEx. Consistent with studies measuring BRAF-CRAF heterodimerization or homodimerization of full-length BRAF (Hatzivassiliou et al., 2010; Karoulia et al., 2016), we detected a marked increase in BRAF V600E DEx co-IP across a range of GDC-0879 doses (Figure 1E). These data show that aC-OUT and aC-IN inhibitors have opposing effects on BRAF splice variant homodimerization or oligomerization. Other mechanisms of resistance to RAF inhibitors, including BRAF amplification and RAS mutation, are characterized by the involvement of additional RAF isozymes (Moriceau et al., 2015; Yao et al., 2015). We did not detect substantial association of BRAF V600E DEx with ARAF, full-length BRAF, or CRAF when we treated with increasing doses of PLX4720 (Figure S1B). By contrast, GDC-0879 treatment, which increased BRAF V600E DEx co-IP (Figure 1E), resulted in a measurable increase in association between BRAF V600E DEx2-8 and other RAF isozymes (Figure S1B). Additional knockdown experiments were consistent with BRAF V600E DEx conferring resistance to PLX4720 independent of full-length BRAF and CRAF (Figures S2A and S2B; Basile et al., 2013; Poulikakos et al., 2011). BRAF V600E DEx Demonstrates Increased Association with Substrate MEK Because treatment with aC-OUT RAF inhibitors uncoupled BRAF V600E DEx homodimerization from MEK-ERK1/2 activation, we sought to understand how additional BRAF binding partners contribute to BRAF V600E DEx-mediated resistance. Increased interaction between full-length mutant BRAF and its substrate, MEK1/2, contributes to RAF-MEK combination therapy resistance (Moriceau et al., 2015). We assayed MEK interaction with BRAF V600E in 1205Lu parental and PLX4720-resistant cell lines (PRT) by co-IP (Basile et al., 2013; Figure 2A). PRT #3 and PRT #4 cells express BRAF V600E DEx3-10 and BRAF V600E D2-8, respectively, at comparable levels to full-length BRAF V600E. Co-IP experiments demonstrated greater MEK1 binding to BRAF V600E DEx compared to full-length BRAF V600E. ARAF and CRAF were not detected in the MEK1 immunoprecipitates (Figure S3A). Next, we inducibly expressed FLAG-tagged BRAF constructs and compared FLAG immunoprecipitates, utilizing FLAG-tagged LacZ as a pull-down control. We compared V600E mutant BRAF isoforms as opposed to wild-type BRAF, as BRAF kinase activity has been shown to affect MEK association (Haling et al., 2014). Consistent with co-IP of endogenous protein, BRAF V600E DEx2-8 demonstrated increased association with total and phospho-MEK1/2 compared to full-length BRAF V600E (Figure 2B; quantitated in Figures S3B–S3D). In this context, we also failed to detect association between BRAF V600E DEx2-8

and either ARAF, full-length BRAF, or CRAF (Figure S3E). Interestingly, low doses of PLX4720 (5–100 nM) that disrupted BRAF V600E DEx homodimerization (Figures 1C and 1D) maintained BRAF V600E DEx-MEK interaction (Figure 2C; quantitated in Figure S3F). Similar results were also noted following treatment with vemurafenib (Figure 2D). These data correlate enhanced MEK association with BRAF V600E DEx-mediated resistance. Disrupting MEK Association with BRAF V600E DEx Increases Sensitivity to PLX4720 To test the requirement of MEK association in BRAF V600E DExmediated resistance, we introduced mutations into BRAF V600E DEx2-8 that are predicted to decrease interaction with MEK (Haling et al., 2014; Moriceau et al., 2015). These mutations did not alter the ability of BRAF V600E DEx2-8 to phosphorylate MEK in 293FT cells (Figure S4A; quantitated in Figure S4B). We expressed R462A/E and I617R/F forms of BRAF V600E DEx2-8 in 1205LuTR cells and observed reduced co-IP of MEK1/2 and phospho-MEK1/2 (Figure 3A; quantitated in Figure 3B). The I617R and I617F mutations profoundly reduced MEK1 association, and a more modest disruption of MEK1 association was measured with the R462A and R462E mutations. I617R/F mutations in BRAF V600E DEx2-8 increased sensitivity to PLX4720 (Figure 3C; quantitated in Figure 3D). Consistent but less profound effects on phospho-MEK levels were measured with the R462A/R mutations, in accordance with the level of MEK1/2 association (Figure S4C; quantitated in Figure S4E). The individual effect of these mutations did not fully recapitulate the sensitivity measured when we inducibly expressed full-length BRAF V600E (Figure S4D; quantitated in Figures 3D and S4E). Importantly, the inclusion of the mutations to the BRAF-MEK interface did not prevent BRAF V600E DEx2-8 homodimerization (Figure 3E; quantitated in Figure S4F). These data suggest that, although enhanced MEK1/2 association alone does not fully account for BRAF V600E DEx-mediated resistance, increased MEK1/2 association is required for sustained BRAF V600E DEx signaling, even when dimerization status is maintained. Serine 729 Contributes to BRAF-MEK Association and BRAF V600E DEx Homodimerization We next sought to identify sites in BRAF V600E DEx that contribute to the observed increase in BRAF-MEK association. We noticed that the protein banding pattern for BRAF V600E DEx was altered when we removed PLX4720 from PRT cells that are normally cultured in the presence of RAF inhibitor (Figure 4A). The presence of a higher molecular weight band during PLX4720 treatment was consistent with increased levels of phospho-S729 on BRAF V600E DEx. This observation was also detected in multiple melanoma cell lines exogenously expressing distinct BRAF V600E DEx isoforms (Figure S5A). S729 is a binding site for the 14-3-3 family proteins, which are reported to associate with BRAF-MEK complexes, although the consequences of this interaction are not fully understood (Haling et al., 2014). We tested whether S729 was involved in BRAF-MEK interaction. Mutation to a non-phosphorylatable alanine (S729A) resulted in a significant reduction in BRAF V600E DEx2-8 association with phospho-MEK and total MEK Cell Reports 25, 1501–1510, November 6, 2018 1503

Figure 2. BRAF V600E DEx Enhanced MEK Association

Displays

(A) 1205LuTR GAL4-ELK1 reporter (parental) cells or PLX4720-resistant cells (PRT) cells, maintained in standard growth conditions, were lysed at confluency and immunoprecipitated with MEK1 antibody or immunoglobulin G (IgG) isotype control. Protein binding was assessed by western blot. Solid arrows indicate BRAF V600E DEx3-10, dashed arrows indicate BRAF V600E DEx2-8, and diamonds indicate background bands. WCL is shown on the left; MEK1 or IgG control immunoprecipitate is shown on the right. (B) 1205LuTR cells were induced with 100 ng/mL DOX for 48 hr to express FLAG-tagged LacZ, BRAF V600E, or BRAF V600E DEx2-8. Cell lysates were immunoprecipitated with anti-FLAG Affinity Gel, and protein binding was assessed by western blot. FLAG immunoprecipitate is shown on the top; WCL is shown on the bottom. (C) 1205LuTR cells were induced with 5 ng/mL DOX for 48 hr to express FLAG-tagged BRAF V600E DEx2-8. FLAG-tagged LacZ was expressed as a control. Cells were dosed with PLX4720 as in Figure 1C. Cell lysates were immunoprecipitated with anti-FLAG Affinity Gel, and protein binding was assessed by western blot. FLAG immunoprecipitate is shown on top; WCL is shown on the bottom. (D) As in (C), except cells were treated with vemurafenib. All displayed blots are representative of three experimental replicates. See also Figure S3.

(Figure 4B; quantitated in Figures 4C and 4D). Consistent with prior findings (Haling et al., 2014; Moriceau et al., 2015; Nan et al., 2013), the R509H mutation to the dimerization domain partially reduced BRAF V600E DEx2-8-MEK1/2 association. We also measured the effect of the S729A and R509H point mutations on BRAF V600E DEx-MEK1/2 association in the presence of PLX4720 (Figure S5B). In this setting, total and phosphoMEK association with BRAF V600E DEx was again partially reduced by R509H and more completely reduced by the S729A mutation. The overall level of MEK association was moderately reduced in the presence of PLX4720, as in Figure 2C. As S729 has also been shown to participate in wild-type RAF dimerization (Garnett et al., 2005; Ritt et al., 2010; Rushworth et al., 2006; Weber et al., 2001), we measured the effects of S729 mutation on BRAF V600E DEx homodimerization and compared the effect to R509H. Both the R509H and S729A mutations disrupted homodimerization, although the R509H mutation did so to a greater extent (Figure 4E; quantitated in Figure 4F). The reduction in dimerization caused by the R509H and S729A 1504 Cell Reports 25, 1501–1510, November 6, 2018

mutations was also observed when cells were treated with PLX4720 (Figure S5C). These results, taken together, suggest that S729 is involved in both BRAF substrate association and homodimerization. S729 is not one of the ERK1/2 feedback phosphorylation sites on BRAF (Ritt et al., 2010), but this site in wildtype BRAF is phosphorylated in response to activation of AMP-activated protein kinase (AMPK) (Shen et al., 2013). To test the extent that AMPK phosphorylates S729 in BRAF V600E DEx, we performed small interfering RNA (siRNA) knockdown of AMPKa1 in cells expressing exogenous or endogenous BRAF V600E DEx (Figures S6A–S6C). AMPKa1 knockdown was sufficient to reduce phosphorylation of the AMPK substrate, acetyl-coenzyme A (CoA) carboxylase (ACC). Following AMPKa1 knockdown, S729 phosphorylation (Figures S6A and S6B) and banding patterns of endogenous BRAF V600E DEx3-10 (Figure S6C) were unaltered. Our data indicate that phosphorylation of S729 in BRAF V600E DEx occurs by an AMPK-independent mechanism. Mutation of Serine 729 in BRAF V600E DEx Impairs MEK-ERK1/2 Activity in the Presence of RAF Inhibitor Given the measured disruption of BRAF-MEK association and BRAF V600E DEx homodimerization by the S729A mutation, we tested the role of S729 in BRAF V600E DEx signaling. In the

Figure 3. Increased MEK1/2 Association Contributes to BRAF V600E DEx Signaling in the Presence of RAF Inhibitors (A) 1205LuTRs were induced with 100 ng/mL DOX for 48 hr to express FLAG-tagged LacZ, BRAF V600E DEx2-8, or BRAF with MEK1 binding site mutations. Lysates were immunoprecipitated with anti-FLAG Affinity gel, and protein binding was assessed by western blot. FLAG immunoprecipitate is shown on top; WCL is shown on the bottom. (B) Protein association was measured using Quantity One software to quantify FLAG, pMEK1/2 (top), and MEK1 (bottom) in the FLAG immunoprecipitate. Displayed values are normalized to BRAF V600E DEx2-8 and represent the mean from three experimental replicates. (C) 1205LuTR cells were induced with 100 ng/mL DOX for 24 hr to express BRAF V600E DEx2-8, BRAF V600E DEx2-8 I617R, or BRAF V600E DEx2-8 I617F. Cells were subsequently dosed with 0, 0.1, 1, and 10 mM PLX4720 for 24 hr. Cell lysates were analyzed by western blot. (D) Protein expression from Figure 3C was measured using Quantity One software to quantify pMEK1/2 and BRAF V600E. Quantitation of fulllength BRAF V600E (western blot in Figure S4D) was included for comparison. Displayed values are normalized to the no-drug condition and represent the mean from three experimental replicates. (E) Similar to (A), except myc-tagged and FLAGtagged BRAF V600E DEx2-8 constructs were coexpressed. Single expression of myc-tagged BRAF V600E DEx2-8 was expressed as a control. An arrow indicates the relevant band for BRAF V600E DEx2-8. FLAG immunoprecipitate is shown on top; WCL is shown on bottom. All displayed blots are representative of three experimental replicates. Error bars represent SEM; significance determined by Student’s t test assuming unequal variance was indicated as follows: *p < 0.05; **p % 0.01; ***p % 0.001; and ****p % 0.0001. See also Figure S4.

absence of RAF inhibitor, the S729A mutation did not change the activity of BRAF V600E DEx when expressed transiently in 293FT cells (Figure 5A) or inducibly in 1205LuTR melanoma cells (Figure 5B). Consistent with others’ data (Poulikakos et al., 2011; Thevakumaran et al., 2015), the R509H mutation also did not alter the activity of BRAF V600E DEx (Figures 5A and 5B). Induced expression of BRAF V600E DEx2-8 maintained phospho-MEK and phospho-ERK1/2 levels in the presence of PLX4720. Interestingly, PLX4720 treatment of the R509H, R462A, or I617R mutants partially reduced signaling, but signaling was nearly ablated in S729A-mutant-expressing cells (Figure 5C). The reduced phospho-MEK and ERK1/2 in BRAF V600E DEx2-8 S729A-expressing cells was evident across a range of PLX4720 doses (Figure 5D). Similar effects were observed with a distinct splice variant, BRAF V600E DEx3-10 (Figures S7A and S7B).

S729 Is Required for RAF Inhibitor Resistance To address the effects of S729 on resistant melanoma growth, we performed 2D growth assays in 1205LuTR cells induced to express either intact, dimerization-inhibited (R509H), MEK-binding-inhibited (R462A and I617R), or S729A versions of BRAF V600E DEx2-8. Full-length BRAF V600E was expressed as a reference. In the absence of PLX4720, cells expressing the various mutated BRAF isoforms grew at the same rate (Figure 6A). In the presence of 1 mM PLX4720, full-length BRAF V600E and R462A, I617R, and S729A versions of BRAF V600E DEx demonstrated reduced growth compared to BRAF V600E DEx. By contrast, the R509H mutation caused a partial reduction in growth. We also compared colony growth between intact, R509H, and S729A versions of BRAF V600E DEx3-10. Although BRAF V600E DEx3-10-expressing cells displayed reduced colony growth compared to non-induced cells, growth Cell Reports 25, 1501–1510, November 6, 2018 1505

Figure 4. S729 Affects BRAF V600E DExMEK Association and Homodimerization (A) 1205LuTR GAL4-ELK1 reporter (parental) cells or PRT cells were cultured in normal media in the absence of PLX4720 for 96 hr. Cells were then treated with or without 1 mM PLX4720 for an additional 24 hr. Cell lysates were analyzed by western blot. Solid arrows indicate BRAF V600E DEx3-10; dashed arrows indicate BRAF V600E DEx2-8. Displayed blots are representative of three experimental replicates. (B) 1205LuTR cells were induced with 100 ng/mL DOX for 48 hr to express FLAG-tagged LacZ, BRAF V600E DEx2-8, BRAF V600E DEx2-8 R509H, or BRAF V600E DEx2-8 S729A. Lysates were immunoprecipitated with anti-FLAG Affinity gel, and protein binding was assessed by western blot. FLAG immunoprecipitate is shown on top; WCL is shown on bottom. (C and D) Protein association was measured using Quantity One software to quantify FLAG, pMEK1/2 (C), and MEK1/2 (D) in the FLAG immunoprecipitate. Displayed values are normalized to BRAF V600E DEx2-8 and represent the mean from six experimental replicates. (E) As in (B), except 1205LuTR cells were induced to co-express myc-tagged and FLAG-tagged BRAF isoforms. Sole expression of myc-tagged BRAF V600E DEx2-8 served as a control. (F) Quantitation of myc and FLAG from the western blot of FLAG immunoprecipitate was performed using Quantity One software to measure homodimerization. Displayed values are normalized to BRAF V600E DEx2-8 and represent the mean from six experimental replicates. Error bars represent SEM; significance determined by Student’s t test assuming unequal variance was indicated as follows: *p < 0.05; **p % 0.01; ***p % 0.001; and ****p % 0.0001. See also Figures S5 and S6.

was maintained in the presence of increasing concentrations of PLX4720 (Figure 6B; representative images in Figure S7C). Expression of R509H mutant forms of BRAF V600E DEx also maintained growth in the presence of RAF inhibitor, consistent with our observations of BRAF V600E DEx homodimerization and signaling in the presence of RAF inhibitor. These data are in contrast to the RAF inhibitor sensitivity of BRAF V600E DEx3-10 S729A, which is both homodimerization and MEK association deficient. Similarly, BRAF V600E DEx2-8 (Figure 6C) or BRAF V600E DEx3-10 (Figure 6D) with the S729A mutation was not sufficient to promote S phase entry in the presence of PLX4720. Although disrupting MEK association with the R462A or I617R mutations affected long-term cell growth, mutation of these sites failed to alter S phase entry in the presence of PLX4720 (Figure S7D). To extend these studies in vivo, we analyzed signaling effects quantitatively and non-invasively using an ERK1/2 reporter system in mutant BRAF melanoma xenografts (Basile et al., 2013). In this model, luciferase levels increased in 1205LuTR xenografts expressing the BRAF V600E DEx3-10 despite treatment with 1506 Cell Reports 25, 1501–1510, November 6, 2018

RAF inhibitors (Figures 6E and S7E). By contrast, ERK1/2 reporter levels remained low in BRAF V600E DEx3-10 S729A-expressing xenografts. This signaling response matched the xenograft growth kinetics (Figure 6F). Thus, BRAF V600E DEx-mediated resistance to RAF inhibitor is dependent on S729. Taken together, our findings support a model whereby MEK association contributes to BRAF V600E DEx-mediated resistance via a mechanism mediated by S729. DISCUSSION BRAF splice variants, translocations, and deletions that express the kinase domain separate from the N-terminal region are involved in malignant transformation and drug resistance (Johnson et al., 2018; Jones et al., 2008; Kemper et al., 2016; Kulkarni et al., 2017; Lin et al., 2012). Selection for BRAF V600E DEx expression frequently drives resistance to RAF inhibitors in melanoma and RAF-MEK inhibitor combinations (Basile et al., 2013; Moriceau et al., 2015; Poulikakos et al., 2011; Shi et al., 2014; Wagle et al., 2014). The mechanism by which BRAF V600E

Figure 5. S729 in BRAF Splice Variants Is Required for the Maintenance of Phosphorylation of ERK1/2 In Vitro (A) 293FT cells were transiently transfected with plasmids encoding myc-tagged BRAF V600E DEx3-10, BRAF V600E DEx3-10 R509H, or BRAF V600E DEx3-10 S729A. Cell lysates were analyzed by western blotting. The levels of phospho-ERK1/2 and exogenous BRAF V600E DEx3-10 were quantitated using Quantity One software. Graphed is the mean of three independent experiments ± SEM. (B) 1205LuTR cells were induced with 100 ng/mL DOX for 24 hr to express FLAG-tagged BRAF V600E DEx2-8, BRAF V600E DEx2-8 R509H, or BRAF V600E DEx2-8 S729A. The levels of pMEK1/2 and exogenous BRAF V600E DEx2-8 were analyzed as in (A). (C) As in (B), except that cells expressing R462Aor I617R-mutated BRAF V600E DEx2-8 were included and cells were subsequently treated with 1 mM PLX4720 for 24 hr. (D) As in (C), except that cells were induced to express BRAF V600E DEx2-8 forms and treated with increasing (0, 0.1, 0.5, 1, 5, and 10 mM) doses of PLX4720 for 24 hr. All displayed blots are representative of three experimental replicates. See also Figure S7.

DEx confer resistance is reported to be dependent on the dimerization domain, including residues R506–M517 (Ro¨ring et al., 2012). Here, we expand upon this model of BRAF V600E DEx signaling to include phospho 729 and increased association with MEK substrate as critical aspects of BRAF V600E DExmediated resistance. BRAF V600E DEx display increased homodimerization compared to full-length BRAF V600E (Poulikakos et al., 2011). It has been postulated that resistance to aC-OUT/DFG-IN RAF inhibitors, such as PLX4720 and vemurafenib, occurs via a mechanism whereby RAF dimers permit inhibitor binding to one protomer that precludes inhibitor binding to the second protomer (Karoulia et al., 2016, 2017). In our study, BRAF V600E DEx protein homodimerization or oligomerization was disrupted at RAF inhibitor concentrations that failed to alter ERK1/2 pathway activity or cell growth. These data support an alternative view that vemurafenib can bind to BRAF complexes and the resulting structural alterations disrupt oligomerization (Hartsough et al., 2018). This is consistent with BRET assays measuring a disruption of in vivo dimerization in the presence of vemurafenib (Thevakumaran et al., 2015). Increased BRAF-MEK association has been observed in resistance to combined RAF and MEK inhibitors mediated by BRAF V600E amplification or MEK1 mutation (Moriceau et al., 2015). Importantly, disrupting BRAF dimerization in these contexts had a partial effect on MEK association and a marginal effect on inhibitor sensitivity. In RAF-inhibitor-resistant cells that express endogenous BRAF V600E DEx, we detected increased association of MEK with BRAF V600E DEx compared to full-length

BRAF V600E. Vemurafenib disrupted MEK association with BRAF at concentrations that inhibited downstream signaling consistent with the inactive structure adopted by BRAF when bound to aC-OUT/DFG-IN inhibitors (Haling et al., 2014). Expressing BRAF V600E DEx with mutations that decreased BRAF-MEK association but preserved BRAF dimerization led to increased sensitivity to RAF inhibitor. Because BRAF is expressed in excess of MEK in most cells (Haling et al., 2014), our data support the disruption of BRAF oligomers that are unbound to MEK at low PLX4720 concentrations with effects on cellular phospho-MEK levels occurring at PLX4720 doses that disrupt BRAF-MEK association. This model allows for low levels of BRAF oligomers to persist until high PLX4720 concentrations disrupt BRAF-MEK complexes. Disrupting BRAF-MEK association may represent a pharmacologic goal in patients with tumors that are resistant to RAF inhibitors. Our data suggest that S729 is involved with the enhanced BRAF-MEK interaction observed in BRAF V600E DEx. S729 is a 14-3-3 binding site that has been previously implicated in wild-type BRAF hetero-dimerization with CRAF (Garnett et al., 2005; Ritt et al., 2010; Rushworth et al., 2006; Weber et al., 2001). In wild-type BRAF, 14-3-3 binding to phosphorylated S729 disrupts KSR1 binding (Shen et al., 2013), and mutation of S729 to a non-phosphorylatable residue impairs activation of downstream signaling (MacNicol et al., 2000); however, loss of S729 in BRAF V600E does not alter potential activity or transformation (Brummer et al., 2006; Ritt et al., 2010). In the context of oncogenic BRAF mutations with low to intermediate kinase activity (G466A and D594G), the S729A mutation impairs BRAF Cell Reports 25, 1501–1510, November 6, 2018 1507

Figure 6. S729 Mutation in BRAF V600E DEx Mediates Growth Resistance to RAF Inhibitor In Vitro and In Vivo (A) 1205 LuTR cells were plated at clonal density and induced with 100 ng/mL DOX to express FLAGBRAF or FLAG-BRAF V600E DEx2-8 with the indicated mutations for 24 hr. Cells were then treated with DMSO (left) or 1 mM PLX4720 (right), and percent confluence was monitored continuously using the IncuCyte Kinetic Imaging System for 5 days. Displayed values represent the mean, and error bars are SEM from three experimental replicates. (B) 1205LuTR cells were plated at clonal density and incubated with or without DOX to express myc-tagged BRAF V600E DEx3-10, DEx3-10 R509H, or DEx3-10 S729A for 24 hr before treating with PLX4720 (0, 0.05, 0.1, 0.2, 0.5, and 1 mM) for 8 days. Growth was quantitated following crystal violet staining with ImageJ software and normalized to nodrug condition. Displayed values represent the mean, and error bars are SEM from three experimental replicates. (C) 1205LuTR myc-BRAF V600E DEx2-8-expressing cells were induced or left un-induced for 24 hr before treatment with DMSO or PLX4720 (1 mM) for 48 hr. EdU was added for final 16 hr. Cells were collected, and S phase entry was assessed by EdU incorporation. Assays were completed in triplicate. Error bars represent SEM; significance determined by Student’s t test assuming unequal variance. (D) As in (C), except that myc-BRAF V600E DEx3-10 cells were analyzed. (E) Female nude mice were injected with 1205LuTR ERK1/2 reporter cells expressing either exogenous myc-BRAF V600E DEx3-10 (n = 8 mice) or mycBRAF V600E DEx3-10 S729A (n = 5 mice). After palpable tumors were formed (day 0), mice were fed chow laced with PLX4720 (417 mg/kg). The graph shows fold change in average luciferase output (ERK1/2 activity) per tumor volume as measured by caliper measurements. The data in each group are normalized to the respective day 0. (F) Graphed are average tumor volume ± SEM from two cohorts until the protocol volume limit was reached. Lines represent trend lines from each dataset with associated p value based on general linear model of log-transformed tumor volumes. Significance was indicated as follows: *p < 0.05 and ***p % 0.001. See also Figure S7.

transforming ability. This is consistent with oncogenic BRAF reliance on dimerization: low and intermediate kinase activity BRAF mutations require dimerization although BRAF V600E can signal independent of dimer status (Ro¨ring et al., 2012; Yao et al., 2015). Importantly, sensitivity to RAF inhibitor closely matched the extent of MEK association. The reduced MEK association with the S729A mutant compared to R509H provides an explanation for the more profound PLX4720 sensitivity of the S729A mutant compared to the R509H mutant. By contrast, homodimerization status did not match the response to RAF inhibitor, as the R509H mutation disrupted homodimerization to a greater level than the S729A mutation but exhibited a partial effect on PLX4720 sensi1508 Cell Reports 25, 1501–1510, November 6, 2018

tivity. How S729 contributes to MEK interaction, either directly or via other protein intermediaries, like 14-3-3 protein family members, represents a future line of inquiry. Because an unmet need in melanoma is second-line treatment options for patients progressing on targeted therapies, our studies highlight the importance of understanding and disrupting the regulation of S729 phosphorylation to thwart acquired resistance that is mediated by BRAF V600E DEx. New RAF inhibitors with distinct modes of action continue to enter clinical trials (Karoulia et al., 2017). These inhibitors have been characterized based on their ability to inhibit signaling or growth of BRAF V600E DEx-harboring cells to infer their ability to inhibit BRAF dimers (Peng et al., 2015;

Yao et al., 2015). Our data suggest that the effectiveness of these RAF inhibitors in this model system may additionally relate to their ability to disrupt BRAF-MEK association or sequester BRAF-MEK in an inactive conformation (Haling et al., 2014; Karoulia et al., 2017). Furthermore, although the disruption of BRAF dimerization has been associated with favorable results in preclinical models (Freeman et al., 2013; Grasso et al., 2016; Sievert et al., 2013), directly testing how these treatments effect substrate binding is necessary to broadly apply such strategies. STAR+METHODS Detailed methods are provided in the online version of this paper and include the following: d d d

d

d

KEY RESOURCES TABLE CONTACT FOR REAGENT AND RESOURCE SHARING EXPERIMENTAL MODEL AND SUBJECT DETAILS B In vivo experiments B Cell lines METHOD DETAILS B Point-mutagenesis and epitope tagging B Western blotting B Transfections B Cell growth assays and S phase entry analysis B Immunoprecipitation QUANTIFICATION AND STATISTICAL ANALYSIS

SUPPLEMENTAL INFORMATION Supplemental Information includes seven figures and can be found with this article online at https://doi.org/10.1016/j.celrep.2018.10.049. ACKNOWLEDGMENTS We thank the members of the Aplin laboratory for the critical feedback that they provided for this article. We also thank Dr. Meenhard Herlyn (Wistar Institute, Philadelphia, PA) for providing 1205Lu cells. Dr. Inna Chervoneva aided with biostatistical analysis. This work was funded by grants from the NIH with the following grant numbers: F30-CA203314 (M.J.V.), K99-CA207855 (E.J.H.), and R01-CA182635 (A.E.A.). A.E.A. is also supported by the Dr. Miriam and Sheldon G. Adelson Medical Research Foundation. The Laboratory Animal shared facility and biostatistics resource in the Sidney Kimmel Cancer Center are supported by the NCI Support Grant P30-CA56036. AUTHOR CONTRIBUTIONS Conceptualization, M.J.V. and A.E.A.; Methodology, M.J.V., E.J.H., and A.E.A.; Investigation, M.J.V., K.L., and E.J.H.; Writing – Original Draft, M.J.V.; Writing – Review & Editing, M.J.V., E.J.H., and A.E.A.; Funding Acquisition, M.J.V., E.J.H., and A.E.A.; Resources, A.E.A. DECLARATION OF INTERESTS A.E.A. received grant funding from Pfizer Inc. from 2013 to 2017. Received: January 23, 2018 Revised: September 11, 2018 Accepted: October 11, 2018 Published: November 6, 2018

REFERENCES Abel, E.V., and Aplin, A.E. (2010). FOXD3 is a mutant B-RAF-regulated inhibitor of G(1)-S progression in melanoma cells. Cancer Res. 70, 2891–2900. Basile, K.J., Abel, E.V., Dadpey, N., Hartsough, E.J., Fortina, P., and Aplin, A.E. (2013). In vivo MAPK reporting reveals the heterogeneity in tumoral selection of resistance to RAF inhibitors. Cancer Res. 73, 7101–7110. Brummer, T., Martin, P., Herzog, S., Misawa, Y., Daly, R.J., and Reth, M. (2006). Functional analysis of the regulatory requirements of B-Raf and the B-Raf(V600E) oncoprotein. Oncogene 25, 6262–6276. Chapman, P.B., Hauschild, A., Robert, C., Haanen, J.B., Ascierto, P., Larkin, J., Dummer, R., Garbe, C., Testori, A., Maio, M., et al.; BRIM-3 Study Group (2011). Improved survival with vemurafenib in melanoma with BRAF V600E mutation. N. Engl. J. Med. 364, 2507–2516. Conner, S.R., Scott, G., and Aplin, A.E. (2003). Adhesion-dependent activation of the ERK1/2 cascade is by-passed in melanoma cells. J. Biol. Chem. 278, 34548–34554. Davies, H., Bignell, G.R., Cox, C., Stephens, P., Edkins, S., Clegg, S., Teague, J., Woffendin, H., Garnett, M.J., Bottomley, W., et al. (2002). Mutations of the BRAF gene in human cancer. Nature 417, 949–954. Flaherty, K.T., Puzanov, I., Kim, K.B., Ribas, A., McArthur, G.A., Sosman, J.A., O’Dwyer, P.J., Lee, R.J., Grippo, J.F., Nolop, K., and Chapman, P.B. (2010). Inhibition of mutated, activated BRAF in metastatic melanoma. N. Engl. J. Med. 363, 809–819. Freeman, A.K., Ritt, D.A., and Morrison, D.K. (2013). Effects of Raf dimerization and its inhibition on normal and disease-associated Raf signaling. Mol. Cell 49, 751–758. Garnett, M.J., Rana, S., Paterson, H., Barford, D., and Marais, R. (2005). Wildtype and mutant B-RAF activate C-RAF through distinct mechanisms involving heterodimerization. Mol. Cell 20, 963–969. Grasso, M., Estrada, M.A., Ventocilla, C., Samanta, M., Maksimoska, J., Villanueva, J., Winkler, J.D., and Marmorstein, R. (2016). Chemically Linked Vemurafenib Inhibitors Promote an Inactive BRAFV600E Conformation. ACS Chem. Biol. 11, 2876–2888. Haling, J.R., Sudhamsu, J., Yen, I., Sideris, S., Sandoval, W., Phung, W., Bravo, B.J., Giannetti, A.M., Peck, A., Masselot, A., et al. (2014). Structure of the BRAF-MEK complex reveals a kinase activity independent role for BRAF in MAPK signaling. Cancer Cell 26, 402–413. Hartsough, E., Shao, Y., and Aplin, A.E. (2014a). Resistance to RAF inhibitors revisited. J. Invest. Dermatol. 134, 319–325. Hartsough, E.J., Basile, K.J., and Aplin, A.E. (2014b). Beneficial effects of RAF inhibitor in mutant BRAF splice variant-expressing melanoma. Mol. Cancer Res. 12, 795–802. Hartsough, E.J., Kugel, C.H., 3rd, Vido, M.J., Berger, A.C., Purwin, T.J., Goldberg, A., Davies, M.A., Schiewer, M.J., Knudsen, K.E., Bollag, G., and Aplin, A.E. (2018). Response and resistance to paradox-breaking BRAF inhibitor in melanomas in vivo and ex vivo. Mol. Cancer Ther. 17, 84–95. Hatzivassiliou, G., Song, K., Yen, I., Brandhuber, B.J., Anderson, D.J., Alvarado, R., Ludlam, M.J., Stokoe, D., Gloor, S.L., Vigers, G., et al. (2010). RAF inhibitors prime wild-type RAF to activate the MAPK pathway and enhance growth. Nature 464, 431–435. Johnson, D.B., Childress, M.A., Chalmers, Z.R., Frampton, G.M., Ali, S.M., Rubinstein, S.M., Fabrizio, D., Ross, J.S., Balasubramanian, S., Miller, V.A., et al. (2018). BRAF internal deletions and resistance to BRAF/MEK inhibitor therapy. Pigment Cell Melanoma Res. 31, 432–436. Jones, D.T.W., Kocialkowski, S., Liu, L., Pearson, D.M., Ba¨cklund, L.M., Ichimura, K., and Collins, V.P. (2008). Tandem duplication producing a novel oncogenic BRAF fusion gene defines the majority of pilocytic astrocytomas. Cancer Res. 68, 8673–8677. Karoulia, Z., Wu, Y., Ahmed, T.A., Xin, Q., Bollard, J., Krepler, C., Wu, X., Zhang, C., Bollag, G., Herlyn, M., et al. (2016). An Integrated Model of RAF Inhibitor Action Predicts Inhibitor Activity against Oncogenic BRAF Signaling. Cancer Cell 30, 485–498.

Cell Reports 25, 1501–1510, November 6, 2018 1509

Karoulia, Z., Gavathiotis, E., and Poulikakos, P.I. (2017). New perspectives for targeting RAF kinase in human cancer. Nat. Rev. Cancer 17, 676–691.

Distinct requirement for an intact dimer interface in wild-type, V600E and kinase-dead B-Raf signalling. EMBO J. 31, 2629–2647.

Kemper, K., Krijgsman, O., Kong, X., Cornelissen-Steijger, P., Shahrabi, A., Weeber, F., van der Velden, D.L., Bleijerveld, O.B., Kuilman, T., Kluin, R.J.C., et al. (2016). BRAF(V600E) Kinase Domain Duplication Identified in TherapyRefractory Melanoma Patient-Derived Xenografts. Cell Rep. 16, 263–277.

Rushworth, L.K., Hindley, A.D., O’Neill, E., and Kolch, W. (2006). Regulation and role of Raf-1/B-Raf heterodimerization. Mol. Cell. Biol. 26, 2262–2272.

Kulkarni, A., Al-Hraishawi, H., Simhadri, S., Hirshfield, K.M., Chen, S., Pine, S., Jeyamohan, C., Sokol, L., Ali, S., Teo, M.L., et al. (2017). BRAF Fusion as a Novel Mechanism of Acquired Resistance to Vemurafenib in BRAFV600E Mutant Melanoma. Clin. Cancer Res. 23, 5631–5638. Lin, A., Rodriguez, F.J., Karajannis, M.A., Williams, S.C., Legault, G., Zagzag, D., Burger, P.C., Allen, J.C., Eberhart, C.G., and Bar, E.E. (2012). BRAF alterations in primary glial and glioneuronal neoplasms of the central nervous system with identification of 2 novel KIAA1549:BRAF fusion variants. J. Neuropathol. Exp. Neurol. 71, 66–72. MacNicol, M.C., Muslin, A.J., and MacNicol, A.M. (2000). Disruption of the 14-3-3 binding site within the B-Raf kinase domain uncouples catalytic activity from PC12 cell differentiation. J. Biol. Chem. 275, 3803–3809. Moriceau, G., Hugo, W., Hong, A., Shi, H., Kong, X., Yu, C.C., Koya, R.C., Samatar, A.A., Khanlou, N., Braun, J., et al. (2015). Tunable-combinatorial mechanisms of acquired resistance limit the efficacy of BRAF/MEK cotargeting but result in melanoma drug addiction. Cancer Cell 27, 240–256. €ney, T.M., Liphardt, J.T., Nan, X., Collisson, E.A., Lewis, S., Huang, J., Tamgu McCormick, F., Gray, J.W., and Chu, S. (2013). Single-molecule superresolution imaging allows quantitative analysis of RAF multimer formation and signaling. Proc. Natl. Acad. Sci. USA 110, 18519–18524. Nazarian, R., Shi, H., Wang, Q., Kong, X., Koya, R.C., Lee, H., Chen, Z., Lee, M.-K., Attar, N., Sazegar, H., et al. (2010). Melanomas acquire resistance to B-RAF(V600E) inhibition by RTK or N-RAS upregulation. Nature 468, 973–977. Peng, S.-B., Henry, J.R., Kaufman, M.D., Lu, W.-P., Smith, B.D., Vogeti, S., Rutkoski, T.J., Wise, S., Chun, L., Zhang, Y., et al. (2015). Inhibition of RAF Isoforms and Active Dimers by LY3009120 Leads to Anti-tumor Activities in RAS or BRAF Mutant Cancers. Cancer Cell 28, 384–398. Poulikakos, P.I., Zhang, C., Bollag, G., Shokat, K.M., and Rosen, N. (2010). RAF inhibitors transactivate RAF dimers and ERK signalling in cells with wild-type BRAF. Nature 464, 427–430. Poulikakos, P.I., Persaud, Y., Janakiraman, M., Kong, X., Ng, C., Moriceau, G., Shi, H., Atefi, M., Titz, B., Gabay, M.T., et al. (2011). RAF inhibitor resistance is mediated by dimerization of aberrantly spliced BRAF(V600E). Nature 480, 387–390. Ritt, D.A., Monson, D.M., Specht, S.I., and Morrison, D.K. (2010). Impact of feedback phosphorylation and Raf heterodimerization on normal and mutant B-Raf signaling. Mol. Cell. Biol. 30, 806–819. Ro¨ring, M., Herr, R., Fiala, G.J., Heilmann, K., Braun, S., Eisenhardt, A.E., Halbach, S., Capper, D., von Deimling, A., Schamel, W.W., et al. (2012).

1510 Cell Reports 25, 1501–1510, November 6, 2018

Shen, C.H., Yuan, P., Perez-Lorenzo, R., Zhang, Y., Lee, S.X., Ou, Y., Asara, J.M., Cantley, L.C., and Zheng, B. (2013). Phosphorylation of BRAF by AMPK impairs BRAF-KSR1 association and cell proliferation. Mol. Cell 52, 161–172. Shi, H., Moriceau, G., Kong, X., Lee, M.-K., Lee, H., Koya, R.C., Ng, C., Chodon, T., Scolyer, R.A., Dahlman, K.B., et al. (2012). Melanoma whole-exome sequencing identifies (V600E)B-RAF amplification-mediated acquired B-RAF inhibitor resistance. Nat. Commun. 3, 724. Shi, H., Hugo, W., Kong, X., Hong, A., Koya, R.C., Moriceau, G., Chodon, T., Guo, R., Johnson, D.B., Dahlman, K.B., et al. (2014). Acquired resistance and clonal evolution in melanoma during BRAF inhibitor therapy. Cancer Discov. 4, 80–93. Sievert, A.J., Lang, S.-S., Boucher, K.L., Madsen, P.J., Slaunwhite, E., Choudhari, N., Kellet, M., Storm, P.B., and Resnick, A.C. (2013). Paradoxical activation and RAF inhibitor resistance of BRAF protein kinase fusions characterizing pediatric astrocytomas. Proc. Natl. Acad. Sci. USA 110, 5957–5962. Sosman, J.A., Kim, K.B., Schuchter, L., Gonzalez, R., Pavlick, A.C., Weber, J.S., McArthur, G.A., Hutson, T.E., Moschos, S.J., Flaherty, K.T., et al. (2012). Survival in BRAF V600-mutant advanced melanoma treated with vemurafenib. N. Engl. J. Med. 366, 707–714. Spofford, L.S., Abel, E.V., Boisvert-Adamo, K., and Aplin, A.E. (2006). Cyclin D3 expression in melanoma cells is regulated by adhesion-dependent phosphatidylinositol 3-kinase signaling and contributes to G1-S progression. J. Biol. Chem. 281, 25644–25651. Thevakumaran, N., Lavoie, H., Critton, D.A., Tebben, A., Marinier, A., Sicheri, F., and Therrien, M. (2015). Crystal structure of a BRAF kinase domain monomer explains basis for allosteric regulation. Nat. Struct. Mol. Biol. 22, 37–43. Wagle, N., Van Allen, E.M., Treacy, D.J., Frederick, D.T., Cooper, Z.A., TaylorWeiner, A., Rosenberg, M., Goetz, E.M., Sullivan, R.J., Farlow, D.N., et al. (2014). MAP kinase pathway alterations in BRAF-mutant melanoma patients with acquired resistance to combined RAF/MEK inhibition. Cancer Discov. 4, 61–68. Wan, P.T., Garnett, M.J., Roe, S.M., Lee, S., Niculescu-Duvaz, D., Good, V.M., Jones, C.M., Marshall, C.J., Springer, C.J., Barford, D., and Marais, R.; Cancer Genome Project (2004). Mechanism of activation of the RAF-ERK signaling pathway by oncogenic mutations of B-RAF. Cell 116, 855–867. Weber, C.K., Slupsky, J.R., Kalmes, H.A., and Rapp, U.R. (2001). Active Ras induces heterodimerization of cRaf and BRaf. Cancer Res. 61, 3595–3598. Yao, Z., Torres, N.M., Tao, A., Gao, Y., Luo, L., Li, Q., de Stanchina, E., AbdelWahab, O., Solit, D.B., Poulikakos, P.I., and Rosen, N. (2015). BRAF Mutants Evade ERK-Dependent Feedback by Different Mechanisms that Determine Their Sensitivity to Pharmacologic Inhibition. Cancer Cell 28, 370–383.

STAR+METHODS KEY RESOURCES TABLE

REAGENT or RESOURCE

SOURCE

IDENTIFIER

Actin

Sigma-Aldrich

Cat# A5316; RRID: AB_476743

AMPKa1

Cell Signaling Technology

#2532

ARAF

Santa Cruz Biotechnology

Cat# sc-408; RRID: AB_630882

BRAF

Santa Cruz Biotechnology

Cat# sc-5284; RRID: AB_626760

BRAF V600E

Spring Bioscience

Cat# E19290; RRID: AB_11203850

CRAF

Santa Cruz Biotechnology

Cat# sc-7267; RRID: AB_628196

Antibodies

ERK1/2

Cell Signaling Technology

Cat# 9102; RRID: AB_330744

ERK2

Santa Cruz Biotechnology

Cat# sc-1647; RRID: AB_627547

FLAG

Cell Signaling Technology

Cat# 2368; RRID: AB_2217020

FRA-1

Santa Cruz Biotechnology

Cat# sc-605; RRID: AB_2106927

Hsp90

Cell Signaling Technology

Cat# 4877; RRID: AB_2233307

Myc-epitope tag

Cell Signaling Technology

Cat# 2278; RRID: AB_490778

MEK1/2

Cell Signaling Technology

Cat# 9122; RRID: AB_823567

MEK1

Cell Signaling Technology

Cat# 9124; RRID: AB_330804

MEK1 (for immunoprecipitations)

Cell Signaling Technology

Cat# 2352; RRID: AB_10693788

MEK2

Cell Signaling Technology

Cat# 9125; RRID: AB_2140644

pan 14-3-3

Cell Signaling Technology

Cat# 8312; RRID: AB_10860606

phospho-ACC S79

Cell Signaling Technology

Cat# 3661; RRID: AB_330337

phospho-BRAF S729

Abcam

Cat# ab124794; RRID_AB_10976055

phospho-ERK1/2 T202/Y204

Cell Signaling Technology

Cat# 9101; RRID: AB_331646

phospho-MEK1/2 S217/221

Cell Signaling Technology

Cat# 9121; RRID: AB_331649

ThermoFisher Sci

C4040-10 E2311

Bacterial and Virus Strains One Shot TOP10 competent cells Chemicals, Peptides, and Recombinant Proteins FuGENE HD Transfection Reagent

Promega Corp

Lipofectamine RNAiMAX

Invitrogen/ThermoFisher

13778030

anti-FLAG affinity gel

Sigma-Aldrich

Cat# A2220; RRID: AB_10063035

protein A/G PLUS-Agarose

Santa Cruz Biotechnology

Cat# sc-2003; RRID: AB_10201400

QuikChange II XL Site-Directed Mutagenesis

Agilent Technologies

#200521

Click-it EdU Alexa Flour 647 Flow Cytometry Assay Kit

Invitrogen

C10634

D-luciferin

Caliper Life Sciences, now Perkin Elmer

122799

A375

ATCC

Cat# CRL-3224; RRID: CVCL_6233

293T

ATCC

Cat# CRL-3216; RRID: CVCL_0063

Jackson Laboratory

Cat# JAX: 007850; RRID: IMSR_JAX: 007850

Critical Commercial Assays

Experimental Models: Cell Lines

Experimental Models: Organisms/Strains Athymic mice, J:NU Oligonucleotides Non-targeting control siRNA

Dharmacon Inc.

D-001810-01

PRKAA1/AMPKa1 siRNAs

Dharmacon Inc.

D-005027-01 and 02

CRAF siRNA

Dharmacon Inc.

D-003601

Software and Algorithms FlowJo

FlowJo, LLC

Quantity One

BioRad

Cell Reports 25, 1501–1510.e1–e3, November 6, 2018 e1

CONTACT FOR REAGENT AND RESOURCE SHARING Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Andrew Aplin ([email protected]). EXPERIMENTAL MODEL AND SUBJECT DETAILS In vivo experiments Seven-week old female nude mice (Jackson Laboratory, Bar Harbor, ME stock# 007850) were injected with 1 3 106 1205LuTR GAL4-ELK1 reporter cells (Basile et al., 2013) transduced to express BRAF V600E DEx3-10 (n = 8 mice) or BRAF V600E DEx3-10 S729A (n = 5 mice). After palpable tumors were formed (day 0) mice were fed chow laced with PLX4720 (417 mg/kg). Mice were randomly assigned to experimental groups. Tumor volumes were recorded every 3-4 days with digital calipers and utilizing the formula: (length x width2) x 0.52. Firefly luciferase was measured using IVIS lumina XR after intraperitoneal injection of 100 mL of 150 mg/ml stock of D-luciferin (Caliper Life Sciences, Hopkinton, MA). The luciferase data in each group is normalized to the respective day 0. All mouse experiments were performed at Thomas Jefferson University (Association for Assessment and Accreditation of Laboratory Animal Care-accredited) and approved by the Institutional Animal Care and Use Committee (IACUC). Cell lines 1205LuTR and PRT cell lines were cultured in 37 C humidified chamber with 5% CO2 in MCDB 153 media containing 20% Leibovitz-L15 media, 2% fetal bovine serum, 0.2% sodium bicarbonate, and 5 mg/ml insulin. A375 and 293FT cells were grown in DMEM supplemented with 10% FBS. 1205Lu cells and A375 cells were derived from a male patient and a female patient, respectively. The generation and use of tetracycline repressor (TR) expressing sublines, 1205LuTR and A375TR, capable of inducible expression of transduced plasmids has been previously described (Abel and Aplin, 2010; Basile et al., 2013). 1205LuTR and A375TR cells expressing doxycycline (DOX)-inducible BRAF isoforms were generated in a similar manner. The generation of the PRT cell lines have been previously described (Basile et al., 2013). Cell lines were validated by Sanger sequence of the BRAF V600 region and by STR analysis. METHOD DETAILS All procedures were performed at Thomas Jefferson University. Point-mutagenesis and epitope tagging Point mutagenesis was performed using the QuikChange II XL Site-Directed Mutagenesis kit (Agilent Technologies, Santa Clara, CA). Template DNA was generated from cDNA libraries (Basile et al., 2013). Epitope tagging was achieved by adding the myc or FLAG nucleotide sequence to the 50 primer. Western blotting Western blotting was performed, as previously described (Spofford et al., 2006). Chemiluminescence was visualized on a Versadoc MultiImager and quantitated using Quantity One software (BioRad, Hercules, CA). Transfections Transient transfections in 293FT cells were performed using the FuGENE HD Transfection Reagent and protocol (Promega Corp, Madison, WI). Briefly, 9 mL of FuGENE reagent was mixed with 150 mL OptiMEM and incubated for 5 minutes at room temperature. 3 mg of plasmid DNA was added to this mixture and incubated for 20 minutes at room temperature. The DNA-Fugene reagent mixture was added drop-wise to a 6-well plate and 2x106 293FT cells suspended in 3ml of media was added. Cells were collected after 48 hr and assessed by western blot. Melanoma cells were transfected with siRNAs at a final concentration of 25 nmol/L using Lipofectamine RNAiMAX (Invitrogen, Carlsbad, CA). Cell growth assays and S phase entry analysis Cells were seeded (1 3 104 in 12-well plates or 2 3 104 cells per 6-well plate) overnight and then treated with DMSO or DOX (100 ng/ml) for 24 hr to induce expression of indicated proteins. Cells were then treated with indicated concentrations of PLX4720; the media and drugs were replenished every 2-3 days. Continuous cell growth was monitored in 12-well plates every 2 hr for 5 days using the IncuCyte Kinetic Imaging System (Essen BioScience, MI, USA). The relative confluency was analyzed using IncuCyte software. 6-well plates were treated for 8 days and subsequently stained with 0.2% crystal violet in formalin. Colonies were imaged on a Nikon Eclipse Ti inverted microscope (Nikon, Melville, NY) with NIS-Elements AR 3.00 software and scanned for quantitation using ImageJ software.

e2 Cell Reports 25, 1501–1510.e1–e3, November 6, 2018

For entry in S-phase entry analysis, cells (2.0 3 105) were seeded in 6-well plates overnight. Cells were treated with DMSO or 100ng/ml doxycycline. 24 hr later, DMSO or 1 mM PLX4720 was added for 48 hr; the thymidine analog, EdU was added for the final 16 hr at a final concentration of 10 mMol/L. Cells were then analyzed using the Click-it EdU Alexa Flour 647 Flow Cytometry Assay Kit (Invitrogen, Carlsbad, CA). EdU staining was quantified on BD FacsCalibur and data were analyzed with FlowJo software. Data points are shown as averages of three experimental replicates. Immunoprecipitation 1205LuTR cells transduced to inducibly express the indicated proteins were seeded (1.0 3 106 cells) on 10 cm plates overnight. Cells were then dosed for 48 hr with 5 ng/ml DOX to induce protein expression. The associated level of expression at this DOX concentration provided a dose response that matched previously published resistant splice variant expressing cell lines (Poulikakos et al., 2011). When indicated, plates were treated with DMSO or indicated inhibitor for an additional 4 hr. When dose curves were indicated, cells were treated with 0.001, 0.005, 0.01, 0.05, 0.1, 0.5, 1, 5, or 10 mM of respective drugs. Cells were collected in NP-40 lysis buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 1% NP40, 10% glycerol, 1 mM EDTA). 20 mL of pre-washed anti-FLAG Affinity gel was used to immunoprecipitate FLAG epitope tagged target at 4 C overnight. Endogenous MEK1 was immunoprecipitated with 5 mL MEK1 antibody for 1 hr at room temperature followed by incubation with 20 mL protein A/G PLUS-Agarose at 4 C overnight. Samples were then washed 3x with cold TBS, resuspended in Laemmli lysis buffer and boiled for 5 minutes. Equal volume was loaded on acrylamide gels for western analysis. QUANTIFICATION AND STATISTICAL ANALYSIS Unless otherwise noted, significance was determined by two-tailed Student’s t test assuming unequal variance. Data presented represent mean values and the error bars represent standard error of the mean (SEM). Significant values are indicated as follows: * < 0.05, ** % 0.01, *** % 0.001, and **** % 0.0001. Statistical details of experiments, including the number of replicates, can be found in the figure legends. For determining significant differences of in vivo growth kinetics, the log transformed tumor volumes were analyzed in a general linear model with day, genotype and their interaction as fixed effects. The model included random effects of animal in both slope and intercept. Model assumptions were validated by examining the residuals.

Cell Reports 25, 1501–1510.e1–e3, November 6, 2018 e3