The c-Jun/RHOB/AKT pathway confers resistance of BRAF-mutant melanoma cells to MAPK inhibitors.

The response of BRAF-mutant melanoma patients to BRAF inhibitors is dramatically impaired by secondary resistances and rapid relapse. So far, the molecular mechanisms driving these resistances are not completely understood. Here, we show that, in BRAF-mutant melanoma cells, inhibition of BRAF or its target MEK induces RHOB expression by a mechanism that depends on the transcription factor c-Jun. In those cells, RHOB deficiency causes hypersensitivity to BRAF and MEK inhibitors-induced apoptosis. Supporting these results, loss of RHOB expression in metastatic melanoma tissues is associated with an increased progression-free survival of BRAF-mutant patients treated with vemurafenib. Following BRAF inhibition, RHOB activates AKT whose inhibition causes hypersensitivity of BRAF-mutant melanoma cells to BRAF inhibitors. In mice, AKT inhibition synergizes with vemurafenib to block tumor growth of BRAF-mutant metastatic melanoma. Our findings reveal that BRAF inhibition activates a c-Jun/RHOB/AKT pathway that promotes tumor cell survival and further support a role of this pathway in the resistance of melanoma to vemurafenib. Our data also highlight the importance of using RHOB tumor levels as a biomarker to predict vemurafenib patient's response and to select those that would benefit of the combination with AKT inhibitors.

Generation of a single chain antibody variable fragment (scFv) to sense selectively RhoB activation.

Determining the cellular level of activated form of RhoGTPases is of key importance to understand their regulatory functions in cell physiopathology. We previously reported scFvC1, that selectively bind to the GTP-bound form of RhoA, RhoB and RhoC. In this present study we generate, by molecular evolution, a new phage library to isolate scFvs displaying high affinity and selectivity to RhoA and RhoB. Using phage display affinity maturation against the GTP-locked mutant RhoAL63, we isolated scFvs against RhoA active conformation that display Kd values at the nanomolar range, which corresponded to an increase of affinity of three orders of magnitude compared to scFvC1. Although a majority of these evolved scFvs remained selective towards the active conformation of RhoA, RhoB and RhoC, we identified some scFvs that bind to RhoA and RhoC but not to RhoB activated form. Alternatively, we performed a substractive panning towards RhoB, and isolated the scFvE3 exhibiting a 10 times higher affinity for RhoB than RhoA activated forms. We showed the peculiar ability of scFvE3 to detect RhoB but not RhoA GTP-bound form in cell extracts overexpressing Guanine nucleotide Exchange Factor XPLN as well as in EGF stimulated HeLa cells. Our results demonstrated the ability of scFvs to distinguish RhoB from RhoA GTP-bound form and provide new selective tools to analyze the cell biology of RhoB GTPase regulation.

RhoB promotes cancer initiation by protecting keratinocytes from UVB-induced apoptosis but limits tumor aggressiveness.

The role of UVB-induced apoptosis in the formation of squamous cell carcinoma (SCC) is recognized. We previously identified the small RhoB (Ras homolog gene family, member B) GTPase, an early response gene to cellular stress, as a critical protein controlling apoptosis of human keratinocytes after UVB exposure. Here we generated SKH1 (hairless immunocompetent mouse) mice invalidated for RhoB to evaluate its role in UVB-induced skin carcinogenesis in vivo. We show that rhob-/- mice have a lower risk of developing UVB-induced keratotic tumors and actinic keratosis that is associated with a higher sensitivity of UVB-exposed keratinocytes to apoptosis. We extend this observation to primary cultures of normal human keratinocytes in which RhoB was downregulated with small interfering RNA (siRNA) and further show that the hypersensitivity to apoptosis depends on B-cell lymphoma 2 (Bcl-2) downregulation. In rhob-/- mice, the UVB-induced tumors were preferentially undifferentiated and highly proliferative. Finally, we show in humans an almost constant loss of RhoB expression in undifferentiated SCCs. These undifferentiated and RhoB-deficient tumors have elevated phosphorylated histone H2AX (γH2AX) and 53BP1, two markers of DNA double-strand breaks. Together, our results indicate that UVB-induced RhoB expression participates in in vivo SCC initiation by increasing keratinocyte survival. Conversely, RhoB may limit tumor aggressiveness as loss of RhoB expression in tumor cells is associated with tumor progression.

RhoA and RhoC differentially modulate estrogen receptor α recruitment, transcriptional activities, and expression in breast cancer cells (MCF-7).

PURPOSE: RhoA and RhoC are closely related, small GTPases that are clearly involved in breast cancer tumorigenesis. Nonetheless, their specific roles in the control of estrogen receptor alpha (ERα) activities have not been elucidated. METHODS: We used siRNA sequences to specifically down-regulate RhoA and RhoC expression in ERα-positive breast adenocarcinoma MCF-7 cells. We then analyzed the consequences of down-regulation on ERα expression, ERα recruitment to the promoters of four target genes, and the mRNA levels of those genes. RESULTS: We demonstrated that RhoA and RhoC clearly and similarly modulated ERα recruitment to the vitellogenin estrogen responsive element (ERE) present in a luciferase reporter gene and to the promoters of progesterone receptor (PR), cathepsin D, and pS2 genes. Besides, RhoA up-regulated the ERE-luciferase reporter gene activity and PR mRNA expression and tended to down-regulate cathepsin D and pS2 mRNA expression. Conversely, RhoC inhibition had no significant effect at the mRNA level. Furthermore, RhoA inhibition, and to a lesser extent RhoC inhibition, increased ERα expression. No alteration in ERα mRNA levels was observed, suggesting potential post-translational control. CONCLUSIONS: Taken together, our results strongly suggest that RhoA and RhoC play different, but clear, roles in ERα signaling. These GTPases are definitely involved, along with RhoB, in ERα recruitment and, to some extent, ERα cofactor balance. We hypothesize a differential role of RhoA in breast cancer tumors that depend on hormone status.

RhoB modifies estrogen responses in breast cancer cells by influencing expression of the estrogen receptor.

INTRODUCTION: RhoB has been reported to exert positive and negative effects on cancer pathophysiology but an understanding of its role in breast cancer remains incomplete. Analysis of data from the Oncomine database showed a positive correlation between RhoB expression and positivity for both estrogen receptor alpha (ERα) and progesterone receptor (PR). METHODS: This finding was validated by our analysis of a tissue microarray constructed from a cohort of 113 patients and then investigated in human cell models. RESULTS: We found that RhoB expression in tissue was strongly correlated with ERα and PR expression and inversely correlated with tumor grade, tumor size and count of mitosis. In human breast cancer cell lines, RhoB attenuation was associated with reduced expression of both ERα and PR, whereas elevation of RhoB was found to be associated with ERα overexpression. Mechanistic investigations suggested that RhoB modulates ERα expression, controlling both its protein and mRNA levels, and that RhoB modulates PR expression by accentuating the recruitment of ERα and other major co-regulators to the promoter of PR gene. A major consequence of RhoB modulation was that RhoB differentially regulated the proliferation of breast cancer cell lines. Interestingly, we documented crosstalk between RhoB and ERα, with estrogen treatment leading to RhoB activation. CONCLUSION: Taken together, our findings offer evidence that in human breast cancer RhoB acts as a positive function to promote expression of ERα and PR in a manner correlated with cell proliferation.

Identification of a GTP-bound Rho specific scFv molecular sensor by phage display selection.

BACKGROUND: The Rho GTPases A, B and C proteins, members of the Rho family whose activity is regulated by GDP/GTP cycling, function in many cellular pathways controlling proliferation and have recently been implicated in tumorigenesis. Although overexpression of Rho GTPases has been correlated with tumorigenesis, only their GTP-bound forms are able to activate the signalling pathways implicated in tumorigenesis. Thus, the focus of much recent research has been to identify biological tools capable of quantifying the level of cellular GTP-bound Rho, or determining the subcellular location of activation. However useful, these tools used to study the mechanism of Rho activation still have limitations. The aim of the present work was to employ phage display to identify a conformationally-specific single chain fragment variable (scFv) that recognizes the active, GTP-bound, form of Rho GTPases and is able to discriminate it from the inactive, GDP-bound, Rho in endogenous settings. RESULTS: After five rounds of phage selection using a constitutively activated mutant of RhoB (RhoBQ63L), three scFvs (A8, C1 and D11) were selected for subsequent analysis. Further biochemical characterization was pursued for the single clone, C1, exhibiting an scFv structure. C1 was selective for the GTP-bound form of RhoA, RhoB, as well as RhoC, and failed to recognize GTP-loaded Rac1 or Cdc42, two other members of the Rho family. To enhance its production, soluble C1 was expressed in fusion with the N-terminal domain of phage protein pIII (scFv C1-N1N2), it appeared specifically associated with GTP-loaded recombinant RhoA and RhoB via immunoprecipitation, and endogenous activated Rho in HeLa cells as determined by immunofluorescence. CONCLUSION: We identified an antibody, C1-N1N2, specific for the GTP-bound form of RhoB from a phage library, and confirmed its specificity towards GTP-bound RhoA and RhoC, as well as RhoB. The success of C1-N1N2 in discriminating activated Rho in immunofluorescence studies implies that this new tool, in collaboration with currently used RhoA and B antibodies, has the potential to analyze Rho activation in cell function and tumor development.

Prenylation inhibitors stimulate both estrogen receptor alpha transcriptional activity through AF-1 and AF-2 and estrogen receptor beta transcriptional activity.

INTRODUCTION: We showed in a previous study that prenylated proteins play a role in estradiol stimulation of proliferation. However, these proteins antagonize the ability of estrogen receptor (ER) alpha to stimulate estrogen response element (ERE)-dependent transcriptional activity, potentially through the formation of a co-regulator complex. The present study investigates, in further detail, how prenylated proteins modulate the transcriptional activities mediated by ERalpha and by ERbeta. METHODS: The ERE-beta-globin-Luc-SV-Neo plasmid was either stably transfected into MCF-7 cells or HeLa cells (MELN cells and HELN cells, respectively) or transiently transfected into MCF-7 cells using polyethylenimine. Cells deprived of estradiol were analyzed for ERE-dependent luciferase activity 16 hours after estradiol stimulation and treatment with FTI-277 (a farnesyltransferase inhibitor) or with GGTI-298 (a geranylgeranyltransferase I inhibitor). In HELN cells, the effect of prenyltransferase inhibitors on luciferase activity was compared after transient transfection of plasmids coding either the full-length ERalpha, the full-length ERbeta, the AF-1-deleted ERalpha or the AF-2-deleted ERalpha. The presence of ERalpha was then detected by immunocytochemistry in either the nuclei or the cytoplasms of MCF-7 cells. Finally, Clostridium botulinum C3 exoenzyme treatment was used to determine the involvement of Rho proteins in ERE-dependent luciferase activity. RESULTS: FTI-277 and GGTI-298 only stimulate ERE-dependent luciferase activity in stably transfected MCF-7 cells. They stimulate both ERalpha-mediated and ERbeta-mediated ERE-dependent luciferase activity in HELN cells, in the presence of and in the absence of estradiol. The roles of both AF-1 and AF-2 are significant in this effect. Nuclear ERalpha is decreased in the presence of prenyltransferase inhibitors in MCF-7 cells, again in the presence of and in the absence of estradiol. By contrast, cytoplasmic ERalpha is mainly decreased after treatment with FTI-277, in the presence of and in the absence of estradiol. The involvement of Rho proteins in ERE-dependent luciferase activity in MELN cells is clearly established. CONCLUSIONS: Together, these results demonstrate that prenylated proteins (at least RhoA, RhoB and/or RhoC) antagonize the ability of ERalpha and ERbeta to stimulate ERE-dependent transcriptional activity, potentially acting through both AF-1 and AF-2 transcriptional activities.