br Introduction HH GLI signaling is one of

Introduction HH-GLI signaling is one of the major pathways involved in both normal and neoplastic development [[1], [2], [3], [4]]. Inappropriate activation of this pathway has been linked to cancer, including medulloblastoma, rhabdomyosarcoma, basal cell carcinoma and cancers of the lung, stomach, pancreas, colon and prostate [5,6]. Medulloblastoma (MB) and rhabdomyosarcoma (RMS) are heterogeneous and aggressive Genomic DNA Isolation Kit of childhood tumors. Subtypes of both MB (Sonic HH (SHH)) [7] and HH-linked RMS tumors [4,8,9] have been characterized with abnormally activated HH signaling. Importantly, both SHH MB and RMS tumors often harbor recurrent mutations in the p53 gene, and a p53-null background dramatically enhances tumorigenesis in mouse models [10,11]. Consequently, in these tumors, in addition to targeting HH signaling, therapeutic strategies aiming at reactivating p53 may have certain advantages. In mammals, the canonical HH signaling pathway is initiated by extracellular HH ligands binding to Patched (PTCH1, PTCH2) [4,12]. This releases the PTCH1 inhibitory effects on the signaling molecule Smoothened (SMO), facilitating downstream activation of the Glioma-associated oncogenes (GLI1, GLI2, GLI3) family of Zn-finger transcription factors. Upon activation, the GLI factors translocate to the nucleus and promote transcription of HH target genes [2,3,12,13]. While GLI1 acts as an activator of the pathway, amplifying the HH signal, GLI2 and especially GLI3 have both activator and repressor functions. Notably, effective inhibition of the GLIs can represent a promising treatment option for HH-activated cancers. Studies on the small molecule inhibitor of GLI1/2, GANT61 (GLI antagonist 61) have proven its efficacy in suppressing HH signaling-dependent cancer cell growth in vitro and in vivo [14], including RMS [15], neuroblastoma [16], pancreas [17], small cell lung [18] and hepatocellular cancers [19]. Another small molecule, NSC652287 [20], later named RITA (reactivation of p53 and induction of tumor cell apoptosis) was identified as a potential tumor suppressor. In early studies RITA was proposed to exert its effects in a p53-dependent manner through activation of both mutant and wild-type p53, resulting in tumor cell growth inhibition and p53-dependent apoptosis in vitro and in vivo [[21], [22], [23], [24]]. However, several recent studies suggest p53-independent effects of RITA, questioning its selective binding to p53 [25], and even suggesting that p53 might be dispensable for RITA activity, as its effects are largely mediated through induction of DNA damage [26], irrespective of the cell's p53 status [27]. Additionally, it was demonstrated that whereas p53 has a central role for RITA-mediated effects in wild-type cells, neither p53, nor the other two homologs of p53 (p63 or p73) are essential for the RITA response in mutant or p53 null cells [28]. Notably, RITA-induced apoptosis is predominantly mediated by the Stress-activated protein kinase/Jun amino-terminal kinase (SAPK/JNK) and p38 Mitogen-activated protein kinase (MAPK) pathways [28,29], which are known to be activated in response to a wide range of extra- and intra-cellular stress stimuli [30]. Other studies have demonstrated reactive oxygen species (ROS)-dependent JNK activation as a possible mode of RITA action [28], which can induce DNA damage, with RITA interacting and inhibiting Thioredoxin reductase 1 (TrxR1), leading to further ROS induction [31,32]. In addition, the JNK pathway has been linked to HH signaling and several studies have indicated interactions between phospho-JNK (activated form) and GLI proteins [[33], [34], [35], [36]]. These studies provide an appealing hint on the mechanism of a possible RITA - > JNK - > HH-GLI axis. In the present study, we focused on tumor cell lines of distinct origin and p53 background, the Rh36 RMS cells with wild-type p53 and the Daoy MB cells carrying a homozygous cysteine to phenylalanine mutation in codon 242 of the TP53 gene [37,38]. We found that RITA can effectively downregulate HH signaling in these cell lines irrespective of siRNA mediated p53 depletion. Moreover, siRNA depletion of the upstream activator of canonical HH signaling SMO did not abrogate the response of HH target genes to RITA treatment, suggesting that RITA acts downstream of SMO. Remarkably, RITA was capable to reduce HH signaling activity even in the context of HH pathway activation by SAG treatment or GLI1 overexpression. In addition, we demonstrated that this downregulation of HH signaling is mediated by ROS-independent activation of JNK kinase, since inhibition of JNK but not of ROS accumulation fully reverted RITA's impact on HH target genes. Furthermore, the cytotoxic effects of RITA were quite distinct compared to the widely used DNA damaging agents doxorubicin and oxaliplatin. Surprisingly, although RITA and GANT61 co-administration enhanced the effects of each drug on cell proliferation and apoptosis, there was a similar reduction of tumor growth in mouse Rh36 xenografts following RITA, GANT61 or combinatorial treatment. Finally, compared to single treatment, co-administration of the drugs resulted in a more stable reduction of tumor volume, reduced tumor cell proliferation and was associated with an expanded spectrum of GO terms and KEGG pathways involved in cellular growth.