Oncogenic B-Raf proto-oncogene, serine/threonine kinase-mediated hedgehog signalling in the pathogenesis and targeted therapy of melanoma

Authors: Rafiq A Rather

Abstract

Melanoma is an aggressive type of skin cancer notorious for its resistance to chemotherapy, radiotherapy and immunotherapy, which greatly impacts its lethality. The hedgehog (HH) signaling cascade, originally known for its roles in embryonic development, regulates growth, proliferation and cancer stem cell (CSC) self-renewal. The glioma-associated oncogene homolog (GLI) transcription factors play crucial roles in melanoma. However, oncogenic B-Raf proto-oncogene, serine/threonine kinase (BRAF) steals the spotlight by driving the aberrant activation of HH-GLI1/2 signaling. Oncogenic BRAF-driven HH-GLI1/2 signaling imparts invasive phenotype to melanoma cells and sustains CSC self-renewal. Interestingly, the transcriptional activities of GLI1 and GLI2 are suppressed by acetylation, a process that is counteracted by the deacetylating actions of histone deacetylase (HDAC) 1/2. Therefore, inhibiting HDAC1/2 might keep GLI proteins in inactive acetylated form, thus representing an attractive druggable target. Notably, both HDAC1 and HDAC2 are induced by HH signaling, creating a positive feedback loop where HH signaling upregulates the expression of both HDAC1 and HDAC2. Selective inhibition of BRAF/HH/HDAC/GLI signaling axis is likely to unravel new therapeutic opportunities in melanoma. However, the precise contribution of oncogenic BRAF-driven HH signaling to therapy resistance and CSC renewal remains unclear and requires thorough investigation. In this article, we endeavored to explore the crosstalk between oncogenic BRAF and HH signaling, and the pivotal role this interaction plays in the self-renewal of melanoma stem cells. A better understanding of the molecular mechanisms governing these interactions is essential for improving melanoma treatment strategies and identifying new therapeutic targets.

Core Tip: Oncogenic B-Raf proto-oncogene, serine/threonine kinase activates the hedgehog signaling pathway, triggering glioma-associated oncogene homolog (GLI) 1/2 transcription factors that promote melanoma cell invasion and sustain cancer stem cell self-renewal, contributing to the malignancy’s therapeutic resistance. Histone deacetylases 1/2 (HDAC1/2) suppress the acetylation of GLI1/2, facilitating their activation. Inhibiting HDAC1/2 may stabilize GLI proteins in their inactive, acetylated form, offering a novel approach to inhibit melanoma progression. Hedgehog signaling induces HDAC1/2 expression, creating a feedback loop that amplifies GLI-mediated transcription, promoting cancer stem cell renewal. Disrupting this loop could unveil new therapeutic avenues for overcoming melanoma’s resistance to treatment.

INTRODUCTION

Melanoma is a significant global public health concern and ranks among the leading causes of cancer-related mortality in Western countries[1]. According to Cancer Statistics 2025, it is estimated that there will be approximately 8430 deaths and 104960 new cases of melanoma in situ in the United States in 2025[2]. This highly lethal disease arises as a consequence of malignant transformation of melanin-producing cells (melanocytes) in human skin through intricate mechanisms[3], making it imperative for the researchers to understand the mechanisms behind such transformation. These melanin-producing cells exhibit phenotypic prominence, yet their histological characteristics remain highly obscure[4]. In human skin, melanocytes are located within the ultraviolet (UV)-permeable range, specifically in the basal layer of the epidermis, where they are exposed to UV light, including UVA and UVB rays[5]. Melanocytes are derived from multipotent neural crest (NC) cells at the dorsal borders of the neural tube, and migrate beneath the epidermis to populate various skin tissues, forming a heterogeneous group of cells responsible for pigmentation and determining skin phototype[6,7]. Tissue-resident melanocyte stem cells (MSCs) and melanocyte precursors play integral roles in the renewal and maintenance of melanocytes[8]. It is believed that the developmental processes governing melanocyte formation have a tendency to reactivate during melanoma formation. For example, embryonic NC stem cells gene expression signature is recapitulated during tumor initiation in zebrafish melanoma model, suggesting the role of NC stem cells in development of melanoma[9]. SOX10 and RAC1, two crucial genes involved in NC development, demonstrate how their disruption can hinder melanoma formation[10]. Likewise, MSX1, MITF, PAX3 and FOXD3 are pivotal in initiating NC development, facilitating NC cell migration and contributing to melanocyte formation[11]. Their activation during the onset of melanoma enhances cellular plasticity and fosters drug resistance[11]. Melanomas have been found to exhibit additional indicators characteristic of embryonic and regenerative melanocyte lineage, such as endothelin and KIT proto-oncogene, receptor tyrosine kinase[12]. Several groups of proteins, transcription factors, extracellular secreted ligands, transmembrane (TM) receptors and intracellular signaling proteins regulate these processes. Among these, the hedgehog (HH) signaling cascade stands out as one of the primary signaling cascades activated during both embryonic development and the onset of melanoma[13]. The canonical pathway of HH signaling comprises of HH ligands, patched (PTCH) receptors, the smoothened (SMO) receptor, and glioma-associated oncogene homolog (GLI) transcription factors[14]. This pathway is known to play a meager role in adult tissues but is highly activated in many tumors including melanoma[15]. HH signaling, which is aberrantly activated in melanoma, is normally used for embryonic development, tissue homeostasis, regeneration, wound healing, stem cell maintenance and intracellular communication[16].

Besides transmitting signals for embryonic development and tumorigenesis, inappropriate activation of HH pathway imparts radio- and chemo-resistant phenotype to many solid tumors, including melanoma. As a result, prognosis of HH-driven tumors is often poor[17]. A better understanding of HH signaling is likely to open new therapeutic avenues for targeted therapy of melanoma and other solid tumors. In fact, to develop an effective anti-melanoma therapy, it is crucial to understand the mechanisms by which melanoma cells acquire chemo-resistance or how resistant cells are formed within the tumor. Consequently, there is a growing focus on conducting studies aimed at discovering novel inhibitors of the HH pathway that could be safely utilized in human intervention trials. In this review, we attempt to critically summarize the important aspects of HH pathway and its involvement in development and progression of melanoma. We also explore the role of the HH pathway in imparting a treatment-resistant phenotype to melanoma, an emerging new area of research. The hope is that understanding the precise role of HH signaling in melanoma pathogenesis may provide a rationale for improving existing cancer therapies and identifying novel targets for therapeutic intervention of melanoma.

The structural architecture of HH signaling cascade is simple. In the absence of HH ligands, PTCH TM receptor inhibits the function of SMO receptor[18]. PTCH and SMO are conserved whereas the mammalian HH proteins are diversified into, Sonic-HH (SHH), Indian-HH (IHH), and Desert-HH (DHH)[19]. These HH proteins encompasses a small family of secreted signaling proteins that jointly regulate multiple aspects of animal development, tissue homeostasis and regeneration[20]. HH signaling is initiated by binding of HH ligands to PTCH (which resides in the non-motile primary cilium) and is mediated by GLI transcription effector family, whose activity is finely tuned by a number of molecular interactions, trafficking and post-translation modifications[21,22]. The HH signaling is strongly influenced by HH interacting protein, HH acyltransferase, growth arrest-specific 1 (GAS1), cell adhesion molecule (CAM)-related/downregulated by oncogenes (CDON), and brother of CDON (BOC) and GLI processing and localization, suppressor of fused (SUFU), protein kinase A, glycogen synthase kinase 3β, casein kinase 1, G-protein-coupled receptor kinase 2, and β-transducin repeat containing E3 ubiquitin protein ligase[23]. These components are involved in HH ligand modification and cell surface binding. Moreover, increasing evidence has suggested that the subcellular localization of HH pathway components is a major regulator of its activity[24].

HH SIGNALING

HH signaling pathway plays a crucial role in intracellular communication during embryonic development, organ formation, tissue maintenance, and regeneration[25]. However, it is often dysregulated in various forms of cancer[22]. Major components of HH signal reception and transduction are as follows.

HH ligands

HH ligands comprise a small family of secreted signaling proteins that are present in majority of metazoans[19]. As morphogens, these secreted proteins are essential during embryogenesis and throughout adulthood, impacting both health and disease. HH proteins achieve their biological functions by inducing gene expression changes in target cells in a concentration-dependent manner, regulating their identity, proliferation, death, or metabolism depending on the specific tissue or organ[26]. The loss of these proteins in Drosophila embryos results in the loss of normal segmented patterns and the formation of a uniform coat of bristles, a feature reminiscent of HH coats. Consequently, Drosophila embryos lacking the Hh gene appear to resemble hedgehogs[27]. In higher animals, HH proteins are involved in tissue formation, wound repair, regeneration, stem cell maintenance and tumorigenesis[28,29]. The processes of wound healing and tumorigenesis share common molecular mechanisms, with both being influenced by HH signaling[29]. HH proteins mediate essential normal shaping (patterning) during many stages of animal development, and abnormal HH function is associated with birth defects and cancer in adult animals[29,30]. HH ligands participate in tissue pattering and growth by acting as classical morphogens during the formation of range of tissues and organs[31,32]. Morphogens are signaling molecules that, based on their concentration, shape cell patterns by triggering gene expression changes in target cells. HH ligands form a gradient of varying concentration from sites of secretion, inducing concentration-dependent differentiation of different cell types[33]. HH proteins consist of two the amino-terminal domain (HhN), responsible for biological signaling activity, and the carboxy-terminal autocatalytic domain, which cleaves HH into two parts through an intramolecular reaction and attaches a cholesterol moiety to HhN. In both flies and mammals, HhN is modified by the addition of cholesterol at its carboxyl terminus and palmitate at its amino terminus. The modified HhN is then released from the cell and moves through the extracellular space. Of the three homologues of HH morphogens expressed by mammalian cells, SHH is the best studied[25,32]. Palmitoylation and cholesterol addition facilitates membrane anchoring and assembly of SHH into multimers[34,35]. Lipid modification and multimerization together remarkably enhance the SHH signaling activity, resulting in changes to downstream effects and cellular functions governed by HH signaling[36]. Given that HH ligands are lipidated and function within the aqueous extracellular space, it is imperative that mechanisms must exist that facilitate the transportation of HH proteins within this environment. To achieve this job, dispatched, signal peptide, CUB domain and EGF-like domain containing 2 and tumor necrosis factor-alpha converting enzyme, act synergistically and cause detachment of cholesterol-modified HH ligands from the plasma membrane, resulting in production of soluble and biologically active HH morphogens[37,38]. Structurally, SHH consists of α + β sandwich core (comprised of two α-helices and a six-stranded mixed β-sheet) and two-stranded antiparallel β-sheet[39]. This type of folding pattern is unique to these proteins. However, the occurrence of tetrahedrally coordinated zinc ion in SHH-N shows close structural resemblance to zinc hydrolases such as thermolysin and carboxypeptidase[40]. In the crystal structure, zinc ion is bound to side chains of three amino acid residues - His 141, Asp 148, and His 183 and a single molecule of water[41].

SMO

SMO is a type of membrane-localized frizzled (class F) G protein-coupled receptor (GPCR) encoded by the SMO gene in humans[42]. This 7-pass integral membrane protein is highly conserved across species, spanning from flies to humans. SMO plays a pivotal role in embryonic development and tissue homeostasis[43]. Recent structural information revealed SMO is regulated by binding of various ligands and signals through G protein-dependent and -independent mechanisms. Structurally, SMO is a GPCR, consisting of N-terminal cysteine-rich domain (CRD), seven-pass transmembrane helices domain (TMD) and an intracellular C-terminal domain (hinge domain)[44]. SMO protein contains two ligand biding sites; one in heptahelical transmembrane domain (TMD) and one in the extracellular CRD[45]. The CRD is stacked atop the TMD, separated by an intervening wedge-like linker domain. Typically, SMO signaling is regulated by its trafficking within the primary cilium[46]. Signal output from SMO is influenced by its functional domains, post-translational modifications, subcellular distribution and trafficking[47]. Oncogenic forms of SMO have been identified in various types of human malignancies[48]. Although, SMO shows similarities in ligand binding with other GPCRs, the molecular basis of SMO activity and its regulation is imprecisely determined and deserves further examination. SMO expression shows positive correlation with tumor size, invasiveness, metastasis and recurrence[49]. Therefore, inhibition of SMO regresses tumor growth in melanoma models. SMO inhibitors have received huge significance in the treatment of skin and brain cancers and many such inhibitors have entered into clinical trials[50,51]. However, SMO mutations induce drug-resistance against SMO antagonists[52].

PTCH

PTCH is encoded by a gene that belongs to the PTCH gene family. Multiple isoforms of PTCH mRNA have been identified and characterized in humans and mice, which are generated by the complex alternative use of several distinct exons[53]. The most important diseases associated with PTCH1 are holoprosencephaly, basal cell nevus syndrome and melanoma[54]. There are two PTCH homolog genes in vertebrates[55,56]. PTCH1 and PTCH2 exhibit overlapping functions in the formation of embryonic structures and in tissue homeostasis[57]. And there is 73% homology between the two subtypes. Both TM proteins function as receptors for SHH, IHH and DHH ligands, and are capable of binding all HH family members with similar affinity. Both PTCH1 and PTCH2 can form a complex with SMO[55]. Physiologically, PTCH functions to repress the activity of SMO[58]. Although PTCH1 and PTCH2 functionally overlap in the repression of SMO, the spatiotemporal expression of BOC (PTCH1 co-receptor) and GAS1 (PTCH2 co-receptor) are believed to determine the outcome of HH signaling through compartmentalization and modulation of SMO-downstream signaling[59]. PTCH seems to exhibit a tumor suppressor function, as inactivation of this protein is a necessary, if not sufficient step for tumorigenesis[60]. PTCH1 is altered across various cancer types, including melanoma, with the American Association for Cancer Research Project Genomics Evidence Neoplasia Information Exchange Consortium reporting mutations in PTCH1 in about 5.54% of melanoma patients[61]. Additionally, high expression of PTCH has been observed in melanoma tumors[62]. Upon HH binding, PTCH relieves its inhibitory influence on SMO, triggering an intracellular signaling cascade that regulates the formation of the zinc finger transcription factors (such as GLI1 and GLI2) and their translocation into the nucleus[63]. In differentiated cells, the expression GLI is generally very low[64]. However, in certain cancer types, aberrant activation of GLI occurs, promoting numerous cancer hallmarks such as proliferation, survival, angiogenesis, metastasis, metabolic rewiring, and resistance to chemotherapy[64].

GLI proteins

In cells, the GLI proteins exist as three distinct zinc-finger transcription factors-GLI1, GLI2 and GLI3-which mediate HH signaling at the distal end of the pathway[65]. While GLI1 and GLI2 operate as transcriptional activators, GLI3 typically serves as a transcriptional repressor[66]. GLI2 has dual functions of activating and inhibiting transcription but mainly functions as a transcriptional activator[67]. The expression of GLI1 is often used as readout of HH pathway activation[68]. In the absence of HH ligands, the activity of SMO protein is negatively regulated by PTCH and the GLI proteins are cleaved proteolytically to form the repressor GLI, majorly obtained for GLI3, which inhibits HH signaling and downstream HH target genes[69]. However, binding of HH binding relieves the inhibitory influence of PTCH on SMO, and promotes the formation of GLIA, majorly obtained from GLI2, which facilitates the expression of HH target genes[70]. The activity of GLI transcription factors is regulated at multiple levels through phosphorylation by multitude of kinases such as SUFU, protein kinase A, phosphoinositide 3-kinases (PI3K)/mammalian target of rapamycin, glycogen synthase kinase 3β, casein kinase 1 α and activators such as dual-specificity tyrosine-phosphorylation-regulated kinase 1, rat sarcoma virus oncogene (RAS, a type of small guanosine triphosphatase) and protein kinase B (AKT). The exact mechanism of non-canonical HH signaling is still not well understood. However, it is likely to serve as an alternative activation pathway when canonical HH signaling is dysfunctional or when transduction must bypass the canonical route due to cytotoxic or inflammatory stress[71]. Both oncogenic BRAF and NRAS are known to activate the GLI proteins non-canonically[64]; however, the precise basis of this interaction is elusive. There is a need to investigate the transcriptional targets of GLI1, GLI2, and GLI3 proteins in melanoma specimens harboring BRAF, NRAS mutations, as well as in those lacking these mutations[72]. Transcriptional targets of GLI proteins could serve as biomarkers for diagnosis, prognosis, and treatment response, improving clinical outcomes for melanoma patients with BRAF mutations[73].

SUFU

SUFU is a highly conserved protein that negatively regulates HH signaling by binding to GLI zinc-finger transcription factors, thereby inhibiting the activation of target gene expression[74]. This protein is encoded by the SUFU gene in humans[75]. In mammals, the mutations in this gene have deleterious effects on embryo development and are associated with cancer-predisposing syndromes and congenital anomalies[76]. Additionally, SUFU can act as a tumor suppressor, as its depletion promotes tumorigenesis in TP53^-/-^ mice[77]. Several vertebrate homologues of SUFU are found in a wide variety of organisms[78]. Structurally, SUFU protein has two domains. However, in eukaryotic SUFU, an additional domain exists at the C-terminus of the protein which interacts with the C-terminal domain of GLI transcription factors, inhibiting their activity[79]. In the human SUFU-GLI complex, SUFU has been observed to switch between “open” and “closed” conformations. The “closed” form of SUFU is stabilized upon binding to GLI, and is inhibited by HH treatment. Conversely, the “open” form of SUFU is promoted by the dissociation of GLI and the activation of HH signaling[80]. In addition to with GLI transcription factors, SUFU has been shown to interact with peroxisomal biogenesis factor 26 (PEX26) which is required for protein import into peroxisomes[81]. The interaction between SUFU and PEX26 suggests a potential link between HH signaling and peroxisomal function. Silencing PEX26, as an unconventional mechanism, has been shown to kill drug-resistant cancer cells and prevent drug resistance in melanoma cells[82].

Co-receptors

In addition to the canonical PTCH receptors, HH ligands signal through three co-receptors: GAS1, CDON and BOC. Together, these co-receptors are needed during embryogenesis to mediate accurate HH signaling[83]. BOC and GAS1 form unique heterogeneous complexes through their interactions with PTCH1 and PTCH2 receptors, respectively, and mediate different kinetic SMO derepression programs through distinct ligand reception patterns[59]. CDON is a TM glycoprotein that binds directly to HH ligand and is required in conjunction with PTCH, BOC and GAS1 to promote HH signaling[84]. Binding of HH ligands to co-receptors GAS1, CDON and BOC leads to SMO derepression and phosphorylation of its cytoplasmic tail. SMO mediates downstream signal transduction by favoring dissociation of GLI proteins from the kinesin-family member 7 and the key intracellular HH pathway regulator SUFU[85]. The manner in which each of these co-receptors interact and contribute to HH signaling is not fully elucidated. However, in mice, knockout of all three causes abrogation of HH signaling[86].

Major protein kinases implicated in HH signaling

HH signaling is regulated by major phosphorylation events through protein kinases, which act as either positive or negative regulators (Table 1)[87-116].

Endogenous modulation of SMO

Although there is no direct physical interaction between PTCH and SMO, steroidal metabolites (such as cholesterol) are believed to mediate communication between the two receptors and act as SMO agonists[117]. The domain architecture of vertebrate PTCH (a 12-pass TM protein) and SMO (a 7-pass TM protein) show intimate links between cholesterol and HH signaling[118]. There are indications that sterol depletion inhibits the accumulation of SMO in the primary cilium[119]. Further, SMO activation is impaired through a cholesterol deficiency either by depleting cholesterol by using methyl-β-cyclodextrin in culture cells or by inhibiting the function of 7-dehydrocholesterol reductase that converts 7-dehydrocholesterol to cholesterol[120]. PTCH1 belongs to a family of resistance-nodulation-cell division pump proteins whose role in bacteria is to transport diverse molecular cargos such as antibiotics and sterols across lipid membranes[58]. It is believed that in the absence of HH ligands, PTCH1 transports the endogeneous modulators (endogenous agonists) to cell outside. PTCH1 also contains a putative sterol-sensing domain that regulates sterol metabolism[121].

An important class of endogenous SMO activators comprises of oxysterols, of which the most potent in SMO activation is 20(S)-hydroxycholesterol [20(S)-OHC][122]. Several clues make it unlikely that 20(S)-OHC is the endogenous SMO modulator. First, 20(S)-OHC could not be detected in cultured cells that are responsive to HH ligands[123]. 20(S)-OHC shows no synergism with SHH ligand in pathway activation[124]. Further, SMO mutations that impair the binding to 20(S)-OHC do not inhibit SMO activity. Recent crystal structure models of SMO action have shown that a cholesterol molecule binds to the cysteine rich domain (CRD) of the SMO and SMO lacking CRD or bearing mutations in key amino acids required for cholesterol binding, does not respond to cholesterol[125]. Cholesterol binding to SMO is competitively inhibited by both 20(S)-OHC and cyclopamine, consistent with a common binding site in the CRD[126]. Both these molecules allosterically regulate SMO protein. The cytoplasmic tail of SMO also participates in endogenous modulation of SMO. Phosphatidylinositol 4-photosphate promotes ciliary accumulation and pathway activation by binding to SMO’s cytoplasmic tail[127].

SMALL-MOLECULE MODULATORS OF HH SIGNALING

Cancer development and progression entail a multitude of biological processes, such as the malfunctioning of pivotal signaling pathways, increased drug efflux via adenosine triphosphate-binding cassette transporters, acquisition of mutations, evasion of apoptosis, activation of DNA damage response mechanisms, epithelial-mesenchymal transition, and adaptation of cancer stem cells (CSCs)[128]. The HH/PTCH/SMO/GLI signaling pathway significantly contributes to both the development of melanoma and its resistance to therapy[129]. A multitude of modulators have been identified as HH pathway antagonists. Fundamentally, these modulators target three key sites in HH HH ligands (SHH neutralizing antibodies, robotnikinin), SMO protein (cyclopamine and its derivatives IPI-926, Cyc-T and synthetic compounds GDC-0449, Cur61414, XL-139 and LDE-225); and GLI transcription factors (HPI-1, HPI-2, GANT-56 and GANT-61).

Cyclopamine, its derivatives, and others

Cyclopamine, a sterol alkaloid obtained from the corn lily plant (Veratrum californicum), acts as an antagonist of the SMO receptor and has been utilized for treating and preventing basal cell carcinoma[130]. It directly binds to the TM helices of the SMO protein, inhibiting HH signaling. At low concentrations (< 10 μmol/L), cyclopamine specifically blocks HH signaling, but at high concentrations, it induces cell death without affecting HH signaling[131]. Several semi-synthetic derivatives, such as KAAD-cyclopamine, IPI-609, IPI-926, and Cyc-T, have been developed[132]. These derivatives exhibit improved stability in both acidic and aqueous environments. IPI-926 has progressed to phase II human intervention trials[133].

Synthetic HH antagonists

Due to dearth of natural products that cause selective and specific inhibition of HH pathway, efforts have been made to identify synthetic HH antagonists with higher potency than that of cyclopamine[133]. SMO agonists such as vismodegib (also known as erivedge or GDC-0449) have been found to be effective in regressing basal cell carcinoma in PTCH^+/-^ mice[134]. This molecule is orally active and has entered into phase I and phase II human intervention trials[135]. Several synthetic compounds have been developed that bind to SMO but show no structural similarity to cyclopamine[136]. Recently, a small molecule inhibitor robotnikinin has been reported to bind SHH protein and inhibits the HH signalling in cultured skins cells. The IC50 of robotnikinin is approximately 3 μmol/L in the Shh-LIGHT2 cells[137].

Due to its pivotal role in melanoma pathogenesis, the HH signaling pathway has attracted considerable attention as a promising therapeutic target. Numerous small-molecule modulators have been identified and studied for their ability to inhibit HH signaling (Table 2)[138-168].

Inhibitors of the HH signaling cascade often show limited efficacy as monotherapy in melanoma due to pathway redundancy and compensatory signaling mechanisms. This limited efficacy is further attributed to several intrinsic (such as genetic mutations and epigenetic alterations within tumor cells) and extrinsic factors (including stromal interactions, immune cell infiltration, and extracellular matrix composition) that undermine their therapeutic potential[169,170]. The primary challenges lies in the redundancy and compensatory nature of cellular signaling transduction pathways. Melanoma cells often circumvent HH pathway inhibition by activating parallel signaling pathways such as PI3K, mitogen-activated protein kinase (MAPK), or transforming growth factor β pathways[170,171]. In particular, the BRAF/MAPK pathway can directly stimulate GLI1, independently of SMO, thereby diminishing the effectiveness of SMO-targeted therapies[101,170]. Resistance also emerges from mutations within SMO or downstream effectors. For instance, specific point mutations like D473H in SMO can hinder drug binding, rendering inhibitors such as vismodegib less effective[172]. Additionally, non-canonical activation of GLI transcription factors - mediated by oncogenes like RAS or signaling pathways such as AKT and transforming growth factor β - further complicates treatment, as this bypasses the conventional SMO-dependent axis of HH signaling. Pharmacokinetic limitations represent another obstacle. Some HH inhibitors have poor tumor penetration due to poor bioavailability, rapid metabolism, or active efflux by membrane transporters like P-glycoprotein[173,174]. Saridegib, for example, is a known substrate of P-glycoprotein, which can lead to diminished drug levels at the tumor site and subsequent resistance[175]. Not all tumors rely on HH signaling for growth and survival; in many cases, only specific subpopulations of cancer cells are HH-dependent[176]. Moreover, interactions within the tumor microenvironment, especially those involving stromal cells, can support paracrine activation of the pathway, which is not always suppressed by conventional HH inhibitors[177]. Lastly, the physiological importance of the HH pathway in normal tissue homeostasis poses a therapeutic challenge. Long-term inhibition may result in toxicities, particularly in tissues where HH signaling is crucial, such as in the maintenance of hair follicles and taste buds[178,179]. These cumulative factors underscore the complexity of HH-targeted therapies and highlight the need for combination strategies or more selective inhibitors to overcome these limitations.

Furthermore, melanoma is notoriously resistant to conventional treatments, primarily due to its high genetic heterogeneity, rapid mutation rate, immune evasion, genomic amplifications, activation of aberrant signaling pathways (such as MAPK and PI3K-AKT), presence of cancer stem-like cells, erroneous DNA repair mechanisms, and enhanced survival strategies including resistance to apoptosis[180,181]. Although immunotherapy offers a promising strategy by harnessing the patient’s immune system to recognize and eliminate tumor cells, its effectiveness in melanoma is limited[182]. This is partly because melanoma has developed sophisticated mechanisms to evade immune detection, including alterations in the tumor microenvironment and upregulation of immune checkpoints[183]. Recent studies have identified the HH signaling pathway as a contributor to tumor immune evasion and poor responses to immunotherapy, further complicating treatment outcomes[184]. Immunotherapy primarily targets immune checkpoints such as cytotoxic T-lymphocyte-associated protein 4 and programmed cell death protein 1 receptor/programmed death ligand 1 to restore T-cell activity[185], yet these interventions alone may not fully overcome the immunosuppressive barriers in melanoma[186]. Consequently, combining immunotherapy with other modalities - such as targeted therapies, chemotherapy, or radiotherapy - has shown potential to synergistically enhance antitumor immune responses, modulate the tumor microenvironment, and improve clinical outcomes in melanoma patients[187].

AN OVERVIEW OF HH PATHWAY IN MELANOMA

The HH signaling is activated in multitude of cancer including melanoma[22]. The mutations in HH pathway genes cause ligand-independent activation of this pathway. However, ligand-dependent activation of HH pathway is potentiated through crosstalk with other signaling pathways such as RAS/Raf proto-oncogene, serine/threonine kinase (RAF)/mitogen-activated protein kinase kinase (MEK)/extracellular signal-regulated kinase (ERK) pathway[188]. The way HH interacts with this pathway is likely to lead development of new combination therapies with improved antitumor efficacy and survival in animal models. The first connecting link between oncogenic BRAF and HH signaling emerged from research indicating that BRAF-mutant melanoma tumors and related cell lines exhibit heightened expression of GLI1 and GLI2[189]. Activating mutations in HH pathway components have been reported in melanoma and other types of skin cancers[190]. In addition, modification of tumor microenvironment by HH signaling in melanoma has been argued to play a key role in melanoma pathogenesis[191]. The interaction between RAS-MEK/AKT pathway and GLI1 is known to play a pivotal role in the pathogenesis of melanoma pathogenesis[192].

Crosstalk between BRAF and HH signaling in melanoma

HH pathway has been reported to be involved in the pathogenesis of melanoma and other skin cancers[193]. This pathway regulates various biological processes including cell differentiation, proliferation, tissue polarity, stem cell maintenance, embryonic patterning, embryonic development and tumor formation[194,195]. In human skin, the SHH pathway participates in the maintenance of MSCs, and in the regulation of hair follicle and sebaceous gland development[196]. Constitutively active RAS-RAF-MEK-ERK pathway has been found to induce HH signaling through translocation of GLI1 and GLI2 into the nucleus (Figure 1)[96]. There is a strong possibility that oncogenic BRAF promotes HH pathway activity in melanoma tumors and associated cell lines. However, the precise mechanism by which oncogenic BRAF enhances HH pathway activity in melanoma tumors and cell lines remains unclear and warrants further investigation. BRAF is a member of the RAF family of serine/threonine protein kinases[197]. An estimated 40% to 50% of mutated melanomas contain BRAF (V600E)[198].

The regulation of HH signaling by BRAF occurs primarily through indirect mechanisms. Studies have demonstrated that activation of the MAPK/ERK pathway can upregulate GLI1 expression and activity in various cancer types[199]. However, the interaction between MAPK/ERK and HH signaling can vary depending on the cellular context. In some instances, activation of MAPK/ERK signaling has been observed following HH pathway inhibition, indicating a complex and context-dependent relationship between these pathways[200]. Oncogenic BRAF typically activates HH signaling via MAPK/ERK pathway through two distinct direct phosphorylation, where BRAF-activated ERKs move into the nucleus to phosphorylate GLIs, and indirect phosphorylation, where ERKs phosphorylate intermediate proteins that subsequently modify GLI transcription factors[94]. Given that most malignant melanomas contain activating mutations in BRAF, many potent inhibitors of oncogenic BRAF such as vemurafenib, dabrafenib and encorafenib were developed over the last two decades[201]. Many of these BRAF inhibitors (for example, vemurafenib) show significantly high clinical response rate in patients with BRAF V600E-mutant melanomas, yet the majority of melanoma patients ultimately develop drug resistance[202]. Although multiple resistance mechanisms are known for melanoma, many of BRAF inhibitor-resistant melanomas are driven by unknown mechanisms[203,204]. Recently, HH signaling has received tremendous attention as a central regulatory pathway that imparts drug resistant phenotype to melanoma tumors. Mechanistically, HH signaling is simple and comprises of HH ligands (such as SHH, IHH, and DHH), TM receptors, and transcription factors[205]. In canonical HH signaling, binding of secreted HH ligands to the TM receptor PTCH activates the GPCR SMO, which triggers an intracellular signaling cascade leading to the formation of transcription factors GLI 1 and 2 (GLI1/GLI2) and their translocation into the nucleus (Figure 1)[101,206]. GLI1/GLI2 transcription factors have been found to play a causal role in both the resistance to BRAF V600E-targeted therapy and the proinvasive behavior of melanoma cells[189]. Consequently, melanoma cell lines with acquired drug-resistance and melanoma tissue samples show increased levels of GLI1 and GLI2 compared to naive cells and normal tissues[189]. This raises the possibility of using GLI1/GLI2 inhibitors to reverse drug-resistance in BRAF driven melanoma cells. Furthermore, aberrant HH signaling has been associated with the maintenance of melanoma stem cells (MSCs)[207]. MSCs, also known as melanoma-initiating cells, are transformed stem cells, which can self-renew, differentiate into diverse progenies, and drive continuous growth of melanoma tumors[208]. HH-driven MSCs are considered the most important source of primary tumor mass and are thought to be responsible for drug resistance phenotype of melanoma tumors. Besides, imparting drug resistance phenotype to melanoma tumors, HH-GLI signaling has been shown to play a pivotal role in self-renewal and tumorigenicity of MSCs[209]. These cells constitute a small subset of cells within a tumor that possess the capacity for unlimited self-renewal, can differentiate into all cell types found in the tumor, and exhibit resistance to many conventional therapeutic modalities that typically target the more differentiated cells comprising the bulk of the tumor[210,211]. The high degree of heterogeneity in melanomas, attributed to differences between tumors (inter-tumor heterogeneity), contributes to the lack of pan-melanoma therapies and underscores the increasingly urgent need to develop novel therapeutic modalities. We anticipate that pharmacological inhibition of GLI1 and GLI2 in the HH signaling pathway could potentially resensitize melanoma tumors and MSCs to apoptosis. Furthermore, at an epigenetic level, the transcriptional activity of GLI1 and GLI2 is inhibited by acetylation, which prevents their recruitment to target promoters, thus representing an attractive and druggable target[212]. GLI acetylation is catalyzed by the histone acetyltransferase p300 and reversed by the histone deacetylases HDAC1 and HDAC2[213,214]. HDACs remove acetyl groups from lysine residues of core histones and nonhistone proteins, resulting in a more closed chromatin structure and repression of gene expression[215]. Notably, both HDAC1 and HDAC2 are induced by HH signaling, establishing a positive loop through HH-induced upregulation of HDAC1 and HDAC2[214]. In agreement with this, HDAC inhibitor suberanilohydroxamic acid and the BRAF inhibitor vemurafenib show cooperative inhibition of BRAF V600E-melanoma xenograft growth in a mouse model[216]. Therefore, promoting GLI acetylation via inhibition of HDACs could be an effective strategy to combat drug resistance in BRAF-mutant melanoma models. However, the exact mechanism by which GLI transcriptional activity is regulated by HDACs in BRAF inhibitor-resistant melanoma cells is not well understood, highlighting the need for further exploration of the precise mechanisms underlying drug resistance in melanoma. We propose that oncogenic HH/GLI signaling plays a critical role in the therapeutic resistance of melanoma tumors and that combining BRAF inhibitors with GLI1/GLI2 inhibitors may improve therapeutic options for BRAF inhibitor-resistant melanomas. Given the critical role of HH/GLI/HDACs signaling axis in determining melanoma pathogenesis, additional studies are needed to expand our understanding of melanoma therapeutic strategies and the identification of novel therapeutic agents.

Oncogenic BRAF, HH signaling and stem cell self-renewal

The self-renewal capacity and differentiation potential of CSCs contribute significantly to the heterogeneity, metastasis, recurrence, and chemo-radioresistance observed in melanoma tumors[217]. CSCs typically originate from normal stem cells through a series of gene mutations, epigenetic modifications, and dysregulated signaling pathways, in conjunction with persistent alterations in the tumor microenvironment[218]. However, the precise origin of MSCs remains undetermined[219]. MSCs, also referred to as malignant melanoma-initiating cells, are experimentally characterized by their ability to self-renew and sustain tumor growth[219]. They express specific biomarkers such as adenosine triphosphate-binding cassette sub-family B member 5, nerve growth factor receptor, CD20, CD133, and aldehyde dehydrogenase[219,220]. Like CSCs, the functions of these stem cells are regulated by several transcription factors such as homeobox protein NANOG, octamer binding transcription factor 4, sex-determining region Y-box 2 (SOX2), Kruppel-like factor 4, and MYC transcription factor[221]. While the subcellular localization of these transcription factors plays a vital role in cellular function, its precise impact on cancer progression remains poorly understood and requires further investigation. In addition to these transcription factors, HH pathway has been identified as a significant regulator of MSC maintenance[68]. Although this pathway is aberrantly activated in several cancers, the mechanisms of activation vary. In melanoma, its precise role in MSC regulation and therapeutic resistance warrants deeper exploration. Notably, transcription factors like SOX2 and NANOG, essential for maintaining stemness, are downstream targets of the HH signaling pathway[222].

The most important biomarkers of MSCs are CD20 (a cell surface marker normally associated with B cells), adenosine triphosphate-binding cassette sub-family B member 5, C-X-C chemokine receptor type 6, CD44, SOX10, CD133, CD271 and aldehyde dehydrogenase (ALDH)[223,224]. Among these MSC biomarkers, the identification and functional validation of CD271 and ALDH have been particularly pivotal in understanding tumor heterogeneity and progression[225,226]. CD271, also known as p75 neurotrophin receptor, has been recognized as a marker of MSCs[227]. Studies have demonstrated that CD271-positive melanoma cells possess self-renewal capabilities and can recapitulate the heterogeneity of the original tumor upon serial transplantation in immunedeficient mice[225]. These cells maintain their tumorigenic potential over multiple passages, suggesting a stable stem-like phenotype. In contrast, CD271-negative cells exhibit limited tumorigenicity and fail to sustain long-term tumor growth, highlighting the functional significance of CD271 expression in MSCs[225]. Similarly, high ALDH activity, particularly involving the ALDH1A1 and ALDH1A3 isozymes, has been associated with MSCs[226]. ALDH-positive melanoma cells demonstrate enhanced tumorigenic potential and resistance to chemotherapeutic agents compared to their ALDH-negative counterparts[226]. Functional assays have shown that silencing ALDH1A expression leads to decreased cell viability, induction of apoptosis, and reduced tumorigenesis in vivo[228]. Moreover, ALDH-positive cells can differentiate into ALDH-negative cells, indicating their role in maintaining tumor heterogeneity. These findings underscore the heterogeneity within melanoma tumors and the critical role of stem cell biomarkers like CD271 and ALDH in tumor initiation, maintenance, and resistance to therapy[229]. Targeting these MSCs may offer novel therapeutic strategies to overcome treatment resistance and prevent tumor relapse in melanoma patients; however, this strategy requires the availability of clinically relevant melanoma models to ensure effective translation into clinical practice. While some models, such as the Tyr-CreER; BRAFCA/+ (BRAF V600E/+); PTEN fl/fl (Tyr-CreER: BRAF: PTEN) murine system, suggests tumor formation from MSCs[230], other studies report contradictory findings, indicating the need for further validation[231].

In melanoma, intratumoral heterogeneity, progression and drug resistance result from the unique characteristics of MSCs[232]. These MSCs harbor distinct protein signatures and tumor growth-driving pathways, whose activation is driven by driver mutation-dependent signals[232]. CSCs are usually identified on their ability to generate tumorspheres in suspension cultures, to possess high invasive behavior, to give rise to the heterogeneous original tumor when inoculated in nude mice, and to express specific surface markers[233]. The maintenance of CSCs in tumors is highly dependent on HH signaling. CSCs respond to the HH ligand, secreted by adjacent stromal cells, tumour cells or CSCs themselves, to maintain their stemness through the regulation of pluripotency genes such as NANOG, SOX2 and BMI1[217]. Aberrant activation of HH signaling is believed to promote the CSC phenotype through the subverted regulation of these stemness-determining genes. NANOG, a transcription factor involved in embryonic stem cell self-renewal and somatic cell reprogramming, is a direct transcriptional target of the HH signaling pathway[234]. Thus, HH signaling sustains the stemness signature across multiple cancers by driving the expression of core stemness-related genes. Knockdown of SMO and GLI1 in SSM2c and A375 melanospheres, as well as engraftment of SSM2c cells transduced with LV-shSMO and LV-shGLI1, significantly reduces the ALDH+ cell fraction, clonogenicity, and tumor growth[235]. Notably, SMO knockdown completely abolishes SOX2 mRNA expression, indicating that SOX2 acts as a downstream mediator of HH signaling in MSCs[222]. Chromatin immunoprecipitation sequencing in M26c cells confirmed that SOX2 is a direct transcriptional target of GLI1 and GLI2[236]. Moreover, WIP1, an oncogenic phosphatase overexpressed in various cancers, is essential for HH-induced MSC growth and self-renewal[237]. HH/GLI signaling axis plays a central role in regulating MSCs and contributes to therapeutic resistance. Targeting this pathway, along with specific stem cell markers, holds promise for more effective melanoma treatment strategies.

Effect of HDACs on HH pathway in BRAF-driven melanoma

HDACs have recently received attention for their role in the drug-resistant phenotype of melanoma, thereby positioning HDAC inhibitors as promising tools for combating this therapy resistance[238,239]. Expression of the class 1 histone deacetylases HDAC8 and HDAC3 are associated with improved survival of patients with metastatic melanoma[240]. HDAC8, in particular, regulates stress response pathways in melanoma to mediate escape from BRAF inhibitor therapy. For example, introduction of HDAC8 into drug-naïve melanoma cells conveyed resistance both in vitro and in vivo[241]. HDAC8-mediated BRAF inhibitor resistance is mediated via activation of receptor tyrosine kinases and MAPK signaling. Inhibitors targeting HDAC8 specifically curb the ability of melanoma cells to adapt to various stressors, including inhibition of BRAF-MEK signaling[241]. Additionally, HDAC inhibitors have been observed to restore sensitivity to BRAF inhibitors by modifying PI3K and survival signaling pathways in a specific subgroup of melanoma cases[242]. Although HDACs function at the histone level, they also regulate nonhistone substrates, and introduction of HDAC8 decreased the acetylation of c-JUN, increasing its transcriptional activity and enriching for an AP-1 gene signature. Using patient- and in vivo-derived melanoma cell lines with acquired BRAF inhibitor resistance, it has been reported that combined treatment with the BRAF inhibitor encorafenib and HDAC inhibitor panobinostat in 2D and 3D culture systems synergistically induced caspase-dependent apoptotic cell death[242]. Similarly, HDAC inhibition overcomes acute resistance to MEK inhibition in BRAF-mutant colorectal cancer by downregulation of cellular FLICE-like inhibitory protein, long isoform. Further, combined HDAC inhibitor/MEK inhibitor treatment resulted in dramatically attenuated tumor growth in BRAF MT xenografts[243].

CONCLUSION

The failure of current therapeutic regimens to fully eradicate MSCs poses a significant challenge to the healthcare system. It is now clear that HH signaling plays a crucial role in maintaining MSCs, and inducing apoptosis in these cells by targeting key signaling pathways represents a promising approach in cancer treatment. Oncogenic BRAF has been shown to activate HH signaling in BRAF-mutant melanoma tumors, drawing considerable attention to the potential of targeting the HH signaling cascade alongside BRAF signaling as a viable option to retard tumor growth and prevent recurrence. Several agents have been developed to specifically target these pathways, and since aberrant activation of HH is associated with enhanced proliferation and cancer development in the skin and other tissues, creating potent small molecule inhibitors for this pathway may provide successful therapeutic interventions for HH pathway-dependent cancers. Developing HH pathway inhibitors for melanoma is crucial due to their potential to target CSCs, overcome therapeutic resistance, and enhance combination therapies, thereby altering the tumor microenvironment and preventing metastasis. Ultimately, these inhibitors offer new avenues for more effective and personalized therapies, improving patient outcomes. However, there is a need to develop clinically relevant models to understand the crosstalk between oncogenic BRAF and HH signaling in sustaining melanoma tumors and to uncover new therapeutic targets.