Demystifying the Druggability of the MYC Family of Oncogenes

Abstract

MYC family of oncogenes (MYC, MYCN and MYCL), encodes a basic helix-loop-helix leucine zipper (bHLHLZ) transcriptional regulator that is responsible for moving the cell through the restriction-point. Through the HLHZIP domain, MYC heterodimerizes with the bHLHLZ protein MAX, which enables this MYC-MAX complex to bind to E-box regulatory DNA elements thereby controlling transcription of a large group of genes and their proteins. Translationally, MYC is one of the foremost oncogenic targets and deregulation of expression of MYC family gene/proteins occurs in over half of all human tumors and is recognized as a hallmark of cancer initiation and maintenance. Additionally, unexpected roles for this oncoprotein have been found in cancers that nominally have a non-MYC etiology. Although, MYC is rarely mutated its gain of function in cancer results from overexpression or from amplification. Moreover, MYC is a pleiotropic transcription factor possessing broad pathogenic prominence making it a coveted cancer target. A widely held notion within the biomedical research community is that the reliable modulation of MYC represents a tremendous therapeutic opportunity given its role in directly potentiating oncogenesis. However, the MYC-MAX heterodimer interaction contains a large surface area with a lack of well-defined binding sites creating the perception that targeting of MYC-MAX to be forbidding. Here we discuss the biochemistry behind MYC and MYC-MAX as it relates to cancer progression associated with these transcription factors. We also discuss the notion that MYC should no longer be regarded as undruggable providing examples that a therapeutic window is achievable despite global MYC inhibition.

Graphical Abstract:

INTRODUCTION

The MYC oncogene family encodes three proteins: c-MYC, N-MYC and L-MYC in mammalian cells that are differentially expressed in terms of tissue type and developmental stage.^{1, 2} While c-MYC is universally overexpressed in rapid proliferation cells, deregulation of N-MYC and L-MYC is restricted to a limited number of carcinomas.^{3, 4} As one of the most studied multifunctional transcription factors, it is implicated in a variety of physiological and pathological processes including: (1) cell proliferation, (2) differentiation, (3) apoptosis, (4) ribosome biogenesis, (5) protein translation, (6) orchestrating biological events like angiogenesis, (7) metabolism, (8) DNA repair, (9) immune surveillance and (10) stem cell formation.^{5, 6} Beyond selectively targeting greater than 15% of the human genome, MYC indirectly acts as a global transcription amplifier likely as a combined effect of the non-selective invasion into active promoters and the feedback loop following MYC induced physiological change.^7–11 Hence, its deregulation induced by gene amplification, chromosomal translocation, as well as mutation in MYC gene or upstream regulators, promotes malignant transformation resulting in hallmark features of over 50% human cancers.^12–15 Studies in transgenic mouse models demonstrated the role of MYC in tumor maintenance, which provides proof of principle of inactivating the oncoprotein as a treatment in reversing tumorigenesis.^{10, 16, 17}

MYC has typically been touted as “undruggable” based upon its nuclear localization, intrinsically disordered structure, protein–protein interaction (PPI) interface and a lack of any cleft like protein pockets.¹⁸ Another general concern in the field has been that MYC inhibition would cause deleterious side effects in normal tissues as MYC responds to virtually every signaling pathway in every cell type. Yet, many of these stereotypes were overturned after the success of applying Omomyc, a MYC dominant-negative peptide in murine cancer models, suggesting a transient inhibition of MYC is highly tolerable.¹⁹ Moreover, other strategies have also appeared in an attempt to “drug MYC”, among which most promise has been seen in with molecules interfering with MYC-MAX protein/protein interactions as well as deterring MYC-MAX DNA association. Particularly promising has been direct targeting of MYC thereby reducing off-target effects, which in turn can effectively impede MYC function. This perspective aims to summarize the biochemistry behind the MYC/MAX network, MYC small molecule targeting strategies, in vivo evaluation of MYC inhibitors and future prospects for clinical trials.

BIOCHEMISTRY OF MYC

As stated MYC proteins belong to the superfamily of basic helix-loop-helix leucine zipper (bHLHLZ) DNA binding proteins. Within this complex is contained three-domains, including an N-terminal transactivation domain (TAD), a central domain, and a C-terminal basic-helix-loop-helix-leucine zipper (bHLHLZ) domain, this interacting with its essential partner, MYC-associated protein X (MAX) (Figure 1A). The central domain consists of a proline rich PEST segment, a calpain cleavage site (CAPN), which results in cytosolic MYC partial cleavage of the C-terminus, producing “MYC-nick”,²⁰ and a nuclear localization sequence (NLS). Six highly conserved regions dubbed MYC homology boxes (MBs 0–IV) are sequentially distributed across TAD and the central domain, which bind to different MYC interactors and thus enable MYC to regulate distinct enzymatic reactions throughout the transcriptional process (Figure 1B).²¹

Figure 1.

MYC and its biochemical impact on the cell. (A) Crystal structure of the c-MYC/MAX dimer bound to E-box DNA (PDB ID 1NKP). (B) Structure components of MYC protein and its interaction partners. (C) MYC stability controlled by the ubiquitination-proteasome system. (D) Scheme showing MYC/MAX/MXD network and their roles in counteracting MYC function. Figure created with BioRender.

Proteomic profiling of the MB interactomes have revealed that half of the MYC interactors require one or more MBs for binding wherein two MBs, MB0 and MBII, are universally required for transformation.²² It should be understood that the interactome MB0 accelerates the transcription process by binding to the general transcription factor IIF (TFIIF); it also recruits PP1 phosphatase via cofactors to regulate MYC activity and stability.²³ Moreover, MBII participates in chromatin remodeling and its modification brings together acetyltransferase-containing complexes through direct interaction with Transactivation/Transformation-Domain Associated Protein (TRRAP) enabling histone acetylation.^{24, 25} In addition, particularly prominent interactions occur between acetyltransferases p300, CBP and the WDR5 protein containing histone lysine methyltransferase complex, recruiting at MBIIIb. In sum these sequences of events all play important roles in directing MYC to chromatin modification.^{26, 27}

From a biochemical perspective MBI is noteworthy for stepwise phosphorylation and dephosphorylation at serine 62 (S62) and threonine 58 (T58) residues and is responsible for ubiquitin-proteasome system (UPS) mediated degradation of the MYC protein (Figure 1C).²⁸ Accordingly, MYC protein is expressed at a low level in normal cells with a fast turnover rate through the ubiquitin proteasome pathway involving interaction between Fbxw7 and its phosphorylation sites before polyubiquitination takes place at lysine 48.^{29, 30} By contrast, MYC binding to Aurora-A, a serine/threonine kinase, antagonizes Fbxw7 mediated degradation.³¹ Moreover, MBI has direct contact with positive transcription elongation factor b (P-TEFb). P-TEFb then phosphorylates the C-terminal domain (CTD) of RNAPII, and stimulates transcription pause release.^{32, 33} In addition to stability control, phosphorylation at S62 by mitogen-activated protein kinases (MAPK), which acts as a downstream signal of oncogenic RAS, or cyclin-dependent kinases (CDK) also plays a critical role influencing activity, target gene selection and intracellular localization of MYC.³⁴ Besides aforementioned phosphorylation within MBI, the whole protein is also subjected to extensive post-translational modification spanning not only phosphorylation but acetylation, ubiquitination, sumoylation, and glycosylation.³⁵ We highlight the modifications inside the bHLHLZ region. Here, MYC is found to be a substrate of Pak2, a serine/threonine kinase that participates in the cellular response to stress. In short Pak2 phosphorylates MYC at three sites, (S373 and T400) and thereby interferes with MYC/MAX complex formation, while (T358) restricts DNA recogniiton.³⁶ By contrast acetylation at lysine has no impact on the assembly but potentially competes with other lysine dependent modifications like sumoylation or ubiquitination.³⁷ Moreover, lysine residues (K389, K392, K398 and K430) within the bHLHLZ are found to be promiscuous SUMO acceptors. Toward this end sumoylation initiates concomitant ubiquitination for MYC degradation while also switching the transcriptional subprogram of MYC from activated to repressed or vice versa.^38–40

Complementing activation, a number of genes have been identified that are transcriptionally silenced by MYC.⁴¹ On the whole, a large fraction is associated with the MYC-interacting zinc finger protein 1 (Miz-1). Here, MYC binding abrogates its individual activation effect on tumor suppressor genes like p15^INK4b, a factor for cell cycle arrest.⁴² Myc also interacts with Sp1/Sp3 to repress the p21 transcription thus affecting the p21-mediated cell cycle checkpoint.⁴³ This dual function is best exemplified in the extensive reprogramming of the miRNA transcriptome concomitantly activating the miR-17–92 cluster while predominantly downregulating other miRs.⁴⁴ Adding another layer of complexity, proliferation is not the only output of MYC overexpression. MYC also sensitizes cells to apoptosis depending on tissue type and its environment, therefore mutations in these cells that elevate pro-survival signals like BCL2 synergize with MYC in tumorigenesis.^45–47 This is compatible with the fact that additional mutagenic events are sometimes necessary to release its tumorigenic potential.⁴⁸

Despite these biochemical complexities in MYC induced tumor occurrence, all pathways converged to the same point where MYC must dimerize with MAX via the bHLHLZ domain. Formation of this master transcriptional regulator is followed by DNA recognition on a specific sequence termed enhancer box (E-box) with high affinity against a canonical sequence of CACGTG.⁴⁹ However, for the moment we need only point out that DNA affinity is not sufficient to account for the occupancies of core promoters by the MYC/MAX complexe.⁵⁰ Moreover, multiple additional transcription factors compete with the same consensus sequence as MYC/MAX, while an active chromatin context is required for MYC/MAX access.^{51, 52} Accordingly, a model has been proposed wherein protein-protein interactions with resident chromatin proteins, such as WDR5, strongly enhance the promoter affinity and the subsequent sequence recognition leading to stabilization in these binding sites.¹¹ The overall occupancy is determined by the kinetics between stratified apparent promoter affinities and the cellular MYC levels to shape the specific gene expression programs.⁵³ Endogenous MYC levels only allows binding at high affinity promoters representing the fundamental role of MYC in cell growth (ribosome biogenesis and protein synthesis), cell cycle control, energy production, anabolic metabolism and DNA replication.^{54, 55} It should be further understood that as MYC levels escalate during transformation, more and more distal sites become occupied, a panoply genes become activated or repressed, eventually breaking the homeostasis and completely rewriting the cell’s behavior.

Within the MYC/MAX/MXD network, MAX functions as an essential dimerization partner. Moreover, MAX is constitutively expressed regardless of the cells stage whereas its interacting partners normally have a very short half-life.⁵⁶ Heterodimers of MAX with MXD family members (MXD 1–4, formally called Mad proteins), Mnt, and Mga, antagonize MYC-dependent cell transformation by competition and transcriptional repression of the same E-box sequences (Figure 1D).⁵⁷ The transcriptional repression is accomplished by association with Sin3, a corepressor complex that recruits HDAC for histone deacetylation.^{58, 59} In addition, loss of Mnt function releases the apoptotic activity of MYC mimicking the outcome of MYC aberrantly expressed at high concentrations, which indicates Mnt’s separate pro-survival capability.⁶⁰ A parallel system centered on another MAX like protein, Mlx, forms a heterodimer with MXD family serving as a potential channel to divert these proteins away from direct participation in the MAX-interacting network.⁶¹ Mlx, when bound to nutrient sensing partners like MondoA and CHREBP, regulates glucose and lipid metabolism depending on the availability of multiple metabolites such as glucose and glucose-6-phosphate, and likely forms competition on similar Myc binding sites.^62,63 Furthermore, MAX can form transcription inert homodimers acting as a tumor suppressor whereas its alternative spliced isoform delta MAX truncated at the C-terminal loses tumor suppressor activity shifting the equilibrium towards tumorigenesis.^{64, 65}

STRUCTURAL STUDIES

Available crystallographic data of the heterodimer bHLHLZ domain to the CACGTG E-box provides a structural basis for DNA recognition and MYC-MAX interactions.⁶⁶ Here, basic regions from the N-terminal insert into the E-box aligned DNA major groove making the major contact with DNA sequence, while the two helical segments (H1 and H2/Leucine zipper) allow dimerization with their counterparts. It has been demonstrated that all of the conserved hydrophobic amino acids within H1 and H2 contribute to the dimerization.⁶⁷ Moreover, it has been shown that extensive hydrophobic and polar interactions between the HLH and LZ regions stabilize both MAX-MAX homodimers and MYC-MAX heterodimers with key residues providing H-bonding between MAX Gln91-Asn92 (monomer numbering), positively charged interactions between Arg-Arg on MYC and the negatively charged interactions found within Glu-Gln on MXD. Critical here is that these charged pairing disfavor MYC/MYC and MXD/MXD homodimer formation.⁶⁸ In addition particularly important are three conserved residues within the basic region making base-specific contacts with the E-box, the His, Glu, and Arg residues at MAX positions 28, 32, and 35, which have been validated through mutagenesis experiments.^{69, 70} Strikingly, a recent X-ray structure and nuclear magnetic resonance (NMR) spectroscopy study of the apo form of Omomyc and MYC/MAX provide critical insights into the binding pattern in the absence of DNA. Here, an extensive network of helical structural interactions was determined in the basic region suggesting that the molecular recognition between MYC/MAX and DNA occurs via a conformational selection rather than an induced fit model.^{71, 72}

MYC TARGETING STRATEGIES

Overall, a tremendous volume of medicinal research has been conducted in terms of targeting MYC; with an array of strategies utilized to target this notorious protein. These tactics are graphically described in Figure 2. Noteworthy efforts include the stabilization of the G-quadruplex,⁷³ which consist of guanine rich segments of DNA, typically present in promoter region of c-MYC gene.⁷³ In short, existence of G-quadruplex in promoter region acts as a transcriptional repressor of c-MYC expression. Hence, stabilization of existing G-quadruplex leads to downregulation of the transcription of c-MYC gene and eventually translation of c-MYC protein.⁷³ Importantly clinical trials have been initiated with this approach.^{74, 75} Another tactic includes of antisense oligonucleotides (ASOs), wherein these synthetic, single stranded nucleotides are used to target RNA. Mechanistically this approach takes advantage of the inhibition of mRNA translation resulting in tumor growth reduction. Efforts here have resulted in ASO’s including INX-3280,^76,77 INX-6295,⁷⁸ and AVI-4126,⁷⁹ which entered in clinical trials; however, they have recently been abandoned.^80,81,82 An additional oligonucleotide agenda involves siRNA, which are well known to regulate gene expression through RNA interference (RNAi).⁸³ These small interfering RNAs are processed by RNA-induced silencing complex and the enzyme named dicer to target RNA.⁸³ In addition an siRNA, named DCR-MYC, reached clinical trials that were later halted due to lack of efficacy.⁸⁴

Figure 2.

Graphical representation of the strategies to target MYC in a cancer cell.

A more chemistry concentric approach involves targeting/inhibition of the bromodomain (BRD) ultimately leading to downregulation of MYC protein. Although, function of bromodomain is not restricted to MYC gene regulation, many bromodomain extra-terminal domain inhibitors (BETis), are currently in clinical trials for MYC inhibition therapy.⁸⁴ An additional indirect approach to regulate MYC that is noteworthy involves the inhibition of the holoenzyme protein complex named PP1/PNUTS, which leads to hyperphosphorylation of MYC. Ultimately this phosphorylation event leads to reduced chromatin binding of MYC as well as elevated proteasomal degradation.⁸⁵

There are other miscellaneous strategies that target MYC through inhibition of its transcription, translation, regulators of MYC stability and its transcriptional cofactors. However, we will focus on direct inhibitors of the MYC-MAX heterodimer or MYC-MAX-DNA complex, encompassing small molecule inhibitors, peptides and miniproteins that have been engaged to inhibit these associations. More specifically, we will feature molecules with known in vitro success that have shown promising in vivo results. Lastly, we will highlight Omomyc, the only MYC-MAX inhibitor that has reached clinical trials.

IN VIVO EVALUATION OF MYC-MAX-DNA INHIBITORS

Cancer animal models have considerably advanced since the introduction of the first xenograft mouse model in the 1950’s. Through decades of cancer research, a better understanding of the advantages and limitations of commonly used animal models have enabled the exploitation of these models to gain insight into the mechanism, tolerability, and efficacy of anticancer immunotherapeutic and pharmacologics.⁸⁶ Xenograft mouse models, including cell line-derived (CDX) and patient-derived (PDX), provide important tools for preclinical evaluation of anticancer therapeutics. The more simplistic CDX approach relies on the implantation of established human cancer cell lines, often commercially available, into immunodeficient animals providing predictive values. Although inexpensive and suitable for numerous types of studies, CDX models lack tumor heterogeneity and tumor micro-environment often diminishing translational value. Alternatively, the multifaceted and more informative PDX approach involves the engraftment of human derived cells or tumor tissue from a patient’s tumor into immunodeficient or humanized mice. Generally, PDX models can accurately depict the biology, heterogeneity, complexity, and molecular diversity of different cancers and recapitulate the tumor microenvironment providing an invaluable platform for translational cancer research. A complimentary approach to xenograft models is allografts that maintain host species, i.e. mouse cancer cell lines or mouse tumor tissue are implanted into mice. Using a combination of both xenograft and allograft mouse models, the therapeutic potential of numerous small molecule and peptide-based inhibitors targeting the MYC-MAX E-box complex have been evaluated in vivo

Myc-Max Dimerization Disruption.

Many small molecule inhibitors that disrupt MYC-MAX dimerization in vitro have been described in the literature, however only a small number of these compounds have shown in vivo potency against MYC-driven or MYC involved cancers, Figure 3. Some of the earliest MYC inhibitor studies identified two unrelated small molecules, 10058-F4 and 10074-G5, among others that bind distinct regions within the c-MYC bHLHLZ domain and stabilize the monomer, thus interfering with MYC-MAX heterodimerization and its association with DNA targets.^87–89 Lack of efficacy in established CDX models, mice bearing human androgen-independent prostate cancer cells or human Burkitt’s lymphoma cells, was attributed to unfavorable pharmacological profiles including rapid metabolism and low tumor penetration.^{90, 91} Although, some success was later reported using 10058-F4 in MYCN transgenic mice.⁹² An improved second generation analog of 10074-G5, termed 3jc48-3, displaying greater potency in vitro and increased half-life underwent preliminary testing in a high grade prostate cancer PDX model.^{93, 94} Albeit a small sample size, 3jc48-3 was well tolerated with evidence of slowing tumor growth reflecting the drugs enhanced properties.

Figure 3.

Structures of non-protein inhibitors of MYC and MYC-MAX as discussed within the text.

Perhaps the first small molecule to display in vivo effectiveness against a MYC-driven human cancer was the novel inhibitor KJ-Pyr-9, discovered through screening efforts of the Janda and Vogt laboratories.⁹⁵ KJ-Pyr-9 was shown to possess a nanomolar dissociation constant for MYC as well as favorable PK properties in vivo with no signs of acute toxicity. Notably, KJ-Pyr-9 was found to cross the blood-brain barrier (BBB) with higher drug concentrations detected in the brain compared to the blood. This latter observation suggests that MYC-targeted pharmacological interventions may have the potential to treat cancers, such as brain cancers, which are not only shielded by the blood-tumor barrier but also the BBB. KJ-Pyr-9 was put to the test against a highly aggressive, invasive triple negative breast cancer (TNBC) human cell line, MDA-MB-231, commonly used to model late-stage breast cancer. Tumor growth was halted in xenograft mice by penetrating tumor tissue and suppressing MYC’s transcriptional activity demonstrating its antitumor activity.

Another compound identified using MYC-targeted high through put (HTS) screening strategies, dubbed Mycro3, precludes MYC-MAX dimerization and binding to DNA, and is highly selective with great specificity.^{96, 97} To evaluate the in vivo efficacy of Mycro3, a K-RAS-induced pancreatic ductal adenocarcinoma (PDA) mouse model with MYC-dependence was developed. The relationship between these two “undruggable” oncoproteins (RAS and MYC) in tumorigenesis is very complex, however a cooperative mechanism between them has been proposed to drive cancer development and progression in several cancer types.⁹⁸ Therefore, the prospect of modulating K-RAS through MYC targeting highlights the tremendous widespread potential of MYC inhibitors. In fact, daily oral administration of Mycro3 to moribund mice resulted in protracted survival and reduced tumor size.⁹⁹ Although tumors did not completely disappear in this study, lack of tumor ablation was not necessarily unexpected and has been noted in other studies with anti-MYC inhibitors. Mycro3 treatment also significantly attenuated tumor growth in xenograft models of lung and breast cancer, illustrating the broad value of such a therapeutic strategy for the treatment of human cancers, including K-RAS-driven PDA.

The small molecule MYCMI-6 was identified through a cellular HTS and shown to disrupt MYC-MAX dimerization by binding of the MYC/MYCN bHLHLZ domain with no observable interaction with MAX.¹⁰⁰ In vivo confirmation was demonstrated wherein MYC-MAX association was significantly reduced in tumor tissue in a MYCN-amplified neuroblastoma xenograft tumor model, correlating with both apoptosis induction and cell proliferation curtailment in tumor areas. MYCMI-7 also identified from this same screen was shown to directly bind MYC/MYCN, induce apoptosis in tumor cells while reducing tumor burden and increasing survival in MYC-driven acute myeloid leukemia, breast cancer and MYCN-amplified neuroblastoma cancer in vivo models.¹⁰¹ An important feature highlighted in these set of studies is the pan-MYC inhibitory activity of MYCMI-6 and −7 involving both MYC and NMYC, findings that are in line with reports involving 10058-F4 vide supra. While MYC has been the main therapeutic target among the MYC family of oncoproteins, NMYC and LMYC have been associated with neuroblastoma, NMYC-driven brain cancer and lung cancer, and are thought to play a role in tumor progression and maintenance. Based on sequence and structural homology between MYC proteins, the development of clinically viable pan-MYC inhibitors appears feasible.

Recently, a MyC-CaP allograft/xenograft mouse prostate model was engaged in a rapid in vivo screening format identifying a set of anti-MYC compounds, including MYCi361 and MYCi975.¹⁰² Strategically, this CDX approach took advantage of an engineered MyC-CaP E-box-Luc cell line that afforded live imaging of bioluminescence in tumors, monitoring MYC-transcriptional activity. Following this initial in vivo screen, a series of studies were conducted shedding light on MYCi361’s mechanism of action and revealing its ability to bind MYC but not MAX, impair MYC-MAX dimer binding to DNA, and promote MYC destabilization through modulating phosphorylation at threonine 58 of the MYC protein (p58). MYCi361 treatment of Myc-CaP xenograft mice revealed an increase in Myc p58 levels and a decrease in the Ki67 proliferation marker in tumor tissue along with remodeling of the tumor microenvironment. The latter observation led to combination studies with MYCi361 and an immune checkpoint inhibitor (anti-PDL) demonstrating the potential synergistic effect of combination therapy involving an anti-MYC inhibitor and immune checkpoint blockade. However, MYCi361 toxicity concerns prompted the development of MYCi975, an analog with an enhanced therapeutic index. MYCi975 was well-tolerated, displayed an exceptional PK profile, and showed in vivo anti-tumor activity including synergistic tumor suppression when administered in combination with a PD-L1 antibody, analogous to MYCi361.

A somewhat different MYC targeting approach sought to covalently modify the MYC protein itself by covalent linkage of a small molecule through reactive cysteine residues. EN4 was uncovered by a cysteine-reactive covalent ligand screen and shown to label a single site, cysteine 171, within an intrinsically disordered region of MYC.¹⁰³ EN4 inhibits MYC-MAX DNA binding, impairs cell proliferation, and diminishes tumor growth in vivo in a 231 MFB breast cancer CDX model. Despite prospective off-targets liabilities and selectivity concerns, EN4 may provide a new avenue towards the treatment of MYC-driven cancers due to the covalent nature of MYC attachment.

Attenuation Of Myc-Max-E-Box DNA Association Through Alternative Mechanisms

Alternative to directly targeting the MYC protein, compounds and mini-proteins that disrupt the MYC-MAX-DNA complex through other mechanisms have been reported to show in vivo efficacy. For example, ME47, an engineered homodimer hybrid consisting of a region of MAX and the HLH domain of transcription factor E47, competes with MYC-MAX for binding to E-box DNA response elements of MYC target genes.¹⁰⁴ Inducible expression of this minimalist protein in a breast cancer CDX model delayed tumor growth, reduced cellular proliferation, and extended survival.

KSI-3716, a small molecule inhibitor discovered by HTS efforts, likewise blocks MYC-MAX binding to target gene promoters.¹⁰⁵ Delivery of KSI-3716 as an intravesical medication curbed tumor growth with no significant systemic toxicity in orthotopic bladder xenografts.¹⁰⁶ KSI-3716’s potential utility in gemcitabine-resistant bladder cancer was also suggested based on results from in vitro resistance studies, although in vivo efficacy has yet to be reported.¹⁰⁷

Lastly, an alternative approach explored perturbation of MYC’s transcriptional activity through targeting its binding partner MAX. KI_MS2-008 was shown to stabilize transcriptionally inert MAX-MAX homodimers, resulting in a decrease in both MYC binding at promoters and MYC protein levels.¹⁰⁸ The in vivo potency of KI_MS2-008 was assessed in allograft mouse models of MYC-induced T cell acute lymphoblastic leukemia and hepatocellular carcinoma mouse models showing a marked reduction in tumor burden.

REACHING CLINICAL TRIALS AFTER DECADES OF MYC RESEARCH

For decades speculation, often skepticism, has centered around whether MYC targeted therapy would become a viable treatment option for MYC-driven and -associated cancers. Given MYC’s central role in a wide array of fundamental cellular processes, the potential for severe consequences caused by hampering MYC’s function were considered inevitable. The multitude of small molecule and peptide-based MYC inhibitor studies have exposed numerous hurdles that still need to be overcome such as lack of specificity, adverse off-targets effects, low potency and tumor penetration, and subpar PK profiles. However great strides have been made as evidenced by the positive in vivo results presented. Importantly, the notion of combination therapies or drug cocktails including MYC targeting drugs as a key component has gained traction. In fact, preclinical studies have revealed drug synergism between an anti-MYC compound and immune checkpoint inhibitor,¹⁰² as well as the mini-protein Omomyc and paclitaxel, vide infra.

Perhaps the best support of a MYC targeting drug becoming a clinically viable cancer treatment option has come from the recent entry of Omomyc (OMO-103) into clinical trials. Omomyc, a first in class MYC inhibitor initially disclosed in 1998, is a dominant negative mutant of the MYC bHLHLZ domain consisting of four mutations that prevent MYC-MAX dimerization.^{109, 110} Through formation of heterodimers with either MYC or MAX or formation of homodimers, Omomyc prevents MYC-MAX dimerization, inhibits E-box DNA binding, and promotes MYC degradation.¹⁰⁹ In early preclinical studies, transgenically-expressed Omomyc showed in vivo efficacy in cancer mouse models without significant side effects, providing proof of concept.¹¹⁰ Possibly one of the most significant findings came later when recombinant Omomyc was shown to have cell penetrating activity.¹¹¹ Intranasal administration of recombinant Omomyc halted tumor progression in a lung adenocarcinoma model, penetrating both cellular and nuclear membranes.¹¹¹ Additionally, intravenously dosed Omomyc was shown to be better than paclitaxel at blunting tumor growth in a lung adenocarcinoma xenograft model and combination therapy with both Omomyc and paclitaxel proved to be far superior to each monotherapy alone, resulting in near complete eradication of tumor growth and prolonging survival without observable side effects.¹¹¹ Overall, these later findings helped solidify Omomyc’s potential as a clinically viable anti-MYC therapeutic catapulting Omomyc into Phase I/II human clinical trials where promising Phase I preliminary results have been reported.

OUTLOOK

The MYC proto-oncogene family has been at the forefront of cancer biology for greater than 30 years, yet it has constantly presented the research community with undeniable biological as well as chemical challenges. For example, MYC is one of the most frequently overexpressed genes in human cancer.¹¹² A major difficulty with targeting MYC driven cancer is that it is also the gatekeeping regulator of cell division.¹¹³ As such a multitude of transcriptional programs are under MYC control with distinct effects on the cellular phenotype. Success in this arena has also had to contend with the disordered nature of Myc, which by the physical standards of the time, vide supra, made it a quite complex and challenging drug target to model.¹¹⁴ Further confidence in the modulation of MYC also waned because the MYC-MAX heterodimer has an estimated protein-protein interaction (PPI) surface area of over 3000 angstroms¹¹⁵ and as well documented PPIs lack well-defined pockets or grooves that could be used for high energy binding interactions with a small ligand.^{68, 116}

At the clinical level therapeutic efforts against MYC and MYC-MAX have also stalled due to off-target effects on other oncoproteins.¹¹⁷ While more efforts on oncogene selectivity is warranted new MYC drug candidates will also need to contend with fast metabolism, nonselective distribution after systematic dosing, and tumor penetrability of these small molecule and peptide inhibitors, all of which will undoubtably hold back their clinical feasibility.^{90, 91} Accordingly, we also envision that it is time to place more weight on the integration of well-designed delivery systems in MYC drug discovery efforts. Studies have shown that cell penetrating peptides and nuclear targeting sequences fused to the peptide inhibitors led to unprecedented enhancement of tumor growth inhibition both in vitro and in vivo.^{114, 118–120} Indeed, a lipase-labile 10058-F4 prodrug carried by integrin-targeted nanoparticles showed reduced tumor burden in integrin overexpressed multiple myeloma animal model against free drugs.¹²¹ It would be interesting to see additional efforts grounded upon nanoparticle platforms making their way to help boost Myc inhibitors in the future.

Apart from all these hurdles, extensive MYC research has generated a profound biomedical arsenal though robust high throughput screening interventions, meticulous computer aided drug design campaigns, screenings of natural products and fastidiously engineered peptides and miniproteins. These efforts have led to the genesis of Omomyc, which has cleared phase I clinical trials with low toxicity and excellent tolerability in patients.¹²² Finally, while not fully insulated from scattered skeptics and still considerable pessimism, we believe it can now be fully acknowledged that after three decades of chemistry centric imagination, diligence and perseverance that the most deregulated oncogene in human cancer can now be delisted from the “undruggable” list of proteins.