Decoding neddylation in malignancies: molecular mechanisms, biological functions, therapeutic resistance, and clinical potential.

Facts

Neddylation is a dynamic and bidirectional post-translational modification.

Neddylation orchestrates protein stability, protein subcellular localization, protein-protein interaction, transcriptional regulation, and protein conformation.

Neddylation exerts a dual effect on sustaining cancer hallmarks.

Neddylation is vital for maintaining cellular homeostasis and is triggered by various forms of cellular stress.

Neddylation is implicated in tumor therapeutic resistance.

Targeting neddylation offers novel avenues for innovative cancer treatments.

Open Questions

What is the role of neddylation in tumor initiation and progression?

Which cellular stress can trigger neddylation?

What is the relationship between neddylation and anti-tumor therapeutic resistance?

Can targeting neddylation effectively inhibit tumors?

Introduction
Neddylation, a dynamic and bidirectional post-translational modification (PTM), involves the covalent attachment of the ubiquitin-like polypeptide neural precursor cell expressed developmentally down-regulated protein 8 (NEDD8) to a specific lysine (K) residue of substrates through a three-step enzymatic cascade, modulating protein stability, subcellular localization, conformation, protein-protein interaction (PPI), and enzymatic activity [1–4]. Similar to ubiquitination, neddylation is initiated by the sequential action of NEDD8-activating enzyme E1 (NAE), NEDD8-conjugating enzyme E2s, NEDD8 E3 ligases, and terminated by deneddylases [5, 6]. Neddylation aberrations play a dual role in tumorigenic properties and cancer hallmark maintenance, such as epigenetic reprogramming, metabolic rewiring, and immune evasion, complicating its influence on tumorigenesis. Multiple cellular stress modalities, including hypoxia, high glucose conditions, and succinate accumulation, trigger neddylation dysregulation in tumors. Additionally, Neddylation drives tumor therapeutic resistance. Consequently, pharmaceutical targeting neddylation represents a promising approach in anti-tumor strategies. This review provides a comprehensive examination of neddylation in tumors, detailing its biological significance, outlining neddylation-mediated cancer hallmarks, investigating the cellular stress that activates neddylation, exploring neddylation-induced therapeutic resistance, and evaluating neddylation inhibition for cancer treatment. Recent advances in elucidating the role of neddylation in oncogenesis may provide a strong rationale for cancer treatment.

Neddylation principles and biological features in tumors

Overview of the Neddylation cascade
Neddylation employs a multi-enzyme cascade, involving NEDD8, NAE, NEDD8-conjugating enzyme E2s, NEDD8 E3 ligases, and deneddylases (Table 1). NEDD8, a conserved protein predominantly localized in the nucleus and widely expressed across eukaryotes [1, 7], was first identified in 1992 as an overexpressed gene in mouse neural precursor cells [8]. In 1993, it was characterized as an 81-amino-acid, 9 kDa ubiquitin-like polypeptide, exhibiting 57.8% sequence identity and 78.9% similarity to ubiquitin [9]. The conserved carboxyl-terminal Gly76 residue in the C-terminal tail of NEDD8 is a requisite for covalent attachment to substrates [1]. NAE, a significant enzyme for NEDD8 activation, is a heterodimer consisting of the regulatory subunit amyloid-beta precursor protein-binding protein 1 (APPBP1/NAE1) and the catalytic subunit ubiquitin-like modifier activating enzyme 3 (UBA3/NAEβ) [10]. E2-conjugating enzymes, including ubiquitin conjugating enzyme E2M (UBE2M/UBC12) and ubiquitin conjugating enzyme E2F (UBE2F), catalyze the transthiolation reaction [11]. Numerous NEDD8 E3 ligases, responsible for NEDD8 positioning and substrate recognition, have been ascertained, such as ring-box 1 (RBX1), RBX2, murine double minute 2 (MDM2), F-box protein 11 (FBXO11), tripartite motif 40 (TRIM40), and others [3, 7]. Deneddylases, principally including constitutive photomorphogenesis 9 (COP9) signalosome, NEDD8 protease 1 (NEDP1/DEN1/SENP8), USP21, and UCH-L3, orchestrate deneddylation by removing NEDD8 from substrates [12–15]. Notably, UCH-L3 and NEDP1 also facilitate the maturation of the NEDD8 precursor.
Neddylation comprises a succession of enzymatic reactions, involving maturation, activation, conjugation, ligation, and deneddylation [16] (Fig. 1). Mechanistically, hydrolases, involving UCH-L3 and NEDP1, proteolytically cleave the C-terminal amino acids of NEDD8 precursor to expose the Gly-Gly motif, resulting in the formation of a 76-amino-acid mature NEDD8 [15, 17]. The mature NEDD8 interacts with the active site cysteine of NAE via its Ala72, shaping a high-energy thioester bond. The activated NEDD8 is subsequently transferred to a cysteine residue on NEDD8-specific E2s through the transthiolation reaction. Under the catalysis of NEDD8 E3 ligase, NEDD8 is transferred from the E2s to the substrates, covalently bonding to the K residue [7, 18, 19]. Deneddylases remove the NEDD8 molecule from substrates, completing the dynamic and reversible neddylation-deneddylation process. The substrates of neddylation are categorized into cullins and non-cullin proteins. The cullin (CUL) family consists of CUL1, CUL2, CUL3, CUL4A, CUL4B, CUL5, CUL7, and CUL9 [20]. Non-cullin neddylation substrates mainly include signaling molecules such as TP53, TP73, E2F1, epidermal growth factor receptor (EGFR), transforming growth factor-β type II receptor (TGFβRII), NF-κB essential modulator (NEMO), and Von Hippel-Lindau (VHL) [3].

Biological features of neddylation in tumors
Neddylation manipulates critical molecular processes in tumors, including gene expression, signaling transduction, and RNA metabolism, by orchestrating protein stability, subcellular localization, and transcriptional regulation (Fig. 1, Table 2). Strikingly, Neddylation exhibits a dual regulatory function in maintaining protein homeostasis within tumors. Typically, neddylation influences protein degradation by improving the activities of Cullin-RING E3 ligases (CRLs). Neddylation also modulates protein function through altering the enzyme activities, binding properties, and protein conformation. Representatively, neddylation of the cullin family enhances the activities of CRLs, which account for approximately 20% of proteasome-mediated protein degradation within the ubiquitin-proteasome system (UPS) [21]. Notably, neddylation governs protein binding properties and conformation in tumors, highlighting its functional diversity. The precise regulation of protein subcellular localization is fundamental for controlling cellular functions, metabolism, and homeostasis. Neddylation modulates tumor initiation and progression through regulating protein nuclear and cytoplasmic localization. Moreover, transcription regulation participates in tumor progression, influencing tumor cell propagation, apoptosis, and metabolism. Neddylation serves as a critical regulatory component in transcriptional regulation, interacting with transcription factors (TFs), nuclear receptors, and multiple signaling pathways. Additionally, Neddylation enables tumor cells to manipulate the DNA damage response (DDR), allowing them to sustain propagation, growth, and survival under internal and external stress. Overall, the neddylation cascade plays versatile roles in tumor cells, administering significant molecular processes that contribute to tumor malignant phenotypes. Therefore, understanding the role of neddylation may help identify effective tumor biomarkers and enable personalized treatment approaches.

Neddylation-maintained cancer hallmarks
Neddylation exerts an important effect on sustaining tumor malignant phenotypes (Fig. 2, Table 3). However, it presents tumor-suppressive effects in certain contexts, depending on the cancer type and the tumor microenvironment (TME) (Fig. 3). Specifically, UBE2F, in conjunction with E3 ligase SAG, neddylates RHEB at K169, boosting its lysosome localization and GTP-binding affinity, thereby exacerbating HCC tumorigenesis [22]. Whereas, UBE2M-mediated neddylation of LSD1 at K63 enhances its ubiquitination and degradation, reducing the stemness and chemoresistance in GC [23]. These observations underscore the intricate role of neddylation in tumorigenicity, with accumulating evidence robustly establishing its role in tumor development. Herein, we investigate the underlying mechanisms by which neddylation sustains cancer hallmarks, enhancing the understanding of cancer biology and providing novel avenues for therapeutic intervention.

Tumor invasion and metastasis
The invasion and spread of tumor cells influence the treatment strategies and patient prognosis. Remarkably, neddylation participates in tumor invasion and metastasis via regulating tumor angiogenesis and epithelial-to-mesenchymal transition (EMT). Targeting neddylation hinders the metastatic process, involving intravascular survival, extravasation, and formation of metastatic clusters [24]. Tumor angiogenesis not only supplies nutrients to tumor cells but also facilitates their dissemination through the bloodstream. Notably, neddylation manipulates tumor angiogenesis. Mdm2-mediated neddylation of VHL tumor suppressor protein at K159 disrupts VHL-p53 complex formation, fostering tumor angiogenesis [25]. Moreover, NAE1-mediated neddylation of CUL1 upregulates vascular endothelial growth factor C (VEGF-C) by augmenting NF-κB transcriptional activity, thereby facilitating tumor angiogenesis and uveal melanoma hepatic metastasis [26]. Furthermore, cullin neddylation-mediated activation of CRL reduces RhoA accumulation and increases the angiogenic activity of vascular endothelial cells, contributing to tumor angiogenesis [27]. These findings demonstrate that neddylation acts as an accomplice in tumor angiogenesis.
EMT enables tumor cells to transition from an epithelial to a mesenchymal phenotype, enhancing their ability to invade and migrate. Neddylation regulates EMT in tumors. RanBP2-mediated neddylation of MKK7 constrains its basal kinase activity and inhibits basal JNK phosphorylation, thus facilitating proliferation and EMT in BC cells [28]. Moreover, neddylation of tuberous sclerosis complex 2 (TSC2) inactivates the mTOR pathway, enhancing migration, invasion, and EMT in head and neck squamous cell carcinoma (HNSCC) [29]. However, neddylation also impedes the EMT process. Concretely, neddylation inactivates hypoxia-inducible factor 1α (HIF-1α) via PI3K-Akt signaling, downregulating zinc finger E-box binding homeobox1 (ZEB1) and inhibiting EMT in cancer cells [30]. Interestingly, neddylation governs tumor cell migration in a manner relying on the p53 status. Specifically, in the presence of wild-type p53, neddylation blockage increases p53 transcriptional activity and enhances p21 and MDM2 expression, ultimately leading to the proteasomal degradation of Slug and impeding EMT-involved tumor cell migration. Conversely, in cells with mutant p53, neddylation inhibition facilitates tumor cell migration by stimulating the PI3K/Akt/mTOR/Slug pathway [31].
Taken together, these findings highlight the dual function of the neddylation cascade in tumor invasion and metastasis, complicating the regulatory mechanisms that control the invasive-metastatic process through neddylation and raising concerns about the use of MLN4924 as an anti-tumor therapeutic agent.

Programmed cell death resistance
Programmed cell death (PCD) sustains physiological homeostasis by eliminating abnormal or harmful cells. Tumor cells resist PCD, endowing them with prolonged survival [32]. Notably, neddylation dysregulates the PCD process in tumors, affecting apoptosis, autophagy, and ferroptosis. Apoptosis inhibition is critical for tumor cell survival. Neddylation governs the apoptotic pathways, manipulating the tumor cell fate. In ESCC, UBC12-involved neddylation of CUL1, CUL2, CUL3, CUL4A, and CUL4B activates CRLs, leading to the depletion of activating transcription factor 4 (ATF4) and transrepression of death receptor 5 (DR5), which inhibits extrinsic apoptosis. ATF4 reduction also downregulates apoptotic protein NOXA, inactivating intrinsic apoptosis [33]. Conversely, blocking neddylation stabilizes ATF4, leading to the transactivation of TF CHOP. This upregulates DR5 and caspase-8, triggering extrinsic apoptosis [34]. Additionally, UBE2F-mediated neddylation activates CRL5, enhancing the ubiquitination and degradation of NOXA, thereby facilitating apoptosis resistance in lung cancer and CRC [35, 36]. Autophagy, which degrades intracellular organelles, proteins, and macromolecules, maintains cellular homeostasis. The neddylation cascade participates in autophagy dysregulation during tumorigenesis. Representatively, neddylation hampers autophagy through activating mTOR signaling in tumors. For instance, UBE2F-induced neddylation of RHEB at K169 attenuates autophagy in liver cancer by fortifying mTORC1 activity, thereby promoting tumor growth and survival [22]. Similarly, CUL-RING E3 ligases-induced DEPTOR depletion prevents autophagy in an MTOR-dependent manner, worsening tumorigenesis [37, 38]. Additionally, neddylation manipulates autophagy in tumors through HIF1-REDD1-TSC1-mTORC1-DEPTOR axis [39], PI3K/AKT/mTOR signaling [40], and NF-κB-catalase-ATF3 axis [41]. These findings highlight that neddylation dampens autophagy in tumors, thereby fostering the malignant phenotypes. Neddylation also governs ferroptosis in tumors. Mechanistically, neddylation E3 CRL-3 decreases cystine availability by destabilizing the cystine/glutamate antiporter SLC7A11, leading to ferroptosis resistance in BC [42]. Overall, PCD resistance is tightly linked to tumor deterioration and therapeutic resistance. Accordingly, elucidating the mechanisms by which neddylation mediates PCD evasion in tumor cells could provide novel therapeutic avenues and diagnostic approaches in oncology.

Cancer stem cell self-renewal and maintenance
Cancer stem cells (CSCs) are a subpopulation of tumor cells with extraordinary abilities to self-renew, differentiate, and maintain tumor malignancy. Strikingly, the overall level of neddylation in tumors significantly influences CSC self-renewal and maintenance. Slug maintains CSC traits by enhancing its stability in a neddylation-dependent manner [26]. Moreover, neddylation controls the expression of SRY-box transcription factor 2 (SOX2), crucial for CSC self-renewal. Mechanistically, neddylation-activated CRL E3 ligase FBXW2 mediates the ubiquitination and degradation of MSX2, a transcription repressor of SOX2, therefore upregulating SOX2 expression and boosting stem cell properties in BC [43]. However, neddylation also attenuates tumor stemness. Specifically, UBE2M-mediated neddylation of LSD1 at K63 enhances its ubiquitination and degradation, thereby reducing GC cell stemness [23]. Similarly, c-MYC, an oncoprotein that promotes CSC self-renewal, is ubiquitylated and degraded by SCF E3 ubiquitin ligases, including SCFFBXW7 and SCFSKP2, hindering tumor sphere formation, stem cell proliferation, and differentiation [44]. CSC self-renewal and maintenance are critical for tumor growth, metastasis, therapy resistance, and recurrence. Targeting CSCs offers potential treatment options for refractory and recurrent cancers. While neddylation inhibition effectively targets tumor heterogeneity, its dual role in CSC self-renewal/maintenance complicates therapeutic application, posing potential risks and challenges.

Epigenetic reprogramming
Epigenetic reprogramming regulates gene expression by altering epigenetic markers without altering the DNA sequence, playing a key role in tumor initiation, progression, metastasis, and therapeutic resistance. In tumors, neddylation serves as an epigenetic modification that coordinates other epigenetic changes, including DNA methylation, transcriptional regulation, and PTMs. Neddylation influences the activity of proteins or enzymes involved in DNA methylation, thus indirectly altering the DNA methylation pattern. Specifically, CUL4A preferentially binds to DNA methyltransferase 3b (DNMT3b) via its C terminus in a neddylation-dependent manner, therefore promoting DNMT3b-mediated DNA methylation in cancer cells [45]. Moreover, neddylation regulates other PTM processes in tumors. Characteristically, neddylation modification enables the CRL complex to recognize target proteins and enhances their degradation via the ubiquitin-lysosome pathway. UBE2F, in association with the neddylation E3 ligase RBX2/SAG, triggers the neddylation of CUL5, thus activating E3 CRL5. Subsequently, CRL5 mediates the degradation of DIRAS2 through K11-linked polyubiquitylation in KRAS mutant pancreatic cancer cells [46]. Similarly, CUL5 neddylation enhances CRL5 activity through the UBE2F/SAG/CUL5 complex, thus ubiquitylating NOXA at K11 in the proteasomal degradation pathway in lung cancer [35]. Neddylation also manipulates transcription by modulating the expression and activity of TFs, nuclear receptors, and transcriptional co-regulators in tumors. Neddylation-driven epigenetic reprogramming promotes tumor cell heterogeneity and clonal evolution. Therapeutic targeting of neddylation may offer a multidimensional strategy to eliminate malignant cells through multiple mechanisms.

Metabolic rewiring
Metabolic rewiring enables tumor cells to modify their metabolic pathways to adapt to the hypoxic, nutrient-deprived, acidic TME, supporting proliferation, immune evasion, and therapy resistance. Remarkably, neddylation supervises metabolic rewiring in tumors, influencing aerobic glycolysis, glutamine metabolism, and fatty acid synthesis. Aerobic glycolysis, or the Warburg effect, is a hallmark metabolic adaptation in tumors, which is regulated by neddylation. Specifically, glucose detaches CRL4 from the deneddylase CSN, contributing to the assembly of the CRL4COP1 E3 complex. This complex facilitates the degradation of wild-type p53, therefore enhancing the Warburg effect and driving mammary tumorigenesis [47]. Moreover, neddylation governs amino acid metabolism in tumors. The CRL3-SPOP E3 ligase, activated by neddylation, mediates the ubiquitination and K48-linked degradation of ASCT2 by interacting with the SPOP consensus motif “GTSSS,” thereby reducing cellular glutamine uptake and glutamate production in BC [48]. Furthermore, neddylation regulates fatty acid metabolism in cancers. Specifically, XIAP-mediated neddylation of PTEN at K197 and K402 reinforces its accumulation in the nucleus, potentiating FASN dephosphorylation and attenuating FASN ubiquitination, which accelerates de novo fatty acid synthesis in BC [49]. Collectively, these findings underscore the essential role of neddylation in controlling metabolic rewiring in tumors. Hence, targeting neddylation may enhance therapeutic outcomes, surmount tumor immunosuppression, and ameliorate the TME by altering the metabolic profile of tumor cells, providing a potential strategy for personalized cancer treatment.

Tumor immune evasion
The immune system recognizes and eliminates abnormal or harmful cells. Tumor cells evade immune surveillance through multiple mechanisms, including the neddylation cascade, thereby expediting tumor growth and metastasis. Strikingly, the Warburg effect exacerbates immune evasion in tumors [50, 51]. Neddylation dysregulates T-cell cytotoxicity within tumors. For instance, neddylation-induced activation of the CUL5-E3 ligase complex, interacting with SOCS-box-containing protein PCMTD2, impedes TCR/IL2 signaling and reduces CD8+ T cell cytotoxicity in tumors [52]. Therefore, pharmacologic blockage of neddylation effectively revitalizes T-cell response in malignancies [53, 54]. Neddylation also governs the cytotoxic activity of Natural Killer (NK) cells in tumors. Neddylation upregulates MICA and MICB, the ligands of NK cell-activating receptor NKG2D, by modulating MICA promoter activity and MICB subcellular localization, contributing to NK cell degranulation [55]. Furthermore, neddylation aggravates the infiltration of tumor immunosuppressive cells. Overexpression of UBA3 facilitates the infiltration of tumor-associated macrophages (TAMs), plasmacytoid dendritic cells (pDCs), Th2 cells, and T-regulatory cells (Tregs) by decreasing phosphorylated nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor (p-IκBα) and improving the gene expression of tumor cell-derived cytokines in lung adenocarcinoma [56]. Contrariwise, neddylation counteracts tumor immune evasion by downregulating the expression of immune checkpoint molecules. For instance, neddylation augments the activity of Cullin1-F-box and WD repeat domain-containing 7 E3 ligase, which destabilizes c-MYC protein, decreasing PD-L1 expression and improving T cell killing in glioblastoma [57]. Intriguingly, deneddylation also contributes to tumor immune evasion by restraining macrophage phagocytosis in colon cancer. Mechanistically, neddylation of Src homology region 2-containing protein tyrosine phosphatase 2 (SHP2) at K358 and K364 disturbs its activity and sustains its autoinhibited conformation. Deneddylation of SHP2 by sentrin-specific protease 8 (SENP8) inactivates αMβ2 integrin, ultimately impairing macrophage phagocytosis in colon cancer [58]. Neddylation exerts dual immunomodulatory effects in tumors, with outcomes determined by its regulation of tumor cells, immune components, and the TME. Understanding the mechanisms underlying tumor immune evasion provides a foundation for tumor immunotherapy.

Cellular stresses triggering neddylation in tumors
Neddylation, maintaining cellular homeostasis, is triggered by various forms of cellular stress. However, the specific stress that induces neddylation and the mechanisms through which it responds remain incompletely understood. This section explores the different stimuli and the mechanisms that induce neddylation in tumor cells (Fig. 4). Neddylation is essential for the cellular response to DNA damage in tumors. In reaction to chemotherapy-induced DNA damage, reactive oxygen species (ROS) induces UBE2F-mediated neddylation of CUL5, facilitating NOXA ubiquitination and degradation, which confers chemoresistance to CRC cells [36]. Furthermore, viruses promote cancer development, in part, by hijacking the neddylation pathway. Specifically, Kaposi’s sarcoma-associated herpesvirus (KSHV), the primary cause of Kaposi’s sarcoma, activates NF-κB signaling by recruiting the neddylated CUL1-activated CRL1/βTrCP E3 ligase to degrade IκBα. More importantly, neddylation also enhances KSHV genome replication during the reactivation of the lytic cycle [59]. In Hepatitis B virus-associated HCC, neddylation of HBx facilitates its nuclear localization, transcriptional activity, and stabilization, thereby activating the transcription of IL-8, MMP9, and YAP, which accelerates tumor growth [60]. High glucose conditions, a known oncogenic factor, stimulate tumor initiation and progression by regulating the neddylation cascade. Under hyperglycemic conditions, PTEN neddylation at K197 and K402 facilitates its nuclear import without altering protein stability, thus exacerbating BC progression [49]. Likewise, high glucose dissociates CRL4 from the deneddylase CSN, assembling the CRL4COP1 E3 complex. This, in turn, leads to the degradation of wild-type p53, accelerating mammary tumorigenesis [47]. Strikingly, metabolites also modulate neddylation by influencing enzyme activity, thereby linking cellular metabolism with protein function in tumors. Concretely, succinate induces the phosphorylation of UBC12 at the serine-6 site, impairing cullin neddylation by undermining UBC12/UBE1C interaction, ultimately stabilizing oncoproteins in AML [61]. Succinate accumulation also enhances the deneddylation of cullin1, impairing the E3 ubiquitin ligase SCF β-TrCP complex, which stabilizes and activates YAP/TAZ in HCC [62]. Additionally, 1-methyl-nicotinamide (1-MNA), a metabolite of nicotinamide N-methyltransferase (NNMT), is critical for CRL1 activation. Mechanistically, NNMT/1-MNA directly binds to UBC12 and impedes UBC12 lysosome degradation, subsequently triggering cullin-1 neddylation and leading to p27 protein decay in BC [63].
These findings indicate that cellular stress within the TME activates adaptive responses that contribute to tumor cell survival and progression. Addressing cellular stress, such as oxidative stress, metabolic pressure, and DNA damage, can help maintain cellular homeostasis, preventing tumor initiation, progression, and resistance. Targeting the cellular stress triggering neddylation in tumors, therefore, holds potential for reducing the adaptability of tumor cells and enhancing treatment efficacy.

Neddylation-involved therapeutic resistance in tumors
Tumor resistance poses a tremendous challenge in oncology, impacting treatment efficacy, patient survival, and prognosis. Resistant tumor cells withstand various therapeutic strategies, including chemotherapy, radiotherapy, targeted therapy, and immunotherapy, raising the risk of recurrence. Remarkably, neddylation contributes to tumor therapeutic resistance (Fig. 5).

Neddylation-mediated chemoresistance
Chemotherapy, a conventional anti-tumor therapeutic modality, faces the issue of chemoresistance, which contributes to tumor progression and relapse, ultimately affecting patient prognosis and survival. In tumor cells, an overactive neddylation cascade is associated with the development of wide-ranging chemoresistance, desensitizing to platinum, etoposide, and paclitaxel. Specifically, UBE2F upregulation enhances the neddylation levels and CUL5 activity, thereby reinforcing CRL5-mediated degradation of NOXA and leading to platinum and etoposide resistance in non-small cell lung cancer (NSCLC) and CRC [36, 64]. Furthermore, UBC12-mediated TRIM25 neddylation at K117 diminishes steric hindrance in its RING domain, enhancing its ability to bind ubiquitylated substrates. This, in turn, allows TRIM25 to facilitate the nuclear translocation of transcription factor EB (TFEB) and the transcription of autophagy-related genes by increasing the K63-polyubiquitination of TFEB, ultimately reducing the sensitivity of TNBC cells to paclitaxel [65]. Intriguingly, neddylation inhibition presents a promising strategy to overcome cisplatin and paclitaxel resistance in tumors [65–68]. Notably, disturbing cullin neddylation by attenuating E2 enzyme UBC12 activity accumulates downstream CRLs, thus inducing chemoresistance in AML [61], indicating that neddylation manipulates chemotherapy sensitivity. Likewise, neddylation blockage hinders the ubiquitination of histone deacetylase 1 (HDAC1) in AML, contributing to its significant upregulation and doxorubicin resistance [69].

Neddylation-mediated radioresistance
Radiotherapy prevents tumor deterioration by utilizing high-energy radiation to damage the DNA of cancer cells, disrupting their ability to divide and replicate, ultimately leading to tumor cell death. Whereas, protein neddylation confers radioresistance to tumor cells. Of note, overexpression of NEDD8 has been deemed a contributor to radioresistance [70], and neddylation blockage boosts radiation-induced DNA damage, aneuploidy, G(2)/M phase cell-cycle arrest, and apoptosis through the accumulation of SCF substrates, such as CDT1, WEE1, and NOXA [71]. Similarly, neddylation inhibition elevates the radiosensitivity of HNSCC by CRL4CDT2-mediated ubiquitination of CDT1, SET8, and p21 [72]. Moreover, CUL1 neddylation has been shown to diminish the radiosensitivity of CRC [73]. UBE2F facilitates CUL5 neddylation and activation, which ubiquitinates and degrades pro-caspase 3, IκBα, and NOXA, thus inducing adaptive radioresistance in lung cancer [74, 75]. Interestingly, cell division cycle (CDC6), a critical molecule assisting tumor cells in resisting radiation, is degraded via the CUL1 neddylation-mediated ubiquitin-proteasome pathway, indicating that neddylation also contributes to radiosensitization [76]. Likewise, overexpression of CUL2, functioning as a scaffold for CRL2VHL E3 ligase, negatively correlates with HIF-1α, VEGF-A, Cyclin B1, and EGFR levels, enhancing radiosensitivity in glioblastoma [77]. Additionally, neddylation of p53 enhances its transcriptional activity, therefore increasing the sensitivity to ionizing radiation in HNSCC [78].

Neddylation-mediated targeted therapy resistance
Tumor-targeted therapy is a highly selective cancer treatment method that impedes tumor growth and metastasis by targeting specific molecules or signaling pathways in cancer cells while minimizing damage to non-tumor cells. Neddylation contributes to targeted therapy resistance, involving tyrosine kinase inhibitor (TKI) resistance and mTOR inhibitor resistance. Concretely, neddylation of BCR-ABL at K500, K739, K802, K1025, K1135, K1590, and K1990, mediated by the NEDD8 E3 ligase RAPSYN, impairs c-CBL-driven ubiquitination, enhancing BCR-ABL protein stability and promoting TKI resistance in Philadelphia chromosome-positive (Ph + ) leukemia [79]. Furthermore, NAE1-mediated neddylation of cullins depletes CRL substrate p27kip1 in the nucleus, thereby retaining leukemia stem cells and conferring imatinib resistance in chronic myeloid leukemia (CML) [80]. In contrast, blocking neddylation sensitizes CML and Ph+ acute lymphoblastic leukemia (ALL) to TKIs targeting the ABL kinase [81]. Moreover, neddylation induces mTOR inhibitor resistance in pancreatic neuroendocrine tumors. Mechanistically, the CUL4B-DDB1-DCAF7 axis, activated by neddylation, enhances the ubiquitination and subsequent degradation of MEN1 at the K135, K151, K201, and K404 residues, therefore contributing to everolimus resistance [82]. Contrariwise, neddylation inactivation ameliorates the resistance of vemurafenib in melanoma, highlighting its inhibitory role in BRAF inhibitors [83]. Additionally, UBE2M-mediated neddylation promotes the decay of multiple substrates of CRLs, thereby shrinking the sensitivity of PARP inhibitor niraparib in castration-resistant prostate cancer [84].

Neddylation-mediated endocrine therapy resistance
Endocrine therapy prevents the growth of tumor cells by altering the hormone levels. Strikingly, neddylation is a trigger of endocrine therapy resistance in breast cancer. For instance, cullin neddylation-induced SOX2 overexpression results in tamoxifen resistance [43]. Furthermore, CAND1, a regulator of cullin-RING E3 ubiquitin ligases, prevents estrogen receptor α (ERα) from CUL2-mediated proteasomal degradation. This preservation of ERα stability in breast cancer with AKTIP loss contributes to the development of tamoxifen resistance [85]. Additionally, overactivated neddylation facilitates the decay of serum and glucocorticoid-induced protein kinase (SGK), which further impedes nuclear export of FOXO3a, transcriptionally upregulating ESR1 expression and leading to fulvestrant resistance [86].

Neddylation-mediated immunotherapy resistance
Tumor immunotherapy activates or enhances the immune surveillance, thereby enabling the recognition and destruction of cancer cells. However, neddylation propels immune evasion via dysregulating the cytotoxicity of T cells [52], impeding the activity of NK cells [55], and intensifying the infiltration of tumor immunosuppressive cells [56], ultimately contributing to immunotherapy resistance. Moreover, mutations that destabilize the DNA mismatch repair (dMMR) system lead to proteome instability in dMMR tumors, accumulating misfolded proteins. In response, dMMR cells activate a NEDD8-dependent degradation pathway to clear these misfolded proteins. Therefore, inhibition of the NEDD8-mediated clearance pathway leads to the buildup of misfolded protein aggregates, which triggers immunogenic cell death and enhances the efficacy of PD1 inhibition in dMMR cancers [87]. In glioblastoma, neddylation reduces the efficacy of immune checkpoint blockade by regulating PD-L1 levels [57]. However, neddylation enhances immunotherapy sensitivity by promoting SHP2 neddylation in CRC [58].
Neddylation serves as a “double-edged sword” in tumor therapeutic resistance, likely due to tumor heterogeneity, treatment specificity, and microenvironmental stress. Consequently, neddylation-targeted therapies should incorporate individualized strategies and combination regimens to achieve optimal therapeutic efficacy.

Antitumor potential of neddylation inhibitors
Malfunctions of neddylation catalytic enzymes frequently occur in tumors [22, 26, 65, 88], exerting a crucial effect on the cancer hallmark sustenance and anti-tumor drug sensitivity. This indicates that neddylation blockage holds significant potential for cancer treatment. Consequently, diverse neddylation inhibitors have been reported, including NAE inhibitors, E2-UBE2M-DCN1 inhibitors, E2-UBE2F inhibitors, and neddylation E3 inhibitors. However, except for MLN4924 (pevonedistat), all other inhibitors remain in the preclinical phase (Table 4).

Neddylation NAE inhibitors
MLN4924, the first-in-class small-molecule inhibitor of NAE, selectively impedes the NAE enzyme activity by establishing a covalent NEDD8-MLN4924 adduct at the NAE active site, therefore inactivating all CRLs and accumulating CRL substrates [89, 90]. Numerous preclinical studies have illustrated that MLN4924 dampens tumor malignant traits [91–94] and sensitizes tumor cells to anti-tumor therapies [95–98]. Multiple clinical trials registered on ClinicalTrials.gov have been initiated to evaluate the therapeutic efficacy and safety of MLN4924 in cancer patients (Table 5).
Uncontrolled tumor cell proliferation is a hallmark of cancer, with cell cycle dysregulation in tumorigenesis [32]. MLN4924 inhibits tumor growth by arresting the cell cycle. It accumulates CRL substrates, involving p21, p27, and CDT1, leading to G2 cell-cycle arrest and suppressing tumor cell proliferation in glioblastoma [99]. Likewise, MLN4924 triggers cell cycle arrest at the G2/M phase by reducing phospho-histone H3 and enhancing phospho-cdc2 levels, leading to the growth inhibition of urothelial carcinoma [100]. Cellular senescence, a permanent and irreversible cessation of the cell cycle and cell division, plays a significant role in the pathological processes of cancer [101]. MLN4924 impedes tumor growth by promoting p21-mediated cellular senescence. Specifically, MLN4924 inhibits the activities of CRL/SCF E3 ubiquitin ligases, involving skp2 and CRL4, thus accumulating p21 and promoting senescence in CRC, lung cancer, glioblastoma, and lymphoma [102–104]. MLN4924 typically eliminates tumor cells through PCD mechanisms. For example, MLN4924-induced CRL/SCFβTrcp stabilizes ATF4, thus stimulating the extrinsic apoptosis via the CHOP-DR5 axis in esophageal cancer cells [34]. In liver cancer cells, MLN4924 triggers autophagy and apoptosis by cullin neddylation blockage and CRL inactivation [38]. Furthermore, MLN4924 inhibits tumor angiogenesis and eliminates CSCs [26, 80]. Remarkably, MLN4924 also enhances the sensitivity of tumor cells to anti-cancer therapies. For instance, MLN4924 sensitizes CRC to topoisomerase I inhibitors by inactivating the DCAF13-CRL4 ubiquitin ligase complex [97]. Collectively, these findings indicate that MLN4924 exerts its antitumor effects by targeting the neddylation-CRL pathway, regulating the stability of tumor-related proteins, and holds potential for clinical applications. However, MLN4924 also exhibits potential tumor-promoting effects. It transcriptionally increases PD-L1 gene expression via upregulating c-MYC, contributing to immune evasion in glioblastoma [57]. Hence, combining MLN4924 with specific target inhibitors could decrease the side effects and prevent the accumulation of cancer-promoting CRL substrates [105].
Besides MLN4924, multiple covalent and non-covalent NAE inhibitors have been identified and evaluated. TAS4464, a selective NAE inhibitor, forms a covalent adduct with NEDD8 within NAE through its sulfonamide structure in an ATP-dependent manner, restricting cullin neddylation and accumulating CRL substrates [106]. Preclinical in vitro and in vivo studies demonstrate TAS4464’s antitumor efficacy in hematologic and solid tumors [106–108]. However, an open-label, phase I clinical trial of TAS4464 in patients with advanced/metastatic solid tumors was discontinued due to its liver toxicity [109]. Multiple non-covalent NAE inhibitors, encompassing natural and synthetic compounds, have been identified and characterized. The second inhibitor of NAE 6,6”-biapigenin, a natural product-like inhibitor, creates hydrogen bonds with the C-terminal Glycine of NEDD8 and the side chain of Arg15 in APPBP1, disturbing the neddylation cascade [110]. Other natural compounds and FDA-approved drugs, such as Flavokawain A and B extracted from kava roots [111, 112], Gartanin identified from the purple mangosteen fruit [113], and FDA-approved drug piperacillin 1 [114], also interact with NAE and impede its enzymatic activity. However, all these inhibitors are merely in the preclinical phase, lacking clinical safety and efficacy evaluation.

Neddylation E2 inhibitors
In mammals, UBE2M and UBE2F are the merely two known neddylation E2 conjugating enzymes, responsible for transferring NEDD8 from NAE to the E2 cysteine active site. Blocking the E2 enzyme activities prevents the binding of NEDD8 to target proteins. Therefore, neddylation E2 inhibitors hold promise as a therapeutic strategy in cancer treatment. UBE2M partners with RBX1 to facilitate the neddylation of CUL1, CUL2, CUL3, CUL4A, and CUL4B, activating the CRLs and manipulating their substrate levels [115]. UBE2M is aberrant in various tumors, and its blockade impedes malignancy [33, 116–120]. Micafungin, an antifungal agent, hinders the conjunction of NEDD8 to UBE2M and disrupts the neddylation cascade. Micafungin forms hydrogen bonds with UBE2M, particularly with its Glu107 and Asn109, suggesting that it may potentially disrupt the catalytic cysteine (Cys111) of UBE2M. In GC cells, Micafungin accumulates CRL substrates, therefore impeding malignant phenotypes and inducing DNA damage [121]. Arctigenin, a natural compound, effectively suppresses UBC12 enzyme activity, reducing cullin neddylation and abrogating tumor progression [122]. Additionally, WS-299 specifically hinders CUL3/5 neddylation by disrupting the PPI between RBX1 and UBE2M, leading to NOXA-mediated apoptosis in GC [123].
Analogous to UBE2M, UBE2F, possessing an N-terminal extension with a conserved catalytic core region, activates CRLs by catalyzing cullin neddylation. UBE2F contributes to tumor progression, making it a potential target for anticancer therapy. Overexpressed UBE2F enhances the growth, survival, and platinum-insensitivity of tumor cells. Mechanistically, UBE2F, integrating with RBX2, facilitates CUL5 neddylation, activating CRL5 and leading to NOXA depletion and apoptosis escape [35, 36, 64]. HA-9104, a small-molecule inhibitor targeting UBE2F, decreases UBE2F protein levels, thus boosting tumor cell apoptosis and radiosensitization through the accumulation of NOXA and deceleration of DNA damage repair [74].
Collectively, the development of high-throughput screening assays has advanced the identification of inhibitors targeting the “undruggable” E2 conjugating enzymes. Nevertheless, their clinical potential still needs to be thoroughly assessed.

UBE2M-DCN1 inhibitors
Defective in cullin neddylation 1 (DCN1), an indispensable scaffold-like co-E3 ligase in the neddylation process, is dysregulated in multiple tumors [124, 125]. DCN1 directly interacts with UBE2M at a region that coincides with the E1-binding site, therefore facilitating the activation of CRLs by selectively catalyzing cullin neddylation and governing the activity of substrate proteins [126]. Consequently, disturbing UBE2M-DCN1 interaction presents another strategy for dampening the neddylation cascade. Three peptidomimetic derivatives, including DI-591, DI-404, and DI-1859, have been identified as selective inhibitors of the UBE2M-DCN1 interaction. DI-591, a high-affinity, cell-permeable small-molecule inhibitor that effectively blocks the interaction between DCN1 and UBE2M, selectively transforms cellular CUL3 into an un-neddylated, inactive form, with minimal or no effect on other cullin family members, contributing to a dose-dependent accumulation of NRF2 across various cancer cell lines [127]. DI-404, a further optimized small-molecule inhibitor derived from DI-591, targets the active sites of DCN1 more precisely. Similar to DI-591, DI-404 effectively and selectively blocks UBE2M-DCN1 interaction and CUL3 neddylation in several cancer cell lines [128]. DI-1548 and DI-1859, two potent covalent inhibitors targeting DCN1, specifically inhibit CUL3 neddylation without significantly affecting other cullins, and robustly increase NRF2 protein levels in the mouse liver. Importantly, DI-1859 effectively reduces acetaminophen-induced liver toxicity in mice [129].
Several piperidinyl urea derivatives have been identified as UBE2M-DCN1 inhibitors. NAcM-HIT specifically prevents the binding of DCN1 to UBE2M via occupying the N-acetyl-methionine binding pocket of DCN1, hindering DCN1-dependent cullin neddylation [130]. Compounds 49 and 52, analogs of NAcM-HIT, demonstrate a 100-fold increase in biochemical potency with improved solubility and permeability, though exhibiting low stability. These analogs selectively decline CUL1 and CUL3 neddylation levels, demonstrating the potential for anti-tumor activity [131]. Likewise, NAcM-OPT, an orally bioavailable inhibitor with satisfactory solubility and permeability, selectively decreases CUL1 and CUL3 neddylation levels and restrains cancer hallmarks in the squamous cell carcinoma line [132].
Through optimization of pyrazolo-pyridone compounds, a new category of inhibitors featuring a pyrazolo-pyridone core structure is developed, proficiently impeding UBE2M-DCN1 interaction. These pyrazolo-pyridone derivatives contain two chiral centers, which enhance their three-dimensional complexity, potentially leading to stronger binding affinity, greater target specificity, and enhanced solubility. Among them, compound 27 exhibits 25-fold greater potency than the initial hit, effectively stabilizing DCN1 and specifically decreasing the CUL1 and CUL3 neddylation levels in tumor cells [133]. Subsequently, compound 40 is identified to hinder UBE2M-DCN1 interaction with enhanced oral bioavailability, hampering tumor cell expansion [134].
Pyrimidine-based UBE2M-DCN1 inhibitors have also been discovered, including WS-291, WS-383, and triazolo[1,5-a]pyrimidine-based inhibitors. WS-383 specifically inhibits CUL1 and CUL3 neddylation via thwarting the UBE2M-DCN1 interaction, thus accumulating p21, p27, and NRF2 [135]. Two novel pyrimidin-based small molecular inhibitors, DC-1 and DC-2, have been unearthed. DC-2, a thiazole-containing 5-cyano-6-phenylpyrimidin-based inhibitor, selectively hinders UBE2M-DCN1 interaction, reducing CUL3 neddylation and accumulating NRF2 [136]. Compound DN-2, a 2-(Benzylthio)pyrimidine-based inhibitor targeting DCN1, is derived from structure-based optimizations [137]. While no UBE2M-DCN1 inhibitors have entered clinical trials, they offer valuable insights for future drug discovery.

Neddylation E3 inhibitors
Neddylation E3 ligases determine the specificity of target substrates and assist in transferring NEDD8 from the E2 to the substrates. Therefore, targeting neddylation E3s holds potential for enhancing selectivity and reducing side effects. However, inhibitors targeting neddylation E3 ligases remain scarce. RBX1 and RBX2/SAG are E3 ligases involved in the neddylation. RBX1 works with UBE2M to facilitate the neddylation of CUL1-4, while RBX2/SAG collaborates with UBE2F to mediate the neddylation of CUL5 [7]. C64, a small-molecule inhibitor, preferentially binds to the “VLYRLWLN” structural motif at the RBX1-binding grooves of CUL scaffolds, impairing RBX1-CULs interaction and impeding cancer cell survival [138]. Moreover, gossypol, a natural compound derived from cottonseed, hinders the neddylation of CUL1 and CUL5 by directly binding to the RBX1-CUL1 or SAG-CUL5 complex, contributing to NOXA and MCL1 accumulation and tumor growth inhibition [139]. Advancing highly specific neddylation E3 inhibitors is crucial for effective pathway disruption.

Conclusions and perspective
NEDD8 covalently modifies substrate lysine, regulating protein stability, function, and localization. Neddylation targets include cullins and non-cullin proteins. Cullin neddylation induces conformational alterations that remove CAND1, enabling CRL activation. This requires NEDD8 binding to a C-terminal cullin lysine, promoting functional CRL assembly and subsequent substrate ubiquitination [140, 141]. Neddylation of non-cullin substrates influences their stability, activity, or PPI. Nowadays, several bioinformatics tools that predict protein neddylation sites using machine learning have been developed, such as DTL-NeddSite (http://dtl-neddsite.bioinfogo.org/) and NeddPred (123.206.31.171/NeddPred/), helping researchers identify potential interaction sites on neddylated proteins. Neddylation mechanisms remain elusive. Some unresolved questions include: How does neddylation precisely select specific substrates? What other non-cullin proteins can undergo neddylation? Beyond the well-known E1, E2, and E3 enzymes, what other enzymes participate in the neddylation cascade? Most importantly, is targeting neddylation for anticancer therapy both safe and effective? Additionally, neddylation does not function in isolation but exhibits intricate crosstalk with other ubiquitin-like modifications (UBLs), particularly ubiquitination and SUMOylation. Recent advances in neddylation proteomics have uncovered the formation of hybrid NEDD8-ubiquitin and NEDD8-SUMO2 chains [142], indicating a previously unrecognized layer of post-translational regulation. These hybrid chains may serve as unique molecular signals, fine-tuning protein stability, localization, and functional interactions in ways distinct from canonical modifications, providing novel molecular mechanisms for proteostasis. Neddylation dysregulation continually occurs in malignancies, exhibiting dual roles in promoting and suppressing tumor progression. Remarkably, neddylation, a non-mutational epigenetic modification, is activated by diverse cellular stress and sustains cancer hallmarks while modulating therapy sensitivity. Therefore, targeting the neddylation enzymes is a potential strategy for anti-tumor treatment. Several small-molecule inhibitors have been established to target neddylation. Representatively, MLN4924 inhibits neddylation, suppressing tumor malignancy, and has entered phase I-III clinical trials. However, MLN4924 also exerts a tumor-promoting effect in some contexts. Since the neddylation cascade participates in essential physiological processes, its inhibition may cause on-target toxicity in normal cells. To address these challenges, the multidisciplinary development of selective neddylation inhibitors is crucial for effective cancer treatment.