Nucleus-targeted Silencer nanoplatform regulating ZEB1-AS1 in head and neck squamous cell carcinoma therapy Authors: Haojie Yang, Yangfan Zhang, Zicong Tan, Zihao Liu, Yingzhe Yan, Qin Li, Phei Er Saw, Ning Liufu, Fengtao Ji Published: 2024 ABSTRACT AbstractLong noncoding RNAs have emerged as key players in the progression of head and neck squamous cell carcinoma (HNSC). Among them, ZEB1-AS1 was identified as an upregulated candidate in HNSC through comprehensive analysis of RNA-sequencing datasets. Here, elevated ZEB1-AS1 expression was correlated with poor prognosis in HNSC patients. Further investigations demonstrated that downregulation of ZEB1-AS1 induced epithelial-mesenchymal transition and increased sensitivity to cisplatin in Cal27 cells, while its upregulation reversed these effects, underscoring its pivotal role in tumor metastasis and cisplatin resistance in Cal27 cells. Mechanistically, ZEB1-AS1, located in cytoplasm and nucleus, directly regulated the expression of ZEB1, thereby influencing the expression of μ opioid receptor (MOR) and implicating in cancer progression. To advance clinical translation, we employed a nucleus-targeting nanoparticle platform for efficient delivery of a mixture of antisense oligonucleotides and siRNA (Silencer), effectively manipulating ZEB1-AS1 expression in vitro and in vivo. Besides, a predictive model for HNSC patients was developed by analyzing the expression levels of ZEB1-AS1, ZEB1, and MOR in the HNSC datasets. Our study underscored the critical role of ZEB1-AS1 in HNSC and its potential as a therapeutic target. By elucidating its functional mechanisms and utilizing a nucleus-targeting nanoparticle platform for efficient delivery, we proved the potential of ZEB1-AS1-targeted therapies in HNSC. Supplementary InformationThe online version contains supplementary material available at 10.1186/s11671-024-04148-9.Keywords: ZEB1-AS1, μ opioid receptor, Nanoparticles, ZEB1-AS1/ZEB1/MOR axis Introduction Tongue squamous cell carcinoma (TSCC) is a prevalent form of head and neck squamous cell carcinoma (HNSC). Despite the widespread use of comprehensive treatment approaches, including surgery, neoadjuvant therapy, immunotherapy, chemotherapy, and radiotherapy, the 5-year overall survival rate remains disappointingly low, at around 40% [1]. Moreover, HNSC patients often face challenging long-term prognoses due to issues such as distant metastasis, chemotherapy resistance, and weak immune responses [2]. Identifying a critical therapeutic target for HNSC patients is an urgent priority. The long non-coding RNA zinc finger E-box-binding protein 1 (ZEB1-AS1), derived from the promoter of zinc finger E-box-binding protein 1 (ZEB1), functions as an antisense transcript without coding ability [3]. Previous researches have consistently shown that ZEB1-AS1 is upregulated in various malignant cancers, including esophageal squamous carcinoma (ESCC) and colorectal cancer [4, 5]. Moreover, elevated ZEB1-AS1 expression has been linked to tumor metastasis, increased tumor proliferation, advanced clinical stage, and unfavorable outcomes such as reduced overall survival (OS), disease-free survival (DFS), and recurrence-free survival (RFS) [6]. These findings underscored the potential importance of ZEB1-AS1 as a critical player in tumorigenesis and a promising biomarker for cancer prognosis. ZEB1-AS1 and its interaction with ZEB1 have been shown to participate in intricate regulatory networks in various cancers. For instance, in non-small-cell lung cancer (NSCLC), the ZEB1-AS1/miRNA-409-3p-ZEB1 positive feedback loop has been implicated in tumor development [7]. In osteosarcoma, ZEB1-AS1 acted as a sponge for miR-200 s and directly targeted the 3 ‘UTR of ZEB1 mRNA, contributing to tumor proliferation and migration through the ZEB1-AS1/miR-200 s/ZEB1 regulatory axis [8]. Similarly, in prostate cancer, ZEB1-AS1 accelerated malignant progression by binding to histone methyltransferase MLL1, leading to H3K4ME3 modification and inducing ZEB1 transcription [9]. Notably, in cervical cancer (CC), downregulation of ZEB1-AS1 inhibited EMT by suppressing ZEB1 expression [10]. However, the specific role of ZEB1-AS1 in HNSC remained to be elucidated. Further exploration of its functions in HNSC could shed light on potential therapeutic targets for this challenging disease. Polymer based nanostructures were commonly used in RNA delivery [11]. To achieve highly selective and effective delivery, nanostructures targeting organelle were developed [12, 13]. For examples, the nanoparticles made with PLGA could be designed to target different organelles. Triphenylphosphonium (TPP) could be used and the PLGA-b-PEG-TPP polymer was designed [14, 15]. Nanoparticles made with this polymer were found to targeting mitochondria effectively. To achieve nuclear targeting, the nuclear location sequence (NLS) functionalized PLGA nanoparticles were designed and proved to enter the nuclear successfully [16, 17]. PLGA nanoparticles could even target endoplasmic reticulum (ER) by conjugating the ER-targeting peptides to its surface [18]. Except for the usage of organelle targeting peptides, the membrane fusion was also an effective approach for organelle targeting [19]. In this study, we utilized the TCGA database of HNSC and identified the crucial LncRNA ZEB1-AS1, which played a role in EMT and cisplatin chemotherapeutic resistance in the Cal27 cell lines by regulating ZEB1 and its downstream target molecule μ opioid receptor (MOR). Based on this regulatory relationship, we established a predictive model using the least absolute shrinkage and selection operator (LASSO) regression analysis, incorporating the expression levels of ZEB1-AS1, ZEB1, and MOR to evaluate the tumor immune microenvironment, immunotherapy responsiveness, and signal pathways associated with tumor progression. Furthermore, we employed a nuclear targeting nanoparticle platform to deliver ZEB1-AS1 Silencer, effectively inhibiting tumor metastasis and reversing chemoresistance. The findings suggested that targeting ZEB1-AS1 might represent a promising therapeutic strategy for HNSC treatment. Material and methods Patients and clinical information Tumor tissue and adjacent para-carcinoma tissue were procured from the Pathology Department of Sun Yat-sen Memorial Hospital. Prior to sample collection, patients provided informed consent, and the study protocol received approval from the hospital’s ethics committee. A total of 101 TSCC tissues and 67 para-cancerous samples were gathered for the purpose of conducting In Situ Hybridization (ISH). These tissue specimens were derived from patients diagnosed before January 2018. Tissue samples from patients with cisplatin-resistant (CRP) tumors were subjected to standardized chemotherapy prior to surgical intervention and all the tissues samples were acquired from the Pathology Department of Sun Yat-sen Memorial Hospital. In accordance with the “Response Evaluation Criteria in Solid Tumors” outlined by the World Health Organization [20], patients demonstrating progressive disease or stable disease were categorized as CRP, while those exhibiting partial or complete sensitivity were categorized as cisplatin-sensitive (CSP). For the RNA Fluorescence In Situ Hybridization (RNA-FISH) assay, a total of 13 patients with cisplatin chemotherapy-resistant cancer tissues and corresponding para-cancerous tissues were included. The tissue samples utilized in this assay were collected from patients diagnosed prior to January 2010. In situ hybridization (ISH) For the ISH procedure, we employed a digoxin-labeled oligonucleotide probe specific to ZEB1-AS1, which was a polyphase oligonucleotide probe sourced from Bosterbio (Wuhan, China). The sequence of the target gene probe for ZEB1-AS1 was as follows: 5’-TGGTACAAGTTTTGGGATTT CTGAAAAGACCATACAGATA-3’, 5’-CAGTATTAAGTAGAATTTAGATTTAGGTTTCCTTCCTGCT-3’, 5’-TTCCGGAATTTTAGAATATTCATTAATAATGTCTTCAAAC-3’, The ISH kit (Product number: MK11112-h), compatible with the aforementioned probe, was also obtained from Bosterbio. The experimental procedures were executed following the instructions provided in the kit. The primary steps encompassed the dewaxing and rehydration of paraffin-embedded specimens, enzymatic digestion using trypsin to expose nucleic acid fragments, and fixation with paraformaldehyde. Sequentially, the following components were added to the specimen: prehybridization solution, hybridization solution along with the probe, sealing solution, biotinylated rat anti-digoxin antibody, streptavidin-avidin–biotin complex (SABC) solution, and biotinylated peroxidase. Ultimately, color development was accomplished using DAB. Staining Intensity Score = Staining Intensity Score × Staining Positivity Rate Score. Staining Intensity Score: 0 (Negative), 0.5 (Visible Weak Staining), 1 (Moderate Staining Intensity), 2 (Slightly Higher Staining Intensity), 3 (Strong Positive). Staining Positivity Rate Score: 1 (1 ~ 25%), 2 (26 ~ 50%), 3 (51 ~ 75%), 4 (76 ~ 100%). The scoring was performed independently by two pathologists, and in cases of discrepancy, the average score was used. The staining images were captured using an IX83 (OLYMPUS, Japan). RNA fluorescence in situ hybridization (RNA FISH) The ZEB1-AS1 probe and corresponding kit were procured from GenePharma (Shanghai, China). The probe sequence for ZEB1-AS1 is as follows: 5’-Cy3-AGAAATCCCAAAACTTGTACCAAGGAC tgccTGCCTGTACTTAAAATGCGGTAGGCGACTGAAAAACATTAGGCCAGAAAGAGAACATGAGAGCCCAAACTAACTAAACCAGAA-3’. For paraffin-embedded TSCC tissues, the initial step involved preheating the specimens in an oven at 60 ℃ until the paraffin melted. Following this, deparaffinization was carried out using absolute xylene (Sigma), and incubation in alcohol solutions of varying concentrations (100%, 95%, 90%, 80%, 70%) was conducted at room temperature. After cleaning with PBS, the protease K working solution was applied. Subsequent steps included Buffer C rinsing, alcohol dehydration (70%, 80%, 90%, 95%, 100%), sequential addition of pre-denaturation solution, Buffer E, and the probe mixture, along with post-incubation washing using the provided washing solution. Nuclei were counterstained with DAPI, and statistical analysis was performed on captured images (IX83, OLYMPUS, Japan). For the cell RNA FISH assay, cells were cultured in a 24-well plate with a glass substrate at the bottom. Cells were fixed using 4% paraformaldehyde. Prehybridization solution, Buffer A, Buffer C, Buffer E, and the probe were successively added for incubation. Subsequent cleaning steps involved the use of Buffer F and Buffer C. Finally, nuclei were stained with DAPI, samples were washed with PBS (Beyotime), and images were captured (IX83, OLYMPUS, Japan). The scoring criteria and process were the same as for ISH staining mentioned above. Animal experiments Male Balb/c nude mice aged 4–5 weeks were procured from the Sun Yat-sen University Laboratory Animal Center. The experimental protocols were granted approval by the Sun Yat-sen University Laboratory Animal Care and Use Committee (AP20220174). The establishment of a lung metastasis model was achieved by injecting the Cal27 cells transfected with Luciferase suspension into the rat tail vein. After a period of 35 days, mice were anesthetized using isoflurane (RWD, Shenzhen, China). Subsequently, abdominal injection of D-luciferin (15 mg/mL, Yeasen, Shanghai) was carried out, and the mice were then subjected to imaging using an In Vivo Imaging System (IVIS 200, Xnogen, Alameda, USA). Following imaging, the mice were humanely euthanized, their lungs were collected, and subsequent fixation was performed using paraformaldehyde for further experimentation. In a distinct set of experiments involving the treatment of tumor-bearing animals with injected nanomaterials, a lung metastatic tumor model was established successfully. Starting from the 18th day post-model establishment, the tumor-bearing mice were randomly divided into four groups. Nanomaterial injections were administered once every 2 days for a consecutive span of 6 treatments. Fluorescence signal intensity of the tumors was assessed on the 26th and 36th days. Following the final imaging session, the nude mice were euthanized, and their lungs were collected for further analysis. For this subcutaneous tumor animal experiment, a total of 1 × 107 Cal27 R cells were suspended in 0.1 ml of PBS and orthotopically injected directly into the dorsal region of mice. Once tumors reached a palpable size, with a diameter of approximately 2 mm, the Cal27 R xenograft mice were randomly divided into eight groups as follows: (1) Negative control group, (2) Silencer group, (3) NPs-Silencer group, (4) NT-NPs-Silencer group, (5) Cisplatin + Negative control group, (6) Cisplatin + Silencer group, (7) Cisplatin + NPs-Silencer group, (8) Cisplatin + NT-NPs-Silencer group. In the cisplatin treatment group, one week after inoculation, cisplatin (5 mg/kg) was administered through intraperitoneal injections every 2 days. Concurrently, nanomaterial or naked Silencer was injected intratumorally every 2 days in all treatment groups, and tumor volumes were regularly measured. Upon completion of the experiment (Day 49), all tumors were collected for immunohistochemistry (IHC) staining. Preparation and characterization of nucleus targeting nanoparticles (NT-NPs) Methoxyl-poly (ethyleneglycol)-b-poly (2-(diisopropylamino) ethyl methacrylate) (Meo-PEG-b-PDPA) was dissolved in DMF at a concentration of 10 mg/ml. Cationic lipid-like compound alkyl-modified polyamidoamine (PAMAM) dendrimer (G0-C14) was synthesized through ring opening of 1,2-epoxytetradecane by generation 0 of (PAMAM) dendrimer according to our previous studies [21].Subsequently, 1 nmol of Silencer (dissolved in a 0.1 nmol/μl aqueous solution), 10 μl of the nucleus targeting peptide DMSO mixture (at a concentration of 5 mg/ml), and 50 μl of G0-C14 in DMF (at a concentration of 5 mg/ml) were combined. This mixture was then mixed with 100 μl of Meo-PEG-b-PDPA buffer. After vigorous stirring at 1000 rpm, 5 ml of deionized water was added to the mixture. The resulting mixture was filtered through an ultrafiltration device (EMD Millipore, MWCO 100 kDa). Following PBS washing, the NT-NPs-Silencer was collected in 1 ml of PBS (pH 7.4) for subsequent use. The size and zeta potential of the NPs were assessed using dynamic light scattering (DLS) with equipment from Malvern, USA. Cy5-labeled Silencer (from Riobio, Guangzhou) was encapsulated within the NPs to determine Silencer encapsulation efficiency (EE%). For this, 5 μl of NPs suspension was mixed with a 20-fold dilution of DMSO. A standard solution was created using 5 μl of naked cy5-Silencer dissolved in 20-fold DMSO. The fluorescence intensity (FI) was measured using a multimode microplate reader (BioTek, USA), and EE% was calculated as follows: EE% = (FINPs/FI standard) × 100. NPs with an EE% exceeding 80% were selected for subsequent experiments. Bioinformatics analysis RNA-Sequencing read counts and standardized FPKM gene expression values were obtained from the TCGA database (https://portal.gdc.cancer.gov/) for TCGA-HNSC dataset. This study enrolled 496 cancer samples and 44 normal samples. R software (version 4.1-7) and R package were used for establishment of prognostic risk model, differentially gene expression and so on. All data is available in Supplementary information. Statistical analysis All experiments were conducted independently, with a minimum of three repetitions for each. Kaplan–Meier method with a log-rank test was used to compare the prognosis of survival events of cancer patients. A two-tailed Student’s t-test and one-way analysis of variance (ANOVA) with a Tukey’s post hoc test were used for the intergroup comparison of normally distributed continuous variables. A Wilcoxon signed-rank test was used for CIBERSORT analyzed immune cell fraction comparison. Statistical significance was indicated as follows: "ns" represents not significant, * indicates p < 0.05, ** denotes p < 0.01, *** signifies p < 0.005, and **** indicates p < 0.001. Results Identifying ZEB1-AS1 as a prognostic marker and its association with cisplatin resistance in HNSC To investigate the role of LncRNAs ZEB1-AS1 in HNSC progression, we collected RNA sequencing data from the TCGA database. We assessed the expression levels of ZEB1-AS1 in 26 human cancer tissue samples and their corresponding normal samples from TCGA. Our findings revealed significant upregulation of ZEB1-AS1 expression in 14 cancer types, downregulation in 5 cancer types, and no significant change in expression in 7 cancer types (sFig. 1A). As depicted in (sFig. 1B and C), the expression of ZEB1-AS1 was significantly upregulated in unpaired and paired tissue samples of HNSC patients, and higher ZEB1-AS1 levels were correlated with poorer DSS (sFig. 1D), DFS (sFig. 1E), and OS (sFig. 1F) compared to lower ZEB1-AS1 levels. Since HNSC dataset mainly comprises of TSCC in TCGA cohort, the roles of ZEB1-AS1 in TSCC remains unclear. In order to investigate the correlation of ZEB1-AS1 with pathological parameters in patient samples, In Situ Hybridization (ISH) was used to detect the expression of ZEB1-AS1 in tumor specimens (n = 101) and adjacent para-cancer specimens (n = 66). The ISH experimental results demonstrated that ZEB1-AS1 expression in tumors was significantly higher than that in para-cancer tissues (sFig. 2A–C), and increased ZEB1-AS1 expression was linked to poorer patient prognosis (sFig. 2D). Furthermore, RNA Fluorescence In Situ Hybridization (FISH) was employed to detect ZEB1-AS1 in cisplatin chemotherapy-sensitive cancer tissues and cisplatin chemotherapy-resistant cancer tissues (n = 13). The levels of ZEB1-AS1 in the tissues of cisplatin-resistant patients (CRP) were found to be higher than those in cisplatin-sensitive patient (CSP) tumor tissues (sFig. 2E). Both the ISH and RNA FISH experiments indicated that ZEB1-AS1 was expressed in the cytoplasm and nucleus. ZEB1-AS1 promoted the progression and cisplatin resistance of Cal27 cells After confirming the higher expression of ZEB1-AS1 in HNSC samples, we investigated its expression and function in cancer cell lines. ZEB1-AS1 showed higher levels in SCC9 cells and Cal27 cells compared to Normal cells, and localized in the cytoplasm and nucleus (Fig. 1A–C). Next, the Transwell assay results indicated that silencing ZEB1-AS1 (sFig. 3A) inhibited the invasion and migration of Cal27 cells, while overexpressing ZEB1-AS1 (sFig. 3B) promoted the invasion and migration of Cal27 cells (Fig. 1D, sFig. 3C and D). Additionally, ZEB1-AS1 knockdown increased E-cadherin expression and reduced N-cadherin and vimentin expression (Fig. 1E), with the opposite effects observed upon ZEB1-AS1 overexpression (Fig. 1F, sFig. 3E and F). We further investigated the relationship between ZEB1-AS1 and chemoresistance in Cal27 cells. Based on drug sensitivity analysis, the higher level of ZEB1-AS1 had a lower sensitivity (Fig. 1G). Overexpression of ZEB1-AS1 resulted in higher IC50 of cisplatin, indicating reduced sensitivity to the drug, while, ZEB1-AS1 silencing increased cisplatin sensitivity (Fig. 1H). The cisplatin-resistant Cal27 (Cal27 R) cells showed higher ZEB1-AS1 expression and increased IC50 compared to parental Cal27 (Cal27 P) cells (Fig. 1I and J). Cal27 R cells showed better motility than Cal27 P cells (sFig. 3G and H). Silencing ZEB1-AS1 restored cisplatin sensitivity and inhibited the motility of Cal27 R cells (Fig. 1L). Moreover, Cal27 R cells exhibited stronger migration and invasion abilities (Fig. 1M), accompanied by decreased N-cadherin and vimentin levels, and increased E-cadherin levels in Cal27 P cells (Fig. 1N, sFig. 3I). Effects of ZEB1-AS1 on Cal27 cells progression and cisplatin resistance A. qRT-PCR analysis of ZEB1-AS1 expression. B. qRT-PCR analysis of ZEB1-AS1 expression in the cytoplasm and nucleus. C. RNA FISH showing the cytoplasm and nuclear localization of ZEB1-AS1 in Cal27 cells. D. Transwell assay evaluating migration and invasion of Cal27 cells after ZEB1-AS1 knockdown or overexpression. E. Western blot analysis of EMT markers following ZEB1-AS1 overexpression in Cal27 cells. F. Western blot analysis of EMT markers following ZEB1-AS1 knockdown in Cal27 cells. G. Prediction of cisplatin sensitivity based on ZEB1 expression levels. H. Cisplatin sensitivity of Cal27 cells manipulated by ZEB1-AS1 upregulation or downregulation. I. Alamarblue assay evaluating cisplatin sensitivity in Cal27 R and Cal27 P cells at different cisplatin concentrations. J. qRT-PCR analysis of ZEB1-AS1 expression in Cal27 R and Cal27 P cells. K. Alamarblue assay evaluating cisplatin sensitivity in Cal27 R cells after ZEB1-AS1 knockdown. L. Transwell assay assessing migration and invasion of Cal27 R cells after ZEB1-AS1 knockdown. M. Transwell assay comparing migration and invasion of Cal27 R and Cal27 P cells. N. Western blot analysis of EMT markers in Cal27 R and Cal27 P cells; *p < 0.05; ** p < 0.01. ZEB1-AS1 promoted Cal27 cells progression and cisplatin resistance through ZEB1. Analysis using the CISTROME database indicated that ZEB1-AS1 likely facilitates tumor malignancy and cisplatin resistance through ZEB1 in other cell lines (Fig. 2A) [22]. Furthermore, we observed a positive correlation between ZEB1-AS1 and ZEB1 expression in tumors (sFig. 4A), and with higher ZEB1 levels in Cal27 R cells compared to Cal27 P cells (sFig. 4B and C). Silencing ZEB1-AS1 led to a significant downregulation of ZEB1 at both the protein (Fig. 2B) and mRNA levels (Fig. 2C). Dual-luciferase assay confirmed that ZEB1-AS1 enhanced the activity of the ZEB1 promoter in Cal27 cells (Fig. 2D). This indicated that ZEB-AS1 was associated with ZEB1 expression and could facilitate cisplatin resistance by regulating ZEB1. Previous studies have confirmed that ZEB1 promotes cisplatin resistance in different types of cancers [23]. Consistently, ZEB1 overexpression (Fig. 2E, sFig. 4D and E) in Cal27 P cells promoted cancer cell cisplatin resistance (Fig. 3F) and promoted cancer cell migration and invasion (Fig. 2G, sFig. 4F and G). The data from the GDSC database also corroborated our experimental findings (Fig. 2H). Furthermore, ZEB1 knockdown (Fig. 2I, sFig. 4H) increased the sensitivity of Cal27 R cells to cisplatin (Fig. 2J), inhibited Cal27 R cells metastasis, while overexpressing ZEB1-AS1 partially reversed the suppressive effect of ZEB1 silencing on tumor metastasis (Fig. 2K, sFig. 4I). These findings suggest that ZEB1-AS1 may modulate HNSC cisplatin resistance through regulating ZEB1. ZEB1-AS1 enhanced ZEB1 expression ZEB1 positively activated the MOR promoter region A. ChIP-Seq data from three cell types were downloaded from the CISTROME database and visualized using IGV software. B. Western blot shows changes in ZEB1 protein levels after ZEB1-AS1 knockdown. C. The quantitative data showed the level of ZEB1 mRNA after knockdown of ZEB1-AS1 in Cal27 R cells. D. Dual luciferase reporter assay suggests that ZEB1-AS1 activates the ZEB1 promoter region. E. Western blot shows changes in ZEB1 protein levels after ZEB1 overexpression. F. Alamarblue assay tests cisplatin sensitivity of Cal27 P cells with ZEB1 upregulation. G. Representative images of Cal27 P cells migration/invasion after ZEB1 overexpression. H. Prediction of cisplatin sensitivity based on ZEB1-AS1. I. Western blot shows changes in ZEB1 protein levels after ZEB1 knockdown. J. Alamarblue assay tests cisplatin sensitivity of Cal27 R cells with ZEB1 knockdown. K. Representative images of migrated and invaded Cal27 R cells after ZEB1 knockdown or ZEB1-AS1 co-transfection; *p < 0.05; ** p < 0.01; ***p < 0.005; ****p < 0.001. ZEB1-AS1 upregulated the expression of MOR in a ZEB1 dependent manner in vivo and in vitro As a transcription factor, ZEB1 ould regulate the expression of various target genes by manipulating their promoter transcriptional activity, such as PDL1, CD47, GPX4 and CCL8 [24–27]. We hypothesized that ZEB1-AS1 may enhance MOR expression through ZEB1, thus promoting tumor metastasis and cisplatin resistance. The JASPER database predicted several ZEB1 binding sites in the MOR promoter region, which were further confirmed through dual-luciferase assay (sFig. 5A and B) [28]. ZEB1 increased MOR promoter activity through specific binding sites (Fig. 3A). Additionally, MOR expression was higher in Cal27 R cells compared to Cal27 P cells (Fig. 3B, sFig. 5C). Knocking down ZEB1 decreased MOR mRNA and protein levels (Fig. 3C, sFig. 5D). Silencing ZEB1-AS1 in Cal27 R cells reduced the expression of both ZEB1 and MOR (Fig. 4D, sFig. 4E and F), and overexpressing ZEB1 partially reversed the downregulation of MOR caused by ZEB1-AS1 knockdown (Fig. 3E, sFig. 5G). Then, stable luciferase-labeled Cal27 P and Cal27 R cells with silenced or overexpressed ZEB1-AS1 were injected into the tail vein of nude mice to establish a lung metastasis model. The results demonstrated that the fluorescence signals in the lungs of the ZEB1-AS1 silenced group were weaker compared to the vector group, whereas the ZEB1-AS1 overexpressed group exhibited stronger fluorescence signals. Moreover, the fluorescence signal in the Cal27 R cells group was significantly stronger than that in the Cal27 P cells group (Fig. 3F, sFig. 5H). Histological analysis revealed that upregulation of ZEB1-AS1 enhanced the metastatic lesions with increased ZEB1 and MOR staining in larger areas, whereas ZEB1-AS1 downregulation led to reduced metastatic lesions and lighter ZEB1, MOR staining (Fig. 3G and H). These findings support our hypothesis that ZEB1-AS1 modulates MOR expression may through ZEB1 regulation. Synthesis and characterization of Silencer-incorporated nanoparticles A. Dual-luciferase assay using constructs with mutations in the three ZEB1 binding sites co-transfected in 293 T cells confirmed the significance of these MOR binding sites for the functional interaction of ZEB1 and MOR. B. MOR expression was significantly higher in Cal27 R cells compared to Cal27 P cells. C. Western blotting demonstrated the downregulation of MOR protein after ZEB1 knockdown in Cal27 R cells. D. Western blotting revealed the changes in MOR and ZEB1 protein levels after ZEB1-AS1 knockdown in Cal27 R cells. E. The western blot shows ZEB1 and MOR expression in Cal27 R cells after ZEB1-AS1 knockdown or ZEB1 co-transfection. F. Fluorescence signals in the lung metastasis model are shown in the images. G. Representative image of hematoxylin and eosin (HE) staining from nude mice with lung metastasis model. H. Representative immunohistochemical images showing the staining signals of ZEB1 and MOR; *p < 0.05; ** p < 0.01; ***p < 0.005; ****p < 0.001. Construction of nuclear-targeted nanomaterials for tumor therapy Nucleic acid-based therapies offered a significant advantage in targeting disease-causing genes, particularly non-coding RNA molecules where conventional drugs may fail [29]. However, siRNA and ASO both faced challenges such as nuclease degradation, limited tumor penetration, and off-target effects, necessitating an ideal delivery system [30]. In our study, we utilized a nanoparticle (NPs) platform, previously used for delivering plasmids, siRNA, and chemotherapeutic drugs, due to its excellent physicochemical properties, biocompatibility, and cyclic stability. Given that ZEB1-AS1 localized in the cytoplasm and nucleus, we employed a nuclear-targeting nanomaterial, NT-NPs-Silencer, developed by our collaborators, to silence ZEB1-AS1 and inhibit tumor growth [29, 31, 32]. NT-NPs-Silencer comprises an endosomal pH-responsive polymer, methoxyl-poly (ethylene glycol)-b-poly(2-diisopropyamino)-ethylmethacrylate) (Meo-PEG-b-PDPA), and an amphiphilic nucleus-targeting peptide, which binds with α/β heterodimer (Fig. 4A). This NPs carrier increases the in vivo stability of siRNA, evades lysosomal capture, and facilitates nuclear release. The average diameter of NT-NPs-Silencer particles is 157 nm (Fig. 4B), and they have a zeta potential of 8.9 mV (Fig. 4C). Notably, NT-NPs-Silencer demonstrated higher Silencer release at pH 6.0 compared to pH 7.0 (Fig. 4D), efficiently delivered Silencer to the cell nucleus (Fig. 4E) and escaped from lysosomal capture (Fig. 4F). Moreover, NT-NPs-Silencer exhibited superior ZEB1-AS1 silencing efficiency compared to NPs-Silencer and Silencer (Fig. 4G), leading to weakened cell migration and invasion ability (Fig. 4H and I). Western blot analysis confirmed reduced expression of ZEB1 and MOR, along with changes in EMT markers in the NT-NPs-Silencer group (Fig. 4J). Pharmacokinetic analysis demonstrated prolonged circulation of NT-NPs-Silencer and NPs-Silencer in the bloodstream due to nuclease protection (Fig. 4K). In the subcutaneous tumor model of nude mice, BioD experiments revealed enhanced signal intensity at the tumor site for NT-NPs-Silencer and NPs-Silencer compared to Silencer. Tumor mass analysis post-euthanization indicated stronger signals in the NT-NPs-Silencer and NPs-Silencer groups (Fig. 4L). A. Illustration of a NT-NPs-Silencer for precise ZEB1-AS1 regulation in cancer therapy. B. Size distribution of NPs in pH 7.4 PBS solution. C. Zeta potential distribution of NPs in pH 7.4 PBS solution. D. Cumulative release of Silencer from NPs in pH 7.4 and pH 6.0 PBS solutions. E. Images displaying Cal27 cells incubated with NPs-Silencer and NT-NPs-Silencer, featuring red fluorescence (Cy5-Silencer) and blue fluorescence (cell nuclei). F. Swift NPs disassembly, effective endosomal escape, and prompt cytosolic Silencer release upon cellular uptake. G. Quantitative data demonstrating ZEB1-AS1 levels after incubation with Silencer or nanoparticles in Cal27 cells. H. Representative transwell assay images of Cal27 cells following incubation with PBS, NC, NAKED, Silencer, NPs-Silencer, or NT-NPs-Silencer. I. Quantification of Cal27 cells migration and invasion after ZEB1-AS1 regulation with Silencer or nanoparticles. J. Western blot analysis revealing protein level changes post-ZEB1-AS1 knockdown in Cal27 cells. K. Comparison of blood circulation profiles for naked Silencer, NPs-Silencer, and NT-NPs-Silencer. L. Fluorescence image capturing subcutaneous Cal27 tumor-bearing mice 24 h after injecting naked Silencer, NPs-Silencer, and NT-NPs-Silencer; *p < 0.05; ** p < 0.01; ***p < 0.005. Nude mouse models with lung metastasis were employed to assess the therapeutic potential of NT-NPs-Silencer. Over a span of 6 consecutive injections, each nude mouse received a 1 nmol dose of Silencer every 2 days. The tumor inhibition achieved by NT-NPs-Silencer was significantly superior compared to the other treatment groups (sFig. 6A). Histopathological examinations through Hematoxylin and Eosin (HE) staining (sFig. 6B) and Immunohistochemistry (IHC) analyses (sFig. 6C) further confirmed that the tumor foci in the NT-NPs-Silencer group were notably smaller than those observed in the other groups. Moreover, the expression levels of ZEB1 and MOR were substantially reduced in the NT-NPs-Silencer group, more so than in the other groups. The above research findings indicated that employing nucleus targeting nanomaterials as carriers for delivering Silencer might represent an effective avenue for treating HNSC. ZEB1-AS1 augmented cisplatin resistance in vivo To explore the impact of ZEB1-AS1 on the regulation of cisplatin resistance in Cal27 cells in an in vivo setting, we introduced Cal27 R cells into the dorsal region of nude mice. Subsequently, mice were subjected to either cisplatin treatment or a control regimen once tumors became palpable. Following this, we administered Silencer, NPs-Silencer, NT-NPs-Silencer, or a control solution into the tumors every 2 days. In line with our in vitro findings, NT-NPs-Silencer exhibited a significant reduction in tumor volumes compared to the other groups. Additionally, treatment with cisplatin further contributed to tumor size reduction, signifying that NT-NPs-Silencer effectively impedes tumor growth and reinstates sensitivity to cisplatin in cisplatin-resistant tumors (Fig. 5A and B). Immunohistochemistry (IHC) staining results indicated a substantial decrease in the expression of ZEB1 and MOR in tumors treated with NT-NPs-Silencer. Furthermore, cisplatin treatment led to a further reduction in the levels of these proteins (Fig. 5C–E). These findings collectively demonstrate that antagonizing ZEB1-AS1 significantly curbs cancer cell proliferation, suppresses ZEB1 protein expression, downregulates the transcription of its downstream target MOR protein, and ultimately reverses cisplatin resistance in an in vivo context. In conclusion, employing NT-NPs to silence ZEB1-AS1 presents a promising strategy for the treatment of HNSC, as demonstrated by the results in this study. ZEB1-AS1 facilitated cisplatin resistance in vivo A. Tumor appearance picture on day 49. B. The comparative tumor growth curves across the various groups. C. ZEB1 and MOR were stained using IHC. D. The quantification of IHC scores for ZEB1. E. The quantification of IHC scores for MOR; *p < 0.05; ** p < 0.01; ***p < 0.005. Illustration of a nucleus-targeted Silencer nanoplatform for precise regulation of nuclear ZEB1-AS1 function in effective cancer therapy. The schematic model illustrated that ZEB1-AS1 is involved in EMT and cisplatin chemotherapeutic resistance of Cal27 cells by regulating ZEB1 and ZEB1 downstream target molecule MOR (Fig. 6). Our nuclear target nanoparticle platform, utilized for delivering ZEB1-AS1 Silencer, resulted in a reduction of tumor metastasis and the reversal of chemoresistance. Schematic diagram of the delivery and function of the NT-NPs-Silencer Construction of ZEB-AS1/ZEB1/MOR risk model in HNSC and identification of high risk model characteristics Previous researches have shown that ZEB1 and MOR were upregulated in various tumors and were associated with poor prognosis in cancer patients [33–36]. To more accurately assess the risk for each tumor patient, a risk model was constructed using three molecules: ZEB1-AS1, ZEB1, and MOR, with an Area Under the Curve (AUC) greater than 0.6. The results revealed that patients in the low-risk group exhibited a better prognosis, whereas those in the high-risk group had a worse prognosis (sFig. 7A). Then we computed individual scores for each patient based on tumor grade, TNM stage, and the expression levels of model genes. Subsequently, we used these scores to determine each patient’s survival probability throughout the entire follow-up period. These factors were derived from trabecular analysis, illustrated in Supplementary sFig. 8. We performed a screening incorporating tumor grade, stage, and the three prognostic risk genes: ZEB1-AS1, ZEB1, and MOR, to ascertain patient survival rates. To evaluate the relationship between the risk prediction model mentioned earlier and tumor immunity, we employed CIBERSORT and ESTIMATE algorithms to explore the composition of immune cells, enrichment of immune-related molecules, and other relevant factors. In the high-risk group, there was a high level of T cells CD8, B cells memory, and T cells follicular helper. Conversely, the low-risk group exhibited high enrichment of macrophages M0, T cells CD4 memory resting, and Neutrophils (sFig. 7B). Furthermore, the high-risk group displayed significantly lower scores in stromal score, immune score, and estimate score compared to the low-risk group (sFig. 7C). Additionally, the high-risk group exhibited significantly higher tumor purity than the low-risk group (sFig. 7D). A heatmap analysis revealed differential expression of immune checkpoint genes (ICGs), including CD274, HLA-G, and TNFRSF18 between the high-risk group and low-risk group (sFig. 7E). Drug sensitivity, hallmarks, GO, and KEGG in the high-risk group and low-risk group. To explore potential gene ontology, we also conducted Gene Set Enrichment Analysis (GSEA). Regarding hallmarks, we found that patients in the high-risk group were enriched in DNA repair (sFig. 9A), E2F target (sFig. 9B), myc target v1 (sFig. 9C), p53 pathway (sFig. 9D). Chemotherapy and immunotherapy remain essential approaches in cancer treatment, and evaluating patient’s sensitivity to these therapies is crucial for improving survival rates. To assist in selecting appropriate chemotherapy agents, we utilized the GDSC database to predict the sensitivity of both patient groups to common chemotherapy drugs. The results indicated that the IC50 values of certain drugs, such as lbrutinib, sapitinib, ulixertinib, erlotinib, alisertib, afatinib, Osimertinib, PD0325901, and trametinib, were higher in the high-risk group than the low-risk group, suggesting that patients in the high-risk group have poorer drug sensitivity (sFig. 9E). To investigate the biological processes, signaling pathways, and key gene changes associated with high- and low-risk group hallmarks, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were conducted using Gene Set Variation Analysis (GSVA). The low-risk group showed enrichment in DNA repair, MYC targets v1, and other hallmarks (sFig. 9F). As shown in sFig. 9G, mitochondrial cytochrome c oxidase assembly and regulation of mitochondrial transcription-related biological processes exhibited lower risk scores. Regarding signaling pathways, the low-risk group demonstrated significant enrichment in base excision repair, DNA replication, and other pathways (sFig. 9H). Discussion This study illustrated a nuclear-targeted nanoplatform and delivered ZEB1-AS1 Silencer in HNSC. By modulating the expression of ZEB1-AS1, the tumor growth was inhibited and the cisplatin resistance was reversed. Since the ZEB1-AS1/ZEB1/MOR axis was found in this process, a risk model for HNSC patients was constructed using ZEB1-AS1, ZEB1, and MOR. ZEB1-AS1 has been demonstrated to promote the proliferation and metastasis of various tumors, including OSCC [37]. Our study demonstrated that ZEB1-AS1 was a critical LncRNA regulating ZEB1 promoter activity, which, in turn, controlled MOR protein transcription by directly binding to the MOR promoter region. ZEB1-AS1 played a pivotal role in tumor initiation and progression by affecting the expression of other proteins through its interaction with ZEB1. Previous studies proved that ZEB1-AS1 disrupted the circuit mechanism, preventing miR-151-3p from inhibiting ZEB1 expression in M0 (acute myeloid leukemia morphological subtypes) [38]. In triple-negative breast cancer, ZEB1 maintained ZEB1-AS1 overexpression and stabilized ZEB1 by directly targeting ELAVL1, forming a circuit mechanism that promotes cancer development [33]. These findings underscored the significance of ZEB1-AS1 in tumorigenesis and its role in modulating critical regulatory pathways. HNSC was characterized by early metastasis and a 30% recurrence rate post-treatment. Previous research identified ZEB1-AS1’s role in paclitaxel and cisplatin sensitivity in ovarian cancer [23]. Our study demonstrated that silencing ZEB1-AS1 inhibits Cal27 cell metastasis and enhances cisplatin sensitivity, while ZEB1-AS1 upregulation showed the opposite effect. Additionally, ZEB1 and MOR, regulated by ZEB1-AS1, also contributed to tumor metastasis and reduced cisplatin sensitivity. Clinical evaluation showed significantly higher ZEB1-AS1 expression in cisplatin-resistant patients, correlating with larger tumor diameter, lymph node metastasis, and poor prognosis. Targeting ZEB1-AS1, ZEB1, and MOR could hold promise for improving outcomes, especially in cisplatin-resistant HNSC patients. Here we introduced a nuclear targeting nanostructure made with PDPA, NTPA peptide and ZEB1-AS1 Silencer. This study was the first to report NTPA binding with LncRNA Silencer to form NTPA/Silencer complex and then assembled with PDPA polymers. Previously, NTPA was proved to bind with siRNA before nanostructure assembly [39, 40]. Since the ZEB1-AS1 Silencer was composed of ASO and siRNA, we introduced NTPA here and the nanostructures showed promising nuclear targeting. Except for nuclear targeting, trapping by endosome was also a drawback for nanostructures [41]. Recently, nanostructures made with PDPA was widely used in the organelle targeting for the satisfying endosomal escape ability [42, 43]. Nuclear targeting nanostructures encapsulating siNEAT2 made with PDPA showed promising endosomal escape [44]. Nanostructures made with PDPA embedded mitochondria-targeting and membrane-penetrating peptide amphiphile (MMPA) and siRNA in the core and successfully delivered siRNA into the mitochondria without trapping by endosome [45]. In this study, the nanostructure showed better release in the low pH solution compared to the high pH solution. The endosomal escape and nuclear targeting were both promising here. Restoring immune cell vitality and increasing their numbers in the body represent a promising approach for cancer treatment [46]. Here we used LASSO regression to analyze TCGA data and established a risk score model using ZEB1-AS1, ZEB1, and MOR genes. The high-risk group showed poor survival prognosis and was associated with specific immune cell types and immune checkpoint gene upregulation, indicating an immunosuppressive tumor state [47, 48]. Previous studies have reported a nomogram prediction model based on ZEB1-AS1 and its associated NUDT3 in colorectal cancer, which shows high predictive value for immunotherapy response [49]. Combining chemotherapy with monoclonal antibody immunotherapy improved the survival rate of HNSC patients, making it significant to explore correlations between risk groups, immune checkpoints, drug sensitivity, and personalized medication [50]. Immune checkpoints such as PD1, PDL1, and CDLA-4 are key targets for regulating immune responses and have shown potential for tumor treatment [47]. In this study, GSEA analysis indicated that the high-risk score group was associated with pathways involving oxidative phosphorylation, reactive oxygen species, DNA repair, MYC-targets, p53 signaling pathway, and fatty acid metabolism. TP53 (Tumor protein p53) was frequently mutated in HNSC and had various biological functions, including oxidative phosphorylation, DNA repair, and cell cycle arrest [51]. Accordingly, the high-risk group in this study was enriched in the p53 signaling pathway. Additionally, the MYC-target pathway was associated with lower overall survival in nasopharyngeal carcinoma patients [52]. Further exploration through GSVA revealed differences in hallmark, GO, and KEGG pathways between high- and low-risk groups, including cell cycle processes and energy metabolism. Low-risk group showed enrichment in DNA repair pathways [53, 54]. This study had several limitations. First, we could not validate our models predictive efficacy for long-term prognosis due to the lack of our own clinical specimens for immunohistochemical experiments. However, we did confirm ZEB1-AS1, ZEB1 and MOR’s molecular functions in Cal27 cells, aligning with previous findings. Secondly, relying on database data for predicting the tumor immune microenvironment without clinical specimen validation might limit conclusion reliability. Nonetheless, other researchers’ findings support our results. Lastly, we did not comprehensively verify nanomaterials physical and chemical properties, biocompatibility, and toxicity, but they were validated by our collaborators. Future focus will be on investigating their effects on tumor inhibition as RNA carriers. Conclusion In conclusion, our study confirmed the significance of ZEB1-AS1 as a potential diagnostic and therapeutic target in HNSC. Nuclear-targeted nanostructures delivering ZEB1-AS1 Silencer inhibited tumor growth and reversed cisplatin resistance both in vitro and in vivo. The prediction models involving ZEB1-AS1, ZEB1 and MOR proteins, may hold promise for evaluating immunotherapy response and the tumor microenvironment in HNSC patients. Supplementary Information Additional file1 Acknowledgements We would like to thank Professor Minghui Cao from the Department of Anesthesia of Sun Yat-sen Memorial Hospital for continuous support on this study. Author contributions H.Y. and Y.Z. finished the experiments except for Nano related experiments and prepared the figures. Z.L. and Q.L. analyzed the data and wrote the original manuscript. Y.Y. analyzed the data and revised the figures. Z.T. and P.S. finished Nano related experiments and revised the original manuscript. N.L. and F.J. prepared the materials for the experiments. All authors reviewed the final manuscript. Funding This work was supported by the Guangzhou Science and Technology Bureau [202002020002]. Data availability The datasets generated during the current study are not publicly available due to the institutional privacy but are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate The animals used in this study had been reviewed and approved by the Institutional Animal Care and Use Committee (IACUC), Sun Yat-Sen University (AP20220174) and the aniaml experiments were performed in accordance with guidelines of the Sun Yat-sen University. All experimental protocols were approved by the review board of Sun Yat-sen Memorial Hospital and all experiments were performed in accordance with guidelines of the Sun Yat-sen Memorial Hospital. Written informed consent was obtained from the individuals for the publication of any potentially identifiable images or data included in this manuscript. Competing interests The authors declare no competing interests. Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Haojie Yang, Yangfan Zhang, Zicong Tan and Zihao Liu have authors contributed equally to this work and share first authorship. Contributor Information Phei Er Saw, Email: caipeie@mail.sysu.edu.cn. Ning Liufu, Email: liufn3@mail.sysu.edu.cn. Fengtao Ji, Email: jift@mail.sysu.edu.cn. References Associated Data Supplementary Materials Additional file1 Data Availability Statement The datasets generated during the current study are not publicly available due to the institutional privacy but are available from the corresponding author on reasonable request.