NPLOC4 is a potential target and a poor prognostic signature in lung squamous cell carcinoma

Few prognostic biomarkers exist for lung squamous cell carcinoma (LUSC), which has a poor five-year survival rate. Using bioinformatics, this study evaluated NPLOC4 as a prognostic marker for patients with lung squamous cell carcinoma. Shorter survival periods and tumor growth were linked to high NPLOC4 expression.Disulfiram (DSF) combined with copper (Cu) targets NPLOC4 to achieve antitumor effects in lung squamous cell carcinoma. Thus, we investigated the effects of DSF with Cu in LUSC. Gene-set enrichment analysis identified ubiquitin-mediated proteolysis as the NPLOC4-associated mechanism influencing LUSC prognosis. In SK-MES-1 cell lines, DSF + Cu increased K48-linked ubiquitinated protein expression and apoptosis. This study identified NPLOC4 as a prognostic biomarker and a potential therapeutic target for LUSC.


Immunohistochemistry
NPLOC4 antibody (ab 224,435; Abcam) was used for immunohistochemical evaluation of NPLOC4.Immunohistochemistry kits were purchased from ZSGB-BIO (China, no: PV-91000).Tissue sections were dewaxed, antigen-repaired for 2 min 15 s at 120 °C, and blocked with peroxidase blockers.After overnight incubation with NPLOC4 antibody (1:45) at 4 °C, reaction enhancement solution was added to the sections.Secondary antibody

Protein-protein interaction network construction
We constructed a network of proteins associated with NPLOC4 using the STRING database (https:// cn.string-db.org/).Nodes with a confidence of interactive relationships larger than 0.4 were used for building the network.Visualization of protein networks using cytoscape software.

Gene enrichment analysis
Based on the median NPLOC4 expression, tumor samples were divided into high-expression and low-expression groups.Whole gene expression data from two groups of patients were substituted into GSEA v4.1.0,and the number of iterations were set to 1000.A KEGG dataset was selected to explore the possible mechanism of the effect of NPLOC4 on pulmonary squamous carcinoma prognosis.Differentially enriched pathways were scored, with higher scores for significant enrichment in the higher-risk group, and lower scores for significant enrichment in the lower-risk group.Enriched pathways were filtered based on P-values (< 0.05) and FDR values (< 0.05).

Correlation of NPLOC4 expression with immune cell infiltration and immune checkpoints
To analyze the relationship between tumor-infiltrating immune cells (TIC) and NPLOC4 expression, the distribution of TIC abundance in tumor samples was estimated using CIBERSORT to select tumor samples with P < 0.05.Correlation between NPLOC4 expression and immune checkpoints was evaluated using the TIMER database (https:// cistr ome.shiny apps.io/ timer/).

Cell viability assay
Cck8 kits were purchased from NCM Biotech (Soochow, China) to measure cell viability.SK-MES-1 cells were seeded onto 96-well plates (12,000/well) and, following cell adherence, treated with DSF, Cu, or a DSF + Cu combination.After 24 h, CCK-8 reagent was added to the wells and cell viability assessed according to the manufacturer's instructions.Cell survival was 50% at 0.25 μM DSF and 0.2 μM Cu, and these concentrations were used for subsequent experiments.

Statistical analysis
Differential expression of NPLOC4 in cancerous and paraneoplastic tissues in the same sample was analyzed using the Wilcoxon nonparametric test and paired t-test.Survival analysis was performed using the Kaplan-Meier method with the log-rank test.p < 0.05 was considered significant.Univariate and multivariate Cox regression analyses were used to assess the effects of clinical variables on patient outcomes.All experimental data are presented as the mean ± standard deviation of at least three independent experiments.Statistical analyses were performed using R software v3.6.3 and GraphPad Prism 7.0.

Ethics approval
The study was approved by the ethics committee of Shanghai Outdo Biotech Company (ID:YB M-05-02).

Differential expression of NPLOC4 and association with clinicopathological variables
Pan-cancer analysis indicated that NPLOC4 was highly expressed in most types of cancer, including clear cell renal cell carcinoma, bladder urothelial carcinoma, and papillary renal cell carcinoma (Fig. 1a).NPLOC4 expression was significantly higher in LUSC samples than in normal lung tissues (p < 0.001) (Fig. 1b).NPLOC4 was highly expressed in 49 samples of lung squamous carcinoma tissue (p < 0.001) (Fig. 1c).The ROC curve showed that NPLOC4 expression could discriminate between LUSC tissues and normal lung tissues with an area under the curve (AUC) of 0.930 (95% confidence interval [CI] = 0.908-0.952)(Fig. 1d).Kaplan -Meier survival curves showed that patients with high NPLOC4 expression had shorter survival times, while those with low expression had longer survival times (Fig. 1e).Validation of the total survival model using ROC curves (Fig. 1f) showed AUCs of 0.544, 0.608, and 0.623 for 1, 5, and 10 years, respectively.To determine the relationship between NPLOC4 expression and clinicopathological characteristics (Fig. 1g,h and Supplementary Fig. 1), we analyzed the corresponding clinical information of LUSC cases from the TCGA database.We found that patients with TNM stages III-IV, as well as patients with lymph node metastasis, had higher NPLOC4 expression.The elevated NPLOC4 expression may be associated with LUSC migration and invasion.

NPLOC4 expression in TMA and clinical sample validation using the TMA database
To further validate NPLOC4 expression, immunohistochemical staining was performed on 90 cancer and paracancer tissue samples.Data was obtained from 83 patients.Cases 1-4 are representative of NPLOC4 immunohistochemical staining.Cases 1 (Fig. 2a) and 2 (Fig. 2b) exhibited high NPLOC4 expression in cancer tissues.Cases 3 (Fig. 2c) and 4 (Fig. 2d) exhibited low NPLOC4 expression in cancer tissues.Evaluation of NPLOC4 expression in conjunction with clinicopathological data from 83 patients revealed higher NPLOC4 expression in tumor tissues, both in the analysis of cancerous and paraneoplastic tissues (Fig. 2e) and in the paired analysis of cancerous and paraneoplastic tissues (Fig. 2f).The ROC curve showed that NPLOC4 expression could discriminate between LUSC and normal lung tissues with an area under the curve (AUC) of 0.995 (95% confidence interval [CI] = 0.998-1) (Fig. 2g).Kaplan-Meier survival analysis revealed that patients with high NPLOC4 expression had shorter survival times than those with low expression (Fig. 2h).Validation of the total survival model using ROC curves (Fig. 2i) showed AUCs of 0.806, 0.717, and 0.727 for 1, 3, and 5 years, respectively.NPLOC4 expression was analyzed according to age, sex, AJCC stage, and TNM stage of the patients (Supplementary Fig. 2).Validation in the organization chip was substantially consistent with analysis results in TCGA database.

Cox regression analyses of the TCGA database and TMA cohort
In TCGA cohort (Table 1), both univariate and multifactorial Cox regression analyses models, NPLOC4 expression and age were significantly associated with overall survival (OS).Most importantly, patients with LUSC have a shorter OS with an increase in NPLOC4 expression.Then verified with the TMA database (Table 2).In univariate Cox regression analysis models, age, TNM stage, AJCC stage, and NPLOC4 expression were significantly associated with OS (p < 0.05).In the multivariate Cox regression analysis, NPLOC4 expression was associated with OS (p < 0.05).Both TCGA and the TMA databases revealed a correlation between NPLOC4 expression and OS.NPLOC4 may be an independent prognostic factor for lung cancer.

Protein interaction networks and predicted signaling pathways
Next, we used TCGA database to identify NPLOC4 pathway mechanisms that may have an impact on pulmonary squamous carcinoma prognosis.The results in TCGA dataset suggested that gene expression in the high-risk group was significantly enriched in ubiquitin-mediated proteolysis (Fig. 3a) and progesterone-mediated oocyte maturation (Fig. 3b).We used the STRING database to analyze the network of protein interactions associated with NPLOC4.Our screening criteria included a maximum of 50 interactors and a minimum interaction score of 0.4.The protein interaction network of NPLOC4 is shown in Fig. 3c.And quantitative analysis of the interaction network using cytoscape found that NPLOC4 is closely related ubiquitinated protein (Fig. 3d).Furthermore, Vol:.( 1234567890

Prediction of drug targets and pathway validation
In our previous research, DSF + Cu was found to inhibit ubiquitin-mediated proteolysis by acting on NPLOC4, resulting in gastric cancer cell apoptosis 12 .However, this has yet to be demonstrated in lung cancer.First, we used Cu (Fig. 4a) and DSF (Fig. 4b) to verify the cytotoxic effect of the two drugs on SK-MES-1 cells.The CCK8 results showed decreased cell survival as the concentration of the two drugs increased, with 50% cell survival at 0.2 μM Cu and 0.25 μ M DSF.DSF combined with Cu had a higher cell inhibition rate than the single drugs in SK-MES-1 cells (Fig. 4c).Cell viabilities were 97% for Cu, 95% for DSF, and 53% for DSF + Cu (Fig. 4d).Apoptosis was detected using TUNEL kits, and we found that DSF + Cu-treated cells had the highest apoptosis rate among the four groups (Fig. 4e).When Cu was combined with DSF, NPLOC4 expression was inhibited compared with that in the control group (Fig. 4g).Compared with the control group, the DSF + Cu group exhibited increased K48-ubiquitinated protein expression (Fig. 4h).Furthermore, we found that NPLOC4 expression is higher in tumor cells than in normal cells (Fig. 4f).

Correlation between NPLOC4 and tumor immunity
With the rise of immunotherapy, its role is extremely important in tumor treatment.At the end of the study, we analyzed the correlation between NPLOC4 expression and immune cells and immune checkpoints.We analyzed the relative proportions of 22 types of immune cells in LUSC samples from TCGA database (Fig. 5a) and constructed an immune cell proportion correlation heat map (Fig. 5b).Next, we evaluated the differences in immune cell proportions associated with high and low NPLOC4 expression (Fig. 5c) and the correlation between NPLOC4 expression and immune cells (Fig. 5d).We identified two types of cells associated with differential NPLOC4 expression, namely naive B cells and neutrophils (Fig. 5e).Using the TIMER database, we analyzed correlations between NPLOC4 expression and immune checkpoints CTLA-4, PD-L1, and PD-1 (Fig. 5f).Among these, PD-L1 expression exhibited the greatest correlation with NPLOC4 expression.NPLOC4 expression affects the immune microenvironment of LUSC.High NPLOC4 expression in LUSC also increased the proportion of naive B cells and decreased the neutrophil proportion.However, the positive correlation between PD-L1 and NPLOC4 expression in the analysis of the database might suggest that patients benefit more from immunotherapy.

Discussion
In this study, we identified NPLOC4 as a prognostic biomarker for LUSC, with elevated NPLOC expression associated with shorter survival and poor prognosis.In addition, we found a positive correlation between NPLOC4 and PD-L1 expression.Finally, in LUSC, DSF combined with Cu might induce tumor cell apoptosis by inhibiting NPLOC4, which increases ubiquitinated protein accumulation.NPLOC4 is an important cofactor of p97 and, together with ubiquitin fusion degradation 1 (UFD1), is involved in many p97-dependent processes 15,16 such as ER-associated protein degradation.p97 is considered a potential cancer biomarker, associated with poor prognosis, increased likelihood of metastasis, or shorter OS in pancreatic cancer 17 , liver cancer 18 , breast cancer 19 , prostate cancer 20 , and lung cancer 21 .In pan-cancer analysis, the expression of NPLOC4 in lung adenocarcinoma differed significantly between cancer and paracancerous, as shown in Fig. 1a.We analyzed the effect of the expression of NPLOC4 in lung adenocarcinoma on the survival of patients, and found that there was no significant different survival between high and low groups (Supplementary Fig. 4).In lung adenocarcinoma, we considered that the presence of some gene hot-spot mutations affected the survival of patients with the expression of NPLOC4.In order to ensure the homogeneity of the study cohort, we selected lung squamous cell carcinoma for study.We found that NPLOC4 expression is associated with shorter  www.nature.com/scientificreports/OS, later AJCC staging, lymph node metastasis, and poor prognosis in LUSC.However, during the study, our research group found that in pancreatic adenocarcinoma, the group with high NPLOC4 expression had longer survival times (Supplementary Fig. 3), suggesting that the effect of NPLOC4 is specific to the tumor type.
In 1977, Lewison published the first clinical case report on DSF as an anticancer therapy.Skrott et al. 10 analyzed medical data from 240,000 patients with cancer in Denmark and reported that patients with alcoholism treated with disulfiram consistently over a long period had a 34% reduction in mortality compared with those who did not receive disulfiram.Studies have shown that disulfiram exerts its antitumor effects through interaction with Cu 14,[22][23][24][25] , and that DSF + Cu exerts antitumor effects by targeting ALDH-positive tumor stem cells 26,27 .However, further investigation indicated that DSF + Cu does not target ALDH 13 , but binds NPLOC4 and induce its aggregation, disabling the p97-NPLOC4-UFD1 pathway and increasing the accumulation of ubiquitinated proteins, leading to cell death 10,12,28,29 .This mechanism has been studied in several tumor types, including gastric cancer 12 , medulloblastoma 30 , and clear cell renal cell carcinoma 14 .Although the effects of DSF + Cu has been studied in non-small cell lung cancer 31,32 , no studies have been conducted to investigate NPLOC4 inhibition or the effect of DSF + Cu on LUSC.In this study, we found that DSF + Cu induces apoptosis in lung squamous carcinoma cells possibly by inhibiting NPLOC4, thereby increasing accumulation of ubiquitinated proteins.We also found a positive correlation between NPLOC4 and PD-L1 expression.DSF + Cu has been shown to upregulate PD-L1 expression, and the combination of DSF + Cu and anti-PD-1 antibody showed better antitumor efficacy than monotherapy in hepatocellular carcinoma 33 and triple-negative breast cancer 34 .The combination regimen is therefore promising for treatment of patients with tumors, but further evidence of the efficacy is required.
Clinical trials of DSF have been conducted for multiple tumor types (https:// clini caltr ials.gov/).A phase IIb clinical trial of non-small cell lung cancer (ClinicalTrials.govNCT00312819) found increased survival in the combined DSF + chemotherapy group compared with the chemotherapy monotherapy group (10 months vs 7.1 months) 35 .
We believe that screening for high NPLOC4 expression levels may provide a greater survival benefit.Future studies to validate the antitumor effect of DSF in lung squamous carcinoma in animal experiments, as well as a larger clinical sample size to evaluate the effect of NPLOC4 expression on the prognosis of patients with LUSC, is required.This study identified a novel prognostic biomarker for patients with LUSC, as well as a new focus for targeted therapy.Furthermore, as the mutation rate of PD-L1 in lung squamous carcinoma is higher than that in lung adenocarcinoma 36 , this study provides a new proposal for combined immune-targeted therapy for lung squamous carcinoma.

Figure 1 .Figure 2 .
Figure 1.(a) NPLOC4 expression in different types of tumors compared with normal tissues in TCGA database.NPLOC4 expression in (b) LUSC tissues and normal tissues in TCGA database, and (c) matched normal tissues in TCGA database.(d) ROC curves for classifying breast cancer versus normal breast tissue in TCGA database.(e) Survival curves show that high NPLOC4 expression in patients with LUSC is strongly associated with shorter overall survival (OS).(f) In TCGA dataset,time-dependent ROC curves predict patient survival at 1, 3, and 5 years according to NPLOC4 expression.Associations between NPLOC4 expression and clinicopathological characteristics.Data are shown for (g) clinical stage, (h) N stage.

Figure 3 .
Figure 3. KEGG enrichment analysis of LUSC.(a) Ubiquitin-mediated proteolysis.(b) Progesterone-mediated oocyte maturation.(c) The PPI network of NPLOC4.A visual network of NPLOC4 and its co-expression genes.(d) Visual Analysis of STRING Protein Interaction Network.

Table 1 .
Univariate and Multifactor Cox regression analysis of OS in TCGA cohort.OS overall survival, TCGA : The Cancer Genome Atlas.

Table 2 .
Univariate and multifactor cox regression analysis of OS in TMA cohort.OS overall survival, TMA:Tissue microarrays were engineered by Shanghai Outdo Biotech Company(Shanghai, China).