Upregulation of NDUFAF2 in Lung Adenocarcinoma Is a Novel Independent Prognostic Biomarker

Background NADH dehydrogenase (ubiquinone) 1 alpha subcomplex assembly factor 2 (NDUFAF2) acts as a molecular chaperone for the assembly of complex I on the mitochondrial membrane, which is involved in the transfer of electrons in the respiratory chain. However, whether NDUFAF2 plays a role in lung adenocarcinoma (LUAD) is largely unexplored. Methods Expression profiles were obtained from the TCGA and GEO databases and integrated via R3.6.3 and several bioinformatics platforms. Western blotting analysis and immunohistochemistry staining were used to examine the expressions of NDUFAF2 in clinical samples. Moreover, the diagnostic and prognostic value of NDUFAF2 expression level was also assessed. GO, KEGG, and gene set enrichment analysis (GSEA) were adopted to investigate NDUFAF2-related molecular functions, signaling pathways, and life activity processes. Results NDUFAF2 was predominantly expressed in LUAD, and it is identified as a promising biomarker in the diagnosis of LUAD and its prognostic prediction. Overexpression of NDUFAF2 was correlated with N stage, T stage, and pathologic stage in LUAD, leading to worse overall survival (OS). Besides, the level of NDUFAF2 was independently associated with OS through a multivariate Cox analysis (HR = 1.538, 95% (1.086-2.177), P = 0.015). GO analysis revealed enrichment in innate immune response in mucosa and mucosal immune response, and GSEA indicated enrichment in G2_M_checkpoints, DNA replication, diseases of mitotic cell cycle, retinoblastoma gene in cancer, cell cycle pathway, and cell cycle. Furthermore, the expression level of NDUFAF2 was negatively correlated with infiltration levels of Tem, Tcm, NK CD56bright cells, and B cells. In contrast, the expression level of NDUFAF2 was positively correlated with the infiltration level of DCs and Th2 cells in LUAD patients. Conclusions Collectively, NDUFAF2 is a promising independent prognostic biomarker and target in LUAD. In addition, NDUFAF2 might affect the prognosis of LUAD via DNA replication, diseases of mitotic cell cycle, cell cycle pathway, and cell cycle.


Introduction
Lung adenocarcinoma (LUAD) is a heavy burden on public health worldwide. According to the report of cancer incidence and mortality by GLOBOCAN 2020, LUAD ranks second in all commonly diagnosed cancers (11.4%), and it causes 18% of the cancer-related death in 2020. The 5-year survival of LUAD is approximately 20% in recent years [1], and the upto-date histopathologic classification shows that LUAD comprises most of all pulmonary malignancies, accounting for approximately 40% of lung malignancies [2]. Notable changes in lung cancer epidemiology, therapeutic intervention, and prevention have occurred in the last few decades owing to the improvement of targeted therapies and immunotherapy. More novel methods are currently being investigated, such as immune checkpoint blockade, oncolytic viruses, stem cellbased medicinal products, cytokine derivatives, CD3bispecific antibodies, vaccine platforms, and adoptive tumor cell therapy [3]. With further research in the field of oncology, the tumor microenvironment (TME) is now generally appreciated to play an indispensable role in tumorigenesis and malignant progression [4]. Until now, the relative prognosis of LUAD remains unsatisfactory. It is imperative that we continue to identify new diagnostic methods and targets.
At present, it has become a common method to search for lung adenocarcinoma prognostic markers through bioin-formatics. By combining high-throughput information of omics with clinical information of patients, effective prognostic markers can be screened, and then, markers can be further analyzed and studied [5][6][7]. Finally, this makes it possible for them to become reliable indicators [8]. Through the analysis of multiomics outcome (Supplementary Materials) from LUAD samples [7], we reported that the level of NDUFAF2 in the tumor tissues is higher than that in adjacent nontumor tissues. Therefore, we speculated that NDUFAF2 could be a promising prognostic indicator in the diagnosis and treatment of LUAD.
In the traditional view, NDUFAF2 is a complex I assembly factor in mitochondria. Functional loss of genes encoding complex I subunits greatly contributes to abnormal function of the respiratory chain, leading to various clinical phenotypes from early onset lethal symptoms to    Computational and Mathematical Methods in Medicine adult-onset exercise intolerance, such as Leigh's syndrome [9]. The homozygous substitution in NDUFAF2 is definitely the pathogenic mutation, which results in the lack of complex I in the fibroblasts of the patient, and the expression and functional activity of complex I can almost entirely be rescued by complementation of the patient's fibroblasts with the baculovirus [10]. Furthermore, depletion of NDUFAF2 in human SK-N-MC neuroblastoma cells can inhibit cell proliferation [9]. Besides, NDUFAF2 is also known as "mimitin" identified in the screening of Myc-regulated genes. It is upregulated by c-Myc in some cell lines, such as glioblastoma T98G, ESCC TE-11, and human embryonic lung (HEL) cells, and specific inhibition of mimitin reduces cell proliferation [11].
Further analysis was carried out with regard to NDUFAF2 in this study. We explored the predictive value of level of NDUFAF2 in LUAD based on clinical information obtained from The Cancer Genome Atlas (TCGA). In addition, Gene Expression Profiling Interactive Analysis (GEPIA) and COX regression analysis were carried out. Simultaneously, the proportions of various tumor-infiltrating immune cells (TIICs) in different TMEs were adopted to explore the correlation between NDUFAF2 and TIICs according to CIBERSORT.
To further explore the potential roles of NDUFAF2, the analyses of biological pathways included in the LUAD tumorigenesis-related NDUFAF2 regulatory network were conducted using gene set enrichment analysis (GSEA) and Gene Ontology (GO) analysis. Furthermore, the probable The HPA showed the expression of NDUFAF2 at the protein level in LUAD tissues and normal lung tissues using IHC staining. * * * indicates P < 0:001, * * indicates P < 0:01, and * indicates P < 0:05.
correlation between NDUFAF2 and TIICs in a variety of TMEs was assessed using single-sample GSEA (ssGSEA). NDUFAF2 might play a key role in cancer cell proliferation, which could be used as a new independent prognostic marker, and its expression level was correlated with immune infiltrates in LUAD.      Affiliated Hospital of Nanchang University. The collected tissue specimens were immediately preserved in liquid nitrogen and then at -80°C. Our current study was approved by the ethics committee of the First Affiliated Hospital of Nanchang University (No. 2022-1-018). Signed written informed consents were provided by all six LUAD patients, and they did not undergo any chemotherapy or radiation therapy before surgery or receive other special treatment prior to the operation. Moreover, these patients had no other malignancies. The flow diagram of this study is shown in Supplementary Figure 1.

Acquisition of Detailed Patient
Data. The mRNA profiles including 513 LUAD specimens and 59 adjacent nontumor tissue specimens (workflow typed: TPM), clinicopathological data, and other data were obtained from the TCGA database (https://cancergenome.nih.gov) in March 2022. In addition, an independent external dataset was extracted, and the GSE116959 including 57 LUAD specimens and 11 nontumor tissue specimens acquired from the GEO database (https:// www.ncbi.nlm.nih.gov/geo/) was adopted to confirm the differential expression of NDUFAF2. Furthermore, the expression of NDUFAF2 at the mRNA level between LUAD and nontumor tissues was compared using the GEPIA (http:// gepia.cancer-pku.cn). Moreover, the expression of NDU-FAF2 at the protein level between the LUAD tissues and nontumor tissues was analyzed using the Human Protein Atlas (HPA) (https://www.proteinatlas.org) and the UALCAN web (http://ualcan.path.uab.edu/index.html). In the present study, our databases were in agreement with the publication guidelines offered by the online database, and approval by ethics committee and informed consent were waived.

GO/Kyoto Encyclopedia of Genes and Genomes (KEGG)
Enrichment Analysis and GSEA. KEGG pathway enrichment analysis and GO term analysis of the differentially expressed genes (DEGs) were carried out using the XIANTAO web (http://www.xiantao.love) based on the cluster Profiler R package (v3.18.0). GSEA was applied to explore the biological functions and enrichment pathways correlated with NDUFAF2 in LUAD. For each analysis, gene set permutation was performed 1,000 times. Normalized enrichment score ðNESÞ > 1, gene sets with false discovery rate (FDR) q value < 0.25, and nominal (NOM) P value < 0.05 were regarded as significant. 2.5. Cell Lines and Culture Condition. Six types of human LUAD cell lines (PC9, H358, A549, HCC827, H1299, and H1975) and pulmonary epithelial cell line (BEAS-2B) were provided by the Procell Life Science &Technology Co., Ltd. (Wuhan, China). These cells were maintained in DMEM (Gibco) or RPMI-1640 (Hyclone) supplemented with 10% fetal bovine serum (FBS), 1% streptomycin, and 1% penicillin at 37°C in a humidified atmosphere containing 5% CO 2 .
2.6. Western Blotting Analysis. The samples of cells and tissues were homogenized in RIPA lysis buffer with 1% protease inhibitor, 1% phosphatase inhibitor, and 1% PMSF (Applygen, Beijing, China). The same amount of proteins  3) was used to analyze the expression profiles based on the XIANTAO platform. Differences between the two groups were compared using the Wilcoxon rank-sum test, and the analysis of variance (ANOVA) was adopted for multigroup comparisons, followed by a Student's t-test. The receiver operating characteristic (ROC) curve was applied to investigate the diagnostic value, and a nomogram was employed to assess the predictive value. The death risk, including age, gender, TNM stage, pathologic stage, smoking history, and NDUFAF2 expression, was assessed using univariate and multivariate Cox regression analyses. All tests were two-sided, and a P value < 0.05 was treated as statistically significant. * , * * , and * * * represent P < 0:05, P < 0:01, and P < 0:001, respectively, in the analysis.   . These data illustrated that the high expression level of NDUFAF2 was remarkably associated with the pathologic stage (P < 0:001), T stage (P < 0:001), N stage (P < 0:001), and OS (P < 0:001).  (Figure 1(a)). Moreover, the average expression level of NDUFAF2 in LUAD tissues was remarkably greater compared with nontumor tissues (P < 0:001, Figure 1(b)). Similar findings were confirmed in paired LUAD tissues and adjacent nontumor tissues (P < 0:001, Figure 1(c)). Similar results showing the higher expression of NDUFAF2 at the mRNA level in LUAD were obtained from the GEO database using GSE116959 datasets (Figure 1(d)). Its high expression in LUAD was also confirmed using the matched TCGA and GTEx data based on the GEPIA platform (Figure 1(e)). We also evaluated the expression of NDUFAF2 at the protein level in LUAD and adjacent nontumor tissues using the UALCAN web and HPA database. The expression of NDUFAF2 at the protein level was higher in LUAD compared with adjacent nontumor tissues (Figure 1(f)). Moreover, its expression was low in type II alveolar cells and medium in macrophages, while it was not detected in type I alveolar cells and endothelial cells using HPA054776 antibodies (Figure 1(g)). In contrast, high expression of NDUFAF2 at the protein level was detected in LUAD tissues using the HPA database, showing high staining intensity in lung tumor cells using HPA054776 antibodies (Figure 1(h)). Furthermore, we selected six pairs of clinical patient samples from our cohort. Western blotting analysis indicated that the expression of NDUFAF2 was higher in LUAD compared with normal lung tissues (Figure 2(a)). In addition, in vitro experiments revealed that the expression of NDUFAF2 at the protein level in NSCLC cell lines, including PC9, H358, A549, HCC827, H1299, and H1975 cells, was markedly greater compared with  (Figure 2(b)). IHC staining of NDUFAF2 was carried out in LUAD tissues and adjacent nontumor tissues. Our results showed that the expression of NDUFAF2 in LUAD was greater compared with adjacent nontumor tissues (Figure 2(c)). Taken together, these findings suggested that the expression of NDUFAF2 at both the mRNA and protein levels was increased in LUAD tissues.

Correlation of NDUFAF2 Expression with Clinical
Features in LUAD Patients. To further reveal the correlation of NDUFAF2 expression with clinicopathologic features in LUAD patients, the expression pattern of NDUFAF2 at the mRNA level in various clinical categories was analyzed using the TCGA database. We found that the high expression of NDUFAF2 was remarkably associated with the N stage, T stage, and pathologic stage, but not age and gender (Figure 3).

Analysis of Survival and Subgroup
Analysis. The Kaplan-Meier survival analysis was employed in the TCGA cohort to understand whether the expression level of NDUFAF2 affected the survival of patients. Figure 4(a) reveals that the high expression level of NDUFAF2 was significantly correlated with the poor OS of LUAD patients (HR = 1:45, P = 0:013). Further subgroup analysis indicated that a high expression level of NDUFAF2 was related to a poor

NDUFAF2
Is an Independent Risk Factor for OS. The association of NDUFAF2 with the clinical features in LUAD was assessed. High expression level of NDUFAF2 was associated with a poor OS, which was significantly correlated with the pathologic stage and TNM stage ( Table 2). Univariate Cox regression analysis showed that the high expression level of NDUFAF2 was markedly related to a poor OS (HR = 1:453, 95%CI = 1:081~1:953, P = 0:013). Table 2 implies that NDUFAF2 was an independent risk factor for OS in LUAD patients (HR = 1:538, 95%CI = 1:086~2:177, P = 0:015), supported by the multivariate Cox regression analysis.

Diagnostic and Predictive Values of NDUFAF2
Expression in LUAD. ROC curve was adopted to analyze the diagnostic value of NDUFAF2 expression level in LUAD patients. The area under the ROC curve (AUC) was 0.738 ( Figure 5(a)). Moreover, the AUC value of NDUFAF2 was 0.738 when discriminating LUAD tissues from healthy lung tissues or adjacent nontumor tissues. To detect the predictive value of NDUFAF2 expression level in LUAD, we constructed a nomogram according to the expression profile of NDUFAF2 in the TCGA database. The clinical variables combined with multivariate Cox analysis results and frequent clinical variables age and gender were selected to establish a nomogram to predict the 1-year, 3-year, and 5year survival probabilities ( Figure 5(b)). The nomogram was calibrated with the calibration curve ( Figures 5(c)-5(e)).

GO and KEGG Analyses and GSEA.
DEGs were identified according to the expression of NDUFAF2. The potential function of NDUFAF2 was predicted based on the enriched GO terms and KEGG pathway analysis ( Figure 6). We discovered that these NDUFAF2-associated genes were mainly enriched in GO : 0002227 innate immune response in mucosa, GO : 0002385 mucosal immune response, GO : 0000786 nucleosome, GO : 0032993 protein-DNA complex, GO : 0031492 nucleosomal DNA binding, GO : 0031490 chromatin DNA binding, hsa00982 drug metabolism-cytochrome P450, and hsa00983 drug metabolism-other enzymes. We, respectively, analyzed the NDUFAF2 expression in LUAD enrichment in G2_M_checkpoints, DNA replication, diseases of mitotic cell cycle, retinoblastoma gene in cancer, cell cycle pathway, and cell cycle using GSEA (Figures 7(a)-7(f)). More information on function and pathway enrichment is shown in Table 3.

The Correlation of NDUFAF2 Expression with Cell Cycle
Regulatory Genes. The GSEA implied that NDUFAF2 was markedly enriched in the diseases of mitotic cell cycle, cell cycle, and cell cycle pathway. Thus, we further analyzed the association of NDUFAF2 with cell cycle regulatory genes. Our data showed that the expression of NDUFAF2 was positively correlated with several cell cycle regulatory genes, such as CCNA2, CCNB1, CCNB2, CDC6, CDC20, CDC25A, CDC25C, CDC45, CHEK1, MCM2, MCM6, PCNA, PLK1, PTTG1, and MCM4 (Supplementary Figure 2).

NDUFAF2 Expression Is Associated with Immune Cell
Infiltration. To clarify the association of NDUFAF2 expression level with tumor immune response, we further performed an enrichment analysis of the LUAD TME using ssGSEA. Figure 8 shows that the Tem, Tcm, B cells, and CD56 bright NK cells were negatively associated with the expression level of NDUFAF2. The dendritic cells (DCs) and Th2 cells were positively correlated with the expression level of NDUFAF2. All the above-mentioned findings suggested that upregulation of NDUFAF2 was concomitant with low immune infiltration, leading to a worse OS.

Discussion
It is well known that NDUFAF2 acts as a mitochondrial complex I assembly factor, and its biological function is conserved from fungi to mammals. The deficiency of NDUFAF2 due to null mutations does not absolutely restrain the assembly or activity of complex I, although it contributes to the development of a progressive encephalopathy [14]. Depletion of NDUFAF2 can impair cell respiration and ATP turnover, increase the output of reactive oxygen species (ROS), and cause oxidative damage, resulting in mitochondrial DNA damage [9]. Some studies have shown that depletion  The expression of NDUFAF2 log 2 (TPM + 1) The expression of NDUFAF2 log 2 (TPM + 1) 17 Computational and Mathematical Methods in Medicine of NDUFAF2 can affect the growth of some cells, such as HEL cells [11]. Abnormality of NDUFAF2 such as mutations causes mitochondrial complex I deficiency [10,15], mitochondrial complex I is very important in tumor metabolism, and defective NADPH production in mitochondrial disease complex I leads to inflammation and cell death [16]. Mitochondrial complex I donates electrons to ubiquinone, which results in the generation of ubiquinol and the regeneration of the NAD+ and FAD cofactors; simultaneously, the mitochondrial electron transport chain (ETC) is necessary for tumor growth [17]. Human lung tumors display robust glucose oxidation by mitochondria [18]. We speculated that NDUFAF2 mainly affects mitochondrial complex I to influence the progression of lung adenocarcinoma. Complex I is the first enzyme of the mitochondrial respiratory chain and consists of 45 subunits in humans; it is one of the largest known multisubunit membrane protein complexes [19]. Complex I is assembled via a series of intermediate assembly modules and requires the involvement of more than 10 known assembly factors; some assembly factors are homologous to structural subunits of complex I and serve as their "place holders" in assembly intermediates; for instance, NDUFAF2 is a homologue of the NDUFA12 subunit [20]. In NDUFA12-knockout cells, the NDUFAF2 substituted for its paralogue NDUFA12, leading to complex I appearing fully assembled [19,21]. Tumor growth is clearly affected by function by mitochondrial complex I [17].
In our present study, through bioinformatics of NDU-FAF2 expression in different tissues, we noticed that the expression of NDUFAF2 was remarkably greater in LUAD compared with adjacent nontumor tissues. IHC staining images acquired from the HPA also supported this conclusion. We externally validated the differences at the protein and pathological levels via western blotting analysis and IHC analysis. Furthermore, the expression of NDUFAF2 was significantly higher in the lung cancer cell lines. A prognostic nomogram was also established, including age, gender, T stage, and the expression of NDUFAF2, which could be used to enhance the accuracy in identifying patients at high risk. A higher point on the nomogram indicated a worse clinical outcome.
The GO and KEGG analyses and GSEA were carried out to deeply investigate the biological functions of NDUFAF2. GSEA showed that G2M checkpoints, DNA replication, cell cycle, and DNA damage response were differentially enriched with the NUDFAF2 expression phenotype, and KEGG/GO indicated that pathways in innate immune response in mucosa, mucosal immune response, nucleosome protein-DNA complex, and nucleosomal DNA binding were mainly enriched. Uncontrolled, unlimited, and accelerated proliferation is one of the primary life behaviors of tumor cells [22]. Tumor cells can cease the cell cycle to repair DNA damage through the G2-M checkpoint. Previous studies have shown the function of WEE1 kinase in mediating the G2/M cell cycle checkpoint, and it is hypothesized that inhibition of WEE1 is most active in a TP53-mutated background, where cells lose the G1/S cell cycle checkpoint, leading to enhanced dependence upon the G2/M checkpoint. The antitumor drug based on WEE1 inhibition has completed phase II clinical trial [23,24].
Additionally, we analyzed the connection of the NDU-FAF2 expression level and immune cells in LUAD by using immune cell infiltration datasets from CIBERSORT. We observed that there was a positive correlation of the expression level of NDUFAF2 with the infiltration of immunosuppressive cells, such as Th2 cells, while there was a negative correlation of its expression level with immune killer cells, such as effector memory T cells (Tem) and central memory T cells (Tcm). This finding was similar to the results of the immune landscape from preneoplasia to invasive LUAD, in

18
Computational and Mathematical Methods in Medicine which tumors are found to be Th2-skewed [25]. Th2 cells release IL-5, IL-4, and IL-13, resulting in tumor cell growth and metastasis. TH1/TH2 ratio is important in the antitumor immune response, indicating that strengthening the TH1 response and inhibiting the TH2 activation and immunity-related cytokines may concurrently help eliminate disseminated tumor cells, preventing cancer metastasis and recurrence [26]. Meanwhile, Tcm cells possess higher persistence and antitumor immunity than Tem cells and effector T (Teff) cells [27]. Besides, the Tcm/Teff ratio has been recommended as a predictive biological indicator of response to the treatment of checkpoint inhibitors in NSCLC patients [28]. The drawback of our work here was that only a small sample was validated, and the actual function of NDUFAFA2 in LUAD needs to be further validated. In summary, the functionality of NDUFAF2 was multiaspect. It played a key role in predicting the prognosis of LUAD. However, the detailed effects and mechanism of NDUFAF2 in LUAD remain to be further explored in the future study.

Conclusions
Our research revealed that NDUFAF2 was predominantly expressed in LUAD, and its expression played a critical role in the diagnosis and prognosis of LUAD. In human LUAD, the expression of NDUFAF2 was upregulated, and overexpression of NDUFAF2 was correlated with N stage, T stage, and pathologic stage in LUAD, leading to a worse OS. Moreover, the expression of NDUFAF2 was an independent prognostic biomarker in LUAD. In addition, NDUFAF2 might affect the prognosis of LUAD via DNA replication, diseases of mitotic cell cycle, cell cycle pathway, and cell cycle. Furthermore, upregulation of NDUFAF2 in LUAD was positively correlated with tumor immune infiltration of DC and Th2 cells. Our current findings provided valuable insights into the potential therapeutic regimens of LUAD.

Data Availability
The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding authors.

Ethical Approval
This study involving human participants were reviewed and approved by the Institutional Research Ethics Committee of the First Affiliated Hospital of Nanchang University (Num.: (2022) yylsd (4-029)).

Consent
The patients/participants provided written informed consent to participate in this study.