Nuclear Respiratory Factor 1 Associated with Prognosis and Immune Inltration in Hepatocellular Carcinoma

Background: Recent studies have shown that functional mitochondria are essential for cancer cells. Nuclear respiratory factor 1 (NRF1) is a transcription factor that activates mitochondrial biogenesis and the expression of the respiratory chain, but little is known about its prognostic value and tumor-inltrating lymphocytes association. Here, we evaluated the association among expression of NRF1, clinicopathological characteristics, survival and immune inltration in hepatocellular carcinoma (HCC). Methods: We used the Tumor Immune Estimation Resource (TIMER) to analyze the difference of NRF1 mRNA expression in human cancers. Clinical-pathological information and follow-up data were collected from HCC (n = 171) and chronic hepatitis (n = 113) patients. NRF1 expression were scored based on the percentage and intensity of immunohistochemical staining in pathological slides. Correlations between clinical features and the expression of NRF1 were evaluated by Chi-square test, Kaplan-Meier curves, logrank tests and multivariate Cox regression analysis. The correlations between NRF1 expression and gene marker sets of tumor inltrating lymphocytes (TILs) were analyzed by TIMER and Gene Expression Proling Interactive Analysis (GEPIA) databases. Results: NRF1 mRNA expression was signicantly higher in HCC than in normal tissue. Compared with chronic hepatitis, more frequency of NRF1 high expression are found in HCC (31.58 % vs 13.27 %, P < 0.001, P < 0.001). In addition, the NRF1 expression was signicantly associated with hepatic cirrhosis (P = 0.021) and vascular invasion (P = 0.025). NRF1 expression was also a signicant independent predictor of survival in HCC (P = 0.003; HR adj = 0.20; 95% CI = 0.09 – 0.44). NRF1 showed positively correlated with TILs, including B cell (r = 0.384, P = 1.68e-13), CD8+ T cells (r = 0.246, P = 3.99e-06), CD4+ T cells (r = 0.535, P = 6.90e-27),

logrank tests and multivariate Cox regression analysis. The correlations between NRF1 expression and gene marker sets of tumor in ltrating lymphocytes (TILs) were analyzed by TIMER and Gene Expression Pro ling Interactive Analysis (GEPIA) databases.
Conclusions: NRF1 expression was a useful independent prognostic factor and correlated with tumor immune in ltration in HCC.

Background
Liver cancer is the second leading cause of cancer death worldwide in men [1]. Most primary liver cancer occurring worldwide is hepatocellular carcinoma (HCC) [2,3]. The early diagnosis of HCC is so far complicated, and there is few effective therapies for HCC [4,5]. Thus, an effective biomarker is urgently needed to estimate the prognosis. Recently instead of chemotherapies and targeted therapies, the immunotherapies have opened a new era of anticancer treatment. Given HCC is a prototype of in ammation-associated cancer, immunotherapies are the promising breakthroughs of anticancer treatment [6].
Hepatocytes which are rich in mitochondria, have developed diverse mechanisms to maintain mitochondrial homeostasis by regulating mitochondrial dynamics, biogenesis and degradation [7,8]. The emerging studies have shown that functional mitochondria are essential for the cancer cell [9]. Beyond the classical role in energy and metabolic mechanisms, mitochondria produce reactive oxygen species, which is capable of increasing tumorigenesis by activating signaling pathways that regulate cellular proliferation, metabolic alterations, and angiogenesis [10]. Mitochondria in cancer cell are different from their normal counterparts in structure and function [11]. Besides, robust reports have proposed links between immune function and mitochondrial processes. Indeed, monogenetic disorders of mitochondrial components may manifest with immune dysfunction [12]. Nuclear respiratory factor 1 (NRF1) is a key transcription factor linked to machinery in mitochondrial biogenesis and the transcriptional expression of the respiratory chain [13,14]. NRF1 also has been identi ed as a valuable biomarker for breast cancer diagnosis and prognosis [15]. However, the effect of NRF1 in HCC progression and tumor immunology remains unclear. The aim of our study was to investigate the effects of NRF1 expression on the prognosis and the relationship with tumor in ltrating lymphocytes (TILs) in HCC.

Study populations
A panel of formalin-xed, para n-embedded HCC and chronic hepatis tissues were excised from fresh surgical samples at the A liated Hospital of Nantong University from 2004 to 2010. Clinical-pathological features and follow-up data were: gender, age at diagnosis, differentiation, vessel invasion, TNM stage, HBV infection, tumor size, AFP value and cirrhosis. None of the patients received radiotherapy, chemotherapy, or immunotherapy prior to surgery. The overall survival duration was the interval from the date of rst biopsy to the date of death from disease.

Immunohistochemistry (ihc)
The tissue microarray (TMA) slides from patients were received for NRF1 staining. The NRF1 staining were performed by Tissue Microarray System (Quick-Ray, UT06, UNITMA, Korea). Core tissue biopsies (2 mm in diameter) were taken from individual para n-embedded sections and arranged in the new recipient para n blocks. IHC analysis was performed as previously described [16]. The slides were incubated with the primary antibodies against NRF1 (Abcam, Cambridge, MA, USA) at 4 °C overnight. Vectra 3 System (PerkinElmer, USA) was used to acquire and analyze the images. There are two parameters estimated: intensity (0 to 3 as negative, weak, moderate or strong) and percentage (0% to 100%). The nal staining score of each tissue sample was generated from intensity multiplied percentage scores. The cutoff value of NRF1 expression was set by X-tile software (http://medicine.yale.edu/lab/rimm/research/software.aspx; Rimm lab at Yale University).
Tumor Immune Estimation Resource (TIMER) and Gene Expression Pro ling Interactive Analysis (GEPIA)

Database Analysis
The level of NRF1 mRNA expression in different tumor types were obtained from TIMER (https://cistrome.shinyapps.io/timer/). TIMER was also employed to analyze the correlation between NRF1 mRNA expression and in ltration levels of immune cells in HCC [17,18]. NRF1 was used for the yaxis, and other genes of interest are represented on the x-axis. The gene expression level was displayed with log2 RSEM. GEPIA (http://gepia.cancer-pku.cn/index.html) were employed to further con rm the results in TIMER. It is a web server for analyzing the RNA-Seq expression data from the TCGA and GTEx projects [19]. The Pearson method was used to determine the correlation coe cient. The tumor and normal tissue datasets were used for analysis.

Statistical Analyses
Correlations between clinicopathologic features and expression of NRF1 were evaluated by Chi-square test. Survival time was calculated from date of diagnosis to date of death/censoring. The log-rank test was used to assess differences between groups and the multivariate survival analysis was performed with Cox regression. All P values reported are from two-sided tests and the threshold for signi cance was set at P = 0.05. The statistical association and survival analyses were performed using STATA version 13.0 (StataCorp, TX, USA).
Given the in uence of tumor purity on immune in ltration analysis, the correlation analysis was adjusted for purity [20]. We determined the relationship between NRF1 and the diverse immune in ltrating cells based on the levels of immune marker gene expression by TIMER (Table 4  * P < 0.05, ** P < 0.01, *** P < 0.001 As expect, the results obtained from GEPIA were consistent with the TIMER analysis results (Table 5).
These ndings strongly suggested that NRF1 was speci cally correlated with TILs in HCC.

Discussion
It is clear that the biology of mitochondria in cancer are important to our understanding of cancer biology, as many classical cancer hallmarks result in altered mitochondrial function in tumor [22,23]. NRF1 serves as an activator that promotes mitochondria biogenesis and helps support mitochondrial function [23,24]. Furthermore, it has been identi ed that NRF1 binding motifs presented in many genes which operated in signaling pathways governing all hallmarks of malignant transformation and progression.
The novel roles of NRF1 has been reported in cancer development and progression through its interplay with the transcription factors E2F4 and MYC [25]. Thus, NRF1 inevitably need to be taken into account when evaluating prognostics and therapeutic options for cancer patients. In our study, we demonstrated that NRF1 had correlation with some clinical variables in HCC, such as cirrhosis and vascular invasion. In the survival analysis, NRF1 functioned independently as a prognostic factor for HCC patients.
Tumor immunotherapy is a promising and transformative therapeutic strategy for patients with advanced cancers [26,27]. Our data also demonstrated that NRF1 expression correlated with the in ltration status of B cells, CD4 + and CD8 + T cells, macrophages, neutrophils, and DCs. It suggested that NRF1 involved in regulating tumor immunity, and therefore in uenced HCC prognosis. T cells have been key mediators of antitumor function in cancer immunotherapy and checkpoint blockade therapies have shown potent therapeutic effects in advanced cancer [27,28]. We observed that expression of exhausted T cells markers, PD-1, CTLA-4 and TIM-3, which are critical inhibitory immune checkpoint proteins positively correlated with NRF1 expression. This suggested that NRF1 may inhibit or promote immune cells in ltrating, although the underlying mechanism is unknown.
Several limitations could in uence the outcomes of this study. Firstly, the analysis of in ltration immune correlation is based RNA-seq data retrieved from in public repositories. Hence, the quality of data can in uence the study outcomes. However, we did not verify TILs outcomes by testing our own clinical samples. Secondly, the sample sizes of HCC patients in our study were somehow small. Larger sample size will be necessary for reliable interpretation of data. Thirdly, racial or ethnic differences that are not explained or discussed in our study.

Conclusion
To summarize, NRF1 is a valuable biomarker for HCC, because it was associated with prognosis and immune cell in ltration. We proposed a rationale for future studies to develop NRF1 signaling-based  Figure 1 NRF1 mRNA expression in HCC and normal tissue. The level of NRF1 mRNA expression in different tumor types were obtained from TIMER database. *P < 0.05, **P < 0.01, ***P < 0.001.