Introduction
Lung cancer is the most prevalent cancer type and the leading cause of cancer-related deaths worldwide [1]. Non-small cell lung cancers (NSCLCs) are the most common lung cancers and known to exhibit diverse pathological features [2]. However, the pathological evaluation of lung cancer has become complex owing to an improved understanding of disease progression and prognosis [3]. Notably, the histologic type and tumor, node, metastasis (TNM) stage assist prognosis. However, regardless of the stage and histology, some cases show rapid progression, accompanied by remarkably distinct genetic alterations compared to common clinical cases.
Lung adenocarcinoma (ADC) is the most common type of NSCLC that accounts for approximately 40% of all lung cancers. It can progress from an indolent in situ carcinoma to an invasive, aggressive, and metastatic tumor [4]. Rapid progression of early lung cancer poses significant challenges to treatment planning and management. Apart from often limiting the feasibility of curative surgical options, more aggressive or targeted therapies, such as systemic chemotherapy, may be required to control the disease.
Ubiquitin C-terminal hydrolase L1 (UCHL1) is an adenosine triphosphate-independent ubiquitin ligase, specifically expressed in the brain, nerve, and neuroendocrine cells [5]. Apart from the expression of UCHL1 in normal tissues, such as neurons, increasing evidence suggests that it is also upregulated in some cancers including NSCLC [6-12] and that it plays a critical role in cell proliferation, migration, invasion, and suppression of apoptosis [13]. However, only a few studies on UCHL1 expression in NSCLC have investigated its expression and inhibition in lung cancer cell lines and the clinicopathological features of UCHL1 expression, and their results have been inconsistent and controversial [14-16]. This study aimed to investigate the expression level of UCHL1 in NSCLC and explore its potential usefulness as a disease-progression biomarker.
Materials and Methods
1. Single-cell RNA sequencing analysis
Single-cell RNA sequencing (scRNA-seq) was performed using 10X chromium v3 chemistry. scRNA-seq process, cluster annotation, and analysis of differentially expressed genes (DEGs) were conducted using R version 4.0.3 (R Foundation for Statistical Computing, Vienna, Austria) and Seurat (version 4.0.2) as described in previous studies [17-19]. scRNA-seq data on normal-appearing lung epithelial versus tumor cells were obtained from initial surgical specimens of a patient who showed rapid progression of lung cancer and died 3 months after diagnosis. Detailed information regarding scRNA-seq is provided in Supplementary Materials and Methods. Among the DEGs, biomarkers that could affect cancer progression and prognosis were identified through literature review and further evaluated by immunohistochemistry (IHC) (Supplementary Table S1).
2. Study cases
To evaluate UCHL1 expression in human lung cancer, specimens were selected from the tissue bank containing samples obtained from patients who underwent surgical resection at Severance Hospital, affiliated with Yonsei University, between January 2005 and March 2009. We conducted a comparative analysis between normal epithelial and tumor cells from a surgical resection to explore the characteristics of ADC cells. Specimens obtained from the operating room were transferred to a pathology laboratory for examination. A total of 50 formalin-fixed, paraffin-embedded, lung-cancer tissue slides were analyzed by IHC using an UCHL1 antibody. All available data, including UCHL1 expression levels, and clinicopathological features, including age, sex, smoking status, and TNM status, were collected. The TNM system of the 7th edition of the International Association for Lung Cancer Research was used for diagnosis and classification. This study was approved by the Institutional Review Board of Gangnam Severance Hospital (approval no. 3-2017-5509) and carried out in compliance with the Declaration of Helsinki. All patients provided written informed consent.
3. IHC
IHC was performed as described elsewhere [20]. Briefly, UCHL1 expression was analyzed in human lung cancer tissue via IHC using the LABS2 System (Dako, Carpinteria, CA, USA). Specimen sections were deparaffinized, rehydrated, immersed in an H2O2/methanol solution, and 200 μL of a 1:200 dilution of UCHL1 antibody (UCHL1/PGP9.5 Antibody 31A3-NB600-1160, Novus Biologicals, Littleton, CO, USA) in blocking solution was applied to each slide at room temperature for 90 minutes. Sections were incubated for 10 minutes with a biotinylated linker and processed using avidin/biotin IHC techniques. 3,3-Diaminobenzidine (DAB) was used as a chromogen in conjunction with the Liquid DAB Substrate kit (Novacastra, Newcastle upon Tyne, UK).
4. scRNA-seq identified UCHL1, BEX3, and MDK as significant DEG
scRNA-seq analysis were performed as described earlier [17-19]. Briefly, after cell clustering and annotating from paired normal and tumor tissue, barcodes of epithelial cells inside the tumor tissue were secured. We extracted epithelial cell barcodes obtained from tumor tissues and designated the cell clusters obtained from normal lung tissues and nonepithelial cell clusters obtained from tumor tissues as the reference. A non-supervised clustering method was applied using the InferCNV package (https://github.com/broadinstitute/inferCNV) to identify cells with distinct chromosomal gene expression perturbations as lung cancer cells. Among the DEGs (Figure 1, volcano plot), we identified 500 genes, which were overexpressed in malignant epithelial cells as compared to normal lung tissue. DEG analysis identified upregulated genes, primarily related to signal transduction (Supplementary Table S1), such as UCHL1, brain expressed X-linked 3 (BEX3), and midkine (MDK), indicating that apoptotic, immune-response, and programmed-cell-death signaling pathways were activated. We conducted IHC staining for UCHL1, BEX3, and MDK, which are associated with tumor growth, and found that only UCHL1 was specifically overexpressed in lung cancer cells.
5. UCHL1 expression in NSCLC tissues
To evaluate UCHL1 expression in human lung cancer cells, we analyzed 50 paraffin-embedded NSCLC tissue samples. Of these, two samples were excluded from the analysis due to inadequate sample amount. UCHL1 expression was evaluated using a scoring system based on the product of intensity and proportion scoring, resulting in the calculation of the histochemical scoring (H-score) for each tissue sample. The UCHL1 staining results of clinical specimens were interpreted by comparing their staining intensity with that of the reference sample. Two independent researchers (Park MK and Kim CY), who were blinded to the pathological reports, analyzed the slides. In cases of discrepancy in the results between the two investigators, a third researcher (Chang YS) also evaluated the slides. We interpreted the H-score and UCHL1 expression as follows: scores ranging from 0 to 100 indicated low UCHL1 expression, scores from 101 to 200 indicated moderate UCHL1 expression, and scores from 201 to 300 indicated high UCHL1 expression. We compared low and moderate expression scores with high expression scores.
6. Statistical analysis
All statistical analyses were performed using R version 4.3.0 (R Foundation for statistical computing). Data are expressed as mean±standard deviation or median with interquartile ranges, as appropriate. Continuous and categorical variables were compared using the Student’s t-test and chi-squared/Fisher’s exact test, respectively. Univariate and multivariate logistic regression analyses were conducted to determine the odds ratio (OR) and corresponding 95% confidence interval (CI) for UCHL1 expression. A significance level of p<0.05 was used to determine statistical significance for all analyses.
To confirm the correlation between UCHL1 expression and survival outcomes, we conducted a survival analysis using the The Cancer Genome Atlas (TCGA)-lung ADC database, accessed through the OncoLnc platform (http://www.oncolnc.org), and used clinical information from the cBioPortal platform (https://www. cbioportal.org). After merging the data using TCGA ID, we excluded entries with missing values for age, sex, smoking amount (pack-years), and cancer stage. For adjusting for confounding factors, we used the Mahalanobis distance matching, followed by a nearest-neighbor search to minimize the impact of potential confounding factors. This process entailed aligning the Mahalanobis distance of each observation within the treatment group with the nearest corresponding Mahalanobis distance of the control group. The covariates considered for this matching included age, sex, smoking status, and cancer stage.
Results
1. Case presentation
A 62-year-old man, who had a 40-pack per year history of active smoking, developed cough, sputum, and blood-tinged sputum 3 months prior to admission. From an initial chest computed tomography (CT) scan (Figure 2A), he was suspected to have early-stage lung cancer. The CT scan revealed a 13×10-mm primary lesion located in the left upper lobe, along with suspicious interstitial lung disease. The lung-cancer stage was determined to be cT1bN0M0. F-fluorodeoxyglucose positron emission tomography/CT 1 month after the initial chest CT scan showed that an additional nodule had developed in the left lobe (Figure 2B), which had not been identified in the first CT scan. This suggested the possibility of a newly developed satellite nodule or interlobar lymph node enlargement (Figure 2B; stage T3N0M0 or T1bN1M0; Figure 2C, D). Surgical resection pathology identified two invasive ADCs in left upper lung with mediastinal lymph node metastasis and pT3N2M0 pathological staging (Figure 2F-I). After undergoing four cycles of adjuvant chemotherapy, the patient exhibited extensive bone and hepatic metastases shortly after surgery, precisely 4 months later (Figure 2E).
2. Demographic characteristics of the patients used for IHC study
The clinical characteristics of the enrolled patients are presented in Table 1. The study population included 32 (66.7%) men, and the average age was 62.5±8.8 years. NSCLC was diagnosed in all 48 patients via pathological examination. Among these patients, the most common pathological diagnosis was ADC, accounting for 26 (54.2%) cases. Additionally, squamous cell carcinoma (SQCC), large cell neuroendocrine carcinoma (LCNEC), and adenosquamous carcinoma were observed in 19 (39.6%), two (4.2%), and one (2.1%) cases, respectively. Table 2 presents the pathological findings stratified by UCHL1 expression levels. Tumor necrosis was found in five (17.9%) cases with low UCHL1 expression, whereas it was present in 12 (60.0%) cases with high UCHL1 expression (p<0.001). The high-UCHL1-expression group had a higher proportion of stage III NSCLC (eight [40.0%] cases) compared to the low-expression group. High-grade differentiation was also more frequently observed in the high-UCHL1-expression group. However, the proportions of patients with lymphovascular and pleural invasion did not show significant differences between the two groups. The expression of UCHL1 in NSCLC tissues and staining intensity was notably higher, particularly strong in SQCC (Figure 3).
3. Pathologic features associated with UCHL1 overexpression
Our univariate analysis indicated that several factors were independently associated with UCHL1 overexpression (Table 3). These factors included sex; smoking status; squamous pathology; and tumor size, necrosis, and differentiation. Table 4 displays the findings of the multivariate logistic regression analysis examining the association between UCHL1 expression and pathological findings, while adjusting for confounding variables. The results showed that high-grade differentiation was significantly associated with high UCHL1 levels (OR, 7.73; 95% CI, 2.00 to 51.5; p=0.008). Additionally, smoking and tumor size were also significantly associated with UCHL1 expression (smoking: OR, 1.06; 95% CI, 1.01 to 1.12; p=0.025; tumor size: OR, 1.07; 95% CI, 1.01 to 1.16; p=0.038). UCHL1 expression was found to be associated with smoking, tumor size, and high-grade dysplasia, which are factors typically associated with poor prognosis. These findings suggest that UCHL1 expression may serve as a potential prognostic marker in NSCLC.
4. Survival analysis with TCGA dataset
For survival analysis, we used TCGA data, which includes age, sex, smoking status, stage, and survival information. Our analysis involved investigating the association between UCHL1 expression and patient survival outcomes using TCGA data. After merging the data using TCGA ID, we excluded entries with missing values for age, sex, smoking amount (pack-years), and cancer stage. We removed missing variables and included a total of 330 participants in the analysis (Figure 4). After cleaning and merging the dataset, we employed a logrank test for survival analysis by categorizing individuals into groups based on gene expression levels— low gene expression (<100.8 fragments per kilobase of transcript per million mapped reads [FPKM]) and high gene expression (≥100.8 FPKM). The Kaplan-Meier curves of UCHL1 expression of ≥100.8 and <100.8 for overall survival (OS) (Figure 3A) demonstrated 10-year OS rates of 23.3% and 48.1% (p=0.047), respectively.
Discussion
This study demonstrated significant associations between UCHL1 expression and various clinicopathological features of patients with NSCLC. UCHL1 overexpression was associated with smoking, tumor size, and high-grade dysplasia, which are typically associated with a poor prognosis and survival outcome. In addition, we investigated additional pathological cases of patients with NSCLC, and externally validated our hypothesis using the TCGA database. These findings suggest that UCHL1 expression may serve as a potential prognostic marker in NSCLC.
Previous studies have reported consistent findings regarding the expression of UCHL1 in different types of lung cancer and its correlation with prognosis. Shimada et al. [21] found that UCHL1 serves as a potential prognostic marker and therapeutic target in high-grade neuroendocrine lung cancer (HGNEC), particularly in small cell lung cancer. Additionally, targeting UCHL1 may enhance therapeutic response in preclinical studies. Elevated levels of UCHL1 mRNA in circulating extracellular vesicles suggest its potential as a diagnostic biomarker for early-stage HGNEC. Another study [22] investigated UCHL1 expression in NSCLC, particularly ADC, and found that it is associated with advanced disease stages and shortened OS, independent of programmed cell death-ligand 1 (PD-L1) expression. Their observations are consistent with our results, as we found that UCHL1 is expressed in all types of highgrade NSCLC, including cases of SQCC.
Although the exact mechanism is unclear, previous in vitro experimental studies have suggested that UCHL1 plays a role in promoting the proliferation and metastasis and inhibiting the apoptosis of human cancer cells [15,23-25]. We attempted to identify functional associations between UCHL1-related proteins using the STRING platform (https://string-db.org). The analysis showed that UCHL1 is associated with 10 additional proteins—lysosomal associated membrane protein 2 (LAMP2), ubiquitin C (UBC), heat shock 70 kDa protein 1 (HSPA8), ubiquitin C-terminal hydrolase L3 (UCHL3), ribosomal protein S27a (RRPS27A), ubiquitin A-52 residue ribosomal protein fusion product 1 (UBA52), synuclein alpha (SNCA), ubiquitin B (UBB), COP9 signalosome subunit 5 (COPS5), and tumor protein P53 (TP53) (Supplementary Figure S1). Using these 10 markers (Supplementary Figure S1), we identified a pathway linked to both processes associated with neuron apoptosis in Parkinson’s disease and proteasomal degradation. This analysis was carried out using the Cytoscape platform (https://cytoscape.org). We identified apoptosis as the primary mechanism of neuronal death in Parkinson’s disease. Apoptosis is characterized by elevated rates of protein degradation [26,27] and increased caspase activity [28]. Furthermore, mutations in UCHL1 could elevate oxidative stress, resulting in apoptosis, similar to what is observed in neuronal cells.
Several studies have demonstrated that the activation of UCHL1 promotes cancer cell invasion by activating various signaling pathways, including mitogen-activated protein kinase (MAPK)/extracellular signal-regulated kinase 1/2 and phosphatidylinositol-3-kinase/protein kinase B (AKT), involved in tumorigenesis of different types of cancer [13,21,23,29-32]. UCHL1 has also been implicated in the stabilization of both wild-type and mutated TP53 in cancer cells. UCHL1 helps maintain the expression of wild-type TP53, which leads to cancer suppression. However, it can also stabilize mutant TP53 protein, thus promoting cancer development. Moreover, UCHL1-mediated activation of the AKT and MAPK signaling pathways phosphorylates mouse double minute 2 (MDM2), which results in p53 downregulation. These findings highlight the complex role of UCHL1 in cancer progression and its involvement in multiple signaling pathways [33,34].
UCHL1 has also been suggested to affect drug response. A study by Hussain et al. [35] indicated that UCHL1 expression levels are positively correlated with a poor prognosis and resistance to pemetrexed in patients with NSCLC. The authors observed a significant upregulation of UCHL1 expression in pemetrexed-resistant NSCLC cells and found that silencing of UCHL1 gene expression or inhibition of the UCHL1 protein suppressed the resistance of NSCLC cells to pemetrexed and other drugs. The study proposed that UCHL1 promotes the resistance of NSCLC cells to pemetrexed by increasing the expression of thymidylate synthase (TS), as evidenced by the reduction of TS expression upon UCHL1 inhibition and the restoration of resistance after recovery of TS expression. Moreover, UCHL1 upregulated TS expression, which mitigated the DNA damage and cell cycle arrest induced by pemetrexed in NSCLC cells, and conferred resistance to pemetrexed and other drugs. The authors suggested that unraveling the precise mechanism of UCHL1 in cancer could potentially lead to the discovery of a novel diagnostic marker and drug target.
Our study had certain limitations. First, it was a retrospective study conducted at a single center, which only focused on a limited number of relatively early-stage NSCLC cases. Furthermore, due to the small number of cases, a lack of internal validation occurred that could demonstrate a correlation between UCHL1 expression and survival outcomes. Finally, discordance between scRNA-seq and protein expression during development can occur under non-steady-state conditions, which may limit their use as biomarkers.
Despite these limitations, our study bore several strengths. First, it identified a biomarker using scRNAseq data from patients with rapidly-progressing lung cancer, which allowed us to suggest UCHL1 as a suitable prognostic marker. We investigated both the correlation between UCHL1 expression and pathology and the patients’ histologic parameters and survival outcome. Nevertheless, our study was particularly designed to investigate the association between UCHL1 expression and relatively early-stage NSCLC. Further investigations involving larger cohorts and longer follow-ups are necessary to unravel the full prognostic potential of UCHL1 in NSCLC.
In conclusion, our study demonstrated significant associations between UCHL1 expression and various clinicopathological features of patients with NSCLC. UCHL1 expression was found to be associated with smoking, tumor size, and high-grade dysplasia, which are typically associated with poor prognosis. These findings suggest that UCHL1 expression may serve as a potential prognostic marker in NSCLC. Further investigations are needed to elucidate the underlying mechanisms and evaluate the clinical implications of UCH-L1 expression in NSCLC.
PDF Links
PubReader
ePub Link
XML Download
Data Sharing Statement
Full text via DOI
Supplement1
Print
Download Citation
