Human lysine oxidase-like 3 high expression as a potential biomarker and association with poor prognosis in pleural mesothelioma

Abstract

Purpose

Although human lysine oxidase-like 3 (LOXL3) is associated with various cancers, its role in pleural mesothelioma (PM) remains uncharacterized. This study investigated the expression level and prognostic association of LOXL3 in PM.

Methods

Tissue specimens were collected from patients with PM. The expression levels of LOXL3 were assessed using immunohistochemistry, Western blot analysis, and quantitative reverse transcription PCR. The clinical correlation analysis was conducted using R software (version 3.6.3), incorporating data from both The Cancer Genome Atlas and Chuxiong cohorts. Univariate and multivariate Cox proportional hazards regression models alongside Kaplan–Meier survival curve analysis were performed to evaluate prognostic significance. Additionally, gene expression correlation studies between LOXL3 and other members of the LOX family were performed using the Gene Expression Profiling Interactive Analysis platform. Finally, Gene Set Enrichment Analysis was conducted to identify the signaling pathways associated with LOXL3.

Results

LOXL3 exhibited significant upregulation in both sarcomatoid and biphasic PM subtypes compared to the control samples. Clinico-pathological analysis revealed the correlations between LOXL3 expression levels and cancer type, and Wilms tumor protein 1 (WT-1) status. Cox regression analysis identified cancer type as an independent prognostic factor. Kaplan–Meier analysis demonstrated obviously poorer survival rates in cohorts with high LOXL3 expression. Notably, coordinated expression patterns were observed between LOXL3 and LOXL4. The protein expression level of LOXL3 exhibits a positive correlation with CD68, CD206, and programmed death-ligand 1 (PD-L1), with this correlation being particularly pronounced in sarcomatoid mesothelioma. Functional enrichment analysis indicated that high LOXL3 expression was primarily associated with pathways related to oxidative phosphorylation, late and early estrogen response, and adipogenesis.

Conclusion

LOXL3 is highly expressed in PM and associated with poor prognosis, and is involved in tumor immune evasion. The expression level of LOXL3 is correlated with cancer types and the expression level of WT-1. Cancer type is an independent prognostic factor for PM. LOXL3 expression is positively associated with LOXL4, and high LOXL3 expression is enriched in oxidative phosphorylation, estrogen response, and adipogenesis pathways, while the low-expression group is enriched in apoptosis, interleukin-2/signal transducer and activator of transcription 5, mammalian target of rapamycin complex 1, and transforming growth factor-β pathways. CD68, CD206, PD-L1, and LOXL3 may collaboratively contribute to the regulation of the PM microenvironment and are closely linked to the invasion and metastasis of PM. Therefore, LOXL3 can be used as both a prognostic marker and a potential therapeutic target for PM.

Introduction

Pleural mesothelioma (PM) is a rare yet aggressive malignancy arising from pleural mesothelial cells, with an incidence of approximately 0.02%–0.04% among all malignancies and accounting for 5% of primary pleural tumors. The World Health Organization (WHO) classifies mesothelioma into three histologic subtypes; epithelioid, sarcomatoid, and biphasic. The clinical presentation often lacks specificity, manifesting with non-distinctive symptoms such as chest pain, dyspnea, and pleural effusion, which frequently leads to delayed diagnosis and diagnostic challenges. ¹ Asbestos exposure remains the primary established etiological factor for PM, inducing chronic inflammatory responses that lead to the malignant transformation of pleural mesothelial cells through three main mechanisms: (a) sustained production of reactive oxygen species, which triggers DNA damage; (b) recruitment of macrophages that create a pro-tumorigenic microenvironment; and (c) direct physical disruption of mitotic spindle formation, ultimately resulting in genomic instability. ² Histologically, it mainly includes epithelial, sarcomatous, and biphasic subtypes. Notably, the sarcomatoid subtype demonstrates significantly enhanced invasive potential compared to other subtypes, as evidenced by its characteristically diffuse growth pattern, mesenchymal marker expression, and clinical correlation with poorer prognosis and limited therapeutic response. ³ Alarmingly, the grim prognosis of PM is underscored by a median overall survival (OS) of <12 months and a 5-year survival rate below 5%, primarily attributable to diagnostic challenges in early-stage disease compounded by the lack of effective therapeutic options. Epidemiological surveillance data reveal a concerning global pattern, with age-adjusted incidence rates increasing by 0.7%–1.2% annually over the past decade, particularly in developing nations where asbestos regulations remain inadequate. ⁴ Significant geographical heterogeneity characterizes the epidemiology of PM, with the highest incidence rates reported in Australia, while hyperendemic clusters are observed in Dayao County, Yunnan Province (China). ⁵ At present, PM diagnosis mainly relies on thoracoscopic biopsy, and more accurate early diagnostic markers are lacking. Noninvasive diagnostic markers are clinically beneficial; however, their clinical use remains controversial, with multiple reports providing contradictory findings, which limits their application in the clinic. A better understanding of the molecular mechanisms and predictive factors of poor patient outcomes in PM is needed for identifying new targets and developing effective therapies.

The LOXL3 gene encodes a secreted glycoprotein comprising 753 amino acid residues with a theoretical molecular mass of 82.7 kDa. Structural characterization reveals two conserved functional domains: an N-terminal scavenger receptor cysteine-rich domain mediating extracellular matrix interactions, and a C-terminal lysyl oxidase homology domain containing the conserved copper-binding motif (His-X-His) essential for catalytic activity. ⁶ LOXL3, a copper-dependent amine oxidase, is one of the four members of the lysyl oxidase family, alongside LOX, LOXL1, and LOXL2. Lysyl oxidases deaminate lysine residues in tropocollagen, leading to the formation of insoluble collagen and elastin fibers. ⁷ Accumulating evidence implicates LOXL3 in diverse human and animal diseases. During malignant invasion, Snail transcription factors interact with LOXL3 (and possibly LOXL2) to downregulate E-cadherin expression, thereby facilitating epithelial–mesenchymal transition (EMT). ⁸ Cancers expressing LOXL1, LOXL3, or LOXL4 exhibit significantly reduced OS compared to those lacking these markers. Highly aggressive diffuse-type gastric cancer cells show markedly elevated LOXL3 and LOXL4 messenger RNA (mRNA) levels. ⁹ LOXL3 upregulation also potentiates the oncogenic BRAF pathway during melanocytic transformation and is essential for modulating DNA repair mechanisms and maintaining genomic stability. ⁸ In a study, Kasashima et al. used immunohistochemistry (IHC) to detect LOX, LOXL1, LOXL2, and LOXL3 proteins in 291 breast cancer cases. The results showed that only 13.4% of patients exhibited positive LOXL3 expression. Furthermore, this expression was associated with intratumoral and peritumoral inflammation, progesterone receptor (PR) and estrogen receptor (ER) status, and molecular subtypes. ⁹ Beyond breast cancer, LOXL3 is expressed in myeloproliferative neoplasms, ¹⁰ ovarian cancer, ¹¹ and circulating tumor cells of colorectal cancer patients—where its levels correlate with treatment response and prognosis. ¹² Notably, glioblastoma demonstrates the highest LOXL3 expression among malignant tumors; its aggressiveness drives recurrence and poor prognosis. LOXL3 silencing via small interfering RNA inhibits adhesion and invasion in glioblastoma cells. ¹³ Collectively, LOXL3 contributes to cancer through three primary mechanisms: (a) interaction with Snail to drive EMT-mediated invasion/metastasis in pancreatic cancer; (b) nuclear localization correlating with reduced survival and induced invasion in gastric cancer; and (c) association with mitotic DNA integrity proteins to sustain tumor proliferation.^9,14,15 Although LOXL2 and LOXL3 levels are elevated in PM compared to ovarian/breast cancers or reactive effusions, ¹⁶ the expression and prognostic significance of LOXL3 in PM remain uncharacterized. Therefore, this study aims to elucidate the association between LOXL3 expression and clinicopathological parameters in PM.

Methods

Patients and tissue specimens

A total of 68 PM tissues and 14 non-neoplastic pleural tissues (control) were collected from the Department of Pathology, The People's Hospital of Chuxiong Yi Autonomous Prefecture (Chuxiong, Yunnan, China) between January 2019 and March 2024. Following surgical resection, tissues were immediately snap-frozen in liquid nitrogen within 15 min and stored at −80 °C until analysis. Cases were first diagnosed without previous chemotherapeutic or radiotherapeutic treatments before surgery. The specimens were obtained from 40 males and 34 females, all newly diagnosed through histopathological examination according to WHO 2021 thoracic tumor criteria. Written informed consent had been obtained from all participants under Declaration of Helsinki protocols.

Immunohistochemistry

Tissues were fixed in 10% formalin, embedded in paraffin, and sectioned at 4 µm. After deparaffinization in xylene and rehydration, antigen retrieval was performed by microwave-heating samples in EDTA buffer (pH 6.0) for 30 min. Following hydrogen peroxide incubation, sections were incubated overnight with rabbit anti-LOXL3 polyclonal antibodies (1:200; GeneTex, USA). Subsequently, horseradish peroxidase (HRP)-conjugated secondary antibodies were applied at room temperature for 2 h. Color development was achieved using 3,3ʹ-diaminobenzidine (DAB; Zhongshan Biotech, Beijing, China) for 15 min, followed by differentiation with 1% HCl-ethanol. Sections were counterstained with hematoxylin, dehydrated, and mounted in neutral resin. Analysis was performed using an inverted microscope, with quantification conducted via ImageJ software. In this study, we analyzed 23 cases of PM tissues to investigate the correlation between LOXL3 and the protein expression levels of CD68, CD206, and programmed death ligand 1 (PD-L1). CD68 and CD206 serve as specific marker antibodies for macrophages, with CD68 localized in the cytoplasm and CD206 situated on the macrophage membrane. The intensity of staining is assessed based on the staining characteristics of the target cells. PD-L1 is found on the tumor cell membrane, where any degree of partial or complete membrane staining is deemed positive. The interpretation criteria utilize the Tumor Proportion Score method, defined as the percentage of PD-L1-positive cells relative to all viable tumor cells, expressed as a percentage. Programmed death-protein 1 (PD-1) is predominantly expressed in the cytoplasm of immune cells, such as T cells and B cells, serving as an indicator of the activation state of these immune cells; however, the scoring criteria for PD-1 expression have not yet been standardized. In this study, the proportion of PD-1-positive cells among immune cells is utilized for statistical scoring.

Quantitative reverse transcription polymerase chain reaction

Total RNA was extracted from PM and control specimens using the RNeasy Mini Kit (Qiagen, 74104, Duesseldorf, Germany). Reverse transcription was performed with M-MLV reverse transcriptase (Promega, M1701, Madison, WI, USA) using 2 µg total RNA. Quantitative reverse transcription polymerase chain reaction(qRT-PCR was conducted on an ABI StepOne Plus Real-Time PCR System (Thermo Fisher Scientific, Waltham, MA, USA) with the LightCycler FastStart DNA Master SYBR Green I Kit (Roche Diagnostics, 03515885001, Basel, Switzerland). LOXL3 primers were: sense, 5′-CGGATGTGAAGCCAGGAAACT-3′; antisense, 5′-AGGCATCACCAATGTGGCA-3′. ACTIN was used for normalization (sense primer, 5′-TGACGTGGACATCCGCAAAG-3′; antisense primer, 5′-CTGGAAGGTGGACAGCGAGG-3′). The amplification process involved an initial 1-min step at 95°C, followed by 40 cycles consisting of 10 s at 95°C, 10 s at 58°C, and 15 s at 72°C. The 2^–ΔΔCt method was used for data analysis.

Western blotting

Following protein extraction, protein concentrations were quantified using a bicinchoninic acid (BCA) assay kit (A55860, Thermo Scientific). Equal amounts of total protein (50 µg) were separated by 12% SDS-PAGE and electrophoretically transferred onto PVDF membranes. The membranes were blocked with 5% skim milk in Tris-buffered saline for 1 h at room temperature, followed by overnight incubation at 4°C with a polyclonal anti-LOXL3 primary antibody (1:1000 dilution). Subsequently, membranes were incubated with HRP-conjugated secondary antibodies for 2.5 h at 4°C. Protein bands were visualized using Pierce™ ECL Plus Western Blotting Substrate (32132, Thermo Scientific), and signal intensities were quantified with Image Lab™ Software (v5.2.1, Bio-Rad, USA).

Data acquisition

The clinical and transcriptomic data for pleural mesothelioma were retrieved from The Cancer Genome Atlas (TCGA) database, which encompasses 86 PM tumor samples. After excluding samples with zero or negative survival times and those lacking survival status information, the cohort was narrowed down to 84 PM samples with comprehensive clinical histories. It is important to note that the TCGA database does not include a dedicated control data. To mitigate this limitation, we utilized 58 histologically normal lung tissue samples from TCGA's RNA-seq cohorts as the control group.

Relationship between LOXL3 and clinicopathological characteristics

The data from TCGA and the Chuxiong database were stratified into high and low expression groups according to the optimal cutoff value of LOXL3 expression. The clinicopathological characteristics of PM, including age (>60 or ≤60 years), gender (female or male), tumor (T) stage (T1, T2, T3, T4, TX), node (N) stage (N0, N1, N2, N3, NX), metastasis (M) stage (M0, M1, MX), AJCC stage (I, II, III, IV), and cancer type (biphasic, epithelioid, sarcomatoid), were incorporated into the analysis to evaluate the association between clinicopathological characteristics and LOXL3 expression levels. Additionally, we collected 68 PM samples and 6 control tissues for analysis in the Chuxiong database.

Prognostic analysis

Univariate and multivariate Cox regression analyses were performed to analyze the effect of LOXL3 expression and clinical data of PM patients on patient prognosis. Additionally, we utilized R version 3.6.3 to conduct Kaplan–Meier analysis on the prognosis of PM within the TCGA dataset, and we separately plotted the OS survival function curves. Currently, a substantial body of literature utilizes the optimal cutoff value for survival difference analysis, primarily because it enhances the discrimination and statistical significance of survival curves. Traditional methods, such as using the median as a threshold, often fail to effectively differentiate survival outcomes between groups. In contrast, employing the optimal cutoff value facilitates a more precise categorization of samples into high and low groups, thereby amplifying the significance of the differences in survival curves between these groups. In scenarios with a considerable amount of censored data or minimal intergroup differences, traditional thresholds may lead to nonsignificant survival analysis results. By adopting the optimal cutoff value, statistical methods, such as the log-rank test, can be employed to improve the detection of intergroup differences, ultimately yielding more reliable conclusions.^17,18

Relevance analysis

To evaluate the functional coordination within the lysyl oxidase family, we conducted a Pearson correlation analysis to assess the coexpression patterns of LOX, LOXL1, LOXL2, and LOXL4 mRNA levels using RNA-seq data obtained from the Gene Expression Profiling Interactive Analysis (GEPIA) database. Additionally, to investigate the relationship between LOXL3 and the tumor microenvironment, we employed Pearson correlation analysis to examine the associations between LOXL3 and the protein expression levels of PD-L1, PD-1, CD68, and CD206.

Gene set enrichment analysis (GSEA)

To investigate the functional pathways associated with LOXL3 expression, TCGA cohorts were stratified into high- and low-LOXL3 expression groups using the cutoff value. The Gene Set Enrichment Analysis (GSEA) v4.3.2 was subsequently performed using the c2.cp.KEGG.v2023.1. Hs.symbols.gmt gene set collection from the Molecular Signatures Database (MSigDB). The analysis parameters included 1000 permutations, with statistically significant pathways defined by a |Normalized Enrichment Score| > 1, nominal P-value <0.05, and false discovery rate < 0.25.

Results

LOXL3 mRNA expression was significantly upregulated in pm tissues

TCGA cohort was used to assess the level of mRNA expression of LOXL3 in PM tissues (n = 87) and control tissues (n = 58). As demonstrated in (Figure 1(a)), LOXL3 mRNA expression was significantly elevated in PM tissues compared to control tissues (P < 0.001). Consistently, qRT-PCR analysis demonstrated significantly elevated LOXL3 transcript levels in both sarcomatoid (P < 0.001) and biphasic (P < 0.01) PM subtypes compared to control. In contrast, the epithelioid subtype exhibited no significant difference in LOXL3 expression relative to normal (P > 0.05) (Figure 1(b)).

Figure 1.

LOXL3 mRNA was upregulated in PM. (a) Analysis of LOXL3 mRNA expression in the TCGA cohort; (b) qRT-PCR assays to detect LOXL3 mRNA expression levels in different subtypes of PM. Data were presented as mean ± SD, and comparisons between the two groups were performed by Independent t-test. One-way ANOVA was employed for comparisons of multiple groups. ***P < 0.001, **P < 0.001.

LOXL3 protein expression was significantly upregulated in PM tissues

Western blotting analysis revealed markedly elevated LOXL3 protein levels in sarcomatoid (P < 0.01) and biphasic (P < 0.05) PM subtypes compared to the control. In contrast, Epithelioid subtype exhibited no significant difference in LOXL3 expression relative to normal controls (P > 0.05) (Figure 2(a) and (b)). Consistent with Western blotting findings, IHC analysis revealed the presence of LOXL3 protein and subcellular localization in 68 PM and 14 nearby noncancerous pleural tissue specimens. PM cell nuclei and cytoplasm were the primary locations for LOXL3 protein expression. 64 PM tissue samples showed strong PM staining, while 14 adjacent noncancerous pleural tissue specimens showed weak or negative staining. IHC analysis revealed the distribution and expression level of LOXL3 protein, LOXL3 was highly expressed in sarcomatoid (P < 0.001) and biphasic (P < 0.05) type of PM tissues. However, there was no significant change in expression in PM tissues of epithelioid type (P > 0.05) (Figure 2(c)). Next, the average density scores of IHC were measured and compared between PM tissues and non-neoplastic pleural tissues (Figure 2(d)).

Figure 2.

LOXL3 protein was upregulated in PM. (a) Western blotting evaluation of LOXL3 protein expression levels in PM. (b) Statistical results of protein expression levels in Western blotting. (c) IHC assessment of LOXL3 protein expression levels in PM. (d) Statistical results of protein expression levels in IHC. Data were presented as mean ± SD, and comparisons between the each groups were performed by ANOVA, **P < 0.01, *P < 0.05.

LOXL3 was associated with the clinicopathological characteristics of PM

To delineate the clinicopathological relevance of LOXL3 in PM, we analyzed its associations with key parameters using the TCGA (n = 87) and Chuxiong (n = 68) cohorts. In the TCGA cohort, high-expression LOXL3 was significantly correlated with histologic subtype (χ² = 9.573, P = 0.008) but showed no association with age, sex, TNM classification, or American Joint Committee on Cancer (AJCC) stage (all P > 0.05) (Supplementary Table 1). Similarly, in the Chuxiong cohort, high-expression LOXL3 was strongly linked to Wilms tumor protein 1 (WT-1) (P = 0.003) and histologic subtype (P = 0.003) (Supplementary Table 2).

Prognostic values of clinicopathological characteristic in PM

To evaluate the prognostic value of LOXL3 in PM, we performed univariable and multivariable analyses using Cox proportional hazards regression models. In univariable analysis, histologic subtype emerged as a significant prognostic factor (hazard ratio (HR) = 0.5, 95% confidence interval (CI): 0.3–0.85, P = 0.01), whereas age, sex, TNM classification, AJCC stage, and LOXL3 expression showed no significant association with OS rate. Multivariable analysis adjusted for all baseline covariates confirmed histologic subtype as an independent prognostic indicator (HR = 0.39, 95% CI: 0.19–0.77, P = 0.007) (Supplementary Table 3).

Lower survival in PM patients with high LOXL3 expression

Survival analysis was performed using RNA-seq data from TCGA PM cohort. Using the survminer R package (v0.4.9) patients were categorized into a high expression group (n = 43) and a low expression group (n = 44) by the optimal cutoff value of 545, and survival analyses were performed. Kaplan–Meier survival curves demonstrated significantly reduced OS in the LOXL3-high group compared to LOXL3-low counterparts (P = 0.015, HR = 0.55) (Figure 3).

Figure 3.

Relationship between LOXL3 mRNA expression and OS in PM.

LOXL3 gene expression is associated with LOXL4，PD-L1, CD68, and CD206 genes

To elucidate potential functional synergies within the LOX family, we interrogated co-expression patterns between LOXL3 and other lysyl oxidase members using transcriptomic profiles from the GEPIA database. Spearman correlation analysis revealed a statistically significant positive correlation between LOXL3 and LOXL4 mRNA levels; (R = 0.26, P = 0.013) (Figure 4(a)); however, no associations were detected with LOX (R = 0.18, P = 0.095) (Figure 4(b)), LOXL1 (R = 0.16, P = 0.13) (Figure 4(c)), or LOXL2 (R = 0.2, P = 0.063)(Figure 4(d)). Furthermore, to investigate the relationship between LOXL3 and the tumor microenvironment, we employed Pearson correlation analysis to examine the associations between LOXL3 and the protein expression levels of PD-L1, CD68, CD206, and PD-1. PD-L1 is expressed on the membrane of tumor cells. CD68 and CD206 are specific marker antibodies for macrophages, with CD68 localized in the cytoplasm and CD206 found on the membrane of macrophages. In contrast, PD-1 is primarily localized in the cytoplasm of immune cells, including T cells and B cells (Figure 5(a)). The analysis indicated a statistically significant positive correlation between LOXL3 and the protein levels of PD-L1 (R = 0.2463, P = 0.016) (Figure 5(b)), CD68(R = 0.3009, P = 0.0067) (Figure 5(c)), and CD206 (R = 0.272, P = 0.0107) (Figure 5(d)), while no associations were found with PD-1 (R = 0.1473, P = 0.706) (Figure 5(e)).

Figure 4.

Correlation analysis of LOXL3 with the mRNA expression level. (a) LOX, (b) LOXL1, (c) LOXL2, and (d) LOXL4.

Figure 5.

Correlation analysis of LOXL3 protein expression level. (a) IHC of PD-L1, PD-1, CD68 and CD206 in PM. (b) Correlation analysis of LOXL3 with the protein expression levels of PD-L1, PD-1, CD68 and CD206 in PM.

Functional enrichment analysis of LOXL3

To identify LOXL3-related signaling pathways, TCGA cohort with high- and low-expression of LOXL3 were used for GSEA. High-expression LOXL3 samples were enriched in oxidative phosphorylation (Supplementary Figure 1(a)), estrogen response late (Supplementary Figure 1(b)), estrogen response early (Supplementary Figure 1(c)), and adipogenesis (Supplementary Figure 1(d)). Low-expression LOXL3 samples were enriched in apoptosis (Supplementary Figure 1(e)), IL2-STAT5 signaling (Supplementary Figure 1(f)), mTORC1 signaling (Supplementary Figure 1(g)), and TGF-β pathway (Supplementary Figure 1(h)).

Discussion

As a member of the LOX protein family, LOXL3 plays a critical role in maintaining genomic stability and promoting mitosis, conferring oncogenic potential in cancer. This study discovered that LOXL3 is highly expressed in PM and significantly associated with poor prognosis. This finding not only uncovers the tumor-promoting role of LOXL3 in PM but also suggests its potential as a novel prognostic biomarker.

LOXL3, a member of the lysyl oxidase family, is closely implicated in tumor invasion, metastasis, and immune microenvironment regulation. For instance, in hepatocellular carcinoma (HCC), LOXL3 is highly expressed in tumor tissues of metastatic patients and correlates significantly with larger tumor size, advanced clinical stage, and worse prognosis. Mechanistic studies revealed that LOXL3 interacts with Snail family transcriptional repressor 1 (Snail1) to activate the Wnt/β-catenin/Snail1 signaling pathway, thereby driving EMT. Further in vivo experiments confirmed that LOXL3 silencing markedly suppresses HCC tumor growth. ¹⁹ Notably, LOXL3 also stabilizes dihydroorotate dehydrogenase (DHODH) to inhibit mitochondrial ferroptosis, thereby promoting chemoresistance in HCC cells. ²⁰ Additionally, LOXL3 expression positively correlates with immune cell infiltration and immune checkpoint gene expression in HCC, ²¹ suggesting its potential role in tumor immune evasion. Similarly, in lung cancer, CCAAT/enhancer binding protein alpha (CEBPA) enhances B-cell lymphoma-2 (BCL-2) stability by promoting LOXL2 and LOXL3 transcription, thereby driving in vitro proliferation and metastasis of lung cancer cells. ²² In melanoma, conditional LOXL3 knockout mouse models demonstrated that LOXL3 deficiency significantly prolongs tumor latency and reduces lymph node metastasis. Further mechanistic investigations revealed that LOXL3 regulates cellular plasticity and survival via the SNAIL1-PRRX1 signaling axis to promote melanoma progression. ²³ Clinical data analysis supports this finding: LOXL3 overexpression is associated with poorer prognosis in primary melanoma patients, and its expression level may serve as a prognostic marker. ²⁴ At the molecular level, LOXL3 silencing not only inhibits melanoma cell proliferation but also induces DNA damage response abnormalities, such as dysregulated ATM/ATR checkpoint activation, leading to double-strand break (DSB) accumulation and aberrant mitosis. Importantly, LOXL3 interacts with genome integrity-associated proteins, and its depletion reduces their protein levels and impairs DSB repair efficiency. ¹⁵ Analogous mechanisms have been observed in other cancers: In glioblastoma, LOXL3 is among the most highly expressed tumor-associated proteins, and its silencing enhances adhesion and invasion of U87MG cells. ²⁵ In gastric cancer, low LOXL3 expression suppresses proliferation, migration, EMT, and potentially hinders tumor progression by promoting ferroptosis. ²⁶ In colorectal cancer, desflurane induces EMT and metastasis by dysregulating the miR-34a/LOXL3 axis. ²⁷ Collectively, these findings demonstrate that LOXL3 exerts pro-tumorigenic effects across multiple cancer types by regulating EMT, genomic stability, and metabolic pathways.

The functional roles of LOXL3 extend beyond cancer. For instance, in osteoarthritis models, LOXL3 expression is upregulated in chondrocytes, and its silencing promotes the expression of autophagy-related proteins by suppressing Rheb/p-p70S6 K signaling. ²⁸ Furthermore, LOXL3 knockout mice exhibit perinatal lethality and skeletal developmental abnormalities. ²⁹ In muscle development, LOXL3 enhances integrin activation at myofiber tips through fibronectin oxidation, ensuring proper myofiber positioning. ³⁰ These findings suggest that LOXL3 plays a broad role in tissue homeostasis, and its dysregulation may simultaneously drive fibrotic and carcinogenic progression. Building on this, the LOXL3 inhibitor PXS-5153A has been developed, which significantly improves liver function in fibrosis models and cardiac output post-myocardial infarction by reducing collagen cross-linking, ³¹ highlighting the therapeutic potential of targeting LOXL3.

Based on histological characteristics, PM is classified into epithelioid (60%–70%), sarcomatoid (10%–20%), and biphasic subtypes. ³² Among these, the epithelioid subtype demonstrates slow growth, sensitivity to therapy, and the most favorable prognosis; the sarcomatoid subtype is highly aggressive with extremely poor outcomes; and the prognosis of the biphasic subtype depends on the proportion of its dominant component.^33,34 This study revealed that LOXL3 expression is highest in the sarcomatoid subtype, intermediate in the biphasic subtype, and lowest in the epithelioid subtype, with both TCGA and Chuxiong databases supporting the significant association between LOXL3 expression and PM subtypes. These results suggest that LOXL3 may serve as an auxiliary marker for distinguishing PM subtypes. The WT-1 gene, located on chromosome 11p13, encodes a transcription factor functioning as a tumor suppressor. ³⁵ In PM, WT-1 is aberrantly overexpressed and serves as a critical diagnostic marker. Immunohistochemically, nuclear staining of WT-1 protein is a hallmark feature of mesothelioma, aiding its differentiation from metastatic tumors such as lung adenocarcinoma. ³⁶ Thus, combined detection of LOXL3 and WT-1 may provide a valuable diagnostic strategy for PM. Recent multi-omics approaches have further illuminated the complex molecular landscape of malignant pleural mesothelioma. ³⁷ Integrative analyses combining genomic, transcriptomic, and proteomic data have identified key signaling hubs and potential therapeutic vulnerabilities in PM. For instance, studies leveraging multi-omics platforms have revealed aberrant activation of oxidative phosphorylation and estrogen response pathways—findings consistent with our GSEA results regarding LOXL3 high-expression phenotypes. These approaches not only enhance our understanding of tumor heterogeneity but also facilitate the identification of robust biomarker panels. Incorporating LOXL3 into such multi-omics frameworks may improve prognostic stratification and reveal synergistic targets for combination therapies. To clarify the prognostic value of LOXL3, univariate/multivariate Cox regression and Kaplan–Meier survival curve analyses were performed. Although LOXL3 expression showed no statistically significant correlation with prognosis, patients with high LOXL3 expression exhibited significantly lower survival rates. This discrepancy underscores the need for future studies to expand sample sizes and refine analytical approaches to validate the prognostic significance of LOXL3.

Finally, through correlation analysis and GSEA, this study uncovered potential mechanisms of LOXL3 in PM: LOXL3 expression positively correlates with LOXL4, and high LOXL3 expression is enriched in oxidative phosphorylation, estrogen response, and adipogenesis pathways, while the low-expression group is enriched in apoptosis, IL-2/STAT5, mTORC1, and transforming growth factor-β (TGF-β) pathways. These pathways have been corroborated in other studies. For example, LOXL3 promotes extracellular matrix remodeling via TGF-β signaling-mediated collagen cross-linking. ³⁸ The miR-449a-5p/LOXL3/mTOR axis drives fibrotic progression. ³⁹ Estrogen may upregulate LOXL3 expression in mice and Ishikawa cells through the TGF-β pathway, ⁴⁰ while combined estrogen–progesterone treatment in ovariectomized rat models significantly increases LOXL3 expression in vascular tissues. ⁴¹ Notably, LOXL3 expression is inversely correlated with ER/PR-positive status and is elevated in hormone receptor-negative or highly aggressive subtypes. ⁴² Moreover, the emergence of mass spectrometry-based plasma/serum proteomics offers a promising non-invasive approach for biomarker discovery and validation in PM. Recent studies have demonstrated the utility of high-throughput proteomic profiling in identifying circulating proteins associated with tumor burden and treatment response. For example, LOXL3, as a secreted glycoprotein, may be detectable in plasma or serum, providing a minimally invasive means for monitoring disease progression or therapeutic efficacy. Future studies integrating tissue-based findings with circulating proteomic data could enhance the clinical applicability of LOXL3 as a biomarker and facilitate its translation into routine diagnostic and prognostic workflows. ⁴³

PD-L1 is an immune checkpoint molecule that inhibits T-cell activity by binding to the PD-1 receptor, thereby assisting tumor cells in evading attacks from the immune system. LOXL3 plays a crucial role in the remodeling of the extracellular matrix and the regulation of the tumor microenvironment, which is closely linked to tumor invasion and metastasis. The observed positive correlation between PD-L1 and LOXL3 in PM suggests a potential synergistic role for these two molecules in tumor immune evasion and progression, particularly pronounced in sarcomatoid mesothelioma, aligning with previous literature reports. ⁴⁴ The elevated expression of LOXL3 may enhance PD-L1 expression by modifying the tumor microenvironment, thereby further augmenting the immune evasion capabilities of tumor cells. ⁴⁵ Additionally, LOXL3 may indirectly upregulate PD-L1 expression by activating specific signaling pathways, such as the Hippo-YAP pathway. ⁴⁶ Research has demonstrated that high PD-L1 expression is significantly associated with shorter OS and progression-free survival. ⁴⁷ The elevated expression of LOXL3 may further exacerbate this trend, indicating that the combined detection of these two markers could enhance the accuracy of predicting clinical outcomes in patients. Immune checkpoint inhibitors targeting PD-L1, such as anti-PD-1/PD-L1 antibodies, may demonstrate greater efficacy in patients exhibiting high LOXL3 expression. PD-1, an immune checkpoint molecule, inhibits T-cell activity by binding to its ligand PD-L1, thereby facilitating tumor evasion of immune system attacks. However, this study did not identify a significant association between PD-L1 and LOXL3. CD68 serves as a specific marker for macrophages, primarily utilized to identify these immune cells within various tissues. In the tumor microenvironment, macrophages exhibit a dual role; they not only promote tumor growth but also contribute to the suppression of tumor progression through their immune responses. In mesothelioma characterized by high CD68 expression, macrophages typically exhibit an M2 phenotype (e.g., CD163+/CD206+), which possesses immunosuppressive functions that inhibit the anti-tumor activity of T cells. ⁴⁸ High CD68 expression is often associated with the expression of other immune checkpoint molecules, such as PD-L1. For instance, the co-expression of CD68 + macrophages and PD-L1 may further exacerbate immunosuppression, leading to a poorer response to chemotherapy or immunotherapy. ⁴⁹ CD206, a marker of M2-type macrophages, is generally associated with immunosuppressive and tumor-promoting functions. M2-type macrophages facilitate tumor angiogenesis and invasion through the secretion of cytokines and growth factors. The elevated expression of CD68 and CD206 indicates a significant presence of macrophages in mesothelioma, particularly M2-type macrophages. These cells may promote the expression of LOXL3 through cytokine secretion or direct interaction with tumor cells. The positive correlation between CD68, CD206, and LOXL3 observed in this study is more pronounced in sarcomatoid mesothelioma, suggesting that macrophages may play a pivotal role in mesothelioma progression, correlating with tumor malignancy and promoting tumor invasiveness by regulating LOXL3 expression. The molecular mechanisms underlying LOXL3 and its crosstalk with other signaling pathways in PM have yet to be fully elucidated. Our future research will delve deeper into the role of LOXL3 in PM and its potential diagnostic significance. Although LOXL3's role in PM was first demonstrated in the present work, the underlying mechanism remains unknown. Large trials with long-term follow-up and basic cell culture and animal assays are warranted for confirming LOXL3's function in PM. There were several limitations in this study. PM is a rare tumor with a poor prognosis. The significance of this observation was limited by the small sample size, especially sarcomatoid malignant pleural mesothelioma. Thus, a further increase in sample size would help unravel the complex genetic architecture of LOXL3 and PM.

Conclusions

LOXL3 exhibits elevated expression levels in PM and is linked to a poor prognosis, playing a role in the evasion of tumor immunity. There is a correlation between the expression level of LOXL3 and various cancer types, as well as with the expression level of WT-1. Additionally, the type of cancer serves as an independent prognostic factor for PM. A positive correlation exists between LOXL3 and LOXL4 expression levels, where high LOXL3 expression is found to be enriched in pathways related to oxidative phosphorylation, estrogen response, and adipogenesis. In contrast, the low-expression group shows enrichment in apoptosis, IL-2/STAT5 signaling, mTORC1, and TGF-β pathways. Consequently, LOXL3 may serve as both a prognostic marker and a possible therapeutic target for PM.