LAR1 promotes breast carcinogenesis by activating NF-κB signaling pathway through binding and enhancing APOC1 expression.

Introduction
Breast cancer (BC) is responsible for the most cancer-related deaths among women globally.1 As the most common cancer in women, it results in millions of new cases and over 400,000 deaths each year.2 While early-stage BC has a high survival rate, patients diagnosed with advanced BC experience a significantly lower survival rate due to metastasis and recurrence.3,4,5 The current therapeutic options for BC encompasses surgery, chemotherapy, radiotherapy, endocrine therapy, and targeted therapy. However, these interventions frequently give rise to severe side effects, including limb dysfunction and pain.6 The poor prognosis associated with advanced BC and the limitations of existing treatments, underscores the critical need to explore novel therapeutic strategies to enhance patient survival and quality of life.
La Ribonucleoprotein 1 (LAR1) is a member of the LARP family that plays a role in regulating both transcription and translation.7 Previous reports have established the significant involvement of LAR1 in tumor progression. Specifically, the elevated LAR1 expression has been observed in non-small cell lung cancer, where overexpression correlated with enhanced cell migration and invasion.8 Conversely, in ovarian cancer, LAR1 was found to be implicated in promoting tumor progression and inducing resistance to chemotherapy, and knockdown of LAR1 in ovarian cancer cells led to decreased cell viability and increased apoptosis.9 Furthermore, database analysis reveals that breast tumors exhibited higher LAR1 expression compared to normal tissues. However, the precise role of LAR1 in BC remains to be fully elucidated.
Apolipoprotein C1 (APOC1) is an apolipoprotein with a small size that is crucial for lipid metabolism.10,11 Recent researches have increasingly linked APOC1 to cancer progression. It has been identified as a contributor to the tumorigenicity of glioblastoma.12 Higher APOC1 expression was relevant to poorer prognosis in patients with renal cell carcinoma, and APOC1 exacerbated malignant behavior of cancer cells.13 More importantly, the APOC1 expression was elevated in BC tissues compared to adjacent normal tissues, and APOC1 knockdown inhibited proliferation, migration, and invasion of BC cells.14 Experimental evidence from the RIP-CHIP assay suggested that LAR1 bound to the APOC1 mRNA.8 Additionally, our mRNA-sequencing identified APOC1 as a key gene differentially expressed in LAR1-overexpressing BC cells, suggesting a potential mechanism whereby LAR1 promotes tumor progression through the regulation of APOC1.
In the study, we investigated the effects of LAR1 on BC cells and determined its BC-promoting effects. Further, LAR1 aggravated malignant progression of BC cells by regulating APOC1 expression. Our findings propose that LAR1 served as a potential therapeutic target for alleviating BC progression.

Results

High LAR1 expression was associated with BC progression
To identify a molecular driver of BC pathogenesis, we performed analysis for GSE113230 dataset and identified a total of 7,840 differentially expressed genes (DEGs) in BC samples compared to adjacent normal tissues, with 3,995 upregulated and 3,845 downregulated (Figure 1A). The expression profiles of these DEGs are displayed in a heatmap (Figure 1B). Next, Gene Ontology-Molecular Function (GO-MF) enrichment analysis showed that TOP 10 pathways enriched by upregulated DEGs (Figure 1C) were more closely related to cancer development than those enriched by downregulated DEGs. Especially, cadherin switching is an important characteristic of epithelial-to-mesenchymal transition, a critical step in cancer progression.15 From the genes within this pathway, we selected LAR1 for further investigation, based on its potential roles in cadherin binding and its RNA-binding characteristics. The LAR1 expression in GSE113230 is shown in a boxplot, which indicated that BC samples had a higher LAR1 expression than adjacent normal tissues (Figure 1D). TNMplot analysis showed that the LAR1 expression in various tumor samples, including BC, was higher than that in normal tissues (Figure 1E). UALCAN analysis also verified highly expressed LAR1 mRNA and protein in BC samples compared with normal samples (Figures 1F and 1G). GEPIA analysis demonstrated that higher LAR1 expression correlated with poorer overall survival in BC patients (Figure 1H). Further investigation using GSEA-KEGG suggested that LAR1 activated the “breast cancer” pathway (Figure 1I). GSEA-GO results indicated that LAR1 was involved in pathways related to cancer development, including positive regulation of cell cycle, regulation of cell growth, regulation of DNA replication, negative regulation of intrinsic apoptotic signaling pathway (Figure 1J). Next, we analyzed the correlation between the LAR1 expression and some clinical features of BC samples using The Cancer Genome Atlas (TCGA) data. We observed that BC samples had a higher LAR1 expression relative to their normal controls (Figure 1K). Furthermore, there were variations in LAR1 expression across different BC subtypes, with the highest levels observed in the LumB subtype (Figure 1L). Univariate Cox analysis revealed that age and tumor stage were independent prognostic risk factors for BC patients. However, the LAR1 expression did not emerge as an independent prognostic risk factor (Figure 1M). More importantly, results from our analysis using the GEPIA database confirmed that higher LAR1 expression corresponded to poorer prognosis. We speculated that this discrepancy might be attributed to the fact that the survival analysis in GEPIA includes a larger sample size. All these results suggested the strong association between the LAR1 expression and BC progression.

LAR1 promotes cell proliferation of BC cells
Next, we investigated the expression pattern of LAR1 in BC cells. The results of qRT-PCR showed that the mRNA expression of LAR1 was higher in BC cells compared to normal breast epithelial cells (Figure S1A). The upregulated LAR1 expression in cancer cells was re-confirmed by WB assays (Figure S1B). Given the similar expression profile of LAR1 in BC cells, we randomly selected MDA-MB-231 and BT-549 cells for further experiments. Following that, gain- and loss-of-function of LAR1 were achieved in MDA-MB-231 and BT-549 cells, respectively, using the lentivirus vectors. The effectiveness of LAR1 knockdown was confirmed by qRT-PCR, showing a 79% knockdown efficiency for shLAR1-1 vs. shNC and 77% knockdown efficiency for shLAR1-2 vs. shNC in MDA-MB-231. For BT-549 cells, the efficiencies were 80% for shLAR1-1 vs. shNC and 79% for shLAR1-2 vs. shNC. The overexpression of LAR1 was confirmed with a 9.32-fold increase for oeLAR1 vs. Vector in MDA-MB-231 cells and a 9.22-fold increase in BT-549 cells (Figure S1C). Western blot assays further validated the effective knockdown and overexpression of LAR1 in the BC cells (Figure S1D). CCK-8 assays suggested that LAR1 knockdown weakened breast cell proliferation but its overexpression exhibited the opposite effects (Figure 2A). Colony formation assays also supported the pro-proliferative effect of LAR1, as evidenced by the increased colony formation rate in LAR1-overexpressing BC cells (Figure 2B). Cell cycle is an important regulator of cell proliferation. By performing flow cytometry, we found that LAR1 knockdown induced more cells arrest at G1 phase, whereas less cells at S phase. By contrast, the proportion of cells in G1 phase was remarkably decreased, while it was increased in S phase after overexpression of LAR1 (Figures 2C and 2D). These results showed that LAR1 promoted proliferation of BC cells by inducing cells arrest at G1 phase. Furthermore, we selected another BC cell line, MCF-7, to further support our findings. Effective LAR1 knockdown and overexpression were confirmed by the qRT-PCR results (an 82.55% knockdown efficiency for siLAR-1 vs. siRNA, 81.2% knockdown efficiency for siLAR-2 vs. siRNA, and a 13.56-fold overexpression efficiency for oeLAR1 vs. Vector) (Figure S3A). Protein expression of LAR1 following transfection of siLAR1s or oeLAR1 displayed a similar trend (Figure S3B). Additionally, CCK-8 assay determined that LAR1 promoted cell proliferation of MCF-7 cells (Figure S3C).

LAR1 encourages cell migration and invasion of BC cells and is negatively regulated by miR-143-3p
Subsequently, Transwell migration assays indicated that LAR1 knockdown decreased number of cells migrating across the Transwell membranes. On the contrary, LAR1 overexpression resulted in the opposite trend. Correspondingly, results of Transwell invasion assays showed that LAR1 enhanced the invasion ability of BC cells (Figure 3). Furthermore, MMP9 is important contributor to cell migration and invasion.16 The chemotactic factor SDF-1 can stimulate migration and invasion of cancer cells.17 WB assays verified that LAR1 knockdown inhibited the protein expression of active MMP9 and SDF-1. On the contrary, LAR1 overexpression resulted in the upregulation of these proteins (Figure S2). Additionally, LAR1 promoted the migration of MCF-7 cells (Figure S3D) Taken together, these results demonstrated that LAR1 overexpression aggravated the migration and invasion of BC cells.
Moreover, a previous study has determined the suppressing role of miR-143-3p in BC progression.18 We found the binding site of miR-143-3p in the 3′-UTR fragments of the LAR1 mRNA. To investigate whether and how miR-143-3p would regulate the effects of LAR1 on BC cells, MDA-MB-231 cells were transfected with the inhibitor (78.28% transfection efficiency) or miR-143-3p mimic (7.67-fold transfection efficiency) (Figure S4A). The qRT-PCR results showed that the LAR1 mRNA expression was increased in cells transfected with the miR-143-3p inhibitor but decreased in cells transfected with the miR-143-3p mimic (Figure S4B). The starBase database also indicated the negative correlation between the miR-143-3p and LAR1 expression (Figure S4C). By performing the luciferase reporter assay, we found that the miR-143-3p mimic lessened the relative luciferase activity of 3′-UTR-WT fragments of the LAR1 but had no effect on the 3′-UTR-MUT fragments (Figure S4D). CCK-8 assays demonstrated that the miR-143-3p mimic reduced the viability of BC cells, which was reversed by LAR1 overexpression (Figure S4E). Furthermore, Transwell assays revealed that LAR1 overexpression rescued the decrease in the number of migrated cells caused by the miR-143-3p mimic. The inhibitory effect of the miR-143-3p mimic on BC cell migration was counteracted by LAR1 overexpression to some extent (Figure S4F). These results indicated that miR-143-3p was involved in BC progression by regulating the LAR1 expression.

LAR1 facilitates xenograft tumor growth and liver metastasis in nude mice
We further assessed the role of LAR1 in tumor growth using xenograft models in nude mice. In detail, mice were injected with BC cells to initiate tumor formation (Figure 4A). Consistent with the in vitro results, the tumors in mice injected with LAR1-knockdown cells were smaller than those in mice receiving negative control cells (Figure 4B). Notably, the prominently reduced tumor volumes were observed in mice in the shLAR1-1 group from the 20th day and in mice in the shLAR1-2 group from the 24th day (Figure 4C). Additionally, LAR1 knockdown led to a decrease in the tumor weights of nude mice treated with BC cells (Figure 4D). Subsequent immunofluorescence staining and quantitative analysis showed that LAR1 knockdown decreased the percentage of Ki67-positive cells in tumor sections (Figures 4E and 4F). Collectively, these results indicated that LAR1 knockdown inhibited tumor growth in nude mice transplanted with human BC cells.
Liver metastasis is one of the most common site of metastasis in patients with BC.19 Thus, we utilized the in vivo model to examine the LAR1’s effect on BC cell metastasis (Figure 4G). Bioluminescence imaging system showed that mice injected with the BC cells with LAR1 knockdown had the significantly decreased metastatic lesions. In contrast, those injected with cells that stably overexpressed LAR1 showed the opposite trend (Figure 4H). Hematoxylin-eosin (HE) staining was used to evaluate liver metastasis in mice following BC cell injection. Images of HE staining showed that LAR1 knockdown inhibited the liver metastasis, whereas its overexpression exerted the opposite effects (Figure 4I). Consistent with this, quantitative analysis showed that LAR1 increased the number of nodules in the liver tissues of mice (Figure 4J). These results suggested that LAR1 promoted the liver metastasis of BC cells in vivo.

APOC1 might be a downstream target of LAR1
To elucidate the mechanistic impact of LAR1 on BC malignancy, we profiled the transcriptomes of LAR1-overexpressing cells and vector controls via mRNA-sequencing. The principal component analysis (PCA) demonstrated a distinct segregation of the two groups, underscoring substantial differences in their global gene expression profiles (Figure 5A). We then identified DEGs (|Log2FC| > 1.5 and p < 0.01) and presented them in a volcano plot (Figure 5B). The expression levels of these DEGs were subsequently visualized in a heatmap (Figure 5C). As an mRNA binding protein, LAR1 forms the complex with the 40S ribosome subunit to stabilize its target proteins.20 To identify downstream targets of LAR1, we overlapped DEGs from our mRNA-sequencing data with a previously published LAR1-mRNA interactome. This identified 14 upregulated and 194 downregulated shared DEGs (Figure 5D). GO enrichment analysis revealed that the upregulated DEGs were enriched in cancer-related pathways, including cellular detoxification, apoptotic mitochondrial changes, and cell redox homeostasis (Figure 5E). Based on their stronger association with these tumor-promoting processes, we focused our subsequent analysis on the 14 upregulated DEGs. To prioritize candidates from this gene set, we ranked them by their Log2FC values. The top four genes were excluded since that there are relatively few reports on the involvement of these genes in cancer development, leading us to select APOC1, which ranked fifth. Additionally, APOC1 was able to regulate the development of BC.14 GO analyses for both Biological Process (BP) and MF terms revealed a strong association between APOC1 and lipid metabolism, processes known to be critically involved in BC development (Figure 5F). In view of these findings, we hypothesized that APOC1 acts downstream of LAR1 in BC progression.

LAR1 binds to the 3′-UTR region and stabilize the APOC1 mRNA
By analyzing the TCGA BC data, we found that the APOC1 expression was highly expressed in tumor samples (Figure 6A). The APOC1 expression was various in different BC subtypes (Figure 6B). Univariate Cox analysis revealed that age and tumor stage were independent prognostic risk factors for BC patients. However, the APOC1 expression did not emerge as an independent prognostic risk factor (Figure 6C). Notably, Gao et al. performed Kaplan-Meier survival analysis and found that BC patients with higher APOC1 expression had a poorer overall survival.21 We speculated that this discrepancy might be attributed to the fact that the survival analysis in Kaplan-Meier survival analysis includes a larger sample size. Subsequently, we performed the RNA Immunoprecipitation-Polymerase Chain Reaction (RIP-PCR) assay and discovered that LAR1 bound to the APOC1 mRNA (Figure 6D). Luciferase reporter assay revealed that LAR1 overexpression markedly enhanced the relative luciferase activity of the pGL3 vector encoding the 3′-UTR fragments of the APOC1 mRNA (Figure 6E). Accordingly, PCR and WB assays showed that LAR1 knockdown inhibited the APOC1 expression, but its overexpression promoted the APOC1 expression (Figures 6F and 6G). LAR1 also upregulated the APOC1 expression in MCF-7 cells (Figures S3E and S3F). We also monitored the APOC1 mRNA stability in LAR1-overexpressing cells and vector control cells upon actinomycin D treatment. LAR1 overexpression prolonged the half-life of APOC1 mRNA, promoting the mRNA stability of APOC1 compared with the vector control (Figure 6H). These results demonstrated that LAR1 regulates the APOC1 expression by enhancing the post-transcriptional activity and mRNA stability of APOC1.

LAR1/APOC1 activated the NF-κB signaling pathway in BC cells
Next, we explored the specific mechanism by which APOC1 regulates cancer cells. A previous study indicated that knockdown of APOC1 took part in suppressing AKT activation.22 Several studies have shown that activation of AKT leads to inhibitor of κB kinase (IKK)-dependent IκBα phosphorylation and subsequent degradation, resulting in the nuclear translocation of nuclear factor κB (NF-κB).23,24 Moreover, APOC1 has been reported to activate the NF-κB pathway,25,26 which is closely associated to the development of various cancers, including BC.27 Therefore, we hypothesized that AKT activation and subsequent NF-κB activation might be the mechanism where LAR1/APOC1 promoted the malignant behavior of BC cells. Results of western blot assays showed that LAR1 knockdown inhibited AKT and IKK activation, suppressed the phosphorylation and degradation of IκBα, and reduced p65 phosphorylation (Figures 7A and 7B). Additionally, immunofluorescence staining and the corresponding quantitative analysis revealed that LAR1 knockdown inhibited p65 nuclear translocation, as demonstrated by the reduced nuclear p65-positive cells, inactivating the NF-κB signaling pathway in BC cells. However, overexpression of LAR1 exhibited the opposite effects (Figures 7C and 7D). Importantly, LAR1 also promoted the AKT and subsequent NF-κB activation in MCF-7 cells (Figures S3G–S3I). To further evaluate whether the NF-κB signaling pathway was involved in the role of LAR1, the NF-κB inhibitor, PDTC, was used to suppress NF-κB activation in BC cells. CCK-8 assay showed that the increase in cell viability induced by LAR1 overexpression was reversed by PDTC (Figure 7E). Consistently, PDTC treatment reduced the invasion ability of LAR1-overexpressing cells (Figure 7F). These resulted indicated that LAR1 affected BC progression by activating the NF-κB signaling pathway.
Further, APOC1 overexpression was achieved in MDA-MB-231 cells via transfection of the overexpressing plasmid. The increased APOC1 expression confirmed the successful transfection with a 7.84-fold increase in the APOC1 mRNA expression (Figures 8A and 8B). Lentiviral vector encoding the shLAR1-1 was used for the subsequent experiments. CCK-8 assays showed that APOC1 overexpression enhanced the cell viability of cells with LAR1 knockdown (Figure 8C). Transwell assays revealed that APOC1 overexpression conversed inhibitory effects of LAR1 knockdown on migration and invasion of BC cells (Figure 8D). The reduction of p65 nuclear translocation in LAR1-silenced cells was reversed by APOC1 overexpression (Figures 8E and 8F). In short, these findings indicate that APOC1 overexpression might partially mediates the tumor-suppressing effects of LAR1 knockdown.

Discussion
Our investigation establishes LAR1 as a driver in BC progression. Functional studies confirmed that elevated LAR1 expression in BC cells promotes tumorigenesis in vitro and in vivo. We further determined the regulatory mechanisms governing LAR1, identifying it as a direct target of the tumor-suppressive miR-143-3p. The function of LAR1 is achieved through its binding and transcriptional upregulation of APOC1. This LAR1/APOC1 axis was found to promote BC progression by activating the NF-κB signaling pathway. In conclusion, our findings elucidate a miR-143-3p/LAR1/APOC1/NF-κB regulatory mechanism that drives BC development, highlighting its significant therapeutic potential.
LAR1 is the largest member of the LARP family that regulates the transcriptional and translational processes.7 A body of literature supports that upregulated LAR1 expression was implicated in different cancers, including hepatocellular carcinoma,28 non-small cell lung carcinoma,29 and prostate cancer.30 In line with these studies, we also found increased LAR1 expression in BC cells and determined its tumor-promoting effects on BC progression. Cell proliferation is severely influenced by cell cycle phases. Importantly, LAR1 was evidenced to assist in cell cycle initiation and progression.31 Our results indicated that LAR1 knockdown induced G1-phase cell-cycle arrest. In the G1 phase, cells undergo protein synthesis, while in the S phase, cells undergo DNA synthesis.32 Blockage of the transition from the G1 phase to the S phase inhibits cell proliferation.33 Therefore, the blockade of this transition provides a possible explanation for the anti-proliferative effect observed upon LAR1 knockdown. In addition, Burrows et al. revealed that loss of LAR1 resulted in decreased lamellipodia in human cervical cancer HeLa cells.34 The formation of lamellipodia plays a crucial role in cell migration and invasion.35 Consistently, Burrows et al. reported the inhibitory role of the loss of LAR1 in cell migration.34 Similar effects of LAR1 on cell migration were raised in BC cells in the present study. Both previous reports and our findings supported the role of LAR1 knockdown in decreasing cell proliferation, migration, and invasion. Furthermore, IκBα binds to p65 and inhibits its phosphorylation and nuclear translocation.36 After being phosphorylated, IκBα is degraded followed by the release of NF-κB.37 Our results revealed that LAR1 promoted this process by activating AKT and IKK, leading to IκBα phosphorylation and degradation and subsequent p65 nuclear translocation. Consistent with the critical role of the NF-κB signaling pathway in cancer cell proliferation,38 we found that the NF-κB inhibitor reversed the effects of LAR1 on BC cells.
Non-coding miRNAs can negatively regulate the expression of target mRNAs.39 The miR-143-3p is the main form of the miR-143, a part of the miR-143/miR-145 cluster.40 Previous studies have announced its tumor suppressor effects on the development of different types of cancers. It was identified as an anti-oncomir in gastric cancer,41 colorectal cancer,42 and pancreatic ductal adenocarcinoma.43 Especially, the inhibitory role of miR-143-3p in BC progression, demonstrated by decreased cell proliferation, migration, and invasion, is well-established.44,45 Consistent with these findings, we confirmed miR-143-3p inhibited these malignant phenotypes in BC cells. Given the predicted binding site for miR-143-3p in the 3′-UTR fragments of LAR1, we investigated their relationship and determined that miR-143-3p bound to the 3′-UTR fragments of LAR1 and inhibited its expression. Furthermore, overexpression of LAR1 partially rescued the changes in phenotypes of BC cells caused by the miR-143-3p mimic, indicating that LAR1 is involved in the effects of miR-143-3p on the development of BC.
As an RNA-binding protein, LAR1 has been reported to regulate the stability and translational efficiency of mRNAs.46 To continue exploring the potential mechanisms by which LAR1 affects the development of BC, LAR1-overexpressing and control vector cells were subjected to mRNA sequencing to identify DEGs. Subsequently, the LAR1-targeted mRNAs were downloaded from previously published literature. By intersection analysis, APOC was ultimately determined as a possible downstream target gene of LAR1 in BC progression. APOC1 took part in the development of metabolic diseases such as Alzheimer’s disease.47 In recent years, increasing evidence revealed the role of APOC1 in cancer progression. Notably, APOC1 was found to promote the migration and invasion of BC cells. Similar to these previous findings, our results revealed the interaction between LAR1 and APOC1 in BC cells. By binding to the 3′-UTR fragments of APOC1, LAR1 stabilized and upregulated the APOC1 mRNA stability and expression. Previous studies have indicated that APOC1 activates the NF-κB signaling pathway. In this study, LAR1 was found to upregulate APOC1 and activate the NF-κB signaling pathway, which is consistent with this trend. APOC1 overexpression rescued the inhibitory role of LAR1 knockdown in the malignant behavior and NF-κB activation in BC cells. It is worth noting that other genes could be bound and regulated by LAR1 in BC cells. Notably, results of GO functional enrichment analysis indicated that several down-regulated genes were enriched in the NF-κB signaling pathway. Among these genes, we noticed the gene MAPKBP1, which was enriched in both the “negative regulation of canonical NF-κB signal transduction” and “canonical NF-κB signal transduction” pathways. Consistently, the GEPIA website indicated that MAPKBP1 was significantly downregulated in BC samples compared with their controls, suggesting that LAR1 might be involved in BC progression by regulating MAPKBP1/NF-κB signaling pathway. Our study cannot rule out whether LAR1 would promote BC progression by regulating other genes.
In conclusion, our findings revealed that LAR1 is a key promoter of the development of BC. We establish a novel mechanism whereby LAR1 post-transcriptionally upregulates APOC1, leading to subsequent activation of the NF-κB signaling pathway. This LAR1/APOC1/NF-κB axis represents a promising new target for therapeutic intervention of BC. There are limitations existed in the present study: (1) given that the number of female BC patients is relatively high and the research focus is on exploring the function of LAR1 in BC progression, only female BALB/c nude mice were used for in vivo tumor formation and metastasis experiments. However, using both genders would confirm whether the role of LAR1 in BC progression is influenced by sex-specific factors, providing a more comprehensive perspective. (2) LAR1-overexpressing cells were used for mRNA sequencing to explore the downstream targets of LAR1, while using LAR1-knockdown cells would more directly simulate the situation where LAR1 is targeted to alleviate the malignant phenotype of BC, providing more specific evidence for clinical translation. (3) The findings suggested that LAR1’s effects might not be exclusively mediated by APOC1, as other targets like MAPKBP1 were identified. To comprehensively elucidate the mechanism of LAR1, it is essential to perform gain- and loss-of-function studies on these putative target genes. Assessing their impacts on the malignant phenotype of BC cells and their specific involvement in the LAR1-modulated NF-κB pathway will clarify whether LAR1 functions through a multi-target network. (4) Although our data indicated that APOC1 promoted AKT phosphorylation, the precise molecular mechanism remains undetermined. Further investigation is required to distinguish whether APOC1 directly interacts with AKT or modulates its phosphorylation indirectly, for instance, through the regulation of phosphatidylinositol (3,4,5)-trisphosphate (PIP3) levels and subsequent AKT recruitment to the plasma membrane. (5) The functional involvement of APOC1 in LAR1-mediated effects was established solely through in vitro rescue experiments. For robust validation of the LAR1/APOC1 axis, future studies should include in tumor xenografts using nude mice, to observe the effects of LAR1 overexpression coupled with APOC1 knockdown on tumor growth and progression. More experimental investigations need to be performed for further exploration in the further study.

Limitations of the study
This study only included female animals for in vivo functional verification of LAR1. Incorporating all genders in future research would provide a more comprehensive understanding for the function of LAR1 in BC progression. Our study identified potential downstream targets of LAR1 in BC through mRNA sequencing in overexpressing cells, though future work using LAR1-knockdown models would better simulate a therapeutic context. The findings suggest LAR1 likely acts via a multi-target network, necessitating gain- and loss-of-function studies on these genes to define their specific roles in the malignant phenotype and the LAR1-modulated NF-κB pathway. While APOC1 was shown to promote AKT phosphorylation, the exact mechanism—whether through direct interaction or indirect modulation—remains unknown and requires further investigation. Finally, to robustly validate the LAR1/APOC1 axis, the in vitro rescue findings must be confirmed by in vivo studies using tumor xenografts to assess the impact of LAR1 overexpression with concurrent APOC1 knockdown on tumor growth and progression.

Resource availability

Lead contact
Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Zitao Li (lizt_li@163.com).

Materials availability
This study did not generate new unique materials or reagents.

Data and code availability

•The shared genes between DEGs from mRNA-sequencing and the previously published LAR1-mRNA interactome has been deposited in the Dryad database (https://doi.org/10.5061/dryad.s4mw6m9kn).

•Original western blot images, microscopic images/data reported here will be shared by the lead contact upon request.

•This paper does not report original code.

•Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

Acknowledgments
This work was supported by the 10.13039/501100001809National Natural Science Foundation of China (82171966); Torch Program of Mudanjiang Medical University (Applied Research Project, 2022-MYHJ-007); Young Innovative Talents Training Program of Regular Undergraduate Colleges and Universities in Heilongjiang Province (UNPYSCT-2020065); Basic Scientific Research Business Fee Project (General Natural Science Project, 2021-KYYWF-0472); and Torch Program of Mudanjiang Medical University (Humanities and Social Sciences Talent Cultivation Project, 2022-MYHJ-020).

Author contributions
J.Y.X., conceptualization, methodology, writing – review and editing; Q.L., Z.L., and L.X.K., investigation, formal analysis; Y.S. and Y.L.X., visualization; M.D. and S.J.Z., writing – original draft; Z.T.L., methodology, supervision, writing – review and editing; C.J.L., conceptualization, supervision, writing – review and editing.

Declaration of interests
The authors declare that they have no competing interests.

STAR★Methods

Key resources table

Experimental model and subject details

Cell culture
Seven BC cell lines (MCF-7, BT-549, T-47D, MDA-MB-453, MDA-MB-231, SK-BR-3, MDA-MB-415) and one normal breast epithelial cell line (MCF-10A) (iCell Bioscience Inc., Shanghai, China) were used in the present study. MCF-7 cells were cultured with Minimum Essential Medium (Cat#41500, Solarbio, Beijing, China) supplemented with 10% fetal bovine serum (FBS) (Cat#11011-8611, Tianhang Biotechnology Co., LTD, Huzhou, China). BT-549 cells were cultured with Roswell Park Memorial Institute-1640 Medium (Cat#31800, Solarbio, Beijing, China) supplemented with 10% FBS. T-47D cells were cultured with Dulbecco’s Modified Eagle Medium (Cat#G4511, Servicebio, Wuhan, China) supplemented with 10% FBS. MDA-MB-453 and MDA-MB-231 cells were cultured with L-15 Medium (Cat#G4570, Servicebio, Wuhan, China) supplemented with 10% FBS. SK-BR-3 cells were cultured with McCOY’s 5A (Cat#G4540, Servicebio, Wuhan, China) supplemented with 10% FBS. MCF-10A cells were cultured with MCF-10A Special Medium (Cat#iCell-h131-001b, iCell Bioscience Inc., Shanghai, China). All cells were cultured in a 37°C incubator. Cell lines have been authenticated by Short Tandem Repeat (STR) profiling and tested to be free of mycoplasma contamination by PCR method.

Mouse models
Four-week-old female BALB/c nude mice were obtained from Huachuang Sino (Taizhou, China). Animal experiments were approved by Laboratory Animal Welfare and Ethics Committee of Mudanjiang Medical University (approval number: 20231115-15) and conducted following the NIH Guidelines for the Care and Use of Laboratory Animals.

Method details

Dataset analysis
GSE113230 (3 BC samples and 3 adjacent normal tissues) was downloaded from the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/). Subsequently, differentially expressed genes (DEGs) were selected from the dataset according to the criteria |Log2FC| > 1 and p < 0.05. Gene Ontology (GO) enrichment analysis of DEGs was performed to identify the latent factors affecting BC progression. In 3 BC samples, the correlation between LAR1 and other DEGs were calculated, and the correlation coefficients were used to perform gene set enrichment analysis (GSEA)-GO and GSEA-KEGG to identify cancer development-associated signaling pathways.

Online website analysis
LAR1 expression was analyzed across various tumor types using the TNMplot platform (https://tnmplot.com/analysis/). Its expression profile in BC samples and its correlation with overall survival of patients were further investigated via UALCAN (https://ualcan.path.uab.edu/index.html) and GEPIA (http://gepia.cancer-pku.cn/), respectively. Finally, the starBase database (https://rnasysu.com/encori/) was utilized to evaluate the potential relationship between miR-143-3p and LAR1 expression.

Cell infection
The sequences of LAR1 or shRNAs targeting LAR1 were inserted into the pLVX-IRES-puro (Cat#BR025) or pLKO.1-EGFP-puro vector (Cat#FH1717) (Hunan Fenghui Biotechnology, Changsha, China), respectively. HEK-293T cells (iCell Bioscience Inc., Shanghai, China) were co-transfected by the lentiviral plasmids, along with pSPAX2 (Hunan Fenghui Biotechnology, Changsha, China), and pMD2.G (Hunan Fenghui Biotechnology, Changsha, China) to produce the recombinant lentiviral particles. For stable overexpression or knockdown of LAR1, BT-549 or MDA-MB-231 cells were infected with the recombinant lentivirus (MOI = 10) for 24 h. Afterwards, puromycin (Cat#P816466, Macklin Inc., Shanghai, China) was added in the culture medium to screen stably infected cells.

Cell transfection
The sequences of LAR1 was inserted into the pcDNA3.1 vector, which was used to transfect MCF-7 cells for LAR1 overexpression. Furthermore, MCF-7 cells were transfected with siRNAs targeting LAR1 to achieve LAR1 knockdown. Additionally, the sequences of APOC1 was inserted into the overexpression vector that was used to transfect MDA-MB-231 cells for APOC1 overexpression. After 48 h of transfection, cells were collected for subsequent experiments.

Cell treatment
MDA-MB-231 cells were treated with the NF- κB inhibitor PDTC (25 μmol/L, Cat#A800469, Macklin Inc., Shanghai, China) for 24 h. For mRNA stability analysis, MDA-MB-231 cells were treated with 10 μg/mL actinomycin D (Cat#HY-17559, MCE, USA) for different periods of time (0, 1, 2, 3, or 4 h). APOC1 mRNA remaining was analyzed by Simple linear regression and half time was determined by the one phase decay of Nonlin fit.

qRT-PCR
Tissues or cells were lysed by TRIpure reagent (Cat#RP1001, Bioteke, Beijing, China) to extract total RNA. The RNA concentration was detected by the spectrophotometer (Thermo Fisher, Waltham, MA, USA). Complementary DNA (cDNA) was synthesized from mRNA using the All-in-One First-Strand SuperMix (Cat#MD80101, Magen Biotech, Guangzhou, China), and from miRNA using a First Strand cDNA Synthesis kit (Cat#B532451, Sangon, Shanghai, China). Subsequent quantification was performed via SYBR Green (Cat#SY1020, Solarbio, Beijing, China)-based qRT-PCR using specific primers on Exicycler 96 (Bioneer, Daejeon, Korea). GAPDH and U6 served as the internal reference genes for mRNA and miRNA, respectively. The 2ˆ(-ΔΔCT) methods were employed to calculate the gene expression. Primer sequences for LAR1: Forward 5'-ACCAAACTACCGCAACA-3', Reverse 5'-TCACGGGAATCCATCAC-3'. Primer sequences for APOC1: Forward 5'-GAGGCTCTTCCTGTCGC-3', Reverse 5'-AACCACTCCCGCATCTT-3'. Reverse primer for miR-143-3p: Forward 5'-TGAGATGAAGCACTGTAGCTC-3', Reverse provided by the reverse transcription kit.

Western blot (WB) assay
Cells were collected to extract proteins using RIPA buffer (Cat#R0010, Solarbio, Beijing, China) containing phenylmethylsulfonyl fluoride (Cat#P0100, Solarbio, Beijing, China). The protein concentration was determined by the BCA Protein Assay Kit (Cat#PC0020, Solarbio, Beijing, China). Protein samples were separated by SDS-PAGE and transferred to polyvinylidene fluoride membranes. After blocking, membranes were incubated with primary antibodies, including antibodies against LAR1 (Cat#13708-1-AP, 1:1000, Proteintech, Wuhan, China), MMP9 (Cat#AF5228, 1:1000, Affinity, Liyang, China), SDF-1 (Cat#AF5166, 1:500, Affinity, Liyang, China), APOC1 (Cat#DF10148, 1:500, Affinity, Liyang, China), p-p65 (Ser536) (Cat#310013, 1:1000, Zenbio, Chengdu, China), p65 (Cat#R25149, 1:1000, Zenbio, Chengdu, China), p-IκBα (ser32/ser36) (Cat#340776, 1:500, Zenbio, Chengdu, China), IκBα (Cat#R380682, 1:500, Zenbio, Chengdu, China), p-IKK (Ser180/181) (Cat#AF3013, 1:500, Affinity, Liyang, China), IKK (Cat#AF6014, 1:500, Affinity, Liyang, China), p-AKT (Ser473) (Cat#AF0016, 1:500, Affinity, Liyang, China), and AKT (Cat#AF6261, 1:500, Affinity, Liyang, China) at 4°C overnight, followed by the incubation of goat-anti rabbit IgG-HRP (Cat#SE134, 1:3000, Solarbio, Beijing, China) at 37°C for 1 h. GAPDH (Cat#60004-1-Ig, 1:10000, Proteintech, Wuhan, China) was used as the internal control and its secondary antibody was goat-anti-mouse IgG-HRP (Cat#SE131, 1:3000, Solarbio, Beijing, China). After that, the protein bands were visualized following treatment with ECL Western Blotting Substrate (Cat#PE0010, Solarbio, Beijing, China).

Cell counting kit (CCK)-8 assay
CCK-8 assay was performed using Cell Proliferation And Cytotoxicity Assay Kit (Cat#CA1210, Solarbio, Beijing, China). Cells were seeded into 96-well plates and cultured in 37°C incubator for different times (0, 24, 48, or 72 h). Subsequently, CCK8 solution was added into culture medium, and cells were cultured in 37°C incubator with 5% CO2 for 2 h. Cell proliferation ability was assessed by the absorbance value at 450 nm using the microplate reader (BioTek, Winooski, VT, USA).

Colony formation assay
For colony formation assay, 300 cells were seeded into dishes. After 14 day of culture, visible colonies were formed and stained with Giemsa stain (Cat#D011-1-2, Nanjing Jiancheng Bioengineering Institute, Nanjing, China), observed by an IX53 microscope (OLYMPUS, Tokyo, Japan). Colony formation rate was calculated by the following formula: number of colonies/number of cells × 100%.

Cell cycle detection
DNA Content Quantitation Assay kit (Cat#CA1510, Solarbio, Beijing, China) was used to analyze cell cycle. Briefly, cells were fixed in the 70% pre-cooled ethanol at 4°C overnight. Fixative solution was washed away with phosphate buffer saline solution. Cells were collected, re-suspended by the RNase A solution, and incubated with propidium iodide (PI) solution at 4°C for 30 min in the dark. Then, samples were subjected to the NovoCyte flow cytometry analysis (Agilent, Santa Clara, CA, USA).

Animal study
To establish xenograft tumors, cells in the logarithmic growth phase were harvested and subcutaneously injected into the mice. Tumor dimensions were measured every 4 days for 32 days to calculate volume. Following euthanasia, tumor tissues were excised and collected for subsequent analysis. For tumor metastasis assay, the labeled cells in the logarithmic growth phase were inoculated into the mouse spleen. Splenectomy was performed on mice 5 min post-inoculation. Following an additional 6 weeks, the mice underwent bioluminescence system detection to access the metastatic lesion area. Subsequently, all mice euthanized and their liver tissues were collected for further experimentation.

Hematoxylin-eosin (HE) staining
Fixed liver tissues were dehydrated in gradient ethanol solution and cleared in xylene (Cat#1330-20-7, Aladdin, Shanghai, China). Tissues were embedded in paraffin and sliced into 5-μm thick sections. Dewaxed and rehydrated section were stained with hematoxylin (Cat#H8070, Solarbio, Beijing, China) for 5 min followed by hydrochloric acid differentiation, and stained with eosin (Cat#A600190, Sangon, Shanghai, China) for 3 min. After dehydration and dewaxing, sections were mounted with neutral gum and imaged using a microscope (OLYMPUS, Tokyo, Japan).

Immunofluorescence staining
Paraffin-embedded tumor tissues were dewaxed, dehydrated, and heated in antigen retrieval solution for 10 min. After blocking, sections were incubated with Ki67 antibody (Cat#AF0198, 1:100, Affinity, Liyang, China) at 4°C overnight and goat-anti rabbit IgG-CY3 (Cat#ab6939, 1:200, Abcam, Cambridge, UK) at room temperature for 60 min. Nucleus were labeled by DAPI stain (Cat#D106471-5mg, Aladdin, Shanghai, China). Sections were then mounted by anti-fluorescence quencher (Cat#S2100, Solarbio, Beijing, China) and viewed using a microscope (OLYMPUS, Tokyo, Japan).
Cells were fixed with 4% paraformaldehyde and incubated with 0.1% tritonX-100. Next, 1% BSA solution was used for blocking, cells were incubated with p65 antibody (Cat#R25149, 1:100, Zenbio, Chengdu, China) at 4°C overnight and goat-anti rabbit IgG-CY3 (Cat#ab6939, 1:200, Abcam, Cambridge, UK) at room temperature for 60 min. After being treated with DAPI (Cat#D106471-5mg, Aladdin, Shanghai, China) and anti-fluorescence quencher (Cat#S2100, Solarbio, Beijing, China), cells were captured using a microscope (OLYMPUS, Tokyo, Japan).

Transwell migration and invasion assays
Transwell chambers (Cat#14341, LABSELECT, Hefei, China) without or pre-coated with Matrigel (Cat#356234, Corning, Corning, NY, USA) were used to detect cell migration and invasion abilities. They were inserted into the 24-well plate. Cells in 200 μL of serum-free medium were seeded into the upper chamber, and 800 μL of culture medium supplemented with 10% FBS was added into the lower chamber. After being cultured in 37°C incubator with CO2 for 24 h, cells were fixed with 4% paraformaldehyde (Cat#C104188, Aladdin, Shanghai, China) for 20 min, stained with 0.5% crystal violet solution (Cat#0528, Amresco, Solon, OH, USA) for 5 min, counted under a microscope (OLYMPUS, Tokyo, Japan). Five views were randomly selected to record the average number of cells.

RNA immunoprecipitation (RIP) assay
RIP assay was performed using the EZ-Magna RIP Kit (Cat#17-701, Millipore, Billerica, MA, USA). Cells were lysed to obtain the supernatant. Magnetic beads were incubated with antibody, and antibody-coated magnetic beads were separated using the magnetic stand and re-suspended in the RIP Immunoprecipitation Buffer solution. Next, the cell lysate was added into this Buffer solution containing antibody-coated magnetic beads. The whole cell lysate was used as the “Input”. All these samples were placed in a rotator at 4°C for overnight. Next, the protein-RNA complex was washed with the RIP Wash Buffer solution, collected, and incubated with the Proteinase K Buffer solution at 55°C for 30 min. However, the Input samples was incubated with the Proteinase K Buffer solution, along with 10% SDS and the RIP Wash Buffer solution. Then, RNA was purified and transcribed to cDNA, which was amplified by 2% agarose gel and visualized by a gel imaging analyzer (Clinx Science Instruments, Shanghai, China).

Luciferase reporter assay
The 3'-UTR sequences, coding sequences (CDS), or 5'-UTR sequences of APOC1 was inserted into the pGL3-promoter. BC cells were co-transfected with the pGL3-promoter and the renilla luciferase reporter pRL-TK (Hunan Fenghui Biotechnology, Changsha, China) using lipo3000 reagent for 48 h. Then, cells were lysed and treated with firefly luciferase assay reagent of Luciferase Assay Kit (Cat#KGE3308, KeyGEN Bio, Nanjing, China) followed by the treatment of luciferase assay reagent. The absorbance value was recorded using a microplate reader (BioTek, Winooski, VT, USA).

mRNA-sequencing analysis
The transcriptomic changes induced by LAR1 overexpression was profiled using mRNA sequencing on LAR1-overexpressing MDA-MB-231 cells and vector controls. DEGs were defined as those with a |Log2FC| > 1.5 and a p value < 0.01. To distinguish direct targets of LAR1, we intersected these DEGs with a previously published LAR1-mRNA interactome dataset generated by RNA immunoprecipitation and microarray profiling (RIP-CHIP).8 Finally, GO enrichment analysis was performed on this overlapping gene set.

Quantification and statistical analysis
All data were provided as the form of the mean ± standard deviation (SD) and analyzed using the GraphPad Prism 9 in the current study. Ordinary one-way ANOVA, Brown-Forsythe and Welch ANOVA tests, or two-way ANOVA was used to compare the means among three or more than three groups. Unpaired t test or Welch's t test was used to compare the means between two groups. The specific statistical tests used are stated in each figure legend. The n values in figure legends means biological replicates. The p value less than 0.05 was considered significant. ∗p < 0.05, ∗∗p < 0.01; #p < 0.05, ##p < 0.01.