Investigation of LGALS2 expression in the TCGA database reveals its clinical relevance in breast cancer immunotherapy and drug resistance

Breast cancer (BRCA) is known as the leading cause of death in women worldwide and has a poor prognosis. Traditional therapeutic strategies such as surgical resection, radiotherapy and chemotherapy can cause adverse reactions such as drug resistance. Immunotherapy, a new treatment approach with fewer side effects and stronger universality, can prolong the survival of BRCA patients and even achieve clinical cure. However, due to population heterogeneity and other reasons, there are still certain factors that limit the efficacy of immunotherapy. Therefore, the importance of finding new tumor immune biomarker cannot be emphasized enough. Studies have reported that LGALS2 was closely related to immunotherapy efficacy, however, it is unclear whether it can act as an immune checkpoint for BRCA immunotherapy. In the current study, changes in LGALS2 expression were analyzed in public datasets such as TCGA-BRCA. We found that LGALS2 expression was associated with immune infiltration, drug resistance and other characteristics of BRCA. Moreover, high LGALS2 expression was closely related to immunotherapy response, and was associated with methylation modifications and clinical resistance for the first time. These findings may help to elucidate the role of LGALS2 in BRCA for the development and clinical application of future immunotherapy strategies against BRCA.


BRCA datasets
The Cancer Genome Atlas (TCGA), the Molecular Taxonomy of Breast Cancer International Consortium (METABRIC), the Gene Expression Omnibus (GEO), the Cancer Cell Line Encyclopedia (CCLE), the cBioPortal database and Connectivity Map (CMap) were used to obtain gene expression and pertinent prognostic and clinicopathological data for BRCA patients.In addition, all data were downloaded from public databases and analyzed using the R (version 4.0.3) and R Bioconductor.

Functional enrichment analysis
In the TCGA dataset (https:// portal.gdc.com), we downloaded RNA-sequencing expression (level 3) profiles and clinical information related to BRCA patients.According to the expression level of LGALS2 gene in the TCGA-BRCA dataset, they were divided into two groups: high (n = 551) and low (n = 550).The R package Limma was used to study the differentially expressed mRNAs.Additionally, Adjusted P < 0.05 and Log2 (Fold Change) > 1 or < − 1 was defined as the threshold for the differential expression of mRNAs.We analyzed Gene Ontology (GO) function of the underlying mRNAs and enriched Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway by using the ClusterProfiler package in R 35 .

Small molecule targeted drugs screening of LGALS2 in BRCA
The Connectivity Map (CMap) (http:// www.broad.mit.edu/ cmap) was used to search for potential small molecule targeted drugs.Those small molecule drugs with |score|> 0.2 and P < 0.05 were recognized as the potential therapeutic drugs targeting LGALS2 in BRCA 40,41 .

Statistical analyses
All statistical analyses were performed using the R (version 4.0.3)software.Kaplan-Meier survival curves were constructed and compared with log-rank tests.The Spearman correlation analysis was used to examine correlations among variables without normal distributions.Data were analyzed between two and multiple groups by Student's t test and one-way ANOVA.P < 0.05 was considered to be statistically significant.(ns, P ≥ 0.05, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

LGALS2 expression associates with diagnosis and prognosis in BRCA patients
The TCGA database was used to analyze the expression of LGALS2 in various normal and tumor tissues.The results showed that LGALS2 expression was statistically significant among 11 groups including BLCA (P < 0.01) (Fig. 1A), and was much lower in tumor groups, especially in BRCA (P < 0.001) (Fig. 1B).In addition, the expression levels of LGALS2 in different tissues (n = 29) and different human breast cancer cell lines (n = 68) were also analyzed based on the CCLE database, which was found to be consistent with the TCGA database results (Fig. S1A, B).Meanwhile, the above cell lines were classified into subtypes and our analyses showed no significant differences between these groups (P > 0.05) (Fig. S1C) [42][43][44] .And then, we further analyzed the diagnostic and prognostic value of LGALS2 in BRCA.The Kaplan-Meier (KM) curve indicated that BRCA patients with high LGALS2 expression had better survival rates (P = 0.014) (Fig. 1C), also in ER, HER2 and PR patients (P < 0.01) (Fig. 1E-G).Moreover, the receiver-operating characteristic (ROC) curve showed the diagnostic value of LGALS2 in BRCA patients (AUC = 0.601) (Fig. 1D) and their different subtypes (Fig. 1H-J).implying the potential value of LGALS2 in the diagnosis and prognosis of BRCA.

Analysis of various clinical factors associated with LGALS2 expression
The clinical patients with different expression of LGALS2 showed the distinct patterns of clinical and pathological characteristics.The change in LGALS2 expression with chemotherapy or radiotherapy, age and gender, pathologic-stage, T-stage, N-stage, M-stage and survival status are shown in Fig. 2A.The expression of LGALS2 was not statistically significant in groups with pathological-, T-, N-or M-stage, radiotherapy, drug treatment (Fig. 2B-G) and gender (Fig. 2I), but except for the age group (P < 0.01) (Fig. 2H).The clinical factors associated with LGALS2 expression (e.g., AJCC-stage, age, gender, race, etc.) were also extracted from the cBipPortal database to complement the results of the TCGA database (Fig. S2) 45 .Consistent with the prior results, LGALS2 expression levels were found to significantly differ only with patient diagnosis age (P < 0.05) and race (P < 0.001).The results of this study did not reveal a statistically significant difference based on the clinical stage.LGALS2 has diagnostic value for TNBC Studies have reported that LGALS2 could be used as a therapeutic target for TNBC 33 .However, there are still some data limitations in the TCGA database, thus, the METABRIC database was used to analyze the expression of LGALS2 in the BRCA subtypes.When molecular markers were detected in various types of BRCA, the expression level of LGALS2 was significantly elevated in ER-negative, HER2-negative, and PR-negative groups (P < 0.0001) (Fig. 3A-C).We then investigated LGALS2 distribution in different subtypes defined by the META-BRIC database.Results showed that LGALS2 was significantly enriched in the Claudin-low subtype compared to the other subtypes (P < 0.0001) (Fig. 3D).
LGALS2 expression specificity was assessed using ROC curves.The area under the curve (AUC) was up to 78.7% in the METABRIC database (P < 0.0001) (Fig. 3E).It appeared that LGALS2 was significantly enriched in TNBC, suggesting its diagnostic potential as a biomarker and providing a reference for the clinical diagnosis of TNBC.LGALS2 regulates immune responses in a T cell-dependent manner LGALS2 expression was investigated in the TCGA database using the LIMMA R package to explore its potential biological function in BRCA.It generated 300 genes, and 279 genes were upregulated, 21 genes were downregulated (|logFC|> 1, adjusted P < 0.05) (Fig. 4A,B).In the GO and KEGG analyses, LGALS2 functions were primarily related to cytokine-cytokine receptor interaction, viral protein interaction with cytokine and cytokine receptor, chemokine signaling pathway, T cell activation, leukocyte cell-cell adhesion, regulation of T cell activation, etc. (Fig. 4C-F).Based on these findings, combined with the previous results related to immune infiltration T cells and its association with LGALS2 expression, we speculated that LGALS2 might be closely associated with immunotherapy outcome and contribute to immune response in BRCA patients.

LGALS2 expression positively correlates with T cell-mediated immune responses
Cancer cells die immunologically due to lymphocyte activation (including NK cells, T cells, and B cells) and the release of chemokines and cytokines 46,47 .Therefore, we investigated the effects of LGALS2 expression on immune pathways and cytokines.TCGA gene set variation analysis (GSVA) was used to determine the immune process enrichment score.Based on correlation analysis between LGALS2 expression and enrichment score, LGALS2 expression was positively correlated with most immune functions, but not with activin receptor signaling pathway and plasma membrane organization (Fig. 5A).
LGALS2 expression was associated with various immune cells as shown in Fig. 5B, and was most closely associated with T cells.Moreover, the immune infiltration data showed a higher degree of enrichment of LGALS2 and T cells (P < 0.001, r = 0.737) (Fig. 5C).The above results indicated that LGALS2 expression was associated with T cell activity and immunotherapy response in BRCA patients.

LGALS2 positively correlates with established cancer immune checkpoints
In previous studies, LGALS2 was shown to be involved in tumor immunity 48 , thus we investigated the relationship between LGALS2 expression in the TCGA dataset and immune checkpoints, including CD200R1, CD47, CTLA4, TIM-3, PD-1, TIGIT, HVEM and CD69.

T cells express high levels of LGALS2
We analyzed public datasets from the GEO database using R for single-cell sequencing.UMAP dimensionality reduction analysis was used to obtain a two-dimensional representation of cell state and 12 clusters were created (Fig. 7A).We found that LGALS2 was mainly enriched and highly expressed in the clusters 4 and 5 (Fig. 7B).Cellular markers enabled us to identify the cluster 4 and cluster 5. Based on the expression of multiple cellular markers, such as CD4, CD68, CD83 and so on [55][56][57] , clusters 4 and 5 were categorized as T cells (Fig. 7C,D).
In the subsequent analyses, we showed that LGALS2 expression was closely associated with the immune system and interacted with T cells.Thus, upregulating LGALS2 expression was associated with the activation and differentiation of T cells, induction of immune responses and promotion of the necrosis or apoptosis of breast cancer cells.

LGALS2 expression level positively correlates with methylation modification
In light of the importance of DNA methylation in regulating gene expression 58,59 , we examined LGALS2-associated methylation level.A significant increase in promoter methylation of LGALS2 was observed in tumor samples as compared to the normal group (P < 0.0001) (Fig. 8A), and this was also significantly associated with the tumor type and patient age (Fig. 8B,C), which were inversely correlated with LGALS2 mRNA expression level (Fig. 1B).At the same time, the expression levels of DNA methyltransferases (DNMT) (including DNMT1, DNMT3A and DNMT3B) in the normal and tumor samples from TCGA database were also determined.The results showed that DNMT was highly expressed in breast cancer (P < 0.0001) (Fig. 8D-F).Furthermore, the results suggested that methylation occurred 2400-2700 bp downstream of LGALS2 transcription start sites (cg23835646, cg11081833 and cg26651950) (P < 0.001) (Fig. 8G-I).Due to this, DNMT might affect the transcription process by adding methyl groups to LGALS2 without affecting its sequence, thereby reducing the expression of LGALS2, and the methylation modification of LGALS2 might be one of the reasons for its decreased mRNA expression.

LGALS2 expression associates with clinical sensitivity to anticancer drugs
To further explore LGALS2 as a potential therapeutic target in BRCA, we performed CMap analysis and obtained 15 compounds with positive and negative correlations, respectively (Fig. S4A) 60,61 .It has been reported that cancer patients often suffer from drug resistance, which leads to relapses and reduced survival rates 62 .The role of LGALS2 expression in breast cancer drug resistance was therefore studied.Surprisingly, we found a negative correlation between LGALS2 expression and resistance in BRCA (P < 0.0001) (Fig. S3).High levels of LGALS2 were associated with a significant reduction in the IC50 values of several clinical anticancer drugs (including camptothecin, parthenolide and paclitaxel) [63][64][65][66][67][68][69][70][71] (Fig. 9).Meanwhile, CMap analysis further evaluated to validate the above drugs.The results also showed that these drugs were significantly associated with LGALS2 expression in BRCA (P < 0.01) (Fig. S4B-E).Therefore, LGALS2 expressed levels might be associated with the increased sensitivity of cancer cells to clinical drugs and longer survival of cancer patients.

Discussion
Breast cancer is the number most common cause of cancer related mortality in women globally, and TNBC accounts for about 10% to 15% of that 72 .Although the patient's condition is improved after surgical resection, radiotherapy, chemotherapy and other therapeutic interventions, adverse reactions such as metastasis and drug resistance contribute to the overall unsatisfactory treatment outcomes 73,74 .Compared with other therapeutic strategies, cancer immunotherapy causes fewer side effects and is more universal, which prolongs the survival time of patients and even achieves clinical cure, and thus is a promising treatment strategy 75,76 .However, there are still some limitations such as tumor immune escape, tumor mutation burden, and adaptive immune resistance, which suppress the therapeutic potential of cancer immunotherapy 38,77 .Therefore, it is crucial to find a new biomarker for evaluating immunotherapy response in breast cancer treatment.
In this study, we found that LGALS2 expression was low in breast cancer patients and was lower in the breast tissue relative to other tissues, and patients with high LGALS2 expression had a better prognosis (P = 0.014).Furthermore, we show that LGALS2 could be used as a diagnostic marker for breast cancer (AUC = 0.601), and has a better diagnostic value in TNBC (P < 0.0001).The ROC analysis showed that the AUC-value was 0.787.The above findings suggest that LGALS2 has a good prognostic value for BRCA patients, especially for TNBC patients.
However, in the analysis of LGALS2 and related clinicopathological factors based on the TCGA and cBioPortal database, there was no statistically significant correlation between LGALS2 expression and the commonly used clinical stages.Braud VM et al. 78 reported that KLRB1 was a predictive marker for the survival status of patients, but was not related to the degree of tumor cell invasion, and our findings were consistent with them.In addition, the current common tumor staging is based on the length of the tumor, the status of the ipsilateral axillary lymph node, and the presence of distant metastasis 79,80 .Therefore, we speculated that LGALS2 might not be directly involved in the migration and development of breast cancer.Of course, data is an important factor affecting the results, including the data collection process, the size and quality of the data, the regional distribution of patients, and the heterogeneity and bias of the data 81,82 .DNA methylation is often dysregulated during tumorigenesis and cancer progression, which are manifested by the reduction of the overall methylation level of the genome and the inactivation of the local DNA regions 83,84 .Whether the low expression of LGALS2 in breast cancer patients was related to DNA methylation modification has not yet been reported.Here, we found that LGALS2 promoter methylation levels were high in the tumor groups (P < 0.001) and were closely related to the tumor subtypes and patient age (P < 0.05).At the same time, DNA methyltransferases (DNMT) (including DNMT1, DNMT3A, DNMT3B) were also highly expressed in tumors (P < 0.001).Namely, LGALS2 expression was negatively correlated with LGALS2 promoter methylation and DNA methyltransferase.As a result, we speculated that DNMT regulated methylation modifications in the LGALS2 promoter region to inhibit the transcription of LGALS2, resulting in decreased LGALS2 expression.Moreover, the results suggested that methylation occurred 2400-2700 bp downstream of LGALS2 transcription start sites (cg23835646[TSS + 2561], cg11081833[TSS + 2683], cg26651950[TSS + 2445]) (P < 0.001).Therefore, reducing the methylation level of the LGALS2 promoter, thereby reversing the occurrence and development of breast cancer, could serve as a new strategy for breast cancer treatment.However, further research is warranted to verify the above findings.
Previous studies reported that LGALS2 induced T lymphocyte apoptosis, improved colitis, and prevented preeclampsia 27,85 .During cancer immunotherapy, T cells play a key role in the antitumor response and are closely associated with the effective inhibition of immune checkpoints 86,87 .In this study, GO and KEGG analysis, GSVA, enrichment score, etc. revealed that LGALS2 was associated with increased T cell infiltration (P < 0.001) and had a positive regulatory relationship with immune response.The data used in these analyses were all obtained from the average of multiple data samples, but we could not obtain information regarding inherent tumor heterogeneity.However, single-cell sequencing solves this problem well, with higher sensitivity and accuracy, and more reliable results.Cell markers such as CD4, CD68, AIF1, PLEK, and others were used to identify T cell clusters 56,88 , and we found LGALS2 to be specifically highly expressed in T cell clusters.The above findings suggested that LGALS2 regulated T cells to induce immune response in the body and participated in tumor immunotherapy.The combination of immunotherapy and chemotherapy (so-called chemo-immunotherapy) has demonstrated excellent therapeutic effects in the clinical treatment of cancer patients 89,90 , but adverse reactions such as drug resistance significantly limit the therapeutic efficacy of the combination strategy 91,92 .The main cause of cancer recurrence is the development of multidrug resistance, which in turn is associated with increased expression of efflux transporters, epithelial-to-mesenchymal transformation and breast cancer stem cell resistance 93 .In addition, it has been demonstrated that the formation of an immunosuppressive niche increased the occurrence of drug resistance and suppressed the antitumor effects of T cells, reducing the efficacy of immunotherapy 94 .We therefore examined the role of LGALS2 in clinical drug resistance based on the GDSC database.The results showed that high LGALS2 expression was associated with a lower IC50 value of 3 clinical antitumor drugs, including camptothecin, which enhanced their drug sensitivity and reduced the occurrence of clinical drug resistance.Subsequently we also used CMap to analyse the correlation between these 3 drugs and LGALS2 in BRCA, and gratifyingly this result also showed that these drugs were significantly correlated with LGALS2 expression.Of course, the results of the database have to be verified by further wet experiments.
Twyman-Saint Victor et al. 95 reported that the up-regulation of PD-L1 in melanoma cells was an important factor in the development of drug resistance and was related to T cell exhaustion.Interestingly, therapy-and resistance-related changes in T cells might constitute biomarkers of tumor response.Moreover, immune checkpoints such as PD-1/PD-L1, TIM-3, and TIGIT are known to play important roles in the treatment of NSCLC, OV, SKCM, BRCA, and have been recognized as biomarkers for immunotherapy response 96,97 .Through enrichment analysis and immune score, Jiang et al. 98 identified that the expression level of STC2 was significantly positively correlated with the infiltration level of B cells, T cells and other immune cells in various cancer cells, and was significantly correlated with the sensitivity towards certain drugs, and suggested STC2 as a promising target for tumor immunotherapy.Therefore, LGALS2 could also serve as a potential marker for breast cancer immunotherapy.
Our study systematically and comprehensively validates that LGALS2 can be used as a diagnostic/prognostic marker for breast cancer, especially a good diagnostic marker for TNBC, and that it participates in tumor immunotherapy by regulating T cells.The result that high LGALS2 expression was associated with reduced occurrence of clinical tumor therapy resistance was even more intriguing for us.However, our study suffered from some limitations.First of all, this study solely relied on existing public data, thus necessitating further experimental evidence to verify and clarify the molecular mechanism of LGALS2 in BRCA.Secondly, the low expression of LGALS2 in BRCA was related to the methylation modification of its promoter region and the ability of LGALS2 to reduce the occurrence of clinical drug resistance was also identified for the first time in this study.However, further experimental and clinical studies are needed to confirm the above findings.Lastly, although the results of this study supported the involvement of LGALS2 in tumor immune regulation and immunotherapy, the underlying molecular mechanism needs further investigation.

Conclusion
In summary, our study shows that LGALS2 could be used as a diagnostic and prognostic marker for breast cancer, and that it regulates the biological activity of T cells to participate in tumor immunotherapy and reduces the occurrence of clinical drug resistance in patients.As a novel molecular biomarker for breast cancer treatment, LGALS2 may enable the development of novel immunotherapy strategies that are of a high clinical relevance in the future.The findings from this study need further validation through in vitro and in vivo experiments to confirm the functions of LGALS2 and unravel the underlying molecular mechanisms in breast cancer.

Figure 1 .
Figure 1.Transcriptional alterations and diagnostic/prognostic value of LGALS2 in BRCA patients.(A) Expression of LGALS2 in paracancers and tumors in TCGA database.The significance of the difference was tested with an unpaired student's t test.(B) LGALS2 expression in BRCA patients based on the TCGA database.The significance of the difference was tested with an unpaired student's t test.(C) KM analysis of the diagnostic value of LGALS2 in TCGA database.(D) ROC analysis of the prognostic value of LGALS2 in TCGA database.(E-G) KM analysis of overall survival of LGALS2 in patients with different subtypes of BRCA.(H-J) ROC analysis of prognostic of LGALS2 in patients with different subtypes of BRCA.ns, P ≥ 0.05, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 2 .
Figure 2. Analysis of various clinical factors associated with LGALS2 expression.(A) The landscape of LGALS2-related clinicopathological features of BRCA in TCGA database.(B-I) LGALS2 and various clinicopathological features of BRCA in TCGA databases.(B-D) LGALS2 was not significantly different in Pathologic-, T-, N-stage in TCGA database.The significance of the difference was tested by one-way ANOVA.(E-I)LGALS2 was not significantly different in M-stage, radiation therapy, pharmaceutical therapy and gender in TCGA database.The significance of the difference was tested with an unpaired student's t test.(H) there was a significant difference between LGALS2 expression and patient age in TCGA databases.The significance of the difference was tested with an unpaired student's t test.

Figure 3 .
Figure 3. Diagnostic analysis of LGALS2 expression with TNBC.(A-C) Expression of LGALS2 in ER, HER2, PR.LGALS2 was enriched in the negative subtype of BRCA in the Metabric database.The significance of the difference was tested with an unpaired t test.(D) LGALS2 was highly expressed in claudin-low subtype of BRCA in the Metabric database.The significance of the difference was tested by one-way ANOVA.(E) The ROC curve showed the high-expression specificity of LGALS2 in TNBC in the Metabric database.

Figure 4 .
Figure 4. GO and KEGG analysis of LGALS2.(A) Volcano plot of differentially expressed genes.Red dots indicate upregulated genes, blue dots indicate downregulated genes, grey dots indicate not significant.(B) Heat map showing the differentially expressed genes, and the different colors represent the trend of gene expression.The top 50 up-regulated genes and top 50 down-regulated genes were shown in this figure (C,E) KEGG enrichment analysis.(D,F) GO enrichment analysis.Colors represent the significance of differential enrichment, the size of the circles represents the number of genes, the larger the circle, the greater the number of genes.(|LogFC|> 1, Adjusted P < 0.05).

Figure 5 .
Figure 5. Correlation analysis between LGALS2 and immune function.(A) Heatmap showing LGALS2 expression and the enrichment scores of immune functions of each patient in TCGA, and the column and line graph on the right shows the R and P values for correlation analysis.(B) Analysis of the correlation between LGALS2 and immune cells.The size of the circle represents the correlation, with the larger the circle the higher the correlation.(C) Scatter plot of LGALS2 and T cell enrichment.LGALS2 expression levels were positively correlated with T cell enrichment.The significance of the difference was tested by Spearman correlation analysis.

Figure 6 .
Figure 6.Analysis of the correlation between LGALS2 and T cell immunity and inflammation.(A) Pearson correlation between LGALS2 and inhibitory immune checkpoints.The color of the band represented the P-value.The correlation was tested by Pearson correlation analysis.(B) Correlation matrix of LGALS2 and inflammatory-related metagenes.The bottom left showed the correlation coefficient.which are shown as a scale of the pie charts.The correlation was tested by Pearson correlation analysis.

Figure 7 .
Figure 7. Single-cell sequencing analysis of LGALS2.(A) Single-cell sequencing analysis showing the cellular subtypes of BRCA, based on the GEO database (GSE161529).(B) LGALS2 is highly expressed in cluster 4 and 5. (C,D) Expression of T cell markers in different subtypes.Verification of cluster 4 and cluster 5 as T cells.

Figure 8 .Figure 9 .
Figure 8. Methylation-related analysis of LGALS2.(A) Promoter methylation level of LGALS2 in BRCA.The significance of the difference was tested with an unpaired student's t test.(B) LGALS2 promoter methylation expression based on BRCA subclasses.The significance of the difference was tested with an unpaired student's t test.(C) LGALS2 promoter methylation expression based on age of BRCA patients.The significance of the difference was tested with an unpaired student's t test.(D-F) Relationship between DNA methyltransferases (DNMT1, DNMT3A and DNMT3B) and LGALS2 expression in BRCA patients.The significance of the difference was tested with an unpaired student's t test.(G-I) Methylation probe linked to LGALS2 in BRCA patients.The significance of the difference was tested by Spearman correlation analysis.ns, P ≥ 0.05, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.