The autoregulatory serglycin/CD44 axis drives stemness‐like phenotypes in TNBC in a β‐catenin‐dependent manner

Dear Editor, Overexpression of serglycin (SRGN) in human breast cancer suggested a poor prognosis. However, its role in triplenegative breast cancer (TNBC) recurrence and metastasis remains unclear.1–3 To explore the clinical value of SRGN, we used 144 breast cancer samples to determine SRGN expression (Figures 1A and 1C). The SRGNexpression level was closely associated with the overall survival of breast cancer (Figure 1B; Figure S1A). Additionally, immunohistochemistry (IHC) staining showed that serglycin was expressed in the tumor stroma (Figure 1F). EGFR and Ki67, which were significantly correlated with worse survival and a poor prognosis in TNBC patients, showed higher expression levels in TNBC tissue.4 The luminal subtype expressed GATA3, which regulates differentiation and suppresses dissemination in breast cancer.5 We found that SRGN was positively correlated with EGFR and Ki67. Additionally, we found that the SRGN mRNA level was negatively correlated with GATA3 (Figures 1G, 1L, and 1M; Figures S1B-S1D). To reconfirm the findings above, we examined SRGN expression in the cell lines of MDA-MB-231 and MCF-7, which are representative cell lines of the basal-like and luminal subtypes, respectively. The results showed that serglycin in MDA-MB-231 was significantly higher either at the mRNA level or protein level compared with MCF-7 (Figures 1D and 1E; Figure S1E). Here, we showed that the migration and invasion and wound healing of TNBC cells could be remarkably suppressed in vitro by stably knocking down SRGN (Figures 1H-1K and 1N; Figures 2A-2C; Figure S1F). In vivo, the results showed a drastically lower rate of lungmetastasis in mice injected with the SRGN knockdown cell lines (MDAMB-231-KD31 andKD32) than in thosewithMDA-MB-231SCR cells (Figures 2D-2G). The stemness of TNBC, such as the self-renewal and tumor-initiating capacities of breast cancer cells, was also mediated by SRGN (Figures 2H-2L; Figures S1G and S1H). Western blotting showed that KD32

Dear Editor, Overexpression of serglycin (SRGN) in human breast cancer suggested a poor prognosis. However, its role in triplenegative breast cancer (TNBC) recurrence and metastasis remains unclear. [1][2][3] To explore the clinical value of SRGN, we used 144 breast cancer samples to determine SRGN expression ( Figures 1A and 1C). The SRGN expression level was closely associated with the overall survival of breast cancer ( Figure 1B; Figure S1A). Additionally, immunohistochemistry (IHC) staining showed that serglycin was expressed in the tumor stroma ( Figure 1F). EGFR and Ki67, which were significantly correlated with worse survival and a poor prognosis in TNBC patients, showed higher expression levels in TNBC tissue. 4 The luminal subtype expressed GATA3, which regulates differentiation and suppresses dissemination in breast cancer. 5 We found that SRGN was positively correlated with EGFR and Ki67. Additionally, we found that the SRGN mRNA level was negatively correlated with GATA3 ( Figures 1G, 1L, and 1M; Figures S1B-S1D).
To reconfirm the findings above, we examined SRGN expression in the cell lines of MDA-MB-231 and MCF-7, which are representative cell lines of the basal-like and luminal subtypes, respectively. The results showed that serglycin in MDA-MB-231 was significantly higher either at the mRNA level or protein level compared with MCF-7 ( Figures 1D and 1E; Figure S1E).
Here, we showed that the migration and invasion and wound healing of TNBC cells could be remarkably suppressed in vitro by stably knocking down SRGN ( Figure S1F). In vivo, the results showed a drastically lower rate of lung metastasis in mice injected with the SRGN knockdown cell lines (MDA-MB-231-KD31 and KD32) than in those with MDA-MB-231-SCR cells (Figures 2D-2G). The stemness of TNBC, such as the self-renewal and tumor-initiating capacities of breast cancer cells, was also mediated by SRGN (Figures 2H-2L; Figures S1G and S1H). Western blotting showed that KD32 This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. The aggressive phenotype of TNBC might be partially due to an abundance of cancer stem cells (CSCs), indicating that TNBC has more CSC-like properties. 6 Therefore, we examined the expression of the CSC marker CD44. [7][8] Real-time quantitative PCR (qPCR) showed that cells in mammospheres generated by MDA-MB-231 and MCF-7 cells expressed a high CD44 level, indicating that mammospheres can enrich CSCs. Thus, CD44 might be a CSC marker for breast cancer and serves as a receptor of serglycin. We found the same with CD44, other CSC markers such as Nanog, ALDH1, and CD133 high expression in MDA-MB-132 than MCF-7. These CSC markers significantly reduced in the silencing of SRGN in MDA-MB-231, but we could not see significantly change in MCF-7 over expression serglycin cells ( Figures S2F, S2H, and S2I;) We examined the expression of CD44 and phosphor-ERK (P-ERK) in two wild-type cell lines. We found that CD44 and P-ERK were both highly expressed in MDA-MB-231 cells ( Figure 3J; Figure S2G), indicating that MAPK signaling was overactivated in MDA-MB-231 cells, and SRGN, CD44, and P-ERK may interact with each other. These data indicate that the expression of SRGN was positively correlated with CD44 and can promote the phosphorylation of ERK to activate MAPK signaling.  To verify the connection between P-ERK and CD44 in TNBC, we used different concentrations of the specific ERK inhibitor selumetinib or U0126 to treat two wild-type cells and found that P-ERK was effectively suppressed, accompanied by decreased CD44 expression in both cell lines (Figures 3K and 3L; Figures S3A and S3B). However, when the MFC-7 cells were treated with U0126 or selumetinib, both CD44 and P-ERK did not significantly increase (Figures S3I and S3J). Additionally, under the treatment of different concentrations of selumetinib and U0126, the numbers of mammospheres formed by MDA-MB-231 cells were markedly decreased, while they in MCF-7 did not show a significant change (Figures S3C-S3H). To investigate how CD44 expression is mediated, we performed the luciferase report assay using a CD44 promoter construct ( Figure 3M). These results revealed that CD44 expression was regulated by a serglycin-activating pathway at the transcriptional level.
Moreover, we demonstrated serglycin promoted nucleus translocation of β-catenin and the β-catenin expression level in two wild-type cells were evaluated by immunofluorescence assay (Figures 3N and 3O; Figure S4A) and Western blotting (Figures 4B and S4C). Serglycin was directly binding with CD44 on the cytomembrane of MDA-MB-231( Figures 4A and 4B; Figure S4D). We found the binding of SRGN to CD44 can be prevented by a CD44-neutralizing antibody in MDA-MB-231 cells ( Figure 4C). The CD44neutralizing antibody could significantly reduce the migration and the formation of mammospheres in a dosedependent manner by blocking the binding of SRGN to CD44 (Figures S4E-S4G). Thus, we demonstrated that combined treatment with DDP and the CD44-neutralizing antibody suppressed the cell proliferation better than single drug treatment separately ( Figures 4D-4F). The effect from binding SRGN with CD44 can be restrained by CD44 neutralizing antibody in a β-catenin-dependent manner ( Figure 4G).
Consequently, the expression level of β-catenin could be degraded by a high concentration of the CD44-neutralizing antibody in a dose-dependent manner. This ubiquitination degradation of β-catenin in MDA-MB-231 cells could be reversed by the proteasome inhibitor MG132 (Figures 4H-4J). These results indicated that the extracellular matrix (ECM) factor SRGN binds to cell surface-adherent CD44 in an autocrine manner, resulting in activation of the MAPK signaling pathway to trigger the translocation of βcatenin into the nucleus, which, in turn, regulates CD44 expression. The CD44-neutralizing antibody competitively inhibits SRGN binding with CD44 and suppresses the MAPK pathway depending on β-catenin ubiquitination degradation pathways. Thus, this result sheds light on breast CSC treatment targeting CD44 ( Figure S4H).

C O N F L I C T O F I N T E R E S T
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

E T H I C S A P P R O VA L A N D C O N S E N T T O PA R T I C I PAT E
All protocols with human specimens were applied under the examination and approval of the Ethical Committee of Sun Yat-Sen University Cancer Center.

A U T H O R C O N T R I B U T I O N S
Fei-Fei Luo and Li Cao conducted the in vitro and in vivo experiment and data analysis and wrote the manuscript. Hong-Bin Huang, Tie-Jun Huang, Li-Xia Peng, and Li-Sheng Zheng collected the patient samples and the followup information and performed the clinical data analysis. Hao Hu and Jing Wang performed the bioinformatics analysis. Chao-Nan Qian and Bi-Jun Huang designed the study and analyzed the manuscript. All the authors read and approved the final manuscript.

D ATA A C C E S S , R E S P O N S I B I L I T Y, A N D A N A LY S I S
All data generated or analyzed during the present study are available via the corresponding author on reasonable request. The data are under review and will link to Research Data Deposit (http://www.researchdata.org.cn/) with a unique deposit ID: RDDB2021001067. Li Cao 2,+ Fei-Fei Luo 1,+ Hong-Bin Huang 3 Tie-Jun Huang 4 Hao Hu 5 Li-Sheng Zheng 1 Jing Wang 1 Li-Xia Peng 1 Chao-Nan Qian 1,6,7 Bi-Jun Huang 1