FAM49B mRNA levels are elevated in BC and correlate with poor prognosis
FAM49B expression has also been observed in several human cancers (Fig. 1A). ONCOMINE analysis revealed that FAM49B mRNA expression was significantly higher in BC samples than in normal samples across a wide variety of datasets covering different types of BC. FAM49B transcripts were elevated 1.664-fold in BC samples compared to normal tissues, in a dataset containing 450 samples derived from The Cancer Genome Atlas (TCGA) database (Fig. 1B). The pooled results of the eight clinical cohorts showed a significant increase in FAM49B expression in BC (p = 0.003, Fig. 1C). Thereafter, we assessed the prognostic effect of FAM49B in BC using the Kaplan–Meier survival analysis plot. This revealed that high FAM49B mRNA expression was correlated with reduced DFS and OS in all BC patients (HR = 1.29, p = 0.016; HR = 1.26, p = 0.034, respectively, Fig. 1D, E).
FAM49B protein expression is elevated in BC tissues and correlates with poor prognosis
The expression of FAM49B protein in eight BC tissues and eight adjacent non-tumor tissues was assessed using western blotting. As shown in Fig. 2A, FAM49B protein expression levels were significantly higher in BC tissues than in normal breast tissues (p < 0.01).
Next, we performed IHC to visualize FAM49B protein expression in BC tissues. This revealed that FAM49B staining occurred mainly in the cytoplasm of BC samples. ER staining was observed in the nucleus of BC cells, whereas HER2 staining was localized in the cellular membrane. Examples of positive protein expression of FAM49B, ER, and HER2 are shown in Fig. 2B–D. IHC analysis showed that the positive rate of FAM49B expression in cancer tissue samples was 52.8% (95/180 cases). The positive rate was significantly higher than that of the adjacent normal breast tissue (28.9%; 26/90 samples) (p = 0.000, Table 1). In addition, positive FAM49B protein expression was positively correlated with tumor size, histological grade, and lymph node metastasis of BC (p = 0.006, p = 0.013, and p = 0.023, respectively, Table 1).
Table 1
The relationship between FAM49B expression and the clinicopathological factors (n = 180)
Varible
|
n
|
FAM49B−
|
FAM49B+
|
p varible
|
Tissue
|
|
|
|
0.000
|
Cancer tissue
|
180
|
85
|
95
|
|
Adjacent tissue
|
90
|
64
|
26
|
|
Age
|
|
|
|
0.927
|
≥ 40
|
152
|
72
|
80
|
|
< 40
|
28
|
13
|
15
|
|
Tumor size
|
|
|
|
0.006
|
T1
|
38
|
25
|
13
|
|
T2
|
112
|
52
|
60
|
|
T3
|
30
|
8
|
22
|
|
Histological grades
|
|
|
|
0.013
|
Ⅰ
|
20
|
12
|
8
|
|
Ⅱ
|
65
|
38
|
27
|
|
Ⅲ
|
95
|
35
|
60
|
|
Lymph node metastasis
|
|
|
|
0.023
|
Negative
|
67
|
39
|
28
|
|
Positive
|
113
|
46
|
67
|
|
“+”, positive; “-”, negative. |
Correlation analysis showed that the FAM49B-positive expression rate was significantly higher in ER-negative (ER-) cases than in ER-positive (ER+) cases (p = 0.007, Table 2). Conversely, FAM49B-positivity was significantly higher in HER2 + cases than in HER2- cases (p = 0.011, Table 2). However, there was no significant difference in FAM49B expression with regard to PR positivity (p = 0.071, Table 2).
Table 2
Correlations between FAM49B expression and immunohistochemical markers
Varible
|
n
|
FAM49B−
|
FAM49B+
|
p varible
|
ER
|
|
|
|
0.007
|
-
|
58
|
19
|
39
|
|
+
|
122
|
66
|
56
|
|
PR
|
|
|
|
0.071
|
-
|
64
|
29
|
45
|
|
+
|
116
|
56
|
50
|
|
HER2
|
|
|
|
0.011
|
-
|
137
|
72
|
65
|
|
+
|
43
|
13
|
30
|
|
“+”, positive; “-”, negative. ER, estrogen receptor; HER2, human epidermal growth factor receptor 2; PR, progesterone receptor |
Furthermore, our patient follow-up analysis showed that 45 of 180 patients died, and the 6-year overall survival rate was 85.0%. FAM49B expression was positive in the BC samples of 29 out of the 45 patients that died, while only 16 cases of death occurred in the group with negative FAM49B expression. Kaplan–Meier analysis showed that compared with FAM49B-negative BC patients, the survival rate of FAM49B-positive BC patients was significantly reduced (log-rank test, p < 0.05, HR = 1.874, 95% CI = 1.045–3.362, Fig. 2E).
FAM49B promotes the proliferation and migration of BC cells in vitro
First, FAM49B mRNA expression levels in BC cell lines were evaluated using real-time PCR, which showed that FAM49B was highly expressed in the five BC cell lines (Fig. 3A). MDA-MB-231 and MCF-7 cell lines were selected for subsequent knockdown or overexpression studies.
We used shRNA to knock out FAM49B in MCF-7 and MDA-MB-231 BC cells and confirmed that the infection efficiency of FAM49B-shRNA and scr-shRNA exceeded 80% at 3 days after infection (Fig. 3B). The results of western blotting and real-time PCR showed that the levels of FAM49B protein and mRNA in FAM49B knockout cells were lower than those in scr-shRNA control cells (p < 0.01, Fig. 3C). FAM49B-shRNA#1 was selected for subsequent analyses. The MTT analysis method was used to analyze the proliferation rate of MCF-7 and MDA-MB-231 cells. MCF-7 and MDA-MB-231 cells were infected with FAM49B-shRNA or scr-shRNA. Within 5 days of FAM49B downregulation, the number of cells decreased and the cell proliferation rate was significantly reduced, as assessed by MTT analysis (p < 0.01, Fig. 3D). However, re-expression of FAM49B in the two FAM49B-shRNA BC cell lines completely restored cell proliferation (p < 0.01, Fig. 3D).
Furthermore, wound healing and transwell assays were performed to evaluate the effect of FAM49B on BC cell migration. MDA-MB-231 cells with downregulated FAM49B migrated much slower than shRNA control cells, indicating that the inhibitory effect of FAM49B effectively inhibited cell migration (p < 0.01, Fig. 3E, F). However, reducing FAM49B back to the FAM49B-shRNA BC cell line completely restored cell migration (p < 0.01, Fig. 3E, F). Therefore, FAM49B plays a significant role in cancer growth and metastasis in vitro. FAM49B knockdown inhibits tumor growth and metastasis in vivo.
We investigated whether FAM49B could regulate the growth capacity of BC cells in vivo. MDA-MB-231 cells expressing scr-shRNA or FAM49B-shRNA were implanted into nude mice (n = 10), and tumor progression was monitored for 7 weeks, after which the mice were sacrificed. Compared with the scr-shRNA control group, the volume of MDA-MB-231 tumors expressing FAM49B-shRNA was significantly reduced (p < 0.05, Fig. 4A, B). At the end of the observation period, the tumors of FAM49B-shRNA MDA-MB-231 were significantly reduced in weight compared with the control group (p < 0.05, Fig. 4C).
To test whether FAM49B regulates metastatic potential in vivo, this study quantified lung metastatic nodules following injection of MDA-MB-231-shFAM49B and their corresponding control cells into the caudal vein of nude mice. Compared with control mice, the lung metastasis nodules of mice injected with MDA-MB-231-shFAM49B cells were significantly reduced (p < 0.05, Fig. 4D, E). These results indicate that FAM49B also plays a role in cancer growth and metastasis in vivo.
FAM49B regulates expression of BC genes
To clarify the mechanism by which FAM49B plays a role in BC, we performed a genome-wide expression microarray on MDA-MB-231 cells expressing scr-shRNA or FAM49B-shRNA. Consequently, we detected 1063 genes that showed differential expression (|fold change| ≥ 1.5 and p < 0.05), including 393 upregulated genes and 670 downregulated genes (Fig. 5A). Using on the IPA database, FAM49B knockdown was found to affect the expression of related genes, such as cancer, cell movement, and cell death and survival (Fig. 5B). Knockdown of FAM49B significantly inhibited tumor cell migration and invasion of tumor cells (Fig. 5C). In addition, FAM49B knockdown had a significant inhibitory effect on several key cancer pathways, such as TWEAK, PPAR, and Toll-like receptor signaling pathways (Fig. 5D), indicating that FAM49B can regulate the malignant phenotype of BC.
FAM49B promoted BC cell proliferation and migration by upregulating Rab10/TLR4 pathway
According to the IPA database, the expression of Rab10 and Toll-like receptor 4 (TLR4) mRNA was inhibited when FAM49B was silenced in Toll-like receptor signaling and TLR4 may be the downstream target of Rab10 (Fig. 5A, E). Rab10 can accelerate the transport of TLR4 to the plasma membrane. Rab10 knockout reduced the expression of membrane TLR4 and reduced the production of inflammatory factors induced by LPS [13]. To further study the regulatory mechanism between FAM49B, Rab10, and TLR4 in BC, FAM49B knockdown in the BC cell lines MCF-7 and MDA-MB-231 was performed using shRNA, and we found that FAM49B knockdown significantly inhibited the protein expression of Rab10 and TLR4 in BC cell lines. However, re-expression of FAM49B back into the two FAM49B-shRNA BC cell lines completely restored Rab10 and TLR4 expression (p < 0.01, Fig. 6A), indicating that FAM49B positively regulates Rab10 and TLR4 expression in BC cells.
To identify whether Rab10 is a key factor in this pathway, endogenous Rab10 was silenced in FAM49B-transfected MCF-7 and MDA-MB-231 cells. BC cells transfected with a non-functional vector were used as controls. It was found that Rab10 could inhibit protein expression of TLR4 in the control group and FAM49B upregulated protein expression of TLR4 in the FAM49B overexpression group (p < 0.05, Fig. 6B). However, TLR4 expression was significantly decreased by silencing Rab10 in the FAM49B overexpression group (p < 0.05, Fig. 6B). These results suggest that Rab10 positively regulates TLR4 expression in BC cells and is required in the FAM49B/TLR4 pathway.
To verify whether FAM49B promotes the proliferation and migration of BC cells through Rab10 regulation, Rab10 was knocked down in FAM49B-transfected MDA-MB-231 and MCF-7 cells. We found that proliferation and migration were promoted in FAM49B overexpressing cells, as assessed by MTT assay, wound healing assay, and transwell assay (p < 0.01, Fig. 7A–C). However, the promotion of proliferation and migration mediated by FAM49B overexpression was reversed by Rab10 knockdown (p < 0.01, Fig. 7A–C). These observations demonstrate that Rab10 is required for the FAM49B pathway-mediated migration and proliferation of BC cells.
FAM49B positively regulates Rab10/TLR4 pathway by stabilizing ELAVL1 protein
According to the IPA database, ELAVL1 may be the downstream target of FAM49B and plays a central role in regulating the Rab10/TLR4 pathway (Fig. 5E). ELAV-like RNA binding protein 1 (ELAVL1) is a member of the ELAVL family of RNA-binding proteins that contain several RNA recognition motifs, and it selectively binds AU-rich elements (AREs) found in the 3′ untranslated regions of mRNAs. It is highly expressed in many cancers and could be potentially useful in cancer diagnosis, prognosis, and therapy [14–19]. To identify target proteins downstream of FAM49B, the Pathway Commons Protein-Protein Interactions dataset (http://amp.pharm.mssm.edu/Harmonizome/) was used. This dataset identified ELAVL1 as a potential interactor of FAM49B [20]. Thereafter, we performed a co-immunoprecipitation assay and found that exogenous FAM49B interacted with ELAVL1 in 293 cells (Fig. 8A). Moreover, FAM49B knockdown reduced protein expression of ELAVL1 in MCF-7 and MDA-MB-231 BC cells (p < 0.01, Fig. 8B). Co-transfection of FAM49B into FAM49B-shRNA BC cells completely restored ELAVL1 expression (p < 0.01, Fig. 8B). In addition, FAM49B knockdown did not alter ELAVL1 mRNA expression (Fig. 8C). Therefore, FAM49B may regulate ELAVL1 protein expression at the posttranslational level. To test this hypothesis, we determined whether FAM49B maintained ELAVL1 stability by treating MDA-MB-231 cells with cycloheximide (CHX) to inhibit protein synthesis. The downregulation of FAM49B induced ELAVL1 degradation in MDA-MB-231 cells (Fig. 8D), suggesting that FAM49B stabilizes ELAVL1 in BC cells. Moreover, FAM49B knockdown promoted ELAVL1 ubiquitination (Fig. 8E).
To further understand the role of ELAVL1 in this pathway, endogenous ELAVL1 was silenced in FAM49B-transfected MCF-7 and MDA-MB-231 cells. BC cells transfected with a non-functional vector were used as controls. We found that ELAVL1 could inhibit protein expression of Rab10 and TLR4 in the control group and FAM49B upregulated protein expression of Rab10 and TLR4 in the FAM49B overexpression group (p < 0.01, Fig. 8F, G). However, Rab10 and TLR4 expression was significantly decreased by silencing ELAVL1 in the FAM49B overexpression group (p < 0.01, Fig. 8F, G). These results suggest that ELAVL1 positively regulates Rab10 and TLR4 expression in BC cells and is required in the FAM49B pathway.
FAM49B promotes anthracycline resistance to chemotherapy in triple-negative BC (TNBC) cells by targeting the ELAVL1/Rab10/TLR4/NF-κB signaling pathway
ROC Plotter showed that the level of FAM49B mRNA in BC samples of anthracycline responders was significantly lower than that in BC samples of anthracycline non-responders (p = 5.3e-07, Fig. 9A) [21]. Furthermore, Kaplan–Meier survival analysis showed that high FAM49B mRNA expression was correlated with reduced RFS in BC patients who received chemotherapy (HR = 1.39, p = 0.019, Fig. 9B). These results suggest that high FAM49B expression may inhibit the chemosensitivity of BC. TNBC is generally malignant, and there are no effective targeted drugs. Therefore, chemotherapy is the main treatment method for TNBC [22]. To evaluate whether FAM49B can directly promote anthracycline resistance to chemotherapy in TNBC cells, MDA-MB-231-shFAM49B cells (or control MDA-MB-231 cells) were treated with doxorubicin. The levels of apoptosis were significantly higher in FAM49B-shRNA MDA-MB-231 cells than in control cells, following treatment with 200 ng/mL doxorubicin (p < 0.01, Fig. 9C). We used FAM49B-shRNA MDA-MB-231 cells (or control MDA-MB-231 cells) in a xenograft tumor model (Fig. 9D). The size of tumors formed by the control group cells was slightly reduced by doxorubicin treatment (p > 0.05, Fig. 9E), whereas the size of the tumors formed by FAM49B-shRNA cells was significantly reduced by doxorubicin treatment (p < 0.05, Fig. 9E). These results show that the expression of FAM49B is directly related to an increase in anthracycline resistance via inhibition of apoptosis.
Next, we determined whether FAM49B expression exerts anthracycline resistance in TNBC cells through ELAVL1 and the Rab10/TLR4 signaling pathway. After FAM49B knockout MDA-MB-231 cells were treated with doxorubicin, the protein levels of ELAVL1, Rab10, TLR4, phosphorylate-p65 (p-p65), XIAP, and survivin decreased in a dose-dependent manner (Fig. 9F). In addition, ELAVL1, Rab10, TLR4, p-p65, XIAP, and survivin protein levels were significantly lower in FAM49B-knockdown cells than in control cells following treatment with the corresponding doses of doxorubicin (p < 0.01, Fig. 9F). However, the protein levels of cleaved caspase 3 (c-caspase 3) and cleaved PARP1 (c-PARP1) were significantly higher in FAM49B-knockdown cells than in control cells following treatment with the corresponding doses of doxorubicin (p < 0.05, Fig. 9F). These results suggest that FAM49B may inhibit the apoptosis and pro-apoptotic protein activation in BC cells through the ELAVL1/Rab10/TLR4/NF-κB signaling pathway, resulting in anthracycline resistance.