A Novel Role for Brain and Acute Leukemia Cytoplasmic (BAALC) in Human Breast Cancer Metastasis

Abstract

Brain and Acute Leukemia, Cytoplasmic (BAALC) is a protein that controls leukemia cell proliferation, differentiation, and survival and is overexpressed in several cancer types. The gene is located in the chromosomal region 8q22.3, an area commonly amplified in breast cancer and associated with poor prognosis. However, the expression and potential role of BAALC in breast cancer has not widely been examined. This study investigates BAALC expression in human breast cancers with the aim of determining if it plays a role in the pathogenesis of the disease. BAALC protein expression was examined by immunohistochemistry in breast cancer, and matched lymph node and normal breast tissue samples. The effect of gene expression on overall survival (OS), disease-free and distant metastasis free survival (DMFS) was assessed in silico using the Kaplan-Meier Plotter (n=3,935), the TCGA invasive breast carcinoma (n=960) and GOBO (n=821) data sets. Functional effects of BAALC expression on breast cancer proliferation, migration and invasion were determined in vitro. We demonstrate herein that BAALC expression is progressively increased in primary and breast cancer metastases when compared to normal breast tissue. Increased BAALC mRNA is associated with a reduction in DMFS and disease-free survival, but not OS, in breast cancer patients, even when corrected for tumor grade. We show that overexpression of BAALC in MCF-7 breast cancer cells increases the proliferation, anchorage-independent growth, invasion, and migration capacity of these cells. Conversely, siRNA knockdown of BAALC expression in Hs578T breast cancer cells decreases proliferation, invasion and migration. We identify that this BAALC associated migration and invasion is mediated by focal adhesion kinase (FAK)-dependent signaling and is accompanied by an increase in matrix metalloproteinase (MMP)-9 but not MMP-2 activity in vitro. Our data demonstrate a novel function for BAALC in the control of breast cancer metastasis, offering a potential target for the generation of anti-cancer drugs to prevent breast cancer metastasis.

Keywords: BAALC; FAK; breast cancer; invasion; migration.

Figure 1

BAALC expression is increased in primary breast cancer and lymph node metastases tissues. (A) Normal breast, (B) primary breast cancer, and (C) lymph node metastases were examined for BAALC expression by immunohistochemistry. (D) Negative control. (E) Staining was quantified and expressed as an H-score in 70 primary breast cancer, 40 matched normal breast, and 10 lymph node metastases cores, (F) H-score comparison between tumors classified as Grade 1 + 2 or Grade 3 + 4 from n=70 primary breast cancer in (E) Photomicrographs are representative of each tissue type. * denotes statistical significance p < 0.05 as determined by one-way ANOVA.

Figure 2

High BAALC predicts for worse progression free and distant metastasis free survival in breast cancer patients. Kaplan-Meier curves showing the (A) overall survival and (B) distant metastasis-free survival in a publicly available 3,935-sample breast cancer data set (32) with elevated (red) or low (black) expression of BAALC in breast cancer tumors when assessing all tumor subtypes together. Kaplan-Meier curves showing the (C) disease-free survival in a publicly available 960-sample breast cancer data set (33), with high (red) or low (blue) expression of BAALC. Kaplan-Meier curves showing the (D) distant metastasis free survival in a publicly available 821-sample breast cancer data set (34), with high (red) or low (gray) expression of BAALC. (E) Multivariate analysis for the Kaplan-Meier analysis shown in (D), using estrogen receptor (ER)-status, tumor size, age, histological grade and low BAALC as covariates, distant metastasis free survival as endpoint with no time-dependent censoring. The hazard ratio and the 95% confidence interval are plotted for each of these covariates. p values were computed by a log-rank test.

Figure 3

BAALC overexpression increases breast cancer cell proliferation and anchorage-independent growth in vitro. MCF-7 cells stably transfected with an empty vector (EV) or BAALC were generated. (A) Top BAALC expression was confirmed by western blot, bottom GAPDH as a loading control. (B) Cell viability was measured at 0, 24, 48, and 72 h post-plating via Cell Titre Blue Assay. n=3 in triplicate. (C) Transduced MCF-7 cells were grown for 14 days. After this time colonies were stained with 0.5% crystal violet/10% methanol/PBS for 30 mins. Photomicrographs are representative of three independent experiments, performed in triplicate. (D) Colonies were counted. (E) Transduced MCF-7 cells were grown for 15 days in soft agar before the colonies were stained with 0.5% crystal violet/10% methanol/PBS for 30 mins. Photomicrographs are representative of four independent experiments, performed in duplicate. (F) After staining, the colonies in each well were counted. * denotes statistical significance p < 0.05, as determined by an unpaired two-tailed t-test.

Figure 4

BAALC overexpression increases MCF-7 migration in vitro. MCF-7 cells stably transfected with an empty vector (EV) or BAALC were generated. (A) Confluent monolayers of MCF-7-EV and BAALC cells were grown to confluence, and a wound was made by scratching with a pipette tip. The wounds were photographed at 4, 7, 17, 24 and 48 h to measure wound closure over time. Photomicrographs are indicative of 6 independent experiments performed in triplicate. (B) Wound widths are expressed as % of 0 h wound width. * denotes statistical significance (p < 0.05). (C) MCF-7 cells expressing BAALC or EV were placed in the upper chamber of a Transwell and allowed to migrate through the uncoated membrane (8 µm pore) for 4 h. n=3 performed in triplicate. (D) MCF-7 cells expressing BAALC or EV were examined for ability to invade through Matrigel plugs. n=3 performed in triplicate. * denotes statistical significance p < 0.05.

Figure 5

Knocking down BAALC expression decreases Hs578T proliferation, migration and invasion in vitro. Hs578T cells were transfected with a pool of siRNA directed against BAALC, or a controlled scrambled sequence. (A) Top BAALC expression at various times post-transfection was examined by western blot, bottom GAPDH as a loading control. (B) Cells were plated immediately following transfection and cell viability was measured at 0, 24, 48, and 72 h post-plating via Cell Titre Blue Assay. n=3 in triplicate. (C) Forty-eight hours post-transfection, cells were placed in the upper chamber of a Transwell and allowed to migrate through the uncoated membrane (8 µm pore) for 4 h. n=3 performed in triplicate. (D) Forty-eight hours post-transfections, cells were examined for ability to invade through Matrigel plugs. (E) Endogenous levels of BAALC protein in Hs578T and MCF7 cells. n=3 performed in triplicate. * denotes statistical significance p < 0.05. *** denotes statistical significance p < 0.001.

Figure 6

FAK, but not ERK, inhibition decreases BAALC-mediated MCF-7 migration in vitro. MCF-7 cells stably transfected with an empty vector (EV) or BAALC were generated. (A) The expression and phosphorylation of FAK and ERK following BAALC overexpression were measured by western blot in MCF-7-EV and BAALC cells. Blots are representative of three independent experiments. (B) Confluent monolayers of MCF-7-EV and BAALC cells were grown to confluence, treated for 1 h with either serum-free media (untreated), vehicle, UO126 or PF-562271, and a wound was made by scratching with a pipette tip. The wounds were photographed at 0 and 48 h to measure wound closure. Photomicrographs are indicative of 4 independent experiments performed in triplicate. (C) Wound widths are expressed as % of 0 h wound width. * denotes statistical significance (p < 0.05).

Figure 7

BAALC can interact with FAK and leads to increased active MMP-9 in vitro. MCF-7 cells stably expressing an empty vector (EV) or BAALC were generated. (A) MCF-7 cells were lyzed, and BAALC was immunoprecipitated from cell lysates with an anti-BAALC antibody, or an IgG isotype control. Co-immunoprecipitated FAK was identified by western blot. Alternatively, endogenous FAK was immunoprecipitated with an anti-FAK antibody, and co-immunoprecipitated BAALC identified by western blot. Blots are representative of three independent experiments. (B) Conditioned media from MCF-7-EV and BAALC cells was electrophoresed on 10% gelatin zymography gels, and gels were stained before being visualized for MMP activity. Gels are representative of three independent experiments.