Follistatin is a metastasis suppressor in a mouse model of HER2-positive breast cancer

Abstract

Background: Follistatin (FST) is an intrinsic inhibitor of activin, a member of the transforming growth factor-β superfamily of ligands. The prognostic value of FST and its family members, the follistatin-like (FSTL) proteins, have been studied in various cancers. However, these studies, as well as limited functional analyses of the FSTL proteins, have yielded conflicting results on the role of these proteins in disease progression. Furthermore, very few have been focused on FST itself. We assessed whether FST may be a suppressor of tumorigenesis and/or metastatic progression in breast cancer.

Methods: Using publicly available gene expression data, we examined the expression patterns of FST and INHBA, a subunit of activin, in normal and cancerous breast tissue and the prognostic value of FST in breast cancer metastases, recurrence-free survival, and overall survival. The functional effects of activin and FST on in vitro proliferation, migration, and invasion of breast cancer cells were also examined. FST overexpression in an autochthonous mouse model of breast cancer was then used to assess the in vivo impact of FST on metastatic progression.

Results: Examination of multiple breast cancer datasets revealed that FST expression is reduced in breast cancers compared with normal tissue and that low FST expression predicts increased metastasis and reduced overall survival. FST expression was also reduced in a mouse model of HER2/Neu-induced metastatic breast cancer. We found that FST blocks activin-induced breast epithelial cell migration in vitro, suggesting that its loss may promote breast cancer aggressiveness. To directly determine if FST restoration could inhibit metastatic progression, we transgenically expressed FST in the HER2/Neu model. Although FST had no impact on tumor initiation or growth, it completely blocked the formation of lung metastases.

Conclusions: These data indicate that FST is a bona fide metastasis suppressor in this mouse model and support future efforts to develop an FST mimetic to suppress metastatic progression.

Keywords: Activin; Breast cancer; Follistatin; Metastasis; Migration.

Fig. 1

Expression of the INHBA/follistatin (FST) axis is altered in human breast cancers. Relative (a) INHBA and (b) FST expression in normal breast versus breast cancer using publicly available datasets from The Cancer Genome Atlas (TCGA). c Relative FST expression in normal breast stroma versus breast cancer stroma in the Finak et al. dataset. d Western blot analysis of FST protein expression in immortalized, nontransformed mammary epithelial cells (NT) and breast cancer cell lines representing luminal, human epidermal growth factor receptor 2-positive (HER2+), and triple-negative breast cancer (TNBC) subtypes

Fig. 2

Follistatin (FST) treatment inhibits migration, and elevated FST expression is correlated with better breast cancer outcomes. Modified Boyden chamber migration assays of (a) MCF10A and (b) 4T1 cells treated with vehicle (Veh), 100 ng/ml recombinant human (rh) activin A (Act), or activin A plus rhFST (400 ng/ml). Migrated cells were quantified in five independent fields per filter and normalized to vehicle (*p < 0.05 compared with vehicle, ^† p < 0.05 compared with activin). c Kaplan-Meier analysis of recurrence-free survival of patients with breast cancer with tumors stratified by FST expression using a 3554-patient cohort representing all breast cancer subtypes (FST low, n = 967; FST high, n = 2587). d Kaplan-Meier analysis of the prognostic significance of FST expression on breast cancer metastasis to the brain/lung in the Bos et al. [24] cohort (FST low, n = 149; FST high, n = 17)

Fig. 3

Follistatin (FST) expression is repressed in mouse mammary tumors and is restored using an Fst-expressing transgene. a Endogenous Fst expression in normal mouse mammary glands compared with human epidermal growth factor receptor 2, receptor tyrosine protein kinase erbB-2, proto-oncogene Neu (Her2/Neu)-induced mammary tumors (*p < 0.05). b Immunohistochemical analysis of endogenous FST expression in mouse mammary epithelia and tumors. c Fst expression in Neu (single) and Neu/FST (bitransgenic) tumors. Fst messenger RNA expression was assessed by quantitative reverse transcription-polymerase chain reaction (**p < 0.01). d Hematoxylin and eosin (H&E)-stained mammary tumors from Neu (single) and Neu/FST (bitransgenic) mice are shown in the upper two panels. Immunohistochemistry for total (endogenous and transgenic) FST expression in lower two panels in single (left) and bitransgenic (right) tumors

Fig. 4

Restoring follistatin (FST) expression abrogates lung metastases in a mouse model of metastatic breast cancer. a Kaplan-Meier plot comparing the percentage of tumor-free mice in a cohort of single transgenic Neu (n = 11) and bitransgenic Neu/FST (n = 8) mice. b Fold change in tumor volume after detection by palpation in single and bitransgenic cohorts. c Representative × 4 magnification images of hematoxylin and eosin-stained lung sections depicting metastases in single and bitransgenic mice. Arrows indicate metastases. d Quantitation of the number of true lung metastases (p < 0.01) and emboli (p < 0.05) in each mouse examined (Neu, n = 10 mice; Neu/FST, n = 8 mice). Statistical significance was determined using Fisher’s exact test