Decreased expression of key tumour suppressor microRNAs is associated with lymph node metastases in triple negative breast cancer

Abstract

Background: Breast cancer is the most common malignancy that develops in women, responsible for the highest cancer-associated death rates. Triple negative breast cancers represent an important subtype that have an aggressive clinical phenotype, are associated with a higher likelihood of metastasis and are not responsive to current targeted therapies. miRNAs have emerged as an attractive candidate for molecular biomarkers and treatment targets in breast cancer, but their role in the progression of triple negative breast cancer remains largely unexplored.

Methods: This study has investigated miRNA expression profiles in 31 primary triple negative breast cancer cases and in 13 matched lymph node metastases compared with 23 matched normal breast tissues to determine miRNAs associated with the initiation of this disease subtype and those associated with its metastasis.

Results: 71 miRNAs were differentially expressed in triple negative breast cancer, the majority of which have previously been associated with breast cancer, including members of the miR-200 family and the miR-17-92 oncogenic cluster, suggesting that the majority of miRNAs involved in the initiation of triple negative breast cancer are not subtype specific. However, the repertoire of miRNAs expressed in lymph node negative and lymph node positive triple negative breast cancers were largely distinct from one another. In particular, miRNA profiles associated with lymph node negative disease tended to be up-regulated, while those associated with lymph node positive disease were down-regulated and largely overlapped with the profiles of their matched lymph node metastases. From this, 27 miRNAs were identified that are associated with metastatic capability in the triple negative breast cancer subtype.

Conclusions: These results provide novel insight into the repertoire of miRNAs that contribute to the initiation of and progression to lymph node metastasis in triple negative breast cancer and have important implications for the treatment of this breast cancer subtype.

Figure 1

miRNAs significantly different in triple negative breast cancer. A) Supervised hierarchical cluster analysis was performed on 71 miRNAs significantly altered between tumour (n = 31, red branches, top) and normal adjacent tissue samples (n = 23, blue branches, top). Similarity in the expression patterns between miRNAs (branches shown on the left-hand panel) and between samples (branches shown on the top panel) was measured using Pearsons correlation. Distances between clustered branches represent the average distances between miRNAs and samples in the cluster. The height of each branch represents the degree of similarity within the cluster. miRNAs are coloured according to their expression level, where up-regulated expression is represented by red, down-regulated expression is represented by blue, and equal expression is represented by yellow. B) Relative quantification of miR-210, miR-100, miR-130a, let-7b and let-7c by real-time RT-PCR in normal (n = 24) and tumour samples (n = 35). Results are shown as a scatter plot of the relative normalised expression (target/RNU44) of the target miRNA (2^-ΔCt). Values represent the median ± interquartile range. *p < 0.0001, **p = 0.0004. C) Correlation between real-time PCR and miRNA array results. Relative quantification of fold change (IDC v NAT) in miR-210, miR-100, miR-130a, let-7b and let-7c between normal and tumour samples by real-time RT-PCR (2^-ΔΔCt, y-axis) or miRNA array (x-axis).

Figure 2

miRNAs significantly different in triple negative breast cancer with lymph node metastases. Supervised hierarchical cluster analysis was performed on miRNAs significantly altered between tumour (red branches, top) and normal adjacent tissue samples (blue branches, top) in A) lymph node negative and B) lymph node positive patients. Similarity in the expression patterns between miRNAs (branches shown on the left-hand panel) and between samples (branches shown on the top panel) was measured using Pearsons correlation. Distances between clustered branches represent the average distances between miRNAs and samples in the cluster. The height of each branch represents the degree of similarity within the cluster. miRNAs are coloured according to their expression level, where up-regulated expression is represented by red, down-regulated expression is represented by blue, and equal expression is represented by yellow. C) Venn diagram representing the overlap between miRNAs regulated in lymph node negative patients (37 miRNAs, left-hand circle), lymph node positive patients (46 miRNAs, right-hand circle) and in the lymph node metastases of lymph node positive patients (63 miRNAs, bottom circle). D) Histogram depicting the fold regulation of the 27 miRNAs (highlighted in red in C) in tumour compared to NAT in lymph node negative tumours (white), lymph node positive tumours (black) and lymph node metastases (grey). Two-fold regulation is depicted by the red line.

Figure 3

Validation of selected miRNAs significantly different in triple negative breast cancer with lymph node metastases. Relative quantification of let-7b, miR-130a, miR-101, miR-26a, miR-100, miR-210, miR-26b, miR-29c and let-7a by real-time RT-PCR in normal and tumour samples from lymph node negative and lymph node positive patients. Results are shown as a box plot of the relative normalised expression (target/RNU44) of the target miRNA (2^-ΔCt). Boxes represent the median ± interquartile range. *p < 0.05, **p < 0.01, ***p < 0.0005 and ⁺p < 0.0001.