Circular RNA hsa_circ_0044234 as distinct molecular signature of triple negative breast cancer: a potential regulator of GATA3

Abstract

Background: Circular RNAs (circRNAs) have been implicated in the initiation and development of breast cancer as functional non-coding RNAs (ncRNA). The roles of circRNAs as the competing endogenous RNAs (ceRNAs) to sponge microRNAs (miRNAs) have also been indicated. However, the functions of circRNAs in breast cancer have not been totally elucidated. This study aimed to explore the clinical implications and possible roles of circ_0044234 in carcinogenesis of the most problematic BC subtype, triple negative breast cancer (TNBC), which are in desperate need of biomarkers and targeted therapies.

Methods: The importance of circ_0044234 as one of the most dysregulated circRNAs in TNBC was discovered through microarray expression profile analysis. Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) was performed to confirm the downregulation of circ_0044234 in triple negative tumors and cell lines versus non-triple negative ones. The bioinformatics prediction revealed that circ_0044234 could act as an upstream sponge in the miR-135b/GATA3 axis, two of the most dysregulated transcripts in TNBC.

Results: Our experimental investigation of circ_0044234 expressions in various BC subtypes as well as cell lines reveals that TNBC expresses circ_0044234 at a substantially lower level than non-TNBC. The ROC curve analysis indicates that it could be applied as a discriminative biomarker to identify TNBC from other BC subtypes. Moreover, circ_0044234 expression could be an independent prognostic biomarker in BC. Interestingly, a substantial inverse expression correlation was detected between circ_0044234 and miR-135b-5p as well as between miR-135b-5p and GATA3 in breast tumors.

Conclusions: The possible clinical usefulness of circ_0044234 as a promising distinct biomarker and upcoming therapeutic target for TNBC have been indicated in this research. Our comprehensive approach revealed the potential circ_0044234/miR135b-5p/GATA3 ceRNA axis in TNBC.

Keywords: Bioinformatics; GATA3; Triple negative breast cancer; ceRNA; circ_0044234; miR135b-5p.

Fig. 1

Volcano plot; Differentially expressed circRNAs (DECs) in triple negative breast cancer in comparison with luminal samples from GEO database GSE101124. The vertical and horizontal dashed lines indicate the fold shift (log₂ scale) and P.value (log₁₀ scale) cut-off points, respectively. Red represents downregulated circRNAs and blue represents upregulated circRNAs

Fig. 2

Schematic presentation of hsa_circ_0044234 biogenesis of CDC27 gene. The verification of back-splice intersection location of hsa_circ_0044234 via Sanger sequencing

Fig. 3

The expression of hsa_circ_0044234 in BC relative to normal mammary tissues. a The relative expression levels of circ_0044234 in TNBC in comparison with non-TNBC; b the relative expression levels of hsa_circ_0044234 among luminal A/B, HER2⁺ and triple negative breast cancer subtypes; c the relative expression levels of hsa_circ_0044234 in triple-negative breast cancer cell lines (MDA-MB-468 and MDA-MB-231) compared to luminal ones (MDA-MB-361 and MCF7); d ROC curve analysis of hsa_circ_0044234 expression value for distinguishing TNBC (N = 40) from non-TNBC samples (N = 55); AUC area under the ROC curve

Fig. 4

Association of circ_0044234 expression status with disease-free survival in breast cancer. The black and red lines indicate patients with the high (N = 42) and low (N = 45) expression level of circ_0044234, respectively

Fig. 5

a Schematic diagram shows the potential function of circ_0044234 in sponging miR-135b-5p that promotes GATA3 gene; b The expression levels of miR-135b-5p and GATA3 in TNBC (N = 40) and non-TNBC (N = 55) relative to normal mammary tissues (N = 10)

Fig. 6

a Hierarchical clustering showing expression values (log₂ transformed) of circ_0044234, miR-135b-5p and GATA3 in different subtype of breast cancers (N = 95) relative to normal mammary tissues (N = 10). Red strips represent low relative expression and blue strips represent high relative expression. Each row shows one tumor sample, and each column demonstrates a transcript; b the negative correlation between circ_0044234 and miR-135b-5p expression levels in BC (N = 95, r = − 0.23, P.value = 0.023); c The positive correlation between circ_0044234 and GATA3 expression levels in BC (N = 95, r = 0.21, P.value = 0.041); d The negative correlation between miR-135b-5p and GATA3 expression levels in BC (N = 95, r = − 0.24, P.value = 0.02); Linear regression analysis was used to achieve the regression equation