Differential microRNA expression signatures and cell type-specific association with Taxol resistance in ovarian cancer cells

Abstract

Paclitaxel (Taxol) resistance remains a major obstacle for the successful treatment of ovarian cancer. MicroRNAs (miRNAs) have oncogenic and tumor suppressor activity and are associated with poor prognosis phenotypes. miRNA screenings for this drug resistance are needed to estimate the prognosis of the disease and find better drug targets. miRNAs that were differentially expressed in Taxol-resistant ovarian cancer cells, compared with Taxol-sensitive cells, were screened by Illumina Human MicroRNA Expression BeadChips. Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) was used to identify target genes of selected miRNAs. Kaplan–Meier survival analysis was applied to identify dysregulated miRNAs in ovarian cancer patients using data from The Cancer Genome Atlas. A total of 82 miRNAs were identified in ovarian carcinoma cells compared to normal ovarian cells. miR-141, miR-106a, miR-200c, miR-96, and miR-378 were overexpressed, and miR-411, miR-432, miR-494, miR-409-3p, and miR-655 were underexpressed in ovarian cancer cells. Seventeen miRNAs were overexpressed in Taxol-resistant cells, including miR-663, miR-622, and HS_188. Underexpressed miRNAs in Taxol-sensitive cells included miR-497, miR-187, miR-195, and miR-107. We further showed miR-663 and miR-622 as significant prognosis markers of the chemo-resistant patient group. In particular, the downregulation of the two miRNAs was associated with better survival, perhaps increasing the sensitivity of cancer cells to Taxol. In the chemo-sensitive patient group, only miR-647 could be a prognosis marker. These miRNAs inhibit several interacting genes of p53 networks, especially in TUOS-3 and TUOS-4, and showed cell line-specific inhibition effects. Taken together, the data indicate that the three miRNAs are closely associated with Taxol resistance and potentially better prognosis factors. Our results suggest that these miRNAs were successfully and reliably identified and would be used in the development of miRNA therapies in treating ovarian cancer.

Keywords:

• microRNA
• ovarian cancer
• Taxol resistance
• Kaplan–Meier survival analysis

Supplementary material

Figure S1 Differential miRNA expression signatures between five ovarian cancer cells and two normal ovarian cells.

Notes: We used 556 filtered miRNA sets using the Illumina human miRNA microarray dataset. This analysis identified 556 up- and downregulated miRNAs, compared with the normal control. To understand the miRNA interactions and visualize the relationship, a heat map was made on the basis of their expression. Each miRNA is presented in matrix format, where rows represent individual miRNA, respectively, and columns represent each ovarian cell. Each cell in the matrix represents the expression level of a miRNA in an individual cell line. Red and green cells reflect high and low expression levels, respectively.
Abbreviation: miRNA, micro rubonucleic acid.

Figure S2 Differential miRNA expression signatures between five ovarian cancer cells and two normal ovarian cells.

Notes: (A) We identified 82 resulting miRNA sets that can directly interact with each probe using the Illumina human miRNA microarray dataset (P<0.01). This analysis identified 45 upregulated miRNAs and 37 downregulated miRNAs, compared with the normal control, and allowed the robust segregation of two groups. To understand the miRNA interactions and visualize the relationship, a heat map was made on the basis of their expression. Each miRNA is presented in matrix format, where rows represent individual miRNA, respectively, and columns represent each ovarian cell. Each cell in the matrix represents the expression level of a miRNA in an individual cell line. Red and green cells reflect high and low expression levels, respectively. (B) In vitro gene expression patterns from six ovarian cancer cells and two normal ovarian cells. The expression of genes’ mRNA in each cell line was examined by qRT-PCR. The results are presented as fold changes (log2) of transcript levels relative to the level in the IOSE386 cells by using the C_T method.
Abbreviations: C_T, threshold cycle; qRT-PCR, quantitative reverse transcription-polymerase chain reaction; miRNA, micro rubonucleic acid.

Figure S3 Fold changes (log2[ovarian cancer cells/normal ovarian cells]) in gene expression measured by qRT-PCR.

Notes: In vitro gene expression patterns using six ovarian cancer cells and two normal ovarian cells. The expression of genes’ mRNA in each cell line was examined by qRT-PCR. The results are presented as transcript levels relative to the level in the IOSE386 cells by using the C_T method.
Abbreviations: C_T, threshold cycle; qRT-PCR, quantitative reverse transcription-polymerase chain reaction; miRNA, micro rubonucleic acid.

Figure S4 Taxol resistance as confirmed by MTT assay.

Notes: Cell viability assay was used to determine Taxol resistance. After incubation with Taxol for 3 days, 100 μL of MTT solution (2 mg/mL) was added to each well and cultured for 4 hours. Taxol (Sigma-Aldrich, St Louis, MO, USA).
Abbreviation: MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide.

Figure S5 Fold changes (ovarian cancer cells/normal ovarian cells) in miRNA expression measured by qRT-PCR.

Notes: (A) Three miRNAs’ expression patterns in vitro using eight ovarian cancer cells and two normal ovarian cells. (B) Three miRNAs’ expression patterns in vitro using two ovarian cancer cells and two normal ovarian cells. The results are presented as transcript levels relative to the level in the IOSE386 cells by using the C_T method. (C) Three miRNAs’ expression patterns in vitro using two ovarian cancer cells. The expression of miRNAs in each cell line was examined by qRT-PCR. The results are presented as transcript levels relative to the level in each parental cell line by using the C_T method.
Abbreviations: C_T, threshold cycle; qRT-PCR, quantitative reverse transcription-polymerase chain reaction; miRNA, micro rubonucleic acid.

Figure S6 Fold changes (Taxol-resistant ovarian cancer cells/parental Taxol-sensitive cells) in gene expression measured by qRT-PCR.

Notes: In vitro gene expression patterns using eight ovarian cancer cells. The expression of genes in each cell line was examined by qRT-PCR. The results are presented as transcript levels relative to the level in each parental cell line by using the C_T method. Taxol (Sigma-Aldrich, St Louis, MO, USA).
Abbreviations: C_T, threshold cycle; qRT-PCR, quantitative reverse transcription-polymerase chain reaction; miRNA, micro rubonucleic acid.

Figure S7 miRNA expression-dominant groups correlated with survival in Taxol resistance.

Notes: (A) Kaplan–Meier analysis showed that miR-136 has a significant difference between over- and underexpression of the resistant group and the sensitive group. (B) The differential expressions of the miRNAs were validated using cross validation using 75% of the dataset (68 patients as chemo-resistant). Red and blue reflect high and low expression levels, respectively. The difference between the two groups was significant when P<0.05. Taxol (Sigma-Aldrich, St Louis, MO, USA).
Abbreviation: miRNA, micro rubonucleic acid.

Figure S8 Inhibition effects of target genes of three miRNA in A2780/A2780-Tx and SKOV3/SKOV3-Tx.

Notes: In vitro gene expression patterns using four ovarian cancer cells were examined by qRT-PCR. The results are presented as transcript levels relative to the level in each parental sensitive cell line by using the C_T method.
Abbreviations: C_T, threshold cycle; qRT-PCR, quantitative reverse transcription-polymerase chain reaction; miRNA, micro rubonucleic acid.

Figure S9 Integrated view of miR-663 regulated by miRNA.

Notes: (A) Integrated Circos plot shows miR-663 regulated by miRNA. An ideogram of a normal karyotype is shown in the outer ring. In the center of the figure, each arc indicates a predicted regulatory relationship between miR-663 and a miRNA. The colored arcs represent predicted regulation by key miRNAs. (B) miR-663-miRNA network shows the relationships that they are predicted to regulate.
Abbreviation: miRNA, micro rubonucleic acid.

Figure S10 Integrated view of miR-622 regulated by miRNA.

Notes: (A) Integrated Circos plot shows miR-622 regulated by miRNA. An ideogram of a normal karyotype is shown in the outer ring. In the center of the figure, each arc indicates a predicted regulatory relationship between miR-622 and a miRNA. The colored arcs represent predicted regulation by key miRNAs. (B) miR-622-miRNA network shows the relationships that they are predicted to regulate.
Abbreviation: miRNA, micro rubonucleic acid.

Acknowledgments

We deeply thank PhD student Pankaj Kumar Chaturvedi for his active advice and support. We are also indebted to PhD student Gantumur Battogtokh for his efforts.

Disclosure

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MEST) (NRF-2012R1A2A1A03670430) and a grant (Industry-Ac ademic Cooperation Foundation program) from the Diatech Korea Co. Ltd, Seoul, Republic of Korea. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The authors report no other conflicts of interest in this work.