This article is mentioned in:

Introduction

Globally, breast cancer is the second most common cancer and the most common cancer in women. In Taiwan, the age-standardized incidence rate (ASIR) of female breast cancer was 70.7 per 100,000 individuals in 2014, and the ASIR gradually increased at a rate of 3.5 per 100,000 person-years (1,2). The age of breast cancer onset is currently 10 years younger than that in Western countries (3). Thus, early detection and early treatment are important objectives for patients with breast cancer (4). At present, diagnosis and disease monitoring primarily involve tissue biopsy and imaging. Tissue biopsy is an invasive method limited to specific regions (5). Mammography is the gold standard for breast cancer detection due to its high sensitivity and specificity; however, numerous women tend to avoid this due the exposure to radiation (6). A novel minimally invasive biomarker for breast cancer is needed.

Liquid biopsy from bodily fluid, such as circulating cell-free DNA (cfDNA), circulating tumor cells and exosomes, is minimally invasive and has proven both convenient and effective in cancer diagnosis (7,8). The quantity of cfDNA in the blood can be used to characterize tumorigenesis, inflammatory disease and stroke (9). The Epi-proColon blood test for circulating methylated DNA has been approved for the detection of colorectal cancer (10,11). It has been revealed that genomic alterations, such as human epidermal growth factor receptor 2 (HER2) amplification, can serve as a predictive biomarker for breast cancer (12); however, the identification of valid diagnostic biomarkers for breast cancer from liquid biopsy is needed.

DNA methylation, occurring at CpG dinucleotides, is a hallmark of cancer (13). As this epigenetic change occurs early in tumorigenesis, it can be used as a biomarker for early detection, disease prognosis and monitoring (14). Several studies have reported the presence of methylated DNA in serum samples from patients with cancer of the gastrointestinal tract, lung, head and neck, liver and breast (15–17). In cases of breast cancer, DNA methylation signature is related to the clinicopathological characteristics of the tumor, such as tumor stage and grade (18,19). It is possible that DNA methylation-mediated epigenetic silencing of tumor suppressors could affect the progression and prognosis of breast cancer (20–22). One previous study demonstrated that hypermethylation of seven biomarkers, including protocadherin β15 (PCDHB15), could differentiate patients with breast cancer into the high- and low-risk groups (23).

PCDHB15 is a member of the cadherin superfamily and calcium-dependent cell-cell adhesion molecules, which encodes for PCDHB15 protein in humans (24,25). Several cell adhesion molecules, such as CDH1 (also known as E-cadherin), act as epithelial-mesenchymal transition suppressors (26). In this regard, the epigenetic silencing of CDH1 has frequently been observed in cases of human cancer, including breast cancer (27–29). Nonetheless, the role played by the epigenetic silencing of PCDHB15 in cases of breast cancer remains unclear. In the present study, PCDHB15 was identified as a potential tumor suppressor gene in breast cancer, based on the observation that PCDHB15 expression is positively correlated with the likelihood of relapse-free survival. The detection of PCDHB15 methylation in serum samples of patients with breast cancer could be a novel minimally invasive biomarker for the diagnosis and prognosis of breast cancer.

Materials and methods

Cell culture

MCF7 and MDA-MB-231 human breast cancer cell lines were maintained in DMEM (Gibco; Thermo Fisher Scientific, Inc.), supplemented with 10% fetal bovine serum (FBS; Invitrogen; Thermo Fisher Scientific, Inc.) and 50 U/ml penicillin/streptomycin (Invitrogen; Thermo Fisher Scientific, Inc.). The cells were incubated under 5% CO2 at 37°C. DNA demethylation involved treating the cells with 0.1 µM DNA methyltransferase (DNMT) inhibitor 5′-aza-2′-deoxycytidine, (5aza; Merck KGaA) or DMSO (as control) at 37°C for 72 h. Culture media and drugs were replenished every 24 h. Following treatment, the cells were harvested for RNA analysis.

Patient samples

All patient samples were collected from the Biobank of the Ditmanson Medical Foundation Chiayi Christian Hospital, Chiayi, Taiwan (Table I). The cancer group (age range, 30–78 years) was comprised of patients with confirmed breast cancer, whereas the control group (age range, 20–53 years) was comprised of patients diagnosed with benign tumors. The inclusion criteria were patients >20 years old and who were undergoing biopsy or mastectomy; exclusion criteria were patients who could not undergo any surgery or blood sampling. Serum samples obtained from patients with breast cancer (n=49) and patients with benign tumors (n=49) were used for quantitative methylation-specific PCR (qMSP) analysis. Briefly, blood samples were drawn into a 10-ml K2-EDTA blood tube (BD Biosciences) and centrifuged at 1,358 × g at room temperature for 10 min, whereupon the serum was collected and stored at −80°C. The present study was approved (approval no. IRB2019006 on 2019/3/5) by the Institutional Review Board of the Ditmanson Medical Foundation Chiayi Christian Hospital (Chiayi, Taiwan) and performed in strict accordance with approved guidelines. Written informed consent was obtained from all participants.

Table I. Summary of cliniopathological data of plasma samples.

Table I.

Summary of cliniopathological data of plasma samples.

	Plasma samples
Clinicopathological characteristic	Cancer (n=49)	Benign (n=49)
Age, years	55.12±10.59a	34.96±9.92
Histological gradeb
Low grade	4
High grade	37
Unknown	8
Estrogen receptor
-	18
+	30
Unknown	1
Progesterone receptor
-	20
+	28
Unknown	1
Human epidermal growth factor receptor 2
-	19
+	28
Unknown	2

a Mean ± SD;

b Grading: Low grade, G1; high grade, G2-3.

Plasmid transfection and colony formation

The full-length human PCDHB15 expression plasmid was a gift from Professor Jun Yu (Chinese University of Hong Kong, Hong Kong). PCDHB15-expressing or empty vectors (5 µg; pCMV6-XL5) were transfected into MDA-MB-231 cells (a triple-negative breast cancer cell line with lower expression of PCDHB15) using TransIT-LT1 Transfection Reagent (Mirus Bio LLC) in accordance with the manufacturer's protocol. After 72 h of incubation at 37°C, the transfection reagents were removed and replaced with fresh medium. Transfected cells were cultured in fresh medium at 37°C prior to further experiments.

For the colony-formation analysis, a total of 1×104 transfected cells per well were seeded in three 6-cm dishes with complete culture medium. Cells were cultured in fresh culture medium at 37°C for 5–7 days, and the culture medium was replaced at intervals of 3 days. Surviving colonies were stained with 0.4% crystal violet (solubilized in 50% methanol) at room temperature for 30 min. The number of colonies (as defined by the size of the colonies >5 pixel2) was then calculated by using ImageJ 1.53e software (National Institutes of Health).

RNA extraction and reverse transcription-quantitative PCR

Total RNA was extracted from the MCF7 and MDA-MB-231 cells using TRIzol® (Invitrogen; Thermo Fisher Scientific, Inc.) in accordance with the manufacturer's protocol. Briefly, 1 µg of total RNA was treated with DNase I (Amplification grade; Invitrogen; Thermo Fisher Scientific, Inc.), before it underwent reverse transcription. First-strand cDNA synthesis was performed using MMLV Reverse Transcriptase (Epicentre; Illumina, Inc.) with oligo dT primers. Briefly, RNA was denatured and the oligo dT primers were annealed at 65°C for 2 min, then chilled on ice for 1 min. The mixture was gently mixed with dNTP, DTT, RNase inhibitor, X10 RT reaction buffer and MMLV Reverse Transcriptase. The final 20-µl mixture was incubated at 37°C for 60 min followed by 85°C for 5 min. qPCR was performed using an ABI Step-One real-time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.) with specific primers and Power SYBR Green Master Mix (Applied Biosystems; Thermo Fisher Scientific, Inc.). The thermocycling conditions were as follows: Initial denaturation at 95°C for 10 min, followed by 40 cycles of 95°C for 15 sec, 60°C for 30 sec and 72°C for 30 sec. The primer sequences were as follows: PCDHB15 forward, 5′-agcctttcaggagaaattcgactaat-3′ and reverse, 5′-gcaccttaacagagacagagcatttt-3′; and GAPDH forward, 5′-ccccttcattgacctcaactacat-3′ and reverse, 5′-cgctcctggaagatggtga-3′. Relative gene expression was calculated by comparing the quantification cycle (Cq) value of PCDHB15 gene against the Cq value of GAPDH in a given sample (i.e., 2−ΔΔCq) (30).

Extraction and bisulphite conversion of DNA

DNA was extracted from serum samples using the QIAamp Circulating Nucleic Acid kit (Qiagen GmbH) in accordance with the manufacturer's protocol. Extracted DNA was bisulphite-modified using the EZ DNA methylation kit (ZYMO Research Corp.) in accordance with the manufacturer's protocol, as previously described (31).

MSP and qMSP

PCDHB15 methylation in serum samples was detected by subjecting the bisulphite-modified DNA to MSP and qMSP, as previously described (32). Briefly, 4 µl bisulphite-converted DNA was subjected to MSP within the specific promoter PCDHB15 region. The thermocycling conditions were as follows: Initial denaturation at 95°C for 10 min, followed by 40 cycles of 95°C for 15 sec, 60°C for 30 sec and 72°C for 30 sec. The PCR products were analyzed by electrophoresis in 10% polyacrylamide gel and subsequent the gel was stained with ethidium bromide. For qMSP analysis, 4 µl bisulphite-converted DNA was subjected to qMSP within the specific promoter PCDHB15 region using an ABI Step One real time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.) and Power SYBR Green Master Mix. The thermocycling conditions were as follows: Initial denaturation at 95°C for 10 min, followed by 40 cycles of 95°C for 15 sec, 60°C for 30 sec and 72°C for 30 sec. The primer sequences for PCDHB15 were as follows: Forward, 5′-acgttttttttaaggaatcg-3′ and reverse, 5′-acgaaccaatatctccga-3′ (130 bp). The presence of cfDNA in serum samples was detected via collagen type II α1 chain (COL2A1) MSP using the forward primer 5′-tct aac aat tat aaa ctc caa cca cca a-3′ and the reverse primer 5′-gggaagatgggatagaagggaatat-3′. The quantity of methylated DNA was determined in terms of Cq value against a standard curve generated using an in vitro methylated DNA-MSP cloned fragment, as previously described (33).

TCGA data analysis

The Cancer Genome Atlas (TCGA) breast cancer (BRCA) Methylation450K dataset was downloaded from UCSC Xena (http://xena.ucsc.edu). The methylation level of CpG sites in PCDHB15 between solid normal tissues and primary tumor tissues were compared and analyzed. Associations between the expression and methylation (cg17023770) of PCDHB15 were analyzed.

Statistical analysis

All statistical analysis was performed using GraphPad Prism Version 5.0 software packages for Windows (GraphPad Software, Inc.). Differences between two groups were analyzed using an unpaired Student's t-test or the Mann-Whitney U test. Pearson's correlation analysis was used to analyze correlation between gene expression, methylation status of a gene or protein expression. Locoregional relapse-free survival was assessed by Kaplan-Meier analysis, and differences between groups were estimated by the log-rank test. P<0.05 was considered to indicate a statistically significant difference.

Results

PCDHB15 is hypermethylated in breast tumor samples compared with that in normal samples

Breast cancer data from TCGA were first used to analyze DNA methylation profiles for PCDHB15. The present study included 785 primary tumors and 98 solid tissue normal samples. The methylation of PCDHB15 was higher in tumor samples than that in normal solid tissue (P<0.001; Fig. 1A and B). As expected, a negative association between the methylation of a particular CG site (cg17023770) in the promoter region and the expression of PCDHB15 was observed in this TCGA cohort (Fig. 1C; r=−0.2, P<0.0001). Kaplan-Meier Plotter (34) revealed that patients with lower PCDHB15 expression were associated with shorter relapse-free survival times (Fig. 1D; HR, 0.74; P=0.00011), but not at all with overall survival times (data not shown).

Figure 1. PCDHB15 may be a tumor suppressor gene that is epigenetically silenced in in breast cancer. (A) Schematic diagram depicting the genomic structure and position of the CG sites (vertical dashes) in the PCDHB15 promoter region (from-1,000 to +400 with respect to the transcriptional start site). The location of the microarray probes (red vertical dashes), bisulphite pyrosequencing (yellow horizontal line) and qMSP primers (blue solid arrows) are indicated. (B) DNA methylation level (β-value from Illumina Infinium 450 K microarray) of the PCDHB15 CpG island from-278 (cg03572772) to +2252 (cg15006101) in solid tissue normal (white) vs. primary tumor (red) in TCGA breast cancer dataset. Primary tumor tissues (n=785) had higher methylation levels than solid tissue normal tissues (n=98). The x-axis indicates the name of the probe on the microarray. (C) Scatter plot showing correlation between PCDHB15 promoter methylation (cg17023770; x-axis) and expression (y-axis) in TCGA breast cancer dataset (n=873). A negative association between promoter methylation and expression was observed. (D) Kaplan-Meier analysis of PCDHB15 mRNA expression in tumor tissues for relapse-free survival of patients with breast cancer. Patients with breast cancer with lower PCDHB15 expression demonstrated shorter relapse-free survival times than patients with higher PCDHB15 expression (log-rank test, P=0.00011). (E) Relative expression level of PCDHB15 mRNA in MCF7 and MDA-MB-231 breast cancer cells. (F) Methylation analysis of PCDHB15 promoter in breast cancer cell lines using bisulphite pyrosequencing. (G) Relative expression level of PCDHB15 in 0.1 µM 5aza-treated MDA-MB-231 breast cancer cells, compared with DMSO control. (H) Ectopic expression of PCDHB15 inhibited tumor proliferation by colony formation assay. MDA-MB-231 breast cancer cells were transfected with empty (control) or PCDHB15 expression vector. Left panel, reverse transcription-quantitative PCR confirmed overexpression of PCDHB15 in MDA-MB-231 cells transiently transfected with PCDHB15 expression vector. Medium panel, MDA-MB-231 breast cancer cells overexpressing PCDHB15 had significantly fewer colonies than the control. Right panel, quantitative analysis of the colony formation assay. Colony formation assay were performed in duplicate and in two independent experiments (mean ± SD). *P<0.05 and **P<0.001. PCDHB15, protocadherin β15; TCGA, The Cancer Genome Atlas; 5aza, 5′-aza-2′-deoxycytidine; qMSP, quantitative methylation-specific PCR.

The association between promoter methylation and the expression of PCDHB15 in breast cancer cell lines was then investigated. It was determined that the expression of PCDHB15 was downregulated in MDA-MB-231 cells compared with that in MCF7 cells (Fig. 1E). Concomitantly, bisulphite pyrosequencing revealed that promoter methylation was higher in MDA-MB-231 cells than that in MCF7 cells at CpG sites located in the upstream promoter region of PCDHB15 (Fig. 1F). Notably, the expression of PCDHB15 in MDA-MB-231 cells was restored upon treatment using DNMT inhibitor (0.1 µM 5aza; Fig. 1G).

To examine the function of PCDHB15, PCDHB15 was overexpressed in MDA-MB-231 breast cancer cells, showing a lower PCDHB15 expression, as compared with MCF7 cells. The induced overexpression of PCDHB15 in MDA-MB-231 cells led to a significant decrease in the number of colonies, compared with the control vector (P<0.05; Fig. 1H). Taken together, these results suggested that PCDHB15 may be a tumor suppressor subject to epigenetic silencing via promoter hypermethylation in breast cancer.

Measuring PCDHB15 methylation in clinical human serum specimens

The present study also sought to determine whether PCDHB15 methylation could be used as a serum biomarker for breast cancer. MSP revealed PCDHB15 methylation in 66% (4/6) of cfDNA extracted from serum samples of patients with breast cancer, but only in 25% of the samples (1/4) from patients with benign tumors (Figs. 2A and S1). The presence of COL2A1 MSP products suggested the presence the cfDNA in serum samples that were devoid of PCDHB15 methylation (Fig. 2A). The PCDHB15 methylation was further examined using qMSP in serum samples from 49 patients with cancer and 49 patients with benign tumors (Table I). The quantity of methylated PCDHB15 was higher in patients with breast cancer than that in samples from patients with benign tumors (P<0.05; Fig. 2B). Based on the cutoff value generated by the area under the receiver operating characteristic curve (0.589; a cutoff value of 1.2 copy number/ml), PCDHB15 methylation in serum samples provided a sensitivity of 40.8% and specificity of 77.6% in breast cancer detection (Fig. 2C; Table II). These results demonstrated the feasibility of using PCDHB15 methylation in the cfDNA of serum samples as a minimally invasive biomarker for breast cancer.

Figure 2. Methylated PCDHB15 level is higher in cfDNA of serum samples of patients with breast cancer. (A) Gel electrophoresis image of PCDHB15 (upper panel) and COL2A1 (lower panel) MSP in cfDNA isolated from serum samples. IVD was a positive control for methylation and H2O was a negative control for PCR. (B) qMSP was performed to determine the amount of methylated PCDHB15 in cfDNA of breast cancer (n=49) and benign tumor samples (n=49). Compared with the benign tumor samples, higher amounts of PCDHB15 were detected in cancer samples. (C) A receiver operating characteristic curve of PCDHB15 methylation in serum samples from 49 patients with breast cancer and 49 patients with benign tumors. The original uncropped gel electrophoresis images can be found in supplementary Fig. S1. *P<0.05. PCDHB15, protocadherin β15; qMSP, quantitative methylation-specific PCR; cfDNA, cell free DNA; IVD, in vitro methylated DNA; AUC, area under the curve; COL21A, collagen type II α1 chain.

Table II. Summary of protocadherin β15 quantitative methylation-specific PCR analysis.

Table II.

Summary of protocadherin β15 quantitative methylation-specific PCR analysis.

Diagnosis	Valid specimens (n)	Positive specimens (n)	Negative specimens (n)	Positive rate (%)
Breast cancer	49	20	29	40.8
Low grade (G1)	4	1	3	25.0
High grade (≥G2)	37	13	24	35.1
Unknown	8	6	2	75.0
Benign	49	11	38	22.4

PCDHB15 methylation is not correlated with other clinicopathological features

Several proteins, including HER2, Ki-67, estrogen receptor and progesterone receptor, are important parameters in subtyping breast cancer and pathogenesis characterization (6). Nonetheless, our analysis did not reveal any correlation between PCDHB15 methylation and any of those clinical parameters (Fig. 3A-D).

Figure 3. PCDHB15 promoter methylation and the clinicopathological features of breast cancer are not associated. (A and B) Scatter plot showing the correlation between PCDHB15 promoter methylation in cfDNA and expression level of (A) Ki67 and (B) PR in breast tumor tissues. (C and D) Dot plot showing the amount of methylated PCDHB15 in cfDNA with expression status of ER (ER+, ER−) and HER (HER+, HER−). PCDHB15, protocadherin β15; PR, progesterone receptor; cfDNA, cell free DNA; ER, estrogen receptor; HER2, human epidermal growth factor receptor 2.

Discussion

PCDHB15 of the protocadherin superfamily is involved in calcium-dependent cell-cell adhesion. The epigenetic silencing of other cadherins (e.g., E-cadherin) has previously been demonstrated; however, the role of PCDHB15 in breast cancer has yet to be fully understood. In the current study, based on the TCGA database and in-house samples, it was determined that PCDHB15 methylation was more pronounced in patients with breast cancer compared with that in patients with benign tumors or in normal controls. Overall, the higher expression level of PCDHB15 was positively associated with relapse-free survival. Further analysis on specific cell lines revealed that PCDHB15 may be a tumor suppressor downregulated via promoter hypermethylation. It is also noteworthy to point out that the correlation between PCDHB15 methylation and relapse-free survival could not be determined, as DNA methylation data was not available from the Kaplan-Meier Plotter. The prognostic significance of PCDHB15 methylation requires further investigation. It was recently reported that PCDHB15, acting as a tumor suppressor through the inhibition of WNT/β-catenin signaling, was epigenetically silenced in KRAS-mutated colorectal cancer (35). This phenomenon can perhaps be attributed to the overexpression of mitochondria glutamate transporter, SLC25A22, in KRAS-mutated colorectal cancer. Nonetheless, determining whether the overexpression of SLC25A22 is responsible for PCDHB15 methylation in breast cancer is worthy of further investigation.

In the present study, PCDHB15 methylation was higher in cfDNA from serum samples of patients with breast cancer than that from patients with benign tumors. Nonetheless, PCDHB15 methylation was not associated with any clinical parameters, thereby suggesting that PCDHB15 methylation may occur early in the carcinogenesis. It also suggested that PCDHB15 methylation is common to all molecular subtypes of breast cancer. The fact that 22.4% (11 out of 49) of the benign tumor samples tested positive for PCDHB15 methylation supports these hypotheses; however, further clinical analysis using a larger sample size and different molecular subtypes will be required to demonstrate whether PCDHB15 methylation is involved in the field defect of breast cancer (36).

The sensitivity of PCDHB15 methylation (40.8%) in breast cancer detection in serum would not allow its use as a sole biomarker; however it could potentially serve as one epigenetic biomarker in a ‘methylation signature panel’ for the diagnosis and/or prognosis of breast cancer (37–39). Previous studies have identified distinct methylation biomarkers indicating the signaling-mediated epigenetic silencing of tumor suppressors, regardless of different subtypes of breast (40–42). Multiple research groups are currently evaluating the methylation signature panel of GSTP1, RASSF1 and RARB for the detection of breast cancer (29). The PITX2 methylation assay, which has been already certified for in vitro diagnosis, has proven effective as a prognostic and predictive biomarker for breast cancer (43). It is possible that methylation of PCDHB15 could be used in conjunction with these markers to form a novel epigenetic panel by which to characterize the progression of breast cancer.

In conclusion, PCDHB15 is a potential tumor suppressor subject to epigenetic silencing via promoter methylation in breast cancer. PCDHB15 methylation in serum cfDNA may provide a novel minimally invasive epigenetic biomarker for the diagnosis and prognosis of breast cancer. Determining whether PCDHB15 methylation is involved in the early carcinogenesis of breast cancer warrants further investigation.

Supplementary Material

Supporting Data

Acknowledgements

Not applicable.

Funding

The present study was supported by grants from the Ministry of Science and Technology, Taiwan (grant no. MOST 108-2314-B-194-003-MY2), the Ditmanson Medical Foundation Chiayi Christian Hospital, Taiwan (grant no. RCN009) and the Center for Innovative Research on Aging Society (grant no. 107-B128-09) from The Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by Ministry of Education in Taiwan.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Authors' contributions

CCC, GLL, CWT, CFW and YCH collected patient samples and analyzed clinical data. GLL, WLH, FCW, PJL, WHH, YMC and YTL performed experiments. YMC, SYY and MWYC performed the bioinformatic analysis. GLL, SYY and MWYC wrote the manuscript. CCY, YCH and MWYC designed the experiments. GLL, WLH and MWYC confirm the authenticity of all the raw data. All authors read and approved the final manuscript.

Ethics approval and consent to participate

The present study was approved (approval no. IRB2019006 on 2019/3/5) by the Institutional Review Board of the Ditmanson Medical Foundation Chiayi Christian Hospital (Chiayi, Taiwan). Written informed consent was obtained from all participants.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References