Skip to main page content
U.S. flag

An official website of the United States government

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2013 Dec 1;19(23):6544-6555.
doi: 10.1158/1078-0432.CCR-12-3224. Epub 2013 Oct 2.

Novel methylation biomarker panel for the early detection of pancreatic cancer

Joo Mi Yi #  1   2 , Angela A Guzzetta #  3 , Vasudev J Bailey  4 , Stephanie R Downing  3 , Leander Van Neste  5 , Katherine B Chiappinelli  1 , Brian P Keeley  4 , Alejandro Stark  4 , Alexander Herrera  3 , Christopher Wolfgang  3 , Emmanouil P Pappou  3 , Christine A Iacobuzio-Donahue  6 , Michael G Goggins  6 , James G Herman  1 , Tza-Huei Wang  4   7 , Stephen B Baylin  1 , Nita Ahuja  1   3
Affiliations

Novel methylation biomarker panel for the early detection of pancreatic cancer

Joo Mi Yi et al. Clin Cancer Res. .

Abstract

Purpose: Pancreatic cancer is the fourth leading cause of cancer deaths and there currently is no reliable modality for the early detection of this disease. Here, we identify cancer-specific promoter DNA methylation of BNC1 and ADAMTS1 as a promising biomarker detection strategy meriting investigation in pancreatic cancer.

Experimental design: We used a genome-wide pharmacologic transcriptome approach to identify novel cancer-specific DNA methylation alterations in pancreatic cancer cell lines. Of eight promising genes, we focused our studies on BNC1 and ADAMTS1 for further downstream analysis, including methylation and expression. We used a nanoparticle-enabled methylation on beads (MOB) technology to detect early-stage pancreatic cancers by analyzing DNA methylation in patient serum.

Results: We identified two novel genes, BNC1 (92%) and ADAMTS1 (68%), that showed a high frequency of methylation in pancreatic cancers (n = 143), up to 100% in PanIN-3 and 97% in stage I invasive cancers. Using the nanoparticle-enabled MOB technology, these alterations could be detected in serum samples (n = 42) from patients with pancreatic cancer, with a sensitivity for BNC1 of 79% [95% confidence interval (CI), 66%-91%] and for ADAMTS1 of 48% (95% CI, 33%-63%), whereas specificity was 89% for BNC1 (95% CI, 76%-100%) and 92% for ADAMTS1 (95% CI, 82%-100%). Overall sensitivity using both markers is 81% (95% CI, 69%-93%) and specificity is 85% (95% CI, 71%-99%).

Conclusions: Promoter DNA methylation of BNC1 and ADAMTS1 is a potential biomarker to detect early-stage pancreatic cancers. Assaying the promoter methylation status of these genes in circulating DNA from serum is a promising strategy for early detection of pancreatic cancer and has the potential to improve mortality from this disease.

PubMed Disclaimer

Figures

Fig. 1
Fig. 1
Methylation frequencies of the 8 genes in a series of pancreatic cancer patients (stage I-IV; n=147); along with normal pancreas from non-cancerous patients (n=4) and pancreatic intraepithelial neoplasia (PanIN) (n=20; ranging from PanIN-1 to -3).
Fig. 2
Fig. 2
(A-B) Silencing of BNC1 and ADAMTS1 genes in pancreatic cancer cell lines. (A) qMSP analysis of BNC1 and ADAMTS1 gene promoter region and correlation with (B) gene expression by qPCR in pancreatic cancer cell lines. IVD= in vitro methylated DNA. DKO=double knockout (DKO) HCT116 cells (DNMT1−/− and DNMT3b−/−). Quantitative PCR expression is shown as fold change ± standard error relative to mock-treated (M) cells during 5 μM 5-Aza-2’-deoxycytidine (DAC; D) and 300 nM Trichostatin A (TSA; T) treatments. Normal pancreas and DKO cells were used as controls. (C-D) Bisulfite sequencing analysis of CpG island in (C) BNC1 and (D) ADAMTS1 gene promoter region. Bisulfite sequencing analysis of the BNC1 (for Panc1) and ADAMTS1 (for MIA-PaCa2) genes in a pancreatic cancer cell line, a primary pancreatic cancer (stage II), normal pancreatic tissue and DKO cell line as a negative control. Open circles represent unmethylated CpG sites and filled circles represent methylated CpG sites with each row representing a single clone. TIS indicate transcriptional start site. Location of CpG sites (BST: upstream region from −331 to +36 for BNC1, +22 to +207 for ADAMTS1 relative to transcriptional start site). The red and blue bars represent MSP amplification and bisulfite amplification region, respectively.
Fig. 2
Fig. 2
(A-B) Silencing of BNC1 and ADAMTS1 genes in pancreatic cancer cell lines. (A) qMSP analysis of BNC1 and ADAMTS1 gene promoter region and correlation with (B) gene expression by qPCR in pancreatic cancer cell lines. IVD= in vitro methylated DNA. DKO=double knockout (DKO) HCT116 cells (DNMT1−/− and DNMT3b−/−). Quantitative PCR expression is shown as fold change ± standard error relative to mock-treated (M) cells during 5 μM 5-Aza-2’-deoxycytidine (DAC; D) and 300 nM Trichostatin A (TSA; T) treatments. Normal pancreas and DKO cells were used as controls. (C-D) Bisulfite sequencing analysis of CpG island in (C) BNC1 and (D) ADAMTS1 gene promoter region. Bisulfite sequencing analysis of the BNC1 (for Panc1) and ADAMTS1 (for MIA-PaCa2) genes in a pancreatic cancer cell line, a primary pancreatic cancer (stage II), normal pancreatic tissue and DKO cell line as a negative control. Open circles represent unmethylated CpG sites and filled circles represent methylated CpG sites with each row representing a single clone. TIS indicate transcriptional start site. Location of CpG sites (BST: upstream region from −331 to +36 for BNC1, +22 to +207 for ADAMTS1 relative to transcriptional start site). The red and blue bars represent MSP amplification and bisulfite amplification region, respectively.
Fig. 3
Fig. 3
Quantitative MSP (qMSP) analysis of BNC1 and ADAMTS1 using real-time PCR. Normal pancreas (n=4 normal tissues, n=10 surrounding normal tissues), chronic pancreatitis samples (n=30), PanIN 1-3 (n=20), and stage II tumors (n=12), respectively. qMSP showed significantly increased frequency of methylation when comparing normal pancreas tissues and invasive cancers (p<0.001; both BNC1 and ADAMTS1) as well as between chronic pancreatitis and invasive cancers (p<0.001, both BNC1 and ADAMTS1). qMSP is shown as fold change for methylated signal relative to unmethylated signal. In qMSP analysis, signals for unmethylated (U) and methylated (M) DNA are shown for each sample. Horizontal bar indicates mean methylation level.
Fig. 4
Fig. 4
Down-regulation of BNC1 and ADAMTS1 gene expression in primary pancreas cancers. (A) Quantitative mRNA gene expression and (B) DNA methylation analysis of BNC1 and ADAMTS1 in pancreas cancers using qPCR and MSP analysis. PC1-3 indicates primary pancreatic cancer samples. Normal pancreas was used as a control. Expression and methylation are matched from each individual samples. MSP analysis, signals for unmethylated (U) and methylated (M) DNA are shown for each sample. IVD= in vitro methylated DNA. DKO=double knockout HCT116 cells (DNMT1−/− and DNMT3b−/). C) Correlation between methylation and expression in 41 samples from The Cancer Genome Atlas for BNC1. The solid line indicates the linear regression model for all points.
Fig. 5
Fig. 5
Functional assays of BNC1 gene in pancreatic cancer cells. (A) Colony formation in Panc1 and MIA-PaCa2 cells transfected with pcDNA3.1 or BNC1-pcDNA3.1 and grown for 2 weeks in medium containing G418. Results are plotted as the mean colony numbers relative to pcDNA3.1 transfectants in three independent experiments (Panc1; *p=0.0275, MIA-PaCa2; *p=0.0294). (B) Cell proliferation and invasion assays in Panc1 cells (C) Cell proliferation and invasion assays in MIA-PaCa2 cells. Left panel: BNC1 transfected Panc1 cells (BNC1-pcDNA3.1) were compared with control cells transfected with empty vector (pcDNA3.1). Results are plotted as the mean cells number in three different independent experiments (Panc1; *p=0.0469; 72 hrs, **p=0.0086; 96hrs: MIA-PaCa2, *p=0.0469; 48hrs, *p=0.0318; 72hrs, *p=0.0389; 96hrs). Middle panel: Cell proliferation measured by 3H-thymidine incorporation (Panc1; *p=0.0286, MIA-PaCa2; *p=0.0256). Right panel: Invasion of Panc1 and MIA-PaCa2 cells through matrigel-coated transwells relative to control cells transfected with empty vector in three independent experiments (NS: not statistically significant).
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Jemal ASR, Ward E, Hao Y, Xu J, Thun MJ. Cancer statistics, 2009. CA Cancer J Clin. 2009;59:225–49. - PubMed
    1. Klapman J MM. Early detection of pancreatic cancer: why, who, and how to screen. Cancer Control. 2008;15:280–7. - PubMed
    1. Siegel RWE, Brawley O, Jemal A. Cancer statistics, 2011: the impact of eliminating socioeconomic and racial disparities on premature cancer deaths. A Cancer Journal for Clinicians. 2011;61:212–36. - PubMed
    1. Gemmel CEA, Helmstädter L, Riemann JF. Pancreatic cancer screening: state of the art. Expert Rev Gastroenterol Hepatol. 2009;3:89–96. - PubMed
    1. Yeo CJCJ, Sohn TA, Lillemoe KD, Pitt HA, Talamini MA, Hruban RH, Ord SE, Sauter PK, Coleman J, Zahurak ML, Grochow LB, Abrams RA. Six hundred fifty consecutive pancreaticoduodenectomies in the 1990s: pathology, complications, and outcomes. Annual of Surgery. 1997;226:248–57. - PMC - PubMed

Publication types

MeSH terms