pubmed.ncbi.nlm.nih.gov

Genome-wide DNA methylation profiles in urothelial carcinomas and urothelia at the precancerous stage - PubMed

. 2010 Jan;101(1):231-40.

doi: 10.1111/j.1349-7006.2009.01330.x. Epub 2009 Aug 27.

Affiliations

Genome-wide DNA methylation profiles in urothelial carcinomas and urothelia at the precancerous stage

Naotaka Nishiyama et al. Cancer Sci. 2010 Jan.

Abstract

To clarify genome-wide DNA methylation profiles during multistage urothelial carcinogenesis, bacterial artificial chromosome (BAC) array-based methylated CpG island amplification (BAMCA) was performed in 18 normal urothelia obtained from patients without urothelial carcinomas (UCs) (C), 17 noncancerous urothelia obtained from patients with UCs (N), and 40 UCs. DNA hypo- and hypermethylation on multiple BAC clones was observed even in N compared to C. Principal component analysis revealed progressive DNA methylation alterations from C to N, and to UCs. DNA methylation profiles in N obtained from patients with invasive UCs were inherited by the invasive UCs themselves, that is DNA methylation alterations in N were correlated with the development of more malignant UCs. The combination of DNA methylation status on 83 BAC clones selected by Wilcoxon test was able to completely discriminate N from C, and diagnose N as having a high risk of carcinogenesis, with 100% sensitivity and specificity. The combination of DNA methylation status on 20 BAC clones selected by Wilcoxon test was able to completely discriminate patients who suffered from recurrence after surgery from patients who did not. The combination of DNA methylation status for 11 BAC clones selected by Wilcoxon test was able to completely discriminate patients with UCs of the renal pelvis or ureter who suffered from intravesical metachronous UC development from patients who did not. Genome-wide alterations of DNA methylation may participate in urothelial carcinogenesis from the precancerous stage to UC, and DNA methylation profiling may provide optimal indicators for carcinogenetic risk estimation and prognostication.

PubMed Disclaimer

Figures

Figure 1
Figure 1

DNA methylation alterations during multistage urothelial carcinogenesis. (a) Microscopic view of normal urothelium obtained from a patient without urothelial carcinoma (UC) (C), noncancerous urothelium obtained from a patient with UC (N), and UC (T). N shows no remarkable histological changes in comparison to C, that is no cytological or structural atypia is evident. Hematoxylin–eosin staining. Original magnification, ×20. (b) Scanned array images obtained by bacterial artificial chromosome (BAC) array‐based methylated CpG island amplification (BAMCA) in C, N, and T. Co‐hybridization was done with test and reference DNA labeled with Cy3 and Cy5, respectively. (c) Scattergrams of the signal ratios (test signal/reference signal) obtained by BAMCA in C, N, and T. In all 18 normal urothelia (C1–C18), the signal ratios of 97% of the BAC clones were between 0.67 and 1.5 (red bars). Therefore, in N and T, DNA methylation status corresponding to a signal ratio of less than 0.67 and more than 1.5 was defined as DNA hypomethylation and DNA hypermethylation on each BAC clone compared to C, respectively. Even though N did not show any marked histological changes in comparison to C (panels C and N in [a]), many BAC clones showed DNA hypo‐ or hypermethylation. In T, more BAC clones showed DNA hypo‐ or hypermethylation, whose degree, that is deviation of the signal ratio from 0.67 or 1.5, was increased in comparison to N. (d) Principal component analysis based on BAMCA data (signal ratios). Progressive alterations of DNA methylation status from normal urothelia (yellow arrows) to noncancerous urothelia obtained from patients with UCs (green arrows), and to UCs (red arrows) were observed.

Figure 2
Figure 2

Correlations between DNA methylation status and clinicopathological parameters. (a) Unsupervised two‐dimensional hierarchical clustering analysis based on bacterial artificial chromosome (BAC) array‐based methylated CpG island amplification (BAMCA) data (signal ratios) in noncancerous urothelia obtained from patients with urothelial carcinomas (UCs). The signal ratio is shown in the color range map. Seventeen patients with UCs were hierarchically clustered into two subclasses, Clusters AN (n = 9) and BN (n = 8). Eight patients (100%) belonging to Cluster BN developed invasive UCs (pT2 or more), whereas five patients (55.6%) belonging to Cluster AN did so (P = 0.0311). (b) Scattergrams of the signal ratios in tissue samples. NS, noncancerous urothelia obtained from patients with superficial UCs. NI, noncancerous urothelia obtained from patients with invasive UCs. TI, invasive UCs. If the average signal ratios in NI were significantly higher than those in NS, the average signal ratios in TI themselves were even higher than (BAC clones RP11‐79K14 and RP11‐29C11), or not significantly different from (BAC clones RP11‐3A9 and RP11‐73G16), those in NI without exception. If the average signal ratios in NI were significantly lower than those in NS, the average signal ratios in TI themselves were even lower than (BAC clones RP11‐210F15 and RP11‐368O13), or not significantly different from (BAC clones RP11‐442N24 and RP11‐65C22), those in NI without exception. a P = 0.001680673, b P = 9.23504e‐7, c P = 0.002197802, d P = 3.64223e‐6, e P = 0.000840336, f P = 0.007692306, g P = 0.004395604, h P = 8.31509e‐6, i P = 0.004395604, j P = 1.10173e‐5, k P = 0.005882353, l P = 0.001098901. (c) Unsupervised two‐dimensional hierarchical clustering analysis based on BAMCA data (signal ratios) in UCs. Forty patients with UCs were hierarchically clustered into two subclasses, Clusters AT (n = 19) and BT (n = 21). All four patients with recurrence belonged to Cluster BT. (d) Unsupervised two‐dimensional hierarchical clustering analysis based on BAMCA data (signal ratios) for noncancerous urothelia obtained from patients with UCs of the renal pelvis or ureter. Thirteen patients with UCs of the renal pelvis or ureter were hierarchically clustered into two subclasses, Clusters ANP (n = 4) and BNP (n = 9). All four patients who developed intravesical metachronous UC belonged to Cluster BNP.

Figure 3
Figure 3

DNA methylation profiles discriminating noncancerous urothelia obtained from patients with urothelial carcinomas (UCs) (N) from normal urothelia (C). (a) Scattergrams of the signal ratios in C and N on representative bacterial artificial chromosome (BAC) clones, RP11‐75C8, RP11‐174B4, and RP11‐89G6. Using the cut‐off values (CV) described in each panel, N in this cohort were discriminated from C with sufficient sensitivity and specificity. (b) Histogram showing the number of BAC clones satisfying the criteria listed in Table S1 in samples C1–C18 and N1–N17. Based on this histogram, we established a criterion that when the noncancerous urothelia satisfied the criteria in Table S1 for 50 (green bar) or more than 50 BAC clones, they were judged to be at high risk of carcinogenesis.

Figure 4
Figure 4

DNA methylation profiles in urothelial carcinomas (UCs) associated with recurrence. (a) Scattergrams of the signal ratios in UCs from patients who did not develop recurrence (n = 36) and UCs from patients who developed recurrence (n = 4) on representative bacterial artificial chromosome (BAC), clones, RP11‐79K6, RP11‐466C12, and RP11‐11A11. Using the cut‐off values (CV) described in each panel, recurrence‐positive patients were discriminated from recurrence‐negative patients with 100% sensitivity. (b) Histogram showing the number of BAC clones satisfying the criteria listed in Table S2 in all 40 patients with UCs. Satisfying the criteria in Table S2 for 17 (green bar) or more than 17 BAC clones discriminated recurrence‐positive patients from recurrence‐negative patients with a sensitivity and specificity of 100%, whereas high histological grade (21), invasive growth (pT2 or more), and vascular or lymphatic involvement were unable to achieve such complete discrimination (data not shown).

Figure 5
Figure 5

DNA methylation profiles in noncancerous urothelia obtained from patients with urothelial carcinomas (UCs) of the renal pelvis or ureter associated with intravesical metachronous UC development. (a) Scattergrams of the signal ratios in noncancerous urothelia obtained from patients who did not develop intravesical metachronous UCs (n = 9) and noncancerous urothelia obtained from patients who developed intravesical metachronous UCs (n = 4) on representative bacterial artificial chromosome (BAC) clones, RP11‐721G11, RP11‐104N11, and RP11‐402N14. Using the cut‐off values (CV) described in each panel, metachronous UC‐positive patients were discriminated from metachronous UC‐negative patients with 100% sensitivity and specificity. (b) Histogram showing the number of BAC clones satisfying the criteria listed in Table S3 in all 13 patients with UCs of the renal pelvis or ureter from whom noncancerous urothelia were obtained. Patients who were negative and positive for metachronous UC were confirmed to show a marked difference in the DNA methylation status of the 11 BAC clones.

Similar articles

Cited by

References

    1. Jones PA, Baylin SB. The fundamental role of epigenetic events in cancer. Nat Rev Genet 2002; 3: 415–28. - PubMed
    1. Eden A, Gaudet F, Waghmare A, Jaenisch R. Chromosomal instability and tumors promoted by DNA hypomethylation. Science 2003; 300: 455. - PubMed
    1. Baylin SB, Ohm JE. Epigenetic gene silencing in cancer – a mechanism for early oncogenic pathway addiction? Nat Rev Cancer 2006; 6: 107–16. - PubMed
    1. Gronbaek K, Hother C, Jones PA. Epigenetic changes in cancer. Apmis 2007; 115: 1039–59. - PubMed
    1. Feinberg AP. Phenotypic plasticity and the epigenetics of human disease. Nature 2007; 447: 433–40. - PubMed

Publication types

MeSH terms