Measurement of relative copy number of CDKN2A/ARF and CDKN2B in bladder cancer by real-time quantitative PCR and multiplex ligation-dependent probe amplification

Abstract

Many tumors have large homozygous deletions of the CDKN2A locus (encoding p14(ARF) and p16) and of CDKN2B (p15). Our aim was to determine which gene is the major target in bladder cancer. We used quantitative real-time PCR (RTQ-PCR) to determine copy number of p15, of p14(ARF) exon 1beta, and p16 exon 2 in 22 tumor cell lines and 83 bladder tumors, some of which had been assessed previously by duplex PCR. Titration experiments showed that homozygous deletion could be detected in the presence of up to 30% normal DNA. Results for cell lines were compatible with previous cytogenetic analyses. Ten cell lines and 32 tumors (38.5%) had homozygous deletion of at least one target. Thirteen tumors (15.7%) had deletion of all three targets. Two tumors had deletion of p14(ARF) exon 1beta alone and four of p16 exon 2 alone. RTQ-PCR detected more homozygous deletions than duplex PCR. Finally we used a multiplex ligation-dependent probe amplification kit to provide independent confirmation of results. We conclude that with appropriate controls RTQ-PCR is a sensitive and robust method to detect copy number changes in tumors even in the presence of contaminating normal cell DNA.

Figure 1

Genomic organization of the CDKN2A locus showing positions of the Taqman probes.

Figure 2

p14^ARF and p16 gene dosage ratios relative to ALDOB for a DNA series containing 0 to 100% normal DNA and 0 to 100% DNA from a cell line with homozygous deletion of both targets.

Figure 3

Pattern of DNA copy number alterations in bladder tumor cell lines. Gene dosage ratios were calculated relative to ALDOB. ▪, homozygous deletion; ░⃞, under-representation of 9p21 gene; ▨, over-representation of 9p21 gene; □, no copy number change.

Figure 4

Pattern of DNA copy number changes in primary bladder tumors. Gene dosage ratios were calculated relative to PFKL. ▪, homozygous deletion; ░⃞, under-representation of 9p21 gene; ▨, over-representation of 9p21 gene; □, no copy number change. Patterns of deletion are shown for 32 tumors with homozygous deletion of one or more genes (a), 23 tumors with under-representation of one or more genes (b), 10 tumors with over-representation of one or more genes (c), 18 tumors with no copy number change (d).

Figure 5

Relative gene dosage ratios for p16 exon 2 and p14^ARF in 83 primary bladder tumors. Gray areas are those scored as homozygous deletion for each gene (ie, GDR ≤0.36). Values are all relative to PFKL control.

Figure 6

MLPA results for 21 probes on 9p21 for 5 TCC samples. a: Tumor with no copy number change in region. b: Tumor with minimal normal DNA contamination showing clear homozygous deletion with no products for probes encompassing p15, p14^ARF, and p16. c: Tumor with homozygous deletion of all three genes but distinct centromeric breakpoint. d: Tumor with homozygous deletion encompassing p14^ARF and p16 but not p15. e: Tumor scored as having homozygous deletion of p16 exon 2 alone by RTQ-PCR using PFKL as control. MLPA result indicates under-representation rather than homozygous deletion but confirms small region of deletion involving p16 only.