Abstract
Background: A genetic syndrome of cutaneous malignant melanoma and nervous system tumors recently has been characterized and shown to be linked to the INK4 locus in the 9p21 region. Hemizygosity at adjacent physically mapped microsatellite markers indicated deletion of p16, p19, and p15 clustered tumor suppressors. Because individuals from this family could benefit from predictive testing in terms of cancer prevention, we developed a direct test without need to analyze parental DNAs to comply with the rules of individual consent and secrecy.
Methods: We developed an assay using TaqManTM real-time quantitative PCR, with p15 as the test sequence and albumin (ALB) as the reference gene. The normalized ratio of p15/ALB is expected to yield a value of ∼1 in individuals without the deletion, whereas a ratio of ∼0.5, indicating p15 haploinsufficiency, is expected in predisposed individuals.
Results: All patients harboring the previously defined at-risk haplotype were correctly identified using this approach. In six individuals with deletions, the p15/ALB ratios were 0.472–0.556 (SD, 0.013–0.078). In the five individuals without deletions, the ratios were 0.919–1.019 (SD, 0.006–0.075).
Conclusions: This is the first report of a high-throughput, automatable gene dosage assay successfully applied to the identification of a germ-line deletion. This approach, not limited by marker informativeness or the need for harvesting live cells, can be applied to any condition with haploinsufficiency and extended to the characterization of most abnormalities of the ploidy.
A genetic syndrome consisting of cutaneous malignant melanoma (CMM)1 and nervous system tumors (NSTs) recently has been defined on the basis of a five-generation family with a predisposing allele to CMM/dysplastic nevi, astrocytoma/glioblastoma, neurofibroma, schwannoma, and meningioma (1). This CMM-NST syndrome recently has been shown to be linked to the INK4 locus, with all patients sharing a common haplotype as defined by genetic markers of the 9p21 region, and has been demonstrated to result from a large germ-line deletion that ablates the whole cluster containing the p16, p19, and p15 tumor suppressor genes (2). Such a large DNA lesion was characterized by hemizygosity mapping based on the segregation of adjacent physically mapped microsatellite markers. We are now addressing the question of direct genetic testing aimed at identification of subjects with individual risk among the 100 initially surveyed. Indeed, a direct test that did not require the analysis of parental DNAs, as opposed to indirect microsatellite-based diagnosis, was a prerequisite for complying with the rules of individual consent and secrecy.
The recent development of real-time quantitative PCR based on the 5′-3′ exonuclease activity of the Taq polymerase (3), referred as TaqManTM, has offered the opportunity to set up an original gene dosage assay using the ABI PRISM 7700 Sequence Detection System (Perkin-Elmer Biosystems), in which p15 was selected as the test sequence and albumin (ALB) as the reference disomic gene used to normalize the amounts of input genomic DNAs. Briefly, during the PCR process, a dual-labeled TaqMan probe annealed to the target sequence is cleaved by the 5′-3′ exonuclease activity of the Taq polymerase, releasing the reporter dye (FAM) from the quencher dye (TAMRA). Upon excitation of by an argon laser, the release of the FAM produces an increase in the fluorescent emission, which is captured at 518 nm by a charged-couple device camera and analyzed through the algorithms of the ABI PRISM 7700 Sequence Detection System computer software (4)(5).
To develop a direct automatable predictive test in the family with INK4 locus haploinsufficiency, this p15 gene dosage assay was applied to a panel of 11 test, haplotype-positive or -negative, individuals from this large kindred.
Patients and Methods
patients, nucleic acids, and reference dna
High-molecular weight DNAs were prepared using standard proteinase K digestions followed by phenol-chloroform extractions (6) from whole-blood leukocytes or lymphoblastoid cell lines from 11 individuals, including 3 clinically affected and 3 healthy, all haplotype-sharing, relatives, as well as in 5 healthy haplotype-negative relatives, including 1 spouse (Fig. 1⇓ ). The haplotypes have been defined elsewhere (1). High-quality human genomic DNA (Boehringer Mannheim) served as reference DNA.
Test individuals from the CMM-NST family.
Squares refer to males, and circles refer to females. Hatched symbols denote the development of CMM or NST. The haplotype status of each test relative is indicated beneath his or her individual symbol, INK4 refers to the at-risk haplotype, and wt refers to any other wild-type haplotype, as previously defined and according to an individual numbering system based on a pedigree published elsewhere (2).
real-time pcr
In the TaqMan approach, one of the main variables, referred as Ct, is defined as the fractional cycle number at which the fluorescence generated by cleavage of the probe (ΔRn) crosses a fixed threshold. The target gene copy number in unknown samples is inferred by plotting the Ct value against a calibration curve. To correctly determine the starting copy number regardless of the precise amounts and qualities of input genomic DNAs, we also quantified an internal control gene (ALB) in each single reaction. The normalized gene dose, N, is given by the following ratio: 
The 7700 Sequence Detection System software automatically determines the Ct value and infers the starting copy number in each sample. The real-time PCR and the calibration curve are presented in Fig. 2⇓ for ALB. Normalized gene doses N were then determined for each sample in three independent assays, and the corresponding means were calculated. In this method, a nondeleted test sequence is expected to yield a ratio of N = 1, as opposed to N = 0.5 when it is heterozygously deleted.
Albumin (ALB) gene dosage by real-time PCR.
Top, amplification plots for reactions with starting ALB gene copy number of 33 000 (A1, 100 ng), 8250 (A4, 25 ng), 2062 (A7, 6.25 ng), or 515 (A10, 1.56 ng). The cycle number is plotted vs the change in normalized reporter signal (ΔRn). For each reaction tube, the fluorescence signal of the reporter dye (FAM) is divided by the fluorescence signal of the passive reference dye (ROX) to obtain a ratio defined as the normalized reporter signal (Rn). ΔRn represents the normalized reporter signal (Rn) minus the baseline signal established in the first 15 PCR cycles. ΔRn increases during PCR as ALB PCR product copy number increases until the reaction reaches a plateau. Ct represents the fractional cycle number at which a significant increase in Rn above a baseline signal (horizontal black line) can first be detected. Three replicates were performed for each reference DNA sample, but the data for only one are shown here. Bottom, calibration curve plotting log starting copy number vs Ct. The black symbols represent the triplicate PCR amplification of the reference DNA samples and red symbols the triplicate PCR amplification of unknown genomic DNA, all included inside the calibration curve. The copy number of ALB (x) can be calculated as follows: y = −3.374x + 40.593, where the Ct value is substituted as y.
primers and probes
Primers and probes were chosen with assistance of the computer programs OligoTM, Ver. 4.0 (National Biosciences) and Primer ExpressTM, Ver. 1.0 (PE Biosystems). Their nucleotide sequences are shown in Table 1⇓ . Primers were purchased from Life Technologies, and probes were purchased from PE Biosystems. The TaqMan PCR Core Reagent Kit, MicroAmp Optical TubesTM, and MicroAmp Optical CapsTM were from PE Biosystems.
Predictive testing based on a p15 TaqMan assay.1
Primers and probes.
pcr amplification
TaqMan amplification reactions were carried out in reaction volumes of 50 μL, using the components as supplied in the TaqMan PCR Core Reagent Kit. Each reaction contained 1× TaqMan buffer; 200 nmol/L each primer; 100 nmol/L each corresponding fluorogenic probe; 5 mmol/L MgCl2, 200 μmol/L each dATP, dCTP, and dGTP; 400 μmol/L dUTP; 1.25 U of AmpliTaqTM Gold, and 0.5 U of AmpEraseTM uracil N-glycosylase. Each sample was analyzed in triplicate, using 20 ng of DNA in each reaction. The reference DNA was serially diluted and run in parallel to establish the calibration curve and to infer copy numbers from the cycle thresholds (Cts), assuming a conversion factor of 6.6 pg of DNA per diploid genome. Thermal cycling was initiated with a 2-min incubation at 50 °C, followed by a first denaturation step of 10 min at 95 °C, and then 40 cycles of 95 °C for 15 s and 65 °C for 1 min.
Results
The results are presented in Table 2⇑ as p15 gene dosage performed in triplicate and in three independent assays. Although the ALB and p15 genes were quantified on the same plate, the low interassay variability, which the use of a calibration curve helped reduce, allowed us to ascertain these genes in different plates. In addition, accurate results can be obtained with duplicate amplification.
The six haplotype-sharing affected or healthy individuals showed INK4 haploinsufficiency, i.e., were heterozygotes for the INK4 locus deletion [N ≈ 0.5 (0.472–0.556; SD, 0.013–0.078)]. This is in contrast to healthy, haplotype-negative individuals who were shown to contain two copies of the p15 gene per cell [N ≈ 1 (0.919–1.019; SD, 0.006–0.075)]. Therefore, all results were clear-cut with relatively small SDs, rendering the deletional status unambiguous.
Discussion
Since 1985, the identification of point mutations has been greatly facilitated by the development of PCR-based assays, but diagnosis of large gene rearrangements still awaited rapid and efficient technology. Recently, analysis of the loss of heterozygosity at a given locus by the use of highly polymorphic microsatellite markers allowed the characterization of a large number of gene deletions, especially in DNAs derived from tumor tissues. Nevertheless, this technology is limited by the presence of polymorphic markers and by their informativeness. In 1998, Cairns et al. (7) published a comparative study between microsatellite and quantitative PCR analyses to detect p16 copy number in primary bladder tumors and concluded that quantitative PCR, which is rapid, accurate, and sensitive, was certainly the most efficient method to identify genetic changes leading to tumorigenesis.
In the aim to identify subjects with individual risk in a large French family diagnosed with a CMM-NST syndrome associated with a whole INK4 locus germ-line deletion, a combination of microsatellite markers of the 9p21 region, flanking the deletion and defining the at-risk haplotype was initially studied. However, this indirect test, which was limited by marker informativeness and required the analysis of nuclear families, possibly including both deceased parents and minor children, was not compatible with the present rules of individual consent and secrecy. Thus, a TaqMan PCR-based gene dosage assay based on the quantification of the p15 gene was developed. This direct and accurate test allowed us to rapidly analyze all the family members willing to benefit from predictive testing and to identify those at risk.
After an initial optimization step for adequate primer pairs and probes, the TaqMan gene dosage method can be used efficiently to assay any chromosome, gene, or exon and thus has a wide variety of applications in both clinical and research settings. This is especially the case with autosomal dominant conditions with haploinsufficiency, such as familial breast and ovarian cancer (8), neurofibromatosis type 1 (9), or in carrier detection in X-linked disorders such as Duchenne muscular dystrophy (10). We believe this high-throughput, fast-turnaround time (up to 96 samples/2 h, which corresponds to 12 individuals tested using our approach), and simple method is clearly a cost-effective alternative to fluorescence in situ hybridization in clinical applications. In this respect, TaqMan lends itself particularly well to routine diagnosis of common microdeletion syndromes. Because of its accuracy, this novel approach, more reliable than end-point PCR gene dosage (11) and not limited by short tandem repeat marker informativeness (12), also strengthens the potentialities of DNA diagnosis in the field of human aneuploidies.
Acknowledgments
This work was supported by the Association pour la Recherche sur le Cancer and the French Ministère de L’Enseignement Supérieur et de la Recherche.
Footnotes
Laboratoire de Génétique Moléculaire, Faculté des Sciences Pharmaceutiques et Biologiques, Université Paris V, 4 Avenue de l’Observatoire, 75006 Paris, France.
↵1 Nonstandard abbreviations: CMM, cutaneous malignant melanoma; NST, nervous system tumor; and Ct, threshold cycle.
1 The clinical status (+, presence of CMM and/or NST; −, healthy), the presence (+) or absence (−) of the predisposing haplotype, the ALB and p15 copy numbers, and the p15/ALB mean ratios are indicated for three independent assays of samples from 11 individuals of the CMM-NST family [pedigree positions as in Ref. (2)], compared with a reference genomic DNA (Boehringer Mannheim). The inferred deletional status (+, deleted; −, not deleted) is indicated based on the mean of the three p15/ALB ratios (±SD) of ∼0.5 or ∼1, respectively. All individuals harboring the predisposing haplotype were correctly found to have the deletion, using the p15 gene dosage assay, irrespective of their clinical status.
b The SDs for ALB and p15 were <1% to 12% of the means (not shown, available from authors).
2 NA, not applicable.
- © 1999 The American Association for Clinical Chemistry
References
In this issue
