Modifications of a telomeric repeat amplification protocol (TRAP) result in increased reliability, linearity and sensitivity.

The ends of vertebrate chromosomes (telomeres) are composed of many kilobases ofTTAGGG repeats (1-4). In normal somatic cells, some of these repeats are lost during eveiy cell division due to the inability of the lagging-strand to replicate the very 5' end of a linear DNA molecule (5,6). The ensuing telomere shortening has been shown to limit cellular proliferative capacity (7-10), and is likely to be the molecular measure (clock) that determines in vitro cellular senescence (11-13). Germline and most immortaltumor cells express an enzyme that maintains telomere length and thus prevents cellular senescence. Telomerase is a ribonucleoprotein enzyme that uses its RNA as a template for the synthesis ofTrAGGG repeats at the ends of the chromosomes (14-16). This polymerizing activity compensates for the failure to completely replicate the 5' end of the lagging strands and consequently results in the stabilization of telomere length. The development of a PCR-based telomerase assay [the telomere repeat amplification protocol; TRAP assay] capable of analyzing small tissue biopsies has permitted a large number of tumor samples to be analyzed (17). Approximately 85% of more than 950 primary tumors have been found to express telomerase (17-21 and unpublished observations), making telomerase a frequent marker for cancer cells. Several weaknesses in the assay have become apparent during these tumor surveys: (i) the assay is non-linear, maldng statements about relative levels of activity difficult; (ii) some tissue samples contain an inhibitor of Taq polymerase, and thus give a false negative results using the standard 6 jg of protein extract per assay but a positive result when diluted 10-100-fold (20) and (iii) the necessity of preparing a cell or tissue extract limits the application of the technique to small but not microscopic samples. We here describe modifications to the original protocol that resolve all of these difficulties. Telomerase can addTTAGGG repeats to a variety of non-telomeric sequences in vitro (16,22). It pauses after adding each repeat, presumably to permit repositioning of its internal RNA template prior to synthesizing the next repeat (15). It thus generates a ladder of 6 bp addition products. These can be visualized following PCR amplification using the telomerase substrate (TS) oligonucleotide as the forward primer and an oligonucleotide able to anneal to the telomeric repeats (CX) as the reverse primer (17). Figure IA (lanes 1-4) shows that the abundance ofproduct increases with increasing amount of Taq polymerase. This demonstrates that the molarity of these multiple amplification products causes TaqDNA polymerase to become limiting, even using a radioactive assay. A 150 bp internal standard, sufficiendy long so that it would not interfere with the visualization of the telomerase ladder, was prepared by synthesizing TS and CX oligonucleotides that contained an additional 15 bases at their 3' ends that overlapped with sequences encoding aa 97-132 of rat myogenin (23): TS-overlap primer, 5'-AATCCGTCGAGCAGAGTTGTGAATGAGGCCTTC-3' CX-overlap primer, 5'-CCCTTACCCTTACCCTTACCCTAATAGGCGCTCAATGTA-3' [TS (18 bases) and CX (24 bases) sequences are in bold type, myogenin sequences (15 bases each) are underlined. TSand CX-overlap primers (30 IM) were only used for the initial amplification after which the 150 bp product was column purified for future use]. Amplification of the myogenin cDNA with these primers generated a 150 bp product which could be reamplified using the same TS and CX primers used to amplify the telomerase ladder in the standard TRAP assay. Figures lB and C demonstrate that nonmalizing the intensity of the telomerase ladder to that of the internal standard permits the assay to become highly linear, so that accurate comparisons between samples becomes possible. The inclusion of the internal standard also immediately identifies false-negative tumor samples that contain Taq polymerase inhibitors (Fig. 1D). We have previously shown that the inclusion of up to 50 ng of contaminating genomic DNA did not interfere with the TRAP assay (20). This suggested that whole cells rather than just cell extracts could be analyzed. Aliquots of a dilute cell suspension containing single cells were identified under direct microscopic examination. These aliquots were then diluted with TRAP assay buffer in which the concentration ofTween-20 detergent had been increased to 0.5% to ensure cellular permeabilization, and then transferred to TRAP assay tubes. Increasing the incubation time 2-fold (from 30 min to 1 h), decreasing the concentration of cold precursors from 50 to 25 piM (to increase 2-fold the specific radioactivity of [32P]dCTP tracer), and increasing the number of PCR cycles from 31 to 34 allowed the telomerase in single cells to be detected (Fig. 2). The central role ofthe derepression of telomerase in the process of cellular immortalization and cancer is becoming increasingly apparent. A multitude of unanswered questions concerning its regulation, prognostic value and utility as a target for therapy remain, and a large number of laboratories are likely to be pursuing these and other questions. The modifications of the TRAP assay described herein should significantly improve the reliability, linearity and sensitivity of these investigations.