Quadruplex ligands are often considered as telomerase inhibitors. Given the fact that some of these molecules are present in the clinical setting, it is important to establish the validity of this assertion. To analyze the effects of these compounds, we used a direct assay with telomerase-enriched extracts. The comparison of potent ligands from various chemical families revealed important differences in terms of effects on telomerase initiation and processivity. Although most quadruplex ligands may lock a quadruplex-prone sequence into a quadruplex structure that inhibits the initiation of elongation by telomerase, the analysis of telomerase-elongation steps revealed that only a few molecules interfered with the processivity of telomerase (i.e., inhibit elongation once one or more repeats have been incorporated). The demonstration that these molecules are actually more effective inhibitors of telomeric DNA amplification than extension by telomerase contributes to the already growing suspicion that quadruplex ligands are not simple telomerase inhibitors but, rather, constitute a different class of biologically active molecules. We also demonstrate that the popular telomeric repeat amplification protocol is completely inappropriate for the determination of telomerase inhibition by quadruplex ligands, even when PCR controls are included. As a consequence, the inhibitory effect of many quadruplex ligands has been overestimated.G-quadruplex Í telomere Í telomeric repeat amplification protocol assay Í direct assay T he unlimited proliferative potential of cancer cells depends on telomere maintenance (1). As a consequence, several classes of telomerase inhibitors have been developed for anticancer purposes. Optimal telomerase activity requires an unfolded single-stranded telomeric overhang (2-4). This overhang may adopt an unusual DNA structure called a G-quadruplex, composed of G-quartets and formed in the presence of cations (5). Therefore, ligands that selectively bind to and stabilize telomeric G-quadruplex structures could act as indirect telomerase inhibitors (Fig. 1A) (6). The number of identified G-quadruplex ligands has grown rapidly over the past few years (for a recent review, see ref. 7); some of these molecules exhibit potent in vivo antitumor activity, and clinical trials started recently.Although various methods are available to measure telomerase activity, the telomeric repeat amplification protocol (TRAP) is most widely used (8). With a few notable exceptions (6, 9-12), TRAP [or a variant called TRAP-G4 (13)] has been the standard method to measure telomerase inhibition by G-quadruplex ligands (reviewed in ref. 7). The original method was improved by the use of an internal control (often called ITAS for internal telomerase assay standard) that coamplifies with the telomerase products and allows the detection of nonspecific Taq polymerase inhibitors. Notably, the ITAS product does not contain any telomeric sequence and therefore is not able to form a G-quadruplex (5,14). We demonstrate here that TRAP is...