DOI: 10.1002/chem.201204440

DNAzyme-Based Probes for Telomerase Detection in Early-Stage CancerDiagnosis

Hui Wang,[a] Michael J. Donovan,[b] Ling Meng,[a] Zilong Zhao,[a, b] Youngmi Kim,[a]

Mao Ye,[b] and Weihong Tan*[a, b]

Introduction

Over the last few decades, telomerase, also known as telo-mere terminal transferase, has gained increasing interest asa potentially sensitive biomarker for early cancer diagnosisand prognosis, as well as a monitor for residual disease. Asan important cancer biomarker, telomerase activity hasbeen observed in approximately 90 % of human tumors,[1]

compared to the absence of telomerase activity in mostnormal somatic cells.[2] This suggests that the immortalityconferred by telomerase plays a key role in cancer develop-ment.[3–12] Telomerase is a reverse transcriptase that tran-scribes single-stranded (ss) RNA into ssDNA at chromoso-mal ends. In order to protect chromosomal linear ends fromthe constant loss of important DNA during cell differentia-tion, telomerase uses its own RNA template to elongate te-lomeres with specific DNA sequence repeats (“TTAGGG”in all vertebrates) at the 3’ ends of eukaryotic chromosomes.Consequently, chromosome ends can never be compro-

mised, irrespective of the number of cell divisions. However,since telomerase also allows cancer cells to divide in per-petuity, tumors are a likely outcome.

To date, several commercially available assays, such as thetelomeric repeat amplification protocol (TRAP), have beendeveloped to evaluate telomerase activity. TRAP is current-ly the most widely used and most sensitive method for telo-merase detection.[13–18] Although quite powerful, TRAP re-quires the use of DNA polymerases and is therefore suscep-tible to PCR-derived artifacts,[19] especially when screeningcompounds for telomerase inhibition. A number of PCR-free assays for telomerase activity, based either on directprobing[20,21] or formation of a sandwich structure containingthe elongated strands,[22] have been developed over the pastdecade. However, the lack of a suitable amplification mech-anism has hindered most of these methods from achievingsensitivity comparable to TRAP.[23]

Here, we study telomerase activity using a molecularsensor based on substrate cleavage by a DNAzyme.[24] Inrecent years, DNAzymes have been reported for a variety ofuses, including assays to detect environmental contaminantsor metal ions in cell-based studies, DNA computing, nano-wire production and drugs in preclinical models of cancer.However, 8–17 DNAzyme, which, in the presence of itsmetal ion cofactor, specifically recognizes and catalyticallycleaves its complementary substrate, has attracted growinginterest for its high selectivity, specificity and sensitivity forbiosensor construction.

We used the PbII-dependent 8–17 DNAzyme as an activecomponent for the analysis of telomerase activity. As shownin Figure 1, this hairpin molecule is engineered by an intra-molecular assembly strategy and consists of three functionaldomains: 1) a DNAzyme catalytic sequence, 2) a primer se-

Abstract: Human telomerase is a poly-merase enzyme that adds tandem re-peats of DNA (TTAGGG) in the telo-meric region to the ends of chromo-somes. Since telomerase can be detect-ed in immortalized, but not normal, so-matic cells, it has been considered aselective target for cancer chemothera-py. Here, we describe a DNAzyme-based probe to detect the presence oftelomerase in cell lysates. Telomerase

elongates the primer site on the probe.Subsequent addition of the PbII cofac-tor activates the DNAzyme, whichcleaves the elongated fragment at theRNA site, releasing the probe for re-petitive cycling and signal amplifica-

tion. The cleaved fragment is detectedby a reporter molecular beacon. Enzy-matic amplification with rapid turnoverallows detection of telomerase in therange of 0.1 to 1 mg cell lysate, with afivefold increase in signal level forcancer cells over normal cells. Thisprobe design can provide a simple, yetrapid and sensitive, measurement of te-lomerase activity.

Keywords: DNAzymes · intramo-lecular probes · fluorescent probes ·lead · telomerase

[a] Dr. H. Wang, Dr. L. Meng, Dr. Z. Zhao, Dr. Y. Kim, Prof. W. TanDepartment of Chemistry and Department of Physiology andFunctional Genomics, Shands Cancer Center andCenter for Research at the Bio/Nano InterfaceUF Genetics Institute and McKnight Brain InstituteUniversity of Florida, Gainesville, FL 32611-7200 (USA)E-mail : tan@chem.ufl.edu

[b] M. J. Donovan, Dr. Z. Zhao, Prof. M. Ye, Prof. W. TanMolecular Science and Biomedicine LaboratoryState Key Laboratory of Chemo/Bio-Sensing and ChemometricsCollege of Biology, College of Chemistry and Chemical EngineeringCollaborative Innovation Center for Chemistry andMolecular Medicine, Hunan UniversityChangsha, 410082 (P.R. China)

Chem. Eur. J. 2013, 19, 4633 – 4639 � 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 4633

FULL PAPER

quence to allow telomere elongation in the presence oftrace amounts of telomerase, and 3) a polythymine (poly-T)linker. In the presence of telomerase the primer sequencesare recognized, and the substrate fragment is generatedthrough telomere DNA elongation. The substrate then hy-bridizes with the catalytic DNAzyme binding domain(Figure 1), and forms a stable hairpin structure. However, ifa trace amount of lead(II) ion is also present, the substratewill be cleaved by the DNAzyme and will dissociate fromthe binding domain.

Signal transduction is achieved through the molecularbeacon (M9). In the absence of the substrate fragment, themolecular beacon forms a stable hairpin structure, and thereporter fluorescence is completely quenched by the prox-imity of the fluorophore and quencher. However, M9 isopened upon complete hybridization with the cleaved sub-strate, and the fluorescence signal is restored. The fluores-cence signal is amplified by repeated cycling of the probe toform a large number of cleaved substrate fragments, therebyallowing determination of trace amounts of telomerase pro-tein for application in early-stage cancer diagnosis.

Here, we report the design, molecular engineering andoptimization of the probe in terms of primer and linkerlength. We characterize the performance of the optimizedtelomerase detection probe and compare its detection abili-ties to those of the conventional TRAP assay. In our design,high sensitivity is achieved without loss of selectivity of the

probe, which maintains a quan-tifiable detection range from0.1 to 20 mg of total protein incell lysate and a selectivity ofover 200-fold for cancer cellsover normal cells. Based onthese results, this probe pro-vides a simple, yet rapid andsensitive, measurement of telo-merase concentration.

Results and Discussion

Design and optimization ofprobe for catalytic reaction todetect telomerase : Several ana-lytical procedures for the de-termination of telomerase ac-tivity have been developed, in-cluding TRAP (Figure 2),which involves intensive PCRamplification, or the function-alization of telomeres with flu-orescent labels.[25] In contrast,we have designed a new probeto monitor telomerase activity

Figure 1. Scheme of the hairpin-structured, DNAzyme-based telomerase sensor and its working principle.I) The TeloD9 DNAzyme–substrate hybrid strand, which is composed of DNAzyme (8–17 DNAzyme, bottom)and telomere primer, upper, hybridize together in Tris-acetate buffer, containing K+ (60 mm). II) After telo-merase-induced elongation, an elongation product is formed and hybridizes to 8–17 DNAzyme, and the 8–17DNAzyme is forced to form its catalytically active structure. III) In the presence of 10 mm PbII, the elongationproduct is cleaved and dissociates. This dissociated ssDNA is then detectedby a complementary molecular beacon, which reports the telomerase activ-ity by fluorescence enhancement.

Figure 2. Gel image for TRAP assays of extracts from cancer cell lines(M7617 and LH76), and one normal cell line (HBE135), which served asnegative control. PCR products were separated by electrophoresis on anondenaturing 12.5 % polyacrylamide gel, stained with ethidium bromide,and visualized on a UV transilluminator.

www.chemeurj.org � 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Chem. Eur. J. 2013, 19, 4633 – 46394634

using a PbII-initiated DNAzyme catalytic reaction. Telomer-ase detection with this assay involves three basic steps(Figure 1). First, telomerase is extracted from cancer cellsand mixed with substrate oligonucleotides, which are thenannealed with 8–17 DNAzyme at the 3’ end. These oligonu-cleotide sequences are both telomerase-binding substratesand strands to partially help 8–17 DNAzyme form its cata-lytically active tertiary structure. Second, telomerase elonga-tion is induced, and an elongation product is formed and hy-bridized to 8–17 DNAzyme, in which the 8–17 DNAzyme isforced to form its catalytically active structure. Specifically,the bound telomerase can catalytically elongate the DNAstrands in TTAGGG repeats in the presence of four deoxy-ribonucleotide triphosphate (dNTP) monomers (dATP,dCTP, dGTP, and dTTP). The 8–17 DNAzyme is formedinto a catalytically active structure because the elongationproduct is able to recognize its complementary target, whichis the flanking binding arm at the 5’ end of 8–17 DNAzyme.Third, in the presence of PbII (10 mm), the telomere productis cleaved and dissociated, whereas the probe returns to itsinitial state and awaits the next round of elongation. Subse-quently, the dissociated cleavage product then hybridizeswith a complementary molecular beacon (M9), and the fluo-rescence signal enhancement of the opened molecularbeacon is measured. To increase cleavage reaction efficiency,we constructed a hairpin-structured telomerase probe withthe substrate in close proximity to the 8–17 DNAzyme,thereby ensuring total hybridization efficiency. With thisdesign, the PbII binding pocket was well formed, thus in-creasing the apparent cleavage efficiency.

In order to maximize probe performance, several designfeatures must be considered. First, once the primer is elon-gated, the linker length must be such that the product willreach a range that allows rapid hybridization with the DNA-zyme, subsequently leading to cleavage and release. Second,the design is initialized by telomerase recognition of theprimer; therefore, the telomerase primer, which is located atthe 3’ end of TeloD9, must have higher binding affinity withtelomerase than with the 8–17 DNAzyme binding arm. Thisallows the primer sequence to be quickly recognized in thepresence of telomerase and then easily dissociated from theintramolecular DNAzyme binding arm, thereby allowingelongation to resume. In order to achieve suitable optimiza-tion toward these requirements, we investigated the probesequences by measuring the background fluorescence andmelting temperatures of the hairpin structures (Table 1) tocompare efficient structure formation (Figure 3). Third, opti-mization of the length of the 5’ end binding arm of DNA-

zyme depends on the nature of the telomere, which is aknown sequence (TTAGGG, human telomere repeatingunit).

To implement this plan, we designed probes termedTeloD6, TeloD9 and TeloD12 containing 6, 9, and 12 nucleo-tides (nt) in telomerase primer sequence fragments, respec-tively. No probe was designed containing a longer substratebecause dissociation with a base pair number in excess of 12would, in all likelihood, have been unfavorable due to thestrong binding effect,[26] resulting in low efficiency of elonga-tion. As shown in Figure 3, the fluorescence backgroundwith TeloD9 probes without PbII was very low. However, inthe presence of PbII, the elongated duplex structures bindthe ions, thereby activating the DNAzyme to catalyze thecleavage reaction. Figure 3 also shows that TeloD9, with amoderate length of telomeric primer fragment, possessedmaximum efficiency in the catalytic amplification reaction,giving an approximately 15-fold signal enhancement whenPbII was added. This outstanding signal enhancement incomparison to other candidates can be explained by en-hanced telomerization efficiency based on the competitionbetween two binding events: 1) the intramolecular hybridi-zation between the primer and DNAzyme, and 2) the inter-molecular binding between primer and the elongation tem-plate within the telomerase protein. Duplex stability (bind-ing event 1) increases while the primer is elongated by telo-merase enzyme, and the elongated product recognizes itscomplementary strand in the 5’ end binding arm of theDNAzyme fragment. In the presence of PbII ions, the elon-gation product is correctly positioned for cleavage by theDNAzyme and dissociates from the 5’ end binding arm.

At room temperature, stable duplex formation of theprimer with the complementary region of the DNAzymeprobe will not be achieved unless the two DNA segmentsare linked by the poly-T in a single probe unit. Comparingthe melting temperature of TeloD9 (57.3 8C) to unlinkedDNAzyme and its substrate (36 8C), the enhanced stability isquite significant. This enhanced duplex stability also resultsin improved cleavage efficiency.

Table 1. Comparison of melting temperature and fluorescence enhance-ment among different design sequences.

Hairpin sequence Tm [8C] S/B ratio

TeloD6 53.2 5TeloD9 57.3 14TeloD12 59.1 8D9+ DS9 36.2 2

Figure 3. Optimization of synthesized probes using fluorescence signalcomparison. Experiments were run in the presence of 2 mg protein after10 min incubation with TeloD9, -6, and -12. Measurements were acquiredafter addition of PbII to a final concentration of 10 mm.

Chem. Eur. J. 2013, 19, 4633 – 4639 � 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemeurj.org 4635

FULL PAPERDNAzyme-Based Probes for Telomerase Detection

The effect of primer length is demonstrated in Figure 3.Unlike TeloD9, TeloD6 showed low fluorescence enhance-ment because of low telomerization efficiency, indicatingthat a 6-nt primer may be too short to efficiently produceenough telomerase elongation product. Lack of elongationproduct would produce incomplete hybridization (no forma-tion of structure II in Figure 1), thereby resulting in an inef-ficient cleavage reaction.[27] On the other hand, no noticea-ble cleavage was observed for TeloD12, because the telo-merase could not bind to the primer. With 12 base pairs, thehybridization between telomeric primer fragment and 8–17DNAzyme binding arm is so strong that the telomerase pro-tein cannot compete, and elongation does not occur. In con-trast, TeloD9, in the absence of PbII, showed a low fluores-cence background signal, but upon the addition of PbII

(10 mm), fluorescence enhancement was observed within5 min. This experiment shows that the increased fluores-cence was indeed a result of DNAzyme-catalyzed cleavagewith PbII as a cofactor.

Because MgII, CuII, and ZnII are common divalent metalions present in biological samples, we also checked for possi-ble cross-reactivity for a series of metal ions. According tothe results in Figure 4, TeloD9 presented significant re-sponse to PbII over other divalent metal ions.

Characterization of the probe with analytical parameters :To further obtain the full profile of the performance ofTeloD9, we determined its dose response and enzymatic re-action kinetics. Figure 5 shows the profile of lead ion de-pendent fluorescence signal enhancement for the three celltypes studied. A linear response is observed at low lead con-centrations. Above the linear region, fluorescence enhance-ment continues to rise for the cancer cell lines (LH86 andM7617), with much less enhancement for the normal cells

(HBE135). However, when the [PbII] was higher than 10 mm,no improvement was observed, and 10 mm was chosen as theoptimum concentration.

One significant feature of the DNAzyme-based telomer-ase sensor is its excellent selectivity, which derives fromunique secondary configurations enabling the probes to dis-criminate between target (cancer cell extracts with telomer-ase) and nontarget (normal cell extracts without telomer-ase). With this capability, TeloD9 possesses the selectivityneeded to discriminate between healthy normal cells (tracelevels of telomerase) and cancer cells (abundant telomer-ase). Thus, to demonstrate the excellent selectivity ofTeloD9 for cancer cells against normal cells, the fluores-cence signal changes of three types of cell lysates at differ-ent concentrations were obtained (Figure 6). As expected,

Figure 4. Selectivity of the TeloD9 probe sensor. Sensor responses to allcompeting metal ions at the eight concentrations tested. The TeloD9 con-centration was 200 nm, that of M9 was 20 mm, and the buffer containedTris-acetate (50 mm, pH 7.2) and NaCl (100 mm). The reaction time was10 min. The values indicate means from three experiments, and the errorbars represent standard derivations.

Figure 5. Fluorescence increase over background fluorescence at varyingPbII concentrations for three cell types. The TeloD9 concentration was200 nm, that of M9 was 20 mm, and the buffer contained Tris-acetate(50 mm, pH 7.2) and NaCl (100 mm). The values indicate means fromthree experiments, and the error bars represent standard derivations.

Figure 6. Monitoring of dose-dependent telomerase elongation in thepresence of 10 mm PbII for different concentrations of total protein. TheTeloD9 concentration was 200 nm, that of M9 was 20 mm, and the buffercontained Tris-acetate (50 mm, pH 7.2) and NaCl (100 mm). The back-ground-corrected intensity in each cell type was normalized relative to0 mg of protein. The values indicate the means from three experiments,and the error bars represent standard derivations.

www.chemeurj.org � 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Chem. Eur. J. 2013, 19, 4633 – 46394636

W. Tan et al.

the normal cell lysate induced little fluorescence change, in-dicating that the intramolecular engineering of TeloD9 ac-complished the goal of in vitro, early cancer diagnosis. Wesuspected that the false positive signal towards normal celllysate may be attributed to the low percentage of M9 bind-ing to the 3’ end of TeloD9 itself, even without telomerase,since they are partially complementary to each other. To de-termine telomerase dependency, we also validated the fluo-rescence signal by adding different amounts of cell extractprotein to the initial reaction mixture. As shown in Figure 6,TeloD9 can provide up to 16-fold signal enhancement in thepresence of 1 mg cancer cell protein extract, compared toonly threefold enhancement for normal cells. This largesignal enhancement allows us to detect telomerase over alarge dynamic range with a low detection limit of telomerasein approximately 0.1 mg total protein from cell extracts.

In addition to the excellent sensitivity and selectivity ofTeloD9 towards telomerase, this probe also shows rapidcleavage kinetics. As shown in Figure 7, the signal reaches aplateau within 20 min. The time to reach this plateau wasmore than 30 min (data not shown) when 8–17 DNAzymeand its substrate (starting with telomere primer) were addedseparately, instead of as a single probe molecule.

The use of DNAzymes for amplified analysis of telomer-ase provides several advantages. First, the high turnover oftelomerase induced elongation reaction assures formation ofa large number of well-formed lead binding pockets.Second, the catalytic cleavage reaction allows a single probeto generate many product segments for detection by the mo-lecular beacon design. Third, the molecular beacon fluores-cence signal is detected within five minutes through fluores-cence measurement—much faster than the total time forcurrent PCR-based assays, which take several hours fromsample preparation to assay generation. Therefore, this newassay rivals the sensitivity and convenience of the conven-tional PCR-based method of telomerase detection.

Characterization of probe activity with PAGE-based assay :A PAGE-based assay was used to assess the multiple turn-over kinetics of the elongation/cleavage reaction of TeloD9in the presence of telomerase. For these experiments,TeloD9 (20 mm) was incubated with protein (2 mg) in Tris-acetate buffer (pH 7.2) at room temperature for a total20 min in vitro reaction. Aliquots were periodically re-moved, and the reaction was terminated by the addition offormamide (90%, v/v) containing EDTA (15 mm).[28] Thecleaved DNA fragments from TeloD9 were separated bypolyacrylamide gel electrophoresis, and the extent of con-version was quantified using a UV imager. The KM and kcat

values were obtained by fitting velocity of cleavage versussubstrate concentration by using the equation V= kcat � [S]/ ACHTUNG-TRENNUNG(KM+[S]). All reactions were terminated and processed asdescribed in the previous section.

The multiple turnover results shown in Figure 8 demon-strate that our probe has the advantage of enzymatic signalamplification according to Michaelis–Menten kinetics, ensur-ing its ability to detect low levels of telomerase, thus promis-ing application in early cancer diagnosis. Under the condi-tions used in these experiments, Vmax =0.97 mm min�1, KM =

0.71 mm, and kcat = 0.017 s�1. Specifically, in the presence of2 mg protein, 20 mm TeloD9 was cleaved, resulting in over700 catalytic turnovers of the DNAzyme used in the de-signed structure. In this experiment, prehybridization of theTeloD9 probe structure is not required. Thus, there is noneed to heat or cool the TeloD9 before hybridization andsubsequent cleavage by the introduction of PbII.

Conclusion

In conclusion, we have demonstrated a PCR-free, yet highlysensitive and selective, DNAzyme-based telomerase sensor.Duplex formation between the PbII-specific DNAzyme frag-

Figure 7. Analysis of the DNAzyme cleavage reaction at different PbII in-cubation times. M7617 and LH86 are cancer cells, and HBE135 is anormal cell line. The TeloD9 concentration was 200 nm, that of M9 was20 mm, and the buffer contained Tris-acetate (50 mm, pH 7.2) and NaCl(100 mm). The values indicate means from three experiments, and theerror bars represent standard derivations.

Figure 8. Catalytic activity of TeloD9 elongation/cleavage. The line repre-sents a nonlinear fit to the data using the Michaelis–Menten expressionwith KM =0.71 mm and kcat =0.017 s�1. Inset: the upper band representsuncleaved TeloD9, and the lower band represents the cleaved fragment.

Chem. Eur. J. 2013, 19, 4633 – 4639 � 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemeurj.org 4637

FULL PAPERDNAzyme-Based Probes for Telomerase Detection

ment and telomere primer results in efficient and specificcleavage of telomere target after telomerase elongation.Our single-molecule probe design allows a quantifiable de-tection range from 0.1–1 mg protein from cancer cell lysates.Not only is this tool for telomerase detection highly sensi-tive and selective, it is simple, cost effective, and can beused in cancer cells. Thus, this novel probe can be applied toboth cancer studies and cancer biomarker discovery. Giventhis degree of sensitivity and selectivity, our molecular engi-neering design may prove useful in the future developmentof other nucleic acid-based probes for clinical toxicologyand cancer therapeutics.

Experimental Section

Chemicals and reagents : All DNA synthesis reagents, including 6-fluores-cein (FAM) phosphoramidite, 5’-4-(4-dimethylaminophenylazo) benzoicacid, (Dabcyl) phosphoramidite, and 2’-O-triisopropylsilyloxymethyl-pro-tected RNA monomers, were purchased from Glen Research. Lead ace-tate and all reagents for buffer preparation and HPLC purification wereacquired from Fisher Scientific. The buffer used for the experiments con-tained Tris-acetate (50 mm, pH 7.2) and NaCl (100 mm).

Synthesis and purification of oligonucleotides : To optimize the design ofthe hairpin probe, multiple candidates were designed and prepared(Table 2). All were synthesized using an ABI 3400 DNA/RNA synthesiz-er (Applied Biosystems) at the 1 mmol scale by standard phosphoramidite

chemistry. HPLC was performed on a ProStar HPLC Station (VarianMedical Systems) equipped with a fluorescence detector and a photo-diode array detector. A C18 reverse-phase column (Alltech, C18, 5 mm,250 � 4.6 mm) with acetonitrile and aqueous TEAA (0.10 m) buffer,pH 7.0, as the mobile phase.

Determination of the melting temperature : The melting temperaturemeasurement for the designed sequences TeloD9, TeloD12, and TeloD6were performed with a Bio-Rad thermal cycler at a heating rate of1 8C min�1, with the fluorescence intensity at 520 nm recorded as a func-tion of temperature every 30 s. The cooling curve of the molecule wasalso obtained by decreasing the temperature at a rate of 3 8C min�1 andrecording the fluorescence intensity. Fluorescence-temperature profilesof the three molecules were recorded in Tris-acetate buffer (50 nm,pH 7.2) with NaCl (100 mm). The Tm values were taken as the tempera-tures corresponding to half-dissociation of the intramolecular duplex.

Cell extraction : One normal cell line, HBE135, and two different livercarcinoma cell lines, M7617 and LH86, were used in experiments. LH86,a human hepatoma cell line, is from a well-differentiated hepatocellularcarcinoma tissue. HBE135, LH86 and M7617 cells were cultured in Dul-

becco�s modified Eagle�s medium (DMEM) supplemented with heat-in-activated fetal bovine serum (FBS, 10 %) and 100 IU mL�1 penicillin/streptomycin at 37 8C in a humid atmosphere with 5% CO2. Cells wereharvested and washed twice with ice-cold phosphate-buffered saline(PBS) (10 mm KPO4, pH 7.5, 140 mm NaCl). Then the cells were lyzed inNP-40 lysis buffer (150 mm NaCl, 1% NP-40, 50 mm Tris-HCl, pH 8.0,5 mm EDTA) with a proteinase inhibitor cocktail (Sigma) for 30 min onice. Lysates were centrifuged at 14000 rpm for 20 min at 4 8C. The super-natant was stored at �80 8C in glycerol (20 %) and NaCl (100 mm). Theprotein content of the supernatant was measured by using the Bio-Radprotein assay.

Telomerase-mediated primer elongation : The reaction buffer was modi-fied from the protocol published by Kim et al. ,[2] consisting of Tris-HCl(20 mm), MgCl2 (1.5 mm), KCl (60 mm), EGTA (1 mm), Tween (0.05 %),and nucleotide (50 mm). Probe (0.15 nmol) and cell extract (1 mg protein)were added to the buffer to generate the final reaction mixture. This re-action mixture was then diluted to a final volume of 50 mL and incubatedfor 1 h at 37 8C in a PCR thermal cycler. Then, the temperature was in-creased to 94 8C and held for 5 min to deactivate telomerase and to ter-minate the elongation.

TRAP assay : Telomerase extracts were prepared and analyzed as descri-bed.[29] Briefly, cells were washed in PBS and then lyzed in ice-cold lysisbuffer (10 mm Tris-HCl, pH 7.4, 1 mm MgCl2, 1 mm EGTA, 0.1 mm phe-nylmethylsulfonyl fluoride, 5 mm 2-mercaptoethanol, CHAPS, 10 % glyc-erol). All cell extracts were immediately frozen and stored at �80 8C.Usually, the whole-cell extracts were thawed immediately prior to theTRAP assay. First, extracts were centrifuged at 14000 g for 30 min at4 8C. Supernatants were transferred to clean tubes, and protein concen-trations were measured using the Bio-Rad protein assay kit. Finally,TRAP, which is a one-tube PCR-based assay, was performed[2] to validatetelomerase activity.

Hybridization assay : The working principle underlying the application ofTeloD9 to monitor telomerase is shown schematically in Figure 1. TeloD9was annealed at a final concentration of 100 nm, by using the same proce-dure performed in the melting temperature determination. The annealedsample was then cooled to room temperature for subsequent assays. Analiquot (80 mL) containing hybridized DNAzyme–substrate (100 nm) solu-tion and 10� molecular beacon M9 was loaded into the wells of a 96-wellplate. An aliquot (10 mL) of concentrated telomerase stock solution wasthen added to the DNA to initiate telomerization. Incubation was per-formed at 37 8C for 30 min. Afterwards, an aliquot (10 mL) of lead stocksolution (20 mm) was added to the DNA to initiate the cleavage reaction.The fluorescence intensity was recorded for 100 mL buffer containingTris-acetate (50 mm, pH 7.2), KCl (100 mm), TeloD9 (100 nm) and M9 sol-ution (1 mm) for three different sets of conditions: without telomerase,with telomerase, and with both telomerase and lead solution. Bound telo-merase was removed with detergent, such as SDS solution. The excitationand emission wavelengths were set to 473 and 520 nm, respectively.Signal enhancement (S/B in Table 1) was calculated with the equation:(Fcleaved�Fbuffer)/(Fannealed�Fbuffer), where Fcleaved is fluorescence signals fromM9 after cleavage of the substrate fragment in the presence of PbII,Fannealed is the fluorescence signals from the hairpin probes without cleav-age, and Fbuffer is the fluorescence signals of buffer.

Gel-based activity assay : TeloD9 (2 mL) samples were annealed in Tris-acetate (50 mm) buffer containing NaCl (100 mm, pH 7.2). After remov-ing an aliquot (5 mL) as a zero time point, PbII was added to the remain-ing solution with a final concentration of 50 nm. Aliquots were removedat 60 s intervals. The TeloD9-only sample was also loaded onto the gel tojudge the effect without telomerase, without PbII, and without both telo-merase and PbII. The cleaved and uncleaved substrates were separatedon a 20 % denaturing polyacrylamide gel, and the gel was analyzed witha fluorescence imager (FLA-3000G; Fuji, Tokyo, Japan) by exciting at473 nm. Gels containing 20 % polyacrylamide were run on a FB-VE10-1electrophoresis unit (Fisher Biotech) at room temperature (200 V, con-stant voltage) for 50 min in TAE/MgII buffer.

Table 2. Names and sequences of DNA fragments used in this study. rAis the adenosine 5’-phosphate monomer. All syntheses were performedwith an ABI synthesizer. The DNA–RNA chimers were purified byHPLC.

Name Sequences (5’!3’)

TeloD12 CTAACCCTAACCTCCGAGCCGGACGAAACCCT-AACCCTTTTTTTTTTGGGTTAGGGTTrA

TeloD9 ACCCTAACCTCCGAGCCGGACGAAACCCTAAT-TTTTTTTTTTTAGGGTTrA

TeloD6 TAACCTCCGAGCCGGACGAAACCCTTTTTTTTT-TGGGTTrA

M9 TTAGGGTTAACCCTAACCCTAACCCTAAD9 ACCCTAACCTCCGAGCCGGTCGAAAACCCTAADS9 TTAGGGTTA

www.chemeurj.org � 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Chem. Eur. J. 2013, 19, 4633 – 46394638

W. Tan et al.

Acknowledgements

This work is supported by grants awarded by the National Key ScientificProgram of China (2011CB911000), the Foundation for Innovative Re-search Groups of NSFC (grant 21221003), China National Instrumenta-tion Program (2011YQ03012412) and by the National Institutes ofHealth (GM066137, GM079359 and CA133086).

[1] D. J. Argyle, L. Nasir, Vet. Pathol. 2003, 40, 1.[2] N. W. Kim, M. A. Piatyszek, K. R. Prowse, C. B. Harley, M. D. West,

P. L. Ho, G. M. Coviello, W. E. Wright, S. L. Weinrich, J. W. Shay,Science 1994, 266, 2011.

[3] K. Hiyama, S. Ishioka, Y. Shirotani, K. Inai, E. Hiyama, I. Muraka-mi, T. Isobe, T. Inamizu, M. Yamakido, Oncogene 1995, 10, 937.

[4] S. Kannan, H. Tahara, H. Yokozaki, B. Mathew, K. R. Nalinakumari,M. K. Nair, E. Tahara, Cancer Epidemiol. Biomarkers Prev. 1997, 6,413.

[5] H. J. Sommerfeld, A. K. Meeker, M. A. Piatyszek, G. S. Bova, J. W.Shay, D. S. Coffey, Cancer Res. 1996, 56, 218.

[6] E. Hiyama, L. Gollahon, T. Kataoka, K. Kuroi, T. Yokoyama, A. F.Gazdar, K. Hiyama, M. A. Piatyszek, J. W. Shay, J. Natl. Cancer Inst.1996, 88, 116.

[7] H. Tahara, T. Nakanishi, M. Kitamoto, R. Nakashio, J. W. Shay, E.Tahara, G. Kajiyama, T. Ide, Cancer Res. 1995, 55, 2734.

[8] E. Hiyama, K. Hiyama, T. Yokoyama, Y. Matsuura, M. A. Piatyszek,J. W. Shay, Nat. Med. 1995, 1, 249.

[9] H. Tahara, H. Kuniyasu, H. Yokozaki, W. Yasui, J. W. Shay, T. Ide,E. Tahara, Clin. Cancer Res. 1995, 1, 1245.

[10] S. Kyo, K. Kunimi, T. Uchibayashi, M. Namiki, M. Inoue, Am. J.Clin. Pathol. 1997, 107, 555.

[11] Y. Lin, H. Miyamoto, K. Fujinami, H. Uemura, M. Hosaka, Y. Iwa-saki, Y. Kubota, Clin. Cancer Res. 1996, 2, 929.

[12] H. Kinoshita, O. Ogawa, Y. Kakehi, M. Mishina, K. Mitsumori, N.Itoh, H. Yamada, T. Terachi, O. Yoshida, J. Natl. Cancer Inst. 1997,89, 724.

[13] A. K. Savoysky, E. Tsuchiya, M. Yamazaki, Nucleic Acids Res. 1996,24, 1175.

[14] H. A. Herbert, W. E. Wright, J. W. Shay, Nat. Protoc. 2006, 1, 1583.[15] M. Hirose, J. Abe-Hashimoto, H. Tahara, T. Ide, T. Yoshimura, Clin.

Chem. 1998, 44, 2446.[16] J. Feng, W. D. Funk, S. S. Wang, S. L. Weinrich, A. A. Avilion, C. P.

Chiu, R. R. Adams, E. Chang, R. C. Allsopp, J. Yu, Science 1995,269, 1236.

[17] T. M. Nakamura, G. B. Morin, K. B. Chapman, S. L. Weinrich, W. H.Andrews, J. Lingner, C. B. Harley, T. R. Cech, Science 1997, 277,955.

[18] T. M. Fletcher, A. Trevino, J. M. Woynarowski, Biochem. Biophys.Res. Commun. 1999, 265, 51.

[19] C. M. Niemeyer, M. Adler, R. Wacker, Trends Biotechnol. 2005, 23,208.

[20] Y. Weizmann, F. Patolsky, O. Lioubashevski, I. Willner, J. Am.Chem. Soc. 2004, 126, 1073.

[21] G. Zheng, F. Patolsky, Y. Cui, W. U. Wang, C. M. Lieber, Nat. Bio-technol. 2005, 23, 1294.

[22] P. M. Schmidt, C. Lehmann, E. Matthes, F. F. Bier, Biosens. Bioelec-tron. 2002, 17, 1081.

[23] K. Piotrowska, E. Kleideiter, T. E. Murdter, S. Taetz, C. Baldes, U.Schaefer, C. M. Lehr, U. Klotz, Lab. Invest. 2005, 85, 1565.

[24] S. W. Santoro, G. F. Joyce, Proc. Natl. Acad. Sci. USA 1997, 94, 4262.[25] F. Patolsky, R. Gill, Y. Weizmann, T. Mokari, U. Banin, I. Willner, J.

Am. Chem. Soc. 2003, 125, 13918.[26] A. J. C. Wolfrum, Nucleosides Nucleotides Nucleic Acids 2005, 24,

1069.[27] G. Zhao, Y. Guan, Acta Biochim. Biophys. Sin. (Shanghai) 2010, 42,

722.[28] J. L. Tae-Jin Yim, Y. Lu, R. S. Kane, J. S. Dordick, J. Am. Chem.

Soc. 2005, 127, 12200.[29] N. W. Kim, C. B. Harley, K. R. Prowse, S. L. Weinrich, M. A. Piatys-

zek, W. E. Wright, J. W. Shay, Science 1995, 268, 1116.

Received: December 13, 2012Published online: February 20, 2013

Chem. Eur. J. 2013, 19, 4633 – 4639 � 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemeurj.org 4639

FULL PAPERDNAzyme-Based Probes for Telomerase Detection