At

YDa

b

a

ARRAA

KVEHAL

1

imiep

tb(cbtcpo

0d

Biosensors and Bioelectronics 28 (2011) 270– 276

Contents lists available at ScienceDirect

Biosensors and Bioelectronics

jou rn al h om epa ge: www.elsev ier .com/ locate /b ios

ptamer beacons for visualization of endogenous protein HIV-1 reverseranscriptase in living cells

u Lianga,b, Zhiping Zhanga, Hongping Weia, Qinxue Hua, Jiaoyu Denga,eyin Guob, Zongqiang Cuia,∗∗, Xian-En Zhanga,∗

State Key Laboratory of Virology, Wuhan Institute of Virology, Chinese Academy of Sciences, Wuhan 430071, China1

State Key Laboratory of Virology and Modern Virology Research Center, College of Life Sciences, Wuhan University, Wuhan 430072, China1

r t i c l e i n f o

rticle history:eceived 11 April 2011eceived in revised form 13 July 2011ccepted 14 July 2011vailable online 23 July 2011

eywords:

a b s t r a c t

Direct visualization of endogenous proteins in living cells remains a challenge. Aptamer beacon is apromising technique to resolve this problem by combining the excellent protein binding specificity ofthe aptamer with the sensitive signal transduction mechanism of the molecular beacon. In this study,aptamer 93del against HIV-1 reverse transcriptase (RT) was engineered into aptamer beacons to rec-ognize and image HIV-1 RT. The constructed aptamer beacons could specifically bind to HIV-1 RT andthe beacon-RT binding showed effective fluorescence signal transduction in homogeneous solution. In

isualizationndogenous proteinIV-1 reverse transcriptaseptamer beaconsiving cells

solutions with 1 �M of the aptamer beacon, the effective fluorescence signal increased with increasingconcentration of HIV-1 RT from 0.5 �M to 5 �M. When the aptamer beacons were delivered into theliving cells that transiently expressed HIV-1 RT, HIV-1 RT could be specifically labeled and imaged. Thedesigned aptamer beacons were further successfully applied for RT imaging in HIV-1 integrated U1 cells.The method developed here may be extended to visualize many other endogenous proteins in living cellsusing appropriate aptamer beacons.

© 2011 Elsevier B.V. All rights reserved.

. Introduction

Imaging and monitoring the behavior of proteins in real times important for understanding protein functions. To date, many

ethods have been developed for protein labeling and monitoringn vitro and in vivo (Kanda et al., 1998; Deka et al., 1998). How-ver, it remains a major challenge to visualize directly endogenousroteins in living cells in real time.

Immunofluorescence has been used extensively to detect pro-ein targets, which allows visualization of a specific protein in cells,y binding with a specific fluorescent-dye-conjugated antibodyDeka et al., 1998). However, the immunofluorescence process canause irreversible damage to cells, and fluorophore-labeled anti-odies are difficult to be used for real-time observation of proteinargets in living cells. Alternatively, many uses of immunofluores-

ence have been outmoded by the development of recombinantroteins that contain fluorescent protein tags, such as green flu-rescent protein (GFP) (Chalfie et al., 1994). Using of such tagged

∗ Corresponding author. Tel.: +86 10 58881508; fax: +86 27 87199492.∗∗ Corresponding author. Tel.: +86 27 87199115; fax: +86 27 87199492.

E-mail addresses: czq@wh.iov.cn (Z. Cui), x.zhang@wh.iov.cn (X.-E. Zhang).1 Affiliated institutions a and b contributed equally to this paper.

956-5663/$ – see front matter © 2011 Elsevier B.V. All rights reserved.oi:10.1016/j.bios.2011.07.031

proteins allows much better localization and less disruption of pro-tein function in living cells (Yuste, 2005; Chudakov et al., 2005).However, because of its big size, the fluorescent protein tags mayinterfere with the structure of target proteins, and consequently,affect their transport and combination with other proteins. Moreimportantly, use of recombinant proteins makes it difficult toachieve real-time visualization of the endogenous proteins that areexpressed by original specimens. Although a genetically encodedfluorescent amino acid has been developed to label proteins inliving cells (Summerer et al., 2006), the procedure is very com-plex because the labeling procedure needs to introduce a specialgene codon into the gene of the target protein and express thecorresponding orthogonal tRNA aminoacyl-tRNA synthetase pairsynchronously. Moreover, the fluorescence produced in this wayseems too weak to visualize endogenous proteins in living cells.

A protein tracer used in real-time visualization of the dynamicbehavior of endogenous viral protein in living cells necessitates: (a)to be high affinity and specificity binding to the target protein; (b)to be able to transduce molecular recognition directly into an easilyacquired signal; and (c) to be easily transported into living cells for

tracing an endogenous protein.

Aptamers are powerful elements for molecular recognition.Aptamers, including DNA/RNA oligonucleotides and peptides, canbe obtained through repeated rounds of selection in vitro from

Bioel

rEuemammaflmaactfl

paftsbctbi

bdutKbscH

d1aRDoibKFttaBwe

2

2

wTG9T

Y. Liang et al. / Biosensors and

andomly designed molecular pools (Tuerk and Gold, 1990;llington and Szostak, 1990). Aptamers can bind to various molec-lar targets such as small molecules, proteins, nucleic acids, andven cells, tissues and organisms. As molecular recognition ele-ents, some aptamers also show remarkable specificity and high

ffinity to their targets, with dissociation constants typically in theicromolar to low picomolar range, comparable to those of someonoclonal antibodies (Jenison et al., 1994). After being selected,

ptamers can be easily produced by chemical synthesis. If labeleduorescently, aptamers can be used as optical sensors to transduceolecular recognition into optical signals. Fluorescently labeled

nti-adenosine aptamers have been designed to detect adenosinend quantify its concentration (Jhaveri et al., 2000). A fluores-ently labeled anti-thrombin DNA aptamer was reported to detecthrombin in solution by monitoring the evanescent-wave-induceduorescence anisotropy of the aptamer (Potyrailo et al., 1998).

Molecular beacons (MBs) are single-stranded oligonucleotiderobes with a fluorophore that is covalently linked to one endnd a quencher to the other end (Tyagi and Kramer, 1996). MBsorm a nonfluorescent conformation in a stem-loop structure whenhey are free. However, when they hybridize to the complementequence, they undergo a conformational change that results inright fluorescence. As an optical sensor, MBs have many importantharacteristics such as high sensitivity, detection without separa-ion, and wide temperature range to hybridize to targets. MBs haveecome a powerful tool for DNA/RNA detection in solution and even

n living cells (Sokol et al., 1998; Cui et al., 2005; Wang et al., 2008).The concept of aptamer beacon is combination use of protein

inding specificity of the aptamer and the sensitive signal trans-uction mechanism of the MB. There are already some reportssing aptamer beacons for protein detection in homogeneous solu-ion, e.g. aptamer beacons for HIV-1 Tat protein (Yamamoto andumar, 2000) and for thrombin (Nobuko et al., 2001; Li et al., 2002),ut not yet for intracellular application. In the current study, wehowed experimentally the successful application of aptamer bea-on in visualization of endogenous protein in living cells usingIV-1 reverse transcriptase (RT) as a demonstration.

HIV-1 RT converts the viral single-stranded RNA genome intoouble-stranded DNA during the early infection stage (Coffin et al.,997). The mature RT holoenzyme is a heterodimer of 66-kDa (p66)nd 51-kDa (p55) subunits and has three enzymatic activities: (a)NA-directed DNA polymerization; (b) RNaseH activity; and (c)NA-directed DNA polymerization. Considering the important rolef RT in the lifecycle of HIV-1, many efforts have been made to selectts aptamers. A wide variety of DNA/RNA ligands to HIV-1 RT haveeen obtained. The selected ssDNA ligands can bind HIV-1 RT withd values as low as 1 nM, and many of them have unique sequences.or example, Andreola et al. (2001) have selected aptamers againsthe HIV-1 RNase H activity associated with RT with a high propor-ion of G and T residues to form G-quartet structures, and suchptamers are able to interact with HIV-1 RT with high affinity.ased on the affinity between G-quartet aptamers and HIV-1 RT,e designed aptamer beacons to achieve real-time visualization of

ndogenous HIV-1 RT in HIV-1-infected living cells.

. Materials and methods

.1. Oligonucleotides, recombinant plasmids, proteins, and cells

All the oligonucleotides, including aptamer beacons,ere purchased from Sangon Biotech (Shanghai) Co. Ltd.

he oligonucleotides used in the present study were 93del,GGGTGGGAGGAGGGT; 93del4d, ACCCGGGGTGGGAGGAGGGT;3del5d, ACCCTGGGGTGGGAGGAGGGT; 93del6d, ACCC-CGGGGTGGGAGGAGGGT. Aptamer beacons (labeled 93del4dMB,

ectronics 28 (2011) 270– 276 271

93del5dMB, 93del6dMB) were synthesized by coupling fluo-rophore (TAMRA) at the 5′-end and a DABCYL-group at the 3′-end.As a control probe, 93del5dF was 93del5d with 5′-TAMRA, butwithout any quencher group on the 3′-end.

Recombinant HIV-1 RT overexpressed in Escherichia coli cellswas purified according to the procedure described by Davies et al.(1991) and Hostomsky et al. (1991). Enzyme was aliquoted andstored at −70 ◦C in HRT buffer (200 mM KOAc, 50 mM Tris–HC1, pH8.0, 6 mM MgCl2, and 10 mM DTT). The purified HIV-1 RT proteinwas thawed and refrozen only once.

Plasmid pEGFP-C1-RT, which expressed HIV-1 RT gene, wasconstructed by ligating RT cDNA in frame into vector pEGFP-C1.The chronically HIV-l-infected promonocytic line Ul was obtainedfrom the AIDS Research and Reference Reagent Program, Divisionof AIDS, NIH. HeLa and U1 cells were grown in DMEM and RPM1640medium, respectively, both supplemented with 10% FBS, 100 U/mlpenicillin and 100 �g/ml streptomycin, and maintained at 37 ◦Cwith 5% CO2 in a cell incubator. To express HIV-1 protein, U1 cellsneed to be activated by TNF-� (100 U/ml) (Sigma–Aldrich) for 48 hat 37 ◦C. U1 cells naturally live in suspension, therefore, beforeobservation, U1 cells need to be adhered to the chambered coverglass treated with poly-l-lysine (100 �g/ml) (Sigma–Aldrich).

2.2. Gel-retardation assay of DNA binding

Oligodeoxynucleotides (93del, 93del4d, 93del5d, 93del6d atfinal concentration of 1 �M) were incubated with increasing con-centration of HIV-1 RT (0, 0.5 �M, 1 �M, 1.5 �M, 2.0 �M, 4.0 �M)in 20 �l DNA-binding buffer that contained 20 mM Tris–HCl (pH7.5), 55 mM KCl, 5 mM DTT, 5% glycerol and 4 mM MgCl2 for30 min at 37 ◦C. And the four oligodeoxynucleotides (final concen-tration of 1 �M) were also incubated with BSA (4 �M) as control.The samples were loaded onto a 9% native polyacrylamide gel(29:1 acrylamide/N,N′-methylenebisacrylamide). Electrophoresiswas performed in 0.5× TBE buffer (45 mM Tris, 45 mM boric acidand 1 mM EDTA) that contained 10 mM MgCl2 for 2–3 h at 4 ◦C.Gels were stained with ethidium bromide (0.7 mg/ml) for 20 min,and scanned using a Bio-Rad Gel Documentation System. All assayswere performed at least three times.

2.3. HIV-1 RT-binding assay using a fluorescence spectrometer

The aptamer beacons (93del4dMB, 93del5dMB, 93del6dMB)were diluted to a concentration of 1 �M in 1× PBS buffer (137 mMNaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.76 mM KH2PO4, pH 7.4), andincubated with increasing concentration of HIV-1 RT (0, 0.5 �M,1 �M, 2 �M, 5 �M, 8 �M, 10 �M) for 30 min at 37 ◦C. The fouraptamer beacons (final concentration of 1 �M) were also incubatedwith BSA (5 �M) as controls. For dissociating any intermolecular G-quartet-mediated structures that might have formed, the aptamerbeacons were heated to 99 ◦C for 3 min and cooled to room tem-perature before the experiment. All of the fluorescence (TAMRA)intensity measurements were performed at 25 ◦C with an excita-tion wavelength of 515 nm, and the emission was monitored at550–650 nm. All assays were performed at least three times andKd values of the binding between the aptamers and HIV-1 RT werecalculated using the method by Stephen and Laura (1991).

2.4. Recombinant plasmids transfection and delivery of aptamerbeacons into living cells

For transfection, HeLa cells were grown on cover slips placed

on 35-mm dishes. Transfection was performed with Lipofectamine2000 (Invitrogen, Carlsbad, CA, USA) according to the manufac-turer’s instructions. HeLa cells were transiently transfected withthe plasmid pEGFP-C1-RT that contained the coding sequence of

2 Bioelectronics 28 (2011) 270– 276

RrwmaeDcf3bfi(

2

pcwpmrr31il1ew

2

wG5ferte3btecMUP

3

3

aHa9ians

Fig. 1. Sequences of ODNs. (A) 93del was aligned according to the G residuesinvolved in the formation of putative G-tetrads (boxes), and was compared withTBA. (B) Schematic representation of the G-quartet ODNs. The structures of 93del

72 Y. Liang et al. / Biosensors and

T protein fused to the EGFP gene. Sixteen hours post-transfection,ecombinant GFP-RT fluoresced in HeLa cells. Aptamer beaconsere delivered into living cells using a reversible permeabilizationethod with streptolysin O (SLO), as described previously (Spiller

nd Tidd, 1995; Santangelo et al., 2004; Walev et al., 2001; Fariat al., 2001). Specifically, SLO was activated first by adding 10 mMTT to 1000 U/ml SLO for 90 min at 37 ◦C. Cells grown in 35-mmell culture dishes were incubated for 30 min in 500 �l Ca2+, Mg2+

ree PBS that contained 25 mM HEPES, 5 U/ml activated SLO and00 nM each aptamer beacon. 30 min later, cells were then resealedy adding 1.5 ml of the typical growth medium and incubatedor 15 min at 37 ◦C before performing fluorescence microscopymaging under TAMRA excitation (545 nm) and EGFP excitation488 nm) beam.

.5. Immunofluorescence

Immunofluorescence experiments were performed as describedreviously (McBeath and Fujiwara, 1984). For immunofluores-ence analysis, U1 cells on 35-mm glass-bottom culture dishesere washed twice with PBS, followed by fixation with 4% coldaraformaldehyde for 15 min at room temperature. Cells were per-eabilized with PBST (PBS and 0.2% Triton X-100) for 10 min at

oom temperature, and then blocked in PBS/2% BSA for 20 min atoom temperature. Thereafter, the cells were incubated for 1 h at7 ◦C with primary antibody (Mouse monoclonal antibody to HIV-

RT) (ab9066; Abcam, Hong Kong, China) at a dilution of 1/100n PBS/2% BSA, followed by incubation for 1 h at 37 ◦C with FITC-abeled secondary antibody (Boster, Wuhan, China) at a dilution of/100 in PBS/2% BSA, and were washed with PBS three times afterach incubation. Cells were washed twice with distilled water. Cellsere analyzed by fluorescent microscopy.

.6. Image acquisition, data collection and analysis

All imaging experiments were carried out using an invertedide-field fluorescent microscope (Axiovert 200; Carl Zeiss,ermany) equipped with a cooled CCD camera (Model Cascade12B; Photometrics, Tucson, AZ, USA). A filter set with 472/30 nmor excitation, 510 nm for dichroic beam splitter, and 520/35 nm formission viewed the green fluorescence from the EGFP channel. Theed fluorescence from the TAMRA channel was viewed with an exci-ation filter of 531/40 nm, a dichroic beam splitter of 600 nm, and anmission filter of 593/40 nm, while the blue fluorescence of Hoechst3258 was viewed with an excitation filter of 365 nm, a dichroiceam splitter of 395 nm, and an emission filter of 420 nm (All fil-ers are from Semrock, Rochester, NY, USA). In the double-labelingxperiment, GFP and TAMRA were sequentially scanned to avoidross talk. Image acquisition and analysis was performed usingetaMorph 6.0 software (Molecular Devices, Downingtown, PA,SA). Images were further analyzed using SoftWoRx 5.0 (Appliedrecision, Inc., Washington, USA).

. Results and discussion

.1. Construction of HIV-1 RT-binding aptamer beacons

The aptamer beacons for HIV-1 RT were constructed based onn oligonucleotide 93del (Fig. 1A) which is truncated from theIV-1 RT aptamer ODNs 93 (Andreola et al., 2001; Soultrait1 etl., 2002). As determined by NMR (Anh Tuan Phan et al., 2005),3del adopts a unique compact dimeric G-quartet structure, which

s similar to the core structure of a reported aptamer beacon TBAgainst thrombin (Nobuko et al., 2001) (Fig. 1B). The 93del was engi-eered into aptamer beacon with the loop-stem structure by addingtem and fluorophore-quencher couple. Considering that the four

and TBA were resolved as previously described (Nobuko et al., 2001; Anh Tuan Phanet al., 2005).

Gs at the 5′-end of 93del are crucial for its binding properties(Andreola et al., 2001), several nucleotides were added to the 5′-endto form the stem by pairing with the 3′-end sequence. As shownin Fig. 3A, 4–6 nucleotides were added to the 5′-end of 93del toconstruct the stem-loop structure, 93del4d, 93del5d, and 93del6d,respectively.

Gel-retardation assay was carried out to evaluate the affinitybetween HIV-1 RT and aptamers 93del or 93del with added extranucleotides. Native gel electrophoresis of 93del binding with HIV-1 RT showed that the magnitude of the band corresponding to theoligonucleotide decreased with increasing HIV-1 RT concentration(Fig. 2A). 93del4d, 93del5d and 93del6d binding with HIV-1 RT hadthe same appearance as 93del (Fig. 2B–D). These indicated that93del and the designed oligonucleotides all interacted with HIV-1 RT and addition of nucleotides to 93del had little effect on theircombination with HIV-1 RT. Fig. 2E showed that there was no inter-action between BSA and the aptamers (93del, 93del4d, 93del5d,and 93del6d), and meant that 93del and the designed oligonu-cleotides represented specificity to HIV-1 RT relative to otherproteins.

Subsequently, fluorophore–quencher pair (i.e. TAMRA-DABCYL)was added to the 5′- and 3′-ends of oligonucleotides 93del4d,93del5d, and 93del6d, respectively, to construct the aptamerbeacons 93del4dMB, 93del5dMB and 93del6dMB (Fig. 3A). Wehypothesized that the designed aptamer beacons would form a sta-ble stem-loop structure when there was no target, and when thetarget was present, the hairpin structure would open and returnto G-quartet structure binding with the target, and the conforma-tional changes could be transformed into observable fluorescencesignals (Fig. 3B).

3.2. Reporting of HIV-1 RT by aptamer beacons in homogeneoussolution

To test whether the designed aptamer beacons reported the pro-tein HIV-1 RT, the aptamer beacon–RT binding assay was carriedout in homogeneous solution. The fluorescence intensity of each

beacon was evaluated for its response to HIV-1 RT with increas-ing concentration. Binding curves for each aptamer beacons wereshown in Fig. 4A. The fluorescence enhancement of each beacon fol-lowed by HIV-1 RT increased, which showed that aptamer beacons

Y. Liang et al. / Biosensors and Bioelectronics 28 (2011) 270– 276 273

Fig. 2. Gel retardation assay. The oligonucleotides 93del and the transformed oligonucleotides (1 �M) were incubated with increasing concentrations of HIV-1 RT (0, 0.5 �M,1 �M, 1.5 �M, 2.0 �M, 4.0 �M) in DNA-binding buffer for 30 min at 37 ◦C. The products were analyzed by electrophoresis on a 9% acrylamide gel in TBE buffer. (A) ODN: 93del;(B) ODN: 93del5d; (C) ODN: 93del4d; (D) ODN: 93del6d. (E) control experiment: 1, 93del5d (1 �M) binding with HIV-1 RT (4 �M); 2–5, 93del, 93del4d, 93del5d, 93del6d( depen

ceha

tc9eitccdatsD

1 �M) binding with BSA (4 �M), respectively. Results are representative of three in

ould be used to report the HIV-1 RT in homogeneous solution, andven to quantify its concentration. 93del5dMB demonstrated theighest response, an approximately 9-fold increase in fluorescencet saturation concentration of HIV-1 RT.

Fluorescence spectrum assay of 93del5dMB (1 �M) binding tohe complementary DNA sequence and HIV-1 RT with varying con-entrations (0–10 �M) was conducted. As shown in Fig. 4B, when3del5dMB was used as a standard molecular beacon, the observednhancement of the fluorescence was approximately 13-fold aftert hybridized with a complementary DNA sequence. In comparison,he fluorescence intensity of 93del5dMB binding with saturatedoncentration of HIV-1 RT was two-thirds of its binding with theomplementary DNA sequence. This may be ascribed to that theistance between the fluorophore and the quencher in G-quartet

fter the aptamer beacon bound to the protein, was shorter thanhat in a fully extended helix. Nevertheless, aptamer beacons didhow effective signal transduction following protein-dependentNA conformational change. As a control, the fluorescence inten-

dent experiments. Standard deviations are within 5–10%.

sity of 93del5dMB had only a slight change when there was BSA insolution (Fig. 4B).

After fluorophore–quencher pairs were added into the oligonu-cleotide probes, the probes are non-fluorescent when there is noDNA/RNA or HIV-1 RT protein target. When there is complementaryDNA or HIV-1 RT protein target in solution, aptamer beacons couldproduce fluorescence signal. This is ideally in accordance with thefunction of the molecular beacons (Tyagi and Kramer, 1996). Theseresults also meant that the presence of fluorescence–quenchingpairs may not interfere with the interaction between aptamer bea-cons and HIV-1 RT. 93del5dMB had best signal response before andafter it bound to the target protein, showing that 5-nucleotide pairin the stem should be appropriate for construction of the aptamerbeacons. According to the method constructed by Stephen and

Laura (1991) for measuring binding affinity, we can calculate thebinding affinity between HIV-1 RT and aptamer beacons using thefluorescence assay. The Kd value of the binding between 93del5dMBand HIV-1 RT is calculated to be 2.5 × 10−8 M.

274 Y. Liang et al. / Biosensors and Bioelectronics 28 (2011) 270– 276

Fig. 3. Construction of aptamer beacons. (A) Predicted secondary structures of threeaptamer beacons with different stem lengths. (B) Schematic drawing of mechanismof signaling by anti-HIV-1 RT aptamer beacon. (Left) Aptamer beacon in quenchedstem-loop conformation. (Right) HIV-1 RT, a heterodimer consisting of p66 and p51subunits, bound to aptamer in G-quartet conformation through the RNase H part inthe p66 subunit. Bulb represents the fluorophore, and the dark square represents theqb

3b

9clPttTeGcpwiaae91tn

Fig. 4. Fluorescence assay of aptamer beacons bonding with HIV-1 RT in homoge-nous solution. (A) Comparison of binding curves for each aptamer beacons(93del4dMB, 93del5dMB, 93del6dMB). Each aptamer beacon (1 �M) bound to HIV-1 RT with concentration of 0, 0.5, 1, 2, 5, 8 or 10 �M for 30 min at 37 ◦C. (B) Bindingcurves of 93del5dMB. Curve 1: 93del5dMB (1 �M) binding with a complementaryDNA sequence (1 �M). Curve 2–8: 93del5dMB (1 �M) binding to HIV-1 RT with vari-

◦

uencher. The emission intensity of the fluorophore is represented by the size of theulb, and increases with the distance between the fluorophore and the quencher.

.3. Visualization of HIV-1 RT in living HeLa cells by aptamereacons

Then, the aptamer beacon with the highest response,3del5dMB, was chosen to detect and image HIV-1 RT in livingells. As seen in Fig. 5A, 93del5dMB and GFP-tagged RT were co-ocalized in the same cell, and from SoftWoRx 5.0 analysis, we gotearson coefficient of correlation of 0.8756. These results indicatedhat 93del5dMB specifically bound to HIV-1 RT, and the interac-ion could be transformed to a fluorescent signal in living cells.he control probe 93del5dF was also delivered into the GFP–RTxpressing cells by SLO. As shown in Fig. 5B, the fluorescence ofFP–RT and 93del5dF was co-localized in the cytoplasm, whichonfirmed that 93del5d interacted with HIV-1 RT. 93del5dF dis-layed stronger fluorescence in the nucleus than in the cytoplasm,hich was caused by excess probe without quencher moiety enter-

ng the nucleus. Many studies have reported that oligonucleotidesnd MBs can be rapidly taken up by the nucleus, although the mech-nism is not well understood (Tyagi and Alsmadi, 2004; Mhlangat al., 2005; Leonetti et al., 1991). There was very little signal of

3del5dMB after MB delivery in HeLa cells after 45 min and even.5 h in the absence of HIV-RT expression (Fig. 5C). Moreover, whenhe control probe 93del5dF, without the 3′ quencher, was added toon-HIV RT-expressing HeLa cells, the fluorescence signals were

ous concentration of 0, 0.5, 1, 2, 5, 8 or 10 �M for 30 min at 37 C. Curve 9: 93del5dMB(1 �M) incubated with BSA (5 �M) as a control. Results were the average of threeindependent experiments. Standard deviations are within 5–10%.

found to be distributed only in the nuclei of living cells, withoutobvious cytoplasmic distribution (Fig. 5D), which indicated thatprobe molecules in non-transfected cells and the excessive probesin plasmid-transfected cells were both distributed in the nuclei.

GFP-RT has the same distribution pattern in HeLa cells withor without 93del5dMB transfection, showing that the aptamerbeacon had no effect on the protein intracellular localization. Asintracellular degradation of molecular beacon remains a prob-lem, the current investigation was mainly focused on the initialphase of probe uptake by cells within 1 h to ensure the reli-ability of the results (Matsuo, 1998; Wang et al., 2008). Usingthis strategy, we optimized the dose of the aptamer beaconsand the delivery system to improve the signal-to-backgroundratio.

3.4. Visualization of HIV-1 RT in HIV-1-integrated U1 cells

93del5dMB was subsequently used for RT imaging in HIV-

1-integrated U1 cells. It was delivered into the activated HIV-1integrated U1 cells. About 45 min later, the resultant fluorescencesignal of 93del5dMB appeared in the cytoplasm, as shown in Fig. 6A,consistent with the cytoplasmic localization of RT in HIV-1-infected

Y. Liang et al. / Biosensors and Bioelectronics 28 (2011) 270– 276 275

Fig. 5. Localization of 93del5dMB and GFP-RT in plasmid-transfected HeLa cells. (A) Co-localization of 93del5dMB and GFP-RT in living plasmid-transfected HeLa cells afterdelivery by SLO for 45 min. (B) 93del5dF was co-localized with GFP-RT in the cytoplasm of living plasmid-transfected HeLa cells, and excess probes were distributed in cellnuclei. (C) Low fluorescence background of 93del5dMB in non-plasmid-transfected HeLa cells delivered by SLO for 45 min. (D) 93del5dF shows concentrated fluorescencesignals in the nucleus after being delivered into non-plasmid-transfected HeLa cells. Bars, 10 �m.

Fig. 6. Visualization of HIV-1 RT in HIV-1-integrated U1 cells. (A) Imaging endogenous HIV-1 RT in living activated U1 cells. (Left) Endogenous HIV-1 RT, detected withTAMRA-labeled 93del5dMB in activated U1 cells, was found in the cytoplasm. (Right) Control non-activated U1 cells showed no obvious fluorescence after being deliveredwith 93del5dMB. (B) 93del5dMB co-localized with HIV-1 RT protein in the fixed activated U1 cells. (C) Low fluorescence background of 93del5dMB and HIV-1 RT protein inthe fixed non-activated U1 cells. Bars, 10 �m.

2 Bioel

lc

sdalswco9asi

tcct

4

ccrpstsa

bbaflma

A

t

2001. Proc. Natl. Acad. Sci. U.S.A. 98, 3185–3190.

76 Y. Liang et al. / Biosensors and

iving cells. There was little fluorescence signal in non-activated U1ells at 45 min after the addition of SLO and 93del5dMB.

Immunofluorescence assay was also carried out to verify theignal specificity of the aptamer beacon. After 93del5dMB beingelivered into the activated U1 cells for 45 min, cells were fixednd HIV-1 RT protein that was expressed by activated U1 cells wasabeled by the specific fluorescent-dye-conjugated antibody. Ashown in Fig. 6B, immunofluorescence signal could be co-localizedith 93del5dMB signal in the cytoplasm of the fixed activated U1

ells, and from SoftWoRx 5.0 analysis, we got Pearson coefficientf correlation of 0.7239. As a control, there was very little signal of3del5dMB or the endogenous HIV-1 RT protein in the fixed non-ctivated U1 cells (Fig. 6C). These results showed that 93del5dMBpecifically bound to the endogenous HIV-1 RT in living HIV-1-nfected cells.

Above all, we concluded that 93del5dMB could specifically bindo the endogenous HIV-1 RT and report fluorescent signal in theytoplasm of living cells. So, as a small and easy-to-use oligonu-leotide probe, the aptamer beacon should be a very good choiceo image the behavior of endogenous proteins in living cells.

. Conclusion

Till now, direct visualization of endogenous proteins in livingells is still a big challenge. In this report, the aptamer beacon wasonstructed by combining the advantage of aptamers with MBs toecognize and image endogenous protein. Aptamer beacons shouldossess good affinity and specificity to target proteins and have sen-itive signal transduction function. Findings in this study indicatehat aptamer beacon is a good system for protein recognition andignal transduction in living cells. The process is straightforwardnd has no obvious adverse impact on virus-infected cells.

As a key protein for HIV-1 infection, HIV-1 RT was used touild the model system for visualization of endogenous proteinsy aptamer beacons. We designed the aptamer beacon 93del5dMBgainst HIV-1 RT. And the aptamer beacon was successfully usedor detecting and imaging HIV-1 RT in solution and in HIV-infectediving cells. The proposed method may be extended to visualize

any other endogenous proteins in living cells by using appropri-te aptamer beacons.

cknowledgments

We thank Prof. Lijun Bi and Dr. Yuanyuan Chen for their help inhis work. ZQ Cui and ZP Zhang were supported by the National

ectronics 28 (2011) 270– 276

Basic Research Program of China (no. 2011CB933600) and theNational Natural Science Foundation of China (no. 30700169 and31070774). The other authors were supported by Chinese Academyof Sciences (no. KSCX2-EW-Q-15).

References

Andreola, M.L., Pileur, F., Calmels, C., Ventura, M., Tarrago-Litvak, L., Toulmeand, J.J.,Litvak, S., 2001. Biochemistry 40, 10087–10094.

Anh Tuan Phan, V.K., Jin-Biao, M., Faure, A., Andréola, M.L., Patel, D., 2005. Proc. Natl.Acad. Sci. U.S.A. 102, 634–639.

Chalfie, M., Tu, Y., Euskirchen, G., Ward, W.W., Prasher, D.C., 1994. Science 263,802–805.

Chudakov, D., Lukyanov, S., Lukyanov, K., 2005. Trends Biotechnol. 23, 605–613.Coffin, J.M., Hughes, S.H., Varmus, H.E., 1997. Retroviruses. C. S. H. L. Press, Cold

Spring Harbor, New York.Cui, Z.Q., Zhang, Z.P., Zhang, X.E., Wen, J.K., Zhou, Y.F., Xie, W.H., 2005. Nucleic Acids

Res. 33, 3245–3252.Davies, J.F., Hostomska, Z., Hostomsky, Z., Jordan, S.R., Matthews, D.A., 1991. Science

252, 88–95.Deka, C., Lehnert, B.E., Lehnert, N.M., 1998. Cytometry (Part A) 25, 271–279.Ellington, A.D., Szostak, J.W., 1990. Nature 346, 818–822.Faria, M., Spiller, D.G., Dubertret, C., Nelson, J.S., White, M.R.H., Scherman, D., Héle‘ne,

C., Giovannangeli, C., 2001. Nat. Biotechnol. 19, 40–44.Hostomsky, Z., Hostomska, Z., Hudson, G.O., Moomaw, E.W., Nodes, B.R., 1991. Proc.

Natl. Acad. Sci. U.S.A. 88, 1148–1152.Jenison, R.D., Gill, S.C., Pardi, A., Polisky, B., 1994. Science 263, 1425–1429.Jhaveri, S., Rajendran, M., Ellington, A.D., 2000. Nat. Biotechnol. 18, 1293–1297.Kanda, T., Sullivan, K.F., Wahl, G.M., 1998. Curr. Biol. 8, 377–385.Leonetti, J.P., Mechti, N., Degols, G., Gagnor, C., Lebleu, B., 1991. Proc. Natl. Acad. Sci.

U.S.A. 88, 2702–2706.Li, J.J., Fang, X.H., Tan, W.H., 2002. Biochem. Biophys. Res. Commun. 292, 31–40.Matsuo, T., 1998. Biochem. Biophys. Acta 1379, 178–184.McBeath, E., Fujiwara, K., 1984. J. Cell Biol. 99, 2061–2073.Mhlanga, M.M., Vargas, D.Y., Fung, C.W., Kramer, F.R., Tyagi, S., 2005. Nucleic Acids

Res. 33, 1902–1912.Nobuko, H., Andrew, E., Martin, S., 2001. Anal. Biochem. 294, 126–131.Potyrailo, R.A., Conrad, R.C., Ellington, A.D., Hieftje, G.M., 1998. Anal. Chem. 70,

3419–3425.Santangelo, P.J., Nix, B., Tsourkas, A., Bao, G., 2004. Nucleic Acids Res. 32, e57.Sokol, D.L., Zhang, X.L., Lu, P., 1998. Proc. Natl. Acad. Sci. U.S.A. 95, 11538–11543.Soultrait1, V.R., de Lozach, P.Y., Altmeyer, R., Tarrago-Litvak, L., Litvak, S., Andréola,

M.L., 2002. Agents J. Mol. Biol. 324, 195–203.Spiller, D.G., Tidd, D.M., 1995. Antisense Res. Dev 5, 13–21.Stephen P.A., Laura T., 1991. US Patent 5,324,633.Summerer, D., Chen, S., Wu, N., Deiters, A., Chin, J.W., Schultz, P.G., 2006. Proc. Natl.

Acad. Sci. U.S.A. 103, 9785–9789.Tuerk, C., Gold, L., 1990. Science 249, 505–510.Tyagi, S., Kramer, F.R., 1996. Nat. Biotechnol. 14, 303–308.Tyagi, S., Alsmadi, O., 2004. Biophys. J. 87, 4153–4162.Walev, I., Bhakdi, S.C., Hofmann, F., Djonder, N., Valeva, A., Aktories, K., Bhakdi, S.,

Wang, W., Cui, Z.Q., Han, H., Zhang, Z.P., Wei, H.P., Zhou, Y.F., Chen, Z., Zhang, X.E.,2008. Nucleic Acids Res. 36, 4913–4928.

Yamamoto, R., Kumar, P.K.R., 2000. Genes to Cells 5, 389–396.Yuste, R., 2005. Nat. Methods 2, 902–904.