Nanoscale

COMMUNICATION

Cite this: Nanoscale, 2015, 7, 8332

Received 16th March 2015,Accepted 8th April 2015

DOI: 10.1039/c5nr01705j

www.rsc.org/nanoscale

NanoCluster Beacons as reporter probes in rollingcircle enhanced enzyme activity detection†

Sissel Juul,‡a Judy M. Obliosca,‡b Cong Liu,b Yen-Liang Liu,b Yu-An Chen,b

Darren M. Imphean,b Birgitta R. Knudsen,c,d Yi-Ping Ho,d Kam W. Leong*e andHsin-Chih Yeh*b

As a newly developed assay for the detection of endogenous

enzyme activity at the single-catalytic-event level, Rolling Circle

Enhanced Enzyme Activity Detection (REEAD) has been used to

measure enzyme activity in both single human cells and malaria-

causing parasites, Plasmodium sp. Current REEAD assays rely on

organic dye-tagged linear DNA probes to report the rolling circle

amplification products (RCPs), the cost of which may hinder the

widespread use of REEAD. Here we show that a new class of

activatable probes, NanoCluster Beacons (NCBs), can simplify the

REEAD assays. Easily prepared without any need for purification

and capable of large fluorescence enhancement upon hybridiz-

ation, NCBs are cost-effective and sensitive. Compared to conven-

tional fluorescent probes, NCBs are also more photostable. As

demonstrated in reporting the human topoisomerases I (hTopI)

cleavage-ligation reaction, the proposed NCBs suggest a read-out

format attractive for future REEAD-based diagnostics.

Introduction

Rolling Circle Enhanced Enzyme Activity Detection (REEAD) isa novel method to detect enzymatic activities by turning singleenzyme catalytic activities into isothermally amplified nucleicacid products.1–4 REEAD has been demonstrated for measur-ing cancer-relevant enzymes in single cells1–3 and for detectingan enzyme activity specific to the malaria-causing Plasmodium

parasites.4 While the REEAD assay is robust, specific, andcapable of multiplexed detection of target enzyme activitieseven in crude cell extracts, the assay cost is relatively high andsample preparation requires multiple steps of washing andseparation, as well as an addition of the anti-fading agent toascertain the detectable fluorescence signal. To push for theuse of REEAD for detection of infectious diseases at point-of-care settings, it is necessary to reduce the cost and simplifythe preparation procedures of this assay while maintaining itsreliability. Current REEAD assays rely on organic dye-taggedlinear probes to report the rolling circle amplification products(RCPs).5 While these organic dye-tagged reporters are easy touse, they are not cost-effective and can only provide modesttarget-to-background (T/B) ratio even with the intervention ofanti-fading agents.

Among all the fluorescent reporters available, activatableprobes,6,7 in particular, offer a high T/B ratio in moleculardetection – they fluoresce only upon binding with specifictarget molecules but otherwise remain dark. While a high T/Bratio makes quantification easier and elimination of the needto remove the unbound probes simplifies the assay,7 activat-able probes are often much more expensive than the organicdye-tagged reporter probes and difficult to prepare. Forinstance, molecular beacons,8 the most widely used activatableprobes for DNA detection, need to be dually labeled (i.e. withan organic dye at one end of the hairpin and a quencher at theother end). Removal of excess dyes and singly labeled impuri-ties during the manufacturing process is necessary, addingpreparation complexity and cost to the molecular beacons.Both semiconductor quantum dots and fluorescent proteinshave been converted to activatable probes for DNAdetection,9–11 but again the material costs are high and thepreparation processes are not straightforward.

Here we present a versatile strategy to design activatableprobes for REEAD assays that are not only simple but cost-effective. Our probes use few-atom silver nanoclusters (Ag NCs)as fluorescent reporters that can be prepared at room tempera-ture via sequential mixing of three inexpensive components ina buffer: a cytosine-rich oligonucleotide, a silver salt, and a

†Electronic supplementary information (ESI) available: The detailed steps ofNCB preparation, REEAD assay and STEM imaging. The sequences of the sNCBand the REEAD substrate. See DOI: 10.1039/c5nr01705j‡These authors contributed equally.

aDepartment of Biomedical Engineering, Pratt School of Engineering,

Duke University, Durham, NC 27708, USAbDepartment of Biomedical Engineering, Cockrell School of Engineering, University of

Texas at Austin, Austin, TX 78712, USA. E-mail: tim.yeh@austin.utexas.educDepartment of Molecular Biology and Genetics, Aarhus University, 8000 Aarhus C,

DenmarkdInterdisciplinary Nanoscience Center (iNANO), Aarhus University, 8000 Aarhus C,

DenmarkeDepartment of Biomedical Engineering, Columbia University, New York, NY 10027,

USA. E-mail: kwl2121@columbia.edu

8332 | Nanoscale, 2015, 7, 8332–8337 This journal is © The Royal Society of Chemistry 2015

Publ

ishe

d on

14

Apr

il 20

15. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

Lib

rari

es o

n 21

/07/

2015

00:

01:2

8.

View Article OnlineView Journal | View Issue

reducing reagent (see ESI† for detailed processes).12–14 Thereis no need to remove any excess reactants as they are essen-tially non-fluorescent, thus eliminating any purification cost.Moreover, upon interaction with a nearby DNA sequence(called an enhancer sequence15), silver clusters exhibit fluo-rescence activatability15–17 and tunability17,18 that cannot beobtained with organic dyes, luminescent nanocrystals andfluorescent proteins.12 These properties have led to the deve-lopment of NanoCluster Beacon (NCB, an activatable probe)for DNA detection16 and chameleon NanoCluster Beacon(cNCB, a multicolor probe) for single-nucleotide polymorph-ism identification.18 In particular, not relying on resonanceenergy transfer as activation or color-switching mechanism,NCBs have achieved a T/B ratio that is five times better thanthat of a conventional molecular beacon in DNA sensing.16

While being explored by researchers in the detection of pro-teins19,20 and metabolites,21 NCBs have not been used insurface-based DNA detection. Due to their low cost and highperformance (high T/B ratio and good photostability), NCBsare perfect reporter probes for REEAD assays.

Results and discussion

In REEAD assays (Fig. 1A),1–4 the enzyme (e.g. human topoiso-merase I, hTopI) under detection first converts a custom-designed linear DNA substrate (the dumbbell structure) into acircularized product through cleavage and ligation. The circula-rized substrate is then used as the template for isothermalrolling circle amplification (RCA) on a glass surface (see ESI†:Materials and Methods and Table S1), leading to multiple(∼103) tandem copies of the circularized substrate called rollingcircle amplification products (RCPs). Traditionally, these RCPsare visualized under a fluorescence microscope by labelingthem with organic dye-tagged DNA probes.5 This organic fluor-ophore labeling allows direct quantification of single enzymaticevents by simply counting the resulting fluorescent RCP dots.

By replacing organic dye-tagged reporter probes with NCBs,we take advantage of the low cost and the “fluorescing-upon-hybridization” nature of NCBs.12,16,17 The conventional NCBhas a binary probe configuration (Fig. 1B),6 having two oligo-nucleotide strands (an NC probe and an enhancer probe) that

Fig. 1 (A) NCB/REEAD detection scheme (not drawn to scale). The detection of enzymatic activity involves three steps: (1) enzyme-mediated (i.e.human topoisomerase I, hTopI) circularization of a synthetic DNA dumbbell hTopI substrate, (2) signal enhancement via an immobilized DNA primerand rolling circle amplification (RCA), and (3) hybridization of RCA products (RCPs) with activatable NanoCluster Beacons (NCBs) for detection. Theright loop of the substrate is designed for primer binding while the left loop carries a guanine-rich enhancer sequence. (B) The conventional binaryNCB consists of an NC probe (the cytosine-rich NC-nucleation sequence is shown in blue) and an enhancer probe (the enhancer sequence isshown in red). When NCB binds to a target, silver clusters interact with the enhancer sequence and turn on. (C) For the single NCB (sNCB) used inthis study, the enhancer sequence is embedded in the RCPs. As a result, RCPs not only serve as a binding target but also an enhancer that turns onthe fluorescence of bound sNCBs. In this configuration, only the NC probe is needed in an sNCB.

Nanoscale Communication

This journal is © The Royal Society of Chemistry 2015 Nanoscale, 2015, 7, 8332–8337 | 8333

Publ

ishe

d on

14

Apr

il 20

15. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

Lib

rari

es o

n 21

/07/

2015

00:

01:2

8.

View Article Online

bind in juxtaposition to a DNA target. Such a juxtapositionbinding enabled the enhancer sequence to interact with thesilver cluster, transforming the cluster from a non-emissivespecies to a highly fluorescent species. Fluorescence thusoccurs only when a specific DNA target is present in thesample.12 Extended from the conventional binary NCB design(Fig. 1B), we have developed a simplified strategy to label RCPsthat further reduces the probe cost. In this new design, which

we termed single NCB (sNCB, Fig. 1C), the enhancer sequenceis programmed to be embedded in the RCPs, which functionsboth as a target and an enhancer (Fig. 1C). In this case, thesNCB consists only of the NC probe as there is no need ofG-rich enhancer probe.

We examined a dumbbell DNA substrate designed for topo-isomerase I activity detection. As shown in Fig. 1A, an sNCB(refer to Fig. S1† for sequence information) was designed to

Fig. 2 Comparison between the conventional REEAD assay and the proposed NCB/REEAD platform. (A) Fluorescence images of a conventional Rh/REEAD assay chip, a sNCB/REEAD chip, a control (no RCPs on the chip, stained with sNCBs), and a chip with in situ synthesized fluorescent Ag2Oparticles. Images shown here are taken from frame 1 and frame 120 of a movie required under continuous illumination (frame rate is 1 Hz). (B) Fluo-rescence decay of representative RCPs labeled with rhodamine probes and sNCBs, respectively. (C) Low- and high-magnification STEM images ofsNCB-labeled RCPs showing an average size of 1 μm with three-dimensional plate-like wavy structures.

Communication Nanoscale

8334 | Nanoscale, 2015, 7, 8332–8337 This journal is © The Royal Society of Chemistry 2015

Publ

ishe

d on

14

Apr

il 20

15. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

Lib

rari

es o

n 21

/07/

2015

00:

01:2

8.

View Article Online

hybridize with the corresponding RCPs. This sNCB design wasfirst tested on a truncated RCP-mimicking synthetic target(60 nt long) in a standard homogeneous solution assay.16

Upon hybridization (i.e. fluorescence activation), sNCB lit upin solution (Fig. S1A†), to produce red emission (Fig. S1B andC†) and displayed an ensemble enhancement ratio greaterthan 1000-fold (see Fig. S1D† for the enhancement ratio defi-nition). A molecular beacon (MB)8,16 using FAM/Dabcyl as theFRET pair was also designed to hybridize to the same synthetictarget for comparison. As shown in Fig. S2,† the signal-to-back-ground (S/B) ratio of sNCB was found ∼30-fold higher thanthat of MB on the same target, making sNCBs a better choicefor our surface-based assay.

The procedure for REEAD has previously been described3,4

and is summarized in the ESI (Materials and Methods andFig. S3†). Initially, RCPs were labeled with multiple rhodamine(Rh)-tagged DNA probes for visualization. However, these Rh-tagged probes bleached rapidly (Fig. 2A). Compared to theorganic dyes, Ag NCs have previously shown a betterphotostability22–24 and stronger single-fluorophore bright-ness.22,23 Therefore we expected that when sNCBs were used asreporter probes, we would see RCPs that are brighter andlonger-lasting. Indeed, under identical illumination andimaging conditions, sNCB-labeled RCPs looked bright andtheir fluorescence faded away much slowly (Fig. 2A and Moviesavailable in ESI†). The average decay times are 28 s for Rhprobes and 186 s for sNCBs (Fig. 2B).

Although the synthetic yield of fluorescent silver cluster canvary between 5%16,24 and 45%25 and the sNCBs used in thisstudy were not purified (i.e. containing non-functional sNCBs),the sNCB-labeled RCPs were about 10× brighter than the Rh-labeled RCPs in frame 1 (Fig. 2A). We believe this difference inthe “appearing” brightness was mainly caused by the differ-ence in the photostability between rhodamine and Ag NCs (inframe 1, many Rh probes had photobleached already). AssNCB-labeled RCPs still bleached after long illumination, weknew that those bright spots were not impurities (e.g. particlesfrom chip dicing) which did not bleach at all.

One may ask why we didn’t use RCPs directly as templatesto synthesize fluorescent Ag NCs and the in situ synthesized AgNCs as reporters for RCPs. Similar ideas have been previouslydiscussed in solution-based isothermal amplification assays.26

To test this idea, we submerged the REEAD chips with andwithout RCPs in a silver nitrate solution and then added withsodium borohydride to the solution. On both chips, manybright spots were seen after silver reduction. Those bright spotswere nonspecific to RCPs and increasing in overall numbersafter long illumination (Fig. 2A). These evidences led us tobelieve that the observed bright dots were silver oxide (Ag2O)nanoparticles27 rather than silver clusters. Photoactivation ofindividual Ag2O nanoparticles has been previously reported.27

We found that fluorescence of silver oxide particles can be acti-vated more effectively with shorter wavelength excitation (e.g.blue light), which agrees well with the literature.27 These silveroxide particles emit as single-quantum systems – they blinkunder the microscope (Fig. 3A) and bleach in a single step(Fig. 3B). On the other hand, fluorescent RCPs bleach gradu-ally (because multiple emitters are incorporated in each RCPrather than a single emitter, see Fig. 3C) and the average dotsize is larger than the diffraction limit spot size. We hypoth-esize that the surface coating of our REEAD chips (CodeLink®activated slides from SurModics), a hydrophilic polymer con-taining N-hydroxysuccinimide (NHS) ester reactive groups, sup-ports the formation of silver islands during the reductionprocess, which then become silver oxide nanoparticles uponexposure to air.27 As a result, in situ synthesis of fluorescent AgNCs cannot be used to report the existence of RCPs here.

The labeled RCPs had a size around 1 µm when observedusing standard fluorescence microscopy. We further examinedthe morphology of RCPs under a scanning transmission elec-tron microscope (STEM). STEM images (Fig. 2C) show RCPswith an average size of 1 μm and an inter-RCP distanceranging from 6 to 20 μm, similar to the inter-dot distance thatwe observed via fluorescence microcopy. We noticed that RCPsdid not have a wool ball-like conformation as shown in the pre-vious report,28 but rather display three-dimensional plate-like

Fig. 3 Fluorescence time traces of representative Ag2O particles showed (A) blinking and (B) single-step photon bleaching. (C) On contrary, thetime trace of an sNCB-labeled RCP exhibited a gradual fluorescence decay.

Nanoscale Communication

This journal is © The Royal Society of Chemistry 2015 Nanoscale, 2015, 7, 8332–8337 | 8335

Publ

ishe

d on

14

Apr

il 20

15. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

Lib

rari

es o

n 21

/07/

2015

00:

01:2

8.

View Article Online

wavy structures (Fig. S4†). Through STEM, we could not tell thedifference in morphology for the unlabeled RCPs, RCPslabeled with Rh probes, and RCPs labeled with NCBs.

We compared the quantification results of endogenoushTopI activity (primary target for several cancer chemothera-peutics) in crude human cell extract using Rh-tagged probesand sNCB as reporters, respectfully (Fig. 4). Since rhodaminebleaches quickly under illumination (Fig. 2A), an anti-fadingmounting medium, Vectashield®, was used to improvethe quantification of Rh-labeled RCPs. Fig. S5† shows the

quantification results from three replicates of 10, 100 and1000 cells in the proposed sNCB/REEAD assay. hTopI signalswere analyzed using Image J software. The number ofhTopI signal increases significantly with the increasingnumber of HEK 293 cells. Although in both cases (rhodamineprobe and sNCB) the hTopl signals are correlated to thecell amount, sNCB detection result is superior to that ofrhodamine probe at the low cell quantities (10 and 100 cells).These results demonstrated that sNCBs can substitute rhoda-mine probes in the REEAD assays. Since Ag NCs are more

Fig. 4 Detection of human topoisomerase I, hTopI, activity in crude human HEK293 cell extracts. Top row: REEAD chips without any labeling.Middle row: the conventional Rh/REEAD assays. To facilitate the quantification of Rh-labeled RCPs, an anti-fading mounting medium was used here.Bottom row: the proposed sNCB/REEAD assays (without the anti-fading medium). Samples were prepared at 3 different cell concentrations (10 cellsper 5 μl, 100 cells per 5 μl and 1000 cells per 5 μl).

Communication Nanoscale

8336 | Nanoscale, 2015, 7, 8332–8337 This journal is © The Royal Society of Chemistry 2015

Publ

ishe

d on

14

Apr

il 20

15. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

Lib

rari

es o

n 21

/07/

2015

00:

01:2

8.

View Article Online

photostable than rhodamine, an anti-fading medium is notneeded when sNCBs are used as reporters. This furtherreduces the cost of the sNCB/REEAD assay.

In conclusion, we have demonstrated a versatile strategy todesign new activatable probes, single NanoCluster Beacons(sNCBs), for imaging and quantifying rolling circle amplifica-tion products (RCPs) in an enzyme activity assay (REEAD). Dis-tinct from the conventional binary NCBs15,16 and thechameleon NCBs18 reported before, single NCBs have theirguanine-rich enhancer sequence directly incorporated intotheir targets (i.e. the RCPs). Therefore, sNCB only contains anNC probe and it lights up upon binding with the target. Themajor contribution of this report is to demonstrate that the“embedded” G-rich sequence can work as the G-rich “over-hang” in the fluorescence activation of silver clusters. The keyadvantages of the proposed sNCB/REEAD assay lie in lowercost (no need of probe purification and anti-fading medium),simpler preparation procedure (no need of stringency wash toremove the unbound probes, see ESI†), and better imagingquality (resistance to photobleaching). We expect that theconcept of sNCBs can be generally applied to a wide variety ofisothermal amplification assays where the enhancer sequencecan be embedded in the amplicons, such as NASBA,29 SDA,30

and LAMP.31

Conflict of Interest

The authors declare no competing financial interest.

Acknowledgements

This work is financially supported by Robert A. Welch Foun-dation (F-1833 to H.-C.Y.) and the Danish Research Council(11-105736/FSS to S. J. and 11-116325/FTP to Y. P. H.). Supportfrom NIH (AI096305) is also acknowledged.

References

1 F. F. Andersen, M. Stougaard, H. L. Jorgensen, S. Bendsen,S. Juul, K. Hald, A. H. Andersen, J. Koch andB. R. Knudsen, ACS Nano, 2009, 3, 4043–4054.

2 M. Stougaard, J. S. Lohmann, A. Mancino, S. Celik,F. F. Andersen, J. Koch and B. R. Knudsen, ACS Nano, 2009,3, 223–233.

3 S. Juul, Y. P. Ho, J. Koch, F. F. Andersen, M. Stougaard,K. W. Leong and B. R. Knudsen, ACS Nano, 2011, 5, 8305–8310.

4 S. Juul, C. J. F. Nielsen, R. Labouriau, A. Roy, C. Tesauro,P. W. Jensen, C. Harmsen, E. L. Kristoffersen, Y. L. Chiu,R. Frohlich, P. Fiorani, J. Cox-Singh, D. Tordrup, J. Koch,A. L. Bienvenu, A. Desideri, S. Picot, E. Petersen,K. W. Leong, Y. P. Ho, M. Stougaard and B. R. Knudsen,ACS Nano, 2012, 6, 10676–10683.

5 P. M. Lizardi, X. H. Huang, Z. R. Zhu, P. Bray-Ward,D. C. Thomas and D. C. Ward, Nat. Genet., 1998, 19, 225–232.

6 D. M. Kolpashchikov, Chem. Rev., 2010, 110, 4709–4723.7 H. Kobayashi, M. Ogawa, R. Alford, P. L. Choyke and

Y. Urano, Chem. Rev., 2010, 110, 2620–2640.8 S. Tyagi and F. R. Kramer, Nat. Biotechnol., 1996, 14, 303–308.9 C. Y. Zhang, H. C. Yeh, M. T. Kuroki and T. H. Wang, Nat.

Mater., 2005, 4, 826–831.10 J. R. Porter, C. I. Stains, D. J. Segal and I. Ghosh, Anal.

Chem., 2007, 79, 6702–6708.11 M. Levy, S. F. Cater and A. D. Ellington, ChemBioChem,

2005, 6, 2163–2166.12 J. M. Obliosca, C. Liu and H.-C. Yeh, Nanoscale, 2013, 5,

8443–8461.13 J. T. Petty, J. Zheng, V. Nicholas and R. M. Dickson, J. Am.

Chem. Soc., 2004, 126, 5207–5212.14 J. T. Petty, S. P. Story, J. C. Hsiang and R. M. Dickson,

J. Phys. Chem. Lett., 2013, 4, 1148–1155.15 J. M. Obliosca, M. C. Babin, C. Liu, Y.-L. Liu, Y.-A. Chen,

R. A. Batson, M. Ganguly, J. T. Petty and H.-C. Yeh, ACSNano, 2014, 8, 10150–10160.

16 H.-C. Yeh, J. Sharma, J. J. Han, J. S. Martinez andJ. H. Werner, Nano Lett., 2010, 10, 3106–3110.

17 H.-C. Yeh, J. Sharma, J. J. Han, J. S. Martinez andJ. H. Werner, IEEE Nanotechnol. Mag., 2011, 5, 28–33.

18 H.-C. Yeh, J. Sharma, M. Shih Ie, D. M. Vu, J. S. Martinezand J. H. Werner, J. Am. Chem. Soc., 2012, 134, 11550–11558.

19 J. Yin, X. He, K. Wang, F. Xu, J. Shangguan, D. He andH. Shi, Anal. Chem., 2013, 85, 12011–12019.

20 J. J. Li, X. Q. Zhong, H. Q. Zhang, X. C. Le and J. J. Zhu,Anal. Chem., 2012, 84, 5170–5174.

21 M. Zhang, S. M. Guo, Y. R. Li, P. Zuo and B. C. Ye, Chem.Commun., 2012, 48, 5488–5490.

22 H.-C. Yeh, J. Sharma, H. Yoo, J. S. Martinez andJ. H. Werner, Proc. SPIE, 2010, 7576, 75760N.

23 C. I. Richards, S. Choi, J.-C. Hsiang, Y. Antoku, T. Vosch,A. Bongiorno, Y.-L. Tzeng and R. M. Dickson, J. Am. Chem.Soc., 2008, 130, 5038–5039.

24 S. Choi, J. Yu, S. A. Patel, Y.-L. Tzeng and R. M. Dickson,Photochem. Photobiol. Sci., 2011, 10, 109–115.

25 P. R. O’Neill, K. Young, D. Schiffels and D. K. Fygenson,Nano Lett., 2012, 12, 5464–5469.

26 Y. Q. Liu, M. Zhang, B. C. Yin and B. C. Ye, Anal. Chem.,2012, 84, 5165–5169.

27 L. A. Peyser, A. E. Vinson, A. P. Bartko and R. M. Dickson,Science, 2001, 291, 103–106.

28 J. Lee, S. M. Peng, D. Y. Yang, Y. H. Roh, H. Funabashi,N. Park, E. J. Rice, L. W. Chen, R. Long, M. M. Wu andD. Luo, Nat. Nanotechnol., 2012, 7, 816–820.

29 J. Compton, Nature, 1991, 350, 91–92.30 G. T. Walker, M. C. Little, J. G. Nadeau and D. D. Shank,

Proc. Natl. Acad. Sci. U. S. A., 1992, 89, 392–396.31 T. Notomi, H. Okayama, H. Masubuchi, T. Yonekawa,

K. Watanabe, N. Amino and T. Hase, Nucleic Acids Res.,2000, 28.

Nanoscale Communication

This journal is © The Royal Society of Chemistry 2015 Nanoscale, 2015, 7, 8332–8337 | 8337

Publ

ishe

d on

14

Apr

il 20

15. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

Lib

rari

es o

n 21

/07/

2015

00:

01:2

8.

View Article Online