Gold nanoprobe assay for the identification of

ORIGINAL ARTICLE

BACTERIOLOGY

Gold nanoprobe assay for the identification of mycobacteria of the Mycobacterium tuberculosis complex P. Costa1,2, A. Amaro2, A. Botelho2, J. Inaćio2 and P. V. Baptista1 1) CIGMH/DCV, Faculdade de Cieˆncias e Tecnologia, Universidade Nova de Lisboa, Campus de Caparica, Caparica and 2) Instituto Nacional de Recursos Biolo´gicos, I.P., Laborato´rio Nacional de Investigaçaõ Veterina´ria (INRB, I.P., LNIV), Estrada de Benfica, Lisboa, Portugal

Abstract Members of the Mycobacterium tuberculosis complex (MTC) are causative agents of human and animal tuberculosis. This complex encompasses several phylogenetically related species, including M. tuberculosis, the main aetiological agent of human tuberculosis, and Mycobacterium bovis, the causative agent of bovine tuberculosis, a relevant worldwide zoonosis. Clear epidemiological evaluation of appropriate and effective treatment requires unambiguous differentiation between MTC members. Routine diagnosis has been increasingly relying on the molecular identification of MTC members. In the present study, we report the use of a gold nanoparticle-based approach for the sensitive, specific and fast identification of MTC and for the differentiation of M. bovis and M. tuberculosis using the gyrB locus as target. This gold nanoprobe strategy relies on the colorimetric differentiation of specific DNA sequences based on differential aggregation profiles in the presence or absence of specific target hybridization. Three nanoprobes were designed and successfully used for the specific identification of members of MTC, M. bovis and M. tuberculosis. Keywords: Gold nanoprobes, Mycobacterium bovis, Mycobacterium tuberculosis complex Original Submission: 4 June 2009; Revised Submission: 3 November 2009; Accepted: 12 November 2009 Editor: M. Drancourt Article published online: 20 November 2009 Clin Microbiol Infect 2010; 16: 1464–1469 10.1111/j.1469-0691.2010.03120.x

Corresponding author: P. V. Baptista, Centro de Investigaçaõ em Gene´tica Molecular Humana/Department Cieˆncias da Vida, FCT/UNL, Campus de Caparica, 2829-516 Caparica, Portugal E-mail: [email protected]

Introduction Tuberculosis is a global health problem associated with significant morbidity, causing epidemics and resulting in the death of approximately two million people per year [1]. This destructive disease is caused by members of the Mycobacterium tuberculosis complex (MTC), a group of closely-related pathogenic species that includes M. tuberculosis, Mycobacterium bovis, Mycobacterium caprae, Mycobacterium africanum and Mycobacterium microti [2] . Despite M. tuberculosis being the most common cause of tuberculosis in humans, M. bovis is responsible for 0.5–7.2% of human tuberculosis cases in industrialized nations and is estimated to be responsible for 10–15% of new cases in the developing world [3]. M. bovis is also the causative agent of

bovine tuberculosis, a zoonosis with high important socioeconomic and public health effects [1,4,5]. Rapid identification of mycobacteria is of paramount importance for unequivocal diagnostics leading to effective infection control. During the past two decades, the development of molecular detection and identification methods using several appropriate loci has allowed the rapid differentiation between the members of the MTC [6–9]. These molecular methods constitute an advantage over the more time-consuming and laborious biochemical characterization [10]. One such molecular assay uses the amplification of the gyrB gene, coding for the B subunit of DNA gyrase (a type II topoisomerase) essential for bacterial replication, followed by analysis of polymorphic sites via PCR-restriction fragment length polymorphism (RFLP) [11,12]. The use of thiol-linked ssDNA-modified gold nanoparticles (Au-nanoprobes) for the colorimetric detection of DNA targets represents an inexpensive and easy-to-perform alternative to common molecular assays [13–16]. The surface plasmon resonance (SPR) of gold nanoparticles is responsible for their intense colours. In solution, monodisperse gold nanoparticles appear red and exhibit a relatively narrow SPR

ª2010 The Authors Journal Compilation ª2010 European Society of Clinical Microbiology and Infectious Diseases

CMI

Costa et al. Gold nanoprobes for identification of mycobacteria

band centered around 520 nm in the UV-visible spectrum. By contrast, a solution containing aggregated gold nanoparticles appears blue, corresponding to a characteristic red shift in the SPR band to higher wavelengths. Our method consists in a visual and/or spectrophotometric comparison of solutions before and after salt-induced Au-nanoprobe aggregation: the presence of complementary target preventing aggregation and the solution remaining red. Noncomplementary/mismatched targets do not prevent Au-nanoprobe aggregation, resulting in a visible change of colour from red to blue. Specific DNA detection based on Au-nanoprobes has proven to be a fast and simple alternative to more cumbersome techniques involving multistep assays. The noncross-linking method for DNA identification has already been successfully applied for the identification of M. tuberculosis in clinical samples based on the rpoB locus [17], gene expression studies and the detection of point mutations [18,19]. In the present study, we report an Au-nanoprobe-based molecular detection assay capable of differentiating between members of the MTC. Based on the gyrB locus, three specific Au-nanoprobes were developed: one for identification of MTC, one for M. tuberculosis and another for M. bovis.

Materials and Methods Strains analysed

In this study, several strains from MTC and non-MTC species were used (Table 1.) The identification of each strain was certified by the National Laboratory of Veterinary Research (INRB, I.P., LNIV), the Portuguese reference laboratory for bovine tuberculosis, where gyrB-PCR-RFLP is routinely used for the identification of MTC species [8]. Strain cultivation and

1465

DNA extraction from cultures were performed according to standard methods described previously [20]. PCR amplification of gyrB

The 1020-bp specific fragment of the gyrB gene, suitable for differentiation of MTC members, was PCR amplified using a procedure adapted from Niemann et al. [12]. Briefly, PCR reactions were carried out in 25 ll using 200 lM of deoxynucleotide triphosphates (Amersham Biosciences, Piscataway, NJ, USA), 1 U of Taq DNA polymerase (Promega, Madison, WI, USA), 2 mM of MgCl2 (Promega), 20 pmol of each primer MTUB-f [5¢-TCG GAC GCG TAT GCG ATA TC-3¢] and MTUB-r [5¢-ACA TAC AGT TCG GAC TTG CG-3¢] and 1 ll of extracted DNA solution. Amplification was performed in a MJmini Thermocycler (Bio-Rad, Hercules, CA, USA) with 35 cycles of 1 min at 95C, 1.5 min at 55C, and 1.5 min at 72C. The PCR products were ethanol precipitated and resuspended in deionized water. The amplified products were electrophoretically analysed in a 1.5% agarose gel (data not shown). Sequence analysis and probe design

Comparative analysis of gyrB gene sequences from mycobacteria was performed using sequence alignment with the CLUSTAL X, version 2.0 [21]. Gene sequences were retrieved from GenBank with accession numbers: AB014184 for M. bovis, AJ276122 for M. caprae, AB014205 for M. microti, Z80233 for M. tuberculosis, AB014202 for Mycobacterium gastri, AB014189 for M. avium subsp. avium and EU029114 for M. avium subsp. paratuberculosis. Probe specificity was tested in silico using the BLAST tools from GenBank. Specific oligonucleotides for MTC, M. tuberculosis and M. bovis were used to functionalize the gold nanoparticles and produce the

TABLE 1. Bacterial strains and clinical isolates used in the present study for the evaluation of specificity of the gold nanoprobe assay Species

Strain*/clinical isolate

gComplex

gMTub

gMBov

Mycobacterium tuberculosis Mycobacterium bovis

ATCC 25177; LNIV 9605 AN5; LNIV 13027, 5530/0/05, 11265, 7230/4, 14421/2, 24497/6, 8855, 5889, 10044, 14577, 13280/6, 13280/4, 34875 and 20564 ATCC 27291 LNIV 17320, 4958/0/05, 8403, 15244 and 20752 ATCC 25291 LNIV 39888 LNIV 23063/4 LNIV 12352

+ +

+ )

) +

+ +

) )

+ )

) ) ) )

) ) ) )

) ) ) )

Mycobacterium bovis BCG Mycobacterium caprae Mycobacterium avium subsp. avium Mycobacterium avium subsp. paratuberculosis Mycobacterium avium subsp. hominissuis Corynebacterium striatum

*ATCC, American Type Culture Collection, USA; LNIV, Laborato´rio Nacional de Investigaçaõ Veterina´ria, Lisbon, Portugal. +, positive (no aggregation); ), negative (aggregation). gComplex, gMTub and gMBov represent the specific Au-nanoprobe text with people specific for mT Complex, M. tuberculosis and M. bovis, respectively.

ª2010 The Authors Journal Compilation ª2010 European Society of Clinical Microbiology and Infectious Diseases, CMI, 16, 1464–1469

1466

Clinical Microbiology and Infection, Volume 16 Number 9, September 2010

Au-nanoprobes: MTC probe (gComplex), 5¢-thiol-CCGAG GACACAGCCTTGTTC-3¢; M. tuberculosis probe (gMTub), 5¢-thiol-TTTGAAGCCAACCCCACCGACG-3¢; and M. bovis probe (gMBov), 5¢-thiol-CGTTTGTGCAGAAGGTCTG TAAT-3¢ (STAB Vida Lda, Portugal).

CMI

opment, the mixtures and the blank were assayed by UVvisible spectroscopy in a microplate reader (Tecan Infinite M200; Tecan, Austria GmbH, Gro¨dig, Austria).

Results and Discussion

Gold nanoparticle and gold nanoprobe synthesis

Gold nanoparticles, with an average diameter of approximately 13 nm, were synthesized by the citrate reduction method described by Lee and Meisel [22]. Briefly, 250 ml of 1 mM HAuCl4 were heated with stirring, 25 ml of 28.8 nM sodium citrate were added, and the solution was refluxed for 15 min. Afterwards, the solution was left at room temperature to cool down. The gold nanoprobes were prepared by incubating the thiol-modified oligonucleotides with the gold nanoparticles for 16 h. The solution was washed with phosphate buffer (pH 7), 10 mM, at increasing salt concentration, aiming to reduce nonspecific bonds between the thiol-modified oligonucleotides and the gold nanoparticles. The solution was centrifuged and the resulting pellet was resuspended in phosphate buffer (pH 7), 10 mM, 0.1 M NaCl, and stored in the dark at 4C until further use. Au-nanoprobe colorimetric assay conditions

Each colorimetric assay was performed in a total volume of 30 ll with Au-nanoprobes at a final concentration of 2.5 nM in phosphate buffer (pH 8), 10 mM, 0.5 M NaCl. For each probe, the assay consisted in the visual/spectrophotometric comparison of a ‘Blank’ [without DNA; phosphate buffer (pH 8), 10 mM, 0.5 M NaCl] with positive samples containing the appropriate amplicon (final concentration of 30 mg/L for the M. tuberculosis probe and 10 mg/L for the MTC and M. bovis probes) and with a negative sample containing noncomplementary DNA (M. avium subsp. avium, M. avium subsp. paratuberculosis, M. avium subsp. hominissuis, Corynebacterium striatum). After 10 min of denaturation at 95C, the mixtures were allowed to stand for 30 min at room temperature and MgCl2 was added to a final concentration of 0.7 M for M. tuberculosis and M. bovis probes and 0.04 M for the MTC probe. After 15 min at room temperature for colour devel-

On the basis of the gyrB gene sequence conserved between species of the MTC, a set of primers was used to PCR amplify a specific 1020-bp fragment of the gene from MTC species only (i.e. no amplicons were generated from nonMTC species) [11,12]. In silico alignment of the gyrB gene sequences showed three regions sufficient to unequivocally discriminate between MTC members (Fig. 1). For proof-ofconcept, we targeted this genomic region shared by all members of the M. tuberculosis complex, and two further probes were designed to specifically identify M. tuberculosis and M. bovis. These sequences were taken for the design of specific thiolated oligonucleotides, which were then used to functionalize the gold nanoparticles. The MTC probe positively identified the members of the MTC used in the assay, at the same time as clearly recognizing the nonmembers (Fig. 2b; Table 1). The M. tuberculosis and M. bovis probes unequivocally identified the respective target (Fig. 2b; Table 1). Furthermore, blind identification of mycobacteria strains isolated from fifteen different clinical samples showed 100% concordance with the results obtained with the gyrB-PCR-RFLP procedure (see Supporting Information, Table S1). Although increasing numbers of mycobacteria belonging to the MTC are isolated from clinical samples, routine identification is not usually carried out to the species level. It is relevant for public health control and treatment of infections caused by members of the MTC that complete molecular characterization is available by means of fast and accurate assays capable of distinguishing among these mycobacteria. Comparative genomics has demonstrated that sequences of the mycobacterium bacillus may provide distinct molecular signatures for the conclusive identification of isolates at the species level [6,11,12,23]. The gyrB gene encodes the B sub-

FIG. 1. Sequence analysis of gyrB genes. Comparative analysis of a segment of gyrB gene sequences from selected mycobacteria. The sequences of the probe targets are typed in bold red and inside boxes and mismatches in relation to the consensus are typed in bold green. GenBank accession numbers are indicated for each sequence. ª2010 The Authors Journal Compilation ª2010 European Society of Clinical Microbiology and Infectious Diseases, CMI, 16, 1464–1469


CMI

1467

(a)

1

2

3

4

5

Absorbance

0.45

FIG. 2. Gold nanoprobe assay. (a) Colorimetric assay (above) and respective spectrophotometry

(below)

for

the

0.40 0.35 0.30 0.25 0.20 400

Mycobacterium tuberculosis complex (MTC) probe (black lines represent the red solu-

500

600 Wavelength (nm)

tions associated with complementary target: non-aggregated Au-nanoprobes; the

1.2

ciated with noncomplementary target: agg-

1.15

The

strains

ATCC 25177 (1); Mycobacterium bovis LNIV 13027 (2) and LNIV 5530/0/05 (3); Myco-

1.1 Abs526/600

Au-nanoprobe).

analysed were: Mycobacterium tuberculosis

800

gComplex probe

(b)

grey line represents the blue solution assoregated

700

1.05 1 0.95

bacterium caprae LNIV 17320 (4); and

0.9

Mycobacterium avium subsp. avium ATCC

0.85

25291 (5). (b) Nanoprobe aggregation as

0.8

1

2

3

4

5

6

7

8

3

4

5

6

7

8

3

4

5

6

7

8

measured by the ratio of aggregation (ratio

gMTub probe

of SPR intensities at 526 and 600 nm) for 1.2

the assay mixtures. The bars represent the

1.15

average of three independent measurendard

deviation.

The

dashed

line

represents the threshold of one considered for discrimination between positive a-

1.1 Abs526/600

ments and the error bars indicate the sta-

1.05 1 0.95 0.9

nd negative detection with M. tuberculosis

0.85

complex probe DNA (10 mg/L) after 15 -

0.8

min of incubation with MgCl2 (0.04 M); detection of M. tuberculosis probe DNA

1.2

MgCl2 (0.07 M); detection of M. bovis pro-

1.15

be DNA (10 mg/L) after 15 min of incuba-

(1); all strains of M. bovis (2); M. bovis BCG (3); all strains of M. caprae (4); M. avium (5); M. avium subsp. paratuberculosis (6); M. avium

subsp.

hominissuis

Corynebacterium striatum (8).

(7);

and

1.1 Abs 526/600

analysed were: all strains of M. tuberculosis

2

gMBov probe

(30 mg/L) after 15 min of incubation with

tion with MgCl2(0.07 M). The strains

1

1.05 1 0.95 0.9 0.85 0.8 1

2


1468

Clinical Microbiology and Infection, Volume 16 Number 9, September 2010

unit of DNA gyrase (topoisomerase II), an enzyme universally present in bacteria and essential for their replication. In the present study, we developed a two-step approach based on the PCR amplification of the gyrB locus and subsequent hybridization with a species-specific gold nanoprobe for the unequivocal species identification via a simple and fast colorimetric assay capable of being introduced into routine microbiology laboratory practice (see Supporting information, Fig. S1) . Specific DNA detection based on Au-nanoprobes has proven to be a fast and simple alternative to more cumbersome techniques involving multistep assays. The non-cross-linking method for DNA identification has already been successfully applied for the identification of M. tuberculosis in clinical samples based on the rpoB locus [17]. The synthesized Au-nanoprobes were evaluated in terms of their capability to identify the respective complementary sequence. The minimum amount of salt (MgCl2) required to cause full aggregation for each nanoprobe was determined (data not shown), and these conditions were used for the detection of targets. The solutions exhibit a strong red colour derived from the localized SPR of the gold nanoparticles. In the presence of the respective complementary synthesized target, the Au-nanoprobes were protected from aggregation and the solution remained red, whereas the presence of noncomplementary target did not protect the Au-nanoprobes from aggregation and the solution turned blue (Fig. 2a). Each probe was used in a minimum of three individual parallel hybridization experiments with the PCR-amplified gyrB gene from each strain. Based on the UV-visible spectra obtained after inducing aggregation, the ratio between the absorbance at 526 nm (peak of the SPR, representing the contribution of the non-aggregated fraction of Au-nanoprobes) and absorbance at 600 nm (contribution from the aggregated fraction of Au-nanoprobes) was calculated. A ratio of one may be considered as the point of equilibrium between non-aggregated and aggregated nanoprobe, and hence the threshold by which to respectively consider the positive and negative discrimination of sequences. The data show that the MTC probe gave a clear positive identification for members of the MTC used in the assay, at the same time as clearly recognizing M. avium complex species and C. striatum as nonmembers (Fig. 2b; Table 1). The M. tuberculosis probe unequivocally identified the respective target, as did the M. bovis probe (Fig. 2b; Table 1). It should be noted that the genomic sequences reveal the difference of only a single nucleotide at the 3¢ end of the probe sequence but are still perfectly capable of clear identification of the respective targets. To evaluate the performance of the method, several reference strains and clinical isolates including mycobacterial

CMI

and nonmycobacterial species were tested and successfully identified (Table 1). The variation between individual experiments showed very small inter-assay deviation (Fig. 2b). This demonstrates the extreme sensitivity of the system and its great potential for use in assays for identification of very closely-related species. The Au-nanoprobe system that we propose is easy to perform without the need for expensive and complex laboratory set up. After PCR amplification, the Aunanoprobe system takes only 15 min to yield a colorimetric result with high specificity and sensitivity [13,17–19,24], and, using a microplate reader, is capable of considerable scalingup, making it suitable for high-throughput analysis. Indeed, using the photometric function of the microplate reader, each sample needs only 1 s to be assayed and the ratio calculated (data not shown). The possibility of use via a portable assay platform is being tested to render this system suitable for in-the-field screening [16]. Future studies will be carried out to optimize the methodology for application in clinical samples, and to extend the range of application to other members of the MTC. The worldwide incidence of tuberculosis is still not completely determined, in particular the incidence of disease as a result of M. bovis and other non-M. tuberculosis members of the MTC. A full understanding of the spectrum of disease caused by MTC members requires the molecular evaluation of mycobacterial isolates to the species level, and assays such as the one described in the present study may provide the additional screening potential required for this task.

Acknowledgements A. Figueiredo and J. Moita (INRB, I.P., LNIV) are acknowledged for providing for the Corynebacterium striatum strain.

Transparency Declaration This work was supported by grants FCT/MCTES (CIGMH); PTDC/BIO/66514/2006; PTDC/SAU-BEB/66511/2006 and PTDC/BIA-MIC/71280/2006 and by INRB, I.P., LNIV. The authors declare that there are no dual/conflicting interests.

Supporting Information Additional Supporting Information may be found in the online version of this article:


CMI


Figure S1. (a) Schematic representation of DNA target detection with gold nanoprobes. The red curve shows the original Au nanoparticles spectra, indicating the strong red colour. Table S1. Blind identification of clinical isolates using the gold nanoprobe assay. Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

References 1. Etter E, Donado P, Jori F, Caron A, Goutard F, Roger F. Risk analysis and bovine tuberculosis, a re-emerging zoonosis. Ann NY Acad Sci 2006; 1081: 61–73. 2. Smith NH, Gordon SV, de la Rua-Domenech R, Clifton-Hadley RS, Hewinson RG. Bottlenecks and broomsticks: the molecular evolution of Mycobacterium bovis. Nat Rev Microbiol 2006; 4: 670– 681. 3. De la Rua-Domenech R. Human Mycobacterium bovis infection in the United Kingdom: incidence, risks, control measures and review of the zoonotic aspects of bovine tuberculosis. Tuberculosis 2006; 86: 77–109. 4. Cosivi O, Grange JM, Daborn CJ et al. Zoonotic tuberculosis due to Mycobacterium bovis in developing countries. Emerg Infect Dis 1998; 4: 59–70. 5. O’Reilly LM, Daborn CJ. The epidemiology of Mycobacterium bovis infections in animals and man: a review. Int J Tuberc Lung Dis 1995; 76 (suppl 1): 1–46. 6. Cheng VC, Yew WW, Yuen KY. Molecular diagnostics in tuberculosis. Eur J Clin Microbiol Infect Dis 2005; 24: 711–720. 7. Drobniewski F, Caws M, Gibson A, Young D. Modern laboratory diagnosis of tuberculosis. Lancet Infect Dis 2003; 3: 141–147. 8. Duarte EL, Domingos M, Amado A, Botelho A. Spoligotype diversity of Mycobacterium bovis and Mycobacterium caprae animal isolates. Vet Microbiol 2008; 130: 415–421. 9. Huard RC, Lazzarini LCO, Butler WR, Van Soolingen D, Ho JL. PCRbased method to differentiate the subspecies of the Mycobacterium tuberculosis complex on the basis of genomic deletions. J Clin Microbiol 2003; 41: 1637–1650.

1469

10. Mishra A, Singhal A, Chauhan DS et al. Direct detection and identification of Mycobacterium tuberculosis and Mycobacterium bovis in bovine samples by a novel nested PCR assay: correlation with conventional techniques. J Clin Microbiol 2005; 43: 5670–5678. 11. Kasai H, Ezaki T, Harayama S. Differentiation of phylogenetically related slowly growing mycobacteria by their gyrB sequences. J Clin Microbiol 2000; 38: 301–308. 12. Niemann S, Harmsen D, Rusch-Gerdes S, Richter E. Differentiation of clinical Mycobacterium tuberculosis complex isolates by gyrB DNA sequence polymorphism analysis. J Clin Microbiol 2000; 38: 3231–3234. 13. Baptista P, Pereira E, Eaton P et al. Gold nanoparticles for the development of clinical diagnosis methods. Anal Bioanal Chem 2008; 391: 943–950. 14. Eaton P, Doria G, Pereira E, Baptista P, Franco R. Imaging gold nanoparticles for DNA sequence recognition in biomedical applications. IEEE Trans Nanobioscience 2007; 6: 282–288. 15. Francisco J, Baptista P, Rueff J. Gold nanoparticle based systems in genetics. Current Pharmacogenomics 2007; 5: 39–47. 16. Silva LB, Baptista P, Raniero L, Doria G, Martins R, Fortunato E. Characterization of optoelectronic platform using an amorphous/ nanocrystalline silicon biosensor for the specific identification of nucleic acid sequences based on gold nanoparticle probes. Sens Actuators B Chem 2007; 132: 508–511. 17. Baptista P, Koziol-Montewka M, Paluch-Oles J, Doria G, Franco R. Gold-nanoparticle-probe-based assay for rapid and direct detection of Mycobacterium tuberculosis DNA in clinical samples. Clin Chem 2006; 52: 1433–1434. 18. Baptista P, Doria G, Henriques D, Pereira E, Franco R. Colorimetric detection of eukariotic gene expression with DNA-derivatized gold nanoparticles. J Biotechnol 2005; 119: 111–117. 19. Doria G, Franco R, Baptista P. Nanodiagnostics: fast colorimetric method for single nucleotide polymorphism/mutation detection. IEE Proc Nanobiotechnol 2007; 1: 53–57. 20. Amaro A, Duarte E, Amado A, Ferronha H, Botelho A. Comparison of three DNA extraction methods for Mycobacterium bovis, Mycobacterium tuberculosis and Mycobacterium avium subsp. avium. Lett Appl Microbiol 2008; 47: 8–11. 21. Larkin MA, Blackshields G, Brown NP et al. Clustal W and Clustal X version 2.0. Bioinformatics 2007; 23: 2947–2948. 22. Lee PC, Meisel D. Adsorption and surface-enhanced Raman of dyes on silver and gold sols. J Phys Chem 1982; 86: 3391–3395. 23. Warren RM, Gey Van Pittius NC, Barnard M et al. Differentiation of Mycobacterium tuberculosis complex by PCR amplification of genomic regions of difference. Int J Tuberc Lung Dis 2006; 10: 818–822. 24. Sato K, Hosokawa K, Maeda M. Rapid aggregation of gold nanoparticles induced by non-cross-linking DNA hybridization. J Am Chem Soc 2003; 125: 8102–8103.
