Communication

pubs.acs.org/JACS

Liposome−Quantum Dot Complexes Enable Multiplexed Detection

of Attomolar DNAs without Target Amplification
Juan Zhou,†,§ Qiang-xin Wang,†,§,‡ and Chun-yang Zhang*,§
§
Single-Molecule Detection and Imaging Laboratory, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences,
Guangdong 518055, China
‡
Zhangjiagang Entry-Exit Inspection and Quarantine Bureau, Zhangjiagang 215600, China
*
S Supporting Information

combines single-molecule detection with the QDs13−16 has

See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
Downloaded via UNIV OF CALIFORNIA SAN DIEGO on October 13, 2020 at 17:12:38 (UTC).

ABSTRACT: Sensitive detection of DNA usually relies shown distinct advantages of a high signal-to-noise ratio,
on target amplification approaches such as polymerase improved sensitivity, low sample consumption, and near-zero
chain reaction and rolling circle amplification. Here we background signal in comparison with the conventional
describe a new approach for sensitive detection of low- ensemble fluorescence measurements. Single-particle detection
abundance DNA using liposome−quantum dot (QD) enables the detection of biomolecules at single-particle
complexes and single-particle detection techniques. This level,13−16 and its sensitivity can reach femtomolar.13a With
assay allows for detection of single-stranded DNA at the involvement of target amplification, its sensitivity can be
attomolar concentrations without the involvement of further improved to attomolar.14 However, the detection of
target amplification. Importantly, this strategy can be biomolecules with attomolar sensitivity without the involve-
employed for simultaneous detection of multiple DNA ment of target amplification has never been reported.
targets. In the conventional QD-based nanosensors, signal enhance-
ment is usually achieved by the assembly of multiple target
molecules on the surface of a single QD,13−16 thus the

S ensitive detection of DNA is of great importance in

biomedical research, clinical diagnosis, and gene expression
studies.1 Target amplification is usually employed to achieve
sensitivity is mainly limited by the availability of the amount of
both target molecules and the QDs. To break through the
bottleneck of signal enhancement, we have developed a new
high sensitivity.2−5 Among the techniques, polymerase chain method for sensitive detection of DNA using liposome−QD
reaction (PCR) is the most widely used technique for the (L/QD) complexes and single-particle detection techniques. As
amplified detection of DNA,2 but PCR involves both multiple shown in Scheme 1, a L/QD complex is composed of hundreds
primers and special DNA polymerases and requires high- of QDs. The presence of target DNA leads to the generation of
precision thermal cycling to separate the two DNA strands, a sandwich hybrid which consists of a L/QD complex-tagged
which limits its practical applications.2c Alternatively, several reporter probe, a magnetic bead-modified capture probe, and a
isothermal amplification techniques, such as rolling circle target DNA. The separation of sandwich hybrids from the free
amplification,3 strand displacement amplification,4 and loop- reporter probes by the magnetic beads and the subsequent
mediated isothermal amplification,5 have been developed. disruption of L/QD complexes result in the release of QDs,
These isothermal amplification methods proceed at a constant which can be sensitively counted by single-particle detection.
temperature and provide high amplification efficiency; however, In this research, liposomes are used to encapsulate the QDs
some special requirements such as the ligation of a padlock to form the L/QD complexes on the basis of two facts: (i)
probe,3 an initial heating denaturation step, and the use of liposomes are prominent cargo carriers which can engulf many
multiple primers5 and special DNA polymerases3−5 increase the hydrophobic nanoparticles due to their unique amphiphilic
experimental complexity and cost. Therefore, the development structures;17 (ii) the unique nature of liposomes such as
of new approaches for sensitive detection of DNA without the accessible functionality and desirable biocompatibility endows
involvement of target amplification is highly desirable. them with promising applications in the biomedical
Due to its remarkable advantages of high signal-to-noise researches.17 In addition, simultaneous detection of multiple
ratio, low sample consumption, and improved sensitivity,6 DNAs can be easily achieved by using two types of L/QD
single-molecule detection has become a promising approach in complexes with different colors.
the research of chemical analysis,7 molecular assembly,8 medical The high-quality green and red QDs were synthesized
diagnosis,9 and dynamic study of biological processes.10 according to a reported procedure.18 The as-obtained QDs
Organic fluorescent dyes are usually employed in single- were evaluated by transmission electron microscope (TEM),
molecule detection.6,7 Recently, owing to their unique optical UV−vis absorption, and steady-state fluorescence spectroscopy
properties of broad excitation, narrow emission, high quantum (see Supporting Information [SI], Figure S1). TEM images
yield, and photochemical stability,11 semiconductor quantum show that both the green and red QDs are highly
dots (QDs) have been widely used in place of organic
fluorescent dyes in biomedical research, biological labeling, and Received: November 9, 2012
in vitro/in vivo imaging.11,12 Single-particle detection which Published: January 30, 2013

© 2013 American Chemical Society 2056 dx.doi.org/10.1021/ja3110329 | J. Am. Chem. Soc. 2013, 135, 2056−2059
Journal of the American Chemical Society Communication

Scheme 1. Schematic Illustration for Sensitive Detection of

Attomolar DNA Using Liposome−QD Complexes and
Single-Particle Detection Techniquesa

a
This method involves three steps: (i) preparation of L/QD
complexes, L/QD complex-tagged reporter probes and magnetic
bead-modified capture probes; (ii) formation of sandwich hybrids in
the presence of target DNA and further purification through a magnet Figure 1. Characterization of two types of L/QD complexes with
separation; (iii) release of QDs from L/QD complex and subsequent different colors. Fluorescence imaging of (a) L/QD green complexes
measurement by single-particle detection. and (b) L/QD red complexes. Size distribution histogram of (c) L/
QD green complexes and (d) L/QD red complexes. (e) Normalized
fluorescence emission spectra of pristine green QDs (green line), L/
monodispersed and uniform in size. Measurement of QD green complexes (black line), pristine red QDs (red line) and L/
fluorescence spectra indicates that the emission peak is 537 QD red complexes (blue line).
nm for the green QDs and 612 nm for the red QDs. The
average size is estimated to be 2.8 ± 0.25 and 4.6 ± 0.34 nm for
the green and red QDs, respectively.19 complexes were characterized by TEM. The TEM images show
Two types of L/QD complexes with different colors were that the L/QD complexes are nearly spherical with the
prepared based on a reported procedure with some encapsulation of hundreds of QDs (see SI, Figure S3). On
midfications.20 Fluorescence images show that both L/QD the basis of the three-dimensional model with the encapsulation
green complexes and L/QD red complexes are spherical in of QDs in the lipid interior of a liposome bilayer and the
shape, remarkably bright, and uniform in size (a and b of Figure calculation using the data obtained experimentally, it was found
1). Size, polydispersity index, and surface charge of L/QD that each liposome can encapsulate either ∼1063 green QDs or
complexes were measured by a Zetasizer Nano-ZS. The L/QD ∼648 red QDs (see SI, Figure S4).
complexes are uniform with a polydispersity index of 0.274 ± For sensitive detection of target DNA, a typical sandwich
0.036 for L/QD green complexes and 0.236 ± 0.016 for L/QD format was constructed. The carboxyl-functionalized L/QD
red complexes (see SI, Figure S2 and Table S1). Analysis of size complexes and carboxyl-modified magnetic beads were
distributions reveals the average size of 82 ± 3.8 nm for the L/ covalently conjugated with the amino-terminated olignucleo-
QD green complexes (Figure 1c) and 90 ± 3.5 nm for the L/ tides,22 producing the reporter probes and the capture probes,
QD red complexes (Figure 1d), much larger than that of 2.8 ± respectively (see details in SI). As a proof of concept, one target
0.25 nm for pristine green QDs and 4.6 ± 0.34 nm for pristine olignucleotide of HIV-1 was sandwich hybridized with a L/QD
red QDs (see SI, Figure S1), suggesting the successful green complex-tagged reporter probe 1 and a magnetic bead-
encapsulation of QDs inside the liposomes. Zeta potential is modified capture probe 1 on the basis of Watson−Crick base
measured to be −32.7 mV for L/QD green complexes and pairing. Another target olignucleotide of HIV-2 was sandwich
−37.2 mV for L/QD red complexes, indicating that the L/QD hybridized with a L/QD red complex-tagged reporter probe 2
complexes are highly dispersible in aqueous solution. The and magnetic bead-modified capture probe 2. The use of two
narrow and symmetrical fluorescence spectra (Figure 1e) types of L/QD complexes with different colors made it possible
further confirm the excellent optical behavior of two types of L/ for simultaneous detection of HIV-1 and HIV-2. Finally, the
QD complexes with different colors, with a red-shift of 5−7 nm separation and purification of sandwich hybrids from the free
in the emission peak as compared with the pristine QDs due to reporter probes was realized using the magnetic beads and an
the interaction of QDs with the lipid layer.21 In addition, L/QD external magnetic field.
2057 dx.doi.org/10.1021/ja3110329 | J. Am. Chem. Soc. 2013, 135, 2056−2059
Journal of the American Chemical Society Communication

In contrast to single QD-based nanosensors,13−16 the

disruption of liposomes and the subsequent counting of
released single QDs are the keys to the improvement of
detection sensitivity in this research. A variety of organic
solvents including methanol, ethanol, 1-propanol, 1-butanol,
and surfactant of triton X-100 were examined for their
capability to disrupt the liposomes.23 Although the liposomes
could be disrupted by these solvents, the fluorescence of QDs
decreased greatly. Chloroform proved to be the most efficient
solvent with the capability of disrupting the liposome and, at
the same time, maintaining the strong fluorescence of single
QDs. The selection of chloroform might be attributed to two
facts: (1) DSPC, DSPE−PEG−COOH, and cholesterol which
form the liposome are easy to dissolve in chloroform;17b,20
consequently, the vesicle structure of liposome cannot be
preserved in chloroform; (2) the encapsulated QDs are prone
to escape upon the disruption of liposome and redisperse in
chloroform due to the presence of hydrophobic ligands on the Figure 3. Simultaneous detection of multiple DNAs. (a−c)
surface of QDs. To characterize the released QDs from the Representative trace of fluorescence bursts from the released QDs in
liposomes, particle size analysis, steady-state fluorescence and the presence of (a) HIV-1, (b) HIV-2, and (c) both HIV-1 and HIV-2.
UV−Vis spectra were investigated. As shown in Figure 2a, the The concentration of HIV-1 and HIV-2 is 5 fM. (d) Variance of burst
counts from the released QDs as a function of the concentrations of
HIV-1 (green) and HIV-2 (red). No change in the burst counts is
observed in the control groups with noncomplementary DNA (black
and blue). (e) Simultaneous detection of HIV-1 (green) and HIV-2
(red). The concentrations of HIV-1 and HIV-2 are each 0.1 fM. The
concentration of L/QD complex-tagged reporter probes is 1.2 μM,
and the concentration of magnetic bead-tagged capture probes is 1.2
μM. Error bars show the standard deviation of three experiments.

released red QDs, respectively. In the presence of HIV-1, only

the fluorescence bursts from the green QDs are observed, but
no fluorescence burst from the red QDs is detected (Figure 3a).
While in the presence of HIV-2, only the fluorescence bursts
from the red QDs are observed, but no fluorescence burst from
the green QDs is detected (Figure 3b). These results indicate
the excellent specificity of the proposed method. In contrast,
the fluorescence bursts from both the green QDs and the red
QDs are observed simultaneously in the presence of both HIV-
1 and HIV-2 (Figure 3c). Thus, this method can be used for
simultaneous detection of both HIV-1 and HIV-2. Moreover,
Figure 2. Release of green and red QDs from the L/QD green
complexes and L/QD red complexes, respectively. (a) Fluorescence
the near-zero background noise warrants the ultrasensitive
imaging of L/QD complexes in PBS solution and the released QDs in detection of target DNA. As shown in Figure 3d, the burst
chloroform under a UV lamp with the excitation wavelength of 365 counts of green QDs increase with the target concentration for
nm. (b) Size distribution histograms of the released green QDs (left) HIV-1 detection, and the burst counts of red QDs increase with
and the released red QDs (right). the target concentration for HIV-2 detection. In contrast, in the
control group with noncomplementary DNA, no obvious
change is observed in the burst counts of either green QDs or
released single QDs display bright fluorescence under UV red QDs. The detection limit can reach 1 aM for HIV-1 and 2.5
excitation, indicating that the optical performance of QDs are aM for HIV-2. Notably, the detection sensitivity of the
well preserved during the incorporation and disruption proposed method has improved by as much as 5 orders of
processes. This is also supported by no obvious change in magnitude as compared with that of fluorescence-tagged
either the fluorescence emission spectra or the maxima microbead-based nanosensors,24 and 3 orders of magnitude as
absorption position of UV−vis spectra between pristine QDs compared with that of single-QD-based nanosensors.13a It
and the released single QDs (see SI, Figure S5). In addition, should be noted that one bead/QD might correspond to
analysis of size distributions reveals the average size of 2.5 ± multiple target DNAs in the fluorescence-tagged microbead-
0.28 nm for the released green QDs and 4.3 ± 0.45 nm for the based biosensors24 and single QD-based nanosensors.13−16 In
released red QDs (Figure 2b), consistent with that of 2.8 ± contrast, the current approach makes one target DNA
0.25 nm for pristine green QDs and 4.6 ± 0.34 nm for pristine correspond to hundreds of QDs upon the release of single
red QDs, suggesting the successful release of single QDs from QDs from the L/QD complexes, thus significantly improving
the liposomes without the existence of QD aggregation. the detection sensitivity even without the involvement of target
The released QDs were further quantified via single-particle amplification.
detection. Panels a−c of Figure 3 show the representative trace To demonstrate the capability of multiplex detection, the
of fluorescence bursts from the released green QDs and the released green QDs and red QDs in the presence of HIV-1 and
2058 dx.doi.org/10.1021/ja3110329 | J. Am. Chem. Soc. 2013, 135, 2056−2059
Journal of the American Chemical Society Communication

HIV-2 were detected simultaneously by single-particle (4) (a) Nycz, C. M.; Dean, C. H.; Haaland, P. D.; Spargo, C. A.;
detection (see SI, Figure S6). As shown in Figure 3e, only Walker, G. T. Anal. Biochem. 1998, 259, 226−234. (b) Guo, Q. P.;
green-QD signals can be detected in the presence of HIV-1, Yang, X. H.; Wang, K.; Tan, W. H.; Li, W.; Tang, H. X.; Li, H. M.
and only red-QDs signals can be observed in the presence of Nucleic Acids Res. 2009, 37, e20.
(5) (a) Notomi, T.; Okayama, H.; Masubuchi, H.; Yonekawa, T.;
HIV-2. While in the presence of both HIV-1 and HIV-2, both Watanabe, K.; Amino, N.; Hase, T. Nucleic Acids Res. 2000, 28, e63.
green-QD and red-QD signals can be detected simultaneously. (b) Tomita, N.; Mori, Y.; Kanda, H.; Notomi, T. Nat. Protoc. 2008, 3,
These results clearly demonstrate that this method can be used 877−882.
for multiple DNA detection. (6) (a) Xue, Q. F.; Yeung, E. S. Science 1997, 275, 1106−1109.
In conclusion, we have developed a new approach for (b) Weiss, S. Science 1999, 283, 1676−1683.
sensitive detection of DNA using L/QD complexes and single- (7) (a) Stoffel, C. L.; Rowlen, K. L. Anal. Chem. 2005, 77, 2243−
particle detection techniques. The release of hundreds of single 2246. (b) Xue, Q. W.; Jiang, D. F.; Wang, L.; Jiang, W. Bioconjugate
QDs from the L/QD complexes and the subsequent counting Chem. 2010, 21, 1987−1993. (c) Knemeyer, J. P.; Herten, D. P.; Sauer,
at single-particle level are crucial to the improvement of M. Anal. Chem. 2003, 75, 2147−2153.
detection sensitivity. Without the involvement of target (8) Yildiz, A.; Forkey, J. N.; McKinney, S. A.; T. Ha, Y.; Goldman, E.;
Selvin, P. R. Science 2003, 300, 2061−2065.
amplification, the detection limit of the proposed method can
(9) Chiu, R. W. K.; Cantor, C. R.; Dennis, Lo, Y. M. Trends Genet.
reach attomolar, which has improved by as much as 5 orders of 2009, 25, 324−331.
magnitude as compared with that of fluorescence-tagged (10) Zhuang, X. W.; Bartley, L. E.; Babcock, H. P.; Russell, R.; Ha,
microbead-based nanosensors,24 and 3 orders of magnitude as T.; Herchlag, D.; Chu, S. Science 2000, 288, 2048−2051.
compared with that of single QD-based nanosensors.13a (11) (a) Alivisatos, A. P. Science 1996, 27, 933−937. (b) Chan, W. C.
Moreover, taking advantage of two types of L/QD complexes W.; Nie, S. M. Science 1998, 281, 2016−2018. (c) Peng, X. G.; Manna,
with different colors, simultaneous detection of two DNA L.; Yang, W. D.; Wickham, J.; Scher, E.; Kadavanich, A.; Alivisatos, A.
targets can be achieved. This method has significant advantages P. Nature 2000, 404, 59−61. (d) Genger, U. R.; Grabolle, Ml; Jaricot,
of high sensitivity, multiplex analysis, and low sample S. C.; Nitschke, R.; Nann, T. Nat. Methods 2008, 5, 763−775.
consumption and might be further extended to sensitive (e) Delehanty, J. B.; Bradburne, C. E.; Susumu, K.; Boeneman, K.;
Mei, B. C.; Farrell, D.; Blanco-Canosa, J. B.; Dawson, P. E.; Mattoussi,
detection of miRNAs and proteins in high-throughput H.; Medintz, I. L. J. Am. Chem. Soc. 2011, 133, 10482−10489.
screening and early clinical diagnosis.

■
(f) Algar, W. R.; Wegner, D.; Huston, A. L.; Blanco-Canosa, J. B.;
Stewart, M. H.; Armstrong, A.; Dawson, P. E.; Hildebrandt, N.;
ASSOCIATED CONTENT Medintz, I. L. J. Am. Chem. Soc. 2012, 134, 1876−1891.
*
S Supporting Information (12) (a) Somers, R. C.; Bawendi, M. G.; Nocera, D. G. Chem. Soc.
Rev. 2007, 36, 579−591. (b) Medintz, I. L.; Uyeda, H. T.; Goldman, E.
Experimental details and supplementary figures. This material is R.; Mattoussi, H. Nat. Mater. 2005, 4, 435−446. (c) Alivisatos, A. P.
available free of charge via the Internet at http://pubs.acs.org.

■
Nat. Biotechnol. 2004, 22, 47−52. (d) Wang, S.; Han, M. Y.; Huang, D.
J. Am. Chem. Soc. 2009, 131, 11692−11694. (d) Medintz, I. L.; Clapp,
AUTHOR INFORMATION A. R.; Mattoussi, H. Nat. Mater. 2003, 2, 630−638. (e) Medintz, I. L.;
Corresponding Author Mattoussi, H. Phys. Chem. Chem. Phys. 2009, 11, 17−45. (f) Mattoussi,
H.; Palui, G.; Na, H. B. Adv. Drug Delivery Rev. 2012, 64, 138−166.
zhangcy@siat.ac.cn
(13) (a) Zhang, C. Y.; Yeh, H. C.; Kuroki, M. T.; Wang, T. H. Nat.
Author Contributions Mater. 2005, 4, 826−831. (b) Zhang, C. Y.; Hu, J. Anal. Chem. 2010,
†
J.Z. and Q.-x.W. contributed equally. 82, 1921−1927.
(14) Zhang, Y.; Zhang, C. Y. Anal. Chem. 2012, 84, 224−231.
Notes
(15) (a) Scholl, B.; Liu, H. Y.; Long, B. R.; McCarty, O. J. T.;
The authors declare no competing financial interest.

■
O’Hare, T.; Druker, B. J.; Vu, T. Q. ACS Nano 2009, 3, 1318−1628.
(b) Long, Y.; Zhang, L. F.; Zhang, Y.; Zhang, C. Y. Anal. Chem. 2012,
ACKNOWLEDGMENTS 84, 8846−8852.
This work was supported by the National Basic Research (16) Zhang, C. Y.; Johnson, L. W. Anal. Chem. 2009, 81, 3051−3055.
(17) (a) Al-Jamal, W. T.; Kostarelos, K. Acc. Chem. Res. 2011, 44,
Program 973 (Grants 2011CB933600 and 2010CB732600),
1094−1104. (b) Mukthavaram, R.; Wrasidlo, W.; Hall, D.; Kesari, S.;
the National Natural Science Foundation of China (Grant Makale, M. Bioconjugate Chem. 2011, 22, 1638−1644.
21075129), the Award for the Hundred Talent Program of the (18) Qu, L. H.; Peng, X. G. J. Am. Chem. Soc. 2002, 124, 2049−2055.
Chinese Academy of Science, and the Fund for Shenzhen (19) Yu, W. W.; Qu, L. H.; Guo, W. Z.; Peng, X. G. Chem. Mater.
Engineering Laboratory of Single-molecule Detection and 2003, 15, 2854−2860.
Instrument Development (Grant No. (2012) 433). (20) Tian, B.; Al-Jamal, W. T.; Al-Jammal, K. T.; Kostarelos, K. Int. J.

■
Pharm. 2011, 416, 443−447.
REFERENCES (21) Pan, J.; Wan, D.; Gong, J. L. Chem. Commun. 2011, 47, 3442−
3444.
(1) (a) Kim, M. S.; Stybayeva, G.; Lee, J. Y.; Revzin, A.; Segal, D. J. (22) Hermanson, G. T. Bioconjugate Techniques; Acdemic Press: San
Nucleic Acids Res. 2011, 39, e29. (b) Zou, B. J.; Ma, Y. J.; Wu, H. P.; Diego, 1996; pp 173−176.
Zhou, G. H. Angew. Chem., Int. Ed. 2011, 50, 7395−7398. (23) (a) Egashira, N.; Morita, S.; Hifumi, E.; Mitoma, Y.; Uda, T.
(2) (a) Heid, C. A.; Stevens, J.; Livak, K. J.; Williams, P. M. Genome Anal. Chem. 2008, 80, 4020−4025. (b) Wang, H. Y.; Sun, D. Y.; Tan,
Res. 1996, 6, 986−994. (b) Heim, A.; Ebnet, C.; Harste, G.; Åkerblom, Z. A.; Gong, W.; Wang, L. Colloids Surf., B 2011, 84, 515−519.
P. P. J. Med. Virol. 2003, 70, 228−239. (c) Gill, P.; Ghaemi, A. (24) (a) Han, M. Y.; Gao, X. H.; Su, J. Z.; Nie, S. M. Nat. Biotechnol.
Nucleosides, Nucleotides Nucleic Acids 2008, 27, 224−243. 2001, 19, 631−635. (b) Ferguson, J. A.; Steemers, F. J.; Walt, D. R.
(3) (a) Lizardi, P. M.; Huang, X. H.; Zhu, Z. R.; Bray-Ward, P.; Anal. Chem. 2000, 72, 5518−5524.
Thomas, D. C.; Ward, D. C. Nat. Genet. 1998, 19, 225−232. (b) Zhao,
W. A.; Ali, M. M.; Brook, M. A.; Li, Y. F. Angew. Chem., Int. Ed. 2008,
47, 6330−6337. (c) Hu, J.; Zhang, C. Y. Anal. Chem. 2010, 82, 8991−
8997.

2059 dx.doi.org/10.1021/ja3110329 | J. Am. Chem. Soc. 2013, 135, 2056−2059