Fluorogenic assay and live cell imaging of HIV-1 protease activity using acid-stable quantum dot-peptide complex

Article abstract of DOI:10.1039/c0cc02702b

A novel QD-peptide complex for detecting HIV-1 protease activity was prepared from simple one step electrostatic interaction. Fluorescence recovery of the pre-quenched QD through fluorescence resonance energy transfer allowed for in vitro assay and live cell imaging of the protease activity in HIV-1 transfected cells, proving the potential for cell-based protease inhibitor screening.

Full text of DOI:10.1039/c0cc02702b

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Fluorogenic assay and live cell imaging of HIV-1 protease activity
using acid-stable quantum dot–peptide complexw
Youngseon Choi,a Junghan Lee,a Keumhyun Kim,a Heeyeon Kim,a Peter Sommerb and
Rita Song*a
Received 21st July 2010, Accepted 7th October 2010
DOI: 10.1039/c0cc02702b
A novel QD-peptide complex for detecting HIV-1 protease
activity was prepared from simple one step electrostatic
interaction. Fluorescence recovery of the pre-quenched QD
through fluorescence resonance energy transfer allowed for
in vitro assay and live cell imaging of the protease activity in
HIV-1 transfected cells, proving the potential for cell-based
protease inhibitor screening.
(pH 7–9) have been used in these enzymatic assays. The
negatively charged QDs are thus not fully applicable to the
family of acid proteases such as the HIV-1 PR.1c Furthermore,
recombinant HIV-1 PR has 106 times lower enzymatic activity
(1 pmol/unit) than most of the aforementioned proteases (e.g.
1 mmol/unit for collagenase).6 These challenges necessitate the
development of more sensitive and robust probe systems for
measuring HIV-1 PR activity in vitro as well as for live cell
imaging applications.
The human immunodeficiency virus protease (HIV-1 PR)
belongs to the family of aspartic proteases which are
characterized by an optimum pH in the acidic range (pH
4–5) due to a conserved region of two aspartates in an active
site.1a HIV-1 PR is essential for processing of newly
synthesized viral polyproteins to create the mature protein
components of an infectious HIV virion. Without effective
HIV-1 PR, viral particles remain uninfectious, which makes
this enzyme one of the key targets for anti-retroviral drug
discovery.1b HIV-1 PR activity has been routinely assayed by
the fluorescence resonance energy transfer (FRET)-based
fluorogenic method using organic fluorophore-labelled
peptide substrates.1c These assays, however, often suffer from
several shortcomings including the instability of the dual dye-
labelled peptides, high background, poor water solubility, and
low quenching efficiency resulting in low sensitivity.2
Here, we demonstrate a novel QD–substrate probe for
sensing HIV-1 PR activity through the fluorescence recovery
of FRET-quenched QD prepared by electrostatic assembly
with peptidic substrate (Scheme 1). The peptide sequence-
specific fluorogenic property of the QD probe was validated
in in vitro assays employing known protease inhibitors, which
allows for determining the inhibition concentration at 50%
(IC50) values for pepstatin A and the FDA (Food and Drug
Administration)-approved anti HIV-1 drug saquinavir (SQ).
Importantly, we could also visualize the effect of SQ on
protease activity in protease expressing live HeLa cells. These
results eventually demonstrated the potential of this QD probe
for the development and implementation of cell-based visual
assays for protease-targeted anti-retroviral drug discovery.
The QDs that are rendered stable in acidic pH were prepared
by exchanging trioctylphosphine oxide (TOPO) molecules on
QD (QD495-TOPO, emission peak at 495 nm; Evidots) with
DPA (DHLA-PEG-Amine) ligand according to the method
Recently, CdSe/ZnS semiconductor nanocrystal, known as
quantum dot (QD), has been widely used for in vitro diagnostics
as well as biological imaging due to its unique fluorescent
properties such as high photostability, broad excitation and
tunable narrow emission spectra.3 The QD, a novel type of
fluorophore, has also been employed as an efficient FRET
donor, which can be conjugated to acceptor dyes or gold
nanoparticle-labelled peptidic substrates for sensing the
activities of various enzymes such as caspases, thrombin,
chymotrypsin,4a,b b-lactamase,4c matrix metalloproteinase,4d
and collagenase.5a–c However, all of these enzymes require
neutral or basic pH ranges for their optimum activity.
Therefore, carboxylated QDs including QD-DHLA
(dihydrolipoic acid), polymer-coated QD, and QD-
streptavidin which are stable in neutral to basic pH ranges
À
À
À
we previously reported.7a,b After the DPA ligand exchange of
QD-TOPO, the QD becomes water-soluble. The excess DPA is
removed by extensive membrane filtration with molecular
weight cut-off (10 kDa) with deionized water. DPA-modifieÀd
+
QDs (1) possessing both ethylene glycol and NH3 ÁCF3CO2
moiety on the surface ligand allowed for the stabilization in
a Nano/Bio Chemistry Laboratory, Institut Pasteur Korea (IP-K), 696
Sampyeong-dong, Bundang-gu, Seongnam-Si, Gyeonggi-Do, 464-400,
South Korea. E-mail: rsong@ip-korea.org; Fax: +82-31-8018-8014;
Tel: +82-31-8018-8230
b Cell Biology of Retroviruses Laboratory, Institut Pasteur
Korea (IP-K), South Korea
w Electronic supplementary information (ESI) available: Detailed
information on the preparation of QD-DPA 1 and FRET study,
in vitro assay and in vivo cellular images. See DOI: 10.1039/c0cc02702b
Scheme 1 Scheme for HIV-1 protease assay based on FRET-
quenched QD–peptide complex. 6E-peptide-dabcyl indicates
6E-GLAib-SQNYPIVQ-K(dabcyl), or 6Glu-Gly-Leu-Aib-2Gly-Ser-
Gln-Asp-Tyr-Pro-Ile-Val-Lys(dabcyl); Aib = amino-isobutyric acid.
c
9146 Chem. Commun., 2010, 46, 9146–9148
This journal is The Royal Society of Chemistry 2010

View Article Online
acidic aqueous media (pH 3–6) and overall positive charges
(B66 amino groups per QD, zeta potential z = +30.6 Æ
5.8 mV). Hydrodynamic size of the QD-DPA was 9.2 Æ 1.4 nm
in diameter based on dynamic light scattering (DLS) analysis;
comparable with B3.5 nm from transmission electron
microscope (TEM) analysis (See the Supporting Information
(S.I.) for the detailed characterization of the QD-DPA).
In order to create a FRET-quenched QD probe specific for
HIV-1 PR, we designed a modular peptide (Pep1) consisting of
six glutamic acid residues (6E) at the N-terminal site, a linker
residue (Gly-Leu-Aib-2G), and HIV-1 PR recognition site
(SQNYPIVQ),8a followed by a lysine residue for labeling
with dabcyl quencher. Due to the presence of negatively
charged 6E residues, the Pep1 (z = À16 mV) immediately
forms an electrostatic complex with the positively charged QD
1 in deionized water, resulting in the DLS size (B12 nm). The
binding of the dabcyl-labeled peptide with the QD-DPA
surface resulted in FRET-induced quenching of the QD
Fig. 2 Determination of IC50 values of the HIV-1 PR inhibitors (left)
pepstatin A (2 mM) and (right) SQ (0.6 mM). 0% inhibition was defined
as the normalized PL recovery ratio of QD-Pep1 after HIV-1 PR
treatment in the absence of inhibitor drug (pepstatin A or SQ), while
100% inhibition was shown in the presence of the highest
concentration of inhibitor used (0.1 mM–1 mM). The data fitting
was performed using a non-linear log (inhibitor) vs. response mode in
GraphPad Prism software (ver 5.0).
buffer formulation (BSA 0–0.2 mg mLÀ1, NaCl 0–100 mM),
which are important factors for the reproducibility of the assay
(See the S.I. for the details). The sequence specificity of the QD
PL recovery was also confirmed with the control QD 1 which
was complexed with non-binding peptide Npep1 (6E-linker-Val-
Asn-Cys-Ala-Lys-Lys-Ile-Val-Lys(dabcyl)).8b The optimized
assay condition and the sensitivity of the QD-Pep1 probe
system allowed us to reproducibly measure the IC50 values of
the HIV-1 PR inhibitors, pepstatin A (2 mM) and SQ (0.6 mM)
(Fig. 2). IC50 values are varied depending on the assay system
used as reported in the literature, for example, 1 mM for
pepstatin A and 0.024–1.4 mM for SQ.1,9
To further test the feasibility of the QD-Pep1 probe for cell-
based drug screening, we attempted to visualize the HIV-1 PR
activity in living cells. To create a HIV-infected cellular model,
HeLa cells were transfected with a HIV-1 plasmid, pCMV-
dR8.74, which encodes for the gag and pol genes of HIV-1.10a,b
The gag genes encode the structural components of the viral
capsid and the pol genes the essential viral enzymes including
HIV-1 PR, respectively. The Gag proteins of HIV-1, like those
of other retroviruses, are necessary and sufficient for the
assembly of virus-like particles, which however remain non-
infectious due to the absence of viral envelope proteins and
genomes. To confirm that the HeLa cells were transfected with
HIV-1 plasmid, we performed the p24 ELISA assay which
identifies the HIV-1 specific antigens.10c The live HeLa cells
containing non-infectious HIV-1 were incubated with QD-
Pep1 probe (50 nM) for 3 h at 37 1C before confocal
microscope imaging. Fig. 3 shows the punctate QD signals
detected in the cytosol of the HIV-1 infected HeLa cells (See
the S.I. for the z-section images in 7 different focal middle
planes at 1 mm interval). These signals were not observed in the
(Forster radius R = 3.7 nm; See the S.I. for the details)
¨
o
(Fig. 1a). The quenching efficiency of the QD probe depends
on the molar ratio of Pep1 to QD 1, saturating FRET efficiency
(over 0.95) at above the molar ratio of 3 (Pep1/QD). To prove
that the quenching phenomenon is based on FRET rather than
non-specific aggregation, we prepared the QD-DPA with
longer emission peak (530 nm) for complexation with the
Pep1. Negligible quenching was observed when compared
with QD500-DPA, due to the lack of spectral overlapping of
QD emission (530 nm) and dabcyl absorption (lmax = 454 nm).
With this QD530-DPA, a similar quenching phenomenon was
observed when we replaced the dabycl with 6-carboxy
tetramethylrhodamine (TAMRA; lmax = 547 nm). These
results strongly suggest that electrostatic interaction between
QD-DPA and 6E-modified peptides resulted in the FRET-
induced quenching (See the S.I. for the details). This result also
shows the simplicity of the preparation for QD–substrate
complex 2 by mixing the two components in a similar
manner to a kit.
To test the fluorogenic properties of the QD-based probe in
the presence of HIV-1 PR, QD-Pep1 2 (0.18 mM) was
incubated with HIV-1 PR (0.35 mg) for 1 h at 37 1C in the
acidic buffer condition (10 mM acetate, 100 mM NaCl, 1 mM
DTT, BSA 10 mg mLÀ1, pH 4.5), resulting in B8-fold increase
of QD photoluminescence (PL) (Fig. 1b). The maximum fold-
increase of PL in the QD-Pep1 probe system was obtained after
optimizing the incubation temperature (25 1C, 37 1C) and
control naıve cells (not shown). Further, to confirm the
¨
sequence specificity of the QD probe in live cells, we also
incubated the QD-Npep1 with the HIV-1 infected cells at the
same conditions, which didn’t show any significant recovery of
QD signals inside the cells, confirming the negligible false-
positive signals.
Fig. 1 (Left) Normalized QD PL intensity at 500 nm in arbitrary units
(a.u.) depending on the molar ratio of Pep1 to QD 1 (black square) and
the corresponding FRET efficiency (blue circle); the red square
indicating QD PL in the presence of an equivalent amount of free
dabcyl dye. (Right) Increase of QD PL after HIV-1 PR digestion of the
QD-Pep1 complex 2 in the acetate buffer (pH 4.5).
With the successful demonstration of the QD-Pep1 probe
for HIV-1 PR activity in live cells, we performed cell-based
visual drug screening ([SQ] = 0.025–10 mM) on the HIV-1
infected HeLa cells. QD PL (green channel) was recovered at
0.025 mM of SQ treatment, indicating the low inhibitory
c
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 9146–9148 9147

View Article Online
(2009-0093618) and by the National R&D Program (No. 2010-
0019107). We are grateful to Dr. Jiyoung Kang, Hwapyung
Kim and Dr. Junwon Kim for fruitful discussions and
experimental advice.
Notes and references
1 (a) S. Seelmeier, H. Schmidt, V. Turk and K. V. D. Helm, Proc.
Natl. Acad. Sci. U. S. A., 1988, 85, 6612; (b) E. De Clercq, Nat. Rev.
Drug Discovery, 2002, 1(1), 13; (c) E. D. Matayoshi, G. T. Wang,
G. A. Kraft and J. Erickson, Science, 1989, 247, 954.
2 J. Zhang, R. E. Campbell, A. Y. Ting and R. Y. Tsien, Nat. Rev.
Mol. Cell Biol., 2002, 3, 906.
3 (a) P. Alivisatos, Nat. Biotechnol., 2004, 22, 47; (b) L. Medintz,
H. T. Uyeda, E. R. Goldman and H. Mattoussi, Nat. Mater., 2005,
4, 435; (c) X. Michalet, F. F. Pinaud, L. A. Bentolila, J. M. Tsay,
S. Doose, J. J. Li, G. Sundaresan, A. M. Wu, S. S. Gambhir and
S. Weiss, Science, 2005, 307, 538; (d) J. B. Delehanty,
K. Boeneman, C. E. Bradburne, K. Robertson and I. L. Medintz,
Expert Opin. Drug Delivery, 2009, 6, 1091.
Fig. 3 Representative confocal microscope images of HIV-1 plasmid
transfected HeLa cells incubated with QD-Pep1 probe (50 nM, 3 h),
which shows the recovered QD signal indicating HIV-1 protease
activity inside living cells. Differential interference contrast channel
(DIC) image, green QD channel (bandpass emission filter 500–525 nm)
with 488 nm laser excitation, and the merged image of the DIC and the
QD channel are shown with scale bar of 10 mm.
4 (a) K. Boeneman, B. C. Mei, A. M. Dennis, G. Bao,
J. R. Deschamps, H. Mattoussi and I. L. Medintz, J. Am. Chem.
Soc., 2009, 131(11), 3828; (b) I. L. Medintz, H. T. Uyeda,
E. R. Goldman and H. Mattoussi, Nat. Mater., 2006, 5(7), 581;
(c) C. J. Xu, B. G. Xing and H. H. Rao, Biochem. Biophys. Res.
Commun., 2006, 344(3), 931; (d) Z. Y. Xia, Y. Xing, M. K. So,
A. L. Koh, R. Sinclair and J. H. Rao, Anal. Chem., 2008, 80(22),
8649.
5 (a) E. Chang, J. S. Miller, J. T. Sun, W. W. Yu, V. L. Colvin,
R. Drezek and J. L. West, Biochem. Biophys. Res. Commun., 2005,
334(4), 1317; (b) L. F. Shi, V. De Paoli, N. Rosenzweig and
Z. Rosenzweig, J. Am. Chem. Soc., 2006, 128(32), 10378;
(c) L. F. Shi, N. Rosenzweig and Z. Rosenzweig, Anal. Chem.,
2007, 79(1), 208.
activity of SQ allowing HIV-1 PR activity. However, the QD
PL can not be recovered when HIV-1 PR activity is
significantly inhibited by the high concentration of SQ
(0.25–10 mM) (See the S.I. for the images). This result
indicates the potential of this probe for cell-based protease
inhibitor screening. The effective concentration of SQ was
found to vary according to the assay system used, for
example, 0.05–0.5 mM using HIV-1 infected macrophages.10a
In conclusion, these data showed the potential of the acid-
stable QD probe for sensing aspartic proteases such as HIV-1
PR which requires an acidic pH for optimum activity. The
simplicity in preparation (simple mixing), reproducibility due
to the photostability of QDs, and enzymatic sensitivity can
truly allow this QD-based probe to be applied to assays of
other aspartic proteases such as b-secretase or cathepsin D.
Moreover, the present approach provided the potential of QDs
to be applied as a cell-based visual screening platform for the
development of novel protease inhibitors as well as visual
target identification in living cells. Finally, the studies to
make this strategy universal for multiplexing and other
enzymatic assays are underway.
6 J. Schneider and S. B. H. Kent, Cell, 1988, 54, 363.
7 (a) J. Lee, Y. Choi, J. Kim, E. Park and R. Song, ChemPhysChem,
2009, 10(5), 806; (b) J. Lee, J. Kim, E. Park, S. Jo and R. Song,
Phys. Chem. Chem. Phys., 2008, 10(13), 1739.
8 (a) N. Kurt, T. Haliloglu and C. A. Schiffer, Biophys. J., 2003,
85(2), 853; (b) N. Ozer, T. Haliloglu and C. A. Schiffer, Proteins:
Struct., Funct., Bioinf., 2006, 64, 444; (c) Y. Wei, C.-M. Ma and
M. Hattori, Eur. J. Med. Chem., 2009, 44(10), 4112.
9 (a) C. F. Perno, F. M. Newcomb, D. A. Davis, S. Aquaro,
R. W. Humphrey, R. Calio and R. Yarchoan, J. Infect. Dis.,
1998, 178(2), 413; (b) R. J. Essey, B. R. McDougall and
W. E. Robinson, Antiviral Res., 2001, 51(3), 189; (c) D. Leung,
G. Abbenante and D. P. Fairlie, J. Med. Chem., 2000, 43(3), 305.
10 (a) D. Farson, R. Witt, R. McGuinness, T. Dull, M. Kelly,
J. P. Song, R. Radeke, A. Bukovsky, A. Consiglio and
L. Naldini, Hum. Gene Ther., 2001, 12(8), 981; (b) C. T. Wang,
J. J. Li, H. Y. Lai and B. S. Hu, J. Med. Virol., 1999, 57(1), 17;
(c) P. Nishanian, K. R. Huskins, S. Stehn, R. Detels and
J. L. Fahey, J. Infect. Dis., 1990, 162(1), 21.
This work was supported by the Korea Research
Foundation Grant funded by the Korean Government
(MEST) and Gyeonggi-do (K204EA000001-09E0100-00110),
and by the Korea Science and Engineering Foundation
(KOSEF) through its National Nuclear Technology Program
c
9148 Chem. Commun., 2010, 46, 9146–9148
This journal is The Royal Society of Chemistry 2010