Patent Grant
8476083
The introduction of luminescent semiconductor nanocrystals or quantum dots (QDs) to biology has provided researchers with novel fluorescent tools for potentially achieving advances in imaging and sensing. See, for example, U.S. Patent Application Publication Nos. 2008/0087843 and 2011/0089241, each of which is incorporated herein by reference. In particular, QDs have been widely adopted as either donors or acceptors in Förster resonance energy transfer (FRET)-based assays and biosensors.
FRET has been used in a wide variety of applications, including: static distance measurements within or between (bio)molecules (i.e. FRET as a “spectroscopic ruler” between ˜1-10 nm); dynamic observation of changes in biomolecular conformation; diagnostic constructs that use changes between FRET “on” (high efficiency) and “off” (low efficiency) states to detect chemical/biological analytes; and light harvesting/photonic wires. In spectroscopic ruler contexts, FRET is generally implemented in its simplest configuration, which comprises a single donor luminophore and single acceptor chromophore. The FRET efficiency can be used to derive the donor-acceptor separation distance and is almost always measured on the basis of quenching of the donor luminescence intensity or decrease in the donor excited state lifetime. When the acceptor chromophore is also fluorescent, the ratio of acceptor and donor luminescence intensities can be a useful qualitative or quantitative measure. Diagnostic probes (for example, molecular beacons, Scorpion primers, or TaqMan probes) also predominately utilize discrete donor-acceptor pairs. Multi-step FRET relays have been described previously, using only prompt (nanosecond scale) fluorescence. The primary purpose has been to extend the net range of FRET or serve as a photonic wire.
Described herein is a time-gated, two-step FRET relay effective to provide temporal transference of a prompt FRET pathway, or provide spectro-temporal encoding.
In one embodiment, a FRET relay assembly includes a long lifetime FRET donor, a semiconductor quantum dot (QD) configured as an intermediate acceptor/donor in FRET, and a fluorescent dye configured as a terminal FRET acceptor, wherein the long lifetime FRET donor has an excited state lifetime of at least one microsecond and the QD and fluorescent dye each have excited state lifetimes of less than 100 nanoseconds.
Before describing the present invention in detail, it is to be understood that the terminology used in the specification is for the purpose of describing particular embodiments, and is not necessarily intended to be limiting. Although many methods, structures and materials similar, modified, or equivalent to those described herein can be used in the practice of the present invention without undue experimentation, the preferred methods, structures and materials are described herein. In describing and claiming the present invention, the following terminology will be used in accordance with the definitions set out below.
As used in this specification and the appended claims, the singular forms “a”, “an,” and “the” do not preclude plural referents, unless the content clearly dictates otherwise.
As used herein, the term “and/or” includes any and all combinations of one or more of the associated listed items.
As used herein, the term “about” when used in conjunction with a stated numerical value or range denotes somewhat more or somewhat less than the stated value or range, to within a range of ±10% of that stated.
FRET refers to Fluorescence Resonance Energy Transfer, also termed Förster Resonance Energy Transfer.
The term “quantum dot” or “QD” as used herein refers to an inorganic semiconductor crystallite of about 1 nm or more and about 1000 nm or less in diameter or any integer or fraction of an integer therebetween, preferably at least about 2 nm and about 50 nm or less in diameter or any integer or fraction of an integer therebetween, more preferably at least about 2 nm and about 20 nm or less in diameter (for example about 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 nm). QDs are characterized by their substantially uniform nanometer size, frequently exhibiting approximately a 10% to 15% polydispersion or range in size. A QD is capable of emitting electromagnetic radiation upon excitation (i.e., the QD is photoluminescent) and includes a “core” of one or more first semiconductor materials, and may be surrounded by a “shell” of a second semiconductor material. A QD core surrounded by a semiconductor shell is referred to as a “core/shell” QD. The surrounding “shell” material will preferably have a bandgap energy that is larger than the bandgap energy of the core material and may be chosen to have an atomic spacing close to that of the “core” substrate.
The core and/or the shell can be a semiconductor material including, but not limited to, those of the groups II-VI (ZnS, ZnSe, ZnTe, US, CdSe, CdTe, HgS, HgSe, HgTe, MgS, MgSe, MgTe, CaS, CaSe, CaTe, SrS, SrSe, SrTe, BaS, BaSe, BaTe, and the like) and III-V (GaN, GaP, GaAs, GaSb, InN, InP, InAs, InSb, and the like) and IV (Ge, Si, and the like) materials, PbS, PbSe, and an alloy or a mixture thereof. Preferred shell materials include ZnS.
A QD is optionally surrounded by a “coat” of an organic capping agent. The organic capping agent may be any number of materials, but has an affinity for the QD surface. The coat can be used to convey solubility, e.g., the ability to disperse a coated QD homogeneously into a chosen solvent, functionality, binding properties, or the like. In addition, the coat can be used to tailor the optical properties of the QD. In general, the capping agent can be an isolated organic molecule, a polymer (or a monomer for a polymerization reaction), an inorganic complex, or an extended crystalline or amorphous structure. The two most common strategies for imparting aqueous solubility are the use of bifunctional ligand coatings and bifunctional polymer coatings. The former coordinate to the QD surface through a chemical function (e.g., thiol, imidazole) and replace the native hydrophobic ligands; the latter have pendant alkyl chains that interdigitate with the native ligands via hydrophobic interactions. In both cases, a second and polar chemical function—such as a carboxylate, amine, or poly(ethylene glycol) (PEG) chain—mediates aqueous solubility. Mixtures of different ligands and the use of copolymers enable the modification of QDs with multiple functional groups. Ligand coatings are advantageous in that more compact QDs are obtained; however, polymer coated QDs tend to have superior brightness and photostability. Standard bioconjugate techniques, such as carbodiimide coupling, are widely used with both ligand and polymer stabilized QDs. Ligand coatings have also frequently enabled the preparation of bioconjugates through self-assembly driven by coordination to the QD surface. Quantum dots herein include those having a simple core with or without a coat, as well as optionally coated core/shell QDs.
As detailed herein, QDs can function in a simultaneous role as acceptors and donors within time-gated FRET relays. Aspects are described in Algar et al., J. Am. Chem. Soc. (2012) 134: 1876-1891, incorporated herein by reference along with its Supplemental Material.
The use of a two-step FRET relay serves to encode information having both a spectral and temporal dimensionality. This is achieved by assembling a configuration with an initial donor luminophore that has an excited state lifetime much longer (>10 μs) than both the intermediary acceptor/donor and terminal acceptor luminophores in the relay (<100 ns). Further, the two FRET steps are designed to be coupled but function approximately independently and resolvable from one another.
The technique described herein provides two novel attributes: (1) The FRET pathway from the QD to the fluorescent dye (termed FRET2), which would normally only be observable within a time-gated window ˜0-250 ns after optical excitation, can be shifted and extended into an observation window of at least 55-1055 μs, via the FRET pathway from the long lifetime FRET donor to the QD (termed FRET1). Significantly longer timescales should be achievable in view of the Tb excited state lifetime of greater than one millisecond. (2) Spectral measurement of the FRET relay configuration in both the 0-250 ns and 55-1055 μs observation windows after excitation enables resolution of the FRET1 and FRET2 processes.
In the embodiment illustrated schematically in
Within the exemplary FRET relay, the QD served as an intermediate acceptor/donor, where: (1) an excited-state Tb donor transferred energy to the ground-state QD following a suitable microsecond delay; and (2) the QD subsequently transferred that energy to an A647 acceptor. A photophysical analysis was undertaken for each step of the FRET relay. The assembly of increasing ratios of Tb per QD was found to linearly increase the magnitude of the FRET-sensitized time-gated QD photoluminescence intensity. The Tb was found to sensitize the subsequent QD-A647 donor-acceptor FRET-pair without significantly affecting the intrinsic energy transfer efficiency within the second step in the relay. The utility of incorporating QDs into this type of time-gated energy transfer configuration was demonstrated in prototypical bioassays for monitoring protease activity and nucleic acid hybridization; the latter included a dual target format where each orthogonal FRET step transduced a separate binding event. Potential benefits of this time-gated FRET approach include: eliminating background fluorescence, accessing two approximately independent FRET mechanisms in a single QD-bioconjugate, and multiplexed biosensing based on spectrotemporal resolution of QD-FRET without requiring multiple colors of QD.
The luminophores (FRET partners) used in this technique may be varied as desired. In addition to the commercially available Lumi4®-Tb (Lumiphore, Richmond, Calif., USA) used in the examples, other long lifetime FRET donors could be used including various luminescent lanthanide complexes (chelates or cryptates) based on Tb3+, Eu3+, Sm3+, Tm3+, and the like. Ruthenium complexes may also exhibit a sufficiently long excited state lifetime to serve as long lifetime FRET donors. Similarly, Cy5, Atto647, and many other fluorescent dyes or any other short-lifetime fluorophore (e.g. fluorescent proteins) may take the place of the Alexa Fluor 647 (Invitrogen by Life Technologies, Carlsbad, Calif., USA) employed in the examples. Different QD intermediates, including CdTe/ZnS and InP/ZnS QDs (etc.) can be used as well as the CdSe/ZnS QDs used in the examples (Invitrogen by Life Technologies).
Methodology
Reagents.
CdSe/ZnS QDs were obtained from Invitrogen by Life Technologies (Carlsbad, Calif.) and functionalized with dihydrolipoic acid-appended poly(ethylene glycol) (PEG, MW-750) ligands (see refs. 24 and 25). Peptides were synthesized as described in refs. 26 and 27 and labeled using A647 maleimide (Invitrogen) or Lumi4® Tb3+ N-hydroxysuccinimide (NHS) complex (Tb; Lumiphore, Richmond, Calif.) (see ref. 28). Probe and complementary target oligonucleotides were obtained from Integrated DNA Technologies (Coralville, Iowa). Targets were labeled with A647 or Tb; probes were modified to hexahistidine (His6)-DNA-peptide chimeras as described in ref. 27. Peptide and oligonucleotide sequences are given in Table 1, and the labeling chemistry is shown in
1Peptide labeled at the cysteine (thiol).
2Peptide labeled at the N-terminus (1° amine), oligonucleotide labeled at a 3′-amino linker.
3Peptide-oligonucleotide chimeras are linked by a disulfide bridge between the peptide cysteine residue and 5′-thiol linker on the oligonucleotide.
Instrumentation and Photoluminescence (PL) Measurements.
PL spectra were acquired using a Tecan Infinite M1000 Dual Monochromator Multifunction Plate Reader equipped with a xenon flash lamp (Tecan, Research Triangle Park, N.C.). Non-gated PL emission spectra: 400 Hz flash frequency, 400 nm excitation, ˜0 μs delay between flash and data acquisition, and a 40 μs integration time. Time-gated PL emission spectra: 100 Hz flash frequency, 339 nm excitation, 55 μs delay, 1 ms integration time. Tb/QD PL decay time measurements were acquired using three different systems (PLD systems 1-3), each with optimized capabilities for different decay.
PLD system 1 measurements were made using a customized spectrometer system with a 355 nm Nd:YAG laser source (60 Hz, 5 ns pulse width; 1 μJ per pulse) and monochromator for emission wavelength selection, analog PMT signals were processed using a digital oscilloscope and the time resolution was 40 μs, and a 55 μs time-gate was used.
PLD system 2 measurements were made using a modified KRYPTOR plate reader (Cezanne, Nimes, France) system with 4000 detection bins at 2 μs integration steps over 8 ms. A nitrogen laser was used for excitation (337.1 nm, 20 Hz, ˜100 μJ per pulse). To detect Tb PL, a 494±20 nm bandpass filter was used; to detect Tb-sensitized QD PL, a 640±14 nm bandpass filter was used.
PLD system 3 measurements used a PicoQuant FT 300 fluorescence spectrometer, with samples at 40 nM concentration in borate buffer, pH 8.5, with a 1 MHz repetition rate and emission collected at 620±5 nm. The QD PL decay lifetimes were measured with excitation at both 330 nm and 405 nm (which are close to the wavelengths used for excitation in time-gated and non-gated PL spectrum measurements). The values reported are the intensity averaged PL lifetime from a triexponential fit.
QD Bioconjugates and Assays.
A647/Tb labeled peptides/oligonucleotides were conjugated to QDs via polyhistidine self-assembly by mixing at the desired stoichiometric ratios, in buffer, for 30-60 min. No purification was necessary, and the well characterized, high-affinity binding resulted in nearly quantitative assembly to the QDs (see refs. 22 and 29-31). Quantitation using the Tb/A647 absorption of the labeled peptides provided knowledge of the number of donors/acceptors per QD. For characterization experiments, the QD conjugate concentration was 45 nM (5 pmol). Time-gated proteolytic assays were done by preparing (PEP B-Tb)10-QD-(PEP A-A647)3 conjugates, adding trypsin, and tracking time-gated PL at 625 and 675 nm over 1.5 h at 2 min intervals. The final QD conjugate concentration was 0.2 μM (20 pmol). Time-gated hybridization assays were done by mixing TGT A-A647 (0-50 pmol) with 60 pmol of PRB A, hybridized 60 min, (PEP B-Tb)10-QD conjugates added, and time-gated PL spectrum measured after 60 min. The final QD conjugate concentration was 45 nM (5 pmol). Two-plex hybridization assays were done similarly, except that TGT A (0-50 pmol) and TGT B (0-80 pmol) were mixed with 50 pmol of PRB A and 80 pmol of PRB B prior to addition of unconjugated QDs. Both non-gated and time-gated PL spectra were measured.
Results
Peptides and peptide-oligonucleotide chimeras used were engineered to display terminal His6 metal-affinity sequences to provide spontaneous self-assembly to the Zn2+-rich QD surface. As noted in refs. 22, 29, and 30, this motif has been used to prepare QD bioconjugates of proteins, peptides, and oligonucleotides with excellent control over the conjugate valence. Characterization of His6 self-assembly to QDs confirmed that an average of 50±10 peptides can be assembled around a ˜6 nm diameter DHLA-capped QD. The 625 nm emitting QDs used here were coated with a similar ligand but were ˜10±1 nm in diameter, suggesting access to an even wider range of conjugate valences. Since these experiments required the assembly of peptides and oligonucleotides, the latter were chemically ligated with a His6-appended peptide as described in ref. 29 (
Donor-Acceptor Pairs and Spectral Overlap.
Red-emitting 625 nm PL QDs (27 nm full-width-at-half-maximum) were paired as both an acceptor for the initial Tb donor, and a subsequent donor for the A647 acceptor.
aExtinction coefficient at peak donor PL emission wavelength, λD.
bExtinction coefficient at first exciton peak.
cSource: Invitrogen by Life Technologies.
dSource: ref.55
eWritten as donor to acceptor.
fGeometric prediction based on QD and peptide dimensions.
gTb → QD measured from Tb PL decay quenching.
hQD → A647 measured from QD PL quenching following direct QD excitation.
iQD → A647 measured from non-gated QD PL quenching following direct QD excitation (PEP B-Tb also present on QD).
jQD → A647 measured from time-gated QD PL quenching following FRET1 sensitization.
The large Förster distance calculated for Tb-to-QD energy transfer (10.1 nm) was a product of the extremely strong absorption of the QD across the emission range of the first three Tb lines. The Förster distance of the QD-A647 FRET pair was 7.5 nm, and this value is among the largest noted when pairing a QD donor with a dye acceptor (typically, R0<6 nm). Examining the Tb-QD-A647 three-luminophore system a priori confirmed the potential for a multistep FRET1+FRET2 relay process wherein excited-state Tb can transfer energy to the QD (acceptor), which subsequently acts as a donor for the A647, resulting in a net energy transfer from the Tb to the A647. The putative Tb-A647 FRET pair had significant spectral overlap and a Förster distance of 5.7 nm (J=2.5×10−10 mol−1 cm6); however, the FRET1 and FRET2 pathways should be more favored since, based on the relative Förster distances, their rates are expected to be 30-fold and 5-fold faster than Tb-to-A647 energy transfer, respectively.
Intensity-Based Analysis of Tb-to-QD Energy Transfer (FRET1).
Initial experiments focused on determining the degree to which the Tb could sensitize time-gated QD PL via FRET1. Increasing ratios of Tb-labeled PEP B (PEP B-Tb) were assembled on the 625 nm QDs and the resulting PL spectra were collected in both non-gated (˜0 μs delay) and time-gated modes (see Materials and Methods section for exact definitions). Time-gating for the measurements was empirically selected to be 55 μs, which corresponded to the minimum delay after flash excitation needed to minimize signal from direct excitation of the QDs. The residual signal was due to an instrumental/electronic echo effect rather than residual PL, as the QD excited-state completely decayed in less than a microsecond. An integration time of 1 ms was selected for time-gated measurements to be commensurate with the typical excited-state lifetimes of Tb complexes (see ref. 33). Without time-gating, the Tb sensitization of QD PL could not be observed over the directly excited QD PL.
As shown in
The Tb functioned as an effective FRET donor for the QD irrespective of whether it was directly labeled onto a peptide terminus (PEP B-Tb) or indirectly through oligonucleotide hybridization (PRB B/TGT B-Tb). The slightly lower rate of QD sensitization or FRET efficiency for the latter is believed to arise from a slightly longer Tb-QD separation in the DNA incorporating configuration. The Tb was attached at the end of a six-carbon aminated linker, which can allow some freedom of movement, in addition to breathing of the oligonucleotide hybrids. Moreover, previous results suggest that dye labels assembled onto QDs using similar peptide-DNA chimeras can have a wide range of movement relative to the QD (see ref. 34). Regardless, assembling greater numbers of Tb around the central QD increased the rate of energy transfer from Tb donors to QD acceptors, and this effect was observed as increases in the time-gated QD PL sensitization.
To further assess FRET1, the excitation spectra of (PEP B-Tb)10-QD conjugates were measured with and without time-gating, as shown in
PL Decay Analysis of Tb-to-QD Energy Transfer (FRET1).
Further characterization and confirmation of Tb-to-QD energy transfer was obtained using PL decay time analyses. Measurements of the Tb PL lifetimes were first collected, using PLD system 1, as the number of PEP B-Tb assembled per QD was increased. As shown in
The appearance of the very fast decay component was consistent with very efficient energy transfer from the Tb to QD, as expected based on the large R0 of 10.1 nm for this FRET pair. It was estimated that the PEP B-Tb places the Tb≦1.2 nm from the QD surface and ˜6.7 nm from the QD center. This value was arrived at by considering: (1) a negligible contribution from the His6-terminus which is in direct contact with the ZnS shell, (2) the Ala3 tract forming a helix that is disrupted by the flanking glycine residues, (3) rotational flexibility in the peptide, and (4) comparison to donor-acceptor distances for similarly sized peptides determined previously (see ref. 38). The QD radius was estimated to be 5.5 nm. As the Tb donor-QD acceptor separation (r≈6.7 nm) was much shorter than the Förster distance (10.1 nm), a FRET efficiency exceeding 92% was expected. The short Tb lifetime component was between ca. 20-200 μs and suggested FRET1 efficiencies of 93-99%. However, these short lifetimes were comparable to the temporal resolution (40 μs) of PLD system 1 measurements; further experiments were done at higher resolution (2 μs bins) using PLD system 2. Representative data measured for different ratios of Tb per QD (490 nm emission line) are shown in
aFRET1 efficiency from fast Tb PL decay component.
bFRET1 efficiency from time-gated sensitized QD PL decay.
cQD in the absence of FRET; measured with PLD system 3 (see Supporting Information).
Due to electronic saturation of PLD system 2 at short time-scales, some residual fast decay component appeared in the PEP B-Tb decay curves (see
QD-to-A647 Energy Transfer (FRET2).
The second QD-to-A647 energy transfer (FRET2) step in the FRET relay was examined. Increasing ratios of A647-labeled PEP A (PEP A-A647) were self-assembled around the central QD, and the non-gated and time-gated PL emission spectra measured. As shown in
Tb-to-QD-to-A647 Time-Gated FRET Relay (FRET1+FRET2).
For the next FRET characterization, PEP B-Tb and PEP A-A647 were co-assembled around the central QD to yield the final Tb-to-QD-to-A647 energy transfer relay. Time-gated PL measurements were again critical for observing Tb-sensitization of the QD during FRET1, and the subsequent energy transfer from the QD to the A647 in FRET2. To allow simple resolution of the effect of FRET1 on FRET2, the PEP B-Tb ratio was fixed at an intermediate value of 10 per QD. This valence corresponded to a significant rate of QD sensitization (see
To further establish the FRET1+FRET2 relay, excitation spectra were collected with PEP A-A647 and (PEP B-Tb)10-QD-(PEP A-A647)4 conjugates, as shown in
Proteolytic Assays.
Protease sensing using the FRET relay assembly was investigated in a time-gated, kinetic mode to examine possibilities of this technique in biosensing. This utilized PEP A-A647 and PEP B-Tb along with trypsin—a prototypical serine protease that cleaves on the C-terminal side of arginine and lysine residues. To enable sensing, PEP A incorporated one lysine (K) and two arginine (R) cleavage sites along its length (see ref. 40). In contrast, PEP B contained no lysine or arginine residues, and was therefore not a potential substrate for trypsin (confirmed experimentally; data not shown). The time-gated QD-FRET relay monitored trypsin activity by following the loss of FRET2 from proteolysis of PEP A-A647. Analogous to previous QD-FRET configurations for sensing proteolytic activity (see refs. 36, 38, and 40) the initial state of this time-gated configuration was “ON” with respect to the QD-A647 FRET2 pair, as illustrated in FIG. 1B(i). Proteolysis decreased the number of A647 proximal to the QD, progressively shifting the system toward a FRET2 “OFF” state with increasing activity, and thus provided a dynamic signal. In parallel, PEP B-Tb provided approximately constant time-gated sensitization of the QD by FRET1. (PEP B-Tb)10-QD-(PEP A-A647)3 conjugates were selected since 10 equivalents of PEP B-Tb provided significant time-gated QD sensitization, and 3 initial equivalents of PEP A-A647 afforded maximal changes in FRET efficiency during subsequent proteolysis. In contrast to previous configurations, time-gated sensing could be accomplished due to the FRET1 pathway. Another novel feature was the measurement of protease activity in a kinetic mode, where the course of proteolysis was followed in real-time using two-color ratiometric measurements.
As shown in
Time-Gated DNA Hybridization Assay.
A second sensing configuration explored using the FRET relay was a time-gated hybridization assay. To this end, QDs coassembled with PEP B-Tb and PRB A were employed. Initial measurements were made without time-gating by mixing QDs with 15 equivalents of PRB A that had been prehybridized with increasing amounts of TGT A-A647, as a QD-(PRB A)15/(TGT A-A647), configuration. As shown in FIG. 6B(i), the result was the expected rise approaching maximum FRET efficiency, and an approximately linear increase in the FRET-sensitized A647/QD PL ratio. The latter provided a more convenient (linear and no reference state needed) and sensitive capacity for quantitative detection. For the time-gated hybridization assay, (PEP B-Tb)10-QD conjugates were co-assembled with 12 equivalents of PRB A to detect an increasing quantity of TGT A-A647, as shown in FIG. 1B(iii). Analogous to the time-gated protease construct, the role of PEP B-Tb was to provide time-gated sensitization of the QD PL in the final (PEP B-Tb)10-QD-(PRB A)12/(TGT A-A647), configuration. The latter time-gated PL spectrum revealed the expected FRET “ON” progression as the amount of hybridized TGT A-A647 increased, indicated by decreases in QD PL and corresponding increases in sensitized A647 PL. Between the non-gated and time-gated formats, the FRET2 efficiency as a function of the number of equivalents of PEP A-A647 did not change (see
Orthogonal Two-Plex DNA Hybridization Assay.
Another sensing configuration focused on exploiting the approximately independent FRET1-FRET2 mechanisms for signal transduction in a multiplexed format. It is apparent that the two different energy pathways could be increasingly sensitized by the assembly of more Tb or A647 per QD. In contrast to previous QD-based biosensing formats (see, e.g., ref. 44), the technique described herein provides a route to multiplexed detection that does not derive its information from the use of multiple QD colors, but rather from the temporal resolution of the FRET1 and FRET2 processes. It was desired to demonstrate that each FRET process could reflect a distinct biorecognition event and provide an orthogonal analytical signal.
The magnitude of FRET1-sensitized time-gated QD PL is linearly proportional to the amount of proximal Tb (see
In conjunction with non-gated and time-gated measurements, a two-plex configuration was created by assembling (PRB B)16-QD-(PRB A)10 conjugates to respond to TGT B-Tb and TGT A-A647. The oligonucleotides were prehybridized with target, the QDs added, and self-assembly allowed to occur yielding (TGT B-Tb)n/(PRB B)16-QD-(PRB A)10/(TGT A-A647)m conjugates for detection. FIG. 1B(ii) illustrates the generic bioconjugate structure, while the schematics in
The orthogonality of the two-plex hybridization assay was evaluated using an array of different mixtures of TGT A-A647 and TGT B-Tb, followed by calculation of the non-gated A647/QD PL ratio (FRET2) and time-gated total QD+A647 PL sensitization (FRET1) from the measured PL spectrum of each mixture. As the amount of TGT A-A647 increased, the non-gated A647/QD PL ratio increased linearly, but was relatively unaffected by the presence or absence of TGT B-Tb (FIG. 7B(i)). Similarly, as the amount of TGT B-Tb increased, the total QD+A647 PL sensitization increased linearly, but was minimally affected by changes in the presence of TGT A-A647 (FIG. 7B(ii)). The data in FIG. 7B(i-ii) were collected simultaneously in the same experiment and from the same samples. These results effectively demonstrated orthogonal quantitative responses, where the A647/QD PL ratio responded to TGT A-A647, and the total QD+A647 PL sensitization responded to TGT B-Tb. For a rigorous demonstration of the concept and analysis, the small amount of overlapping Tb PL were numerically deconvolved from the QD PL in the measured two-plex PL spectra (alternative analyses are discussed later). The LOD for TGT A-A647 and TGT B-Tb were estimated to be 17 nM (1.9 pmol) and 29 nM (3.2 pmol), respectively. The LOD thresholds were taken as three standard deviations above the average A647/QD PL ratio (0 nM TGT A-A647, 0-727 nM TGT B-Tb, time-gated) or total QD+A647 PL sensitization (0 nM TGT B-Tb, 0-454 nM TGT A-A647, non-gated). Despite the two-plex format, these LODs compare favorably to the time-gated, one-plex hybridization assay.
Due to the emission rate of the long lifetime FRET donor, the FRET1 pathway is poorly observed promptly after interrogation (<100 ns), but readily observed over extended observation times (105-106 ns). This aids directly in distinguishing between the two FRET steps. The number and proximity of long lifetime FRET donors and fluorescent dyes assembled to the QD determines the rate of FRET1 and FRET2, respectively. Control over these parameters thus enables the modulation of FRET1 and FRET2, thereby permitting spectral-temporal encoding of information.
Applications
Potential applications areas utilizing luminescent probes or reporters in non-biological or biological applications requiring, or benefiting from, extending the observation window of FRET pair. Other applications include labeling, assays, or chemo/biosensing on the surface of or within cells, tissues (in vitro or in vivo), environmental samples, and/or other complex sample matrices prone to high levels of optical source scattering and/or autofluorescence that can be ameliorated by time-gating.
Potential applications also exist in areas where luminescent probes or reporters in biological applications requiring, or benefiting from, multiplexed detection in a spectro-temporal format. This includes: labeling, assays, or chemo/biosensing in vitro or within cells and tissues.
Further potential application areas are those where luminescence can be used for unique identification, tracking, or validation/authentication. This includes: optical barcodes for commercial/shipping use, and/or anti-counterfeit measures/forgery deterrent.
This technique may also be used advatageously in areas where traditional FRET has been employed, including measurements of binding and/or dissociation, enzymatic activity, protein folding, and cellular processes (such as endocytosis and protein synthesis).
Nanoparticles such as QDs also show great promise for a new generation of diagnostic probes, biomedical technologies, therapeutics, and single-vector combinations thereof, sometimes called “theranostics.”
Advantages
QDs as intermediate acceptors/donors in FRET. Several properties of QDs are uniquely advantageous for assembling the two-step energy transfer relay, particularly when the QD is configured as an intermediary. The strong, broad QD absorption was resonant with the three strongest Tb emission lines and was characterized by extinction coefficients approaching an order of magnitude greater than that of most fluorescent dye acceptors (see ref. 45). The use of a fluorescent dye intermediary in the FRET relay configuration would also be hindered by a narrower absorption band that would only be resonant with two of the Tb emission lines. As a consequence, the QD acceptor offered a much larger spectral overlap integral and Förster distance with the Tb. The QD PL was also well separated from the Tb PL (except for the Tb emission line at 620 nm, which was the least intense of all the lines). In turn, when the QD functioned as the donor for the final A647 acceptor in the relay, the narrow QD PL provided strong overlap with the A647 absorption, while its peak remained spectrally well-resolved from the A647 emission at 675 nm. The latter would not have been possible with a fluorescent dye intermediary, given the characteristic broad and red-tailed dye emission profile.
The non-trivial surface area of the QD was of almost equal importance to its optical properties, and its utilization as a central nanoscaffold greatly facilitated the physical assembly of the energy transfer relay. In contrast, the use of a fluorescent dye intermediary in the relay would require an extrinsic scaffold, such as a protein or double-stranded DNA, to provide the necessary proximity with the initial donor(s) and final acceptor(s). Such a scaffold would not readily provide the approximately centrosymmetric distribution of Tb and A647 that was achieved with the QD intermediary, and which enabled straightforward analysis. Further, since the QD was its own intrinsic scaffold, it enabled the use of peptides and nucleic acids in a biological rather than structural motif. A maximum of 20 total peptides (PEP A+PEP B) were assembled per QD here, which is less than half of the maximum packing expected for even smaller QDs. This “extra” surface availability could allow the assembly of one or more other biomaterials to the QD in order to provide utility beyond the FRET relay. For example, cell penetrating or other targeting peptides could be added to induce cellular uptake or in vivo targeting of the final bioconjugate as described in refs. 46-51.
The properties of His6 self-assembly for binding to the QD also provided advantages. Utilizing the His6 motif offered simplicity, efficiency, and reproducibility. QD bioconjugation required only mixing of the QD and the desired equivalents of biomolecule(s). This level of control over QD-bioconjugate valence effectively permitted tuning of the relative efficiency of the FRET1 and FRET2 pathways, regardless of the use of Tb/A647 labeled His6-appended peptides, or His6-appended oligonucleotide probes hybridized with labeled target. Such versatile and incremental tuning of two distinct FRET pathways—solely through assembly valence—may be unique to the use of QDs and His6-bioconjugation in these types of assemblies. In contrast, assembling biotinylated DNA to streptavidin-coated QDs results in a large amount of heterogeneity in separation distance, hindering control and characterization of FRET, as noted in ref. 34. Purely synthetic chemical approaches to deriving similar architectures based on, for example, a dendrimeric substrate, are extremely labor intensive, and neither spin coating nor layer-by-layer approaches would provide the same levels of control and/or precision as His6 (see refs. 17 and 52).
Time-Gated Biosensing.
With few exceptions, multistep FRET relays incorporating QD-bioconjugates have been based on QD-to-dye-to-dye energy transfer configurations (see refs. 20-22). The role of the relay in these instances was primarily to extend the range of energy transfer and/or to allow for wavelength shifting. Herein, the former consideration was addressed by the large Förster distances of the FRET1 (10.1 nm) and FRET2 (7.5 nm) pairs, while the latter was addressed by pairing a red-emitting QD with the deeper-red A647 acceptor. The unique features of the Tb-to-QD-to-A647 FRET relay were the added ability to make time-gated QD PL measurements, and sensitization of QD-to-A647 energy transfer over a millisecond timescale. With time-gating, undesirable direct excitation of the final A647 acceptor—regardless of the excitation wavelength used—was avoided due to its ˜1 ns fluorescence lifetime. However, a potentially more important advantage arises from the <20 ns characteristic decay times of cellular and tissue autofluorescence (see ref. 56), such that the time-gating afforded by Tb-to-QD FRET is expected to permit the separation of analytical PL signals from unwanted background PL in almost any complex biological matrix. While the protease sensing demonstrated in this work was primarily proof-of-concept for demonstrating utility of the FRET relay, it should be noted that abnormal protease activity is associated with many diseases, including ischemia, autoimmune and neurodegenerative disease, as well as several types of cancers (see ref. 57). This relevance suggests, for example, the possibility of sensitive, time-gated, in situ measurements of protease activity associated with over-expressed extracellular matrix metalloproteinases in complex tumor milieu. Use in complex biological matrices will be predicated on the fidelity of the QD-conjugate assemblies therein. Complexation of the Tb is highly stable and selective for lanthanide ions, polyhistidine is not an endogenous motif, and PEG coatings are largely biocompatible—all of which suggest that preassembled QD-conjugates should remain functional in biological matrices.
In addition to the advantages of time-gating via the Tb-QD FRET pair, there are advantages associated with the ratiometric detection afforded by the QD-A647 FRET pair. Ratiometric analyses tend to be relatively insensitive to dilutions and small variations in excitation intensity or between different instruments. This format is also particularly well suited to assays in a kinetic format, where donor and acceptor PL intensities dynamically change over extended time periods (e.g. hours) and are highly susceptible to instrumental drift and noise.
Orthogonal Spectrotemporal Multiplexing.
In previous FRET sensing configurations based on QD donors, multiplexed information has been encoded by using two different colors of QD with either a common or different FRET acceptors (refs. 60 and 35, respectively). Multiplexing configurations where QDs have been used as FRET acceptors also rely on multiple QD colors (refs. 11 and 14). In practice, ratiometric methods based on use of different fluorescent acceptor(s) have been preferred, and the number of colors measured (i.e. wavelength bands) was always greater than the number of targets. Moreover, multiplexing in these formats was usually limited to the ensemble since each individual QD-bioconjugate still only transduced one type of target. Compared to this state-of-the-art, the Tb-to-QD-to-A647 FRET relay offers an important and unique possibility in multiplexed bioanalysis: the ability to simultaneously transduce the activity of two different biomolecular targets by using a single color of QD in combination with orthogonal PL signals that are sensitive to different luminophores (the latter are bound to the QDs via biomolecular recognition). Furthermore, since these signals were proportional to the number of proximal acceptors, the viability of this assay format was demonstrated in a semi-quantitative, two-plex hybridization assay (
The use of full PL spectra and numerical deconvolution to resolve overlapping Tb and QD PL was performed for characterization and validation of the two-plex strategy, and is not generally expected to be required in the majority of practical applications of the FRET relay assembly. The small degree of Tb crosstalk in measuring QD PL only appeared in the time-gated measurement and systematically added to the apparent magnitude of the latter, such that it was readily calibrated into quantitative results. Alternatively, the use of 605 nm PL QDs, which would also be a suitable intermediary for the Tb and A647 in a relay, could potentially avoid this limited crosstalk since its PL maximum falls between the 585 and 620 nm Tb emission lines. In either case, the Tb PL intensity need not factor into the analysis, thereby permitting the use of two-color detection centered on the QD and A647 PL. Importantly, this two-color advantage does not come at the cost of ratiometric measurements. Since the A647 and QD PL manifests in both the non-gated and time-gated PL spectra, the FRET1 pathway can be measured relative to the directly excited FRET2 pathway by dividing the time-gated A647+QD PL sensitization (FRET1 signal) by the non-gated A647+QD PL, which is a reference state independent of FRET1. Thus, both FRET1 and FRET2 can be analyzed ratiometrically. The caveat of this FRET-relay strategy is that only static biological processes, or those with slow dynamics (≧10−2 s), can be monitored due to the need for microsecond time-gating and a millisecond integration time.
The orthogonal calibration curves in
The advantages of the techniques described herein include the following:
1. The use of a long-lifetime (microsecond to millisecond) initial donor for FRET1 enables time-gated observation, with good fidelity, of an (approximately) independent FRET2 process (intermediary-terminal acceptor pair), which would otherwise only be visible with prompt observation. Time-gated observation allows for the collection and accumulation of large amounts of data, with improved signal-to-noise with weak signals, and can be useful in situations when multiple events might contribute to the same signal.
2. The time-gated observation window that is suitable for the amelioration of scattered source light and sample/matrix autofluorescence such as that found in biological samples and in vivo cellular and tissue formats.
3. The FRET2 pair can be spectrally observed at their characteristic wavelengths in both prompt and time-gated observation after pulse/flash excitation.
4. The two-step FRET relay is a discrete entity (compared with multiple one-step FRET pairs).
5. Approximately independent tuning of FRET1 and FRET2 by controlling the number of initial donors and terminal acceptors, respectively, in association with the intermediary acceptor/donor.
6. Approximately independent tuning of FRET1 and FRET2 by controlling the proximity between the intermediary acceptor/donor and the initial donor(s) and terminal acceptor(s), respectively.
7. Three-fold flexibility in tuning the FRET1 and FRET2 efficiencies by the selection of the initial donor, intermediary acceptor/donor, and/or terminal acceptor on the basis of spectral overlap.
8. The two-step FRET relay requires less spectral bandwidth and/or channels to encode, for example, two-plex information when compared to two discrete FRET pairs.
9. Spectro-temporal information is more difficult to replicate than solely spectral information (e.g. as an authenticator/forgery deterrent).
10. The two-step FRET relay can potentially be used in solution, at an interface, within cellular environments, or embedded within matrices such as host nanoparticles or polymeric films/coatings.
11. Applicability and compatibility with both biological and non-biological environments.
12. Beyond the long lifetime FRET donor- or dye-labeled peptide and oligonucleotide assembly to QDs demonstrated herein, two-step FRET relays can be comprised of fluorescent dyes and long lifetime FRET donors labeled/conjugated to a variety of other biomolecules, organic, or inorganic scaffolds that assemble to, or incorporate, a suitable intermediary acceptor/donor.
All documents mentioned herein are hereby incorporated by reference for the purpose of disclosing and describing the particular materials and methodologies for which the document was cited.
Although the present invention has been described in connection with preferred embodiments thereof, it will be appreciated by those skilled in the art that additions, deletions, modifications, and substitutions not specifically described may be made without departing from the spirit and scope of the invention. Terminology used herein should not be construed as being “means-plus-function” language unless the term “means” is expressly used in association therewith
| Number | Name | Date | Kind |
|---|---|---|---|
| 20060148104 | Marini et al. | Jul 2006 | A1 |
| 20080087843 | Medintz et al. | Apr 2008 | A1 |
| 20080213780 | Butlin et al. | Sep 2008 | A1 |
| 20090162861 | Mathis et al. | Jun 2009 | A1 |
| 20110089241 | Medintz et al. | Apr 2011 | A1 |
