|Home | About | Journals | Submit | Contact Us | Français|
We sought to develop a novel competitive fluorescence resonance energy transfer (FRET)-aptamer-based strategy for detection of foot-and-mouth (FMD) disease within minutes. A 14-amino-acid peptide from the VP1 structural protein, which is conserved among 16 strains of O-serotype FMD virus, was synthesized and labeled with Black Hole Quencher-2 (BHQ-2) dye. Polyclonal FMD DNA aptamers were labeled with Alexa Fluor 546-14-dUTP by polymerase chain reaction and allowed to bind the BHQ-2-peptide conjugate. Following purification of the FRET–aptamer–peptide complex, a “lights off” response was observed within 10 minutes and was sensitive to a level of 25–250 ng/mL of FMD peptide. Ten candidate aptamers were sequenced from the polyclonal family. The aptamer candidates were screened in an enzyme-based plate assay. A high- and low-affinity aptamer candidate were each labeled with Alexa Fluor 546-14-dUTP by asymmetric polymerase chain reaction and used in the competitive FRET assay, but neither matched the sensitivity of the polyclonal FRET response, indicating the need for further screening of the aptamer library.
Although there has not been an outbreak of foot-and-mouth disease (FMD) in the United States since 1929 and all of North America is deemed FMD-free, the 2001 and 2007 outbreaks in the United Kingdom and other outbreaks in Asia and South America in recent years are poignant indicators of the need for early and rapid detection of FMD outbreaks.1–3 An agrobioterrorist attack employing FMD could have a devastating economic impact on the United States and other economies.2 While several rapid immunoassays4 and reverse transcriptase polymerase chain reactions (RT-PCR) or other nucleic acid amplification-based tests5–7 with broad FMD serotype specificity are marketed or in development, no portable field tests exist for all FMD serotypes and strains and “rapid” typically means at least 2 hours for results from the existing tests.5 Hence, we sought to develop a novel, competitive fluorescence resonance energy transfer (FRET)-based DNA aptamer test that might detect FMD within minutes by simply binding and increasing or decreasing fluorescence intensity. Since several companies are now marketing or developing portable or handheld fluorometers,8–10 we reasoned that lyophilized competitive FRET-aptamer assays might be valuable for rapid assessment and containment in the unfortunate event of an FMD outbreak.
Unlike signaling aptamers11 or aptamer beacons,12 our competitive FRET-aptamers attempt to place a known quencher–target conjugate within the Förster distance of fluorophores in the aptamer-binding pocket and then compete it off with unlabeled analyte. The approach is based on competitive immuno-FRET assays which one of the authors developed previously.13 We reasoned that such an approach might be highly sensitive if the fluorophore and quencher were spectrally matched to reduce background fluorescence. By this we mean that a large degree of overlap should exist between the fluorophore’s emission spectrum and quencher’s absorption spectrum. Therefore, we selected AF 546 and BHQ-2 as our fluorophore–quencher pair, because Alexa Fluor (AF; Invitrogen, Carlsbad, CA) 546 exhibits an emission peak around 573 nm, which is near the point where BHQ-2 absorbs maximally at 579 nm.
In the present work, we describe our preliminary attempts to study and characterize a polyclonal family of aptamers that bind an accessible FMD peptide (from a known antigenic region of the virus3) and individual candidate aptamer clones from the polyclonal family in competitive FRET assays. The data provide proof-of-principle for our approach and demonstrate that further screening of the aptamer library is warranted to ultimately enable development of an FMD FRET–aptamer assay for early detection of FMD outbreaks on farms or ranches. If successful, the competitive FRET-aptamer format could provide a platform for rapid and portable detection of many environmental analytes including other pathogens and pollutants.
For safety reasons we could not work with intact FMD virus. Instead, we chose a highly conserved peptide from the VP1 structural polypeptide of O-serotype FMD having the primary amino acid sequence RHKQKIVAPVKQLL.3 This sequence corresponds to amino acids 200 through 213 of 16 different O-serotype FMD viruses and represents a “neutralizable” antigenic region wherein antibodies are known to bind.3 High performance liquid chromatography (HPLC)-purified peptide was obtained from GenScript (Piscataway, NJ). BHQ-2-succinimide ester was obtained from Biosearch Technologies (Novato, CA) and dissolved in DMSO. Equimolar amounts of BHQ-2-succinimide ester and FMD peptide were mixed together in phosphate buffered saline (PBS; pH 7.2) and incubated for 2 hours at 37°C. The BHQ-2-labeled peptide was purified through Sephadex G25 and refrigerated until used for binding and complex formation with fluorophore-labeled aptamers.
FMD peptide was conjugated to Dynal 2.8 μm-diameter tosyl M-280 magnetic beads (MB; Invitrogen) via the three lysine residues in its structure. The beads were washed in 1X binding buffer (BB) (1XBB; 0.5M NaCl, 20 mM Tris HCl, and 2 mM MgCl2, pH 7.2–7.4) and five rounds of SELEX were conducted according to Bruno and Kiel’s protocol,14 except that a 72-base template (5′-ATCCGTCACACCTGCTCTN36-TGGTGTTGGCTCCCGTAT-3′, where N36 represents the randomized region) was employed with appropriate primers. PCR reactions were supplemented with 0.5 μL of Perfect Match Escherichia coli single strand binding protein (SSBP, Stratagene, La Jolla, CA) to inhibit high molecular weight concatamer formation according to Crameri and Stemmer.15
Chromatide AF 546-14-dUTP (14-carbon spacer between AF 546 and dUTP) was obtained from Invitrogen and incorporated at a concentration of 8 μM into the polyclonal aptamer family by conventional PCR16,17 and into the FMD 1F and 12F aptamer candidates (Table 1) by using their cDNAs (FMD 1R and 12R) as templates with a 100:1 forward:reverse ratio of primers18 to yield a predominately single-stranded AF 546-labeled product. Incorporation of AF 546-dUTP with Taq was relatively poor in both PCR protocols, but Deep Vent exo-polymerase (5 U; New England Biolabs, Ipswich, MA) gave intensely fluorescent bands in polyacrylamide gels indicating efficient incorporation.16 PCR conditions for AF 546-14-dUTP labeling were: 5 min at 95°C followed by 10 cycles of 95°C for 45 seconds, 53°C for 45 seconds, 72°C for 45 seconds, followed by 7 minutes at 72°C and indefinite holding at 4°C.
Aptamers were cloned into chemically competent E. coli using a GC kit from Lucigen (Middleton, WI). Several white and blue colonies were sent to Sequetech (Mountain View, CA) for rolling circle amplification-based sequencing with proprietary chemistry to relax secondary structures of some of the aptamers (Table 1). In theory, blue colonies should not contain the aptamer inserts, but because aptamers are quite short (72 bases), they often do not interrupt the lacZ gene sufficiently to disrupt color formation. We have, therefore, discovered a number of aptamer candidates in blue clone colonies.
A modified version of the Bruno and Kiel procedure14 was employed in which a Corning Costar N-oxy-succinimide (NOS) plate (No. 2525) was used to immobilize the FMD peptide in phosphate buffered saline (PBS, pH 7.2) for 1 hour at 37°C. The peptide was added to the top row of the plate at 50 μg per well in 100 μL of PBS and serial twofold diluted down the rows of the plate (Figure 1) except for the last row which served as a zero peptide deletion control. One hundred micrograms of Fraction V bovine serum albumin (BSA) from Sigma Chemical (St. Louis, MO) was added to several wells to assess cross-reactivity as well. The peptide and BSA were decanted and the plate was blocked for 1 hour at 37°C with 10% ethanolamine in PBS. Next, 100 μL of 5′-biotinylated aptamers were added at a concentration of 1.0 mg/mL (100 μg/well) for 1 hour at room temperature with gentle rotary mixing. Wells were washed three times with 200 μL of 1XBB for 5 minutes per wash. One hundred microliters of a 5-mg/mL streptavidin-peroxidase (Southern Biotech, Birmingham, AL) solution diluted 1:2000 was added per well for 30 minutes at room temperature with gentle mixing. The wells were decanted and washed three more times in 200 μL of 1XBB followed by addition of 100 μL of 1-Component ABTS (Kirkegaard Perry Laboratories, Gaithersburg, MD) per well for 15 minutes at room temperature. Absorbance was read on a microplate reader at 405 nm.
In the case of polyclonal AF 546-labeled aptamer families, the aptamers were first heated to 95°C for 5 minutes to render them single-stranded. The AF 546-labeled polyclonal or monoclonal aptamers were then added to an equimolar amount of BHQ-2-succinimide ester and allowed to combine at room temperature for 1 hour with gentle mixing. The conjugates were passed through a Sephadex G25 column and the absorbance of each 1-mL fraction was evaluated at 212 nm (peak absorbance for the FMD peptide), 260 nm (peak for DNA), 555 nm (peak for AF 546), and 579 nm (peak for BHQ-2). The fraction with the highest collective absorbance at these wavelengths (Figure 2B) was referred to as the “FRET reagent” and diluted 1:10 in 1XBB for use in the competitive FRET assays. FMD peptide was either serial twofold or tenfold diluted in 1-mL volumes of 1XBB across a series of polystyrene cuvettes except for the final cuvette which received no peptide and served as a zero control. One milliliter of diluted FRET reagent was added to each of the cuvettes to produce a 2-mL final volume. The cuvettes were gently swirled for at least 30 minutes at room temperature and fluorescence emission spectra were obtained from 560 to 600 or 650 nm with a Cary Varian Eclipse spectrofluorometer using an excitation of 555 nm, 5-nm slit widths, and photomultiplier voltage setting of 650 V. Mouse serum albumin (MSA; Sigma Chemical) was also used as a competitive target to assess general cross-reactivity in the FRET assays.
The polyclonal aptamer family produced a “lights off” response to the unlabeled FMD peptide in the competitive FRET format (i.e., exhibited decreasing fluorescence as a function of increasing peptide concentration) as shown in Figure 3. This FRET response was noted within 10 minutes, but grew in magnitude over 2 hours as illustrated by the difference between Figures 3A and 3B. Figure 3A demonstrates a sensitivity of at least 250 ng/mL of FMD peptide (curve c in Figure 3A). Curiously, the 25-ng spectrum matched the 250-ng spectrum at 2 hours after peptide addition (curves b and c in Figure 3B) and the fluorescence spectra were more broadly distributed at 2 hours. We also evaluated reproducibility of the fluorescence scans as presented in Figure 3C, which reveals a very high degree of reproducibility among three separate scans using 2.5 μg/mL of unlabeled FMD peptide at 2 hours after addition. The high level of reproducibility verified that the polyclonal FRET–aptamer assay was truly sensitive to the 25–250 ng/mL level.
Based on the initial competitive FRET results, we cloned and sequenced several aptamers from the polyclonal aptamer pool in an effort to optimize detection. This effort resulted in the forward (F) and reverse (R) aptamer candidate sequences listed in Table 1. These sequences were subjected to free energy minimization analysis with web-based Vienna RNA software19 using DNA parameters20 to produce the secondary stem-loop structures shown in Figure 4. This analysis was performed to determine if any similarities in secondary structures could be discerned, and several morphologic clusters appeared to emerge, but these could not be correlated with binding affinity from the colorimetric plate assays shown in Figure 1.
We hypothesized that only one of the forward or reverse primed complementary sequences from the cloned plasmids could be true aptamers, but entertained the possibility that both forward and reverse oligonucleotides may exhibit some affinity for the FMD peptide, because the peptide may have multiple “epitopes” or binding sites. We screened the 10 candidate aptamers in a colorimetric peroxidase-based plate assay as shown in Figure 1 and discovered that clones 1F, 1R, and 10F had superior affinity over the range of FMD peptide concentrations examined, while clones 10R and 12R, 13F, and 13R exhibited little or no affinity for the immobilized peptide. The remaining aptamer candidates appeared to be of intermediate to high affinity for the FMD peptide, but none of the aptamer candidates showed any noteworthy affinity for BSA (Figure 1).
It is interesting to note that both the polyclonal and monoclonal competitive FRET schemes generated “lights off” responses to increasing amounts of the unlabeled peptide (Figures 3A, 3B, and and2C).2C). This fact suggests that the polyclonal aptamer family is probably predominately composed of individual aptamer sequences in which the BHQ-2 portion or portions of the quencher-peptide conjugate may be held out away from the fluorophores in the binding pocket. In addition, it suggests that when the unlabeled peptide competes off the quencher-labeled peptide, it is better able to quench the fluorophores in the aptamer structure from its new position in the bulk solution.21
The plate binding results led us to choose clone 1F as our first aptamer for competitive FRET screening. Figure 2A illustrates that asymmetric PCR with Deep Vent exo-polymerase is effective at incorporating AF 546-14-dUTP into the 1F and 12F aptamer clones. Figure 2B demonstrates that the AF 546-aptamer-BHQ-2-peptide complex can be purified by Sephadex G25 size-exclusion chromatography and emerges in the fourth milliliter of eluate, because the highest aggregate absorbance peaks for each of the complexed components (fluorophore, DNA, peptide, and quencher) can be found in the fourth fraction. Unfortunately, as Figure 2C shows, this process did not improve the sensitivity of the competitive FRET assay with the FMD 1F aptamer candidate versus the FRET assay observed for the polyclonal aptamer family (Figures 3A and 3B).
We hypothesized that aptamer affinity would be a factor in the sensitivity of our competitive FRET assays and attempted the assay again with the 12F candidate, since this clone demonstrated intermediate to low affinity for the target peptide. However, it performed worse than 1F (data not shown), probably because of its general lack of affinity. Based on the sensitive results obtained in Figure 1, we remain hopeful that an optimal aptamer for our competitive FRET format exists in the FMD aptamer pool and simply needs to be discovered. If necessary, we will perform more rounds of selection and sequence a new set of aptamers, but this may increase affinity to a point which is detrimental to our competitive FRET approach (i.e., if the aptamer binds the quencher–peptide conjugate too tightly, it may not be competed off significantly by the unlabeled peptide).21
Theoretically, several factors probably influence the sensitivity of our competitive FRET assays. These factors include (1) aptamer affinity for the target; (2) number of fluorophores incorporated into the aptamer by PCR, especially if the fluorophores are not clustered and cannot self-quench; (3) proximity of fluorophore to quencher in the binding pocket; and (4) purity of the FRET complex (FRET reagent) following chromatographic separation. By piecing together data such as how FRET performance correlates with binding affinity in plate assays and studying potential binding pocket locations with the aid of computer-generated secondary or tertiary (3-dimensional) structures, we hope to optimize the placement of fluorophores and quenchers within the Förster distance in binding pockets to optimize FRET and maximize sensitivity of our assays. We intend to continue cloning, sequencing, and screening candidate aptamers by plate assay and competitive FRET from our polyclonal FMD aptamer pool until we find an optimal aptamer and conditions for rapid and sensitive detection of FMD in the field. We are currently working on a high-throughput means to accomplish FRET screening much more rapidly, thereby facilitating the search for an optimal competitive FRET–aptamer sequence.
This work was partly funded by U.S. Department of Agriculture grant no. 2006-33610-16757.