Viruses.

Snow Mountain virus (SMV), the prototype genogroup II, genotype 2 (GII.2) HuNoV and the target for aptamer selection, and Norwalk (NV) the prototype genogroup I, genotype 1 (GI.1) strain were obtained as stool specimens originating from a human challenge study (courtesy of C. L. Moe, Emory University, Atlanta, GA). The SMV human challenge study was conducted at the University of North Carolina at Chapel Hill (UNC-CH) and was approved by the UNC-CH Biomedical IRB. The NV study was conducted at Emory University and approved by the Emory University IRB, Biomedical Committee. Both studies had written consent, with each participant signing an informed consent witnessed by trained study staff. The informed consent documents were approved by the IRBs. The Emory University group also supplied pre-challenge stool samples confirmed (by RT-qPCR) as negative for HuNoV. These were used for counter selection and as negative controls in some studies. Additional fecal specimens associated with previously confirmed HuNoV outbreaks (representing strains GI.6, GII.1, GII.3, GII.4, GII.7 and an untypable GII) were also used in detection assays (courtesy of S.R. Greene, North Carolina Department of Health and Human Services, Raleigh, NC). All stool samples were suspended 20% in phosphate buffered saline (PBS). In some cases, these suspensions were used without further purification, designated as crude suspensions. In other cases, the suspensions were partially purified using chloroform extraction [10]. Hepatitis A virus HM175 (cell culture adapted) and poliovirus 1, routinely used in our laboratory, were partially purified (chloroform extracted) from cell culture lysates and used in exclusivity studies. All virus suspensions were stored at −80°C until use.

Virus-Like Particles (VLPs).

Self-assembled non-infectious Virus-Like Particles (VLPs), produced using the recombinant expression of HuNoV capsid proteins were used as purified candidate proteins for the characterization of aptamer binding affinity. The VLP panel consisted of representatives of genogroup I [GI.1 (Norwalk virus), GI.4, GI.6, GI.7 and GI.8)] and genogroup II [GII.1, GII.2 (SMV), GII.3, GII.4 (Houston and Grimsby virus), GII.6, GII.7, GII.12 and GII.17] HuNoV (kindly provided by R. Atmar, Baylor College of Medicine, Houston, TX). VLPs were stored at 4°C until use.

Preparation of the DNA Library.

All primer and probe sequences used in this study were obtained from Integrated DNA Technologies (Coralville, IA). The 81-base combinatorial DNA library consisted of 40-mer random regions flanked by forward and reverse constant regions [5′-AGTATACGTATTACCTGCAGC-(N)₄₀-CGATATCTCGGAGATCTTGC-3′]. Preparation of the dsDNA library for SELEX was done in accordance with the method of Dwivedi et al. (2010) [11]. Specifically, the library was amplified using a fluorescein (FAM)-labeled forward constant region primer [5′-56-FAM/-AGTATACGTATTACCTGCAGC-3′] and a biotinylated reverse constant region primer [5′-/5Bios/-GCAAGATCTCCGAGATATCG-3′]. Briefly, a 50 µl reaction master mix containing 5 µl of aptamer library (10 µM), 1x GoTaq buffer (Promega Corp., Madison, WI), 0.2 mM dNTPmix (Applied Biosystems, Foster City, CA), 5 U Go Taq DNA polymerase (Promega), and 500 nM of each primer were amplified as follows: 95°C for 5 min, followed by 30 cycles of 95°C (1 min), 55° (1 min), 72°C (1 min) and a final extension at 72°C for 10 min, using a DNA Engine (PTC-200) Peltier Thermal Cycler-200 (MJ Research/Bio-Rad Laboratories, Hercules, CA). The labeled dsDNA library was conjugated to Streptavidin MagneSphere Paramagnetic particles (SA-PMPs) (Promega) according to manufacturer instructions and harvested with an MPC-M magnetic particle concentrator (Invitrogen Dynal AS, Oslo, Norway). The library-magnetic bead conjugate was washed three times in 0.1X SSC. The FAM-labeled ssDNA moieties were then separated from the immobilized biotinylated strands by alkaline denaturation using 50 µl of 0.15 M NaOH for 7 min at room temperature (RT), followed by magnetic separation of the beads. The supernatant obtained was mixed with 1 ml of nuclease free water followed by filtration to remove residual NaOH using a YM-30 filter device (Millipore, Billerica, MA). It was then purified by ethanol precipitation with resuspension in 50 µl of nuclease free water. The presence of FAM-labeled DNA was confirmed using a fluorescent plate reader (Tecan Safire, Tecan Group Ltd., Männedorf, Switzerland) and its concentration determined using a NanoPhotometer Pearl (Implen GMbH, Munich, Germany).

Preparation of the target for SELEX.

The target for SELEX was produced by immobilizing SMV to antibody-bead conjugates. Briefly, M-280 Tosylactivated Dynabeads (Invitrogen Dynal) were cross-linked to mouse monoclonal antibodies against SMV (Abcam, Cambridge, MA) as per manufacturer instructions. One hundred µl of the partially purified SMV stock (consisting of approximately 10⁵–10⁶ RT-qPCR amplifiable units/ml) was mixed with 10 µl of the antibody-bead conjugate suspended in 500 µl of binding buffer [consisting of phosphate buffered saline (pH 7.1) supplemented with 100 mg/L CaCl₂, 100 mg/L MgCl₂ and 0.05% Tween 20] followed by RT incubation for 2 h. After washing thrice with PBS supplemented with 0.05% Tween 20 (PBST), the conjugate was resuspended in 20 µl of PBS and used for SELEX.

SELEX.

An aliquot of 300–500 pmoles of FAM-ssDNA pool suspended in 500 µl of binding buffer was added to 10 µl immobilized SMV target followed by gentle rotation for 45 min at RT. Aptamer-bound SMV was recovered by magnetic capture, washed thrice in PBST, and the aptamers eluted from the bead-bound virus particles with 200 µl of nuclease free water followed by heating at 90°C for 5 min. The resulting aptamer pool was purified by ethanol precipitation, re-amplified using the labeled constant region primers, and the FAM-ssDNA aptamers were recovered as described above. This constituted one selection round and a total of nine such iterations of the selection process were performed. To avoid non-specific amplification of the recovered aptamer pool, single stranded binding protein (Promega) at a concentration of 0.1 µg/µl was added to the PCR reactions and the pool was amplified using the appropriate annealing temperatures (obtained by running a temperature gradient within a range of 55°C to 65°C on the recovered aptamer pool).

Counter-SELEX.

Two sequential counter-SELEX rounds were done after each round of SELEX round to impart specificity to the aptamer candidates against (1) the components of a 20% HuNoV-negative human stool suspension; and (2) the bead-antibody complex (without SMV). In counter-SELEX, exposure of the aptamer pools to the negative stool specimens or bead-antibody complex was done as described above, but in this case, the aptamers bound to the complex were discarded, while the unbound aptamer pool (supernatant) was recovered. This was purified by ethanol precipitation, re-amplified by PCR, and used in another round of SELEX.

Identification of Aptamer Sequences.

After the 4^th, 7^th, and 9^th rounds of sequential SELEX and counter-SELEX, final PCR product was cleaned using the QIAquick PCR purification kit (Qiagen, Valencia, CA) and cloned into TOPO vector (Invitrogen, Carlsbad, CA) according to manufacturer instructions. The DNA insertion in each clone was sequenced by Genewiz (South Plainfield, NJ). The unique aptamers were identified by sequence analysis.

Dissociation Constant and Structural Analysis of Aptamers.

Consistent with the approach of others [12] [13], an ELISA-like assay (Enzyme-Linked Aptamer Sorbant Assay, or ELASA, described below) was used to determine equilibrium dissociation constants (K_d) for the candidate aptamers. This was done using SMV VLPs (3 µg/ml) and different concentrations (10 nM, 100 nM, 500 nM, 1 µM, 2 µM) of each aptamer. To estimate K_d, plots of the ratio between absorbance of Test samples/absorbance of Negative controls (T/N ratios, Y axis) as a function of the aptamer concentration (X axis) were generated using a non-interacting binding sites model in Sigma Plot (Jandel, San Rafael, CA). Common sequence motifs were identified using the online MEME server (http://meme.sdsc.edu). Structural folding analyses of the selected aptamers were done using the DNA Mfold online server (http://mfold.rna.albany.edu/?q=mfold/DNA-Folding-Form) [14]. To identify potential binding epitopes, the major capsid protein (VP1) sequences of each of the VLPs tested were aligned using Clustal W alignment in the Molecular Evolutionary Genetics Analysis program (MEGA 6.0) (http://www.megasoftware.net/). Aligned sequences were matched with the aptamers that showed the strongest (+++) binding affinity to each VLP.

Binding Affinity Studies Using Enzyme-Linked Aptamer Sorbent Assay (ELASA).

Binding affinity studies were performed with candidate aptamers and VLPs using a protocol adapted from a previously reported ELISA-based antigen detection assay [15]. In this case, the antibody was replaced with an aptamer and the resulting procedure termed Enzyme-Linked Aptamer Sorbent Assay (ELASA). For this assay, the selected aptamers were labeled with a 5′ biotin moiety. One hundred µl of pure VLP suspension (3 µg/ml) was placed in each well of a covered, flat-bottomed polystyrene 96-well plate (Costar 3591, Fisher, Pittsburg, PA) and incubated overnight at 4°C. After coating the wells with VLPs, the plates were blocked with 200 µl of 5% skim milk in PBST containing non-related DNA sequences (i.e., L. monocytogenes primers hlyQF/R and L23SQF/R) [16], followed by overnight incubation at 4°C. After washing three times with PBST, 100 µl of biotinylated aptamer (1 µM) was added to each well and the plate was incubated for an hour at RT with gentle mixing. After removing excess aptamers, the plates were washed three times with PBST. One-hundred µl of ELISA-grade streptavidin-horse radish peroxidase conjugate (1 mg/ml, 1∶5000, Invitrogen) was added to the plate and incubated at RT for 15 min. The unbound conjugate was removed and the plate was washed five times with PBST before applying 100 µl of 3,3′5,5′tetramethylbenzidine (TMB) peroxidase substrate (KPL, Gaithersburg, MD) to each well. The plate was incubated for 5 min at RT after which 100 µl of 1 M phosphoric acid was added to stop the reaction. Absorbance at 450 nm was recorded using a microplate reader (Tecan Infiniti M200pro). Negative controls consisted of no VLPs. As per convention [17], binding affinity was interpreted based on the ratio between the absorbance readings for the test samples (to which labeled aptamer had been added) versus those for the negative control (PBS alone), (T/N ratios). Ratios ≤2.0 were considered negative [17]. Ratios between 2.0 and 5.0; >5.0 and 10.0; and >10.0 were interpreted as low, medium or strong binding, respectively. Ratios obtained for the negative control were in the range of 0.1–0.3. To evaluate binding inclusivity, ELASA was performed using 1 µM (100 µl) of each candidate aptamer as applied to a panel of virus-like particles (VLPs) corresponding to genogroup I [GI.1 (Norwalk virus), GI.4, GI.6, GI.7 and GI.8)] and genogroup II [GII.1, GII.2 (SMV), GII.3, GII.4 (Houston and Grimsby), GII.6, GII.7, GII.12 and GII.17] HuNoV. Exclusivity analyses were done by ELASA using hepatitis A virus (HAV) and poliovirus. A scrambled random sequence for aptamer 25 (25 S) was generated by the Random Nucleic Acid Sequence Generator server on line http://molbiol.ru/eng/scripts/01_16.html and used to evaluate background binding in the ELASA. In all cases, PBS and SMV-VLPs were included as negative and positive controls, respectively.

Binding affinity Studies Using Enzyme-Linked Immunosorbent Assay (ELISA).

The original ELISA protocol adapted for ELASA was used in these experiments [15]. Briefly, SMV-VLPs were used to coat 96-well polystyrene plates as described above. The wells were blocked with 200 µl of 5% skim milk in PBST and incubated for 2 h at RT. After washing 3 times with PBST, 100 µl of diluted (1∶5000) GII.2 antibody (Abcam, Cambridge, MA) was added and incubated for 1 h at 37°C. After washing 3 times with PBST, 100 µl of the secondary antibody (anti-mouse conjugated with horseradish peroxidase; Abcam, Cambridge, MA) at a 1∶5000 dilution was added to each well and incubated for 1 h at 37°C. The plate was washed 3 times with PBST with subsequent development using TMB peroxidase substrate and absorbance reading was recorded as described above for ELASA.

Detection of HuNoV in Outbreak-Derived Stool Samples.

The ELASA assay was used to assess the performance of select aptamer candidates for detection of HuNoV in outbreak-derived fecal suspensions. Specimens evaluated included those representing strains GI.1 (Norwalk), GI.6, GII.1, GII.2 (SMV), GII.3, GII.4, GII.7 and an untypable GII. Briefly, 100 µl of ten-fold serially diluted partially purified fecal suspensions were used to coat plates followed by incubation overnight at 4°C. After three washes with PBST, the ELASA assay was done as described above. PBS alone and fecal suspensions derived from uninfected individuals were used as negative controls. Due to limited availability of SMV VLPs, we used GII.4 (Houston) VLPs, which were also highly reactive to aptamer 25, as the positive control.

Detection of HuNoV in Artificially Contaminated Lettuce Samples Using a Combined Pre-Concentration-Aptamer Magnetic Capture (AMC)-RT-qPCR Assay.

For virus inoculation and pre-concentration, 3 g of lettuce (approximately 3×3 cm square) was disinfected by UV light and inoculated with 200 µl of serially diluted crude GII.4 virus stock suspension (due to limited availability of SMV) at inoculum concentrations ranging from 1–5 log₁₀ RNA copies per lettuce sample. The inoculum was allowed to dry for 30 min. Virus pre-concentration was done using a previously reported elution-concentration method [18]. Briefly, the samples were mixed with 25 ml of 0.5 M glycine-0.14 M NaCl buffer (pH 9.0), placed in sterile Whirl-Pak-filter bags (Nasco, Fort Atkinson, WI) and stomached (Stomacher 400 Circulator, Seward, Norfolk, UK) for one min. The recovered filtrate (containing the eluted viruses) was adjusted to 0.9 M NaCl and supplemented with 1% bovine serum albumin (Sigma Aldrich, St. Louis, MO), after which 12% polyethylene glycol (PEG) MW 8,000 (Sigma Aldrich) was added. After incubation for 2 h at 4°C, samples were centrifuged at 10,000×g for 20 min at 8°C and the recovered pellet was resuspended in 1 ml of PBST.

For aptamer magnetic capture (AMC), the resuspended pellet was incubated with 1 µM of biotinylated aptamer for an hour at RT with rotation. This was followed by the addition of 10 µl of Streptavidin (SA)-C1 magnetic beads (from 10 mg/ml stock solution) (Invitrogen Dynal AS, Oslo, Norway) previously blocked overnight with 5% skim milk in PBST followed by incubation for 25 min at RT. The bead-aptamer-virus conjugates were recovered using the magnetic particle concentrator. The conjugates were washed twice with PBST, suspended in 100 µl PBS, and the RNA extracted using the easyMAG automated system (bioMerieux SA, Marcy l'Etoile, France) according to manufacturer's instructions. The viral RNA was eluted in 40 µl of proprietary elution buffer and stored at −80°C until used for amplification.

RT-qPCR was carried out using the Superscript III Platinum One-Step RT-qPCR system (Invitrogen) according to manufacturer instructions. Specifically, 25 µl RT-qPCR reactions consisted of 12.5 µl of 2X reaction mix, 5.5 µl DNase-RNase free water, 200 nM of each primer [JJV2F (5′-CAAGAGTCAATGTTTAGGTGGATGAG-3′) and COG2R (5′-TCGACGCCATCTTCATTCACA-3′)], and probe RING2-TP [5′-56-FAM TGGGAGGGCGATCGCAATCT-1BHQ-3′] [19] [20], 0.5 µl of the enzyme mix (SuperScript III RT/Platinum Taq Mix) and 5 µl of the RNA template. Amplification was done under the following conditions: 50°C for 15 min, 95°C for 2 min followed by 45 cycles of 95°C for 15 s, 54°C for 30 s, and 72°C for 30 s in a SmartCycler (Cepheid, Sunnyvale, CA). The RNA copy number was extrapolated from a standard curve based on Ct values obtained by RT-qPCR amplification of serially diluted synthetic RNA as previously reported [21]. Capture efficiency, expressed as a percentage, was estimated from the standard curve and calculated as the ratio of the extrapolated RNA copies (after capture and detection by RT-qPCR) to the total input RNA copies per sample, multiplied by 100 [22]. Negative controls consisted of capture by blocked beads in the absence of aptamer.