このコンテンツを視聴するには、JoVE 購読が必要です。 サインイン又は無料トライアルを申し込む。

概要

We provide a protocol that can be generally applied to select aptamers that bind to infectious viruses only and not to viruses that have been rendered non-infectious by a disinfection method or to any other similar viruses. This opens the possibility of determining infectivity status in portable and rapid tests.

Abstract

Virus infections have a major impact on society; most methods of detection have difficulties in determining whether a detected virus is infectious, causing delays in treatment and further spread of the virus. Developing new sensors that can inform on the infectability of clinical or environmental samples will meet this unmet challenge. However, very few methods can obtain sensing molecules that can recognize an intact infectious virus and differentiate it from the same virus that has been rendered non-infectious by disinfection methods. Here, we describe a protocol to select aptamers that can distinguish infectious viruses vs non-infectious viruses using systematic evolution of ligands by exponential enrichment (SELEX). We take advantage of two features of SELEX. First, SELEX can be tailor-made to remove competing targets, such as non-infectious viruses or other similar viruses, using counter selection. Additionally, the whole virus can be used as the target for SELEX, instead of, for example, a viral surface protein. Whole virus SELEX allows for the selection of aptamers that bind specifically to the native state of the virus, without the need to disrupt of the virus. This method thus allows recognition agents to be obtained based on functional differences in the surface of pathogens, which do not need to be known in advance.

Introduction

Virus infections have enormous economic and societal impacts around the world, as became increasingly apparent from the recent COVID-19 pandemic. Timely and accurate diagnosis is paramount in treating viral infections while preventing the spread of viruses to healthy people. While many virus detection methods have been developed, such as PCR tests^1,2 and inmunoassays³, most of the currently used methods are not capable of determining whether the detected virus is actually infectious or not. This is because the presence of components of the virus alone, such as viral nucleic acid or proteins, does not indicate that the intact, infectious virus is present, and levels of these biomarkers have shown poor correlation with infectivity^4,5,6. For example, viral RNA, commonly used for the current PCR-based COVID-19 tests, has very low levels in the early stages of infection when the patient is contagious, while the RNA level is often still very high when patients have recovered from the infection and are no longer contagious^7,8. The viral protein or antigen biomarkers follow a similar trend, but typically appear even later than the viral RNA and thus are even less predictive of infectability^6,9. To address this limitation, some methods that can inform on the infectivity status of the virus have been developed, but are based on cell culture microbiology techniques that require a long time (days or weeks) to obtain results^4,10. Thus, developing new sensors that can inform on the infectability of clinical or environmental samples can avoid delays in treatment and further spread of the virus. However, very few methods can obtain sensing molecules that can recognize an intact infectious virion and differentiate it from the same virus that has been rendered non-infectious.

In this context, aptamers are particularly well-suited as a unique biomolecular tool^11,12,13,14. Aptamers are short, single-stranded DNA or RNA molecules with a specific nucleotide sequence that allows them to form a specific 3D conformation to recognize a target with high affinity and selectivity^15,16. They are obtained by a combinatorial selection process called systematic evolution of ligands by exponential enrichment (SELEX), also known as in vitro selection, that is carried out in test tubes with a large random DNA sampling library of 10¹⁴-10¹⁵ sequences^17,18,19. In each round of this iterative process, the DNA pool is first subjected to a selection pressure through incubation with the target under the desired conditions. Any sequences that are not bound to the target are then removed, leaving behind only those few sequences that are able to bind under the given conditions. Finally, the sequences that have been selected in the previous step are amplified by PCR, enriching the population of the pool with the desired functional sequences for the next round of selection, and the process is repeated. When the activity of the selection pool reaches a plateau (typically after 8-15 rounds), the library is analyzed by DNA sequencing to identify the winning sequences exhibiting the highest affinity.

SELEX has unique advantages that can be exploited to gain increased selectivity against other similar targets^20,21, such as for infectivity status of the virus²². First, a wide variety of different types of targets can be used for the selection, from small molecules and proteins to whole pathogens and cells¹⁶. Thus, to obtain an aptamer that binds to an infectious virus, an intact virus can be used as the target, instead of a viral surface protein¹⁹. Whole virus SELEX allows for the selection of aptamers that bind specifically to the native state of the virus, without the need for disruption of the virus. Second, SELEX can be tailor-made to remove competing targets^21,23, such as other similar viruses or non-infectious inactivated viruses, using counter selection steps in each round of selection²². During the counter selection steps, the DNA pool is exposed to targets for which binding is not desired, and any sequences that bind are discarded.

In this work, we provide a protocol that can be generally applied for selecting aptamers that bind to an infectious virus but not to the same virus that has been rendered non-infectious by a particular disinfection method or to another related viruses. This method allows recognition agents to be obtained based on functional differences of the virus surface, which do not need to be known in advance, and so offers an additional advantage for the detection of newly emerged pathogens or for understudied diseases.

Protocol

1. Preparation of reagents and buffers

1. Prepare 10x Tris-borate EDTA (10x TBE) by adding 0.9 M Tris-base, 0.9 M boric acid, 20 mM EDTA (disodium salt), and deionized water to a final volume of 1 L. Mix until all the components are dissolved.
2. Prepare a 10% denaturing polyacrylamide stock solution as follows. In a 250 mL glass bottle, add 120 g of urea (8 M), 25 mL of 10x TBE, 62.5 mL of 40% acrylamide/bisacrylamide (29:1) solution, and enough distilled water to reach the final volume of 250 mL. Mix until all the components are dissolved.
3. Prepare 2x loading buffer as follows. In a 50 mL tube, add 21.636 g of urea (8 M), 1.675 g of EDTA (1 mM), and 4.5 mL of 10x TBE. Fill to 45 mL with distilled water and mix until all components are dissolved.
4. Prepare the extraction buffer by adding 100 mM sodium acetate and 1 mM EDTA (disodium salt). Set the final pH to 5.
5. Prepare ethanol precipitation solutions as follows. Prepare 3 M sodium acetate, adjust to pH 5.2, and store at 4 °C. Prepare 70% ethanol v/v and store at -20 °C. Prepare 100% ethanol and store at -20 °C.
6. Prepare 500 mL of SELEX buffer containing 1x PBS, 2.5 mM MgCl₂, and 0.5 mM CaCl₂. Adjust to pH 7.4. Prepare 100 mL of SELEX buffer with 8 M urea by adding 1x PBS, 2.5 mM MgCl₂, 0.5 mM CaCl₂, and 8 M urea. Adjust to pH 7.4.
   NOTE: The choice of SELEX buffer is critical to ensure the selected aptamer will work on the final sample where the aptamer will be applied. For instance, aptamers specific to SARS-CoV-2 will be used in biological samples (e.g., saliva or nasopharynges swabs). Thus, it is important to choose ion concentrations, pH, and buffer that closely mimic those conditions in the intended biological samples.
7. Prepare the binding and washing buffer as follows. In a 50 mL tube, prepare 5 mM Tris, 0.5 mM EDTA (disodium salt), and 2 M NaCl. Adjust to pH 7.5.

2. Design and synthesis of DNA library and primers

1. Design the initial ssDNA library and primers with the following criteria (see Table 1 for an example ssDNA library and set of primers, where N indicates a random nucleotide position).
   1. Ensure the library contains a central random region of 35 to 60 nucleotides flanked by two constant sequences at the 3' and 5' ends that act as primer regions for amplification. A typical library length is 45 random nucleotides.
   2. Ensure the reverse primer contains a biotin modification to separate ssDNA from amplified double stranded PCR products using streptavidin-coated beads during the in vitro selection process.
2. Purchase the DNA oligos from commercial sources with standard desalting purification. Purify the ssDNA library and primers by 10% denaturing polyacrylamide gel electrophoresis (dPAGE), followed by ethanol precipitation according to standard procedures²⁴.
   CAUTION: Acrylamide monomer and ethidium bromide are hazardous chemicals (human carcinogens), so use appropriate personal protective equipment (lab coats, gloves, and eye protection) while performing PAGE. When possible, avoid using ethidium bromide and replace it with a safer dye.
   NOTE: It is possible to order the oligos with HPLC purification to avoid performing this purification step, though this will not remove short oligonucleotides as effectively.
3. Add nuclease free water to resuspend ssDNA after ethanol precipitation up to the desired concentration (100 µM). Determine the concentration of the ssDNA library and primers by UV absorption at λ = 260 nm. Store at -20 °C.
4. Purchase an unmodified reverse primer to use for high-throughput sequencing library preparation and qPCR quantification.

3. Infectious and non-infectious virus samples

CAUTION: The infectious and non-infectious virus samples are biosafety level 2 (BSL2) samples that require extra care to handle safely and appropriately. All steps of the procedure that include these samples must be performed in a biosafety cabinet, or the virus solution must be in a sealed container (e.g., capped plastic tubes).

1. Obtain a stock solution of infectious viruses or prepare them, following the corresponding protocol for each virus. The SARS-CoV-2, SARS-CoV-1, and H5N1 pseudoviruses used here were provided by Prof. Lijun Rong's lab (UIC) in PBS buffer, and were prepared following the reported protocols^22,25,26.
   NOTE: For viruses that require biosafety level 3 facilities to handle the intact infectious virus, it is possible to work with pseudoviruses. A pseudotyped virus is generated from a lentivirus (HIV) that displays the surface proteins of the virus within the viral envelope, and thus closely mimics the surface and entry mechanism of the virus but is defective in continuous viral replication.
2. Obtain a non-infectious virus stock solution or prepare it by following the disinfection procedure. The UV-inactivated SARS-CoV-2 pseudovirus (p-SARS-CoV-2) stock used here was provided by Prof. Rong lab (UIC) in PBS buffer, and the previously reported protocol was used for UV-inactivation²².
   NOTE: If possible, perform an assay to test the infectivity of the virus to ensure that the virus has been completely inactivated.
3. Quantify the virus stocks
   1. Quantify the virus using a plaque assay according to standard procedures^27,28,29. This assay not only allows the quantification of the virus but also indicates the infectivity status of the virus, confirming the concentration of infectious virus.
   2. For pseudoviruses or viruses where a plaque assay is not available, quantify the virus using a commercially available quantitative lentivirus ELISA kit.
   3. Prepare 1 x 10⁸ copies/mL of stock solution for p-SARS-CoV-2, p-SARS-CoV-1, and p-H5N1. Separate each stock solution in aliquots containing 50 µL each and keep at -80 °C. Thaw a new aliquot before each experiment.

4. In vitro selection or SELEX process: initial round

NOTE: For all the steps using the infectious virus, work in a BSL2 cabinet.

1. Denature the ssDNA library. Take 10 µL of 100 µM of the ssDNA library (1 nmol) and mix with 240 µL of SELEX buffer. Heat at 95 °C for 15 min in a dry bath, and then place the tube on ice for 15 min.
2. Incubate the ssDNA library with the infectious virus (positive selection step) as follows. Mix 250 µL of ssDNA library in SELEX buffer from step 4.1 with 50 µL of infectious virus (1 x 10⁸ copies/mL). Incubate for 2 h at room temperature.
3. Block the nonspecific sites in centrifuge filters as follows. While waiting on step 4.2, prepare the centrifuge filters for the next steps.
   1. Add 400 µL of 1 mM T20 sequence (Table 1) to each 0.5 mL centrifuge filter that will be used-one centrifuge filter with a cut-off of 100 kDa and one with a cut-off of 10 kDa.
   2. Incubate for 30 min at room temperature and centrifuge at 14,000 x g for 10 min to remove the T20 solution in the filter. Wash 3x with SELEX buffer to remove excess T20 sequences.
4. Wash unbound sequences as follows. Add the mix incubated in step 4.2 to the blocked 100 kDa centrifuge filter and centrifuge at 14,000 x g for 10 min. Wash 3x, adding 400 µL of SELEX buffer and centrifuging at 14,000 x g for 10 min. Keep the fraction in the filter and discard the flow-through.
5. Elute bound sequences as described below.
   1. Change the collection tube of the centrifuge filter and add 300 µL of SELEX buffer containing 8 M urea to the centrifuge filter in step 4.4. Heat the centrifuge filter at 95 °C for 15 min in a dry bath and then centrifuge at 14,000 x g for 10 min.
   2. Collect the fraction that flowed through the filter containing the eluted sequences. Repeat another three times. Work in a BSL2 cabinet and wash all materials with 10% bleach to inactivate the virus before discarding the materials.
6. Concentrate and desalt as follows. Add the solution collected in 4.5 to a blocked 10 kDa centrifuge filter. Centrifuge at 14,000 x g for 15 min. Discard the fraction that flowed through the filter.
   NOTE: As the viruses are retained and inactivated with 8 M urea in the filter in the previous step, this and the following steps do not need to be performed in the biosafety cabinet. This procedure needs to be repeated if a 0.5 mL centrifuge tube is used until all the solution is collected as in step 4.5, 1.2 mL flows through the filter.
7. Wash 3x with 300 µL of SELEX buffer to remove the urea by centrifuging at 14,000 x g for 15 min. Discard the fraction that flowed through the filter. Recover the solution in the filter by turning it upside down in a clean collection tube and centrifuging for 5 min.
8. Measure the final volume of the recovered solution from 4.7 using a pipette in a plastic tube. This solution is named R_ia, where i corresponds to the round number. Typically, volumes between 30 µL to 50 µL are obtained. Take 1 µL of the eluted ssDNA to quantify the amount of DNA by qPCR.
   NOTE: This is a possible pause point, where the R₁a sample can be kept at −20 °C to continue at another time.
9. Perform PCR amplification as described below.
   1. In the first round, take 90% of the R₁a sample as a PCR template. In the following rounds, take 60% of the total volume in step 4.8 to perform PCR and store the other fraction at -20 °C.
   2. Set up a 50 µL PCR reaction with the following having the specified final concentration: 1x PCR reaction buffer, 200 µM dNTPs, 10 µL of DNA template (R₁a sample), 200 nM each primer, and 1.25 U polymerase (2.5 U/µL stock).
   3. Optimize the PCR conditions including number of cycles and annealing temperature for each set of primers and ssDNA library. Avoid using too many cycles, which will produce undesired PCR subproducts such us primer-dimers. Test different annealing temperatures based on the melting temperature of the primers.
   4. Run the PCR using optimized conditions at 95 °C for 5 min followed by 18 cycles of 1 min at 95 °C, 30 s at 52 °C, and 1 min at 72 °C, with a final extension step at 72 °C for 10 min, to obtain the dsDNA pool.
      NOTE: This is another possible pause point, where the PCR product can be kept at −20 °C to continue at another time.
10. Recover ssDNA using streptavidin-modified magnetic beads.
    1. Take 80% of the PCR product. Store the remaining fraction at -20 °C.
    2. Split the PCR product in aliquots of 50 µL and add them to microfuge tubes containing 50 µL of streptavidin-modified magnetic beads (MB). Incubate for 30 min with mild agitation (for instance, use a rotating shaker) at room temperature. Then, place the microfuge tube on the magnetic rack to isolate the MB and remove the supernatant by pipetting (it must be a clear solution).
    3. Wash by adding 200 µL of binding and washing buffer to the tube. Remove the tube from the magnetic rack, tap the tube, and resuspend the MB to a homogeneous solution by pipetting. Place the microfuge tube back on the magnetic rack to isolate the MB and remove the supernatant by pipetting. Repeat this washing step twice.
    4. Once washing is complete, resuspend the MB in a total of 100 µL of SELEX buffer and heat the solution to 95 °C for 10 min. Immediately place the tube in the magnetic rack before the tube cools down and take the supernatant containing the ssDNA pool. Repeat this recovery step, adding 50 µL of SELEX buffer.
11. Measure the final volume of the recovered fraction from 4.10 using a pipette. This fraction is named R_ix, where i corresponds to the round number. In general, volumes between 140 to 150 µL are obtained. Take 1 µL of R₁x to quantify the amount of DNA by qPCR.

5. Subsequent selection rounds

1. Denature the ssDNA pool. Take 60% of sample R₁x and mix with 50 µL of SELEX buffer. Heat at 95 °C for 15 min in a dry bath. Place the tube on ice for 15 min.
2. Incubate the ssDNA pool with the non-infectious virus and other potential interfering viruses (counter selection step). Mix the denatured ssDNA pool in SELEX buffer from 5.1 with 50 µL of each virus (5 x 10⁹ copies/mL; e.g., non-infectious p-SARS-CoV-2, p-SARS-CoV-1, and p-H5N1). Incubate for 1 h at room temperature.
3. Block nonspecific sites in centrifuge filters. While waiting on step 5.2., prepare the centrifuge filters for the next steps. Typically, use two centrifuge filters, one with a cut-off of 100 kDa and one with a cut-off of 10 kDa. Follow the same procedure as in 4.3.
4. Wash away the sequences bound to non-target viruses. Add the mix incubated in step 5.2 to a blocked 100 kDa centrifuge filter. Centrifuge at 14,000 x g for 10 min and recover the fraction that flows through the centrifuge filter. Wash 2x, adding 100 µL of SELEX buffer and centrifuging at 14,000 x g for 10 min. Keep the fraction that flowed through the filter.
5. Incubate the ssDNA pool with the infectious virus (positive selection step). Take the 300 µL of fraction from step 5.4 and mix with 50 µL of the infectious virus (1 x 10⁸ copies/mL). Incubate for 2 h at room temperature. Follow steps 4.4 to 4.11.

6. Monitoring the SELEX process

NOTE: To monitor the enrichment of the pool, qPCR is used in two ways. First, by absolute quantification, it is possible to test the enrichment of the pools (elution yield). Second, by monitoring the melting curve, the diversity of the pools (convergence of the aptamer species) can be evaluated³⁰.

1. Perform all of the qPCR assays with a 10 µL reaction volume in 96-well plates designed for qPCR.
2. Prepare a standard qPCR mixture containing 5 µL of master mix, 0.3 µL of 500 nM of each unlabeled primer, 3.4 µL of H₂O, and 1 µL of DNA template.
   1. Prepare dilutions of purified ssDNA library (step 2.3) to run the standard curve.
   2. Dilute the DNA samples to make their concentration fit in the standard curve. For R_ia samples, add 1 µL of the sample without dilution and 1 µL of 1:10 dilution to the qPCR mixture. For R_ix samples, include 1:10 or 1:100 dilutions.
3. Run qPCR by setting the following protocol. First, set an initial denaturation step at 98 °C for 2 min. Then, set 40 cycles of denaturation at 98 °C for 5 s and annealing and extension at 52 °C for 10 s. Finally, set a melting curve analysis from 65 to 95 °C. Determine threshold cycle (Ct) values by automated threshold analysis.
4. Using the standard curve, quantify the amount of ssDNA in R_ia and R_ix samples.
5. Calculate the elution yield as the amount of bound DNA (number of moles in R_ia) divided by the amount of initial DNA (number of moles in R_i-1x). Plot the elution yield vs round number to see if an enrichment over the rounds is observed.
6. Plot the melting curves for different rounds. A shift to higher melting temperatures in the peak near 75/85 °C is expected as the number of rounds increase. This means the diversity of the pool decreases.

7. High-throughput sequencing

1. Consult with the high-throughput sequencing facility about what sequencing types and scales are available, which will determine the library preparation method.
   NOTE: This decision will take into consideration both the length of the pools to be sequenced (including the primers) and the number of pools to be submitted (generally, aim for at least 1 x 10⁵-10⁶ sequences per pool). If a particular sequencing scale cannot read the entire length of the pool, paired-end sequencing can be used to read from both ends of the sequence, as opposed to single-end which reads from only one end.
2. Select an appropriate commercially available kit based on the type of sequencing, in consultation with the sequencing facility. Prepare multiple pools of selection rounds representing high, medium, and low enrichment, as well as the final pool, using a different pair of indexes in the adapters for each pool that allowed the sample analysis of all rounds using one lane.
   NOTE: PCR amplification can also be used to incorporate the adapters and indexes required for high-throughput sequencing, but this generally costs similar to the commercial kits and does not perform any better.
   1. Perform library preparation by following these steps: end repair of fragmented DNA, adaptor ligation, and PCR amplification to produce the final libraries.
   2. Purify the PCR product with magnetic DNA clean-up beads following the manufacturer's instruction.
   3. Quantify the DNA using a fluorescence-based quantification kit. Avoid using less accurate methods, such as small volume UV measurements.
   4. Mix approximately equal amounts (by amount of DNA, not volume) of each pool library containing specific indexes.
   5. Check the quality of the final library with qPCR quantification and a DNA fragment analyzer. This step is often performed by the sequencing facility.
      NOTE: Sequencing facility personnel can provide useful information on how to prepare the library and which quality controls are suggested. Discussing with them is required before preparing the libraries.
3. Send the prepared final library to the sequencing facility following their requirements.

8. Sequencing analysis

1. Most software available for sequencing analysis is designed to be run on the command-line on a Linux operating system. For small datasets, set up the desired Linux operating system and install required programs on a personal computer; for larger datasets, HTS analysis is run on high-performance computing resources or super computers (recommended).
   NOTE: Access and training will be needed to use high-performance computing resources, which may take some time to obtain. Additionally, basic knowledge of UNIX command-line usage will be required. Many free resources on basic command-line usage are available online, and the computing resources will likely provide additional information for using their systems.
2. Access the sequencing files, usually in a compressed FastQ file format, using the information provided by the sequencing facility.
   1. Use an FTP client to transfer the files to your own computer and/or the high-performance computing resource. Many sequence analysis programs can directly use compressed files, so keep the files compressed, if possible, to limit their file sizes (large sequence reads of 50 M+ sequences can take up 100 Gb+ of file space when uncompressed).
   2. If sequence files need to be uncompressed, use the gzip function which is standard on most Linux installations. Obtain additional information about gzip and its options from the installed manual by typing "man gzip" on the command-line.
3. Clean-up the sequences before analysis by demultiplexing, removing the sequencing adapters, removing low quality reads, and including both forward- and reverse-direction reads.
   1. Use the FastQC program³¹ after each of the following steps to get basic quality control information on the sequences to assess the effectiveness of each clean-up step.
   2. Demultiplex the sequences to separate all sequences from a single read file into the different pools based on the indexes included in the sequencing adapters. This step is often performed by the sequencing facility, who should be consulted as the exact procedure will depend on the sequencing machine used.
   3. Remove the sequencing adapters using the command-line program, Cutadapt³². Use the selection primers (rather than sequencing adapters) as the input in Linked Adapter mode. Use the –discard-untrimmed option to discard any sequences that do not contain the selection primers.
   4. Set options for the maximum error rate to match the adapters, and the minimum and maximum length of sequences to accept. If using paired end reads, input specific parameters available and give both the Read 1 and Read 2 sequencing files. Set additional parameters as needed by referring to the Cutadapt manual³².
      NOTE: The ligation of sequencing adapters to pools is direction independent and will incorporate approximately 50% of sequences in the forward direction and 50% in the reverse direction. To avoid losing the 50% of sequences in the reverse direction, Cutadapt can be run a second time using the reverse-complement of the selection primers as the inputs.
   5. (Optional) If paired-end sequencing is used, merge the Read 1 and Read 2 files with the program PEAR³³.
   6. Use the FASTX-Toolkit program³⁴ to remove low-quality sequences that have a quality below a certain threshold at any nucleotide. A quality cut-off score of 30 is a good starting point. If the overall quality is generally good, this step can be skipped.
   7. To merge the sequence files in the reverse direction with those in the forward direction, use the FASTX-Toolkit fastx_reverse_complement function on the reverse direction files. Then use the built-in cat program (standard on most Linux installations) to merge the forward and reverse-complement – reverse files into a single file.
4. To analyze the enrichment of sequences across different selection pools, use the FASTAptamer analysis toolkit³⁵.
   1. Use FASTAptamer-Count to count the number of times each sequence appears. Then, rank and sort the sequences by abundance. The count output file is required as input for all other FASTAptamer programs.
   2. Use FASTAptamer-Clust to group all sequences in a file into clusters of closely related sequences. Use the "distance" parameter to define the number of single nucleotide mutations or indels allowed to group into a cluster.
      NOTE: The FASTAptamer-Clust program can be very computationally expensive if a high number of unique sequences are present (especially for early rounds), which could take days or even weeks to fully complete. The cut-off filter parameter can be used to exclude low abundance sequences from clustering. Generally, it is good to start with a higher cut-off, such as 10 or 5 reads per million (RPM) and decrease it if the program finishes quickly. If the pools are very heterogenous, it may not be feasible to cluster the sequences in a reasonable timeframe.
   3. Use FASTAptamer-Enrich to calculate the enrichment of each sequence present in more than one round of the selection. Compare the RPM of the sequence from one population with the RPM in another. Use the Enrich program on either the Count or Cluster files. The output is a tab-separated-value (.tsv) file that can be opened in any spreadsheet software for further analysis and easy sorting.
      NOTE: The FASTAptamer-Enrich program can also be very computationally expensive if a high number of unique sequences are present. The cut-off filter parameter can be used to exclude low abundance sequences from the output to avoid files that are too big to open in spreadsheet software.
   4. (Optional) If many pools are too heterogenous (many unique sequences and few duplicate sequences), it may not be feasible to use either the Clust or Enrich programs. In this case, perform a manual enrichment check using the Count output file which sorts sequences by abundance (most abundant the at top) and renames the sequence identifier to include important information such as RPM.
      1. Use the standard Linux "head" program on the Count files to get a list of the most abundant sequences. Then use the standard Linux "grep" program to search each of the most abundant sequences in the Count files of all of the other relevant pools. If there are any matches, use the modified sequence identifier to determine the RPM in that pool, to manually construct the enrichment information.
         NOTE: Scripts for all the sequencing analysis steps can be found in the Supplementary Coding Files 1-11.

9. Aptamer binding validation and assays

1. From the enrichment information obtained through sequence analysis, identify candidate aptamer sequences based on sequences that have the most enrichment over subsequent selection rounds and/or the most abundant sequences in the final selection round. Select several different candidate sequences (at least 10-20) for further testing as the most highly enriched may not necessarily be the best performing aptamers.
2. Perform initial screening of candidate aptamers using a binding assay that can be performed quickly for multiple samples. A column-based ultrafiltration binding assay³⁶ or enzyme-linked oligonucleotide assay (ELONA) are good methods for this initial screening.
3. Analyze the specificity and affinity of the sequences that show the best binding activity from the initial screening, using at least two independent binding assays (e.g., MicroScale Thermophoresis (MST)^37,38, Enzyme-Linked Oligonucleotide Assay (ELONA)³⁹, surface plasmon resonance⁴⁰, or biolayer interferometry (BLI)⁴¹).

Representative Results

Since DNA aptamers can be obtained using SELEX in a test tube¹⁵, this SELEX strategy was carefully designed to include both positive selection steps toward the intact, whole infectious virus (i.e., retain the DNA molecules that bind to the infectious virus), as well as counter selection steps for the same virus that has been rendered non-infectious by a particular disinfection method, specifically UV-treatment, by discarding the DNA sequences that can bind to the non-infectious virus. A schematic representation of the selection process is shown in Figure 1.

As representative results of this protocol, we chose the selection of an aptamer for infectious SARS-CoV-2. Due to the requirement of biosafety level 3 (BSL3) working conditions to handle the intact, infectious SARS-CoV-2, we have chosen instead to work with a pseudotyped virus. Pseudotyped viruses are generated from a relatively harmless backbone virus, in this case a type of lentivirus (HIV), that has been modified to display the surface protein of a different virus of interest within its viral envelope, allowing it to closely mimic the surface and entry mechanism of the desired virus without the risks associated from working with dangerous human pathogens. Importantly, these pseudoviruses are modified to be defective in continuous viral replication^{25, 42}. In this work, three different pseudotyped viruses were used: p-SARS-CoV-2, where SARS-CoV-2 spike (S) proteins are incorporated in the envelope; p-SARS-CoV-1, containing SARS-CoV-1 spike (S) protein; and p-H5N1, containing hemagglutinin and neuraminidase isolated from the highly pathogenic avian influenza virus A/Goose/Qinghai/59/05 (H5N1) strain.

In the first two rounds of selection, no counter selection step was included to minimize the nonspecific removal of DNA binders that are typically only present as single copies in beginning rounds. After the first two rounds, positive and counter selection steps were included in each round to reach a high selectivity. For counter selection, p-SARS-CoV-2 that had been rendered non-infectious by UV-treatment, as well as p-SARS-CoV-1 and p-H5N1 were used to gain selectivity against other viruses.

To monitor the selection progress, quantitative polymerase chain reaction (qPCR) was used. This technique represents a simple method that does not require labelling of the pool and can be generally applied to virtually any target³⁰. It takes advantage of two features of the qPCR technique. First, the possibility of absolute quantification using a standard curve to calculate the elution yield, defined by the ssDNA bound to the infectious virus over the total added ssDNA. Our results showed that the elution yield initially increased with each early round of SELEX and leveled out at rounds 8 and 9 (R8 and R9), suggesting an enrichment of the pools in sequences that bind to the infectious virus (Figure 2A). Second, melting curves of every round of the DNA pool provide further evidence of the sequence diversity of the pools³⁰. By comparing the melting curves, a shift from the peak at high melting temperature (T_m) from 77 °C to 79 °C can be observed, suggesting that the DNA pool has converged from random sequences with low T_m to more conserved sequences with higher T_m (Figure 2B). Moreover, in the last rounds (R8 and R9), a peak at low T_m appears (~70 °C). This could be related to the production of primer-dimers during the PCR. One possible solution to avoid this is to reduce the number of cycles during PCR (for instance, from 18 cycles to 12 or 15 cycles).

After confirming an enrichment of the pool by qPCR, we used high-throughput sequencing (HTS) for rounds 3, 5, 7, 8, and 9 of the SELEX to find out which sequences are responsible for the binding of the virus target. Compared with traditional cloning-sequencing procedures, HTS allows the evolution of individual sequences over multiple selections rounds to be monitored, to finally identify the aptamer sequences that are enriched over subsequent rounds. Figure 3A shows the relative abundance (in reads per millions) for the sequence SARS2-AR10, obtained over consecutive selection rounds. The secondary structure of SARS2-AR10 is predicted by Mfold software, and the results (Figure 3B) show a structured secondary structure that contains a stem loop region which may be involved in recognizing the virus. Further characterization of the affinity binding of this sequence has been reported previously²². Microscale thermophoresis (MST) and an enzyme-linked oligonucleotide assay (ELONA) demonstrated that the SARS2-AR10 aptamer binds to the infectious p-SARS-CoV-2 with a K_d = 79 ± 28 nM, and does not bind to the non-infectious p-SARS-CoV-2 or to other coronaviruses such as p-SARS-CoV-1 and 229E.

Figure 1: Schematic representation of the SELEX process of aptamers that distinguishes infectious from non-infectious viruses. Positive and counter selection steps were added in each round after the second round to reach high specificity toward the infectious virus. Please click here to view a larger version of this figure.

Figure 2: Monitoring the progress of SELEX of infectious SARS-CoV-2-specific aptamers. (A) Quantification of the elution yield (i.e., the bound ssDNA over the added ssDNA) for each round of SELEX using qPCR. (B) Melting curve for the different pools during SARS-CoV-2 aptamer selection. Please click here to view a larger version of this figure.

Figure 3: Enrichment and secondary structure of SARS2-AR10 aptamer. Reads per million (RPM) obtained by analysis of the HTS data for SARS2-AR10 sequence as a function of the selection rounds, using FASTAptamer-Count. Inset: The predicted most stable secondary structure of the SARS2-AR10 sequence based on the UNAFold software. Calculations were made at 25 °C, 100 mM NaCl, and 2 mM MgCl₂. Please click here to view a larger version of this figure.

Name	DNA sequence (5′ to 3′)
T20	TTTTTTTTTTTTTTTTTTTT
DNA library	ACCGTCAGTTACAATGCT- N45 -GGCTGGACTATCTGTGTA
Forward primer (FwdP)	ACCGTCAGTTACAATGCT
Reverse primer (RevP)	Biot-TACACAGATAGTCCAGCC
SARS-AR10	CCCGACCAGCCACCATCAGCAACTCTTCCGCGTCCATCCCTGCTG
N: represent a random nucleotide; Biot: Biotin modification in the 5´end.

Table 1: List of DNA sequences.

Supplementary Coding File 1: Cutadapt_script.txt identifies any sequences in a given input file that have the specified forward and reverse primers, discards any that do not have the primers, and removes the primers from the sequences that did have them. See reference³¹ for more information on Cutadapt. Please click here to download this File.

Supplementary Coding File 2: FASTAptamer_cluster_script.txt will cluster all sequences from a single count file into families/clusters of closely related/similar sequences. This is useful if a candidate sequence is identified, so that other similar sequences in the same cluster can also be identified as potential candidates as well. See reference³⁴ for more information on FASTAptamer. Please click here to download this File.

Supplementary Coding File 3: FASTAptamer_count_script.txt will count the number of occurrences of each unique sequence from a given input FASTA/FASTQ file and produce an output file of each unique sequence in descending order (highest abundance to lowest abundance). The Count output files are required as inputs for all other FASTAptamer programs. See reference³⁴ for more information on FASTAptamer. Please click here to download this File.

Supplementary Coding File 4: FASTAptamer_enrich_script.txt compares all sequences from any two or three input files, gives the abundance of all unique sequences between all of the files, and gives all combinations of pairwise enrichment values (in RPM) of all sequences found in multiple files. See reference³⁴ for more information on FASTAptamer. Please click here to download this File.

Supplementary Coding File 5: fastqc_script.txt is a quality control tool for FastQ files that provides easy to read quality information for high-throughput sequencing data, such as duplication levels, read lengths, and quality scores. See reference³⁰ for more information on FastQC. Please click here to download this File.

Supplementary Coding File 6: FASTX_fwd_rev_merge_script.txt uses the FASTX-Toolkit to output the reverse-complement of all sequences from a given reverse file. It then uses the cat command (standard in most UNIX operating systems) to merge this reverse reverse-complement file with a given forward file of sequences, to give a single file with all sequences. See reference³³ for more information on FASTX-Toolkit. Please click here to download this File.

Supplementary Coding File 7: FASTX_quality_filter_script.txt uses the FASTX-Toolkit to check the quality of sequences in a given FastQ file and can discard any sequences that are of low quality. This method is generally better than trimming methods of quality filtering for analysis of in vitro selection sequences. Trimming involves removing bases (typically near the ends) of sequences based on low quality, which is acceptable for genomic sequencing, but because the entire sequence is needed for functional DNA, trimming the ends is not useful. Instead, if too many bases of a sequence are low quality, the entire sequence is discarded. See reference³³ for more information on FASTX-Toolkit. Please click here to download this File.

Supplementary Coding File 8: grep_searcher_script.txt is used to search for a specific input sequence in a specific input file, and output both the ID and the sequence in an output file (if it was found in the input). This script is useful for pools that are very heterogenous (i.e., have a lot of unique sequences), which make FASTAptamer Clust take too long to run properly and result in FASTAptamer Enrich files that are too large to open in typical spreadsheet programs for analysis. Please click here to download this File.

Supplementary Coding File 9: gzip_compress_decompress_script.txt uses Gzip to compress or uncompress files. A compressed file will typically have a .gz extension appended to the file name. It is recommended to compress large files to save on file space. Please click here to download this File.

Supplementary Coding File 10: PEAR_script.txt is used only for paired-end sequencing files and will merge the Read 1 file with the corresponding Read 2 file. See reference³² for more information on PEAR. Please click here to download this File.

Supplementary Coding File 11: tar_extraction_creation_script.txt will extract or create Tar archives, which are used to collate multiple files and/or directories into a single file. They are also often compressed to reduce file size, such as with bz2 compression, as included in this script. Please click here to download this File.

Discussion

SELEX allows not only the identification of aptamers with high affinity, in the pM-nM range^22,43,44,45, but also with high and tunable selectivity. By taking advantage of counter selection, aptamers with challenging selectivity can be obtained. For instance, the Li group has demonstrated the ability to obtain sequences that can differentiate pathogenic bacterial strains from non-pathogenic strains²¹. Also, Le et al. identified an aptamer able to differentiate serotypes of Streptococcus pyogenes bacteria⁴⁶. This opens the possibility of integrating aptamer molecules with unique selectivity in different functional DNA sensors^{12,47,48,49,50} with new nanotechnologies^22,51,52 to construct sensors for different applications, such as for portable and rapid diagnostic tests or for environmental detection.

The protocol presented here allows aptamers to be obtained with high selectivity for an infectious virus over the same virus that has been inactivated and is thus non-infectious. To obtain such selectivity, the SELEX approach in this work is based on the design of the counter selection step. First, for a successful counter selection step, a correct choice and characterization of non-target samples needs to be performed. Thus, this method depends on starting with well-known infectious and non-infectious intact virus stocks. Second, a counter selection step is incorporated in each round of the selection after the first round, to apply stringent conditions from the beginning of the selection and to maximize the probability of identifying aptamers that can differentiate between the slight differences in the surface of the virus particle from the infectious and non-infectious state of the same virus.

In this protocol, the focus is on the selectivity of the aptamer. If stronger aptamer affinities are required, it is possible to increase the stringency of the process after certain SELEX rounds. For instance, by decreasing the infectious virus concentration and incubation time during the positive selection, it is possible to obtain aptamers with a stronger binding affinity⁴⁴.

Another important consideration when designing the SELEX protocol is the final sample in which the aptamer will be applied. Aptamers specific to viruses are usually incorporated in complex samples (e.g., serum, saliva, and real-water samples), and this need to be considered during the aptamer selection to increase the chance of obtaining functional aptamers in the final samples. Therefore, it is important to perform SELEX in a buffer that closely mimics these samples, particularly with regard to cation concentration and pH. For instance, for the SARS-CoV-2 aptamer, we chose PBS buffer (pH 7.4) containing 2 mM MgCl₂ and 0.5 mM CaCl₂ to closely mimic biological samples. In addition, if other species in the samples where the aptamer will be applied can potentially compete with the target to bind the aptamer, it is possible to minimize interference from these species by including them in the counter selection step, to eliminate sequences that could bind these species.

The successful results of this SELEX process²², which has been applied to different viruses (adenovirus and SARS-CoV-2) and different inactivation methods (UV-treatment and free chlorine) indicate that this method can be widely applied for other viruses and other inactivation methods, opening a broad range of applications in the future. Aptamers that distinguish infectious viruses have potential not only in diagnostic applications to identify patients that are still contagious, but also in environmental applications where the sample is defined as contaminated if the pathogens in these samples are still active. Thus, the incorporation of aptamers with the ability to identify infectious viruses in rapid sensors offers opportunities to monitor disinfection treatments.

Furthermore, it was possible to differentiate the infectivity status of the virus without any information on the structural differences between the two infectivity states. The only information that is required is to know that the differences between both states are related to differences in the structure of molecules on the surface of intact viral particles. Thus, we are confident that the same approach can be used to differentiate variants and serotypes of the same virus, where the differences are due to small mutations in the residues of proteins on the surface of viral particles, similar to those that differentiate the infectivity status of the virus. For instance, we envision it will be possible to obtain aptamers specific to variants of concern of SARS-CoV-2 or other viruses, such as influenza, opening the possibility of rapid monitoring of variants that are the biggest threat for society.

Finally, because the SELEX strategy to obtain aptamers that can differentiate infectious from non-infectious viruses does not require any foreknowledge about what specific structural differences are present between them, it may be possible to use aptamers with this selectivity to identify the specific surface changes responsible for the loss of infectivity from various disinfection methods by a detailed characterization and identification of the binding targets of our aptamers.

Materials

10% Ammonium persulfate (APS)	BioRad	1610700
100% Ethanol	Sigma-Aldrich	E7023
1x PBS without calcium & magnesium	Corning	21-040-CM
40% acrylamide/bisacrylamide (29:1) solution	BioRad	1610146
Agencourt AMPure XP Beads	Beckman Coulter	A63880	DNA clean-up beads – Section 7.2.2
Amicon Ultra-0.5 Centrifugal Filter Unit	Merck	UFC501024	cut-off 10 kDa
Amicon Ultra-0.5 Centrifugal Filter Unit	Merck	UFC510024	cut-off 100 kDa
Boric Acid	Sigma-Aldrich	100165
C1000 Touch Thermal Cycler with Dual 48/48 Fast Reaction Module	BioRad	1851148
Calcium Chloride	Sigma-Aldrich	C4901
CFX Connect Real-Time PCR Detection System	BioRad	1855201
Digital Dry Baths/Block Heaters	Thermo Scientific	88870001
Dynabeads MyOne Streptavidin C1	Thermo Fisher	65001	streptavidin-modified magnetic beads – Section 4.9
EDTA disodium salt	Sigma-Aldrich	324503
Eppendorf Safe-Lock microcentrifuge tubes	Sigma-Aldrich	T9661	1.5 mL
Lenti-X p24 Rapid Titer Kit	Takara Bio USA, Inc.	632200	Lentivirus quantification kit – Section 3.3.2.1
MagJET Separation Rack, 12 x 1.5 mL tube	Thermo Scientific	MR02
Magnesium chloride	Sigma-Aldrich	M8266
Microseal 'B' PCR Plate Sealing Film, adhesive, optical	BioRad	MSB1001	non-UV absorbing
Mini-PROTEAN Tetra Cell for Ready Gel Precast Gels	BioRad	1658004EDU
Mini-PROTEAN Short Plates	BioRad	1653308
Mini-PROTEAN Spacer Plates with 0.75 mm Integrated Spacers	BioRad	1653310
Molecular Biology Grade Water	Lonza	51200
Multiplate 96-Well PCR Plates, high profile, unskirted, clear	BioRad	MLP9611
Nanodrop One	Thermo Scientific	ND-ONE-W
OneTaq DNA Polymerase	New England BioLab	M0480S
Ovation Ultralow v2 + UDI	Tecan	0344NB-A01	High-troughput sequencing library preparation kit – Section 7.2.
PIPETMAN G (100-1000 µL, 20-200 µL, 2-20 µL and 0.2-2 µL)	Gilson	F144059M, F144058M, F144056M, F144054M
Purifier Logic+ Class II, Type A2 Biosafety Cabinets	Labconco	4261
Qubit dsDNA BR Assay Kit	Invitrogen	Q32850	fluorescence-based  dsDNA quantification  kit – Section 7.2.3
SHARP Classic Low Retention Pipet Tips (10 uL, 200 uL, 1000 uL)	Thomas Scientific	1158U43, 1159M44, 1158U40
Sodium acetate	Sigma-Aldrich	S2889
Sodium chloride	Sigma-Aldrich	S7653
Sorvall Legend Micro 17R Microcentrifuge	Thermo Scientific	75002440
SsoFast EvaGreen Supermix	BioRad	1725201	qPCR mastermix – Section 6.2.
Tris(hydroxymethyl)aminomethane	Sigma-Aldrich	T1503
Tubes and Ultra Clear Caps, strips of 8	USA scientific	AB1183	PCR tubes
Urea	Sigma-Aldrich	U5128