Production, Purification, and Quality Control for Adeno-associated Virus-based Vectors

Table 1: Composition of the required solutions.

1. Tri-transfection of HEK293T Cells

NOTE: Please refer to Table 1 for the composition of the buffers and solutions used in the protocol.
NOTE: Performance of this section of the protocol takes approximately 4 days.

1. Thaw a vial of human embryonic kidney (HEK) 293T cells in a water bath set at 37 °C.
   NOTE: Use only cells that have been passaged less than 20x to guarantee optimal transfection efficiency.
2. Seed HEK293T cells at a density of 2 x 10³ to 6 x 10³ cells/cm² in DMEM10 in 15 cm diameter cell culture dishes.
3. Grow the cells to 70–80% confluence in a standard incubator set at 37 °C, with 95% humidity and 5% CO₂.
   NOTE: The production of one batch of AAV vectors using this protocol requires 18 cell culture dishes (15 cm diameter). A cell confluence of 70% – 80% corresponds to 6 x 10³ to 7 x 10³ cells/cm², maintained in 17–20 mL of DMEM10 culture medium.
4. Prepare polyethylenimine (PEI)/DNA mix at a concentration ratio of 1/3.5 (w/w).
   1. Prepare the DNA mix for 18 cell culture dishes in one 50 mL conical tube by mixing 360 µg of pΔF6, 180 µg of pCapsid, and 180 µg of pTransgene in 18 mL of 150 mM NaCl.
   2. Distribute the DNA mix over three 50 mL conical tubes (6 mL of DNA mix per conical tube).
   3. Prepare the PEI mix for six cell culture dishes in a new 50 mL conical tube by mixing 840 µg of PEI (1 µg/µL) in 6 mL of 150 mM NaCl.
   4. Prepare the PEI/DNA mix by adding 6 mL of the PEI mix (prepared in step 1.4.3), drop by drop, to one of the conical tubes containing the DNA mix (prepared in step 1.4.2) and incubate for 20 min at room temperature.
      NOTE: After 20 min of incubation, the PEI/DNA mix will become slightly turbid.
5. Take six cell culture dishes out of the incubator and completely aspirate the medium from each culture dish in a laminar flow hood. Remove the traces of the medium by rinsing the dishes with 5 mL of prewarmed Dulbecco’s phosphate-buffered saline (DPBS).
6. Ensure the distribution of DPBS over the entire surface by gently tilting the dish.
7. Gently aspirate the DPBS and add 12 mL of DMEM1 to each dish.
   NOTE: Avoid detaching the cells by adding the medium slowly, from a pipette placed at the edge of the dish.
8. Mix the PEI/DNA by pipetting up and down 3x – 5x. Add 2 mL of the PEI/DNA mix to each of the six cell culture dishes in a drop-by-drop fashion, carefully distributing it over the entire surface. Once the mix is added to each dish, place the dishes back in the incubator. Repeat steps 1.4.3– 1.8 for the remaining culture dishes.
9. Incubate the transfected cells for 5 h at 37 °C, with 95% humidity and 5% CO_2.
10. Add an additional 12 mL of DMEM10 to each culture dish without removing the pre-existing medium (total medium volume = 25 mL).
11. Incubate the transfected cells up to 72 h post-transfection at 37 °C, with 95% humidity and 5% CO₂.

Supplementary Figure 1: HEK 293T cell morphology visualized by phase contrast microscopy (left) with confirmation of GFP expression visualized by fluorescence imaging (right). (A) Successful transfection of HEK293T cells with a GFP-encoding pTransgene is confirmed by fluorescence imaging. (B) HEK293T cells treated solely with transfection reagents show no GFP expression. Please click here to view a larger version of this figure.

12. Harvest the medium and the cells 72 h post-transfection. Use a cell scraper to carefully detach the cells from the culture dish. Collect the medium and the cells in a 50 mL conical tube kept on ice.
    NOTE: The contents of two cell culture dishes can be collected into a single 50 mL conical tube. At the end of this step, each of the nine tubes will contain approximately 50 mL of medium.
13. Centrifuge the conical tubes at 420 x g for 10 min at 4 °C with the acceleration and deceleration of the centrifuge set to maximum.
14. Carefully discard the supernatant from each tube. Do not use pipettes to prevent the loss of material. Instead, gently pour the supernatant into a waste disposal container and place the tubes containing the cell pellets back on the ice.
15. Resuspend each cell pellet in 2 mL of lysis buffer directly in the 50 mL conical tube by pipetting up and down 5–10x. Do not vortex. Pool the lysates from three tubes together.
    NOTE: At this point, there will be three 50 mL conical tubes, each one containing 6 mL of the resuspended cells in lysis buffer.

2. AAV Vector Purification

NOTE: Performance of this protocol section takes approximately 1 day. Perform the following steps simultaneously on each of the three 50 mL conical tubes containing the cells resuspended in lysis buffer (see the previous note).

1. Freeze and thaw the resuspended cells 3x to lyse them and release the AAV particles. Perform the freezing step by placing the tubes in a bucket containing dry ice mixed with ethanol. Perform thawing by immediately placing the cells in a water bath set at 37 °C.
2. After the third thawing step, centrifuge at 1,167 x g for 15 min at 4 °C.
   NOTE: A noticeable pellet composed of cell debris will be formed. When handling the tubes, avoid sudden movements as these can cause the detachment of the pellet into the supernatant, which will compromise the purity of the final vector.
3. Carefully transfer the supernatants to clean 50 mL conical tubes and then add nuclease to each tube, to a final concentration of 50 U/mL supernatant.
4. Incubate for 30 min at 37 °C. Swirl the 50 mL conical tubes by hand every 10 min to ensure that the nuclease is thoroughly mixed with the supernatant.
5. Clarify the supernatant by centrifugation at 13,490 x g for 20 min at 4 °C.
6. Attach a 0.45 µm filter to a 10 mL syringe and place it on top of a clean 50 mL conical tube. Carefully remove the plunger and fill the syringe with the supernatant from step 2.5.
7. Use the plunger to force the lysate through the filter. Use a new filter and syringe for each tube of supernatant obtained in step 2.5.
   NOTE: The obtained fraction is known as ‘crude lysate’.
8. Prepare each of the 15%, 25%, 40%, and 60% iodixanol fractions in four separate 50 mL conical tubes, according to the instructions in Table 1.
9. Prepare the iodixanol gradients in three ultracentrifugation tubes using the following order of iodixanol fractions: 8 mL of 15% iodixanol, 5.5 mL of 25% iodixanol, 5 mL of 40% iodixanol, and 4.5 mL of 60% iodixanol.
   CAUTION: Iodixanol can cause irritation to eyes, skin, and the gastrointestinal and respiratory tracts. When handling iodixanol gradients, wear gloves and work under a laminar flow hood.
   1. Pipette 8 mL of 15% iodixanol into each ultracentrifugation tube.
   2. Pipette 5.5 mL of 25% iodixanol solution into a clean 50 mL conical tube (Figure 1A).
   3. Use a non-graduated Pasteur pipette (as the neck of the ultracentrifugation tube is too narrow for conventional graduated pipettes) to carefully layer 5.5 mL of the 25% iodixanol solution below the 15% iodixanol solution (Figure 1B).
      NOTE: This can be successfully achieved by adding the iodixanol in three steps since the Pasteur pipette can only hold around 2 mL.
   4. Add the 40% and 60% iodixanol solutions as described in step 2.9.3. Do not disturb the different iodixanol interfaces during layering.
      NOTE: The proper preparation of the different iodixanol fractions can be ensured by visual confirmation thanks to the phenol red added to the iodixanol fractions (see the instructions in Table 1). Since each fraction has a specific density, they will not intermix during the layering step, if it is performed properly (check Figure 1C).
10. Layer the crude lysate on top of the 15% iodixanol gradient with a Pasteur pipette. Proceed drop by drop to avoid disturbing the interface between the crude lysate and the iodixanol solution.
11. Fill the ultracentrifugation tube up with lysis buffer until the meniscus reaches the base of the tube neck to ensure the tube does not collapse under the very high forces generated during the ultracentrifugation (Figure 1C).
12. Close the ultracentrifugation tubes using appropriate lids. Use a digital scale to make sure all three ultracentrifugation tubes have the same weight. Adjust the weight, if necessary, by adding more lysis buffer on top of the ‘crude lysate’.
    NOTE: The weight difference between the ultracentrifugation tubes must be below 0.1 g to ensure the safe operation of the ultracentrifuge. Ultracentrifuges are potentially dangerous pieces of equipment and should only be used by properly trained personnel.
13. Centrifuge the tubes at 301,580 x g, using a fixed-angle titanium rotor, for 1 h and 40 min at 12 °C, using maximum speed of acceleration and deceleration.
14. Carefully insert a stainless-steel blunt needle at the 40% and 60% iodixanol interface.
    1. Attach the needle to a 5 mL syringe (Figure 1E).
    2. Aspirate only the clear fraction, containing the vector particles.
       NOTE: The total volume collected is approximately 2.5–3 mL. Avoid the collection of material from the 25–40% interface, since this will increase the level of contamination in the final vector batch, due to the presence of non-desirable proteins.
    3. Process the collected fraction (desalting and concentration) or store it overnight at 4 °C.

Figure 1: Setup for iodixanol gradient purification and subsequent vector collection. (A) Before pipetting the different iodixanol gradients into the ultracentrifugation tube, pipette an adequate volume of each iodixanol solution into a separate conical tube. (B) Then use a Pasteur pipette to sequentially transfer each iodixanol solution to the ultracentrifugation tube: layers of an increasingly high iodixanol concentration should be added at the bottom of the tube underneath the previous layer(s). (C) Layer the crude vector lysate on top once the gradient has been prepared. This vector collection system does not use sharp needles, which present a risk of 'needlestick' injuries. (D) A stainless-steel 316-syringe needle is inserted through the iodixanol gradient up to the 40%/60% iodixanol interface. (E) Vector particles are found in the 40% iodixanol phase and are collected. Please click here to view a larger version of this figure.

3. Desalting and Concentration of the AAV Vector

NOTE: Performance of this section of the protocol takes approximately 2 h.

1. Prerinse the filter membrane of the centrifugal filter by adding 5 mL of 1x PBS-MK and centrifuge for 5 min at 4,000 x g.
2. Add 5 mL of 1x PBS-MK to the collected AAV vector fraction, mix and then add the total volume to the filter. Upon addition of 1x PBS-MK, turbidity is observed. Centrifuge at 4,000 x g until the volume present on the cone-shaped filter has been reduced to 1 mL. Discard the flow-through accumulated in the outer collection tube.
3. Add 13 mL of 1x PBS-MK to the cone-shaped filter and repeat the centrifugation step as many times as necessary (minimum 3x), until there is no more turbidity observed when fresh 1x PBS-MK is added.
4. In the final wash step, add 13 mL of PBS + 0.01% (v/v) non-ionic surfactant and centrifuge at 4,000 x g until the volume is reduced to 300–350 µL.
5. Collect the remaining fraction present on the filter by pipetting the whole volume into a sterile skirted microcentrifuge tube.
   NOTE: This fraction contains the desalted and concentrated AAV vector. In the following steps of this protocol, this fraction is referred to as the ‘primary’ fraction. Make sure to handle the tube under the hood and store it at 4 °C.
6. Rinse the filter with 200 µL of 1x PBS-MK to collect the residual vector particles retained on the filter. Collect in a sterile microcentrifuge tube.
   NOTE: This is a diluted vector stock, which may be suitable for some applications requiring lower vector titers. In future steps, this fraction is referred to as the ‘secondary’ fraction.
7. For short-term storage, store the vector fraction(s) at 4 °C (less than 2 weeks). Aliquot and store the vector at ‑20 °C when long-term storage is required.
8. Perform quality control as described in section 4.

4. Titration of the Vector by Quantitative Polymerase Chain Reaction

NOTE: Performance of this section of the protocol takes approximately 3 h.

1. Standard curve
   1. Linearize the transgene-expressing plasmid (pTransgene) used for the transfection of HEK-293T cells in section 1.
   2. Prepare the restriction digest mix in a 0.5 mL microcentrifuge tube (refer to Table 2 for the composition of the restriction digest mix).
      NOTE: Adjust the composition of the restriction digest mix according to the enzyme used and the related guidelines from the manufacturer.
   3. Incubate the restriction digest mix for 1 h at 37 °C.
   4. Check the efficiency of the restriction digestion by running the linearized plasmid on a 0.8% (w/v) agarose gel at 100 V for 1 h. Complete digestion of the plasmid should produce a single fragment of defined size.
   5. Purify the linear plasmid DNA using a PCR purification kit, following the manufacturer’s instructions²⁴, and measure the DNA concentration with a spectrophotometer by measuring the absorption at 260 nm. After the titration, store the remaining aliquot of the linear plasmid at ‑20 °C for further use.
   6. Calculate the molecules (i.e. DNA copies) of the plasmid stock per microliter as follows.
      1. First, calculate the molecular weight of the plasmid. Assuming that the average weight of a DNA base pair (bp) is 650 Daltons (Da) and that one mole of a bp weighs 650 g, the molecular weight of any double-stranded DNA template can be estimated by taking the product of its length (in bp) and weight per base pair.
         ​Plasmid molecular weight [g/mol] = plasmid size [bp] x 650 [Da/bp]
      2. Next, calculate the number of moles of plasmid per microliter.
         Moles of plasmid per microliter = plasmid concentration [g/µL] / plasmid molecular weight [g/mol]
         ​NOTE: The inverse of the molecular weight is the number of moles of plasmid present in 1 g of the material.
      3. Then, calculate the number of plasmid molecules per microliter, using Avogadro’s number (6.022 x 10²³ molecules/mole).
         Molecules of plasmid per microliter = moles of plasmid per microliter [moles/µL] x Avogadro’s number [molecules/mole]
      4. Finally, dilute the plasmid stock (molecules/µL) to obtain a 100 µL solution with the desired concentration of 1 x 10⁹ molecules (vector genomes (vg) per microliter).
         Plasmid stock (100 µL) = (desired concentration [molecules/µL] x 100 µL) / plasmid molecules [molecules/µL]
   7. Make serial dilutions of the plasmid stock (1 x 10⁹ vg/µL) in triplicates:
      10 µL of 1 x 10⁹ vg/µL plasmid stock + 90 µL of H₂O = 1 x 10⁸ vg/µL solution
      10 µL of 1 x 10⁸ vg/µL dilution + 90 µL of H₂O = 1 x 10⁷ vg/µL solution
      10 µL of 1 x 10⁷ vg/µL dilution + 90 µL of H₂O = 1 x 10⁶ vg/µL solution
      10 µL of 1 x 10⁶ vg/µL dilution + 90 µL of H₂O = 1 x 10⁵ vg/µL solution
      ​Continue to obtain a 1×10¹ vg/µL solution.
   8. Keep the serial dilutions of the standard plasmid stock on ice until loading them on the qPCR plate (section 4.3).

Table 2: Restriction digest mix composition.

Table 3: Stock plasmid volume calculator. Please click here to download this table.

2. DNA extraction from the AAV vector
   1. Mix 2 µL of the AAV vector stock (the primary fraction from step 3.5) with 198 µL of DNase I buffer (1x) in strip tubes (PCR tubes) and add 2 µL of DNase I.
      NOTE: DNase I will degrade any genetic material that is not contained inside a vector capsid (which would distort the qPCR results). This solution is referred to as dilution ‘dil1 x 10^-2’.
   2. Incubate for 30 min at 37 °C, followed by 10 min at 95 °C.
      NOTE: The protocol can be stopped at this point and the material can be stored indefinitely at 4 °C, to avoid product deterioration.
   3. Add 2 µL of proteinase K to the dil1 x 10^-2 solution (step 4.2.1) and incubate for 60 min at 50 °C, followed by 20 min at 95 °C.
      NOTE: This step will disassemble the AAV vector capsid and release the AAV vector genome into the solution. Add proteinase K in excess as the protein (capsid) content of the sample is not known. Note, it is essential to ensure that all proteinase K activity is removed by denaturation, prior to qPCR, to avoid issues with (partial) polymerase degradation influencing the final result.
   4. Prepare 1:10 serial dilutions of the proteinase K treated dil1 x 10^-2 solution (step 4.2.3) in 1.5 mL microcentrifuge tubes as follows:
      10 µL of dil1 x 10^-2 + 90 µL of H₂O = dil1 x 10^-3 dilution
      10 µL of dil1 x 10^-3 + 90 µL of H₂O = dil1 x 10^-4 dilution
      ​10 µL of dil1 x 10^-4 + 90 µL of H₂O = dil1 x 10^-5 dilution
   5. Keep the serial dilutions of DNA extracted from the vector on ice until loading them on the qPCR plate (section 4.3).
3. Titration by SYBR Green detection-based qPCR
   1. Prepare the qPCR master mix in a 1.5 mL microcentrifuge tube, for both the sample and the standards. Use 10 µL of SYBR Green Master Mix, 1 µL of forward primer (10 µM stock), 1 µL of reverse primer (10 µM stock), and 3 µL of H₂O per reaction. See the protocol in Table 4 for primer sequences.
   2. Pipette the qPCR master mix up and down, but do not vortex.
   3. Add 15 µL of the qPCR master mix, followed by either 5 µL of the standard curve prepared in section 4.1 or the DNA extracted from the AAV vector in section 4.2, into each well. Include three wells containing only the qPCR master mix, as a negative control. Refer to Figure 2 for the plate layout.
   4. Seal the plate with a sealing film and briefly centrifuge the qPCR plate at 1,500 x g for 30 s at 4 °C.
   5. Run the qPCR reaction on a plate-based real-time PCR amplification and detection instrument, using the conditions suggested in Table 5.

Table 4: Primer sequences designed against the CBA promoter.

Figure 2: Plate layout for qPCR-based vector titration. The samples are color-coded: green = standard curve; blue = H₂O control; grey = primary fraction; orange = secondary fraction. Please click here to view a larger version of this figure.

Table 5: Thermal cycling protocol for SYBR green-based qPCR titration.

4. Data analysis to determine the AAV vector titer
   1. Fill the spreadsheet data cells (Table 6A) with the Ct values obtained for the different dilutions of the standard sample prepared in section 4.1 to generate a standard curve.
      ​NOTE: The equation of the standard curve will be shown (y = a ln(x) + b) together with the R² efficiency (Table 6B). A qPCR must have an efficiency close to 100% and R² close to 1.0 (≥ 0.96).
   2. Use the calculated values of a and b to fill in the corresponding data cells in the spreadsheet (Table 6C).
   3. Complete the spreadsheet with the Ct values obtained for the different dilutions of the AAV sample prepared in section 4.2.
   4. Calculate the average AAV vector titer in vector genomes per microliter of the 'primary fraction' by using the following formula:
      
      NOTE: The factor 400 is used for single stranded genomes, while sc genomes use a multiplication factor of 200²⁵. In fact, as DNA quantities double during each qPCR cycle, sc DNA is detected 2x in comparison to an ss genome. The aim of the titration is to calculate the concentration in terms of vector genomes per microliter. A correction is necessary for running the qPCR when using 2 µL of starting material (step 4.2.1). dil represents the dilution factors from step 4.2.4. The titer of the 'secondary' AAV vector fraction (step 3.6) can be calculated simultaneously.

Table 6: Template for qPCR data analysis. Please click here to download this file.

5. Purity Control by SDS-PAGE and Silver Staining

NOTE: Performance of this section of the protocol takes approximately 5 h.

1. Use 70% (v/v) ethanol to thoroughly clean the glass plates used for casting the gels.
2. Assemble the gel casting system. Ensure that the bottoms of the glass plates are not chipped, to prevent leakage of the acrylamide mixture when casting the gel.
3. Prepare the stacking and the resolving gel solutions in two separate 50 mL conical tubes. Omit the tetramethylethylenediamine (TEMED) and the 10% (w/v) ammonium persulfate (APS), as they are responsible for polymerization of the gel. Add them immediately before casting the gel.
   NOTE: The final percentage of acrylamide in the gel influences the separation profile of the proteins: typically, a 10% (v/v) final concentration of acrylamide will be sufficient to test the purity of an AAV vector preparation.
4. Mix gently by swirling the tube containing the gel components. Do not vortex, since excessive oxygenation can impair polymerization.
5. Add TEMED and 10% (w/v) APS to the resolving gel solution. Mix and then pour into the gel holder until it reaches 1-2 cm below the top of the small glass plate.
6. Place a layer of water-saturated butanol on the top of the acrylamide mixture. This will ensure the formation of a flat surface during polymerization. Do not leave the gel under alcohol for longer than 30 min as this will dehydrate the gel and impair its function.
   CAUTION: Butanol is hazardous in case of skin contact. It also presents a fire risk. Handle with care.
7. Wait for the gel to polymerize and then pour off the butanol. Tip: Check to see whether the excess gel mix in the tube has solidified. Polymerization occurs in less than 20 min. Wash the surface of the gel with H₂O and then dry it using paper towels, taking care not to disturb the surface of the gel.
8. Add TEMED and 10% (w/v) APS to the stacking gel and pour the gel into the gel holder on top of the separating gel.
9. Place the provided comb in the gel. Perform this action with one steady movement to avoid the formation of air bubbles inside the wells.
10. Wait for at least 20 min for the gel to polymerize. Tip: Check if the excess gel mix in the conical tube has solidified.
11. Prepare the two mixes (Table 7) for both the primary and the secondary AAV vector fractions (steps 3.5 and 3.6, respectively).

Table 7: Composition of the sample mixes required for silver staining.

12. Fully denature the sample mixes by heating them for 5 min at 95 °C. Assemble the electrophoresis tank, paying attention to the electrode orientation.
13. Fill the tank with 1x cathode buffer on the inside of the electrode assembly and 1x anode buffer on the outside.
    NOTE: Anode buffer can be recycled from run-to-run. Fresh cathode buffer must always be used.
14. Remove the comb from the gel and clean the wells of the gel with 1x cathode buffer.
15. Load 1 µL of protein ladder in a well. Load the sample mixes in different wells on the same gel (Figure 3). Run the gel at 50 V until the samples enter the resolving gel. Then increase the voltage to 100 V until the dye front reaches the lower limit of the gel.
16. Extract the gel carefully from the glass plates and use the silver staining kit (according to the manufacturer’s instructions²⁶) to visualize the viral proteins (VP1, VP2, and VP3 subunits) that comprise the AAV capsid, as well as to check for possible protein contamination (Figure 3).

Figure 3: Evaluation of vector purity using SDS-PAGE and silver staining. Using a Tricine-SDS gel, 5 µL of various vector preparations were separated. Proteins were subsequently detected by silver staining. Vectors are considered pure when VP1 (82 kDa), VP2 (67 kDa), and VP3 (60 kDa) are visible in a 1:1:10 ratio (lane 1), without excessive background (lane 2) or non-specific bands (lane 3). Please click here to view a larger version of this figure.