Quantitative and Digital Droplet-Based AAV Genome Titration

Abstract

The adeno-associated viral vector (AAV) platform has developed into a primary modality for efficient in vivo, and in more limited settings, in vitro or ex vivo gene transfer. Its applications range from a tool for experimental purposes to preclinical and clinical gene therapy. The ability to accurately and reproducibly quantify vector concentration is critical for any of these applications. While several quantification assays are available, here we outline a detailed protocol for the quantification of DNase-I protected vector genomes reliant on the polymerase chain reaction (PCR) as a measure of the active component of the vector, namely its transgene cargo. With the emergence of droplet digital PCR (ddPCR), we provide side-by-side protocols for traditional TaqMan™ real-time, quantitative PCR (qPCR) and ddPCR, as well as comparative data generated with both methods. Lastly, we discuss the importance of the use of surfactant (here, Plutonic® F-68) in the execution of the assay to limit DNA and AAV adherence to various carriers during the titration, particularly at low concentrations. We believe these protocols can lead to reduced variability and increased comparability between AAV studies.

1. Introduction

Recombinant AAV is a complex biologic derived from a nonpathogenic virus that is rendered replication-defective through elimination of all viral open reading frames [1], It is composed of a proteinaceous viral capsid and a single-stranded DNA (ssDNA) vector genome. The vector genome is fully user-defined to deliver transgenic content to a target cell, although the transgene cargo must be flanked by inverted terminal repeats (ITRs), the only element retained from the native virus in the vector genome. In most applications, these ITRs are derived from AAV serotype 2 [2]. These AAV2-based ITR genomes can be packaged into a plethora of natural or engineered AAV capsids [3–5]. Several methods to produce AAV are established that result in vector particles of high concentration and purity, as described in Chapters 3, 7, 19, 21, 22, and 23 [6].

Following production, AAV preparations traditionally undergo various assessments of quantification and qualification, such as vector titration. Titration is a process aimed at quantifying the number of vector particles in a given volume. For AAV, as for many viruses, various assays are available, many of which rely on different principles or measure distinct subpopulations. Specifically, infectious titers determine the concentration of infectious units, which is dependent on the specific conditions of the infectivity assay. Physical titers are analytical measures of the number of physical particles within a set volume. Physical titers for AAV can differ based on the principle that is relied on or the component of the particle that is probed. These include (1) measures of the number of genome containing particles determined by probing for components of the vector genome, (2) measures of the number of assembled capsids as determined by probing for the viral capsid protein, (3) calibrated methods identifying unique physical properties of AAV particles [7–9], and (4) automated imaging methodologies relying on a particle-by-particle assessment and counts. These methods each have their own benefits and limitations, and consequently each may be appropriate depending on the question asked. For example, an assessment of the total amount of assembled capsids is relevant for evaluating safety, and thus in this case measurements of viral capsid protein are preferred.

For most applications, however, the active agent in AAV is considered to be the vector genome, irrespective of its use in experimental laboratory or preclinical, translational, or clinical settings. For this reason, most dosing is based on titration methods probing the viral genome, leading to a determination of viral genomes (vg) or genome containing particles (GC) within a given volume. Genome titration is therefore the focus of the protocols below, which leverage the polymerase chain reaction (PCR). Here, we outline and detail a workflow schematically depicted in Fig. 1, leading to the determination of titer using TaqMan™ quantitative PCR (qPCR) or droplet digital PCR (ddPCR) [10]. These methods have been described in various levels of detail previously [11, 12]; however, here we further optimize and compare these different methods. Previous studies have highlighted the potential for large discrepancies among the results of different laboratories using analogous protocols [13]. This protocol aims to provide minute detail to ensure a greater level of standardization across the AAV field.

Fig. 1.

Protocol schematic and workflow

2. Materials

1. Nuclease-free water (see Note 1).

2. GeneAmp® 10× PCR Buffer I (Thermo Fisher).

3. Sheared Salmon Sperm (SSS) DNA (Thermo Fisher) (see Note 2).

4. Certified DNase- and RNase-free 0.65 mL microcentrifuge tubes, clear.

5. Certified DNase- and RNase-free 0.65 mL microcentrifuge tubes, colored.

6. Certified DNase- and RNase-free 1.7 mL microcentrifuge tubes, clear (see Notes 3 and 4).

7. L20, L200, and L1000 manual micropipettes.

8. E3–20 and E3–200 LTS electronic pipettes (Rainin Instruments) or electronic equivalent.

9. Electronic pipette rapid charge stand.

10. Sterile low-retention filter micropipette tips.

11. Pipet-Lite™ Multi Pipette L8- or L12–50XLS+ (Rainin Instruments) or electronic equivalent (see Note 5).

12. Compact PCR Racks for 0.2 mL tubes and strips.

13. 0.2 mL PCR 8-tube strips with attached clear flat caps.

14. Microcentrifuge tube racks.

15. Dedicated Microcentrifuge.

16. Dedicated 8-well PCR tube strip mini centrifuge.

17. DNase-I: RNase-free with 10× incubation buffer (Roche) (see Note 6).

18. 100× Plutonic® F-68 surfactant (10% solution) (Thermo Fisher) (see Notes 7–9).

19. 10× Plutonic F-68 (1% solution) diluted in nuclease-free water, prepared fresh each time.

20. 100× SSS DNA (200 ng/μL solution) in 1× PCRBuffer I and 0.1% Plutonic® F-68.

21. 10× SSS DNA (20 ng/μL solution) in lx PCR Buffer I and 0.1% Plutonic® F-68.

22. AAV sample dilution buffer: lx PCR Buffer I, 2 ng/μL SSS DNA, 0.1% Plutonic® F-68 (see Note 10).

23. Blank buffer (Mock negative control): PBS, 35 mM NaCl, 0.001% Plutonic F-68, or nuclease-free water.

24. AAV validation sample for quantitative positive control (see Note 26).

25. DNase-I linear plasmid control (1E+8 copies/5 μL), serially diluted from stock using AAV sample dilution buffer.

26. AAV experimental samples.

27. Forward sequence detection primer working stock solution: 9 μM for ddPCR, 3 μM for qPCR (see Notes 11 and 12).

28. Reverse sequence detection primer working stock solution: 9 μM for ddPCR, 3 μM for qPCR (see Notes 11 and 12).

29. TaqMan™ fluorescent probe working stock solution: 2.5 pM for ddPCR, 2 μM for qPCR (see Note 13).

30. Applied Biosystems SDS7500 Sequence Detector (Thermo Fisher), or equivalent.

31. Restriction enzymes with reaction buffer.

32. Agarose gel with Ethidium Bromide or equivalent.

33. QIAquick PCR Purification Kit (QIAGEN) with EB Buffer or Plasmid Purification Kit.

34. Linearized plasmid standard set, 1E+8 to 1E+1 copies/5 pL, serially diluted from stock in AAV dilution buffer.

35. Calibrated spectrophotometer (see Note 14).

36. Qubit™ Fluorometer (Invitrogen), or equivalent.

37. Qubit™ dsDNA HS Assay Kit (Invitrogen), or equivalent.

38. 2× ddPCR™ Supermix for Probes, no dUTP (Bio-Rad).

39. 2× TaqMan™ Universal Master Mix for qPCR (Thermo Fisher).

40. Auto DG™ Automated Droplet Generator (Bio-Rad) [14].

41. Automated Droplet Generation Oil for Probes (Bio-Rad).

42. DG32™ AutoDG™ Cartridges (Bio-Rad).

43. AutoDG™ droplet cooling box.

44. QX100™or QX200™ ddPCR System (Droplet Reader) and QuantaSoft™ Software (Bio-Rad) [15, 16].

45. C1000 Touch™ Thermocycler (Bio-Rad), or equivalent.

46. PX1 PCR Plate Sealer and corresponding Plate Support Block (Bio-Rad).

47. Pierceable Foil Heat Seal (Bio-Rad).

48. ddPCR 96-well plates (Bio-Rad) or equivalent.

49. Optical 96-Well Reaction Plate for qPCR.

50. Optical Adhesive Film Kit for qPCR (Thermo Fisher).

51. Splash-Free 96-Well Base (Thermo Fisher).

52. TempPlate® sealing film, non-sterile (USA Scientific).

53. Corning™ Costar™ 50 mL Sterile Disposable Reagent Reservoirs (Fisher Scientific), or equivalent.

54. Powder-free Nitrile Gloves.

55. Clean Lab Coat.

56. BSL-2 Biological Safety Cabinet with UV light.

57. 15 mL conical tubes.

58. Kimwipes™ Delicate Task Wipers, or equivalent.

59. Paper towels.

60. Wet ice and ice bucket.

61. Timer.

62. Vortex Mixer.

63. Lysol.

64. 70% Ethanol.

65. Lab notebook and pen.

66. Adhesive tape.

67. Color markers, fine and extra-fine point.

3. Methods

3.1. Preparation of the Work Area (See Note 15)

1. Work in a biosafety cabinet or HEPA-filtered PCR station with UV light capability, preferably a dedicated unit for quantification of high copy number targets.

2. Wear gloves at all times and change them often, particularly after leaving (and returning to) the work area.

3. Apply a liberal amount of Lysol or equivalent to the work surface area inside the cabinet and wipe it clean with a paper towel. Do the same again with 70% Ethanol.

4. Apply a small amount of Lysol or equivalent to a paper towel and wipe clean the microcentrifuge tube racks that will be used. Repeat with 70% Ethanol. Place the clean racks well inside the clean biosafety cabinet.

5. Apply a liberal amount of Lysol or equivalent to pipette shafts (manual or electronic) over a biohazard trash bag. Repeat with 70% ethanol and wipe the pipette shafts dry with a lint-free Kimwipes™. Place the pipettes in the clean biohazard cabinet.

6. Protect the LED screen of electronic pipettes with aluminum foil while the units are exposed to UV light. Exposure to UV light will crack LED screens over time (Fig. 2).

7. Turn on the UV light for 15 min (see Note 16).

Fig. 2.

Protect the LED screens of electronic pipettes from UV light damage

3.2. Preparation of the Real-Time PCR (qPCR) Standard

1. Prepare the real-time PCR standard in a biological safety cabinet or PCR station, rather than on the bench, where reagents or standards are more likely to become contaminated. Use dedicated reagents.

2. The standard is usually a cis-plasmid (i.e., ITR-flanked transgene construct) used in the manufacturing of the virus. There are two conditions that need to be met in advance of standard preparation: the plasmid must be a clean, clonally pure species (OD 260/280 between 1.80 and 1.90) and an accurate electronic map must be available (see Note 17).

3. Confirm the clonal purity of the plasmid by performing an analytical restriction digestion assay of the DNA. Ideally, the assay will target both the recombinant AAV genome embedded in the plasmid and the plasmid backbone. Set up several reactions and choose enzymes that will target several components of the recombinant genome.

4. Linearize the plasmid with 10–20 units of enzyme in the appropriate reaction buffer, preferably overnight. Digest 10 pg of plasmid per 50 μL reaction. For long-term use, 40 pg of linearized plasmid will last for months or years if properly stored (see Note 18).

5. Use 1% of each reaction (0.5 μL) to verify the efficiency of linearization in separate wells of an agarose gel. There must be a single band per lane in the gel. Run appropriate controls, as needed, side by side.

6. Clean up the linearization reactions using QIAGEN’s QIAquick® PCR Purification Kit (or a plasmid purification kit). Elute twice with 50 μL of EB Buffer in a DNase- and KNase-flee 1.7 mL tube. Pool the eluates into a single tube and label (see Note 19).

7. Quantify the pooled linearized plasmid using spectrophotometry or fluorometry, but preferably both (see Note 20).

8. Determine the copy number concentration of the linearized plasmid and the volume needed for the first standard dilution (Fig. 3). Round off all the estimated volumes for pipetting to two decimal points. A final concentration of 0.1% Plutonic® F-68 must be included at the outset in the first dilution of the standard (lE+10 or 1E+09 copies per 5 μL, depending on the yields from steps 6 and 7). The “1E+10” and/or the “1E+09” dilution constitute the working standard stock solutions of the linearized plasmid (see Notes 21–23).

9. Serially dilute the working stock oflinearized plasmid standard (lE+10 or 1E+09 copies/5 μL) in tenfold dilutions to 1E+01 copies/5 μL in AAV sample dilution buffer. Vigorously vortex each previous dilution before moving on to the next, and always rinse the tip carrying the plasmid by pipetting up and down 5–10 times after addition of plasmid to diluent (Fig. 4) (see Note 24).

10. Discard 10 μL from the “1E+01 copies/5 μL” dilution to keep the volumes of the standard set equal across all tubes. The “1E+08 copies/5 μL” through the “1E+01 copies/5 μL” dilutions comprise the standard set to be used downstream for quantification purposes in real-time PCR.

11. From the “1E+09” working stock dilution, prepare an additional “1E+08 copies/5 μL” dilution of the linear standard in AAV dilution buffer to serve as the DNase-I plasmid control in all future quantification runs. Store the linearized plasmid standard dilutions and stocks at −20 °C (see Note 25).

Fig. 3.

Workflow for determining linear plasmid copy number and dilution. Calculate the copy number concentration of the linearized plasmid and the volume needed for a dilution of 2E+9 copies per μL (1E+10 copies per 5 μL), based on the known length of the plasmid, “Plasmid Length (bp),” and the mass concentration of the linearized plasmid as determined in Subheading 3.2, steps 6 and 7, “measured mass concentration (g).” Choose pipettes that maximize the accuracy and precision of reagent delivery. *See Subheading 3.2, steps 6 and 7 and Notes 19 and 20. **The final concentration of the stock linear plasmid dilution may be adjusted according to the linear plasmid yields obtained. For example, this value may be reduced from 2E+09 to 2E+08 copies/μL to create a stock of 1E+09 copies/5 μL, rather than 1E+10 copies/ 5 μL, if the yield of linear plasmid is low. ***See Subheading 2 for exact formulation

Fig. 4.

Serial dilution for MV quantification. Follow the same protocol whether diluting virus or plasmid

3.3. Sample Preparation: DNase-I Treatment

1. Prepare the surface area of a dedicated biosafety cabinet as specified in Subheading 3.1, above. In addition, clean a compact PCR rack with Lysol and 70% ethanol and add to the work space.

2. Wear a lab coat and gloves for this procedure. Change gloves often, especially after leaving (and returning to) the work area.

3. Thaw and equilibrate the AAV sample dilution buffer to room temperature.

4. Pulse-centrifuge the AAV test samples and controls to collect them at the bottom of the tube (see Note 26).

5. Transfer as many 0.2 mL 8-well tube strips as needed to the compact PCR rack in the biohazard cabinet for the DNase-I treatment step. Close all the lids. The DNase-I incubation will be carried out in a programmable thermocycler that can accommodate the strips.

6. Organize the AAV test samples in ascending order followed by the controls and assign a number to them. Label the flat caps of the 0.2 mL tube strips in advance with the assigned number (see Note 27).

7. The DNase-I treatment step will be the first tenfold dilution of the AAV test samples and controls (see steps 3 and 4 in Subheading 3.4). All dilutions made going forward will be tenfold serial dilutions.

8. Prepare 10× Pluronic® F-68 (1%) reagent (5 pL per reaction). Make enough for all samples and controls, with sufficient extra volume to offset pipetting errors.

9. Prepare the DNase-I reaction mix for all samples and controls in a 1.7 mL microcentrifuge tube as follows (45 μL per reaction):

   Reagent	Volume per sample
   Nuclease-free Water	33 μL
   10× DNase Buffer	5 μL
   10× (1%) Pluronic F-68	5 μL
   DNase I, RNase-free	2 μL (20 units)
   Sample (AAV)	5 μL
   Total Volume	50 μL

   Multiply the “volume per sample” from above by the number of samples (AAV test samples and controls) to determine the total volume of reagents that will be needed for this step. Include two additional volumes in the calculation to offset pipetting errors (see Note 28).

10. Before adding DNase-I to the reaction mix, dispense 43 μL of the DNase-free reaction mix into the DNase-I plasmid control tube that will not include the enzyme. Bring to 45 μL with 2 μL of nuclease-free water (see Note 29).

11. Add the total volume of DNase-I needed for the remaining reactions to the 1.7 mL tube with the reaction mix, plus two extra volumes (see Note 30).

12. Invert the (complete) DNase-I reaction mix several times and only gently vortex it once. Dispense 45 μL to the remaining tubes in the strips, one at a time, using the multi-dispense function of an electronic 200 μL pipette. Immediately close the lid of each tube after dispensing the mix before moving on to the next one. Centrifuge the strips.

13. Add the test samples and controls to the DNase-I reactions (see Note 21). Vigorously vortex one sample or control at a time and add 5 μL to the corresponding reaction in the 8-tube strips in the compact PCRrack. Pipet up and down 5–10 times in the reaction mix to rinse the pipette tip (see Note 31).

14. Place the tube strips in the C1000 Touch thermocycler block and incubate the DNase-I reactions at 37 °C for 1 h (see Note 32).

15. Return all original samples and controls to the refrigerator or freezer (see Note 33).

3.4. Sample Preparation: Serial Dilution

1. Prepare the work area as described in Subheading 3.1.

2. Allow the DNase-I reactions to equilibrate to room temperature after taking them out of the thermocycler. Centrifuge the strips and vigorously vortex to mix.

3. AAV test samples, Mock controls, and Validation controls will be serially diluted 100,000-fold more (1,000,000-fold total) with AAV sample dilution buffer. Prepare five 0.65 mL microcentrifuge tubes for each sample or control for the serial dilution step.

4. The DNase-I plasmid controls will be diluted 10,000-fold more (100,000-fold total) with AAV sample dilution buffer. Prepare four 0.65 mL tubes for each plasmid control for the serial dilution step.

5. Set up the 0.65 mL microcentrifuge tubes so that they follow the same organization as the samples in the 0.2 mL strip tubes along the bottom length of a microcentrifuge tube rack(s).

6. For brevity, label only the last 0.65 mL tube in each dilution series (the dilution that will be used to deliver the sample to the PCR reaction).

7. Dispense 45 μL of the room temperature-equilibrated AAV sample dilution buffer to each 0.65 mL dilution tube using the multi dispense mode of an electronic 200 μL pipette (see Note 34).

8. Serially dilute each DNase-I reaction (Fig. 4).

9. Because most of the 0.65 mL tubes in the rack will not be labeled, use the empty slots in the microcentrifuge tube racks to track your progress, avoid confusion during pipetting, and minimize errors due to slapped tubes. Move each tube up a slot in the rack after it has been used to track progress until completion (see Note 27).

10. Place the most dilute samples (the labeled tubes) on ice until the next step. Repeat until all the samples or controls are diluted. Discard all unlabeled tubes. Return the initial DNase-I reactions to the freezer and do not discard them until results are obtained (see Notes 35 and 36).

11. For ddPCR, proceed to Subheadings 3.5 and 3.6. For qPCR, proceed to Subheading 3.7.

3.5. ddPCR Reaction Setup and Droplet Generation (See Note 37)

1. Prepare the work area as described in Subheading 3.1. Only one clean microcentrifuge tube rack will be needed to hold the serially diluted samples and controls, as well as the bulk reaction mixture preparations. Also include a splash-free 96-well base to hold a Bio-Rad 96-well plate (or equivalent) where the reactions will be ultimately assembled.

2. While the interior of the biohazard cabinet is exposed to UV light, create a map of the 96-well plate indicating the position of all samples and controls in each TaqMan™ assay that will be included (Fig. 5).

3. Set aside a 1.7 mL microcentrifuge tube for each TaqMan™ assay. Prepare all ddPCR reactions for each assay in a master mix as follows (18 μL per reaction):

   Reagent	Vol. per reaction (μL)
   2× ddPCR™ Supermix for Probes	10
   Forward Primer (9 μM)	2
   Reverse Primer (9 μM)	2
   TaqMan™ Fluorescent Probe (2.5 μM)	2
   Nuclease-free water	2
   Sample; (AAV or control)	2
   Final Volume	20

   Calculate the number of reactions that are needed to quantify all samples and controls in triplicate. Include additional 6–8 reactions per assay to offset pipetting error. The suggested sample order for each TaqMan™ assay is as follows: Test samples (in ascending order), DNase I (−) Plasmid Control, DNase I (+) Plasmid Control, Mock Control, AAV Validation Control, and NTC (No Template Control). A maximum of 32 unique samples may be analyzed in triplicate per plate (see Note 38).

4. Mix the equilibrated 2× ddPCR Supermix (No dUTP) by inversion and prepare the ddPCR master mix(es) according to the instructions above. Vortex gently to mix.

5. Carefully transfer 18 μL of master mix to the bottom of each reaction well following the map generated in step 2; use the multi-dispense function of a 200 pL electronic pipette (single or multichannel) (see Note 39).

6. Transfer 2 μL of each DNA sample or control to the side of each designated well using the multi-dispense function of a p20 electronic pipette and following the map generated in step 2 (see Note 21). Open one tube at a time while dispensing and keep tube lids closed when not in use.

7. After completion, seal the plate with lightly applied TempPlate® sealing film (see Note 40).

8. Start the heat-sealing program in a PX1 PCR Plate Sealer (180 °C for 5 s) and wait for it to reach maximum temperature. The program will not be able to start until maximum temperature is reached. The program will hold until the start button is pressed in step 11.

9. Spin down the plate in a dedicated PCR 96-well plate centtifuge at 226× g for 30 s.

10. Transfer the plate to the plate support block in the PX1 PCR Plate Sealer.

11. Carefully remove the TempPlate® sealing film while holding the plate down, and replace it with a sheet of pierceable foil heat seal (with red marker line up toward the top row of the plate). Close the PX1 Plate Sealer and start the heat-sealing program that has been holding since step 8.

12. Remove the plate and briefly vortex to mix the reactions. Centrifuge the plate again.

13. Program the AutoDG™ by matching the position and number of columns on the plate that hold reactions to their exact location in the droplet generation map provided by the instrument. The AutoDG™ will calculate the amount of materials and the volume of droplet generation oil that will be needed for the plate you have programmed in the system.

14. Set the plate inside the AutoDG™ system in the designated location for droplet generation.

15. Populate the AutoDG™ with the necessary materials requested by the instrument to complete the task: DG32™cartridges (up to three), boxes of tips (up to two), droplet cooling box, and a new 96-well plate to contain the newly formed droplets.

16. Once everything is in place, lower the AutoDG™ door hatch and start the instrument. For a full plate of reactions, the AutoDG™ may take up to 45 min to complete.

17. Droplet formation is affected by temperature. Ideally, the room where the AutoDG™ is kept should be between 20 and 25 °C. If the temperature is lower than this, the AutoDG™ may generate an error during the execution of the droplet generation program. To date, in our experience such errors have not impacted results (see Notes 41 and 42).

Fig. 5.

Schematic view of a 96-well plate layout for the Bio-Rad QX200® Quanta Soft™ ddPCR software. Sample and TaqMan™ assay locations are mapped in advance to assist in the calculation of reagent volumes and determination of loading strategy

3.6. Droplet Digital PCR and Plate Reading

1. Droplets are collected in a new plate over a cooling block. The plate must be heat-sealed promptly without disrupting the droplets, then transferred to a thermocycler for PCR. This should be done within an hour of droplet formation. Keep the plate on the cooling box at all times. The droplets are fragile at this stage, but will be stabilized during PCR.

2. Use the following PCR Reaction conditions:

   1. 95 °C, 10 min
   2. 40 cycles of 94 °C, 30 s and 60 °C, 1 min
   3. 98 °C, 10 min
   4. 4 °C, indefinitely

   The PCRramp rate should be between 2 °C/3 °C per s. Set the final reaction volume to 40 μL (see Note 43).

3. Open the QuantaSoft™ application and use the map generated in step 2 of Subheading 3.5 to create a droplet readout file. Assign the quantification protocol to be used (ABS), supermix type, sample names, and TaqMan™ assay to all wells.

4. Insert the plate in the QX200™ reader and initiate the droplet count. It will take an additional 2.5 h to read a full plate.

5. After completion, click OK and analyze the run. Select the ID amplitude chart from the available options under the “Analyze” tab, and manually assign a threshold to all wells with the same TaqMan™ assay. The “Analyze” tab allows the operator to choose whether to assign thresholds on an individual reaction basis or as a group. Set the threshold value such that it falls just above the negative droplets at the bottom of the chart. The software will automatically calculate a target concentration per μL of reaction.

6. After the threshold values have been assigned, save the document and export the .csv file to a USB drive for further analysis and final calculations.

7. From the “Concentration” column in the .csv file (Copies/ μL), calculate AAV Titer (Genome Copies/mL) using the following formula (see Note 44):

   $$\mathbf{AAV}\mathbf{Titer}=\mathrm{Concentration}\times 10\times 1000\mu L/\mathrm{mL}\times \mathrm{1,000,000}$$
8. For the plasmid DNase-I controls, data should be reported as “Copies per DNase-I reaction” rather than as a concentration, in order to determine the recovery of total plasmid input. Calculate using the following formula (see Note 45):

   $$\mathbf{Plasmid}\mathbf{Copies}\mathbf{per}\mathbf{DNase-I}\mathbf{Reaction}=“\mathrm{Concentration}”\times 20\mu L\times \mathrm{100,000}\times 2.5$$
9. The background signal from the “No Template Control” (NTC) wells should also be reported as “Copies per reaction.” Calculate using the following formula:

   $$\mathbf{NTC}\mathbf{Copies/reaction}=\mathrm{Concentration}\times 20\mu L$$

10. Generate a report from the .csv data and save the file as an Excel sheet or workbook (Table 1). Distribute the report to the AAV production team and quality assurance supervisors for review (see Note 46).

Table 1.

Example ddPCR Report

Sample ID	Titer #1	Titer#2	Titer #3	Average	Stnd Dev	Notes
Sample A TFF	2.52E+12	2.43E+12	2.14E+12	2.36E+12	1.99E+11
Sample A Final	2.45E+13	2.17E+13	2.03E+13	2.22E+13	2.16E+12
1E8 IE-5 (Plasmid DNase-I Control)	9.25E+07	8.85E+07	6.35E+07	8.15E+07	1.57E+07	Copies per reaction
1E8D IE 5 (Plasmid DNase-I Control)	6.00E+05	1.55E+06	1.50E+06	1.22E+06	5.35E+05	Copies per reaction
M/B (Mock Control)	3.70E+09	2.30E+09	2.40E+09	2.80E+09	7.81E+08
XXXX (Validation Control)	1.20E+13	1.08E+13	9.71E+12	1.08E+13	1.16E+12
NTC	2.40E+00	8.00E+00	3.60E+00	4.67E+00	2.95E+00	Copies per reaction

In-process samples (TFF) and final concentrates are routinely quantified in our labs, as well as crude lysates. The latter require less dilution. The validation control samples are reproducible to within 10–30% with very low background each run (<0.01%). Intra-assay reproducibility is within 10–30% among different operators

3.7. Real-Time PCR (qPCR) (See Note 47)

1. Prepare the work area as described in Subheading 3.1.

2. Thaw the qPCR standards.

3. Thaw the working TaqMan™ assay stock for qPCRin advance. The oligos should amplify the same target sequence that was used to quantify test samples and controls by ddPCR.

4. Remove the 2× TaqMan™ Universal master mix from storage. Equilibrate all reagents to room temperature as before.

5. Prepare 0.65 mL colored tubes for the real-time PCR reactions: one for NTC, eight for standards, and one for each test sample or control.

6. All real-time PCR reactions will be carried out in a 50 μL final reaction volume in triplicates for each standard, sample, or control (see Note 48).

7. Prepare the real-time PCR master mix as follows:

   Reagent	Vol. per reaction (μL)
   TaqMan™ Universal PCR Mix (2×)	26.25
   Forward Primer (3 μM)	5.25
   Reverse Primer (3 μM)	5.25
   Fluorescent Probe (2 μM)	5.25
   Nuclease-free water	5.25
   Test Sample or control	5.25
   Total	52.50

8. Multiply the volume of each reagent by the total number of NTCs, standards, and samples and controls in triplicate. Include an additional two volumes to offset pipetting errors. Prepare the master mix in a 1.7 mL microcentrifuge tube or 15 mL conical tube. Mix by vortexing.

9. Take the same precautions for qPCR as for ddPCR when dispensing reagents and samples. Keep all lids closed (tubes, tip boxes) when not in use. Avoid aerosol formation by limiting blowouts and change gloves often, especially after leaving (and returning to) the work area.

10. Transfer 141.75 μL of PCR master mix to each 0.65 mL tube. Add 15.75 μL of nuclease-free water to the NTC reaction, then add 15.75 μL of each standard, sample, or control DNA to the predesignated tube. Mix each tube by vortexing.

11. Dispense 3 × 50 μL of the reactions to designated wells in the qPCR plate using the multi-dispense function of an electronic pipette, starting with the standards, followed by samples or controls, and ending with the NTC.

12. Seal the plate with optical film and spin it down at 226 × g for 30 s.

13. Set up a dedicated file for the run using the SDS software and load the plate into the qPCR instrument.

14. Use the following PCR Reaction conditions:

    1. (a) 10 min at 95 °C
    2. (b) 40 cycles of 95 °C, 15 s and 60 °C, 1 min

15. Allow the instrument to establish the threshold automatically.

16. Calculate the final titers (GC/mL) using the following formulas.

    For ssAAV and mock control (see Note 49):

    $$\mathrm{Titer}(\mathrm{GC/mL})=(\mathrm{Qty}\mathrm{Mean}÷5\mu L)\times 2\times \mathrm{1,000,000}\times 1000\mu L/\mathrm{mL}$$
    For scAAV:

    $$\mathrm{Titer}(\mathrm{GC/mL})=(\mathrm{Qty}\mathrm{Mean}÷5\mu L)\times \mathrm{1,000,000}\times 1000\mu L/\mathrm{mL}$$
    For plasmid controls:

    $$\mathrm{Total}\mathrm{Copies}=\mathrm{Qty}\mathrm{Mean}\times \mathrm{100,000}$$

    For NTC, no “Qty Mean” is provided in the SDS7500 system. The Ct values are used as an estimate of background signal. Alternatively, a “Qty Mean” value may be calculated independently from the standard curve data if desired.

3.8. Limitations of the Protocol

1. This protocol relies on the denaturation temperature and incubation time of the PCR to break open vector capsids and make AAV genomes therein accessible for quantification. AAV serotypes are differentially stable and we may well be underestimating the titers of the most stable capsids [8, 17].

2. Some AAV genomes or target sequences therein may be refractory to amplification despite our best efforts for efficient quantification, due to entrenched secondary structures, potential methylation of CpG islands at oligonucleotide binding sites, repeat sequences, or our inability to adequately visualize the AAV genome in solution [18]. Unknown, distant interactions may lead to incorrect placement of TaqMan™ assay oligonucleotides. In this case, the combination of qPCR and ddPCR for the same target sequence may yield valuable insights regarding amplification efficiency for the improvement of AAV genome quantification by qPCR. Alternatively, methods that rely on AAV capsid (protein) quantification (semiquantitative SDS-PAGE, ELISA, and AUC) are needed to complement AAV genome quantification by PCR and achieve an analytical understanding of vector lot composition. In the absence of empty capsids and spurious DNA encapsidation products, the successful assembly of recombinant virions is expected to yield a 1:1 capsid-to-genome ratio, and this ratio should serve as a benchmark for the appraisal of quantification methods and packaging efficiency of the recombinant vectors on a lot-tolot basis. The use of surfactant during genome quantification is crucial for this assessment.

3. Limitations may be imposed by the biology of AAV production itself, where heterogeneity of packaged genomes can make it difficult to confidently assess the concentration of the genome sequence ofinterest [19]. In this case, neither PCRmethod nor AAV capsid quantification can reliably assess the integrity of AAV preps; modifications to the biology of AAV production must be implemented for these highly specialized constructs with subsequent evaluation by next generation sequencing (see Chapter 5).

3.9. Conclusion

Our data indicate that the adaptation of Plutonic® F-68 from an AAV manufacturing setting [20] to vector quantification work markedly increases quantification accuracy, and moves us closer to a true representation of vector concentrations in recombinant AAV preparations. It is now possible to offer critical quality control support to vector manufacturing facilities from the perspective of genome quantification, as it is possible to arithmetically track AAV genome content in a highly specific manner and with minimal loss of sample. This detailed protocol for AAV quantification by qPCR and ddPCR has been presented in the spirit of current guidelines for publication [21, 22] and to address concerns by the gene therapy community [23]. Our data also suggest that ddPCR and qPCR are equivalent technologies for accurate AAV quantification if the physical properties of the viral capsid are observed. Such considerations may need to be extended to other quantitative assays that are currently hindered by the adherence of the AAV capsid to plastic surfaces, most notably the semiquantitative SDS-PAGE assay [24], which together with PCR provide an overall view of AAV genome packaging efficiency and product quality. Further optimization work is in progress.

Fig. 6.

ddPCR titration is improved by inclusion of Pluronic® F-68 according to AAV serotype. The addition of F-68 during AAV guantification confers to a significant increase in signal that improves vector guantification. The degree of gain ranges from modest (two- to threefold) for AAV2/9 and AAV2/8, to large (five- to tenfold) for AAV2/Anc80 and AAV2/2. The same TagMan™ assay was used for these measurements, and all five serotypes carried the same single-stranded genome. Based on these observations, it appears that AAV capsids differentially attach to plastic

Fig. 7.

Pluronic® F-68 optimization for ddPCR. Increasing the amount of F-68 does not significantly improve AAV titration of several serotypes. The same TagMan™ assay was used for all measurements, and all vectors carried the same single-stranded genome. In light of these results, it was decided to abide by the manufacturer’s recommendations for future measurements, and to use a final F-68 concentration of 0.1% (1×) for untested AAV capsid variants

Fig. 8.

Quantification of a linearized qPCR standard plasmid. Fluorometry (Qubit™) was compared against Spectrophotometry (BioTek® Take3™). When the DNA being measured is clonally pure and clean, these two types of measurements closely agree. Fluorometry measurements may be lower than spectrophotometer O.D. measurements, because the former is highly specific for double-stranded DNA

Fig. 9.

qPCR amplification plot showing the effect of 0.1% Pluronic® F-68 on qPCR standard performance. Standards of the same copy number concentration, serially diluted from the same source material, demonstrate lower Ct values (a “shift to the left”) in the presence of F-68, indicating their higher abundance in solution due to reduced attachment to plastic. This shift translates to higher titers for AAV quantification

Fig. 10.

ddPCR vs. qPCR comparison across AAV serotypes carrying the same genome. The cis-plasmid used to manufacture these vectors was used for the qPCR standard. The same TaqMan™ assay was used for all measurements and exactly the same sample dilutions were assayed by both technologies. Pluronic® F-68 improves qPCR-based AAV genome quantification to a level comparable to ddPCR by making more target available for quantification and correcting qPCR standard performance

Fig. 11.

ddPCR intra-laboratory assessment of quantification variability. Trainees should first observe a run set up by a more experienced user. They should then set up their own run under supervision before taking up the assay independently. Samples of known concentration should be used to ensure accurate quantification by new users

Table 2.

Plasmid control samples for qPCR or ddPCR

Sample ID	CMV2 Titer #1	CMV2 Titer#2	CMV2 Titer #3	CMV2 Average	CMV2 Stnd Dev	Notes
1E8 IE-5 (Plasmid DNase-I [−] Control)	8.45E+07	8.85E+07	8.00E+07	8.43E+07	4.25E+06	Copies per reaction
1E8 IE 5 (Plasmid DNase-I [+] Control)	3.25E+06	4.70E+06	5.50E+06	4.48E+06	1.14E+06	Copies per reaction

DNase-I efficiency is >95%. Note that contrary to observations made previously [10], we are able to quantify plasmid DNA very accurately