A Step-By-Step Method to Detect Neutralizing Antibodies Against AAV using a Colorimetric Cell-Based Assay

Adeno-associated viruses (AAV) are increasingly used as vectors for the delivery of gene therapies to trial treatments for various health conditions that impact the cardiovascular, pulmonary, circulatory, ocular, and central nervous systems^1,2,3,4,5. The popularity of AAV vectors as a leading gene therapy platform stems from their positive safety profile, long-term transgene expression, and wide-ranging tissue-specific tropisms^1,6. Successful outcomes in animal studies have paved the way for over fifty AAV gene therapy clinical trials that have successfully reached their efficacy endpoints⁷, as well as the release of the first commercially available AAV gene therapy drug approved by the US Food and Drug Administration⁸. Following initial successes, AAV has continued to gain traction in the basic and clinical research sectors as a vector of choice and is currently the only in vivo gene therapy approved for clinical use in the US and Europe⁹. Nonetheless, the presence of pre-existing neutralizing antibodies (NAbs) against AAV vector capsids remains a hindrance to both preclinical research and the efficacy of clinical trials. NAbs are present in both naïve human and animal populations and inhibit gene transduction following in vivo administration of an AAV vector¹. AAV seropositivity is an exclusion criterion for most gene therapy trials, and therefore preliminary screening for host immunity is crucial in both the laboratory and the clinic. Establishing an assay that can detect the presence of NAbs against AAV is an essential step in the pipeline of any AAV gene therapy-based research project. This report focuses on AAV6 which has been of interest to researchers due to its efficient and selective transduction in striated muscle (heart and skeletal muscle)^1,10,11,12. Gene therapy is considered a promising strategy for targeting the heart because it is difficult to specifically target the heart without invasive open-heart procedures.

Neutralizing activity is usually determined using either a cell-based in vitro or in vivo transduction inhibition assay. In vivo NAb assays usually involve administering serum from a test subject (e.g., human or large animal) into mice, followed by an AAV with a reporter gene, followed by testing for the expression of the reporter gene or corresponding antigen. In vitro assays determine NAb titers by incubating serum or plasma from a human or large animal in serial dilutions with a recombinant AAV (rAAV) that expresses a reporter gene. Cells are infected with the serum/virus mixture, and the extent to which the reporter gene expression is inhibited is assessed compared with controls. In vitro assays are widely used for NAb screening due to their comparatively lower cost, rapidity in testing, and greater capacity for standardization and validation^13,14 compared with in vivo assays. In vivo assays are often reported to have greater sensitivity^15,16, but the same claim has been made concerning in vitro assays^14,17.

To date, in vitro NAb assays have mainly used luminescence (luciferase) as the reporter gene to detect neutralization. Although a light-based method has merit in many contexts, a colorimetric/chromogenic NAb assay may be advantageous in some circumstances. Colorimetric assays to assess neutralization have been successfully employed for other viruses such as influenza and adenovirus^18,19. Their attractiveness stems from their simplicity, lower cost, and the requirement for only everyday laboratory apparatus and tools²⁰. NAb assays that use a luminescence-based reporter gene require costly substrate kits, a luminometer, and corresponding software for analysis²¹. This colorimetric assay has the advantage of only requiring a light microscope and a very cheap substrate. Reporting of the sensitivity of colorimetric versus luminescent assays has yielded conflicting results. One study suggested luminescence-based ELISA assays display greater sensitivity and comparable reproducibility to colorimetric assays²², while another found colorimetric-based ELISA assays to confer greater sensitivity²³. Here, a detailed protocol for an in vitro NAb assay against AAV that utilizes the chromogenic reaction between an AAV encoding an alkaline phosphatase (AP) reporter gene and a nitro blue tetrazolium /5-bromo-4-chloro-3-indolyl phosphate (NBT/BCIP) substrate is provided. This step-by-step protocol was developed based on a previous report that utilized an hPLAP (human placental alkaline phosphatase) reporter gene (AAV6-hPLAP) to detect neutralizing activity against AAV²⁴. This assay is cost-effective, time-efficient, easy to set up, and requires minimal technical skills, laboratory equipment, and reagents. Moreover, the simplicity of this assay gives it the potential to be optimized for broad applications across different types of cells, tissues, or viral serotypes.

All aspects of animal care and experimentation were conducted following Florey Institute of Neuroscience and Mental Health guidelines and the Australian Code for the Care and Use of Animals for Scientific Purposes following Reference²⁵. 1.5-3-year-old Merino ewes were used for the study. A schematic overview of the assay protocol is provided in Figure 1.

Figure 1: Schematic diagram of NAb assay protocol. (A) Visual representation of the NAb assay illustrating the primary steps involved in the three-day protocol. Briefly, cells are grown and plated overnight. The following day, serial dilutions of serum are prepared, incubated with AAV, and then incubated with the cells overnight. The next day, cells are fixed, washed, incubated, combined with the substrate, and incubated again, followed by imaging and quantitation. (B) Representative images of a minimum signal control (complete AAV inhibition), a maximum signal control (no inhibition), and an ovine serum sample with ~50% signal inhibition. Scale bar = 0.5 mm. Please click here to view a larger version of this figure.

1. Initial preparation

1. For assessment in sheep: collect blood in 8 mL serum separator clot activator tubes (see Table of Materials), leave the blood sample at room temperature (RT) for 20-30 min, and subsequently spin down at 2,100 x g for 15 min. The clear supernatant that forms at the top of the tubes is serum. Aliquot the clear aqueous phase into microcentrifuge tubes and store at -80 °C.
   NOTE: The serum at -80 °C remains stable for ~5 years. Blood was collected from the carotid vein using a 16 G needle (tip cut-off) and syringe from conscious animals.
2. Heat inactivate fetal bovine serum (FBS) by placing it in a water bath at 56 °C for 30 min and swirl intermittently. For precision, place a thermometer in a second bottle containing an equivalent volume of water and add it to the heat bath at the same time as the FBS bottle. Begin timing when the thermometer reaches 56 °C.
3. Employ proper aseptic technique and cell culture practice for all subsequent steps performed in the cell culture hood^26,27. Spray 70% ethanol on all objects and the hood before use and clean with 1% sodium hypochlorite upon completion.
4. Make complete Dulbecco's Modified Eagle Medium (DMEM) by combining high glucose (4.5 g/L) DMEM (89%) with heat-inactivated FBS (10%) and Penicillin Streptomycin (1%). Combine and filter using a sterile vacuum filtration system (0.22 µm pore size, polyethersulfone membrane) (see Table of Materials). Store complete DMEM wrapped in foil at 4 °C.
5. Establish HT1080 cells (see Table of Materials) and passage in a 75 cm² square flask as described in Reference²⁸. Create multiple frozen stocks of cells. Do not use cells after 20 passages as further passaging may influence the assay results.

2. Day 1 - Plating of cells

1. Passage HT1080 cells when they reach ~80% confluency.
2. Pre-warm complete DMEM (prepared in step 1.4), 0.05% trypsin-EDTA, and 1x phosphate-buffered saline (PBS) to 37 °C in a water bath.Remove the growth medium from passaged cells using an aspiration system.
   NOTE: All aspiration in this protocol uses a vacuum system with a tube attached to a sterile 5 mL serological pipette.
3. Wash the cells in 10 mL of pre-warmed (37 °C) 1x PBS and trypsinize cells for 3-4 min in 4 mL of pre-warmed 0.05% trypsin-EDTA to detach the cells from the flask.
4. Inactivate the trypsin by adding 6 mL of pre-warmed complete DMEM and pipette the cells into a 50 mL tube. Calculate the number and concentration of viable cells using a hemocytometer and the trypan blue exclusion method²⁹.
5. Dilute the cells to a concentration of 1 x 10⁵ cells/mL in pre-warmed complete DMEM. Seed 100 µL of cells/well into clear 96-well flat-bottomed plates (1 x 10⁴ cells per well). Incubate the plate at 37 °C, 5% carbon dioxide (CO₂) overnight for 16-22 h.

3. Day 2 - Infecting the cells

1. Remove plate/s from the incubator and use a light microscope to confirm that cells are evenly dispersed within the wells and that the confluency is ~50%. If cells are not within a range of 45%-55% confluency, repeat the 'Day 1' protocol and adjust initial cell concentration accordingly.
2. Generate serial dilutions of the serum samples of interest in 1.5 mL microcentrifuge tubes using pre-warmed complete DMEM as the diluent. Table 1 demonstrates the generation of a dilution cascade for triplicate samples.
   1. To perform the assay in triplicate, prepare a 7.5 x 10⁶ vector genomes (vg)/µL of working solution of AAV6-hPLAP (see Table of Materials) by diluting a virus stock solution in 1x PBS.
   2. Add 66 µL of the 7.5 x 10⁶ vg/µL virus working solution to each tube containing 264 µL of serum/media dilution (330 µL of total volume/dilution, see Table 1).
      NOTE: This is a robust assay that does not require perfect culture conditions. However, to accurately quantitate and ensure each assay run is reliable, it is necessary to include the following: (1) a virus and media only control, (2) a media only control, and (3) a NAb positive control sample on all plates under the same experimental conditions. The volume described (330 µL) accounts for triplicate samples +10% of the serum and virus mixture. Performing replicates is highly recommended for the accurate determination of neutralizing activity.
3. Mix the virus/serum dilutions by pipetting and place the tubes containing the virus/serum mixtures in an incubator at 37 °C, 5% CO₂ for 30 min to allow potential neutralization to occur.
4. Pipette 100 µL of the virus/serum mixture to each well on the 96-well plate containing 1 x 10⁴ cells/well for each dilution.
   NOTE: This will generate a final viral concentration of 15k viruses/cell multiplicity of infection (MOI) in each well. Table 2 provides an example 96-well sample plate layout for assessing samples to a 1/512 dilution.
5. Wrap the 96-well plate containing cells, serum, and AAV-hPLAP in foil and place in an incubator at 37 °C, 5% CO₂ overnight for 16-24 h to allow AAV entry into the cells.

Table 1: Volumes of serum and diluent required to generate serial dilutions of serum in triplicate.

Table 2: Example 96-well plate layout for assessing naïve serum samples in dilutions ranging from 1/2 to 1/512. Higher dilutions are incorporated into the assay if assessing a sample known to be positive for AAV NAbs (post-administration samples) or if a higher titer is required. MO (-C): Media-only control. VO (+C): Virus and media only control. mAb: Monoclonal antibody against AAV (NAb positive control).

4. Day 3 - Fixing and adding substrate to the cells

1. Pre-warm an aliquot of 1x PBS to 37 °C (~25 mL/96-well plate). Cool separate aliquots of PBS (~25 mL/96-well plate) and double-distilled H₂O (DDW, ~25 mL/96-well plate) to 4 °C. Dissolve a pellet of BCIP/NBT (see Table of Materials) in 10 mL of DDW in a 50 mL conical centrifuge tube by vortexing (10 mL is enough for 2 x 96 well plates).
2. Aspirate the media from the wells of the 96-well plate using a serological pipette or similar attached to a suction-based aspiration system or fume hood vacuum. Gently place the tip of the serological pipette into the well and remove the media taking caution not to disrupt the adhered cells.
   1. Add 50 µL of RT 4% PFA to each well using a pipette. Wrap the plate in foil and leave it at RT for 10 min to fix the cells.
      CAUTION: Paraformaldehyde (PFA) is a probable carcinogen and is toxic from skin or eye contact or inhalation. Handle in a fume hood with proper personal protective equipment as well as a facemask. Make fresh 4% PFA diluted in PBS (~7 mL required per 96-well plate).
3. Wash and aspirate the cells with 200 µL of RT 1x PBS. Repeat this step twice.
   NOTE: A multichannel pipette is an efficient option for the pipetting steps.
4. Pipette 200 µL of pre-warmed PBS into each well, wrap the plate in foil and incubate at 65 °C for 90 min to denature endogenous alkaline phosphatase activity³⁰.
5. Aspirate wells and wash cells with 200 µL of cold (4 °C) PBS. Aspirate again, wash in 200 µl of cold DDW, and aspirate again.
6. Pipette 50 µL of the dissolved BCIP/NBT (prepared in step 4.1) into each well.
7. Wrap the plate in foil and incubate at RT for 2-24 h.
   NOTE: Be consistent with incubation time between runs; the time flexibility allows users to photograph wells either on day 3 or the following day.
8. Using a light microscope camera, take photos of each well using a 4x objective lens, ensuring the same exposure, white balancing, and light settings are used consistently for all assays performed.
   1. Position each well identically and ensure the edges of the well are not visible in the photos. Save photos in TIF format or similar.
      ​NOTE: Specific settings will vary between microscopes, but quantitation will be most effective if the background lighting is high and consistent throughout the wells (Figure 1B).

5. Quantitation to determine the neutralizing activity using ImageJ

1. Download and install the freely available software "ImageJ" (see Table of Materials).
2. Open the image to be analyzed in ImageJ by selecting File > Open (Figure 2).
3. If using colored images, convert to grayscale by selecting Image > Type > 8-bit.
4. Click on Image > Adjust > Threshold. Adjust the threshold until all colored areas are colored in red, but the background is not. Upon adding NBT/BCIP, the colored product will deposit in the area around the cells expressing hPLAP.
   NOTE: It is recommended to use the same threshold setting for all images captured on the same plate.
5. Click on Analyze > Set Measurements and tick the Area, Limit to Threshold, Area Fraction, and Display label checkboxes and click on Ok.
6. To determine the signal reading (percentage of coloration) of a given well, click on Analyze > Measure. The '% Area' column of the pop-up window displays the signal reading.
7. Perform quantitation for all sample replicates. Exclude any contaminated wells, wells showing uneven cell distribution, or wells varying in cell density or lighting.
   ​NOTE: See Supplementary Figure 1 for examples of wells that should be considered for exclusion. Typically, 3-4 wells may require exclusion from a 96-well plate. Figure 2 provides a visual representation of the quantitation process using ImageJ.

Figure 2: Steps for determining percentage coloration using ImageJ software. (A) Open the image to be analyzed with ImageJ software. (B) Convert the image to 8-bit grayscale. (C) Open the threshold window. (D) Adjust the maximum threshold so all colored areas are covered, but the background area is not (this threshold should be consistent across an entire plate). (E) Select the 'Analyze' dropbox, click on 'Set measurements' and tick 'Area', 'Area fraction', 'Limit threshold' and 'Display label', and click on 'OK'. (F) Click on 'Measure' to measure the covered area. The % area indicates the proportion of the image that was colored. This can then be used with the control samples to determine the TI₅₀ titer. Please click here to view a larger version of this figure.

6. Determination of Transduction Inhibition (TI₅₀) titer

1. Determine the average readout from replicates (using the steps described in step 5) for the following: (1) Media-only control (baseline signal reading). (2) Virus + media only control (maximum signal reading). (3) Virus + serum samples of interest.
2. Calculate the percentage of inhibition using the following formula:
   100 - [(Test sample signal readout (virus + serum sample of interest) - baseline signal readout (media only control)) / (maximum signal readout (media and virus only) - baseline signal readout) x 100] = % Transduction inhibition¹³.
3. Calculate the % transduction inhibition from all replicates of each dilution for all samples using the formula in 6.2. Determine the average transduction inhibition between the technical replicates for each dilution for all samples and controls.
4. Calculate the 50% transduction inhibition titer (TI₅₀ titer) of a sample of interest by determining the lowest dilution of the sample that yields 50% or greater transduction inhibition of hPLAP activity. e.g., if a 1/8 dilution of a sample has greater than 50% transduction inhibition based on the calculation performed in 6.2 (and a 1/4 dilution does not), report the TI₅₀ titer as 1/8.

7. Determination of neutralized AAV particles

1. Calculate the number of neutralized AAV particles per µL of serum for a given sample by employing the following formula:
   ((MOI x cell count/well) / (volume of serum / dilution factor of TI₅₀ titer)) / 2 = neutralized AAV particles / µL of serum⁹.
   NOTE: Dividing by 2 accounts for the TI₅₀ measuring 50% of neutralized particles. For a sample that gives a TI₅₀ titer of 1/4 (25% serum, 75% diluent) in which the assay used 80 μL of undiluted serum and an MOI of 15k plated onto 1 x 10⁴ cells, the following calculation would be used: ((15000 x 10000) / (80/4)) / 2 = 3.75x10⁶ neutralized particles / µL of serum.

Transduction assay to establish the optimal viral dosage for plate coverage
HT1080 cells, a well-established fibrosarcoma cell line, were selected for this assay. A concentration of 1 x 10⁴ HT1080 cells/well provided ~50% cell confluency in each well of a 96-well plate. To determine the optimal viral concentration for the assay, an rAAV encoding an hPLAP (human placental alkaline phosphatase) reporter gene (AAV6-hPLAP)³¹ was added in triplicate at a range of concentrations of vg containing particles per cell (MOI: 0, 150, 500, 1500, 5000, 15000, 50000 & 150000 (Figure 3A)). An MOI of 15000 (1.5 x 10⁸ vg/well) conferred 36% plate coloration and was selected as the optimal viral dosage. A positive correlation was observed between coloration and viral concentrations for all MOI between 0 and 15000 (n = 6, r = 0.995, P<0.001). This concentration adequately displayed the reporter gene signal above the background in the presence of high concentrations of NAbs (low plate coloration) while not losing sensitivity due to color saturation in the absence of NAbs (high coloration, Figure 3B).

The efficacy of the assay, when exposed to neutralizing elements, was trialed using serial dilutions of an anti-AAV6 mouse monoclonal antibody (mAb) in triplicate. The standard approach of using the first dilution to display 50% or more transduction inhibition (TI₅₀) was applied to determine the neutralizing titer of a given sample (step 6). Assessing log₁₀ dilutions, the anti-AAV6 mAb displayed a TI₅₀ titer of ~10 ng/mL (1 ng total mAb), while a concentration of 500 ng/mL (50 ng total Ab) and above completely inhibited reporter gene expression (Figure 3C).

Figure 3: Optimization of viral dose and assessment of assay efficacy against an anti-AAV6 monoclonal antibody (mAB). (A) The proportion of coloration (% of total area) for individual wells at different multiplicities of infection (MOI) and representative images (below) displaying the corresponding chromogenic reaction between the alkaline phosphatase (hPLAP) and NBT/BCIP for each viral dose (left). Percentage plate coverage is also shown in tabulated form (right). Each data point represents a technical replicate (n = 3 replicates per MOI). (B) A representation of the correlation between coloration and MOI is shown in Figure 3A. The red dotted line represents the highest concentration tested that did not affect the linear correlation between coloration and viral concentration. (C) Neutralizing activity against AAV6 from an anti-AAV6 monoclonal antibody at log₁₀ dilutions. 50% inhibition of AAV6 transduction (TI₅₀) is observed at a ~10 ng/mL concentration. n = 3 replicates per dilution. Data represent mean ± SEM. Please click here to view a larger version of this figure.

Assessment of NAbs against AAV6 in sheep samples
Serum was collected from the carotid vein using a 16 G needle (tip cut-off) and syringe from conscious naïve healthy adult sheep (1.5-3-year-old Merino ewes) (n = 11) and screened to determine the TI₅₀ titer using the colorimetric NAb assay. Two-fold serial dilutions ranging from a 1/2 to a 1/512 dilution were assessed for each serum sample in duplicate or triplicate. The 1/2 dilution contained a total of 40 µL of serum, which corresponded to a concentration of 1.88 x 10⁶ AAV particles per µL of serum. The degree of AAV neutralization varied within the naïve sample population, with TI₅₀ titer values ranging from as low as 1/2 (blue line) to as high as 1/80 (green line, 1.88 x 10⁶ to 7.5 x 10⁷ neutralized AAV particles/µL of serum) (Figure 4A).

Subsequently, direct cardiac injection (n = 5) of AAV6 was performed at doses ranging between 5 x 10¹² and 3 x 10¹³ vg to naïve sheep. Briefly, sheep were anesthetized as previously described³². The heart was exposed from the left lateral position. The pericardium was opened, and AAV was administered via 10-40 ~20 µL injections into the left ventricular myocardium (anterior) in the region around the second branch of the left anterior descending coronary artery (LAD). The pericardium, intercostal muscle, subcutaneous tissue, and skin were closed, and the anesthetic was removed. Serum was collected from all animals before and six to eight weeks after AAV administration from the pre-cannulated right jugular vein using a 16 G needle cut-off and syringe (~5 mL serum/animal). The colorimetric NAb assay was employed to determine the change in NAb titer following AAV6 administration. Post-AAV serum samples were screened in triplicate as 2- to 4-fold serial dilutions ranging from a 1/2 to a 1/32768 dilution. Assay results indicated that AAV inhibition TI₅₀ titer values before AAV administration ranged from 1/4 to 1/80 (3.75 x 10⁶ to 7.5 x 10⁷ neutralized AAV particles/µL of serum; Figure 4B). Following AAV cardiac injection, a clear and consistent contrast in the NAb titer was observed compared with pre-AAV serum titers (Figure 4B, Table 3). The lowest AAV dose (5 x10¹² vg) displayed a 1/2048 TI₅₀ titer (dashed blue line, 1.92 x 10⁹ neutralized AAV particles/µL serum) and the remainder of the doses (1-3 x 10¹³ vg) displayed TI₅₀ titers ranging from 1/12000 to >1/32768 (1.13 x10¹⁰ to >3 x 10¹⁰ neutralized AAV particles/µL of serum).

Figure 4: Example of neutralizing antibody (Nab) assay results using ovine serum samples. (A) Adeno-Associated Virus (AAV) neutralizing serum samples were collected from 11 naïve sheep, measured in 2-fold serial dilutions ranging from 1/2 to 1/512. Colored lines represent samples with low and high neutralizing activity; grey lines represent the nine additional samples. The dotted line represents 50% transduction inhibition (TI₅₀) and the corresponding TI₅₀ titers for the low (blue) and high (green) representative samples. (B) AAV neutralizing activity of serum samples collected from five sheep before and after receiving a dose of AAV via direct cardiac injection. Neutralizing activity was assessed in 2-fold serial dilutions ranging from 1/2 to 1/32768. Each color represents serum from a different animal, filled lines represent pre-AAV administration and dotted lines represent post-AAV administration. n = 2-3 replicates per dilution for each sample. Data represent mean ± SEM. Please click here to view a larger version of this figure.

Table 3: Impact of AAV exposure on neutralizing activity. Neutralizing activity for sheep was assessed before and after receiving a direct cardiac injection of a rAAV6 at varying doses. The dose received pre and post TI₅₀ titers and fold change following administration are displayed.

Supplementary Figure 1: Visual examples of different reasons for excluding sample wells. (A) The presence of contamination can be seen by clumps in the center. (B) High cell density. (C) Uneven lighting of well (left image), corresponding thresholding of the same well displaying excess coverage in the bottom right corner (right image). (D) Technical replicate images, the image on the left representing results with normal cell density, the image on the right reflecting results with low cell density. (E) Cells are unevenly distributed across a well. Please click here to download this File.

Supplementary Figure 2: Simplified plasmid map displaying hPLAP insert. CMV: Cytomegalovirus promoter. hPLAP: human placental alkaline phosphatase. SV40: Simian virus 40 polyadenylation signal. ITR: Inverted terminal repeat sequences. Please click here to download this File.

This report describes a colorimetric assay that assesses the extent of AAV neutralization in a given serum sample by evaluating a chromogenic reaction corresponding to the degree of in vitro viral transduction. The development of the protocol was based on the known chromogenic reaction between the enzyme alkaline phosphatase and NBT/BCIP, which has been widely utilized as a staining tool for the detection of protein targets in applications such as immunohistochemistry and as a reporter tool for evaluating viral transduction^24,33,34,35. Its merit stems from its time and cost-effectiveness, accessibility, ease of setting up and performing while still demonstrating a high degree of efficacy. The rAAV6 employed in this assay (AAV-hPLAP) carries the reporter gene human placental alkaline phosphatase (hPLAP) and is driven by a cytomegalovirus (CMV) promoter³⁴ (Supplementary Figure 2). NBT/BCIP is an hPLAP substrate that is initially dephosphorylated by alkaline phosphatase and sequentially undergoes oxidization to form a dimer, resulting in an insoluble product that is a vibrant purple color³⁶.

In selecting the optimal MOI for this assay, the aim was to establish a viral concentration that would sufficiently express the AAV reporter gene through virus-cell binding, and in conjunction with the NBT/BCIP substrate, provide coloration within a range that could be accurately measured. An MOI of 15,000 was selected, as it was the highest concentration tested that did not affect the linear correlation between coloration and viral concentration. Higher concentrations (50,000 and 150,000 MOI) caused the concentration-color response curve to plateau, indicating color saturation (Figure 3B). Assessment of viral MOIs between 0 and 15,000 (n = 6) and their corresponding level of coloration resulted in r = 0.995 (P < 0.001), validating the sensitivity of the assay by establishing a very strong positive correlation between viral concentration and reporter-gene driven coloration. Given potential variability in factors such as cell culture conditions, laboratory technicians, techniques, and equipment, as well as variations in viral batches, it is recommended that any new user perform a preliminary trial assessment of the optimal MOI when establishing a NAb assay.

The MOI chosen for a given NAb assay is a major contributing factor in the overall titer observed for a given serum sample. If an MOI of 5,000 instead of 15,000 had been selected, a 3-fold difference in titer value would be anticipated. This has historically been problematic in the field of AAV gene therapy, as different preclinical and clinical trials have implemented AAV NAb assays with MOI values ranging from less than 1,000 to as high 25,000³⁷, meaning any kind of cross-study comparison of NAb titers for a given AAV serotype is of little to no value. It has recently been suggested that reporting titers as neutralized AAV particles per µL of serum can provide more comparable values across different studies^9,38. Numerous other factors may contribute to variation in titers between studies, such as the choice of the cell line, reporter gene, incubation times, and culture conditions. To facilitate the standardization of AAV NAb assays, both the titer values and neutralized AAV particles/µL of serum have been reported.

It is essential to include a serial dilution of a known NAb sample on every plate to act as both a positive control and a common sample between plates. This is important to identify any possible variability between separate runs. A neutralizing monoclonal antibody against the AAV of interest is an ideal positive control and standard, but a serum sample that is positive for NAbs is also acceptable. The efficacy of the assay was validated by demonstrating that a monoclonal antibody specific to intact AAV6 particles (ADK6) can quantitatively inhibit transduction in a concentration-dependent manner.

Based on the manufacturer's material datasheet, the BCIP/NBT substrate system produces the insoluble blue-purple product within ~10 min and is very stable. However, it is noted that procedures may affect the length of incubation time. Based on prior reports, time frames of 1-24 h have been used^30,39. For this assay, incubation times must be consistent between runs. The time flexibility allows users to photograph wells either on day 3 of the protocol or the following day.

Preclinical trials using large animal models provide a crucial stepping-stone between the laboratory and the clinic due to the physiological resemblance that animals such as sheep and pigs share with humans^1,40,41. Historically, most candidate AAV gene therapies that have made it to clinical trials have undergone preliminary trials in large animals¹. Multiple studies have demonstrated that both humans and a range of large animals, including sheep, pigs, dogs, rabbits, and non-human primates, can harbor neutralizing antibodies against AAV6 as well as many other AAV serotypes^42,43. This highlights the importance of preliminary screening for NAbs before trials in both large animal models and humans. NAb status of serum samples from sheep that had no previously known exposure to AAV was assessed, of which 10 of 11 displayed TI₅₀ titers <1/30 (<3 x 10⁷ neutralized AAV particles/µL of serum). In contrast, one displayed a TI₅₀ titer of 1/80 (7.5 x 10⁷ neutralized AAV particles/µL of serum). Five of the sheep went on to receive direct cardiac muscle injection of rAAV. Amongst all samples, administration of AAV dramatically changed the neutralizing activity, with fold change increases ranging from 125 to a >10,000 fold in TI₅₀ titer values between pre and post AAV exposure (Table 3). Of note, the lowest AAV6 dose administered (5 x 10¹² vg) corresponded to the lowest post-AAV TI₅₀ titer value (TI₅₀ titer 1/2000) and fold change increase (125 fold). In comparison, no clear evidence of pre-existing NAbs within the 11 naïve sheep was observed at levels that would prevent AAV transduction (all TI₅₀ titer values <1/100). Data from the 5 sheep that received a direct AAV injection would suggest a cut-off for NAb positivity would be >1/1000 (based on pre-and post-AAV values). The stark difference between the pre-and post-NAb titer and the capacity to detect titers >1/32,000 provides further validation of the efficacy and sensitivity of the assay. Establishing a cut-point at which a sample is deemed positive for neutralizing activity is an essential subsequent step when determining a threshold for NAb positive animals. This can be determined by assessing the variability in a group (~n ≥ 30) of naïve samples from a specific population of interest. This allows for establishing criteria to choose a statistically derived cut-point in which a sample is deemed positive for neutralizing activity. Alternatively, positive control samples from animals both pre-and post-AAV administration, as shown in Figure 4B, can indicate the positive titer range for neutralizing activity. Recommendations regarding establishing cut-point thresholds and validation and optimization of in vitro neutralizing assays have been described extensively in the literature^44,45,46,47.

The technique involves certain limitations. The use of the microscope camera to image wells is helpful because it provides very high-quality images that can accurately differentiate the degree of neutralization. However, microscopy can be labor intensive and may not be practical if processing many (>100) samples. A high-resolution flatbed or plate scanner may provide a more rapid approach to imaging the wells if the quality and lighting of images can be maintained. High MOI's have been reported to reduce assay detection sensitivity. It has been suggested that high AAV doses may evade neutralizing activity or that AAV transduction may occur in the presence of NAbs^48,49,50. This assay uses a moderately high MOI (15,000); however, the sensitivity does not appear to be impeded as it can detect neutralizing activity at very high dilutions (>1:32,000) (Figure 4B). Lastly, as this assay uses a viral vector, appropriate approval by a governing body is generally required to use recombinant AAV; this will differ from country to country.

In summary, this in vitro assay provides a rapid, cost-effective, easily accessible, and simple method to detect the presence of neutralizing antibodies against rAAV6. This assay can be adjusted and optimized to perform with different AAV serotypes with relative ease. The AAV6-hPLAP vector can be provided for this assay upon request.