(T1430-06-36) Immunogenicity Assessment: A Comparative Study between a Chemically Labeled and an Enzymatically Biotinylated rAAV9 Vector

Purpose: Adeno-associated viruses (AAVs) have gained significant attention in the field of gene therapy due to their desirable characteristics as gene delivery vectors. AAVs stand out particularly for their non-pathogenic disposition and versatile ability to infect a wide range of cell types. While the immune response to AAVs in humans and other primates is generally mild, there are still concerns related to the transgene expression, safety, and efficacy of this gene delivery platform thereby making the immunogenicity assessment essential. To date, several methods have been established for this purpose which include anti-AAV TAb and NAb assays. Yet, the effective labeling of rAAV capsid-derived reagents for utilization in these assays remains a significant challenge. Chemical labeling of the capsid lacks standardization and is prone to significant batch-to-batch variations which can impact assay cut-point determination, sensitivity and reproducibility, and other critical parameters thus substantially increasing assay repeats, troubleshooting interruptions, and overall costs. rAAV vectors which are engineered to be enzymatically biotinylated have also been developed. However, despite their enhanced biotinylation efficiency, known biotin labeling site, comprehensive characterization, and cost-effectiveness, these biotinylated rAAV vectors are yet to be tested for use in pre-clinical ADA assays. The aim of this project was to directly compare a chemically labeled AAV9 capsid (Biotin-AAV9) to an enzymatically labeled rAAV9-biotin vector (Biotion-rAAV9-VC007) for use in pre-clinical Total Antibody (TAb) and Neutralizing Antibody (NAb) Assays and to assess its feasibility for future ADA assay applications.

Methods: Chemical Labeling of AAV9 Capsid was performed in-house. Briefly, the ruthenylated AAV9 Capsid was labeled at 1 mg/mL at a challenge ratio 1:20, whereas the biotinylated AAV9 Capsid was labeled at 1:10 ratio. The biotinylation of the enzymatically labeled rAAV is achieved via insertion of an Avi-tag between 139-140 amino acid position of VP1 on AAV9 Rep/Cap plasmid as previously reported. Avi-tag is a short 15 amino acid peptide that is capable of biotin modification by biotin ligase (BirA) to the lysine residue without effecting the protein's function. A quatral plasmid co-transfection method was used to produce the Biotin-rAAV9 vectors used in this study. Briefly, one transgene plasmid pVC007 that expresses GFP, one AVI-rAAV9 Rep/Cap plasmid, one BirA-EGFP plasmid, and one helper plasmid, were co-transfected into HEK293 cells at a ratio of 1:1:1:2. The purification methods used were identical to those used to produce nonbiotinylated rAAVs. The peak fractions of purified vector were pooled. Biotin was quantified utilizing a commercially available Biotin Quantitation Kit. The chemically and enzymatically labeled AAV9-Biotin vectors were evaluated in three different total anti-AAV9 antibody binding (TAB) assay formats as shown in Figure 1A. A previously developed cell-based assay to measure activity of anti-AAV9-Neutralizing Antibody (NAb) in mouse serum was also utilized to compare the chemically and enzymatically labeled AAV9-Biotin vectors as indicated in Figure 1B. Briefly, this method utilizes a AAV9 vector containing a luciferase reporter gene and HEK 293T cells. The AAV9-transduced cells are lysed in buffer containing the luciferase substrate (ONE-Glo^TM EX reagent) and read for luminescence signal. In this assay, the presence of anti-AAV9 neutralizing antibodies in the test serum sample results in a reduction of AAV9-CMV-luciferase transduction efficiency, and hence, a lower expression of the luciferase reporter gene, leading to a lower luminescence signal readout

Results: The enzymatically labeled rAAV9-VC007 vector demonstrated an enhanced signal-to-noise (SNR) ratio across all assay formats surpassing that of the chemically labeled AAV9. Of note here is that the highest relative SNR was generated when measuring immunoglobulin G (IgG) that could bind to AAV capsids. These findings indicate that the enzymatically biotinylated vector performs on par with the chemically labeled AAV9 while also preserving its inherent ability to transduce cells as evidenced by the results of the neutralizing antibody (NAb) assay.

Conclusion: Labeling of AAV capsids continues to be a challenge during pre-clinical ADA assay development especially due to the multiplicity of AAV vector characteristics. This study demonstrates the successful utilization of an enzymatically biotinylated rAAV9 vector generated through the quatral plasmid co-transfection method in pre-clinical ADA assays. The findings support the use of these vectors as a promising alternative to overcoming challenges associated with chemical labeling of AAV capsids for use in ADA assays. The utilization of these vectors can potentially reduce the inherent variability observed across different testing sites thereby enabling the development of considerably more robust assays for immunogenicity assessment.

References: 1. Molecular Therapy: Methods & Clinical Development (2015) 2, 15047; doi:10.1038/mtm.2015.47
2. Molecular Therapy: Methods & Clinical Development (2022) 25:74-83; doi: 10.1016/j.omtm.2022.02.011

Acknowledgements: This study was sponsored and funded by Takeda Pharmaceutical Company Limited. J.R., M.Q., and X.Z. are employees of Takeda Development Center Americas, Inc., Cambridge, MA, USA. X.S. was an employee of Takeda Development Center Americas, Inc., Cambridge, MA, USA while contributing to this study.

(A) Total anti-AAV9 Antibody (TAb) Assay Formats. Typically, a bridging assay that detects all binding antibodies. From left to right: detection utilizing ruthenylated-AAV, ruthenylated-Streptavidin, ruthenylated Anti-Mouse IgG.
(B) Neutralizing Antibody Binding (NAb) Assay format. An in vitro cell-based neutralizing assay that detects neutralizing antibodies (NAbs) that interfere with transduction by measuring outstanding expression of Luciferase following transduction with AAV9.

Anti-AAV9 monoclonal antibody was used as a positive control (PC) in all AAV9 TAb assays. The PC was diluted in negative mouse serum at 0, 100,300, and 900 ng/mL. Final concentrations after 50-fold dilutions in assay buffer are shown on the x-axis. Top Panel: the y-axis shows RLU signal/mean negative control as SNR ratio. Bottom Panel: the y-axis shows % Inhibition which captures the specificity of the positive anti-AAV9 antibodies identified during screening assay (Top Panel). The diluted PC was incubated with or without an excess of unlabeled AAV9 vector to determine the percentage of inhibition of the specific binding.

Anti-AAV9 monoclonal antibody was used as a positive control in the NAb assay. It was diluted in negative mouse serum at 0, 25,100, and 400 ng/mL. The AAV9 Nab assay was conducted in HEK293T cells. The presence or absence of AAV9-specific NAb is determined by the inhibition of AAV9 transduction in HEK293T cells with either the engineered Biotin-rVAA9-VC007* or with chemically labeled Biotin-AAV9**.