Analysis of Adeno-Associated Virus Quality Attributes with LC-FLD-(MS)

Significance of the topic

Adeno-associated viral (AAV) vectors are central to modern gene therapies due to their high transduction efficiency and low immunogenicity. Accurate characterization of their critical quality attributes (CQAs), including aggregation state, full versus empty capsid ratio, and capsid protein composition, underpins the safety, efficacy, and regulatory approval of gene therapy products.

Objectives and study overview

This work presents optimized liquid chromatography–fluorescence detection coupled, where indicated, with mass spectrometry (LC-FLD-(MS)) workflows for simultaneous assessment of three key AAV CQAs: aggregation, full/empty capsid ratio, and capsid protein identity and stoichiometry. The study evaluates method linearity, precision, sensitivity, and serotype versatility.

Methodology and instrumentation

The following analytical strategies were developed:

• Aggregation analysis by size-exclusion chromatography (SEC) on Bio SEC-5 columns (4.6×300 mm, 5 μm, 1000 Å) using 50 mM phosphate, 400 mM NaCl, pH 7.4, flow rate 0.4 mL/min. Detection by intrinsic fluorescence (Ex 280 nm, Em 340 nm).
• Full/empty ratio assessment on Bio SAX NP5 PK weak anion-exchange columns (2.1×50 mm, 5 μm) with a tetramethylammonium chloride gradient (150–275 mM) in 70 mM Bis-Tris Propane + 2 mM MgCl₂, flow rate 0.1 mL/min.
• Capsid protein separation and identification on Zorbax RRHD StableBond columns (C3, diphenyl, C18; 2.1×100 mm, 1.8 μm) at 75–80 °C following heat denaturation (70 °C, 3 M GdHCl + 10 mM DTT). Mobile phase comprised IPA, ACN, water, formic and trifluoroacetic acids. High-resolution MS was used for intact mass and MS/MS confirmation.

Main results and discussion

• Aggregation: Unstressed AAV-2 and AAV-9 exhibited low aggregate levels (1.2–1.3%), whereas heat- and DNA-stressed AAV-9 showed 7.5% aggregates and 8.8% fragments. Total peak area dropped by ~40%, indicating insoluble species formation.
• Full/Empty Capsid Ratio: Calibration with mixed standards of AAV-1 and AAV-6 yielded linear responses (R²>0.966) for relative peak height and area. Inter-day RSD for full/empty ratios was ≤10% (AAV-1) and ≤11.2% (AAV-6) across 5–90% full capsids, establishing a 5% limit of quantitation.
• Capsid Protein Identity and Stoichiometry: Optimal resolution of VP1–VP3 varied by serotype; C18 columns provided baseline separation for AAV-2, while diphenyl chemistry was required for AAV-9, rh10, and Anc80. Measured VP1:VP2:VP3 ratios approximated the theoretical 1:1:10, with greater variability observed in VP1 abundance. A previously unreported Ala→Tyr mutation in AAV-2 VP1 was detected by intact mass shift and confirmed by MS/MS.

Benefits and practical applications

These LC-FLD-(MS) methods enable rapid, sensitive, and reproducible multiattribute monitoring of AAV vectors. They support in-depth quality control throughout process development, formulation, and batch release to ensure vector consistency, potency, and safety.

Future trends and opportunities

Advances may include integration of native MS for intact capsid characterization, microflow and ultrahigh-throughput LC formats, AI-driven data analysis, novel stationary phases for even finer serotype discrimination, and automation to streamline QC workflows.

Conclusion

The described LC-FLD-(MS) platform delivers comprehensive assessment of AAV aggregation, full/empty capsid ratio, and capsid protein composition with high precision and sensitivity. It is adaptable across multiple serotypes, facilitating robust CQA monitoring in gene therapy development.

References

1. Backovic A.; et al. Capsid Protein Expression and Adeno-Associated Virus-like Particles Assembly in Saccharomyces cerevisiae. Microbial Cell Factories 2012, 11, 124.
2. Trilisky E. I.; Lenhoff A. M. Flow-Dependent Entrapment of Large Bioparticles in Porous Process Media. Biotechnology and Bioengineering 2009, 104, 127–133.