Mechanistic Insights into AAV Capsid–Stationary Phase Interactions Governing Native Stability and Chromatographic Separation Using AAV8 as a Model System
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe author studied the interaction of AAV to stationary phase, optimized conditions for AAV E-F separation balancing the virus capsid stability and separation efficiency, and developed new 2D in-line buffer exchange system. The manuscript is well written, and their work is informative to AAV purification workflows. There are few minor comments:
- The chromatography signals presented in Figures 7, S3, and S4 are redundant and do not provide additional information relevant to this study. UV 280 and fluorescence signals reflect protein concentration, while UV 260 is more indicative of DNA concentration. Combining these signals can allow deconvolution of protein and DNA content in each peak, and even for partially packaged AAVs. Light scattering is typically used to determine molecular mass and size of particles and is highly sensitive to aggregates. However, such analyses are not performed or discussed in the manuscript and appear outside the scope of the study. Therefore, it is not necessary to include all these data.
- The calculation of recovery using CEX-MALS is not clearly described. The manuscript states "by combining chromatographic peak areas," but it would be more helpful for readers if the authors provided more details about the calculations. Specifically, please clarify which signal was used and whether an extinction coefficient was applied. In Figure S3, the main peaks (thicker lines) appear on a changing baseline with shoulders; how the baseline was defined and how peak boundaries were determined are important for accurately calculating the recovery.
- In Figure S3, each panel contains both thicker and finer lines. Please clarify what these lines represent.
- Please clarify the exact pH conditions used in Figures S3 and S4. The text may suggest Figure S4 was performed at pH 6.5, but it is not explicit for Figure S3, which is described only as "lower pH values." Stating the precise pH used for each figure would improve clarity.
- On page 30, the figure citation at the end of the second paragraph may need to be corrected to S2.
Author Response
Comment 1: The chromatography signals presented in Figures 7, S3, and S4 are redundant and do not provide additional information relevant to this study. UV 280 and fluorescence signals reflect protein concentration, while UV 260 is more indicative of DNA concentration. Combining these signals can allow deconvolution of protein and DNA content in each peak, and even for partially packaged AAVs. Light scattering is typically used to determine molecular mass and size of particles and is highly sensitive to aggregates. However, such analyses are not performed or discussed in the manuscript and appear outside.
Response 1: We thank the reviewer for this comment. We have carefully reassessed the figures. As suggested, we have revised two of the three figures (Figures 7 and S4) and removed redundant detector signals to ensure that only data directly relevant to the study’s scope and conclusions are presented. For Figure S3, however, we have intentionally retained the original detector characteristics. The purpose of this figure is to illustrate the analytical value and complementarity of multiple detectors within the chromatographic workflow for recovery determination. Although detailed deconvolution of protein/DNA content and molecular mass determination via light scattering are beyond the scope of the present study, including these signals provides important methodological context. To ensure clarity, the function and relevance of each detector are explicitly described in the Supplementary Material, providing readers with a clear basis for interpretation.
Comment 2: The calculation of recovery using CEX-MALS is not clearly described. The manuscript states "by combining chromatographic peak areas," but it would be more helpful for readers if the authors provided more details about the calculations. Specifically, please clarify which signal was used and whether an extinction coefficient was applied. In Figure S3, the main peaks (thicker lines) appear on a changing baseline with shoulders; how the baseline was defined and how peak boundaries were determined are important for accurately calculating the recovery.
Response 2: We thank the reviewer for this comment. We have revised the manuscript to provide a clearer, more detailed description of the recovery calculation using CEX-MALS. Specifically, we now state explicitly which detector signal was used for quantification and clarify whether an extinction coefficient was applied in the calculation. It should be noted that recovery determination was not a primary focus of this study and is included solely to provide additional context and background for the presented findings. Moreover, the following sentence was added to improve clarity on this topic: “Moreover, the LS chromatographic signal of the initial harvest sample largely overlaps with that of the purified DSP sample, exhibiting minimal or no peak shoulders compared with the UV or fluorescence signals, in agreement with previous studies.”
Comment 3: In Figure S3, each panel contains both thicker and finer lines. Please clarify what these lines represent.
Response 3: We thank the reviewer for noticing this. The lines have been clarified under the Figure S3 caption.
Comment 4: Please clarify the exact pH conditions used in Figures S3 and S4. The text may suggest Figure S4 was performed at pH 6.5, but it is not explicit for Figure S3, which is described only as "lower pH values." Stating the precise pH used for each figure would improve clarity.
Response 4: We thank the reviewer for pointing out this lack of clarity. We have revised the manuscript to explicitly state the exact pH conditions used in both Figures S3 and S4. The precise pH values are now clearly indicated in the corresponding sections of the text to avoid any ambiguity. In particular, the pH used for Figure S3, previously described only as “lower pH values,” is now specified explicitly.
Comment 5: On page 30, the figure citation at the end of the second paragraph may need to be corrected to S2.
Response 5: We thank the reviewer for this comment. However, the submitted manuscript contains 29 pages, and there is no page 30 in the current version. We have carefully rechecked the figure citations throughout the manuscript and did not identify any incorrect reference to Figure S2. We would appreciate further clarification regarding the specific location, if needed.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Comments and Suggestions for AuthorsThe authors expand current knowledge on AAV purification by investigating the use of DEAE AEX for empty and full AAV8 separation. Special attention is given to understand the mechanisms behind AAV interaction with strong and weak AEX ligands. The authors also provided helpful information on purification process development with screening of buffers and pH during chromatography.
The paper presents discussions on potential stability issues of AAV at pH ~ 9.0 and low salt concentration as a reason to test weak AEX chromatography. However, very limited information is provided regarding AAV8 stability. More information by citing other studies or by conducting stability studies would help to make a stronger argument.
Page 4. Line 3. Check citation style. Make sure it is consistent on the paper.
Page 6. Section 2.3.1. Since the main focus of the paper is AAV purification, it includes more details on chromatographic methods. For example, in the passage
“In brief, samples were loaded onto the ion-exchange column and eluted using a linear gradient elution from 100% MPA to 100% MPB over 16 minutes. This was followed by a hold at 100% MPB, a high salt wash step (100% MPC), and subsequent column regeneration, equilibration, and conditioning.”
How much AAV8 material loaded or approximate range?
What is the resident time for loading and elution steps? How many monolith volumes were included in the linear gradient? Are all monolithic columns 0.1 mL?
Includes how R was calculated for resolution of empty and full AAV peaks.
Please fix “Error! Reference source not found!” throughout the manuscript.
Page 7. Please include column volume of SEC Methacrylate-2000 Å column.
For section 2.3.2. Consider adding a figure or diagram to show 2D chromatography workflow.
Page 12. In Figure 4, consider including until 3600 cm-1. Highlighting some wavenumber regions in the ATR-FTIR graphs will help the reader.
Page 15. Indicate in Figure 5, how %F was determined, meaning of %HS ad CIP.
Page 21. Consider listing other techniques that could be used to assess AAV stability. Although the lack of “extra” peaks during SEC analysis indicates no degradation, AAV transduction potency may have been affected.
Author Response
REVIEWER 2
Comments and Suggestions for Authors
The authors expand current knowledge on AAV purification by investigating the use of DEAE AEX for empty and full AAV8 separation. Special attention is given to understand the mechanisms behind AAV interaction with strong and weak AEX ligands. The authors also provided helpful information on purification process development with screening of buffers and pH during chromatography.
Comment 1: The paper presents discussions on potential stability issues of AAV at pH ~ 9.0 and low salt concentration as a reason to test weak AEX chromatography. However, very limited information is provided regarding AAV8 stability. More information by citing other studies or by conducting stability studies would help to make a stronger argument.
Response 1: We thank the reviewer for this valuable comment. While the primary focus of this study is the development and mechanistic understanding of analytical AEX separation methods rather than comprehensive formulation or long-term stability studies, we agree that a stronger contextualization of AAV8 stability under alkaline and low-conductivity conditions is important. To address this, we have substantially revised Section 3.2.2 to include additional literature reports on serotype-dependent AAV stability, with particular emphasis on AAV8 behaviour at alkaline pH and varying ionic strength. These references provide a broader framework for understanding potential stability risks and support the rationale for exploring weak AEX chromatography under milder conditions.
In addition, we performed a complementary time-dependent stability experiment under the exact analytical conditions applied in this study. A bulk-diluted AAV8 sample was incubated in AEX loading buffer and repeatedly analysed by 1D AEX after defined residence times while stored in an autosampler at 8 °C. These new data, now presented in Section 3.2.3, demonstrate that AAV8 maintains chromatographic integrity over analytically relevant time windows, with no evidence of aggregation, fragmentation, or loss of empty–full resolution. The results directly support the robustness of the applied analytical conditions and clarify that the observed separation performance is not confounded by rapid capsid degradation.
We believe that the combination of expanded literature context and experimentally verified short-term stability under the applied alkaline, low-conductivity conditions sufficiently strengthens the manuscript and addresses the reviewer’s concern, while remaining within the study’s intended analytical scope.
Comment 2: Page 4. Line 3. Check citation style. Make sure it is consistent on the paper.
Response 2: The citation style has been amended.
Comment 3: Page 6. Section 2.3.1. Since the main focus of the paper is AAV purification, it includes more details on chromatographic methods. For example, in the passage
“In brief, samples were loaded onto the ion-exchange column and eluted using a linear gradient elution from 100% MPA to 100% MPB over 16 minutes. This was followed by a hold at 100% MPB, a high salt wash step (100% MPC), and subsequent column regeneration, equilibration, and conditioning.”
Response 3: We agree that a more detailed description of the chromatographic methods is essential for clarity, reproducibility, and accurate interpretation of the results. We have therefore maintained an appropriate level of methodological detail to fully support the study.
Comment 4: How much AAV8 material loaded or approximate range?
Response 4: This information was added to the text.
Comment 5: What is the resident time for loading and elution steps? How many monolith volumes were included in the linear gradient? Are all monolithic columns 0.1 mL?
Response 5: The resident time for loading and elution steps was revised and added to the text accordingly.
Comment 6: Includes how R was calculated for resolution of empty and full AAV peaks.
Response 6: Information about R was also included and added to the text.
Comment 7: Please fix “Error! Reference source not found!” throughout the manuscript.
Response 7: We carefully reviewed the manuscript and could not identify any “Error! Reference source not found!”. All references appear to be correctly formatted. We would be grateful for clarification on the exact location of the reported error.
Comment 8: Page 7. Please include column volume of SEC Methacrylate-2000 Å column.
Response 8: The column volume of the SEC column, as well as its additional characteristics, were appropriately included.
Comment 9: For section 2.3.2. Consider adding a figure or diagram to show 2D chromatography workflow.
Response 9: We thank the reviewer for this helpful suggestion. To improve clarity and facilitate understanding of the analytical setup, we have added a schematic diagram illustrating the 2D chromatography workflow. This figure provides a clear overview of the system configuration and experimental sequence.
Comment 10: Page 12. In Figure 4, consider including until 3600 cm-1. Highlighting some wavenumber regions in the ATR-FTIR graphs will help the reader.
Response 10: We thank the reviewer for a very helpful comment. In response, we have extended the spectral range in Figure 4 to 3600 cm⁻¹. In addition, key wavenumber regions relevant for interpretation have been highlighted in the ATR-FTIR spectra to improve clarity and guide the reader.
Comment 11: Page 15. Indicate in Figure 5, how %F was determined, meaning of %HS and CIP.
Response 11: We have revised Figure 5 and its caption to clearly indicate how %F was determined and to define the meaning of %HS and CIP. These explanations are now explicitly provided to ensure transparency and facilitate reader understanding.
Comment 12: Page 21. Consider listing other techniques that could be used to assess AAV stability. Although the lack of “extra” peaks during SEC analysis indicates no degradation, AAV transduction potency may have been affected.
Response 12: Thank you so much for your helpful suggestion! We agree that exploring additional techniques could really enhance the assessment of AAV stability. We mentioned some of these methods to give helpful context (see Section 3.2.3). However, evaluating biological potency—like how well the virus transduces cells—was beyond what we aimed to cover in this study, which mainly focused on analytical separation methods and physicochemical characterisation. So, our conclusions about stability are limited to structural and chromatographic integrity. We even added a paragraph to clarify this: “Nonetheless, subtle effects on biological activity, such as transduction efficiency—particularly relevant at preparative scales—cannot be excluded and were beyond the scope of this work.”
Author Response File: Author Response.docx
