Chronologically distributed transfection improves AAV2 and AAV2/8 capsid filling and reveals assembly schedule divergence

Figures

• 

  Figure 1. IIlustrative diagrams of AAV genome and capsid production and assembly plasmids and transfection timelines (1) Simplified diagrams of Payload (red), RepCap (blue), and Helper (green) plasmids. (2) Occam schema of zeta (ζ) procedures, leading to correct payload genome production, and theta (θ) procedures, which require at least one payload genome in order to commence, for production of capsids correctly loaded with payload genomes. (3) Diagram of time points for the transfection chronologies used in this study. At 0 h, host cells are seeded, with transfection 1 (T1) 24 h post-seeding, T2 44 h post-seeding, and harvest 96 h post-seeding. The duration of time periods between those four events are also depicted in hours and as a percentage of the 96 h total. (4) Graphical depictions of the eight transfection chronologies used in this study and their predicted consequences. For chronologies A–F the following applies: plasmids are depicted as being partially opaque at T2 if they were present at T1. Payload genomes or correctly packaged capsids are absent if their formation is not conventionally understood to be likely or possible. A prediction of empty capsid formation is depicted for…

• 

  Figure 2. AAV2 and AAV2/8 viral genome yield performance as a function of transfection chronologically Viral genome titers were determined by qPCR for AAV2 and AAV2/8 crude lysates arising from the transfection chronologies illustrated in Figure 1.4 by letter code, at 10 mL and 1 mL scales, respectively. Plots (A) and (C) show AAV2 results on log and linear scales respectively, while plots (B) and (D) show AAV2/8 results on log and linear scales, respectively. Crude lysate from non-transfected cells (Ntf) were used as qPCR template and interpreted as the limit of quantitation (LOQ) for the assay. The dotted line highlights the lowest yield achieved by any transfection chronology. Error bars represent standard deviation of three biological replicates.

• 

  Figure 3. AAV2 and AAV2/8 transducing unit yield performance as a function of transfection chronologically Volumetric transducing unit titers (TU/mL) were determined by flow cytometry of target cells, to detect GFP expression, for AAV2 and AAV2/8 crude lysates arising from the Figure 1.4 transfection chronologies, at 10 mL and 1 mL scales, respectively. Plots (A) and (C) show AAV2 results on log and linear scales, respectively, with the equivalent for plots (B) and (D) for AAV2/8. Crude lysate from non-transfected cells (Ntf) was incubated with target cells and the resulting fluorescence signal interpreted as the LOQ for the assay. The dotted lines and plots in (A) and (B) highlights the lowest yield achieved by any transfection chronology. Error bars represent standard deviation of two to four biological replicates.

• 

  Figure 4. Transfection chronology F produces particles of standard morphology for AAV2 and AAV2/8 Transmission electron microscopy (TEM) analysis of AAV2 and AAV2/8 viral particles. Representative TEM micrographs showing AAV2 (A and D) and AAV2/8 (C and F) particles. Benchmark (Bmk) preparations of both serotypes (A and C) show characteristic ∼28 nm spherical capsids. AAV2 produced using transfection chronology F (D) (labeled “2F”) yielded intact particles similar to benchmark. AAV2/8 produced using transfection chronology F (F, labeled “8F”) also demonstrated morphology consistent with the benchmark preparation. (B) (labeled “X”) is the HEK293F cell lysate that had undergone the same production and sample preparation process as Bmk but without the addition of plasmids and (E) (labeled “Y”) is PBS. No particle structures were observed for either. Scale bars represent 200 nm and is applicable to all panels.

• 

  Figure 5. Correctly assembled AAV2 capsid yield and filling as a function of transfection chronology (A) is a plot of ELISA-based quantification of total correctly assembled capsids (capsids/mL) in crude lysates arising from the Figure 1.4 AAV2 transfection chronologies, performed at 10 mL scale. The dotted line highlights the lowest measurable yield achieved. Non-detectable (N.D.) signal resulted from the indicated chronologies. ELISA data were combined with the AAV2 qPCR data plotted in Figure 2 to calculate inferred levels of capsid filling with payload genomes. The full/empty capsid ratios (qPCR/ELISA) were then plotted in (B), for the chronologies with measurable capsids/mL. Asterisk (∗) denotes statistical significance (p < 0.05). Error bars represent standard error of the mean from four biological replicates.

• 

  Figure 6. Inferred levels of correct capsid filling using the ELISA-free A-CUE procedure The novel, “A-CUE”, ELISA-free procedure for inferring the level of correct capsid filling with payload genomes is set out in detail in the results section. (A) illustrates a workflow whereby the A-CUE method can be validated against ELISA for AAV2. In (i), comparison of Figure 5A ELISA data and a western blot image arising from the A-CUE procedure in (C) reveals matching patterns of signal as a function of transfection chronology. As such, when western blot images (C) are quantified by antibody binding signal in a given lane (step ii), they can be combined with Figure 2 qPCR data (step iii) to calculate an inferred level of correct capsid filling. (B) plots this inferred correct filling level as a function of transfection chronology, giving a comparable AAV2 filling pattern to Figure 5B. (C) is the western blot image, annotated by red, purple, and black left-pointing triangles, to indicate the VP1, VP2, and VP3 bands. These sizes were inferred by comparison with band migration in the ladder (unlabeled, leftmost lane). (D) is the A-CUE data analysis workflow for AAV2/8, which omits step (i) due to the…