Human Gene Therapy - April 2023 - 336

336
TROXELL ET AL.
dilutions. Except for one sample in the dilution curve, the
interassay precision was £15% for both capsid and genome
concentrations (Table 4).
Compared to the capsid and genome concentrations
obtained by SEC-MALS, mean CP/mL values were within
30% across the sample curve; however, VG/mL values
were within 20% and ddPCR was capable of accurately
quantifying the concentration of the target genome
within the empty preparation (Table 4). The full-toempty
ratio for the full preparation was 90% and strikingly
close to the SV-AUC value of 89.9%. Apart from
one sample, the full-to-empty ratio was within 30% of the
SV-AUC. ddPCR/ELISA results were linear compared to
the SV-AUC value (R2 = 0.933; Fig. 5). Compared to the
data from SEC-MALS, all samples tested by ddPCR/
ELISA were within 30% of the full-to-empty ratio
(Supplementary Table S4).
Bulk absorbance methods have been used to quantify
AAV capsid and genomes.17 The Stunner instrument from
Unchained Labs offers the combination of UV/Vis measurements
coupled with DLS and SLS. Similar to SVAUC,
cryo-EM, and SEC-MALS, this approach is capable
of measuring multiple CQAs. A major benefit of the
Stunner instrument is a small sample requirement (2 lL)
with minimal sample preparation, which allows for an
orthogonal method to titration assays with minimal impact
on the sample consumption. With the exception of the full
sample, all full-to-empty values obtained with the Stunner
were more than 30% lower than the corresponding SVAUC
value (Supplementary Table S4). Because the fullto-empty
value for SEC-MALS, ddPCR/ELISA, and
Stunner are based on the measurement of both capsid and
genome concentrations, measurements from these three
approaches were compared.
The VG/mL values across all three methods were very
similar with each level from SEC-MALS and Stunner
within 20% of the DNase-resistant ddPCR value (Table 4).
This indicates that each method is accurate for the quantification
of AAV genomes with purified samples. The
concentration of capsids with these three methods exhibited
more variance. The CP/mL values with each level
from SEC-MALS and Stunner were within 50% of the
ELISA value (Table 4). SEC-MALS measurements were
closer to the ELISA values and Stunner CP/mL measurements
tended to overestimate the number of capsids
present. Nevertheless, full-to-empty data from Stunner
were linear when compared to SV-AUC (R2 = 0.961;
Fig. 5).
Stability-indicating performance
of SEC-MALS and ddPCR/ELISA
Given the alignment ofSEC-MALS with SV-AUC fullto-empty
data and the ability to quantify CP/mL and VG/
mL, we conducted a series of experiments to assess the
method's stability-indicating performance. To test this, the
Table 5. Size exclusion chromatographywith UVand multiangle
light scattering results from pH stress compared with other
stability-indicating assays
SEC-MALS
Treatment
Neutralized
Control
4.60E+13
0.01 N HCl 2.72E+13
0.01 N NaOH 1.24E+13
CP/mL %CV
3%
1%
10%
VG/mL %CV VG/CP Ratio
4.29E+13 3% 0.933
2.37E+13 1% 0.873
4.14E+12 5% 0.336
ddPCR/ELISA
Treatment Mean CP/mL %CV Mean VG/mL %CV VG/CP Ratio
Neutralized
Control
4.54E+13
0.01 N HCl 3.35E+13
0.01 N NaOH 2.75E+13
TCID50
TCID50/mL
Treatment
Neutralized
Control
Operator 1 Operator 2 Mean TCID50/mL %CV Mean P:I Ratio
4.93E+08
1.16E+09
0.01 N HCl 2.57E+08
4.33E+08
0.01 N NaOH <8.00E+06 <8.00E+06
8.28E+08 57% 7.13E+04
3.45E+08 36% 1.24E+05
<8.00E+06 N/A >5.00E+06
P:I, particle to infection; TCID50, 50% tissue culture infectivity.
STRX-330 material used for the non-GLP animal study
was subjected to pH stress at room temperature for 3 h.
Aliquots of the test article were treated with HCl or NaOH
to a final concentration of 0.01 N. As a control, the concentrated
HCl and NaOH were premixed to neutralize and
then added to a separate aliquot. After 3 h of incubation,
the opposing treatment was added to neutralize the pH and
stop the stress. For instance, the 0.01N HCl treated sample
received equimolar NaOH to neutralize. Then, the samples
were tested by SEC-MALS, ddPCR/ELISA, and infectivity
was assessed by TCID50.
TCID50 and ddPCR/ELISA were used since these are
known to indicate capsid function and can be compared
with results from SEC-MALS. As expected, the control
treated material exhibited SEC-MALS profile similar to
untreated and earlier tested samples (Fig. 6A). However,
treatment with either HCl or NaOH resulted in very different
chromatograms. Specifically, the HCl treatment
produced two major peaks; one corresponding to the
monomer peak and the other corresponding to an aggregated
peak (Fig. 6B). Treatment with NaOH dramatically
altered the SEC elution profile of AAV with a
pronounced reduction in the monomer peak. Instead, a
prominent peak eluting slightly before the typical aggregated
peak followed by a minor monomer peak was
observed (Fig. 6C).
In response to alkaline stress, there was a clear shift in
the population of nucleic acid content (determined from
7%
13%
6%
4.58E+13 3% 1.008
2.61E+13 2% 0.779
1.73E+12 11% 0.063
Human Gene Therapy - April 2023

Table of Contents for the Digital Edition of Human Gene Therapy - April 2023
