Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 May 4.
Published in final edited form as: J Virol Methods. 2022 Aug 5;309:114598. doi: 10.1016/j.jviromet.2022.114598

Optimized formulation buffer preserves adeno-associated virus-9 infectivity after 4 °C storage and freeze/thawing cycling

Angela Chan 1,*, Carola J Maturana 1, Esteban A Engel 1
PMCID: PMC10157798  NIHMSID: NIHMS1891700  PMID: 35940276

Abstract

Adeno-associated virus (AAV) have long been one of the most common and versatile vectors for in vitro and in vivo gene transfer. AAV production protocols are complex and time consuming, one key concern is the recovery and infectivity of viral vector after purification. The buffer used in the storage of AAV at 4 °C and −80 °C is a crucial factor and methods to improve it have been thoroughly investigated. Viral core facilities have developed formulas using either 0.001% Pluronic F68 or 5% sorbitol in their storage buffers based on the results of this research. Interestingly, few use formulations that include both a non-ionic surfactant and cryopreservative. In this study, AAV9 stored at 4 °C and at −80 °C in the standard buffers is compared to a buffer that contains 5% glycerol and 0.001% Pluronic F68. By viral genome quantitation with qPCR, all three formulations show the same extent of viral titer loss at 4 °C, while after several cycles of freeze/thaws at −80 °C, the viral recovery and infectivity in the preparation with both glycerol and Pluronic F68 was most stable compared to the other buffers.

Keywords: AAV, Sorbitol, Glycerol, Pluronic F68, Formulation buffer

In recent years adeno-associated virus (AAV) vectors have been successfully used for gene transfer. It was first discovered in the 1960’s and since then there have been significant improvements on its design, gene expression, and in vivo and in vitro applications (Srivastava et al., 2021; Wang et al., 2019). AAVs are part of the Parvoviridae family and are non-enveloped viruses composed of single strands of DNA, encapsulated in small, ~25 nm, icosahedral protein capsids. AAV production is increased when using a helper-free system in which three plasmids are co-transfected in mammalian HEK 293 cells. These plasmids consist of 1) a transgene with a gene promoter, polyA sequences and flanking inverted terminal repeats (ITRs) that function as the origin of replication, 2) contain the rep and cap genes that are involved in the assembly of the virus, and 3) composed of adenoviral helper genes required for AAV replication.

As the demand for AAVs continue to grow, storage buffer formulas have been refined to increase viral stability and infectivity. It has been demonstrated that pH (Bee et al., 2022a; Croyle et al., 2001, 1998; Evans et al., 2004; Pacouret et al., 2017; Rodrigues et al., 2018), ionic strength (Hoganson et al., 2002; Rodrigues et al., 2018; Wright et al., 2005), cryoprotectors and non-ionic surfactants (Bee et al., 2022b, 2022a; Bennicelli et al., 2008; Evans et al., 2004; Hoganson et al., 2002; Pacouret et al., 2017; Patrício et al., 2019; Rodrigues et al., 2018; Van Den Berg and Soliman, 1969; Wright et al., 2005; Xu et al., 2022) in buffer formulations can prevent aggregation and loss of infectivity of the virus. Furthermore, the type of storage buffer can optimize AAV transduction and transgene expression (Bennett et al., 2017; Croyle et al., 2001, 1998; Hoganson et al., 2002; Pacouret et al., 2017). Previous works have evaluated the viability of AAV when stored at 4 °C as well as −80 °C, however discrepancy in the findings shows that the AAV production is an iterative process and the development of new formulations that improve the viability are required (Bee et al., 2022b, 2022a; Chen et al., 2006; Croyle et al., 2001, 1998; Gruntman et al., 2015; Hoganson et al., 2002; Howard and Harvey, 2017; Rieser et al., 2022; Xu et al., 2022). Based on these studies, viral core facilities have developed their own storage buffer formulations. Here, three different PBS-based preparations were examined and the stability of AAV9 during storage at 4 ° C and at −80 °C was analyzed. Most of these facilities use PBS and 0.001% Pluronic F68 to preserve AAV. Pluronic F68 is a non-ionic surfactant that has been shown to prevent loss of titer due to contact with surfaces and has been FDA approved for human use (Bee et al., 2022a; Bennicelli et al., 2008; Pacouret et al., 2017; Patrício et al., 2019; Wright et al., 2005). Other viral facilities use PBS and 5% sorbitol as their storage buffer. Sorbitol is one of several cryoprotectants used to inhibit loss of virus after freezing (Croyle et al., 1998; Pacouret et al., 2017; Wright et al., 2005). It is noteworthy that these formulas use either a non-ionic surfactant or a cryoprotectant, but not both. Recent publications have revealed that buffers containing both protect AAV capsids from rupture during several cycles of freeze/thaws (Bee et al., 2022b, 2022a; Xu et al., 2022). In this current study, we investigate whether a PBS buffer that consists of both a cryoprotectant and Pluronic F68 would prevent loss of AAV9 comparable to the two buffers mentioned. AAV9, with a broad tissue tropism, is one of the most commonly studied serotypes in the research and clinical setting (Kuzmin et al., 2021; Schuster et al., 2014). It has also been shown by differential scanning fluorimetry (DSF) to have very little TM variation between different types of buffers (Bennett et al., 2017). Glycerol was selected as the cryopreservative as it has traditionally been used as such for decades and have been investigated intensively in regards to AAV stability (Croyle et al., 2001; Evans et al., 2004; Hoganson et al., 2002; Van Den Berg and Soliman, 1969; Wang et al., 2019; Xie et al., 2004; Xu et al., 2022). It should be noted that although PBS is the base of the formulation, the concentration of NaCl is different in each of these buffers. The buffers used in this work were labeled as follows: buffer A (172 mM NaCl, 0.001% Pluronic F68, 5% glycerol), buffer B (337 mM NaCl, 0.001% Pluronic F68), and buffer C (350 mM NaCl, 5% sorbitol). The analysis shows that after the first week, the stability of AAV9 in all three buffers at 4 °C remain unchanged but that after a sequence of freeze/thaws at −80 °C, a formulation containing both types of excipients improved the stability and reduced the loss of AAV9.

To study the effect of these buffers on AAV titer, a triple transfection with AAV9 serotype (University of Pennsylvania Vector Core), EF1α-GFP plasmid (Addgene 60058), and pHelper vector (Agilent) were transfected into HEK 293 cells using PEI Max (Polyscience). AAV production and purification was performed as previously reported, with slight modifications (Maturana et al., 2022). Briefly, cells were collected 72 h post transfection and pelleted by centrifugation at 3000 x g for 5 min at RT. The supernatant was discarded, and the pellet was stored at −80 °C. For AAV purification, cells were lysed by freezing in liquid nitrogen and thawing at 45 °C three times. Turbonuclease (250 U/μl, BPS Bioscience) was added to the lysate and incubated at 37 °C for 30 min. Supplemented with 5 M NaCl, the mixture was centrifuged at 10,000 x g for 30 min at 4 °C. The supernatant was layered onto an Iodixanol density step gradient (Zolotukhin et al., 1999; Addgene, [WWW Document]) and was then centrifuged at 350,000 x g for 75 min at 18 °C. The viral particles were removed from the 40% iodixanol portion of the gradient and divided into 3 equal aliquots. A different PBS (Hyclone) formulation buffer was added to each of the aliquots. Buffer A: 1X PBS, 172 mM NaCl, 0.001% Pluronic F68 (Sigma), 5% glycerol; Buffer B: 1X PBS, 337 mM NaCl, 0.001% Pluronic F68; or Buffer C: 1X PBS, 350 mM NaCl, 5% sorbitol. Each AAV aliquot was further purified and concentrated through a Vivaspin 20 column (GE Healthcare, 100, 000 MWCO). The virus was recovered and 50 μl was aliquoted into five 0.5 ml silicone tubes (BioPlas). Four tubes were stored at −80 °C and the fifth tube was stored at 4 °C. The AAV9 purification was repeated on 3 separate occasions to generate 3 biological replicates for each buffer (Fig. 1A).

Fig. 1.

Fig. 1.

Open in a new tab

Workflow for comparative analysis of the stability of AAV9 in different storage conditions. A) AAV9 production. GFP was packaged in AAV9 serotype under control of the EF1α promoter. The first step in packaging AAV9 is a triple transfection of HEK 293 cells with AAV9, pEF1α-GFP and pHelper vector. Three days after co-transfection, the AAV-containing cells are harvested and purified. The viral particles were removed from the 40% iodixanol gradient and divided into three equal aliquots. A different PBS formulation buffer (A, B, C) was added to each of the aliquots. B) Experimental condition. After concentration of each aliquot, the AAV9 was separated into five tubes. Four tubes were stored at −80 °C and one tube at 4 °C. For each freeze/thaw cycle, four vials at −80 °C were removed, thawed at RT for 1 h and then replaced at −80 °C for 24 h. Similarly, with the 4 °C stability experiment, an aliquot of AAV9 in the three different buffers was stored at 4 °C for 3 weeks. Each week thereafter, for the next 3 weeks, 5 μl was removed for qPCR. Comparative analysis of AAV9 titer was assayed by qPCR. The image was created with biorender.com.

To determine the amount of titer loss, the initial genome copy number of AAV9 was quantified by qPCR. Each sample of 5 μl was treated with DNaseI (Roche) for 15 min at 37 °C and then placed at 95 °C for 10 min. The DNaseI treated samples were diluted 1:1000 and with the addition of a Taqman probe against WPRE used in the qPCR reaction. The samples were measured in a Thermo Fisher QuantStudio 3 System and the titer in genome copy per ml (gc/ml) was calculated. The subsequent titers from the stability experiments were compared to this original starting titer. For each freeze/thaw trial 4 vials stored at −80 °C were removed, thawed at RT for 1 h, then replaced at −80 °C for 24 h. The cycle was repeated until the final freeze/thaw was completed for a total of 9 freeze/thaws. After the final freeze/thaw, 5 μl was quantitated by qPCR to determine titer at 1, 3, 6, and 9 freeze/thaws. Similarly, with the 4 °C stability experiment, an aliquot of AAV9 in the three different buffers was stored at 4 °C for 3 weeks. Each week thereafter, for the next 3 weeks, 5 μl was removed and the titer was assayed and compared to the initial titer (Fig. 1B).

Additionally, the purity of the AAV9 was determined by running the freeze/thaw samples of each buffer on a SDS polyacrylamide gel. Labs routinely silver stain polyacrylamide gels to visualize capsid proteins of AAV (Sonntag et al., 2011; Steinbach et al., 1997; Wang et al., 2019). There are 3 subunits termed VP1, VP2, and VP3 that form these proteins, with molecular weights of 87 kDa, 73 kDa, and 62 kDa respectively at a 1:1:10 ratio. The capsid proteins (5 μl) were separated in a 4–12% Bolt Bis-Tris polyacrylamide gel (Thermo Fisher). The gel was run in MOPS buffer for 32 min at 200 volts and a SilverXpress staining kit (Thermo Fisher) was used to detect the protein bands.

To investigate whether the changes in stability of AAV stored at −80 °C could be recapitulated in live cells, Neuro-2a (N2a) cells, a mouse neuroblastoma cell line that exhibit brain neuronal cell morphology, were used for a transduction assay. The cells were grown in DMEM with 2% FBS and 1% penicillin-streptomycin. N2a cells were seeded onto 12 well tissue culture plates (Fisher Scientific) with each row transduced with 4 μl AAV9 from each buffer and the number of freeze/thaws. This was repeated for each biological replicate. The plates were incubated at 37 °C with 5% CO2 for 3 days and GFP fluorescent cells was imaged with a Nikon Ti-E inverted epifluorescence microscope (Nikon Instruments), with a CoolSNAP ES2 camera (Photometrics) and the Nikon NIS-Elements software. To quantify the efficiency of AAV9 GFP expression after Neuro-2a infection, we marked cells with every image (average 5000 cells per image) obtained from three replicate dishes per freeze/thaw conditions. Cell counting was preformed manually using an ImageJ plugin Cell Counter. Each click marked the cell with a colored number, while the cells expressing GFP were counted in a separate group with a different colored number to determine the percentage of GFP positive cell from the total cell population for each dish.

The stability of AAV9 at −80 °C was determined by comparing the titer after several cycles of freeze/thaws to the original titer. It should be noted that the starting titer for buffer C (1X PBS, 350 mM NaCl, 5% sorbitol) was approximately half a log lower than buffers A (1X PBS, 172 mM NaCl, 0.001% Pluronic F68, 5% glycerol) and B (1X PBS, 337 mM NaCl, 0.001% Pluronic F68) (Fig. 2A). Since AAV9 was equally aliquoted into each buffer before further purification through columns, it is postulated that the AAV adhered to the surface of the column in the buffer without Pluronic F68. Alternatively, sorbitol could have influenced the recovery of the titer itself. The results show that after an initial decline in titer after one freeze/thaw, AAV9 in buffer A and C remained constant, whereas the titer in buffer B significantly dropped at the ninth freeze/thaw when compared to 0 (p < 0.002) and 1X (p < 0.033) freeze/thaws (Fig. 2A). The average percent change from the starting AAV9 titer was then calculated. It was determined that the titer in buffer A fell 32% at the first freeze/thaw and then remained steady after subsequent freeze/thaws. Although the titer in buffer B consistently dropped, starting at a decrease of 24% from the starting titer at one freeze/thaw, it was not significant until the final loss of 88% at the ninth freeze/thaw. The titer in buffer C remained unchanged at the first freeze/thaw, after which it decreased to 40% at the third freeze/thaw, where it stabilized thereafter. Interestingly, when the individual data points of the percentage change were studied, greater variability in titer loss was seen in buffers B and C than in A (Fig. 2B). To determine if freeze/thaw influenced the AAV capsid integrity, AAV9 was run on polyacrylamide gels and developed with silver stain. From the results the three subunits, VP1, VP2, and VP3 were intact and the ratio between subunits for each freeze/thaw of all three buffers were equivalent to one another (Fig. S1). Thus, from the data we can infer that having a cryoprotectant in the formulation is more important to the stability of AAV9 but the inclusion of Pluronic F68 also augments it.

Fig. 2. :

Fig. 2. :

Open in a new tab

Comparative analysis of the stability of AAV9 after freeze/thaw cycles in the different formulation buffers, AAV9 titer quantitated by qPCR after 1, 3, 6, and 9 freeze/thaw cycles at −80 °C, representing 3 independent experiments. A) AAV titer (gc/ml) in buffer A (black), buffer B (orange), and buffer C (blue). B) Plotted graph of percentage change compared to starting titer (dotted line) in buffer A (black), buffer B (orange), and buffer C (blue). Statistical analysis was performed using GraphPad Prism Statistics Software version 9.0. A one-way ANOVA test with Tukey’s post-hop was performed for multiple comparisons. All data represented as Mean ± SEM with *p < 0.033, **p < 0.002, ***p < 0.001. Buffer A: 1X PBS, 172 mM NaCl, 0.001% Pluronic F68, 5% glycerol; Buffer B: 1X PBS, 337 mM NaCl, 0.001% Pluronic F68; Buffer C: 1X PBS, 350 mM NaCl, 5% sorbitol.

In the assessment of AAV9 stability in the three buffers at 4 °C, one vial of each was stored at 4 °C for 3 weeks. The titer was quantitated by qPCR each week and compared to the starting titer. An immediate, though not statistically significant, decrease in titer was observed in all three buffers after storage at 4 °C for 1 week. The titers remained consistent over the following weeks (Fig 3A). When comparing the average percentage change of the titers to the starting titer, it was determined that all three buffers lost approximately 40% after the first week and then remained constant for the succeeding weeks. Examination of the individual data points of percentage change, showed that there was wider spread of variability in buffers B and C than in buffer A, comparable to the freeze/thaw experiments (Fig. 3B, Table 1). These results would suggest that using both glycerol and Pluronic F68 also improves the stability of AAV9 at 4 °C.

Fig. 3. :

Fig. 3. :

Open in a new tab

Comparative analysis of the stability of AAV9 at 4 °C in the different formulation buffers, AAV9 titer quantitated by qPCR after 3 weeks of storage at 4 °C; representing 3 independent experiments. A) AAV titer (gc/ml) in buffer A (black), buffer B (orange), and buffer C (blue). B) Plotted graph of percentage change compared to starting titer (dotted line) in buffer A (black), buffer B (orange), and buffer C (blue). Statistical analysis was performed using GraphPad Prism Statistics Software version 9.0. A one-way ANOVA test with Tukey’s post-hop was performed for multiple comparisons. All data represented as Mean ± SEM with *p < 0.033, **p < 0.002, ***p < 0.001. Buffer A: 1X PBS, 172 mM NaCl, 0.001% Pluronic F68, 5% glycerol; Buffer B: 1X PBS, 337 mM NaCl, 0.001% Pluronic F68; Buffer C: 1X PBS, 350 mM NaCl, 5% sorbitol.

Table 1:

Summary of Experimental Results for each Formulation Buffer

−80°C Freeze/Thaw Cycle - Titer Buffer A Buffer B Buffer C
1 NC NC NC
3 NC NC NC
6 NC NC NC
9 NC decrease NC
−80°C Freeze/Thaw Cycle - % Change (Variability)
1 NC Increase Increase
3 NC Increase Increase
6 NC Increase Increase
9 NC Increase Increase
 
4°C Storage (Weeks) - Titer
1 NC NC NC
2 NC NC NC
3 NC NC NC
4°C Storage (Weeks) - % Change (Variability)
1 NC Increase Increase
2 NC Increase Increase
3 NC Increase Increase
Open in a new tab

NC: No significant change from the starting buffer; Buffer A: 1X PBS, 172mM NaCl, 0.001% Pluronic F68, 5% Glycerol; Buffer B: 1X PBS, 337mM NaCl, 0.001% Pluronic F68; Buffer C: 1X PBS, 350mM NaCl, 5% Sorbitol

The effect of each buffer on AAV stability was evaluated in vitro. AAV9 was transduced into N2a cells and cells expressing GFP were counted after 3 days (Fig. S2). In buffer A, a significant reduction of the number GFP cells was observed after the first freeze/thaw compared to the rest (3X, p < 0.33; 6X, p < 0.002; 9X, p < 0.002). However, in the subsequent freeze/thaw cycles, the percentage of GFP cells seem to have stabilized. Although there was also a loss in GFP cells from the first to third freeze/thaw round in buffer B, it was insignificant. Yet, there was a considerable decrease at the sixth freeze/thaw compared to 1X (p < 0.001) and 3X (p < 0.002). Interestingly, there was a notable increase of GFP cells at the ninth freeze/thaw, nonetheless, the change was minor when compared to 1X (p < 0.002). Like buffer A, in buffer C, there was diminished number of GFP cells after the first freeze/thaw compared to the other rounds (3X, p < 0.033; 6X, p < 0.002; 9X, p < 0.002), but the cycles thereafter displayed consistent number of GFP cells (Fig. 4A). From the average percentage change, it was determined that there was approximately 25% reduction in GFP expressing cells in buffer A and B and a 40% drop in buffer C after the first freeze/thaw. There was another 10% decrease in expression at the sixth freeze/thaw and no change at the ninth in buffer A. In contrast, it was observed that the percentage dropped to 70% at the sixth cycle in both buffers B and C. (Fig. 4B). The data demonstrates that in the N2a cell lines, there was moderate loss of AAV9 infectivity and greater stability in the formulations consisting of glycerol and Pluronic F68.

Fig. 4. :

Fig. 4. :

Open in a new tab

Evaluation of AAV9 GFP expression in Neuro2A cell line with the different formulation buffers, Mouse neuroblastoma cell line, Neuro2a were transduced with AAV9 from each formulation buffer with 1, 3, 6, and 9 freeze/thaw cycles at −80°C. A) The percent of GFP expressed cells from 3 independent experiments were counted from buffer A (black), buffer B (orange), and buffer C (blue). B) Plotted graph of percentage change after the 1st freeze/thaw (dotted line) for buffer A (black), buffer B (orange) and buffer C (blue). Statistical analysis was performed using GraphPad Prism Statistics Software version 9.0.1. A one-way ANOVA test with Tukey’s post-hoc was performed for multiple comparisons. All data represented as Mean ± SEM with *p < 0.033, **p < 0.002, ***p < 0.001. Buffer A: 1X PBS, 172 mM NaCl, 0.001% Pluronic F68, 5% glycerol; Buffer B: 1X PBS, 337 mM NaCl, 0.001% Pluronic F68; Buffer C: 1X PBS, 350 mM NaCl, 5% sorbitol.

Although there have been numerous studies about the components in AAV formulation, they do not directly examine the buffers that are used today in viral core facilities. In general, the majority use PBS and 0.001% Pluronic F68 as the AAV storage buffer, though there are a few that use PBS and 5% sorbitol. It was hypothesized that a formulation containing both a cryoprotector and a non-ionic surfactant may improve AAV recovery. Since both FDA approved gene therapy drugs Luxturna and Zolgensma (LUXTURNA® (voretigene neparvovec-rzyl) HCP [WWW Document]; ZOLGENSMA® (onasemnogene abeparvovec-xioi) [WWW Document]) use Poloxamer 188 (Pluronic F68) to store their AAV and it has been shown to be efficacious in vector delivery in dogs and mice (Bennicelli et al., 2008), the same non-ionic surfactant was used in the buffers tested. In considering the cryoprotector, glycerol was elected instead of sorbitol. Glycerol has been used extensively in human clinical trials in patients with stroke, cerebral edema and Reye’s syndrome, suggesting that small amounts of glycerol are safe for human use and gene therapies (Frank et al., 1981; Sloviter, 1958; Wang et al., 2021). Researchers have also demonstrated that the addition of 2.5% glycerol to high ionic strength Tris buffer can prevent AAV aggregation and provide long term stability of AAV at > −65 °C and 2–8 °C (Hoganson et al., 2002).

PBS, the base of these buffers, is supplemented with NaCl as studies have shown that the high ionic strength of a buffer can prevent AAV aggregation (Hoganson et al., 2002; Rodrigues et al., 2018; Wright et al., 2003). In fact, buffer B and C have an additional 200 mM and 213 mM NaCl respectively added to PBS. For buffer A, a lower concentration of NaCl (37 mM) was added to approximate the amount used in the application of AAV in Luxturna (LUXTURNA® (voretigene neparvovec-rzyl) HCP [WWW Document]). The reduced concentration of NaCl in buffer A did not adversely affect the recovery of AAV9 titer as the starting titer was comparable to buffer B, which contained almost twice the amount of NaCl (Fig. 2A and 3A). Other labs have also used formulation buffers containing less than 200 mM NaCl and have not observed loss in viral recovery (Bee et al., 2022b, 2022a; Rieser et al., 2022).

When the starting titers were quantified by qPCR, an unexpected lower yield of AAV9 was observed in the buffer that contained 5% sorbitol (buffer C) in all three purifications (Fig. 2A and 3A). When Wright and collaborators in 2005 examined AAV recovery in formulations consisting of different ionic strengths it was compared to a control that was of low ionic strength with 5% sorbitol. The authors noted that AAV recovery in the control buffer was significantly decreased compared to other formulas, concluding that this effect was due to low ionic strength of the buffer. It would be interesting to explore whether sorbitol also contributed to the low recovery. Howell and Miller in 1983 found that in 4% sorbitol, there was a greater loss of cytomegalovirus (CMV) and varicella-zoster virus (VZV) infectivity when evaluated against a buffer containing sucrose during long term storage at 4 °C and at −20 °C. This may explain the data that is seen at −80 °C and 4 °C, where AAV9 recovery was less than expected.

The non-ionic surfactant, Pluronic F68, was added to buffers A and C. This is a customary practice in many viral core facilities. From the experiment assessing titer loss after several freeze/thaws at −80 °C, it was observed that after an initial drop, AAV9 titer remained steady in buffers that also contained a cryoprotectant when compared to a buffer with only Pluronic F68 (Figs. 2B and 4B; Table 1). This agrees with research that looked at AAV stability in the presence of cryoprotectants (Croyle et al., 2001, 1998; Evans et al., 2004; Hoganson et al., 2002; Rodrigues et al., 2018). In several recent 2022 publications, researchers exposed AAV serotypes to several consecutive freeze/thaws in either a DPBS buffer consisting of 0.001% Pluronic F68 or 0.001% Pluronic F68 and a cryoprotectant such as sucrose and determined that the AAVs were more stable in buffers that contained the cryoprotectant (Bee et al., 2022b, 2022a; Xu et al., 2022). It was also noted that there was an increase of free single stranded DNA from sheared capsids in the buffer with Pluronic F68 only, leading them to surmise that the addition of a cryoprotectant shields the AAV capsids from rupture during the freeze/thaw cycles. Some have reported that non-ionic surfactants were not effective in preventing aggregation (Wright et al., 2005), whereas others have shown it to be the opposite case (Bennicelli et al., 2008; Evans et al., 2004; Patrício et al., 2019). This study shows that supplementing with Pluronic F68, as the non-ionic surfactant, in conjunction with 5% glycerol exhibits added stability after several freeze/thaw cycles (Fig. 2B, Table 1). After 3 weeks at 4 °C, all three buffers show relatively stable AAV9 titers after a loss of ~ 40% in the first week (Fig. 3A, Table 1), consistent with observations by multiple groups (Hoganson et al., 2002; Howard and Harvey, 2017; Wright et al., 2003). In a recent finding, researchers demonstrated that AAV remained stable for 6 months between 2 and 8 °C in formulation buffers comprising of either Pluronic F68 or Pluronic F68 and trehalose. (Rieser et al., 2022). Comparable to the −80 °C stability experiments, there was less instability in buffer A, which included both types of excipients than in either buffer B or C, which only contained one (Fig. 3B, Table 1). This suggests that having both glycerol and Pluronic F68 in the formulation is advantageous for AAV storage.

Howard and Harvey in 2017 transduced AAV1 stored in Tris buffer into rat primary cortical neurons with samples that were consecutively frozen and thawed 10 times. By measuring the iRFP expression they could determine the transduction efficiency. A similar experiment was performed here by transducing AAV9 from freeze/thaw samples of each buffer into N2a cell lines and counting cells with GFP expression (Fig. S2). Comparable with Howard and Harvey, greater variability in transduction efficiency was observed with the increase in number of freeze/thaw cycles. Also, consistent with this study, approximately 25% drop in expression after the third freeze/thaw was observed in Buffer A and B. In buffer C, there was a 40% drop in expression (Fig. 4B). Although, initially buffer B showed the same percent decrease as buffer A, by the sixth freeze/thaw the decline was much greater. This agrees with current data which demonstrate that the amounts of free DNA from ruptured capsids increase with the number of freeze/thaws in buffers containing only Pluronic F68 compared to the buffer consisting of two excipients. (Bee et al., 2022a; Xu et al., 2022).

Interestingly, it has been observed that different buffers had various effects on the thermostability of AAV serotypes but the transduction efficiency of the serotypes in these buffers remained unchanged (Bennett et al., 2017). In vivo studies would increase our knowledge of how well AAV recover after long-term storage. Gruntman and collaborators in 2015 measured transgene expression from sera and determined that there was no loss of viral potency when AAV stored long term at 4 °C and after two to four freeze/thaws at −80 °C was administered into mice. We know through our own investigations, that AAV stored in 5% glycerol and 0.001% Pluronic F68 buffer does not have a detrimental effect in mice (Maturana et al., 2021, 2020). This does not preclude further investigations to validate whether a formulation with both glycerol and Pluronic F68 would be a better choice for AAV storage.

In summary, AAV storage buffers that contain a cryopreservative is better suited than one with only Pluronic F68 in the recovery of the virus. With the addition of 5% glycerol to the non-ionic surfactant buffer, AAV9 infectivity is preserved. We have also shown that reduced concentrations of NaCl in PBS do not have a deleterious effect on the recovery of AAV9 or the long-term stability of AAV at 4 °C or at −80 °C. The results indicate that undiluted AAV9 formulated in buffer containing 5% glycerol and 0.001% Pluronic F68 can be stored at 4 °C for a few weeks (short term) and can undergo multiple freeze/thaw cycles from −80 °C storage (long term). Testing other commonly used AAV serotypes in this buffer would further substantiate these results. The assessment of this buffer could potentially influence the way AAV is administered in the research and clinical setting in the future.

Supplementary Material

Supplementary material Figure S2
Supplementary material Figure S1

Acknowledgements

We would like to thank Dr. H. Huang for comments and suggestions on this manuscript.

Funding

This work was supported by the Princeton Neuroscience Institute Research Innovator grant (USA), Princeton IP Accelerator grant (USA), New Jersey Alliance for Clinical and Translational Science (USA) UL1TR003017, and National Institutes of Health (USA) grants P40OD010996 & NIH-1U01NS113868 (EAE).

Footnotes

CRediT authorship contribution statement

Angela Chan: Conceptualization, Data curation, Formal analysis, Methodology, Investigation, Project administration, Resources, Supervision, Visualization, Writing – original draft, Writing – review & editing. Carola J. Maturana: Conceptualization, Data curation, Formal analysis, Methodology, Validation, Visualization, Writing – review & editing. Esteban A. Engel: Conceptualization, Data curation, Funding acquisition, Project administration, Resources, Supervision, Validation, Writing – review & editing.

Declaration of Competing Interest

Esteban A. Engel is an employee and equity holder of Sparks Therapeutics, a Roche company located in Philadelphia, PA 19104, USA. This work was performed while he was a Princeton University investigator.

Appendix A. Supporting information

Supplementary data associated with this article can be found in the online version at doi: 10.1016/j.jviromet.2022.114598.

References

  1. Addgene [WWW Document], n.d. URL 〈https://www.addgene.org/protocols/aav-production-hek293-cells/〉 (Accessed 5.12.20).
  2. Bee JS, Zhang Y, Finkner S, O’Berry K, Kaushal A, Phillippi MK, DePaz RA, Webber K, Marshall T, 2022a. Mechanistic studies and formulation mitigations of adeno-associated virus capsid rupture during freezing and thawing: mechanisms of freeze/thaw induced AAV rupture. J. Pharm. Sci 10.1016/j.xphs.2022.03.018. [DOI] [PubMed] [Google Scholar]
  3. Bee JS, Zhang Y., (Zoe), Phillippi MK, Finkner S, Mezghebe T, Webber K, Cheung WD, Marshall T, 2022b. Impact of time out of intended storage and freeze-thaw rates on the stability of adeno-associated virus 8 and 9. J. Pharm. Sci 10.1016/j.xphs.2022.01.002. [DOI] [PubMed] [Google Scholar]
  4. Bennett A, Patel S, Mietzsch M, Jose A, Lins-Austin B, Yu JC, Bothner B, McKenna R, Agbandje-McKenna M, 2017. Thermal stability as a determinant of AAV serotype identity. Mol. Ther. Methods Clin. Dev 6, 171–182. 10.1016/j.omtm.2017.07.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bennicelli J, Wright JF, Komaromy A, Jacobs JB, Hauck B, Zelenaia O, Mingozzi F, Hui D, Chung D, Rex TS, Wei Z, Qu G, Zhou S, Zeiss C, Arruda VR, Acland GM, Dell’Osso LF, High KA, Maguire AM, Bennett J, 2008. Reversal of blindness in animal models of leber congenital amaurosis using optimized AAV2-mediated gene transfer. Mol. Ther. J. Am. Soc. Gene Ther 16, 458–465. 10.1038/sj.mt.6300389. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Chen S, Wasserfall C, Kapturczak MH, Atkinson M, Agarwal A, 2006. Freeze-thaw increases adeno-associated virus transduction of cells. Am. J. Physiol. -Cell Physiol 291, C386–C392. 10.1152/ajpcell.00582.2005. [DOI] [PubMed] [Google Scholar]
  7. Croyle M, Cheng X, Wilson J, 2001. Development of formulations that enhance physical stability of viral vectors for gene therapy. Gene Ther. 8, 1281–1290. 10.1038/sj.gt.3301527. [DOI] [PubMed] [Google Scholar]
  8. Croyle MA, Roessler BJ, Davidson BL, Hilfmger JM, Amidon GL, 1998. Factors that influence stability of recombinant adenoviral preparations for human gene therapy. Pharm. Dev. Technol 3, 373–383. 10.3109/10837459809009865. [DOI] [PubMed] [Google Scholar]
  9. Evans RK, Nawrocki DK, Isopi LA, Williams DM, Casimiro DR, Chin S, Chen M, Zhu D-M, Shiver JW, Volkin DB, 2004. Development of stable liquid formulations for adenovirus-based vaccines. J. Pharm. Sci 93, 2458–2475. 10.1002/jps.20157. [DOI] [PubMed] [Google Scholar]
  10. Frank MSB, Nahata MC, Hilty MD, 1981. Glycerol: a review of its pharmacology, pharmacokinetics, adverse reasctions, and clinical use. Pharmacotherapy 1, 147–160. [DOI] [PubMed] [Google Scholar]
  11. Gruntman AM, Su L, Su Q, Gao G, Mueller C, Flotte TR, 2015. Stability and compatibility of recombinant adeno-associated virus under conditions commonly encountered in human gene therapy trials. Hum. Gene Ther. Methods 26, 71–76. 10.1089/hgtb.2015.040. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hoganson D, Ma J, Asato L, Ong M, Printz M, Huyghe B, Sosnowski B, D’Andrea M, 2002. Development of a stable adenoviral vector formulation. BioProcess J. 1, 43–48. [Google Scholar]
  13. Howard DB, Harvey BK, 2017. Assaying the stability and inactivation of AAV serotype 1 vectors. Hum. Gene Ther. Methods 28, 39–48. 10.1089/hgtb.2016.180. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Howell CL, Miller MJ, 1983. Effect of sucrose phosphate and sorbitol on infectivity of enveloped viruses during storage. J. Clin. Microbiol 18, 658–662. 10.1128/JCM.18.3.658-662.1983. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Kuzmin DA, Shutova MV, Johnston NR, Smith OP, Fedorin VV, Kukushkin YS, van der Loo JCM, Johnstone EC, 2021. The clinical landscape for AAV gene therapies. Nat. Rev. Drug Discov 20, 173–174. 10.1038/d41573-021-00017-7. [DOI] [PubMed] [Google Scholar]
  16. LUXTURNA® (voretigene neparvovec-rzyl) HCP [WWW Document], n.d. URL 〈http://luxturna.com/image〉 (Accessed 6.4.20).
  17. Maturana CJ, Verpeut JL, Engel EA, 2022. Single-cell quantification of Triple-AAV vector genomes coexpressed in neurons. Curr. Protoc 2, e430 10.1002/cpzl.430. [DOI] [PubMed] [Google Scholar]
  18. Maturana CJ, Verpeut JL, Kooshkbaghi M, Engel EA, 2021. Novel tool to quantify with single-cell resolution the number of incoming AAV genomes co-expressed in the mouse nervous system. Gene Ther. 1–6. 10.1038/s41434-021-00272-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Maturana CJ, Verpeut JL, Pisano TJ, Dhanerawala ZM, Esteves A, Enquist LW, Engel EA, 2020. Small alphaherpesvirus latency-associated promoters drive efficient and long-term transgene expression in the CNS. Mol. Ther. Methods Clin. Dev 17, 843–857. 10.1016/j.omtm.2020.04.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Pacouret S, Bouzelha M, Shelke R, Andres-Mateos E, Xiao R, Maurer A, Mevel M, Turunen H, Barungi T, Penaud-Budloo M, Broucque F, Blouin V, Moullier P, Ayuso E, Vandenberghe LH, 2017. AAV-ID: a rapid and robust assay for batch-to-batch consistency evaluation of AAV preparations. Mol. Ther 25, 1375–1386. 10.1016/j.ymthe.2017.04.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Patrício MI, Cox CI, Blue C, Barnard AR, Martinez-Fernandez de la Camara C, MacLaren RE, 2019. Inclusion of PF68 surfactant improves stability of rAAV titer when passed through a surgical device used in retinal gene therapy. Mol. Ther. Methods Clin. Dev 17, 99–106. 10.1016/j.omtm.2019.11.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Rieser R, Menzen T, Biel M, Michalakis S, Winter G, 2022. Systematic studies on stabilization of AAV vector formulations by lyophilization. J. Pharm. Sci 10.1016/j.xphs.2022.03.004. [DOI] [PubMed] [Google Scholar]
  23. Rodrigues GA, Shalaev E, Karami TK, Cunningham J, Slater NKH, Rivers HM, 2018. Pharmaceutical development of AAV-based gene therapy products for the eye. Pharm. Res 36, 29. 10.1007/s11095-018-2554-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Schuster DJ, Dykstra JA, Riedl MS, Kitto KF, Belur LR, Mclvor RS, Elde RP, Fairbanks CA, Vulchanova L, 2014. Biodistribution of adeno-associated virus serotype 9 (AAV9) vector after intrathecal and intravenous delivery in mouse. Front. Neuroanat 8, 42. 10.3389/fhana.2014.00042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Sloviter HA, 1958. Effects of the intravenous administration of glycerol solutions to animals and man1. J. Clin. Invest 37, 619–626. 10.1172/JCI103644. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Sonntag F, Köther K, Schmidt K, Weghofer M, Raupp C, Nieto K, Kuck A, Gerlach B, Böttcher B, Müller OJ, Lux K, Hörer M, Kleinschmidt JA, 2011. The assembly-activating protein promotes capsid assembly of different adeno-associated virus serotypes▿. J. Virol 85, 12686–12697. 10.1128/JVI.05359-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Srivastava A, Mallela KMG, Deorkar N, Brophy G, 2021. Manufacturing challenges and rational formulation development for AAV viral vectors. J. Pharm. Sci 110, 2609–2624. 10.1016/j.xphs.2021.03.024. [DOI] [PubMed] [Google Scholar]
  28. Steinbach S, Wistuba A, Bock T, Kleinschmidt JA, 1997. Assembly of adeno-associated virus type 2 capsids in vitro. J. Gen. Virol 78, 1453–1462. 10.1099/0022-1317-78-6-1453. [DOI] [PubMed] [Google Scholar]
  29. Van Den Berg L, Soliman FS, 1969. Effect of glycerol and dimethyl sulfoxide on 0 changes in composition and ph of buffer salt solutions during freezing. Cryobiology 6, 93–97. 10.1016/S0011-2240(69)80469-4. [DOI] [PubMed] [Google Scholar]
  30. Wang D, Tai PWL, Gao G, 2019. Adeno-associated virus vector as a platform for gene therapy delivery. Nat. Rev. Drug Discov 18, 358–378. 10.1038/S41573-019-0012-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Wang J, Ren Y, Zhou L-J, Kan L-D, Fan H, Fang H-M, 2021. Glycerol infusion versus mannitol for cerebral edema: a systematic review and meta-analysis. Clin. Ther 43, 637–649. 10.1016/j.clinthera.2021.01.010. [DOI] [PubMed] [Google Scholar]
  32. Wright JF, Le T, Prado J, Bahr-Davidson J, Smith PH, Zhen Z, Sommer JM, Pierce GF, Qu G, 2005. Identification of factors that contribute to recombinant AAV2 particle aggregation and methods to prevent its occurrence during vector purification and formulation. Mol. Ther 12, 171–178. 10.1016/j.ymthe.2005.02.021. [DOI] [PubMed] [Google Scholar]
  33. Wright JF, Qu G, Tang C, Sommer JM, 2003. Recombinant adeno-associated virus: formulation challenges and strategies for a gene therapy vector. Curr. Opin. Drug Discov. Dev 6, 174–178. [PubMed] [Google Scholar]
  34. Xie Q, Hare J, Turnigan J, Chapman MS, 2004. Large-scale production, purification and crystallization of wild-type adeno-associated virus-2. J. Virol. Methods 122, 17–27. 10.1016/j.jviromet.2004.07.007. [DOI] [PubMed] [Google Scholar]
  35. Xu Y, Jiang B, Samai P, Tank S-M, Shameem M, Liu D, 2022. Genome DNA leakage of adeno–associated virus under freeze–thaw stress. Int. J. Pharm 615, 121464 10.1016/j.ijpharm.2022.121464. [DOI] [PubMed] [Google Scholar]
  36. ZOLGENSMA® (onasemnogene abeparvovec-xioi) [WWW Document], n.d. URL 〈https://www.zolgensma.com〉 (Accessed 6.4.20).
  37. Zolotukhin S, Byrne BJ, Mason E, Zolotukhin I, Potter M, Chesnut K, Summerford C, Samulski RJ, Muzyczka N, 1999. Recombinant adeno-associated virus purification using novel methods improves infectious titer and yield. Gene Ther. 6, 973–985. 10.1038/sj.gt.3300938. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary material Figure S2
NIHMS1891700-supplement-Supplementary_material_Figure_S2.jpg (104.3KB, jpg)
Supplementary material Figure S1
NIHMS1891700-supplement-Supplementary_material_Figure_S1.jpg (80.8KB, jpg)

RESOURCES