Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Jul 1.
Published in final edited form as: J Virol Methods. 2020 May 1;281:113863. doi: 10.1016/j.jviromet.2020.113863

A Simple, Two-Step, Small-Scale Purification of Recombinant Adeno-Associated Viruses

Shih-Heng Chen a,b, Amy Papaneri a, Mitzie Walker a,b, Erica Scappini c, Robert D Keys d, Negin P Martin a,b,e
PMCID: PMC7293145  NIHMSID: NIHMS1596847  PMID: 32371233

Abstract

Recombinant adeno-associated viruses (rAAVs) are robust and versatile tools for in vivo gene delivery. Natural and designer capsid variations in rAAVs allow for targeted gene delivery to specific cell types. Low immunogenicity and lack of pathogenesis also add to the popularity of this virus as an innocuous gene delivery vector for gene therapy. rAAVs are routinely used to express recombinases, sensors, detectors, CRISPR-Cas9 components, or to simply overexpress a gene of interest for functional studies. High production demand has given rise to multiple platforms for the production and purification of rAAVs. However, most platforms rely heavily on large amounts of starting material and multiple purification steps to produce highly purified viral particles. Often, researchers require several small-scale purified rAAVs. Here, we describe a simple and efficient technique for purification of recombinant rAAVs from small amounts of starting material in a two-step purification method. In this method, rAAVs are released into the packaging cell medium using high salt concentration, pelleted by ultracentrifugation to remove soluble impurities. Then, the resuspended pellet is purified using a protein spin-concentrator. In this protocol, we modify the conventional rAAV purification methods to eliminates the need for fraction collection and the labor-intensive steps for evaluating the titer and purity of individual fractions. The resulting rAAV preparations are comparable in titer and purity to commercially available samples. This simplified process can be used to generate highly purified rAAV particles on a small scale, thereby saving resources, generating less waste, and reducing a laboratory’s environmental footprint.

Keywords: adeno-associated virus, rAAV, gene delivery, virus purification, neurobiology

1. Introduction

Recombinant adeno-associated viruses (rAAVs) are replication-defective parvoviruses that readily infect dividing and non-dividing mammalian cells (Daya and Berns, 2008). The AAV genome consists of a single-stranded DNA with self-complementary ends that form high-molecular-weight head-to-tail circular monomer, dimer, or concatemers (Yang et al., 1999). These concatemeric circles are often maintained in transduced cells for lasting gene expression in vivo (Penaud-Budloo et al., 2008). rAAVs can efficiently deliver up to 4.5 kilobase pairs (kbp) of genetic material. Unlike the wild-type virus, the genome of rAAV is typically maintained episomally in the nucleus and does not integrate onto specific sites in the host chromosome, whereas random integration events are observed at regions of host chromosomal instability and breakage with a low frequency of 0.1–1% transduction events (Li et al., 2011; Smith, 2008; Valdmanis, Lisowski, and Kay, 2012).

rAAVs are promising agents for gene therapy due to their varying capsid tropism, low immunogenicity, and lasting gene expression (Daya and Berns, 2008; Mingozzi and High, 2011). In research, they are ideal for functional studies in vivo. The genetic load can accommodate constitutive or inducible promoters to regulate the expression of genes of interest in addition to selectable and/or fluorescent markers. Many new techniques such as Cre-recombination-based rAAV targeted evolution (CREATE) have given rise to a myriad of rAAV capsids to achieve the desired performance and cell/tissue targeting (Deverman et al., 2016; Kotterman and Schaffer, 2014; Kotterman, Vazin, and Schaffer, 2015). Biotech manufacturing has responded to the demand for AAVs by increasing the quality and quantity of produced AAV vectors while maintaining a reasonable cost.

Currently, there are several platforms used for the production and purification of rAAVs (Aponte-Ubillus et al., 2018; Lock et al., 2010). Biotech manufacturing and virus core facilities used to rely on helper viruses such as adenovirus and herpesvirus for en mass delivery of complementary genes for large-scale and cost-efficient rAAV production. However, contamination of the helper virus in samples could become a health and safety concern (Clement, Knop, and Byrne, 2009; Naso et al., 2017; Thomas et al., 2009; Thorne, Takeya, and Peluso, 2009). Some biotech companies such as Applied Viromics and Virovek use baculovirus expression system which incorporates the complementary genes into the baculovirus genome and infects the insect cells sf9 to produce rAAV very efficiently. Baculovirus expression system offers several advnatages: (1) Sf9 cells can be cultivated at high density in suspensions with volumes of up to 200 liters, which could be far more efficient than adherent cells; (2) Sf9 cells can be grown under serum-free conditions, which eliminate the presence of potential immunoreactive or toxic animal-derived proteins; and (3) the added safety feature of baculovirus since it cannot replicate in human cells. The disadvantages of using the system includes the unstability of baculovirus genome at high viral concentrations and loss of rep genes that greatly decrease the rAAV production (Cecchini, Virag, and Kotin, 2011; Mietzsch et al., 2014; Smith, Levy, and Kotin, 2009). Most commonly, the complementary genes are delivered as plasmids to eukaryotic cells with a variety of transfection reagents (Clement and Grieger, 2016; Zolotukhin et al., 1999). Human embryonic kidney 293 cells (HEK293 and HEK293T) are a preferred cell line for packaging rAAVs since they constitutively express adenovirus E1A/B factors that are needed for packaging, are economical, and are transfected efficiently (Clement and Grieger, 2016). In conventional methods, to produce rAAVs, eukaryotic cells are transfected with three plasmids: 1) an rAAV transfer vector carrying the gene of interest flanked by Inverted Terminal Repeats (ITRs), 2) Rep/Cap genes, and 3) Helper plasmid delivering VA RNAs, E2A, and E4 OEF6 genes (Clement and Grieger, 2016). Following transfection, rAAVs are collected from the media or cell lysate and subjected to numerous purification steps (Grieger, Choi, and Samulski, 2006; Hagedorn et al., 2017; Lock et al., 2010; Xiao, 2010; Zolotukhin et al., 1999). Gradient centrifugation with iodixanol or cesium chloride (CsCl) provides flexibility for rAAV purification since it can be used to purify various rAAV serotypes. However, using gradients to purify rAAVs are time consuming and requires multiple purification steps to produce high purity rAAVs. The formed gradients are typically fractionated and individually evaluated for the amount and purity of rAAV particles. Moreover, gradient chemicals such as CsCl can exert toxic effects in animals and the resulting rAAV fractions have to be dialyzed with a physiologically balanced solution before use in vivo (Ayuso et al., 2010; Grieger et al., 2006; Lock et al., 2010). As expected, rAAV particles are lost after each purification step, and therefore, large amounts of starting material are needed to ensure sufficient amount of rAAV recovery. The complexity of purification and length of time spent on steps often prevents small laboratories from preparing their own rAAV samples. Moreover, budget limitations restrict research plans to what is achievable based on commercially available rAAV stocks, since customized rAAV preparations are commercially packaged at a significant cost.

Here, we describe a method for a small-scale production of rAAVs that allows novice research laboratories to produce purified rAAVs efficiently and economically for use in their research. Although a number of very efficient protocols for large scale rAAV production are published and currently in use by large virus production core laboratories (e.g. Addgene), the described method simplifies the purification process to enable small laboratories that are not accustomed to handling rAAVs to design and construct plasmids, and package particles for gene delivery in their own laboratories.

There are currently several hundred rAAV variants isolated by creating synthetic chimeras and/or mutagenizing wildtype rAAV capsid genes. We purified and tested rAAVs with several commonly used serotypes (that are parental serotypes for synthetic/chimera/mutagenized serotypes) using the two-step purification protocol. Using this method, we compared and analyzed yields for rAAV preparations with serotypes 1, 2, 5, 6, 8, and 9. The process effectively isolated infectious rAAV particles for all serotypes except for serotype 2. The reduced number of steps in the purification process minimized virus particle loss and saved resources. Moreover, the fraction collection step used in conventional rAAV purification methods, and the need to evaluate individual fractionated aliquots for titer and purity was eliminated. The resulting rAAV preparations are comparable in titer and purity to commercially available samples. This simplified process can be used to readily generate highly purified rAAV particles on a small scale, thereby saving resources, generating less waste, and reducing a laboratory’s environmental footprint.

2. Materials and Methods

2.1. Animals

C57BL/6J mice were obtained from Jackson Laboratories (Bar Harbor, ME, USA). Mice were housed in polycarbonate cages in animal facilities with controlled environmental conditions with a 12-hour artificial light-dark cycle and were provided fresh deionized water and NIH 31 chow ad libitum. All animal procedures were approved by the Institutional Animal Care and Use Committee and conducted in strict accordance with the National Institutes of Health animal care and use guidelines.

2.2. Cell culture

Mycoplasma-free HEK293-AAV cells (Cell Biolabs Inc., Cat. # AAV-100) were maintained in Dulbecco’s modified Eagle’s medium (DMEM, Life Technologies, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS, HyClone Laboratories, South Logan, UT, USA), 2 mM L-glutamine, and 1 mM sodium pyruvate. Cells were passaged three times per week to maintain them in exponential growth phase.

2.3. Plasmids and viruses

The plasmids used for transfection were (1) cis plasmid pAAV.hSyn.eGFP.WPRE.bGH (Addgene Cat# 105539) and pAAV-hSyn-eGFP (Addgene Cat# 50465); (2) trans plasmids pAAV2/1 (Cell Biolabs Inc., Cat# VPK-421), pAAV2/2 (Cell Biolabs Inc., Cat# VPK-422), pAAV2/5 (Cell Biolabs Inc., Cat# VPK-425), pAAV2/6 (Cell Biolabs Inc., Cat# VPK-426), pAAV2/8 (Cell Biolabs Inc., Cat# VPK-428), or pAAV2/9 (UPenn Vector Core); (3) pHelper plasmid containing adenovirus E2A, E4 and VA genes (Cell Biolabs Inc., Part No. 340202). AAV2-hSyn-EGFP (Addgene Cat# 50465-AAV2) and AAV9-hSyn-EGFP (UPenn Vector Core Cat # AV-9-PV1696) were purchased from Addgene and University of Pennsylvania Vector Core, respectively, and used for comparison of stock quality in this study.

2.4. AAV production and two-step purification

For transfections, each 15-cm dish was seeded with HEK293-AAV cells at 6 × 106 cells in 20 ml of DMEM with 10% FBS without antibiotics. Cells were incubated at 37°C in 5% CO2 for 24 hours before transfection. The AAV cis, AAV trans and pHelper plasmids (33.3 ug of each) were added to 2 ml of sterile 150 mM NaCl solution. Polyethylenimine MAX (PEI “MAX”, Polysciences, Warrington, PA, USA) stock solution was prepared at 16 mg/ml in sterile water and the pH was adjusted to 4.5 with sodium hydroxide. 12.5 μl of PEI stock was added to the 2 ml plasmid solution and mixed by vortexing. After 10 minutes of incubation at room temperature, 2 ml of solution was added dropwise to a 15-cm plate. Cultures were incubated at 37°C in 5% CO2 incubator for 24 hours and then the media was changed to 14 ml of fresh serum-free DMEM containing 2 mM L-glutamine, and 1 mM sodium pyruvate. 120 hours post-transfection, 2ml of 5M NaCl was added to the plates and incubation was resumed for an additional 2 hours before collecting the culture medium into 50ml conical tubes. Turbonuclease (Eton Bioscience, San Diego, CA, USA) was added to the culture supernatant to a final concentration of 50 units/ml and incubated at 37°C for 1 hour. To remove the cellular debris, the collected culture medium was centrifugated at 3,000 × g for 20 minutes at 4°C. After centrifugation, the supernatant was collected and filtered through a 0.45 μm Durapore PVDF filter (EMD Millipore, Billerica, MA, USA). Filtered supernatant was aliquoted into 30 ml conical tubes (Beckman Coulter, Brea, CA, USA) underlaid with a 4 ml sucrose cushion (40% sucrose in Tri-sodium chloride-EDTA buffer/TNE, sterile filtered) and centrifuged for 16 hours at 100,000 × g in a Beckman Coulter SW32Ti rotor at 4°C to pellet the AAV virus. After centrifugation, supernatant was gently removed, and the viral pellet was resuspended in 0.5 ml of cold PBS and mixed on a nutator at 4°C overnight. The next day, the virus was mixed by pipetting up and down in the tubes, combined, and then diluted with additional PBS to a final volume of 10 ml. Pooled virus solution was cleared of debris by centrifugation at 500 × g at 4°C for 10 minutes. The remaining supernatant was concentrated in a 100-kDa molecular weight cutoff (MWCO) protein concentrator (Pierce, Rockford, IL, USA) by centrifugation at 3,000 × g at room temperature for 10-minute intervals until the volume was reduced to 100–500 μl. The virus was aliquoted and stored at −80°C.

2.5. Analyzing transfection efficiency/GFP expression using flow cytometry

To measure the transfection efficiency, we replaced the cis plasmid from pAAV-hSyn-eGFP to pAAVEF1-eGFP (Addgene Cat# 60058) because the HEK293 cells do not possess synapsin activity. Cis plasmid together with helper plasmid and different serotypes of trans plasmids were transfected onto HEK293 cells in 6-well plates. Approximately 24–48 hours post-transfection, cells were imaged with a fluorescence microscope (Zeiss Colibri LED inverted epifluorescent microscope). After taking images, cells were harvested and the GFP expression was measured using a Sony ec800 flow cytometer.

2.6. rAAV titer by Q-PCR

The rAAV genome was titered as previously described (Lock et al., 2010). Briefly, rAAV samples were serially diluted from 10−2 to 10−8-fold, and the standard curve was generated by diluting the rAAV plasmid containing the ITR2 sequence (from 105 to 32 copies per 5 μl). The Q-PCR was performed using SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA, USA) and a LightCycler 96 System (Roche, Indianapolis, IN, USA) according to manufacturer’s protocols. The ITR forward primer (5’ GGAACCCCTAGTGATGGAGTT 3’) and the ITR reverse primer (5’ CGGCCTCAGTGAGCGA 3’) were added to the reaction with final concentration of 300 nM in 25 μl of total volume (5 μl of sample or standard curve plus 20 μl of primers and SYBR mix), and under the following PCR conditions: 95°C for 10 minutes, 95°C for 10s, 60°C for 60s, for 45 cycles. The virus titer was calculated by LightCycler 96 SW1.1 software using the parallel standard curve in the reaction, and the titer was given in genome copy/ml (GC/ml).

2.7. Silver staining

The purity of rAAV was determined via polyacrylamide gel electrophoresis (PAGE) by resolving 3–5 × 1010 GC of each rAAV sample boiled in NuPage sample buffer, loaded onto a 4–12% NuPage Bis-Tris gels (Thermo Fisher Scientific, Rockford, IL, USA). The proteins were revealed using Pierce Silver Staining kit (Thermo Fisher Scientific, Rockford, IL, USA) according to manufacturer’s protocols. The image was acquired by GeneFlash Bio Imaging system (Syngene, Frederick, MD, USA).

2.8. Organotypic mouse brain slice culture and rAAV transduction

Brain slices were isolated from C57BL/6J mice (postnatal 6–8 days). Brain slices were prepared according to previously described methods (Bastrikova et al., 2008; Gu, Lamb, and Yakel, 2012; Stoppini, Buchs, and Muller, 1991). 350 μm slices of brain were sectioned using a vibratome (Leica VT1200 S, Leica Biosystems, Buffalo Grove, IL, USA) in ice-cold cutting solution (GIBCO MEM Cat#: 61100, NaHCO3 26 mM, HEPES 25 mM, Tris 15 mM, Glucose 10 mM, MgCl2 3 mM). The organotypic brain slices were placed on a 0.4-μm pore size membrane insert (Transwell 3450-Clear, Corning Incorporated, Corning, NY, USA) in a 6-well plate. Slices were cultured at 37°C and 5% CO2 with the medium changed three times per week. AAVs were added as a drop on top of brain slices at 24 hours post dissection and incubated for 2 weeks. To visualize cells infected with rAAVs, fluorescence images were captured on a Zeiss LSM710 (Carl Zeiss Inc, Oberkochen, Germany) using an EC Plan-Neofluar 10x/0.3 objective. The 488 nm laser line from an Argon laser at 5% power was used for excitation of the GFP labeled cells; after which, a 493–577 nm band pass emission filter was used to collect the images of the GFP signal.

2.9. TEM electron microscopy

Transmission electron microscopy was performed on equal volumes (5 μL) of rAAV9 samples that were either prepared by the two-step purification method or purchased from the UPenn Core. Samples were prepared for negative staining by dropping 5 μl aliquot of viral suspension on a formvar coated 300 mesh grid and blotting off the grid at the edge with Whatman’s #1 filter paper until not quite dry. The wet grid was immediately negative stained with a drop of 1% phosphotungstic acid, pH 7.1, for 1 minute, and blotted dry at the edge with Whatman’s #1 filter paper. Grids were examined on a Technai T12 electron microscope equipped with a Gatan Orius camera system and the Digital Micrograph software. Negative stained images were obtained at 150000x.

2.10. Statistics

Data are presented as the mean ± SEM. Comparisons between three groups were conducted using one-way ANOVA, followed by Bonferroni’s post hoc multiple comparison test. Data were analyzed using Prism (v8.00, GraphPad, San Diego, CA). P-values less than or equal to 0.05 were considered statistically significant.

3. Results

3.1. Two-step purification of rAAVs eliminates the need for fraction collection and validation of individual fractions

Here, we describe a rapid and economical two-step rAAV purification method for novice researchers to purify quality rAAV preparations from a small amount of starting material to minimize waste and save resources. In order to assess the efficiency and yield of the two-step purification protocol, the same rAAV transfer vector, pAAV-hSyn-GFP, was packaged and purified using serotypes 1, 2, 5, 6, 8, and 9 as described in the Materials and Methods section. Synapsin promoter, a common promoter used to express genes and reporters in neurons, was used to express GFP. The described purification steps from seeding the HEK293-AAV cells to aliquoting and freezing the samples were completed in 8 days (Fig. 1a). The number of processing days could be shortened by reducing the incubation time following transient transfection from 4 to 2 days (described below) and replacing Day 7 overnight incubation with a 1-hour shaker/vortexer resuspension of the AAV pellet. However, reducing the number of purification process days results in longer processing time/steps per day as described below. In our protocol, polyethylenimine MAX (PEI) was used as an economical and efficient method to transfect and deliver packaging plasmids to HEK293-AAV cells (Lock et al., 2010). Other transfection methods, including calcium phosphate precipitation of DNA, can be substituted for PEI transfection (Martin and Raphael, 2017; Wright, 2009). Cells can be grown in attached monolayers, multilayered flasks, or in suspension depending on the preparation size and user preference. PEI offers the versatility of transfection in adherent or cell suspension cultures. We also measured transfection efficiencies among AAV serotypes by determining the amount of GFP expression after transfection with rep/cap, helper and AAV transfer plasmids. Since synapsin promoter does not express in HEK293 cells, we changed the cis transfer plasmid from pAAV-hSyn-eGFP to pAAV-EF1-eGFP. At 24–48 hours post-transfection, the cells were imaged and the GFP expression was determined by flow cytometry. The transfection efficiency among different serotypes is 43–52% based on GFP expression with no statistically significant difference among rAAV serotypes (Fig. 1b).

Fig. 1. Protocol outline and transfection efficiency.

Fig. 1

Fig. 1

Open in a new tab

Two-Step purification procedure and the evaluation of transfection efficiency. (a) The NIEHS Viral Vector Core AAV preparation protocol outline, including the two-step AAV purification process. (b) Transfection efficiency compared by imaging and measuring GFP expression using fluorescence microscopy and flow cytometry. Values are mean ± SEM with triplicates for each serotype. Data were analyzed with one-way ANOVA followed by Bonferroni’s post hoc multiple comparison test.

Transfected cells were allowed to express rAAV packaging genes and incubated at 37°C for 5 days. Recent studies have shown that extending incubation time from 3 to 5 days after transfection results in release of rAAV particles from cells and their accumulation in culture medium (Adamson-Small et al., 2016; Lock et al., 2010; Xiao, 2010). In order to minimize serum protein content in the cell medium, the media was switched to serum-free medium 24 hours post-transfection. Incubation with high salt (> 500 mM) has been shown to increase the release of rAAV particles from the cells (Adamson-Small et al., 2017; Lock et al., 2010). Salt’s effect on protein-protein interactions is often responsible for increased protein solubility (Kenney and Hunt, 1990) and mature rAAV particles are stable under high salt and in a wide range of pH conditions. Therefore, salt concentration was increased to enhance rAAV particle release from cells without damage to the particles. rAAV particles can also be harvested 3 days post-transfection. However, to release the rAAV particles from cells and into the media, four freeze-thaw cycles are necessary. During the freeze-thaw cycle, cellular components are also released into the media as impurities. Therefore, the addition of 2 incubation days for the release of rAAV particles into media reduces the amount of impurities in samples. Since no sample processing is necessary during the incubation time, we highly recommend the extended incubation time after the transfection step.

After transfection, rAAV particles were collected in supernatant media, cleared from debris, and treated with Turbonuclease to remove RNA, chromosomal DNA, and the packaging plasmid remnants. At this stage, the unpurified rAAV particles were suspended in a pool of protein impurities. In traditional purification methods, the unpurified rAAV solution is subject to two rounds of gradient centrifugation on cesium chloride or a single centrifugation step in a more inert medium such as iodixanol (Lock et al., 2010). The collected fractions are then individually assayed for purity, pooled, and then further concentrated. We observed that pelleting the rAAV particles and then passing them through a protein concentrator could result in purified rAAVs with comparable quality – while eliminating the fraction collection and evaluation step. In this two-step purification protocol, crude rAAV particles are pelleted by ultracentrifugation over a 40% sucrose cushion overnight (16–20 hours). Although the ultracentrifugation time is extended from 2 hours in iodixanol gradient to overnight over a sucrose cushion, the elimination of fraction collection and validation step saves a significant amount of processing time. In addition, since the rAAV particles are pelleted, they can be resuspended in desired solutions (for example, in special media for use with stem cells). Moreover, pelleting rAAVs allows for resuspension in a small volume. So, small-scale rAAV pellet production resulting from transfection of few plates could be resuspended in small amount of media for higher concentration.

3.2. AAV titers and purity are comparable to commercial preparations

The number of rAAV genocopies (GC) and the titer for individual preparations were determined by standard Q-PCR protocols as described in Material and Methods (Lock et al., 2010). These titers were originated from small-scale rAAV preparations from five or six 15-cm plates (Table 1). rAAV serotypes 1, 5, 8, and 9 yielded titers of greater than 1E+12 genome copies per ml (GC/ml) in volumes greater than 100 μL per preparation. This titer is comparable to most commercially available rAAV stock. Greater than 1500 genome copies were recovered per HEK293 cell. rAAV serotypes 2 and 6 had low yields in similar volumes. rAAV with serotype 8 produced the greatest yield of 1.02E+05 genome copies per cell (Table 1).

Table 1.

Virus yield of rAAV serotypes using two-step purification method

rAAV serotype Transgene Virus titer (1012 GC/ml) Final volume (μl) Total virus yield mean (GC) Virus yield/cell mean (GC)
rAAVI eGFP 1.04~4.87 100~150 3.02E+11 8.39E+03
rAAV2 eGFP 0.11~0.78 150 5.97E+10 1.74E+03
rAAV5 eGFP 1.89~9.65 150 7.05E+11 1.96E+04
rAAV6 eGFP 0.15~0.84 150~200 1.02E+11 2.82E+03
rAAV8 eGFP 10.6~61.8 100~150 3.67E+12 1.02E+05
rAAV9 eGFP 9.92~11.8 150~250 2.00E+12 5.55E+04
Open in a new tab

AAV-hSyn-eGFP was packaged with serotypes 1, 2, 5, 6, 8, and 9 using the two-step purification method in small scale. Viral titers were determined by Q-PCR as described in Materials and Methods. The value is mean value of three independent repeats. The titer of rAAV2 from Addgene is 4.07E+12 GC/ml and the rAAV9 from U Penn vector core is 4.54E+14 GC/ml.

To test the purity of samples, the same number of genome copies for each sample were resolved via SDS-PAGE and stained for protein content with silver stain (Fig. 2a). Silver-staining assay is typically used to detect the major rAAV capsid proteins VP1, VP2, and VP3; and to compare the ratios of rAAV capsid proteins to impurities, bands other than VP1, VP2, and VP3. The amount of VP1, VP2, and VP3 is also indicative of the amount of rAAV particles present in the preparation (Fig. 2b). Similar amounts of genome copies for each virus for AAV1, 5, 8, and 9 (3E+10 to 1E+11 GC), and 2E+9 and 5E+9 GC for rAAV2 and 6 (due to low titer and limited volume that could be loaded onto a well) were loaded onto the gel in Fig. 2a. Equal volumes of rAAV preparations (3ul of each preparation) were loaded onto the gel in Fig. 2b. Two samples of AAV-hSyn-eGFP were purchased from commercial sources (Addgene.org and UPenn Vector Core) and resolved side-by-side on silver-stained gels for comparison of stock quality. The same rAAV transfer plasmids for rAAV2 and rAAV9 were purchased from Addgene and UPenn Vector Cores and used for preparing rAAV particles using the two-step purification method. The low genome copies of two-step purification for rAAVs with serotypes 2 and 6 are evident on the gel in Fig 2b. However, rAAV serotypes 1, 5, 8, and 9 had robust VP1, VP2, VP3 presentations. All two-step preparations had comparable purities to purchased commercial rAAV preparations (Fig. 2a).

Fig. 2. Comparison between purity and yield of AAV serotypes.

Fig. 2

Open in a new tab

AAV-hSyn-eGFP was packaged with serotypes 1, 2, 5, 6, 8, and 9 using the two-step purification method. a. Compare similar amounts of viral genocopies (GC) on the gel. About 3E+10 to 1E+11 GC of rAAV1, 5, 8, 9 produced by two-step purification method, and rAAV2 and 9 purchased from Addgene and UPenn Vector Core were loaded onto the gel. Due to low titer and limited volume that could be loaded onto a well, only 2E+9 and 5E+9 GC of AAV2 and 6 purified with the two-step purification were loaded. Three major protein bands VP1, VP2, and VP3 are the most abundant proteins resolved by silver-staining. b. Equal volumes of AAV preparations (3ul) were loaded onto the gel and silver-stained. Two samples of AAV-hSyn-eGFP were purchased from commercial sources (Addgene.org and UPenn Vector Core) and loaded side-by-side on silver-stained gels for comparison.

Transmission electron microscopy (TEM) was used to visualize the structural features of the rAAV particles and sample impurities. rAAV9 sample prepared by the two-step purification method and its commercially purchased equivalent rAAV9 from the UPenn Core were imaged using TEM (Fig. 3a). As shown in the negative stain images, when the stain fills the empty center of the virus particles, it gives a darker contrast shadow – red arrows. The images and the quantification result shows that there is no significant difference between stained particles and impurities in samples. The empty/full ratio of rAAV9 from UPenn is 4.29 ± 1.77% and is 6.09 ± 2.82 % for the two-step purification method (Fig. 3b). The purchased AAV9 from UPenn contained more particles and higher titer.

Fig. 3. Transmission Electron Microscopy (TEM) images of AAV samples.

Fig. 3

Fig. 3

Open in a new tab

(a) Equal volumes (5 μL) of AAV9 sample preparation by the two-step purification method and the AAV9 samples purchased from the UPenn Core were imaged by TEM as described in Materials and Methods. Negative stained images were obtained at 150,000x. Red arrows indicate the empty rAAV particles. (b) Empty/full capsid ratio was counted and analyzed using Prism 8.0 software. Values are mean ± SEM with eight images for each sample. Data were analyzed with t-test.

3.3. rAAVs prepared using two-step purification method are effective ex-vivo transduction vectors

In addition to determining the genome copies for each rAAV preparation and resolving the samples on silver-stained gels, we tested the transduction efficiency of each sample in organotypic mouse brain slice cultures. Human synapsin promoter is a robust promoter in central nervous system neurons and provides an effective way of comparing GFP expression in rAAV transduced living tissue. Many small-scale rAAV preparations are used for gene delivery to animal tissue in research laboratories (Aschauer, Kreuz, and Rumpel, 2013; Rost et al., 2017).

Two microliters of viral preparations were added dropwise on top of cultured mouse brain tissues and monitored for expression of GFP by confocal microscopy. GFP expression was detected one week after infection. rAAV preparations for serotypes 1, 5, 6, 8, and 9 transduced mouse neurons effectively (Fig. 4). Despite low titer, the rAAV sample with serotype 6 transduced neurons as effectively as other serotypes. rAAV with serotype 2, had the lowest yield and expressed the lowest levels of GFP. Samples of AAV-hSyn-eGFP with serotypes 2 and 9 from commercial sources (Addgene.org and UPenn Vector Core, respectively) had comparable robust expressions similar to the two-step purified rAAV preparations. The differences in GFP expression, and hence transduction, could also be due to tropism of different serotypes.

Fig. 4. AAV transduction of cultured mouse brain slices.

Fig. 4

Open in a new tab

Organotypic mouse brain slice cultures were prepared and transduced with various AAV serotypes. Two microliters of each serotype of rAAVs were added to the brain slices. Confocal images depict GFP expression in cultured mouse neurons after two weeks. Despite low yield, AAV6 transduced cells as efficiently as higher yield AAV preparations. Commercially purchased samples were used as comparison of stock quality.

All two-step purified rAAV samples, except for serotype 2, demonstrated effective transduction efficiencies with no observed cytopathic effect (CPE) based on brightfield imaging. GFP expression in neurons was consistently monitored for 4 weeks with no observed changes in expression.

4. Discussion

Rapid expansion of the rAAV toolbox has availed a wide range of vectors with designer capsids, promoters, and/or selectable markers to researchers (Berns and Muzyczka, 2017; Deverman et al., 2016; Rost et al., 2017). rAAVs are preferred vectors for many in vivo gene transfer applications to non-dividing cells. In order to facilitate gene delivery for basic and clinical research, manufacturing practices for rAAV production have been optimized to provide research and clinical-grade preparations (Aponte-Ubillus et al., 2018). A myriad of rAAV samples are commercially packaged and available for purchase. However, preparation of rAAVs from customized vectors carrying genes with single nucleotide polymorphisms, dominant-negative mutations, chimeric proteins, etc. are often costly and time-consuming to prepare. Here, we describe a method that can be used to purify many rAAV serotypes economically and on small-scale.

In this method, cell transfection and recovery of rAAV particles from cell media is performed according to standard protocols (Grieger et al., 2006; Lock et al., 2010). The virus is collected from the culture medium that has been incubated for five days as compared to the traditional method of collecting rAAV particles from freeze-thawed cell lysate in three days. During the five-day incubation following transfection, rAAV particles are released and accumulated in media devoid of serum. Therefore, fewer impurities are present in the starting material compared to freeze-thawed cell lysates. After ultracentrifugation, less than 15% of rAAV particles (genocopies determined by Q-PCR, data not shown) are discarded in the medium and the remainder of particles are pelleted below the sucrose cushion. The resulting pellet is then resuspended in phosphate-buffered saline (PBS) or any desired media overnight. We recommend gentle resuspension of rAAV particles overnight to maintain the integrity of the capsid protein since the hydrodynamic shear forces from vortexing or shaking could result in protein structure destabilization (Bekard et al., 2011), although placing the samples in vortexer/shaker reduces the incubation time from overnight to one hour. In the last step of the two-step purification, we used a 100-kDa molecular weight cutoff (MWCO) protein concentrator to concentrate and remove the impurities. This step only takes about 30–60 minutes to concentrate the rAAV from 10 ml to 100–500 μl. As shown in Fig. 2, the small molecular weight impurities can be removed as good as samples purified with cesium chloride or Iodixanol gradient purification. Approximately 10% of rAAV genocopies (determined by Q-PCR, data not shown) are lost through the protein concentrator pores during centrifugation. The volume of rAAV solution can be continuously reduced by centrifugation until the desired volume and concentration is reached. This allows for the preparation of small amounts of rAAV from a reduced amount of starting material. The purification steps are simplified and shortened without jeopardizing quality. In this method, since rAAVs are not purified through multiple steps, less sample is lost during purification steps, and less starting material is required (Lock et al., 2010). This protocol allows research laboratories to produce multiple rAAV preparations from small amounts of starting material. Less waste also improves the environmental footprint of research that involves small amounts of rAAVs. In neurobiological studies, rAAVs are often diluted several folds and delivered to rodent brain via stereotactic injections for in vivo studies. The delivered amounts are frequently measured in nano and picoliters (Hocquemiller et al., 2016; Stoica et al., 2013). However, rAAVs are typically purchased from commercial sources or virus cores in 100 ul aliquots or more. The remainder of rAAVs, if stored properly, shows almost no loss of efficacy for years. However, rapid advances in design of neurobiological tracers, sensors, and actuators, cause the stored rAAVs to become obsolete quickly. Most neurobiology laboratories require multiple small-scale preparations of the latest tools. This two-step purification could be a useful tool to rapidly produce multiple small-scale viruses samples at one time.

The standard rAAV purification protocols could be classified into three categories, gradient-based, column-based, and combined methods. In the gradient-based procedure, the purity of rAAVs relies on collecting rAAV fractions from gradients and examination of each individual fraction for quality and quantity. Samples purified using cesium chloride also require an additional dialysis step to remove toxic cesium chloride. Affinity column chromatography and ion-exchange column chromatography systems are available to purify rAAVs. The flexibility of this method is low because it is not suitable for a broad range of rAAV serotypes. Although the combined gradient and column method provides better quality and flexibility in rAAV purification, the cost of this method could be high. In this protocol, no toxic material is introduced into the collected samples and a protein concentrator device is used as a filtration device to remove impurities. This method could be used in a variety of rAAV serotypes. Using this two-step purification protocol, we have prepared and tested rAAVs with serotypes 1, 2, 5, 6, 8, and 9. Low recovery yields were observed for serotypes 2 and 6. Furthermore, serotype 2 preparation showed low levels of neuronal transduction in organotypic mouse brain slice cultures. Low yield may be due to the binding of surface proteins of rAAV serotype 2 to cell surface heparan sulfate proteoglycan (HSPG) that has been shown to reduce the number of rAAV2 particles released into the culture media during virus production (Perabo et al., 2006). Each chimeric/designer rAAV serotype should be assessed and validated individually for yield and transduction efficiency after purification with the described two-step purification method. Meanwhile, The rAAV samples prepared by the two-step purification method were resolved by SDS-PAGE, silver-stained, and imaged by TEM to demonstrate the purity of samples as compared to similar commercially obtained rAAV preparations. The simplification of the purification protocol saves resources and less labor-intensity. The materials used in the two-step purification protocol are inexpensive compared to the purification after cesium chloride or Iodixonal gradient using FPLC or chromatographic columns, the protein concentrator column is cheaper. The equipment used, such as centrifuges, are typically available in most departments as shared resources. This method also reduces the amount of generated hazardous waste by scaling down the preparation size. In addition, the AAV samples in this method were prepared from 6 × 15cm plates. In comparison, conventional rAAV purification methods use 20 or more 15cm plates or multi-layer cell flasks that can house equivalent number of cells. The protocol also utilizes 200ug of each plasmid as compared to several milligrams of plasmids used in conventional methods (Lock et al., 2010). Total amount of time per day spent on the purification process is significantly less than traditional purification methods especially since the need for labor-intensive fraction collection, assaying each fraction, and further concentration and dialysis to remove chemicals and impurities is eliminated.

Lastly, the purified virus stocks using the two-step method may also contain empty rAAV viral capsids. The empty/full virus particle ratio determined by TEM revealed that rAAV9 produced by the two-step purification method doesn’t have a significantly higher number of the empty virus particles compared to commercially purchased rAAV9 using iodixanol gradient. The effect of empty capsids on transduction outcomes is still unclear. The presence of empty capsids may increase the unwanted immune response; yet, some studies report that the existence of empty capsids is beneficial and increases transduction (Wright, 2014a; Wright, 2014b).

In conclusion, the two-step rAAV purification method is an effective and easy-to-use method for research laboratories to produce small-scale, yet high-quality, rAAV preparations. This technique enables researchers to rapidly create and test new rAAV constructs in animal models and can be scaled up for bulk preparations.

Highlights.

  • A simplified AAV purification procedure for novice user

  • Less labor-intensive, saves resources, eliminates waste

  • Eliminates multiple steps used in conventional AAV purification such as fraction collection

  • Protocol produces the commercially comparable quality of rAAV for use in vivo

Acknowledgements

This research was supported by the Intramural Research Program of the National Institute of Health (NIH), National Institute of Environmental Health Sciences (NIEHS). We are grateful to Dr. Carl Bortner and Dr. Jesse Cushman for critical reading of the manuscript and helpful advice. We would also like to acknowledge and thank Dr. Jerrel Yakel, Dr. Zhenglin Gu, Ms. Pattie Lamb, Ms. Deloris Sutton, and Dr. Bernd Gloss for their support, in addition to intellectual and technical contributions.

Funding: This research was supported by the Intramural Research Program of the National Institute of Health (NIH), National Institute of Environmental Health Sciences (NIEHS).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Conflict of interest: All authors declare no competing financial interests.

Ethical Approval: Animals used in this study were ordered from Charles River and Jackson Laboratories, USA. All animal procedures complied with the institutional guidelines, NIH/NIEHS, animal care guidelines and were approved by the Animal Care and Use Committee (ACUC) at the NIH/NIEHS, animal Protocol # 2012–0004.

References

  1. Adamson-Small L, Potter M, Byrne BJ and Clement N, 2017. Sodium Chloride Enhances Recombinant Adeno-Associated Virus Production in a Serum-Free Suspension Manufacturing Platform Using the Herpes Simplex Virus System. Human gene therapy methods 28, 1–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Adamson-Small L, Potter M, Falk DJ, Cleaver B, Byrne BJ and Clement N, 2016. A scalable method for the production of high-titer and high-quality adeno-associated type 9 vectors using the HSV platform. Molecular therapy. Methods & clinical development 3, 16031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Aponte-Ubillus JJ, Barajas D, Peltier J, Bardliving C, Shamlou P and Gold D, 2018. Molecular design for recombinant adeno-associated virus (rAAV) vector production. Applied microbiology and biotechnology 102, 1045–1054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Aschauer DF, Kreuz S and Rumpel S, 2013. Analysis of transduction efficiency, tropism and axonal transport of AAV serotypes 1, 2, 5, 6, 8 and 9 in the mouse brain. PloS one 8, e76310. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Ayuso E, Mingozzi F, Montane J, Leon X, Anguela XM, Haurigot V, Edmonson SA, Africa L, Zhou S, High KA, Bosch F and Wright JF, 2010. High AAV vector purity results in serotype- and tissue-independent enhancement of transduction efficiency. Gene Ther 17, 503–10. [DOI] [PubMed] [Google Scholar]
  6. Bastrikova N, Gardner GA, Reece JM, Jeromin A and Dudek SM, 2008. Synapse elimination accompanies functional plasticity in hippocampal neurons. Proceedings of the National Academy of Sciences of the United States of America 105, 3123–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bekard IB, Asimakis P, Bertolini J and Dunstan DE, 2011. The effects of shear flow on protein structure and function. Biopolymers 95, 733–45. [DOI] [PubMed] [Google Scholar]
  8. Berns KI and Muzyczka N, 2017. AAV: An Overview of Unanswered Questions. Human gene therapy 28, 308–313. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cecchini S, Virag T and Kotin RM, 2011. Reproducible high yields of recombinant adeno-associated virus produced using invertebrate cells in 0.02- to 200-liter cultures. Human gene therapy 22, 1021–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Clement N and Grieger JC, 2016. Manufacturing of recombinant adeno-associated viral vectors for clinical trials. Molecular therapy. Methods & clinical development 3, 16002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Clement N, Knop DR and Byrne BJ, 2009. Large-scale adeno-associated viral vector production using a herpesvirus-based system enables manufacturing for clinical studies. Human gene therapy 20, 796–806. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Daya S and Berns KI, 2008. Gene therapy using adeno-associated virus vectors. Clinical microbiology reviews 21, 583–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Deverman BE, Pravdo PL, Simpson BP, Kumar SR, Chan KY, Banerjee A, Wu WL, Yang B, Huber N, Pasca SP and Gradinaru V, 2016. Cre-dependent selection yields AAV variants for widespread gene transfer to the adult brain. Nature biotechnology 34, 204–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Grieger JC, Choi VW and Samulski RJ, 2006. Production and characterization of adeno-associated viral vectors. Nat Protoc 1, 1412–28. [DOI] [PubMed] [Google Scholar]
  15. Gu Z, Lamb PW and Yakel JL, 2012. Cholinergic coordination of presynaptic and postsynaptic activity induces timing-dependent hippocampal synaptic plasticity. The Journal of neuroscience : the official journal of the Society for Neuroscience 32, 12337–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hagedorn C, Schnodt-Fuchs M, Boehme P, Abdelrazik H, Lipps HJ and Buning H, 2017. S/MAR Element Facilitates Episomal Long-Term Persistence of Adeno-Associated Virus Vector Genomes in Proliferating Cells. Human gene therapy 28, 1169–1179. [DOI] [PubMed] [Google Scholar]
  17. Hocquemiller M, Giersch L, Audrain M, Parker S and Cartier N, 2016. Adeno-Associated Virus-Based Gene Therapy for CNS Diseases. Human gene therapy 27, 478–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Kenney PB and Hunt MC, 1990. Effect of water and salt content on protein solubility and water retention of meat preblends. Meat science 27, 173–80. [DOI] [PubMed] [Google Scholar]
  19. Kotterman MA and Schaffer DV, 2014. Engineering adeno-associated viruses for clinical gene therapy. Nature reviews. Genetics 15, 445–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kotterman MA, Vazin T and Schaffer DV, 2015. Enhanced selective gene delivery to neural stem cells in vivo by an adeno-associated viral variant. Development 142, 1885–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Li H, Malani N, Hamilton SR, Schlachterman A, Bussadori G, Edmonson SE, Shah R, Arruda VR, Mingozzi F, Wright JF, Bushman FD and High KA, 2011. Assessing the potential for AAV vector genotoxicity in a murine model. Blood 117, 3311–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lock M, Alvira M, Vandenberghe LH, Samanta A, Toelen J, Debyser Z and Wilson JM, 2010. Rapid, simple, and versatile manufacturing of recombinant adeno-associated viral vectors at scale. Human gene therapy 21, 1259–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Martin DM and Raphael Y, 2017. It’s All in the Delivery: Improving AAV Transfection Efficiency with Exosomes. Mol Ther 25, 309–311. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Mietzsch M, Grasse S, Zurawski C, Weger S, Bennett A, Agbandje-McKenna M, Muzyczka N, Zolotukhin S and Heilbronn R, 2014. OneBac: platform for scalable and high-titer production of adeno-associated virus serotype 1–12 vectors for gene therapy. Human gene therapy 25, 212–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Mingozzi F and High KA, 2011. Therapeutic in vivo gene transfer for genetic disease using AAV: progress and challenges. Nature reviews. Genetics 12, 341–55. [DOI] [PubMed] [Google Scholar]
  26. Naso MF, Tomkowicz B, Perry WL 3rd and Strohl WR, 2017. Adeno-Associated Virus (AAV) as a Vector for Gene Therapy. BioDrugs 31, 317–334. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Penaud-Budloo M, Le Guiner C, Nowrouzi A, Toromanoff A, Cherel Y, Chenuaud P, Schmidt M, von Kalle C, Rolling F, Moullier P and Snyder RO, 2008. Adeno-associated virus vector genomes persist as episomal chromatin in primate muscle. Journal of virology 82, 7875–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Perabo L, Goldnau D, White K, Endell J, Boucas J, Humme S, Work LM, Janicki H, Hallek M, Baker AH and Buning H, 2006. Heparan sulfate proteoglycan binding properties of adeno-associated virus retargeting mutants and consequences for their in vivo tropism. Journal of virology 80, 7265–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Rost BR, Schneider-Warme F, Schmitz D and Hegemann P, 2017. Optogenetic Tools for Subcellular Applications in Neuroscience. Neuron 96, 572–603. [DOI] [PubMed] [Google Scholar]
  30. Smith RH, 2008. Adeno-associated virus integration: virus versus vector. Gene Ther 15, 817–22. [DOI] [PubMed] [Google Scholar]
  31. Smith RH, Levy JR and Kotin RM, 2009. A simplified baculovirus-AAV expression vector system coupled with one-step affinity purification yields high-titer rAAV stocks from insect cells. Mol Ther 17, 1888–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Stoica L, Ahmed SS, Gao G and Sena-Esteves M, 2013. Gene transfer to the CNS using recombinant adeno-associated virus. Curr Protoc Microbiol Chapter 14, Unit14D 5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Stoppini L, Buchs PA and Muller D, 1991. A simple method for organotypic cultures of nervous tissue. Journal of neuroscience methods 37, 173–82. [DOI] [PubMed] [Google Scholar]
  34. Thomas DL, Wang L, Niamke J, Liu J, Kang W, Scotti MM, Ye GJ, Veres G and Knop DR, 2009. Scalable recombinant adeno-associated virus production using recombinant herpes simplex virus type 1 coinfection of suspension-adapted mammalian cells. Human gene therapy 20, 861–70. [DOI] [PubMed] [Google Scholar]
  35. Thorne BA, Takeya RK and Peluso RW, 2009. Manufacturing recombinant adeno-associated viral vectors from producer cell clones. Human gene therapy 20, 707–14. [DOI] [PubMed] [Google Scholar]
  36. Valdmanis PN, Lisowski L and Kay MA, 2012. rAAV-mediated tumorigenesis: still unresolved after an AAV assault. Mol Ther 20, 2014–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Wright JF, 2009. Transient transfection methods for clinical adeno-associated viral vector production. Human gene therapy 20, 698–706. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Wright JF, 2014a. AAV empty capsids: for better or for worse? Mol Ther 22, 1–2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Wright JF, 2014b. Product-Related Impurities in Clinical-Grade Recombinant AAV Vectors: Characterization and Risk Assessment. Biomedicines 2, 80–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Xiao X, 2010. Adeno-associated viral vectors found free in media. Human gene therapy 21, 1221–2. [DOI] [PubMed] [Google Scholar]
  41. Yang J, Zhou W, Zhang Y, Zidon T, Ritchie T and Engelhardt JF, 1999. Concatamerization of adeno-associated virus circular genomes occurs through intermolecular recombination. Journal of virology 73, 9468–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Zolotukhin S, Byrne BJ, Mason E, Zolotukhin I, Potter M, Chesnut K, Summerford C, Samulski RJ and Muzyczka N, 1999. Recombinant adeno-associated virus purification using novel methods improves infectious titer and yield. Gene therapy 6, 973–85. [DOI] [PubMed] [Google Scholar]

RESOURCES