Novel Approaches for Production of
Recombinant Adeno-Associated Virus
This article describes how to produce a recombinant
adeno-associated virus (rAAV) stock of high purity.
The adeno-associated virus (AAV) is a human parvovirus
that was first engineered as a recombinant
viral vector for gene delivery by Hermonat and
Muzyczka in 1984. The unique biology and life cycle
of rAAV make it a popular choice as a gene delivery
system as it satisfies the main criteria for successful
gene vectors. These criteria include but are not limited
to (1) efficient transduction of the target cell; (2) stable
and long-term expression of the transgene of interest,
especially with the use of promoters and enhancers
that are not inactivated in the transduced cell; and (3)
a lack of stimulation of a cytotoxic immune response
to the vector or transduced cell, resulting in a very
good safety profile for clinical application. Wild-type
AAV has not previously been associated with disease
in healthy adult humans and is classified as a risk
group 1 agent under the NIH's recombinant DNA
guidelines (rev. 04/02). Recombinant AAV vectors
retain only the inverted terminal repeat (ITRs)
sequences from the wild-type AAV genome, with 96%
of the DNA genome removed. This includes all viral
coding genes. Normally recombinant AAV is considered
nonpathogenic, noninfectious, and nonhazardous.
However, the incorporation of oncogenes or
toxin encoding genes into the vector genome may alter
this status. Therefore, laboratory facilities used to
produce rAAV may be required by local institutions to
operate in accordance with biosafety level 2 guidelines,
despite rAAV not being a risk group 2 agent in
typical circumstances. Wild-type AAV requires helper
functions provided in trans
by a helper virus such as adenovirus or herpes virus for AAV replication. Early
generation rAAV preparations were produced using a
helper adenovirus that was then almost completely
eliminated in the purification process. More recently,
trace levels of helper virus in rAAV stocks have been
shown to elicit a cellular immune response to the AAVtransduced
tissue (Monohan et al.
, 1998). Soon after
this observation, efforts were made to improve the
procedure for generating rAAV vectors, with our laboratory
developing a packaging procedure that uses
nonoverlapping plasmid constructs to produce rAAV
vectors free of contamination by wild-type AAV or
helper adenovirus (Xiao et al.
, 1998). All AAV vectors
utilize a plasmid substrate carrying the viral ITR
sequences flanking the therapeutic gene of interest. For
efficient packaging, the rAAV insert size must be
~4.6kb or smaller, consistent with wild-type genome
size ~4.7kb. The AAV plasmid is then cotransfected
into human embryonic kidney (HEK) 293 cells, along
with a plasmid(s) that provides AAV and adenovirus
helper functions. HEK 293 cells contain an adenovirus
5 EIA gene integrated into the genome that activates
the AAV Rep and Cap, as well as other essential Ad
genes required for productive AAV infection. In this
setting, only the gene insert along with the flanking
ITRs is then packaged into rAAV virions. Major
advances in AAV production have been directly
related to better understanding the unique biology of
this virus. For example, Summerford and Samulski
(1998) identified the primary receptor for AAV type 2
as heparin sulfate proteoglycan. As a result, a novel
purification procedure using affinity chromatography
was developed to generate virus stocks with a very
high level of purity. In addition to the affinity chromatography
step, this protocol also uses an iodixanol
gradient in place of the cesium chloride step used in earlier protocols to significantly shorten the highspeed
centrifugation step and improve the quality of
the vector preparations. The aim of this article is to
discuss methods for quantifying the purified vectors
using the most current approaches that are reproducible
from laboratory to laboratory.
II. MATERIALS AND
A. Cell Culture
Human 293 cells are from American Type Culture Collection
(ATCC, Rockville, MD; CRL 1573).
Phosphate-buffered saline (PBS)(No. D-5837;
Dulbecco's modified Eagle's medium (DMEM) (No. D-
Penicillin-streptomycin (No. 15140-122) (Gibco-BRL
Fetal bovine Serum (No. F-2442) (Sigma-Aldrich)
Trypsin-EDTA (No. T-4049) (Sigma-Aldrich)
Falcon integrid tissue culture dish (No. 08-772-6)
Plasmid with transgene of interest
Plasmid pXX2, the AAV helper plasmid (Samulski laboratory);
map and sequence are available on the
internet at http://www.med.unc.edu/genether/
Plasmid pXX6, the adenoviral helper plasmid
(Samulski laboratory); map and sequence are
available at the just-listed web address
C. Production of Adenovirus-Free
Monolayers of 293 cells at approximately 80%
Corning 50-ml concical centrifuge polystyrene tubes
(No. 05-538-55A, Fisher)
2X HeBs (HEPES-buffered saline): Mix 16.4 g of NaCl,
11.9g of HEPES, and 0.21 g of Na2
Adjust to 1 liter and filter sterilize.
Nalgene nitrocellulose filter sterilization unit (0.45 µm)
(1000 ml, No. 09-761-40; Fisher)
Restriction enzymes Xba
l and Hind
Sure bacteria (Stratagene)
D. Purification of Recombinant Virus
Sorvall RT 6000B and Sorvall GS3 rotor
Beckman Ultracentrifuge and SW-41 and Ti-70
Ultrasonic processor (Cole-Palmer) with 1/8-in diameter
probe (processor No. U-04711-30; microtip No.
U-04710-46; Cole Palmer)
Pump Pro (Watson-Marlow) (No. 14-283-13; Fisher)
Ethanol/dry ice bath
Opti-seal tubes (No. 361625; Beckman)
Corning 50-ml conical centrifuge tubes
PBS-MK: Mix 50ml of 10X PBS, 0.5 ml of 1M
and 0.5 ml of 2.5 M
KCl and adjust to a final volume
of 0.5 liter with ddH2
Ultraclear tubes (12.5 and 32.4ml) for SW-41 rotor
Heparin sepharose column (1 ml) Hi Trap No. 17-0407:
Amersham Pharmacia Biotech
Optiprep (No. 103-0061) (Gibco-BRL Life
Opti-mem 1 (No. 31985-013) (Gibco-BRL Life
Sodium deoxycholate (No. D-5670; Sigma)
Benzonase (No. E-8263; Sigma)
AKTA FPLC (Amersham-Pharmacia Biotech)
15% iodixanol with 1M
NaCl: Mix 5ml of 10X PBS,
0.05 ml of 1 M
, 0.05 ml of 2.5 M
KCl, 10 ml of
NaCl, 0.075 ml 0.5% stock phenol red, and 12.5
ml of Optiprep. Adjust to a final volume of 50ml
O and filter sterilize.
25% iodixanol: Mix 5ml of 10X PBS, 0.05ml of 1M
, 0.05ml of 2.5M KCl, 0.1ml of 0.5% stock
phenol red, and 20ml of Opti-prep. Adjust to a final
volume of 50ml with ddH2
O and filter sterilize.
40% iodixanol: Mix 5ml of 10X PBS, 0.05ml of 1M
, 0.05 ml of 2.5M
KCl, 33.3 ml of Opti-prep.
Adjust to a final volume of 50ml with ddH2
60% iodixanol: Mix 0.05 ml of 1M
, 0.05ml of
2.5 M KCl, and 0.025 ml of 0.05% stock phenol red in
50ml of Opti-prep. Filter sterilize.
Phenol red (Gibco-BRL Life Technologies)
Slide-A-Lyzer 10,000 MWCO (No. 66451; Pierce)
Syringes (5 ml)
Needles (18 gauge)
Rubber policeman wings (no. 14-110; Fisher)
E. Dot Blot Assay
DNasel digestion mixture: 10mM
Tris-HCl, pH 7.5,
, 50U/ml DNase 1
Proteinase K digestion mixture: 1M
NaCl, 1% (w/v)
Sarkosyl, 200µg/ml of proteinase K
Whatman 3MM paper
Dot blot apparatus
Gene-Screen Plus membrane (New England Nuclear)
Random primer labeling kit (Roche Biomedical)
p-dCTP (Amersham Pharmacia Biotech)
Church buffer: Mix 5 g of bovine serum albumin, 1 ml
EDTA, 33.5 g of Na2
O, 1 ml of
, 35g of sodium dodecyl sulfate (SDS).
Adjust to a final volume of 0.5 liter with ddH2
Heat at 65°C to dissolve. Store on the laboratory
Hybridization low-stringency wash solution: 2X saline
sodium citrate (SSC), 2X 0.1% (w/v) SDS
Hybridization medium-stringency wash solution: 0.5X
SSC, 1X 0.1% (w/v) SDS
Hybridization high-stringency wash solution: 0.1X
SSC, 0.5X 0.1% (w/v) SDS
2X SSC: Mix 17.5g of NaCl and 8.8g of trisodium
O. Adjust the final volume to 1 liter and
adjust the pH to 7.0.
Hybridization bottles (Gibco-BRL Life Technologies)
Phosphoimager or scintillation counter
A. Ad-Free Production of Recombinant Virus
1. Construction of rAAV Plasmid Vector
2. Transfection of 293 cells
- Modify the plasmid psub201 by digestion with
enzymes XbaI and HindIII to remove the Rep and Cap genes. This should produce a fragment less
than 4.6 kb and should contain the AAV ITRs.
- Insert the foreign gene cassette made of the transgene
and its promoter between the XbaI sites.
: Warm up 2X HeBS buffer, 2.5 M
CaCl, and filtered
distilled water to room temperature.
3. Cell Harvesting
- Seed 293 cells 24 h before transfection at 22 × 106 cells per 15-cm dish in complete DMEM. Generally a
total of 20 dishes are transfected for the viral preparation
- Three hours before the transfection, aspirate
media and replace with 25ml prewarmed complete
- Prepare transfection mixture. Each 15-cm dish is
transfected with 25 ml of preformed DNA-CaPO4 precipitate.
A total of 37.5 µg DNA per dish will give the
optimal precipitate. This should include 7.5 µg vector
plasmid, 7.5 µg helper plasmid (i.e., XX2), and 22.5 µg
of Ad plasmid (i.e., XX6-80). The precipitates should
be formed in reactions of 40-ml. A 40-ml reaction will
transfect 20 × 15-cm dishes.
- Aliquot DNA mixture into a sterile 50-ml
Corning conical tube. Bring volume to 18ml with
sterile ddH2O. Add 2ml of 2.5M CaCl2. Aliquot 20ml
of 2X HeBS to a separate 50-ml Corning conical tube.
- Add the DNA mixture by pipette dropwise to the
tube containing the 2X HeBS. Gently mix by inversion.
Allow 3-5 min for precipitate to form.
- Once precipitate has formed, quickly but gently
add 2ml of transfection solution dropwise in a circular
motion around the plate. Swirl gently to mix.
- Sixteen hours posttransfection, aspirate the
media of each plate and replace with 25 ml of complete
: Harvesting should occur within 64 h but not
less than 48 h of transfection.
4. Purification of Recombinant Virus
- Use a rubber policeman to detach cells from
dishes using a scraping motion. Pipette the entire
mixture into two sterile 500-ml centrifuge bottles
(maximum 250ml per bottle). Spin at 5000rpm for
10rain. Pour off the supernatant. (Note: Cell pellets
may be frozen in a -20°C freezer overnight if
5. Purification Using an Iodixanol
- Resuspend the cell pellet in 1X PBS in a sterile 50-
ml conical tube.
- Add 0.35ml of stock 10% DOC (sodium deoxycholate).
Final concentration should be 0.5%.
- Add 1.4µl of stock 250µg/µl benzonase. Final concentration
should be 50µg/µl.
- Incubate at 37°C for 30min.
- Spin for 30rain at 6500rpm (8400g). Transfer the
supernatant to a sterile 50-ml conical tube.
- Freeze/thaw the cell suspension using a dry iceethanol
bath and a 37°C H2O bath.
- Repeat the freeze/thaw step two more times.
- Spin for 20 min at 6500 rpm. Decant the supernatant
to a new tube. (Note: Sample may be stored in
-80°C freezer at this point).
6. Purification Using a Heparin Column
- Dissolve the pellet into PBS-MK. Place half
(7.5ml) of the resuspended pellet into each of two
Optiseal tubes. Using the Pump Pro, create a step gradient
by underlying the sample with 6 ml of 15% iodixanol,
6 ml of 25% iodixanol, 7 ml of 40% iodixanol, and
5ml of 60% iodixanol. Carefully remove the tubing
without disturbing the gradient layers. (Note: A
syringe and small diameter tubing may also be used to
layer the gradients.)
- Balance the tubes and fill them completely by
slowly adding 1X PBS dropwise to the tube to the
uppermost layer. Insert a plug and centrifuge at 4°C for 1 h using a Beckman Ti-70 rotor at 70,000rpm
- Carefully remove the Optiseal tubes from the
rotor. In a viral hood, remove the plug from the top of
the tube. Use a 5-ml syringe with an 18-gauge needle
to puncture the tube at the 40-60% iodixanol interface.
Remove 75% of the 40% iodixanol layer.
: A viral preparation made from twenty 15-cm
dishes of 293 cells can be purified on a 1-ml column
with a single injection.
7. Delivery of Recombinant Virus in Vitro
a. Determination of rAAV titer by Dot Blot Assay
- Sterilize the lines of the AKTA FPLC with 0.5M NaOH and 1X PBS.
- Run the "pump wash" program.
- Set up the heparin column and check it for leaks.
- Inject the sample and autozero the UV.
- Run the 1-ml heparin column program.
- Collect 0.5-ml fractions.
- Sterilize the lines again using 0.5M NaOH and 1X
- Discard the column.
- Test the fractions for the presence of virus using the
dot blot hybridization assay and combine the fractions
with the highest concentration of virus.
- Place 5µl of each fraction collected from the
heparin sulfate column into a well of a 96-well
microtiter plate. Assay duplicate samples of each
- Add 50µl of DNase 1 digestion mixture and
incubate for 1 h at 37°C. This treatment digests
any viral DNA that has not been packaged into
- After 1 h, stop the digestion by adding 10µl of
0.1 M EDTA to each reaction. Mix well.
- Add 60µl of proteinase K digestion mixture to
each sample to release the viral DNA from the capsid.
Incubate for 30rain at 50°C.
- To create a set of DNA hybridization standards,
use plasmid DNA that was used for the transfection.
Linearize the plasmid and do serial dilutions in 10 mM
Tris-HCl (pH 8.0) and 1 mM EDTA. A volume of 25 µl
is convenient for each standard in wells of a 96-well
microtiter plate. A suitable standard working range is
500 ng to 10 fg.
- Denature the samples and control DNA by
adding 100µl of 0.5M NaOH to each.
- Equilibrate a nylon membrane and one piece of
Whatman blotting paper in 0.4M Tris-HCl (pH 7.5)
and place them between the upper and lower blocks
of a dot blot manifold apparatus; membrane should be
on top of the Whatman paper.
- Add the denatured DNA from the 96-well
microtiter plate to wells of the dot blot manifold apparatus
in the absence of a vacuum. After all the DNA
has been transferred into the manifold, apply a
vacuum for 3-5 min.
- Radiolabel a transgene cassette-specific probe.
(Note: The probe should not contain plasmid backbone
or ITR sequences.)
- In a hybridization bottle, prehybridize the
nylon membrane with 5 ml of Church buffer for 5 min
at 65°C. Discard the prehybridization Church buffer
and replace with 5 ml of fresh Church buffer. Place at
- Boil the 32p-dCTP-radiolabeled probe for 5 min,
place on ice, and add to the hybridization bottle containing
the dot blot. Hybridize overnight at 65°C.
- Remove the hybridization solution and add
10 ml of low-stringency wash solution. Wash for 10 min
at 65°C. Repeat the wash with 10ml of fresh solution.
- Wash the dot blot for 10min at 65°C with the
medium-stringency wash solution and discard the
- Monitor the dot blot with a Geiger counter. Continue
the washes if needed using the high-stringency
wash solution. Do not let the membrane dry out or the
probe will permanently adhere to the membrane.
- To quantitate each spot on the dot blot, expose
the filter to a Phosphoimager cassette. Alternatively,
employ X-ray film to identify labeled regions on the
nylon membrane, excise each sample, and quantitate
using a scintillation counter.
- Plot a standard curve of DNA concentration vs
integrated intensity per counts per minute for the
DNA standards and employ the curve to determine
the concentration of DNA in fractions obtained from
the heparin sulfate column. (Note: The replication
center assay is also a useful method to calculate the
rAAV titer. The rAAV particle number of each fraction
can be calculated. Remember to take into consideration
that plasmid standards are double stranded
whereas rAAV virions harbor only a single strand.)
- HindIII is used in the digest to cut the rep and cap fragment in half for easy isolation of the plasmid
- DNA preparation should be pure. Purify your
viral fragment by agarose gel separation and running
onto Whatman DEAE-8 1 paper or a preparation of
equivalent high quality.
- Alternatively, blunt-end ligation may be used to
construct the rAAV vector plasmid.
- For efficient packaging into AAV capsids, the size
of the rAAV construct (including the 190-bp ITRs)
must be 4.6 kb or less.
- Plasmids are grown in the Sure strain of Escherichia coli. The literature shows that AAV ITRs are
unstable in bacteria. To avoid deletion, restrict bacterial
growth in the stationary phase. If you still obtain
deletions, grow the plasmids at 30°C for only 12 h. The
integrity of the plasmids can be assayed by restriction
- Polypropylene and glass attract ionic strengthdependent
aggregates more than polystyrene. For this
reason, mixing containers made of polystyrene are preferred
- The total DNA is equal to 37.5µg/plate, and the
ratio of rAAV construct to the pXX2 and pXX6 is equal
to a molecular ratio of about 1:1:1.
- If a coarse precipitate forms, decrease the incubation
time. If a precipitate forms too quickly (i.e., less
than I min) check the pH of the 2X HeBS. The pH range
of 2X HeBs should be between 7.05 and 7.12.
- Only use sterile ddH2O. Do not autoclave!
- After 24 h, the 293 cells (when viewed through
a microscope) should have a rounded appearance,
indicating viral replication. If detached cells are noted,
the incubation was too long.
- Cell suspensions or cell precipitates may be
stored at -20°C for up to 6 months.
- The NaCl in the 15% iodixanol layer will separate
viral aggregates that may form due to the high
concentration of virus.
- Collecting fractions from the flow-through and
washing steps ensures that all the virus was bound to
the column and eluted as the salt gradient increased.
- Using a new heparin column for each purification
ensures that you will not cross-contaminate viral
Amiss, T. J., and Samulski, R. J., (2000). Methods for adenoassociated
virus-mediated gene transfer into muscle. In "Methods
in Molecular Biology"
(M. P. Starkey, R. Elaswarapu, ed.), Vol. 175,
pp. 455-469. Humana Press, Totowa, NJ.
Berns, K. I., and Giraud, C. (1995). Adeno-associated virus (AAV)
vectors in gene therapy. Curr. Topics Microbiol. Immun
Bartlett, J. S., and Samulski, R. J. (1996). Production of recombinant
adeno-associated viral vectors. In "'Current Protocols in Human
, pp. 12.1.1-12.1.24. Wiley, Philadelphia.
Daly, T. M., Okuyama, T., Vogler, C., Haskins, M. E., Muzyczka, N.,
and Sands, M. S. (1999). Neonatal intramuscular injection with
recombinant adeno-associated virus results in prolonged betaglucuronidase
expression in situ and correction of liver pathology
in mucopolysaccharidosis in type VII mice. Hum.Gene Ther
Ferrari, E K., Xiao, X., McCarty, D., and Samulski, R. J. (1997). New
developments in the generaion of Ad-free, high-titer rAAV gene
therapy vectors. Nature med
Hermonat, P. L., Mendelson, E., and Carter, B. J. (1984). Use of adenoassociated
virus as a mammalian DNA cloning vector: Transduction
of neomycin resistance into mammalian tissue culture
cells. Proc. Natl. Acad. Sci. USA 81
Kotin, R. M., Siniscalco, M., Samulski, R. J., Zhu, X. D., Hunter, L.,
Laughlin, C. A., Mclaughlin, S., Muzyczka, N., Rocchi, M., and
Berns, K. I. (1990). Site-specific integration by adeno-associated
virus. Proc. Natl. Acad. Sci. USA 87
Monahan, P. E., Samulski, R. J., Tazelaar, J., Xiao, X., Nichols, T. C.,
Bellinger, D. A., and Read, M. S. (1998). Direct intramuscular
injection with recombinant AAV vectors results in sustained
expression in a dog model of hemophilia. Gene Ther
Samulski, R. J. (1993). Adeno-associated virus: Integration at a specific
chromosomal location. Curr. Opin. Gen. Dev
Samulski, R. J., Sally, M., and Muzycka, N. (1999). Adeno-associated
viral vectors. In "The Development of Human Gene Therapy"
Friedmann, ed.), pp. 131-172. Cold Spring Harbor Laboratory,
Cold Spring Harbor, NY.
Samulski, R. J., Zhu, X., Xiao, X., Brook, J. D., Housman, D. E.,
Epstein, N., and Hunter, L. A. (1991). Targeted integration of
adeno-associated virus (AAV) into human chromosome 19 [published
erratum appears in EMBO J 11
(3), 1228 (1992)]. EMBO J.
Song, S., Morgan, M., Ellis, T., Poirier, A., Chesnut, K., Wang, J.,
Brantly, M., Muzyzcka, N., Byrne, B. J., Atkinson, M., and
Flotte, T. R. (1998). Sustained secretion of human alpha-1-
antitrypsin from murine muscle transduced with adenoassociated
virus vectors. Proc. Natl. Acad. Sci. USA 95
Summerford, C., and Samulski, R. J. (1998). Membrane-associated
heparin sulfate proteoglycan is a receptor for adeno-associated
virus type 2 virions. J. Virol
Summerford, C., and Samulski, R. J. (1999). Viral receptors and
vector purification: New approaches for generating clinicalgrade
reagents. Nature Med
Xiao, X., Li, J., and Samulski, R. J. (1996). Efficient long-term gene
transfer into muscle tissue of immunocompetent mice by adenoassociated
virus vector. J. Virol
Xiao, X., Li, J., and Samulski, R. J. (1998). Production of high-titer
recombinant adeno-associated virus vectors in the absence of
helper adenovirus. J. Virol
Zolotukhin, S., Byrne, B. J., Mason, E., Zolotuknin, I., Potter, M.,
Chesnut, K., Summerford, C., Samulski, R. J., and Muzyczka, N.
(1999). Recombinant adeno-associated virus purification using
novel methods improves infectious titer and yield. Gene Ther