Original Article

Gene Therapy (2011) 18, 709–718; doi:10.1038/gt.2011.19; published online 10 March 2011

Longevity of rAAV vector and plasmid DNA in blood after intramuscular injection in nonhuman primates: implications for gene doping

W Ni1,5, C Le Guiner2,3,5, G Gernoux2, M Penaud-Budloo1,2, P Moullier1,2,3 and R O Snyder1,2,4

  1. 1Department of Molecular Genetics and Microbiology, College of Medicine, University of Florida, Gainesville, FL, USA
  2. 2Laboratoire de Thérapie Génique, INSERM UMR649, IRT UN, Nantes, France
  3. 3GENETHON, Evry, France
  4. 4Department of Pediatrics, College of Medicine, University of Florida, FL, USA

Correspondence: Dr RO Snyder, Department of Molecular Genetics and Microbiology, College of Medicine, University of Florida, 1600 SWArcher Road, Gainesville, FL 32610-0266, USA. E-mail: rsnyder@cerhb.ufl.edu

5These authors contributed equally to this work.

Received 13 December 2010; Revised 31 January 2011; Accepted 31 January 2011; Published online 10 March 2011.



Legitimate uses of gene transfer technology can benefit from sensitive detection methods to determine vector biodistribution in pre-clinical studies and in human clinical trials, and similar methods can detect illegitimate gene transfer to provide sports-governing bodies with the ability to maintain fairness. Real-time PCR assays were developed to detect a performance-enhancing transgene (erythropoietin, EPO) and backbone sequences in the presence of endogenous cellular sequences. In addition to developing real-time PCR assays, the steps involved in DNA extraction, storage and transport were investigated. By real-time PCR, the vector transgene is distinguishable from the genomic DNA sequence because of the absence of introns, and the vector backbone can be identified by heterologous gene expression control elements. After performance of the assays was optimized, cynomolgus macaques received a single dose by intramuscular (IM) injection of plasmid DNA, a recombinant adeno-associated viral vector serotype 1 (rAAV1) or a rAAV8 vector expressing cynomolgus macaque EPO. Macaques received a high plasmid dose intended to achieve a significant, but not life-threatening, increase in hematocrit. rAAV vectors were used at low doses to achieve a small increase in hematocrit and to determine the limit of sensitivity for detecting rAAV sequences by single-step PCR. DNA extracted from white blood cells (WBCs) was tested to determine whether WBCs can be collaterally transfected by plasmid or transduced by rAAV vectors in this context, and can be used as a surrogate marker for gene doping. We demonstrate that IM injection of a conventional plasmid and rAAV vectors results in the presence of DNA that can be detected at high levels in blood before rapid elimination, and that rAAV genomes can persist for several months in WBCs.


plasmid; rAAV; white blood cells; gene doping; nonhuman primates; Taqman PCR



The use of performance-enhancing drugs in sports is commonly referred by the term ‘doping’, particularly by those regulatory bodies overseeing sports that have made efforts to fight doping. Through the efforts of WADA (World Anti-Doping Agency) and country-specific agencies, the total tested sample number in sport games has increased because of more athletes being screened and a larger repertoire of tests for various nongene transfer agents.1 Through these surveillance efforts, the integrity and fairness of sports can be protected. However, new kinds of doping approaches are possible, such as gene doping. Advances in gene transfer technology used for legitimate gene therapy and vaccination, are capable of providing the means to enhance the physical performance of athletes. According to WADA, gene doping is defined as the nontherapeutic use of genes and genetic elements that are capable of enhancing athletic performance. If gene doping is being pursued,2 it would undermine principles of fair play in athletics and, most importantly, could present major health risks to athletes who partake in gene doping. One attraction of gene doping lies in the apparent difficulty in detecting its use. Gene doping is now considered a real threat to the world of sports, and it is anticipated that the few athletes seeking an advantage will want to use the latest gene transfer technology.

For legitimate gene transfer used for gene therapy and vaccination, vector biodistribution is an important parameter used to evaluate safety in both pre-clinical and clinical studies.3, 4 The type of vector, the delivery method, injection schedule, route of administration and administrated dose are the main variables to affect biodistribution and shedding.5, 6, 7, 8, 9 Intramuscular (IM) injection is a convenient method owing to physical accessibility, mass of the tissue and access to the vasculature.10, 11, 12 Moreover, recombinant adeno-associated viral (rAAV) vectors and naked plasmid are two different gene transfer systems used for IM delivery in animal models13, 14, 15, 16, 17 and in humans.18, 19 Recently, regional vascular infusion of a vector to achieve skeletal muscle transduction has been reported for plasmid DNA (pDNA)20, 21 and for rAAV vectors.8, 22 Among the rAAV serotypes analyzed to date, rAAV1 and rAAV8 are among the most efficient for muscle transduction.8, 23, 24, 25, 26 After IM administration, it was demonstrated that rAAV DNA resides as episomal circles27, 28, 29 in a chromatin structure;30 and from a biosafety perspective, the inefficient integration of rAAV into the host genome is an attractive feature for the legitimate use of this gene transfer system.

Real-time PCR is a new technology, which has been widely used in recent years,31 and can detect specific DNA signals, even at a very low concentration with reliability and specificity. In addition, real-time PCR is amenable to automation and remote data collection. Here, we report the development of sensitive real-time PCR methods useful for detecting vector sequences present in blood after IM administration, and determine the short- and long-term profiles after naked plasmid and rAAV vector gene transfer in nonhuman primates (NHPs). We show that rAAV sequences can be found in white blood cells (WBCs) of NHPs for several months after low-dose IM administration, but not when pDNA is administrated IM at high doses. Data generated in this study will be the basis for developing a legally defensible commercial Taqman PCR assay for the detection of rAAV-mediated gene doping.



Evaluation of our screening approach to be used in humans was conducted in NHPs to closely mimic the biodistribution and longevity predicted in humans. For naked plasmid injection, a plasmid pKanaORI/PGKcmEPO was constructed. It has the cynomolgus macaque erythropoietin (cmEPO) gene under control of the human phosphoglycerate kinase (PGK) promoter and SV40 polyA (Figure 1a). The PGK promoter was chosen, as it is a weak promoter that is active in the muscle, to slightly increase hematocrit at a high dose of plasmid. Furthermore, the high dose of plasmid was injected to investigate the ability to transfect NHP WBCs in vivo.

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Plasmid and rAAV vector structures. (a) hPGK, human phosphoglycerate kinase promoter; cmEPO, cynomolgus macaque erythropoietin cDNA; SV40 pA, simian virus 40 polyadenylation site; Kan, kanamycin-resistance gene. (b) CMV, cytomegalovirus immediate early promoter; β-gbIVS, β-globin intron; cmEPO cDNA and SV40 pA. The expression cassette was cloned between AAV2-inverted terminal repeats. (c) The cmEPO genomic DNA contains four exons. The intron between exon 2 and exon3 is the longest (583bp) in this gene. Genetic elements are not drawn to scale.

Full figure and legend (90K)

The rAAV vector transgene cassette has cmEPO under control of the CMV promoter and SV40 polyA (Figure 1b); the promoter is a robust promoter in the muscle known to increase hematocrit at a low vector dose.12 In contrast to plasmid, previous in vivo studies have shown that after a single injection, rAAV genomes can be detected in peripheral blood mononuclear cells (PBMCs) for several months.8, 19 Our goal in this study is to determine whether under stringent conditions (very low doses of rAAV), rAAV genomes can be detected in WBCs, which were targeted because they are easily accessible and have the potential of carrying vector sequences in the long term. Detection of low copies of plasmid and rAAV sequences in blood required the evaluation and optimization of blood collection, DNA isolation and handling, as well as PCR amplification and analysis.

Total DNA extraction from WBCs

DNA was extracted from WBCs instead of isolating purified PBMCs on Ficoll gradients. Our goal was to decrease the number of sample manipulations, and to increase the yield and maintain the quality of the extracted DNA, as different testing laboratories are expected to use these methods. We excluded kits that included columns or resins, because we were concerned about the competition between genomic DNA (gDNA) and the plasmid or rAAV episomes for binding. On the basis of these criteria, we decided to evaluate the Gentra Puregene kit obtained from Qiagen (Courtaboeuf, France) and the Wizard kit from Promega (Charbonnières-les-Bains, France).

For kit evaluation, total DNA was extracted from the WBC fraction of 3ml of whole blood of three naive macaques (n=2 for each animal). The total gDNA yield was similar with both kits. For all extracted DNA samples, the A260/280 ratios were between 1.86 and 1.91, and the A260/230 ratios were ~2.2, which is consistent with the absence of proteins and other contaminants. We obtained gDNA that seems to remain intact, although a small amount of subchromosomal size fragments can be seen by agarose gel electrophoresis (data not shown). This quality of DNA is suitable for PCR amplification, as determined by the ability to amplify by conventional PCR, an endogenous gene (ε-globin, see below) from 50ng to 1μg of DNA (data not shown).

As published previously, in NHPs, rAAV genomes persist as chromatinized large concatemeric and small monomeric circles in the skeletal muscle.30 The rAAV genome structure (episomal or integrated) is unknown in WBCs; thus, we wanted to determine whether episomes could be extracted along with gDNA. We carried out blood gDNA extractions spiked with a large plasmid (35200bp mimicking rAAV concatemeric circles) or a small plasmid (5700bp mimicking rAAV monomeric circles) to evaluate the efficiency of capture in the presence of excess gDNA. In brief, DNA extractions were performed from whole blood of the same noninjected NHPs, using the two kits. Overall, 3ml blood was spiked with 3E6 copies of large or small plasmids, either at the first step of the extraction, that is, in whole blood mixed with a red blood lysis solution, or at the second step of the extraction, that is, on the white cell pellet after the first centrifugation step. The DNA yield of our extractions was between 100 and 150μg of total DNA for 3ml of whole blood. A quantitative PCR analysis that is specific for plasmid sequences was performed on 10 and 50ng of DNA, and quantified using a plasmid standard curve. Recovery of the initial spiked plasmid was then calculated (Table 1). To confirm the reliability of each extraction, the quality of the extracted DNA was checked by electrophoresis, spectrophotometry and by amplification of the ε-globin gene (data not shown).

When pDNA was spiked during the first step of extraction, recovery was very low, regardless of the extraction kit. It is likely that the vast majority of the plasmid was lost after lysis of red blood cells and the subsequent centrifugation step, when the supernatant (serum) was removed from the white cell pellet. This was confirmed with the results obtained when the plasmid was spiked directly into the WBC pellet in which recovery of both spiked plasmids was higher, and the Gentra Puregene DNA extraction kit was used for subsequent analyses as it was more efficient. Furthermore, capture of the larger plasmid was more efficient than that of the smaller plasmid. DNA extracted from WBCs using this method was comparable in quality and yield with DNA purified from PBMCs isolated on Ficoll gradients (data not shown).

Evaluation of DNA storage and transport

In our study, NHP injections, blood collection and DNA extraction were performed in Nantes (France) and the PCR analyses in Gainesville (FL, USA); hence, it was necessary to validate the stability of our DNA samples during storage and shipping. After extraction, total DNA was resuspended in Tris-EDTA buffer, stored one night at +4°C to dissolve and then frozen at −20°C. We favored aliquoting into 100μl to avoid repeated thawing and freezing cycles. One aliquot of each sample was sent to Gainesville and backup samples were archived in Nantes. The shipping was carried out at ambient temperatures at different times of the year through overnight courier to reduce the cost associated with frozen or refrigerated shipping, and samples were delivered in <72h. Upon arrival, integrity of the DNA was determined by spectrophotometry at A260 and electrophoresis. To further confirm the integrity of the DNA, all samples were subjected to real-time PCR using an assay targeting the ε-globin genomic locus (Table 2), in which all gDNA samples produced similar ε-globin signals before and after shipment (data not shown).

Taqman real-time PCR assay design

To detect the biodistribution of injected DNA sequences, sensitive and specific assays that can distinguish exogenous DNA from NHP gDNA were developed. The SV40 polyA real-time PCR assay was adapted with changes from the assay developed by Lock et al.32 for the rAAV2 Reference Standard Material. The SV40 polyA site, which is not present in the NHP genome but is present in the naked plasmid and rAAV vectors used here, is an ideal target as there is no competition with endogenous sequences. In addition, the kanamycin-resistance cassette in the pKanaORI/PGKcmEPO plasmid provides an additional target for pDNA. Although these two sequence targets can be used to detect naked plasmid and rAAV vectors, we also want to detect the transgene (cmEPO) sequence, because when screening athletes, the other genetic elements incorporated in a vector illicitly used for performance enhancement would be unknown. As the administrated cmEPO sequence is a cDNA, the exon–exon junctions are targets to distinguish exogenous DNA from endogenous gDNA. A Taqman primer-probe set was designed to span the EPO Exon 2–3 boundary with the probe hybridizing to the exon–exon junction (Figure 1). The Taqman primer-probe sets were designed and validated to specifically detect naked plasmid and rAAV vectors in the presence of excess gDNA (Table 2). To verify the amount of DNA that was analyzed in each reaction, the ε-globin PCR assay was used to normalize the input DNA. As expected, EPO Exon 2–3, SV40 and Kan primer-probe sets specifically detect cmEPO cDNA, SV40 polyA and kanamycin targets, respectively, and do not produce a signal for NHP gDNA alone. Conversely, the ε-globin primer-probe set detects only gDNA.

Efficiency and linearity of Taqman primer-probe sets

The input DNA amount was based on careful spectrophotometric measurement using an Implen nanophotometer (Implen, München, Germany). An amount of 500ng of gDNA, equivalent to 75000 WBCs, was analyzed in each PCR assay. To determine the possible competition from gDNA, pDNA was spiked into naive gDNA and testing was performed using Exon 2–3, SV40 and Kan assays. Standard curves were established using the pKanaORI/PGKcmEPO plasmid ranging from 10 to 109 copies in the absence or presence of 500ng naive gDNA. As shown in Table 3, linearity is maintained over eight logs with all three primer-probe sets in the presence or absence of 500ng gDNA. The efficiency of the EPO Exon 2–3 assay was inhibited by 10% in the presence of 500ng gDNA and may be due to competition with gDNA for the EPO Exon 2–3 primers. Meanwhile, no competition was observed with the Kan and SV40 assays because none of the Kan and SV40 primer- and probe-binding sites exist in gDNA (Table 3). These two assays were considered to have the best sensitivity and serve as a reference for the EPO Exon 2–3 assay.

False-positive and false-negative testing

After establishing these vector-specific Taqman PCR assays, the lower limit of quantitation was determined. On the basis of the work above, we reliably detected three copies of the pKanaORI/PGKcmEPO plasmid spiked into 500ng gDNA within 40 PCR cycles. A lack of signal at 40 cycles was defined as ‘negative’, and a Ct signal before 40 cycles of amplification was considered ‘positive’. Overall, 500ng of naive gDNA and the Tris-EDTA buffer by itself were used as ‘no template controls’. Table 4 demonstrates that when 20 replicates were executed, the false-positive rates of each of the three assays are equal to zero. Meanwhile, when three copies of plasmids in 10μl Tris-EDTA were tested in 15 replicates of the assay, the false-negative rates of the EPO Exon 2–3, SV40 and Kan assays in the presence of 500ng gDNA were 13, 0 and 7%, respectively. Pipetting error most likely is the cause of false-negative results, as the Poisson distribution yields a 5% probability to pipet zero copies of the plasmid in a test sample. The results demonstrate that three copies of target plasmid in the presence of 500ng gDNA, which is equivalent to 75000 WBCs, are detected by the three assays, with the copy number obtained in all three assays having means of approximately three copies (Table 4).

Naked plasmid expression and biodistribution

Two NHPs were injected IM with 10mg pKanaORI/PGKcmEPO at pre-tattooed sites along the tibialis anterior muscle. In vivo plasmid expression was evaluated by measuring hematological parameters (such as hematocrit and reticulocytes) as indirect measurements of in vivo cmEPO expression. A slight increase in the number of reticulocytes was seen in Mac1 during the first week post injection (p.i.) (Figure 2a), but this increase had no effect on the hematocrit, and at 4 weeks p.i., the number of reticulocytes decreased until reaching the baseline value. No significant increase in the number of reticulocytes or hematocrit was observed in Mac2 (Figure 2b). Mac2 received some Fercobsang (a vitamin cocktail against anemia containing iron, copper, cobalt and vitamins B1, B12 and PP) between week 2 and week 10 p.i., as its basal hematocrit during this period was low (near 30%) and the animal was subjected to repetitive bleedings for our analyses. This treatment is expected to affect the hematocrit and reticulocytes levels, and could explain the variations observed in the hematological parameters of Mac2.

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Expression and biodistribution of the naked plasmid after IM injection. Hematological parameters (hematocrit in solid circles and reticulocytes in dashed open triangles) in (a) Mac1 and (b) Mac2 before and after injection. (c) Serum EPO levels of Mac1 and Mac2 before and after plasmid injection. The results are the average of three independent measurements. Copy number of the naked plasmid in white blood cells of (d) Mac1 and (e) Mac2 at different sampling times. WBC samples from Mac1 and Mac2 were analyzed by real-time PCR. The plasmid copy number is extrapolated from Exon 2–3, SV40 and Kan quantitative PCR assay standard curves. Plasmid copy numbers were normalized to gDNA using the ε-globin primer-probe set. (f) Plasmid copy number from the serum of Mac1 and Mac2 using the Exon 2–3 assay.

Full figure and legend (123K)

To directly determine whether the plasmid injection resulted in transgene expression, cmEPO protein levels were measured by enzyme-linked immunosorbent assay in the serum of Mac1 and Mac2 (Figure 2c). No significant increase in the cmEPO protein was observed after injection of pDNA, and the levels of cmEPO remained at basal values throughout the study. Variation in EPO levels in Mac2 could be due to anemia of this animal and due to Fercobsang treatment. As we injected 10mg (a very high dose equal to 2.75E15 molecules) of a functional plasmid, these results were quite surprising, considering that the human PGK promoter used in this expression cassette is functional in the macaque skeletal muscle (Le Guiner C and Moullier P, unpublished data).

To track pDNA in blood, EPO Exon 2–3, SV40 and Kan assays were used to analyze WBC DNA samples obtained from Mac1 and Mac2 at different time points. All three primer-probe sets showed a similar trend in WBCs (Figures 2d and e). Copy number was detected as soon as 30min p.i. and reached maximum at 1 day p.i., and then fell back to the pre-injected level by 1 week. The data demonstrated that the pKanaORI/PGKcmEPO plasmid enters the blood shortly after IM injection. Negative control samples (blood samples obtained from a noninjected animal sampled at the same time points as injected animals) were used to determine the background and showed no PCR signal (data not shown). Comparing the copy number from injected and noninjected animals, we can conclude that the plasmid cannot be detected in WBCs more than 3 weeks after IM injection. Serum samples obtained from Mac1 and Mac2 were also tested for the presence of pDNA. Similar to the signal from WBCs, the copy number in the serum of both animals reached the maximum at 1 day p.i. However, the peak copy number of Mac1 is 20 times higher than that of Mac2 (Figure 2f) and pDNA in serum is cleared within 4 weeks.

To evaluate the transfection of myofibers at the site of plasmid injection, we performed PCR analysis at 3 months p.i. on muscle biopsies in the injected legs of Mac1 and Mac2. Total DNA was extracted from muscle biopsies and analyzed using the EPO Exon 2–3 assay, and we observed the presence of pDNA in the injected muscle at 3 months p.i. at copy numbers ranging from 20 to 200 per 500ng gDNA.

rAAV vector expression

Four NHPs were injected IM with low doses of rAAV1 or rAAV8 vectors harboring the same expression cassette (CMV-cmEPO-SV40pA). For rAAV1, Mac3 received 2.5E10vgkg−1 (7.75E10vg total) and Mac4 received 2.5E11vgkg−1 (9E11vg total). For rAAV8, Mac5 was injected with 5E9vgkg−1 (1.3E10vg total) and Mac6 received 2.5E10vgkg−1 (1.01E11vg total). The in vivo cmEPO expression of rAAV vector-injected animals was evaluated indirectly by measuring hematological parameters (such as hematocrit and reticulocytes) in each rAAV-injected animal (Figures 3a–f), and directly by measuring cmEPO secretion in the serum of each rAAV-injected animal (Figures 3c–h). In contrast to the inefficient expression from the naked plasmid shown in Figure 2, rAAV1 and rAAV8 vectors efficiently expressed EPO in all four animals. Mac3, Mac5 and Mac6 exhibited long-term cmEPO expression, with a rapid increase in the number of reticulocytes and hematocrit levels that increased ~15% for Mac3 and Mac5, and >20% for Mac6 after rAAV injection. This 20% increase in Mac6 required scheduled bleeds each week to maintain the hematocrit level under 60% (an ethically acceptable level). For Mac4 (that received the highest vector dose), the hematocrit level increased until week 4, and the number of reticulocytes increased until week 2 and then decreased; this is in correlation with a transient increase in cmEPO in the serum of this animal in this timeframe. This animal exhibited autoimmune anemia, with the development of an immune response against autologous cmEPO (data not shown), as seen in previous studies.33, 34 At week 17 p.i., the hematocrit decreased below 15%, and as this phenomenon was not transient, the animal was killed at 17.5 weeks p.i.

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Hematological parameters and EPO levels in the serum of rAAV vector-injected NHP. Hematological parameters (hematocrit in solid circles and reticulocytes in dashed open triangles) in (a) Mac3, (b) Mac4, (e) Mac5 and (f) Mac6 before and after rAAV administration. Serum EPO levels of (c) Mac3, (d) Mac4, (g) Mac5 and (h) Mac6 before and after rAAV injection. The results are the average of three independent measurements. For rAAV1, Mac3 received 2.5E10vgkg−1 (7.75E10vg total) and Mac4 received 2.5E11vgkg−1 (9E11vg total). For rAAV8, Mac5 was injected with 5E9vgkg−1 (1.3E10vg total) and Mac6 received 2.5E10vgkg−1 (1.01E11vg total).

Full figure and legend (168K)

rAAV vector DNA in blood

DNA samples were extracted from WBCs, serum and urine of the four rAAV-transduced macaques. Overall, 500ng WBC gDNA was analyzed using the EPO and SV40 Taqman PCR assays. Negative control samples (blood samples obtained from noninjected animals sampled at the same time points as injected animals) were used to establish a background PCR signal. No signal was obtained from all the negative control animal samples (data not shown). Taqman PCR results illustrate (Figures 4a and d) that pre-injection samples obtained from all four rAAV-injected animals were negative for both assays and that the rAAV vector appears in WBCs 30min after IM injection; this likely represents vector in the serum (see Figure 4e), which may contaminate WBCs at a low level. The maximum copy number in WBCs and serum occurred at 1 day p.i. and likely includes infectious rAAV detectable until 7 days p.i.8 The copy number in WBCs and serum decreased sharply until 3 weeks p.i., and then approximately three copies of the vector were maintained in WBCs for >20 weeks in all primates, but not serum. Although the amount of vector DNA is low, the sensitive Taqman PCR assays developed here are capable of reliably detecting the vector genomes. As mentioned before, the SV40 assay is more efficient than the EPO Exon 2–3 assay (Table 3), so that at some time points, only the SV40 assay is capable of detecting a signal.

Figure 4.
Figure 4 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

rAAV vector copy number in white blood cells, serum and urine after IM injection. Copy number of the rAAV1 vector in white blood cells of (a) Mac3 and (b) Mac4. Copy number of the rAAV8 vector in white blood cells of (c) Mac5 and (d) Mac6. The copy number of DNA is extrapolated from Exon 2–3, and SV40 quantitative PCR assay standard curves. (e) rAAV copy number in the serum of Mac3, Mac4, Mac5 and Mac6. (f) rAAV copy number in urine of Mac3, Mac4, Mac5 and 6. Vector in serum and urine was detected using the Exon 2–3 assay.

Full figure and legend (120K)

Similar to plasmid results (Figure 2), most of the rAAV vector genomes were cleared from serum in 5 weeks (Figure 4e). In addition, a significant amount of vector was detected in urine within the first 5 days, but no target DNA signal was observed in urine samples later than 5 weeks (Figure 4f). At 1 day p.i., vector copies in WBCs for Mac3 (rAAV1) were nearly one log greater than those seen for Mac6 (rAAV8), which received the same dose; thus, it seems that rAAV1 is more efficient to enter the circulation. Moreover, muscle gDNA samples near the injection sites from the four rAAV-injected animals were extracted and analyzed by real-time PCR to demonstrate the successful transduction of the skeletal muscle with copy numbers ranging from 50 to 2000 per 500ng gDNA.



Detection of gene doping requires methods that are reliable, sensitive and specific, using samples obtained noninvasively. Although injected plasmid and rAAV genomes can be maintained in the injected muscle for at least 3 months, it is not feasible to perform muscle biopsies on athletes, especially without knowing the location of the injection. Thus, the ideal sample should be blood, urine or serum, which the athletes are required to provide before competing. Although Baoutina et al.35 have demonstrated that real-time PCR can be used to detect cDNA spiked into human blood; it is unknown from their study how sensitive the assay is for samples taken from animals transduced in vivo. The efficiency of vector escape from the injection site into the circulation, and the detection of low doses in NHPs provide the best model to test the limit of detection related to dose. Our real-time PCR assays have been used to test samples from six NHPs transduced in vivo, and these data provide sufficient evidence that suggests that the rAAV vector genome can be maintained in WBCs up to 57 weeks after IM injection of rAAV8 and up to 26 weeks with rAAV1 (the longest points sampled). Meanwhile, the rAAV vector in serum and urine is mostly cleared within 5 weeks p.i., similar to the clearance profile of pDNA. From these results, DNA samples extracted from WBCs should be considered as a testing target in future sport competitions.

gDNA isolated from rAAV1- and rAAV8-injected animals showed a positive signal at 21 months p.i. in purified PBMCs (data not shown), and Toromanoff et al.8 also observed rAAV1 in PBMCs at 34 months and rAAV8 in PBMCs at 20 months after IM injection. However, the methods used here to extract DNA from WBCs do not require PBMC isolation, which reduces (1) variability, (2) required sample volume to 3ml (versus 5–10ml with PBMC isolation), (3) required processing time and (4) risk of sample cross-contamination. Meanwhile, the detection methods we are applying are based on single-step real-time PCR assays, which simplify the testing procedures by avoiding both the electrophoresis detection from conventional PCR analysis and the multiple amplification steps of nested PCR.36

The biodistribution analysis of rAAV vectors in animal models and humans is usually limited to the short-term relative to possible life-long expression. Flotte et al.14 demonstrated that rAAV1 vector genomes can be detected in mouse and rabbit blood up to 90 days after IM injection of doses greater than or equal to1E13vgkg−1 (40–2000-fold higher doses than those used here). Beiter et al.36 also showed that rAAV vectors can be detected in mouse blood up to 56 days using a nested PCR strategy. However, comparing the mouse and rabbit models, NHPs more closely compare with human beings in the muscle structure, blood volume to mass ratio, viral receptors and immune response.16 Here, we demonstrated rAAV1 and rAAV8 tropism for NHP WBCs after IM injection, although we did not assess whether transgene expression originates from infected WBCs. For Mac4, autoimmune anemia may have been facilitated by transduction of dendritic cells by the rAAV1 vector, which has been shown to be efficient at transducing this cell type.37

Manno et al.19 successfully transduced the human liver using a rAAV2-Factor IX vector. They observed the presence of the rAAV2 vector up to 20 weeks in human PBMCs, up to 14 weeks in serum and up to 4 weeks in urine, with their highest dose (2E12vgkg−1), which is 8–80-fold higher than the doses administrated to the NHPs in this report. Brantly et al.38 detected rAAV1 vector sequences in human blood up to 90 days after IM injection with the highest dose (1E12vgkg−1). These groups used PCR assays with a sensitivity of 50–100 copies in the presence of 1μg of gDNA.

The assays developed here have high sensitivity with reliable detection of three copies of target DNA sequences in the presence of 150000 copies of host genomes. As a result, the combination of DNA isolation methods together with these assays is capable of detecting the minute amounts of vector sequences in 75000 WBCs of NHP genomes. Five copies of vector in 500ng gDNA, which is extracted from the equivalent of 10–15μl of whole blood, indicate that 150000 rAAV vector genomes are in the circulation (~300ml), demonstrating that the rAAV vector genome can be maintained in a sub-population of WBCs in the long term. We detected the rAAV vector genome in WBCs at the lowest dose for each serotype (Mac3: rAAV1 at 2.5E10vgkg−1 (=7.75E10vg total dose) and Mac5: rAAV8 at 5E9vgkg−1 (=1.3E10vg total dose)) and achieved an increase in hematocrit. Athletes would potentially target similar vgkg−1 equal to a 1.75E12vg dose for rAAV1 and a 3.50E11vg dose for AAV8 per 70kg human.

Unlike the rAAV signal, the plasmid signal is not maintained in WBCs in the long term and fell back to baseline within a week p.i. and may not have been associated with WBCs, but instead confined to serum. A significant amount of DNA localizes to the blood before rapid elimination by the liver,39 and an additional mechanism for plasmid elimination from the blood may involve kidney secretion into the urine. Furthermore, the low amount of pDNA detected 3 months after injection at the muscle site could be the result of inefficient transfection, or transfected myofiber elimination after an immune cell infiltration in response to CpG sequences found in prokaryotic sequences (that is, the plasmid backbone used here).40, 41, 42

The long-term maintenance of rAAV sequences in WBCs provides an easily accessible target for the surveillance of gene doping. The methods developed here to test gene doping also provide the basis for detecting other WADA-prohibited gene-doping targets. Furthermore, the assays developed here performed similarly in different matrices of specimens: blood, urine, muscle and serum. Our experiments in NHPs were designed to test the feasibility of detecting vector sequences in blood in the long term after IM injection and to gain insights into testing humans. Storage and shipping of DNA required validation, as athlete samples will be shipped to testing laboratories and results sent to central databases. The data generated here are being used to develop similar assays to test homologous human transgenes. Given that gene transfer technology encompasses various vectors, routes of administration, as well as injection formulations, vector biodistribution can vary widely. Thus, another outcome of our work will be to better define assays that can be used to elucidate vector distribution for legitimate gene therapy applications.


Materials and methods

Recombinant plasmid

The pKanaORI/PGKcmEPO plasmid contains the cmEPO cDNA under the control of the human PGK promoter and the SV40 polyA sequence. The backbone of the plasmid contains a kanamycin-resistance cassette and the DNA replication origin from pUC. The integrity of the plasmid was checked by complete sequencing.

For animal studies, the plasmid was amplified in Escherichia coli (DH5α from Invitrogen, Cergy Pontoise, France), and purified using the ‘Nucleobond PC 10000-EF’ Endotoxin-Free DNA Plasmid kit obtained from Macherey-Nagel (Hoerd, France). Plasmid was formulated at a concentration of 7.9mgml−1 in normal saline solution. Concentration and purity of the plasmid were checked on a nano-spectrophotometer obtained from Implen. The identity of the plasmid was verified by restriction digestion, using restriction enzymes from New England Biolabs (Saint Quentin Yvelines, France). The sterility of the plasmid was verified using the Hemoline Performance Duo kit obtained from Biomérieux (Craponne, France). The functionality of the plasmid was checked on DAE7 cells, which are sensitive to cmEPO. In brief, 48h after transfection of the plasmid in 293 cells, the supernatant containing cmEPO was harvested and different dilutions were added to DAE7 cells. Forty-eight hours later, a cytotoxic test with MTT (Thiazolyl Blue Tetrazolium Bromide, Sigma, Lyon, France) was performed on DAE7 cells. The positive control was recombinant human EPO (Eprex, Janssen-Cliag, Issy-les-Moulineaux, France), and the negative control was an untransfected 293 cell supernatant.

rAAV vector production

The pSSV9-MD2-cmEPO rAAV vector plasmid harbors the cmEPO transgene under control of the CMV promoter and SV40 pA. The integrity of the plasmid was checked by complete sequencing.

The rAAV1-MD2-cmEPO and rAAV8-MD2-cmEPO vectors were prepared by transient transfection of 293 cells and purified by cesium chloride density gradients, followed by extensive dialysis against phosphate-buffered saline.43 Appropriate quality control was performed to evaluate viral vector purity, vector genome titer and infectious titer.

Animal administration and analysis

Experiments were conducted on captive-bred cynomolgus macaques purchased from BioPrim (Baziège, France). The Institutional Animal Care and Use Committee of the University of Nantes (France) approved the protocol. Animals were prescreened for the presence of anti-AAV1 or anti-AAV8 antibodies, SV40 and other pathogens. All injections and blood samples were collected under ketamine-induced anesthesia (10mgkg−1). Creatine phosphokinase levels were monitored in the six primates and all stayed in the normal range before injection and throughout the protocol after administration of the rAAV or plasmid.

Plasmid administration

The pKanaORI/PGFcmEPO pDNA was delivered IM to Mac1 and Mac2. The total plasmid dose was 10mg split between two pre-tattooed injection sites along the tibialis anterior muscle in a volume of 630μl per injection site.

rAAV vector administration

For direct rAAV IM injections, the total dose was split over one or two pre-tattooed injection sites along the tibialis anterior muscle in a maximal volume of 600μl. Mac3 and Mac4 were injected with 2.5E10 and 2.5E11vgkg−1 of the rAAV1-MD2-cmEPO vector, respectively. Mac5 and Mac6 were injected with 5E9g and 2.5E10vgkg−1 of the rAAV8-MD2-cmEPO vector, respectively.

DNA extraction from WBCs

Whole blood was collected in tubes containing ethylenediaminetetraacetic acid as an anticoagulant and stored for a maximum of 4h at room temperature or at +4°C before DNA extraction. The first step of the protocol is lysis of red blood cells, followed by a centrifugation step to recover the WBC pellet. The supernatant (containing the serum) is removed and DNA is extracted from the WBC pellet according to the manufacturer's instructions using the Gentra Puregene kit (category no. 158467) from Qiagen or using the Wizard Genomic Purification kit (category no. A1120) from Promega, which includes an RNase treatment. Concentration and purity of the gDNA were determined using a nano-spectrophotometer obtained from Implen. Integrity of the DNA was verified by migration of 3μg of total DNA on a 0.8% agarose gel, followed by ethidium bromide staining, and by real-time PCR of the endogenous macaque ε-globin gene. PCR analyses were performed using GoTaq Flexi DNA polymerase from Promega and a GeneAmp PCR System 9700 thermocycler from Applied Biosystems (Carlsbad, CA, USA). The PCR program was as follows: initial denaturation at 94°C for 5min, followed by 30 cycles at 94°C for 30s, 60°C for 30s, 72°C for 30s and final extension at 72°C for 10min. Amplified products were analyzed by electrophoresis on a 3% agarose 1000 gel (Invitrogen), followed by ethidium bromide staining.

DNA extraction from serum or urine

Total DNA was extracted from 140μl of serum or urine using the Qiamp Viral RNA minikit (category no. 52904) obtained from Qiagen. One-tenth of the extraction (8μl) was then analyzed by real-time PCR.

DNA extraction from tissues

Muscle biopsies were performed under ketamine (8mgkg−1)/metedomidine (20ìgkg−1)-induced anesthesia, and marbofloxacin and meloxicam were administered for the next 3–5 days, so as to avoid discomfort for the animal. During its anemia, Mac2 received a daily regiment of Fercobsang (0.1mlkg−1). gDNA was extracted as described earlier.7 Concentration and purity of the gDNA were determined using a nano-spectrophotometer from Implen.

In vivo cmEPO expression analysis

Serum cmEPO levels were measured by enzyme-linked immunosorbent assay using the Quantikine IVD kit (category no. DEP00) obtained from R&D Systems (Lille, France).

Anti-cmEPO response analysis

Overall, 500ng of recombinant human EPO protein (Eprex, Janssen-Cliag) was subjected to electrophoresis on a 10% polyacrylamide SDS-PAGE gel, and then transferred to a nitrocellulose membrane. The membrane was then cut into strips, and after saturation, each slice was incubated with a dilution of individual macaque serum. The presence (or absence) of an anti-EPO antibody was revealed using a secondary antibody specific from NHPs (Rhesus macaca) IgG conjugated with peroxidase. The result was obtained after chemiluminescence detection. As negative controls, the serum of each animal before rAAV injection was used. As positive controls, sera from a previous animal (Denis) that has developed autoimmune anemia was used.

Taqman PCR development

Primers and probes were designed using ABI Primer Express 3.0. However, due to constraints of the cmEPO exon–exon junction sequences and the required Tm, screening multiple primer–probes combinations was required. Conventional PCR was first applied to test the efficiency of primer pairs and to confirm the proper PCR product (specificity). The PCR was performed in a final volume of 25μl using Amplitaq Gold Master Mix (Applied Biosystems category no. 4318739), the forward and reverse primers described in Table 2 at a concentration of 400nM (1 × ), and 500ng of noninjected naive macaque gDNA was spiked with different amounts of pDNA. After one cycle at 95°C for 8min to denature the DNA and to activate the Amplitaq Gold, the three-step program of 60s at 95°C, a 30s at 56°C and 60s at 72°C was used. One cycle for 2min at 72°C for the final extension was then performed. The PCR products were analyzed using UltraPure agarose-1000 (Invitrogen, Carlsbad, CA, USA) to verify the correct size of the PCR product and the absence of other bands.

Taqman real-time PCR conditions were optimized with primers and their corresponding fluorescent probes for each of the four assays. Concentrations of the 250-nM probe and 900-nM primers were found to be optimal. Overall, 500ng of each DNA sample was amplified in a final volume of 25μl containing 1 × Taqman Universal PCR Master Mix (Applied Biosystems category no. 4304437) Amplification was performed using an ABI StepOnePlus PCR machine with an initial incubation at 50°C for 2min, a denaturation at 95°C for 10min, then 40 cycles of denaturation at 94°C for 15s and an annealing/extension step at 60°C for 1min. During thermal cycling, emission from each sample was recorded, and the ABI StepOne software v2.0 processed raw fluorescence data to produce threshold cycle (Ct) values for each sample.


Conflict of interest

ROS is an inventor on patents related to recombinant AAV technology. ROS owns equity in a gene therapy company that is commercializing AAV for gene therapy applications. To the extent that the work in this manuscript increases the value of these commercial holdings, ROS has a conflict of interest.



  1. Baoutina A, Alexander IE, Rasko JE, Emslie KR. Potential use of gene transfer in athletic performance enhancement. Mol Ther 2007; 15: 1751–1766. | Article | PubMed | ISI | ChemPort |
  2. Wells DJ. Gene doping: the hype and the reality. Br J Pharmacol 2008; 154: 623–631. | Article | PubMed | ISI | ChemPort |
  3. Verdier F, Descotes J. Preclinical safety evaluation of human gene therapy products. Toxicol Sci 1999; 47: 9–15. | PubMed | ISI | ChemPort |
  4. Gonin P, Gaillard C. Gene transfer vector biodistribution: pivotal safety studies in clinical gene therapy development. Gene Therapy 2004; 11(Suppl 1): S98–S108. | Article | PubMed | ISI | ChemPort |
  5. Provost N, Le Meur G, Weber M, Mendes-Madeira A, Podevin G, Cherel Y et al. Biodistribution of rAAV vectors following intraocular administration: evidence for the presence and persistence of vector DNA in the optic nerve and in the brain. Mol Ther 2005; 11: 275–283. | Article | PubMed | ISI | ChemPort |
  6. Lai L, Davison BB, Veazey RS, Fisher KJ, Baskin GB. A preliminary evaluation of recombinant adeno-associated virus biodistribution in rhesus monkeys after intrahepatic inoculation in utero. Hum Gene Ther 2002; 13: 2027–2039. | Article | PubMed | ChemPort |
  7. Favre D, Provost N, Blouin V, Blancho G, Cherel Y, Salvetti A et al. Immediate and long-term safety of recombinant adeno-associated virus injection into the nonhuman primate muscle. Mol Ther 2001; 4: 559–566. | Article | PubMed | ISI | ChemPort |
  8. Toromanoff A, Cherel Y, Guilbaud M, Penaud-Budloo M, Snyder RO, Haskins ME et al. Safety and efficacy of regional intravenous (r.i.) versus intramuscular (i.m.) delivery of rAAV1 and rAAV8 to nonhuman primate skeletal muscle. Mol Ther 2008; 16: 1291–1299. | Article | PubMed | ISI | ChemPort |
  9. Schenk-Braat EA, van Mierlo MM, Wagemaker G, Bangma CH, Kaptein LC. An inventory of shedding data from clinical gene therapy trials. J Gene Med 2007; 9: 910–921. | Article | PubMed | ISI |
  10. Fisher KJ, Jooss K, Alston J, Yang Y, Haecker SE, High K et al. Recombinant adeno-associated virus for muscle directed gene therapy. Nat Med 1997; 3: 306–312. | Article | PubMed | ISI | ChemPort |
  11. Xiao X, Li J, Samulski RJ. Efficient long-term gene transfer into muscle tissue of immunocompetent mice by adeno-associated virus vector. J Virol 1996; 70: 8098–8108. | PubMed | ISI | ChemPort |
  12. Snyder RO, Spratt SK, Lagarde C, Bohl D, Kaspar B, Sloan B et al. Efficient and stable adeno-associated virus-mediated transduction in the skeletal muscle of adult immunocompetent mice. Hum Gene Ther 1997; 8: 1891–1900. | Article | PubMed | ISI | ChemPort |
  13. Wu Z, Asokan A, Samulski RJ. Adeno-associated virus serotypes: vector toolkit for human gene therapy. Mol Ther 2006; 14: 316–327. | Article | PubMed | ISI | ChemPort |
  14. Flotte TR, Conlon TJ, Poirier A, Campbell-Thompson M, Byrne BJ. Preclinical characterization of a recombinant adeno-associated virus type 1-pseudotyped vector demonstrates dose-dependent injection site inflammation and dissemination of vector genomes to distant sites. Hum Gene Ther 2007; 18: 245–256. | Article | PubMed | ISI | ChemPort |
  15. Wolff JA, Williams P, Acsadi G, Jiao S, Jani A, Chong W. Conditions affecting direct gene transfer into rodent muscle in vivo. Biotechniques 1991; 11: 474–485. | PubMed | ISI | ChemPort |
  16. Jiao S, Williams P, Berg RK, Hodgeman BA, Liu L, Repetto G et al. Direct gene transfer into nonhuman primate myofibers in vivo. Hum Gene Ther 1992; 3: 21–33. | Article | PubMed | ISI | ChemPort |
  17. Vigen KK, Hegge JO, Zhang G, Mukherjee R, Braun S, Grist TM et al. Magnetic resonance imaging-monitored plasmid DNA delivery in primate limb muscle. Hum Gene Ther 2007; 18: 257–268. | Article | PubMed | ISI |
  18. Manno CS, Chew AJ, Hutchison S, Larson PJ, Herzog RW, Arruda VR et al. AAV-mediated factor IX gene transfer to skeletal muscle in patients with severe hemophilia B. Blood 2003; 101: 2963–2972. | Article | PubMed | ISI | ChemPort |
  19. Manno CS, Pierce GF, Arruda VR, Glader B, Ragni M, Rasko JJ et al. Successful transduction of liver in hemophilia by AAV-Factor IX and limitations imposed by the host immune response. Nat Med 2006; 12: 342–347. | Article | PubMed | ISI | ChemPort |
  20. Hengge UR, Dexling B, Mirmohammadsadegh A. Safety and pharmacokinetics of naked plasmid DNA in the skin: studies on dissemination and ectopic expression. J Invest Dermatol 2001; 116: 979–982. | Article | PubMed | ISI | ChemPort |
  21. Hagstrom JE, Hegge J, Zhang G, Noble M, Budker V, Lewis DL et al. A facile nonviral method for delivering genes and siRNAs to skeletal muscle of mammalian limbs. Mol Ther 2004; 10: 386–398. | Article | PubMed | ISI | ChemPort |
  22. Su LT, Gopal K, Wang Z, Yin X, Nelson A, Kozyak BW et al. Uniform scale-independent gene transfer to striated muscle after transvenular extravasation of vector. Circulation 2005; 112: 1780–1788. | Article | PubMed | ISI | ChemPort |
  23. Rivera VM, Gao GP, Grant RL, Schnell MA, Zoltick PW, Rozamus LW et al. Long-term pharmacologically regulated expression of erythropoietin in primates following AAV-mediated gene transfer. Blood 2005; 105: 1424–1430. | Article | PubMed | ISI | ChemPort |
  24. Van Vliet KM, Blouin V, Brument N, Agbandje-McKenna M, Snyder RO. The role of the adeno-associated virus capsid in gene transfer. Methods Mol Biol 2008; 437: 51–91. | PubMed | ChemPort |
  25. Gao GP, Alvira MR, Wang L, Calcedo R, Johnston J, Wilson JM. Novel adeno-associated viruses from rhesus monkeys as vectors for human gene therapy. Proc Natl Acad Sci USA 2002; 99: 11854–11859. | Article | PubMed | ChemPort |
  26. Stroes ES, Nierman MC, Meulenberg JJ, Franssen R, Twisk J, Henny CP et al. Intramuscular administration of AAV1-lipoprotein lipase S447X lowers triglycerides in lipoprotein lipase-deficient patients. Arterioscler Thromb Vasc Biol 2008; 28: 2303–2304. | Article | PubMed | ISI | ChemPort |
  27. Duan D, Sharma P, Yang J, Yue Y, Dudus L, Zhang Y et al. Circular intermediates of recombinant adeno-associated virus have defined structural characteristics responsible for long-term episomal persistence in muscle tissue. J Virol 1998; 72: 8568–8577. | PubMed | ISI | ChemPort |
  28. Schnepp BC, Clark KR, Klemanski DL, Pacak CA, Johnson PR. Genetic fate of recombinant adeno-associated virus vector genomes in muscle. J Virol 2003; 77: 3495–3504. | Article | PubMed | ISI | ChemPort |
  29. Vincent-Lacaze N, Snyder RO, Gluzman R, Bohl D, Lagarde C, Danos O. Structure of adeno-associated virus vector DNA following transduction of the skeletal muscle. J Virol 1999; 73: 1949–1955. | PubMed | ISI | ChemPort |
  30. Penaud-Budloo M, Le Guiner C, Nowrouzi A, Toromanoff A, Cherel Y, Chenuaud P et al. Adeno-associated virus vector genomes persist as episomal chromatin in primate muscle. J Virol 2008; 82: 7875–7885. | Article | PubMed | ISI | ChemPort |
  31. Espy MJ, Uhl JR, Sloan LM, Buckwalter SP, Jones MF, Vetter EA et al. Real-time PCR in clinical microbiology: applications for routine laboratory testing. Clin Microbiol Rev 2006; 19: 165–256. | Article | PubMed | ISI | ChemPort |
  32. Lock M, McGorray S, Auricchio A, Ayuso E, Beecham EJ, Blouin V et al. Characterization of a recombinant adeno-associated virus type 2 reference standard material. Hum Gene Ther 2010; 21: 1273–1285. | Article | PubMed | ISI | ChemPort |
  33. Gao G, Lebherz C, Weiner DJ, Grant R, Calcedo R, McCullough B et al. Erythropoietin gene therapy leads to autoimmune anemia in macaques. Blood 2004; 103: 3300–3302. | Article | PubMed | ISI | ChemPort |
  34. Chenuaud P, Larcher T, Rabinowitz JE, Provost N, Cherel Y, Casadevall N et al. Autoimmune anemia in macaques following erythropoietin gene therapy. Blood 2004; 103: 3303–3304. | Article | PubMed | ISI | ChemPort |
  35. Baoutina A, Coldham T, Bains GS, Emslie KR. Gene doping detection: evaluation of approach for direct detection of gene transfer using erythropoietin as a model system. Gene Therapy 2010; 17: 1022–1032. | Article | PubMed | ISI |
  36. Beiter T, Zimmermann M, Fragasso A, Hudemann J, Niess AM, Bitzer M et al. Direct and long-term detection of gene doping in conventional blood samples. Gene Therapy 2010 (in press).
  37. Lu Y, Song S. Distinct immune responses to transgene products from rAAV1 and rAAV8 vectors. Proc Natl Acad Sci USA 2009; 106: 17158–17162. | Article | PubMed |
  38. Brantly ML, Chulay JD, Wang L, Mueller C, Humphries M, Spencer LT et al. Sustained transgene expression despite T lymphocyte responses in a clinical trial of rAAV1-AAT gene therapy. Proc Natl Acad Sci USA 2009; 106: 16363–16368. | Article | PubMed |
  39. Liu F, Shollenberger LM, Conwell CC, Yuan X, Huang L. Mechanism of naked DNA clearance after intravenous injection. J Gene Med 2007; 9: 613–619. | Article | PubMed | ISI | ChemPort |
  40. Hughes TS, Langer SJ, Johnson KW, Chavez RA, Watkins LR, Milligan ED et al. Intrathecal injection of naked plasmid DNA provides long-term expression of secreted proteins. Mol Ther 2009; 17: 88–94. | Article | PubMed | ISI |
  41. Mayrhofer P, Blaesen M, Schleef M, Jechlinger W. Minicircle-DNA production by site specific recombination and protein-DNA interaction chromatography. J Gene Med 2008; 10: 1253–1269. | Article | PubMed | ISI | ChemPort |
  42. Gill DR, Pringle IA, Hyde SC. Progress and prospects: the design and production of plasmid vectors. Gene Therapy 2009; 16: 165–171. | Article | PubMed | ISI |
  43. Chenuaud P, Larcher T, Rabinowitz JE, Provost N, Joussemet B, Bujard H et al. Optimal design of a single recombinant adeno-associated virus derived from serotypes 1 and 2 to achieve more tightly regulated transgene expression from nonhuman primate muscle. Mol Ther 2004; 9: 410–418. | Article | PubMed | ISI | ChemPort |


We thank Michael L Nickerson, PhD (NCI Frederick), Mahajoub Bello-Roufai, PhD (UF CERHB), Oumeya Adjali, MD, PhD, James Baus for helpful discussions, Yan Chérel, DVM, PhD, Béatrice Joussemet, DVM, PhD, and Delphine Nivard for technical assistance, the personnel at the Boisbonne Centre (large animal facility, ONIRIS, Nantes) and the Vector Core at the University Hospital of Nantes for providing the rAAV1 and rAAV8 stocks. This project was funded by the World Anti-Doping Agency (ROS and PM), United States Anti-Doping Agency (ROS) and the Agence Francaise de Lutte contre le Dopage (PM). This work was performed under a Cooperative Agreement between INSERM, AFM, l’Etablissement Francais du Sang (EFS) and the University of Florida Center of Excellence for Regenerative Health Biotechnology.