HUMAN GENE THERAPY 22:112 (XXXXX 2011) Mary Ann Liebert, Inc. DOI: 10.1089/hum.2011.

034

Universal Real-Time PCR for the Detection and Quantication of Adeno-Associated Virus Serotype 2-Derived Inverted Terminal Repeat Sequences
1,* 1,* Christine Aurnhammer, Maren Haase, Nadine Muether,2 Martin Hausl,2 Christina Rauschhuber,2 1 2 1 1 Ingrid Huber, Hans Nitschko, Ulrich Busch, Andreas Sing, Anja Ehrhardt,2,3 and Armin Baiker1

Abstract

Viral vectors based on various naturally occurring adeno-associated virus (AAV) serotypes are among the most promising tools in human gene therapy. For the production of recombinant AAV (rAAV) vectors, researchers are focusing predominantly on cross-packaging an articial AAV genome based on serotype 2 (AAV2) into capsids derived from other serotypes. Within the packaged genome the inverted terminal repeats (ITRs) are the only cisacting viral elements required for rAAV vector generation and depict the lowest common denominator of all AAV2-derived vector genomes. Up to now, no quantitative PCR (qPCR) for the detection and quantication of AAV2 ITRs could be established because of their extensive secondary hairpin structure formation. Current qPCR-based methods are therefore targeting vector-encoded transgenes or regulatory elements. Herein we establish a molecular biological method that allows accurate and reproducible quantication of AAV2 genomes on the basis of an AAV2 ITR sequence-specic qPCR. Primers and labeled probe are located within the ITR sequence and have been designed to detect both wild-type AAV2 and AAV2-based vectors. This method is suitable for detecting single-stranded DNA derived from AAV2 vector particles and double-stranded DNA derived from vector plasmids. The limit of detection has been determined as 50 ITR sequence copies per reaction, by comparison with a plasmid standard. In conclusion, this method describes the rst qPCR system facilitating the detection and quantication of AAV2 ITR sequences. Because this method can be used universally for all AAV2 genome-based vectors, it will signicantly simplify rAAV2 vector titrations in the future.

Introduction

deno-associated viruses (AAVs) are small (25 nm) nonenveloped viruses that package a linear, singlestranded DNA (ssDNA) of approximately 4.7 kilobases (Lusby et al., 1980; Astell et al., 1983; Daya and Berns, 2008). The ssDNA genome of wild-type AAV comprises two inverted terminal repeats (ITRs) of 145 nucleotides anking the two open reading frames (ORFs) of structural (cap) and packaging (rep) genes. rep ORFs are essential for viral replication and the packaging of viral genomes, whereas the cap ORFs encode the viral capsid proteins VP1, VP2, and VP3 and the assembly-activating protein (AAP), necessary for the generation of viral particles. During capsid assembly, both plus and minus strands are encapsidated into separate virions
1 2

at equal frequency (Mayor et al., 1969; Lusby et al., 1980; Srivastava et al., 1983; Knipe and Howley, 2001; Sonntag et al., 2010). According to the current understanding of AAV replication, the ITR sequences play a critical role as origins of replication, serving as priming sequences for the host cell DNA polymerase during second-strand synthesis, as well as being essential for genome packaging (Yan et al., 2005; McAlister and Owens, 2007; Daya and Berns, 2008). The rst 125 nucleotides of the ITRs contain palindromic sequences, which fold into T-shaped hairpin structures for maximized basepairing (see Fig. 1). Folded ITRs contribute to primaseindependent self-priming during DNA replication (McAlister and Owens, 2007) and enable packaging of viral DNA. In addition, these hairpin structures at each end of the AAV

Bavarian Health and Food Safety Authority, 85764 Oberschleissheim, Germany. Department of Virology, Max von Pettenkofer Institute, 80336 Munich, Germany. 3 Institute of Microbiology and Virology, University of Witten/Herdecke, 58453 Witten, Germany. *These authors contributed equally to this work.

2 genome are also substrates for recombination that give rise to persistent circular and concatemeric DNA episomes through intramolecular and intermolecular recombination, respectively (Choi et al., 2005). After its original discovery as a contaminant of adenovirus preparations (Atchison et al., 1965), AAV has gained growing interest as a human gene therapy vector for more than two decades. This is due mainly to its nonpathogenicity in humans, its high transduction efciency, low inammatory response, and broad host cell tropism. Therefore, recombinant AAV (rAAV) gained enormous interest in experimental and clinical gene therapy applications (Yang et al., 1997; Monahan and Samulski, 2000; Knipe and Howley, 2001; Monahan et al., 2002; Carter, 2005; Goncalves, 2005; Daya and Berns, 2008; Asokan, 2010). rAAV vectors have been generated from several AAV serotypes, but so far recombinant AAV serotype 2 (rAAV2) represents the most extensively studied rAAV-based gene transfer vector. To date, more than 60 clinical trials have been conducted with rAAV2 for gene delivery (Mitchell et al., 2010). Over the years vectors based on serotypes other than rAAV2 gained increasing attention because of their different host cell tropisms. For instance, rAAV vectors derived from serotype 8 (rAAV8) stand out because of their broad tropism in mice and their efcient gene delivery into many different organs in vivo, whereas rAAV derived from serotype 9 (rAAV9) displays strong tropism toward the heart (Bish et al., 2008). Moreover, AAV serotype 7 shows a preference for transducing muscle cells, serotypes 1 and 5 prefer endothelial cells, and serotype 6 prefers airway epithelial cells (Gao et al., 2006; Wang et al., 2011). Because the host cell tropism of rAAV vectors is determined by their capsid structures, researchers are focusing on cross-packaging the well-characterized genome of rAAV2 into the capsid of the respective serotype of interest. Typically, rAAV vector genomes consist of approximately 4.5 kilobases of ssDNA (Grimm et al., 1999, 2005; Carter, 2005). Within rAAV vector genomes, the ORFs encoding the rep and cap gene products are deleted and replaced by the transgene and its regulatory elements. Therefore, the two anking AAV ITR sequences are the only genomic elements of wild-type AAV present in the rAAV vectors and despite several variations in viral vector design, the majority of rAAV genomes contain these essential cis-active sequences. Standard production procedures for rAAV vectors are based on transient triple plasmid transfection of the producer cell line, which is usually based on human embryonic kidney cells (Grieger et al., 2006). For triple transfection one plasmid carries the rAAV2 vector genome with the respective transgene expression cassette to be packaged, another plasmid contains the rep and the serotype-dependent cap coding sequence, and the third plasmid provides the adenovirus helper functions required for efcient replication of the rAAV vector. After purication of the virions nal rAAV vector preparations are precisely characterized and quantied for proper applications. Current titration methods for AAV are based on three rationales. One is the determination of the physical titers of genome-containing particles within rAAV vector preparations as assayed by quantitative PCR (qPCR) or dot-blot analysis (Grieger et al., 2006). A second rationale is based on the detection of specic capsid proteins by enzyme-linked

AURNHAMMER ET AL. immunosorbent assays (ELISAs) (Grimm et al., 1999; Wobus et al., 2000). Other procedures such as infectious center assays are aiming for the enumeration of the infectious or transducing units present within rAAV vector preparations. For determination of rAAV vector genome copies by dotblot analysis, dilution series of both isolated genomic rAAV vector DNA and linearized plasmid standard are transferred to a membrane and subjected to Southern blot hybridization. The number of rAAV vector genome copies can be calculated from the dilution series of the plasmid standard (Grieger et al., 2006). Currently available qPCR strategies for the detection and quantication of rAAV vector preparations are designed to target transgene-specic genome sequences. For instance, Rohr and colleagues (2002) demonstrated the applicability of promoter-specic primers in combination with LightCycler technology. However, a clear limitation of this strategy is the fact that the primers used for PCR amplication of a transgene or promoter-specic sequence need to be newly designed for each rAAV vector carrying a different transgene expression cassette. Therefore, a universal method allowing the detection and quantication of rAAV vector genomes independent of their respective transgene expression cassette would be highly desirable. Herein, we describe a reliable method to detect and quantify both single-stranded DNA derived from AAV2 vector particles and double-stranded DNA derived from vector plasmids by using an AAV2 ITR sequence-specic qPCR. This method can be used in a one-for-all based manner and, therefore, it will signicantly simplify rAAV2 vector titrations in the future. Materials and Methods Plasmids and vectors The transgene-containing plasmid for packaging of the rAAV2 genome was based on the plasmid pZac2.1 (provided by J.M. Wilson, University of Pennsylvania, Philadelphia, PA). rAAV vectors AAV2.8 mTRIF and AAV2.8 EGFP were produced by polyethylenimine (PEI)-mediated triple transfection of HEK293 cells. The transgene-containing plasmid was cotransfected with the helper plasmid pAdDeltaF6 providing the three adenoviral helper genes (provided by J.M. Wilson) and p5E18-VD28 (provided by J.M. Wilson) for packaging into the AAV8 capsid. AAV vectors were puried by iodixanol gradient centrifugation and stored in Dulbeccos phosphate-buffered saline. Details about the DNA sequences and transgenes contained in the rAAV vectors AAV2.8 mTRIF and AAV2.8 EGFP can be obtained on request. The AAV2 vector plasmid pZACmSB containing the AAV ITRs used to generate a standard curve for the qPCR was based on the plasmid pZac2.1. The concentration of isolated plasmid DNA was determined with Quant-iT PicoGreen DNA reagent and kits (Invitrogen, Karlsruhe, Germany) according to the manufacturers instructions. Primers, probe, and oligonucleotides The AAV2 ITR qPCR is based on the forward primer (fwd ITR primer, 5-GGAACCCCTAGTGATGGAGTT-3) and the reverse primer (rev ITR primer, 5-CGGCCTCAGTGAGCGA-3).

AAV2 ITR SEQUENCE-SPECIFIC qPCR The hydrolysis probe was labeled with uorescein (FAM) and quenched with BlackBerry quencher (BBQ) (AAV2 ITR probe, 5-FAM-CACTCCCTCTCTGCGCGCTCG-BBQ3). Primers and probe are suitable to detect both wild-type AAV2 and AAV2-derived vectors. For the analysis of AAV2 ITR qPCR cross-reactivity with the respective ITR sequence elements of other AAV serotypes, the oligonucleotides AAV1 oligo (5-CGGCCCCAC CGAGCGAGCGAGCGCGCAGAGAGGGAGTGGGCAAC TCCATCACTAGGGGTAAT-3), AAV2/6 oligo (5-CGGCC TCAGTGAGCGAGCGAGCGCGCAGAGAGGGAGTGGCC AACTCCATCACTAGGGGTTCC-3), AAV3 oligo (5-CGGC CCCACCGAGCGAGCGAGTGCGCATAGAGGGAGTGGC CAACTCCATCACTAGAGGTATG-3), AAV4 oligo (5-AG GGCCGAGTGAGTGAGCGAGCGCGCATAGAGGGAGTG GCCAACTCCATCATCTAGGTTTGC-3), AAV5 oligo (5-A CGAGCCAGCGAGCGAGCGAACGCGACAGGGGGGAG AGTGCCACACTCTCAAGCAAGGGGGT-3), AAV7 oligo (5-CGGCCCCACCGAGCGAGCGAGCGCGCATAGAGG GAGTGGCCAACTCCATCACTAGGGGTACC-3), AAV8 oligo (5-CAGCGTGGAAATTGAATGGGAGCTGCAGAAG GAAAACAGCAAGCGCTGGAACCCCGAGATCC-3), and AAV9 oligo (5-CAGCGTGGAAATCGAGTGGGAGCTG CAGAAAGAAAACAGCAAGCGCTGGAATCCAGAGAT CC-3) have been used. Primers, probe, and oligonucleotides were synthesized by TIB MOLBIOL (Berlin, Germany). Qualitative PCR analysis Qualitative PCR analysis for detection of the 62-bp AAV2 ITR amplicon out of pZACmSB was performed with a Mastercycler gradient PCR device (Eppendorf, Hamburg, Germany). PCRs were performed in a 25-ll volume comprising 2 mM MgCl2, 0.4 mM dNTP mixture (Applied Biosystems, Foster City, CA), 0.2 lM fwd ITR primer, 0.2 lM rev ITR primer, 50 ng of pZACmSB, and 1 U of Platinum Taq DNA polymerase (Invitrogen). The PCR prole contained an initial denaturation step at 95C for 10 min followed by 35 cycles of denaturation at 95C for 30 sec, annealing at 55C to 65.5C (gradient) for 30 sec, and extension at 72C for 1 min, with a nal extension after the last cycle at 72C for 10 min. The 62-bp amplication product was analyzed on a 4% agarose E-Gel (Invitrogen), using 25-bp DNA ladder (Invitrogen) as DNA size marker. Quantitative PCR analysis Quantitative PCR (qPCR) analysis was routinely performed with a Stratagene Mx3000 real-time PCR system device (Agilent Technologies, Palo Alto, CA). PCRs were performed in a 25-ll nal volume, using a QuantiTect Multiplex PCR NoROX kit (Qiagen, Hilden, Germany) supplemented with 340 nM rev ITR primer, 100 nM fwd ITR primer, 100 nM AAV2 ITR probe, and 5 ll of template DNA (either plasmid standard or extracted sample DNA) according to the manufacturers instructions. The PCR prole contained a uracil-N-glycosylase (UNG) treatment at 50C for 2 min, an initial denaturation step at 95C for 15 min, followed by 40 cycles of denaturation at 95C for 1 min and annealing/extension at 60C for 1 min. Each qPCR run was controlled by a no template negative control performed in quadruplicate. As positive controls and

3 for the calculation of standard curves, four serial dilutions of the plasmid standard pZACmSB (containing 105, 104, 103, and 102 plasmid copies per 5 ll) were prepared and subjected to qPCR analysis in triplicates. Data analysis was performed with Stratagene MxPro version 4.10 software (Agilent Technologies). For investigation of the robustness of this method, the performance of the qPCR was additionally assayed with a Brilliant III ultra-fast QPCR master mix kit (Agilent Technologies) and the TaqMan universal PCR master mix (Applied Biosystems) according to the manufacturers instructions. For further studies on the robustness of this method, qPCRs were additionally performed with a 7900 HT sequence detection system (Applied Biosystems) and a LightCycler 480 device (Roche, Mannheim, Germany). To analyze the inuence of genomic DNA on the robustness of our ITR PCR, dened numbers of pZACmSB DNA molecules were spiked with various amounts of genomic DNA extracted from HEK293 cells. Establishment and validation of this qPCR method were performed in accordance with listed requirements on appropriate experimental planning and data analysis in the MIQE (Minimum Information for Publication of Quantitative Real-Time PCR) guidelines (Bustin et al., 2009). Transgene (EGFP) qPCR A transgene-specic qPCR to detect enhanced green uorescent protein (EGFP) cDNA within AAV vector genomes and vector plasmids was developed with the LightCycler system (Applied Biosystems), using a LightCycler FastStart DNA MasterPLUS SYBR green I kit (Roche). For amplication the primers EGFP fwd (5 ACA GGG AAT TCC ACC ATG GTG AGC AAG 3) and EGFP rev (5 TTG AAG AAG ATG GTG CGC TCC 3) were used. The thermal prole contained a preamplication step at 95C for 10 min followed by 40 amplication cycles (95C for 15 sec and 60C for 60 sec). A standard curve was generated, using the corresponding AAV vector plasmid pZAC-EGFP. Extraction of viral nucleic acids Viral nucleic acids from AAV2.8 mTRIF and AAV2.8 EGFP vector preparations were extracted with the High Pure viral nucleic acid kit (Roche) according to the manufacturers instructions. Nucleic acids were eluted in a 50-ll volume. Bioinformatics analysis To investigate putative cross-reactivity, BLAST analyses for the 62-bp AAV2-specic amplicon sequence were performed against all NCBI-listed sequences (http:/ /blast .ncbi.nlm.nih.gov/Blast.cgi). For the generation of alignments between the AAV2 amplicon sequence and the respective sequences of other AAV serotypes (see Fig. 5) the HUSAR software tool was used (http:/ / genius.embnet.dkfz-heidelberg.de/menu/w2h/w2hdkfz/). Results Design of AAV2 ITR qPCR primers and probe The inverted terminal repeats (ITRs) of AAV2 consist of 145 nucleotides that form an extensive secondary hairpin

4 structure. This secondary structure formation has long impaired the design of real-time PCR-based detection methods within this region. To realize this AAV2 ITR qPCR both primers and probe were designed to target a 62nucleotide sequence within the stem region of the ITR sequence, as depicted in Fig. 1 for the 5 end of the wild-type AAV2 genome. In this case, the reverse primer (rev ITR primer) represents a 16-mer oligonucleotide directing from the ITR loop toward the inside of the AAV2 genome. In contrast, the respective forward primer (fwd ITR primer) represents a 21-mer oligonucleotide directing from the hypothetically extended 5 end of the AAV2 genome toward the ITR loop. The probe (AAV2 ITR probe) represents a 21mer oligonucleotide located on the C-rich strand of the AAV2 stem region in between the forward and reverse primers (Fig. 1). An initial qualitative PCR experiment was designed to prove the functionality of the combination of forward and reverse primers for the generation of the respective 62-bp PCR product (Fig. 1, bottom) and to determine the optimal annealing temperature. For this purpose the AAV2 vector plasmid pZACmSB was used as a template. As depicted in Fig. 2, the PCR amplied a 62-bp fragment out of pZACmSB, as expected for this primer combination. The integrity of this PCR product was veried by sequencing (data not shown). The amplication of the 62-bp product was robust at various annealing temperatures from 55C to 64.2C (Fig. 2). With respect to the further establishment of the AAV2 ITR qPCR an annealing temperature of 60C was

AURNHAMMER ET AL. chosen, because this temperature was additionally recommended by the QuantiTect Multiplex PCR NoROX kit manual (Qiagen). Performance of AAV2 ITR qPCR for quantication of AAV2 vector plasmids For the establishment of the AAV2 ITR qPCR the AAV2 vector plasmid pZACmSB was used as a template. This plasmid comprises four target sites for qPCR amplication (Fig. 3A). After the exact determination of the pZACmSB plasmid concentration, using the Quant-iT PicoGreen DNA reagent and kits (Invitrogen), a stock solution with a concentration of 1 1012 plasmids per 5 ll and 10-fold serial dilutions thereof were generated for further analyses (Fig. 3B). For determination of the linear dynamic range of the AAV2 ITR qPCR, serial dilutions of pZACmSB were assayed within 6 logs (in the range of 102 to 107 plasmid copies), with each dilution being assayed in 10 replicates. Within this range the AAV2 ITR qPCR exhibited a linear correlation coefcient (RSq) of 0.997, a slope of - 3.220, and an efciency of 104.4% (Fig. 3C and D). To specify the sensitivity of the AAV2 ITR qPCR, 10, 20, 40, 60, and 80 plasmid copies of pZACmSB were assayed in triplicates. For 60 and 80 plasmid copies all triplicates could be detected below a cycle threshold (Ct) of 40. When using 40 plasmid copies, two of three triplicates exhibited a Ct value > 40 (Fig. 3E), indicating a limit of detection (LOD) between 40 and 60 plasmid copies. For exact determination

FIG. 1. Design of AAV2 ITR qPCR primers and probe. Top: 5 ITR secondary hairpin structure of (wild-type) AAV2 and localization of the AAV2 ITR qPCR-specic primers and probe. Bottom: Corresponding 62-bp PCR product.

AAV2 ITR SEQUENCE-SPECIFIC qPCR

FIG. 2. Qualitative analysis of AAV2 ITR PCR. A qualitative PCR analysis was performed to investigate whether the combination of fwd ITR primer and rev ITR primer is able to amplify the desired 62bp amplicon fragment of pZACmSB and to determine the optimal annealing temperature. The amplication of the 62-bp product (arrow) was robust at various annealing temperatures from 55C to 64.2C. M, marker.

of the LOD, 40 and 50 plasmid copies were assayed in 20 replicates. Ninety percent of the replicates exhibited Ct values < 40 when 40 plasmid copies were used for qPCR (data not shown). However, 100% of the replicates exhibited Ct values < 40 when 50 plasmid copies were used (Fig. 3F). Hence, the LOD could be detected as approximately 50 vector plasmid copies of pZACmSB. This corresponds to approximately 200 individual (single-stranded) AAV2 ITR target sites for PCR amplication, and to approximately 100 (single-stranded) vector genomes. Notably, no linear regression of the AAV2 ITR qPCR could be observed below 102 assayed plasmid copies. To assess the intraassay variance (repeatability) of the AAV2 ITR qPCR, 103 plasmid copies were quantied within 20 replicates on one PCR plate. The standard deviation (SD) of the median of quantication cycle (Cq) values could be determined as 0.7% (data not shown). To assess the interassay variance (reproducibility) of the AAV2 ITR qPCR, 103 plasmid copies were quantied (in triplicates) on eight individual PCR plates. The coefcient of variation (CV) for the copy numbers could be determined as 14.8% (data not shown). For investigation of the robustness of the performance of the AAV2 ITR qPCR the performance of the AAV2 ITR qPCR was additionally assayed with the Brilliant III ultra-fast QPCR master mix kit (Agilent Technologies) or the TaqMan universal PCR master mix (Applied Biosystems) in combination with the Stratagene Mx3000 real-time PCR device (Agilent Technologies), respectively. Both alternative qPCR kits yielded similar amplication results (Supplementary Fig. S1; supplementary data are available online at www .liebertonline.com/hum). The AAV2 ITR qPCR showed similar results, when the 7900 HT sequence detection system (Applied Biosystems) or the LightCycler 480 device (Roche) was used in combination with the Quantitect Multiplex PCR NoROX kit (Qiagen) (Supplementary Fig. S2). In future studies our novel PCR approach could be used to determine vector genome copy numbers or wild-type AAV2 genomes in transduced cell lines or tissues. Furthermore, self-complementary AAV (scAAV) vectors with enhanced uptake into the respective target cell and an improved expression prole in vivo are of broad interest

(McCarty et al., 2003; Thomas et al., 2004). Therefore, we rst aimed to test the inuence of genomic DNA on the robustness of our AAV2 ITR qPCR protocol. In the experimental PCR setup we spiked 104 pZACmSB DNA molecules with various amounts of DNA derived from HEK293 cells, ranging from 10 to 2500 ng per reaction. As depicted in Supplementary Table S1, no signicant inuence of genomic DNA background on the robustness of the qPCR could be detected. To investigate whether our method is also suitable for quantication of wild-type AAV2 genomes and scAAV genomes we performed additional PCR analyses. This was important because scAAVs contain one mutated terminal resolution site (TRS) in the ITR partly overlapping with our PCR target sequence and in naturally occurring wild-type AAV2 genomes differ in the 5 and the 3 ITR sequences. Therefore, we performed ITR-specic PCRs with the plasmids pSUB 201 (Samulski et al., 1987), pTAV 2-0 (Heilbronn et al., 1990), and pAV2 (Laughlin et al., 1983), into which wild-type AAV genomes were inserted, and the plasmid dSAAV CMV EGFP (Wang et al., 2003), which represents a plasmid for production of scAAV. As described in Supplementary Table S2 our novel PCR strategy for quantication of AAV genomes worked for all AAV DNA templates tested. Performance of AAV2 ITR qPCR in quantication of AAV2 vector genome copies To investigate the performance of the AAV2 ITR qPCR for the quantication of linear single-stranded AAV2 vector genome copies within AAV2 vector preparations, two different samples of AAV2.8 mTRIF and AAV2.8 EGFP vector preparations were analyzed. Both AAV2 vector preparations had been preanalyzed by dot-blot quantication as described (Rohr et al., 2002; Grieger et al., 2006), with genome titers of 5.6 109/ll and 6.7 109/ll, respectively. In contrast to AAV2 vector plasmids, linear AAV2 vector genomes comprise two target sites for qPCR amplication (Fig. 4A). For preanalytics, each AAV2 vector sample was 10-fold serially diluted and viral nucleic acids were extracted in triplicates from 5 ll of each dilution, using the High Pure viral nucleic acid kit (Roche), with a nal elution volume of 50 ll. For

AURNHAMMER ET AL.

FIG. 3. Performance of AAV2 ITR qPCR within quantication of AAV2 vector plasmids. (A) Schematic drawing of pZACmSB. This AAV2 vector plasmid comprises four target sites for qPCR amplication. (B) Pipetting scheme of the serial 10-fold dilutions of pZACmSB for determination of the linear dynamic range. (C) Amplication plots for 10-fold dilution series of pZACmSB. (D) Respective standard curve: within the range of 102 to 107 plasmid copies the AAV2 ITR qPCR exhibited an RSq of 0.997, a slope of - 3.220, and an efciency of 104.4%; n = 20. (E) Approaching the sensitivity of the AAV2 ITR qPCR by using 10, 20, 40, 60, and 80 plasmid copies per qPCR; n = 3. (F) Determination of the LOD at 50 plasmid copies of pZACmSB per reaction; n = 20. Cq, quantication cycle; Eff, efciency; LOD, limit of detection; RSq, linear correlation coefcient.

qPCR quantication 5-ll volumes of the eluted AAV2 nucleic acids were assayed in duplicates (Fig. 4B), using dened serial dilutions of pZACmSB for calculation of a standard curve. The AAV2 ITR qPCR exhibited a linear correlation coefcient (RSq) of 0.999, a slope of - 3.405, and an efciency

of 96.6% for the 10-fold serial dilutions of the AAV2.8 mTRIF vector sample (Fig. 4C and D). In the case of AAV2.8 EGFP vector sample dilution, a linear correlation coefcient (RSq) of 0.969, a slope of - 3.056, and an efciency of 112.9% could be observed (Fig. 4E and F). The calculation of AAV2.8

AAV2 ITR SEQUENCE-SPECIFIC qPCR

FIG. 4. Performance of AAV2 ITR qPCR in quantication of AAV2 vector genome copies. (A) Schematic drawing of AAV2 vector genomes. Linear and single-stranded vector genomes comprise 2 target sites for qPCR amplication. (B) Pipetting scheme for the AAV2 ITR qPCR preanalytics: each AAV2 vector sample (AAV2.8 mTRIF or AAV2.8 EGFP) was 10-fold serially diluted and viral nucleic acids were extracted in triplicate from 5ll of each dilution followed by elution in a 50ll volume. For qPCR quantication 5ll of the eluted AAV2 nucleic acids were assayed in duplicate. (C) Amplication plots for 10-fold dilution series of AAV2.8 mTRIF. (D) In the case of AAV2.8 mTRIF the AAV2 ITR qPCR exhibited a RSq of 0.999, a slope of - 3.405 and an efciency of 96.6%. (E) Amplication plots of a 10-fold dilution series of AAV2.8 EGFP. (F) In the case of AAV2.8 EGFP, a RSq of 0.969, a slope of - 3.056 and an efciency of 112.9% could be observed. Cq, quantication cycle; Eff, efciency; RSq, linear correlation coefcient.

mTRIF and AAV2.8 EGFP genome titers within the various AAV vector dilutions is depicted in Table 1. It is worth mentioning that for the quantication of single-stranded viral genome copies from the pZACmSB standard a correction factor of two has been introduced. In the case of AAV2.8 mTRIF a genome titer of 2.9 1010/ll, and in case of AAV2.8 EGFP a genome titer of 2.08 1010/ll (calculated as the

median genome titer of all assayed dilutions), could be detected. A separate experiment was designed to compare genomic titers obtained by the new AAV2 ITR qPCR method with titers obtained by currently available qPCR strategies targeting the transgene expression cassette. For this purpose the AAV2.8 EGFP vector preparation was additionally

AURNHAMMER ET AL. Table 1. Calculation of Adeno-Associated Virus Serotype 2 (AAV2) Genome Titers by AAV2 Inverted Terminal Repeat Quantitative PCR

Volume (ll) of sample used for extraction 5 100 5 10 - 1 5 10 - 2 5 10 - 3 5 10 - 4 5 10 - 5 5 10 - 6 5 10 - 7 5 10 - 8 5 10 - 9

Relative volume (ll) of sample used for qPCR 5 10 - 1 5 10 - 2 5 10 - 3 5 10 - 4 5 10 - 5 5 10 - 6 5 10 - 7 5 10 - 8 5 10 - 9 5 10 - 10 AAV2.8 mTRIF

Copy number of AAV genomes per qPCR 3.30 109 3.65 108 5.20 107 6.18 106 7.55 105 6.95 104 1.26 104 1.45 103 1.52 102 1.92 101

Calculated AAV genome titer (/ll) 1.32 1010 1.46 1010 2.08 1010 2.47 1010 3.02 1010 2.78 1010 5.04 1010 5.78 1010 6.08 1010 7.68 1010 Median: 2.9 1010 4.51 1010 2.28 1010 2.14 1010 1.47 1010 1.25 1010 8.89 109 1.36 1010 2.02 1010 1.53 1011 7.59 1011 Median: 2.08 1010

5 100 5 10 - 1 5 10 - 2 5 10 - 3 5 10 - 4 5 10 - 5 5 10 - 6 5 10 - 7 5 10 - 8 5 10 - 9

5 10 - 1 5 10 - 2 5 10 - 3 5 10 - 4 5 10 - 5 5 10 - 6 5 10 - 7 5 10 - 8 5 10 - 9 5 10 - 10

AAV2.8 EGFP

1.13 109 5.70 108 5.36 107 3.67 106 3.12 105 2.22 104 3.40 103 5.05 102 3.82 102 1.90 102

Depicted here is the calculation of the median AAV2.8 mTRIF and AAV2.8 EGFP genome titers within various analyzed AAV vector dilutions. In the case of AAV2.8 mTRIF a median genome titer of 2.9 1010/ll, and in the case of AAV2.8 EGFP a median genome titer of 2.08 1010/ll, could be detected. For the quantication of single-stranded viral genome copies from the pZACmSB standard a correction factor of two has been introduced. AAV2.8, articial AAV genome based on serotype 2 (AAV2) pseudotyped with AAV serotype 8 capsid; EGFP, enhanced green uorescent protein; ITR, inverted terminal repeat; qPCR, quantitative PCR.

quantied according to a customized Transgene (EGFP) qPCR protocol. Both methods quantied within the same order of magnitude, with the EGFP qPCR exhibiting approximately 3.6-fold higher titers as compared with the AAV2 ITR qPCR (data not shown).

Investigation of putative AAV2 ITR qPCR cross-reactivity with other AAV serotypes To investigate putative cross-reactivity, BLAST searches for the 62-bp AAV2-specic amplicon sequence were carried out against all NCBI-listed sequences (http:/ /blast.ncbi .nlm.nih.gov/Blast.cgi). Highly similar sequences could be identied only within deposited sequences of wild-type AAV2 (e.g., NC_001401; Max Score, 115), AAV2-derived vectors (e.g., EU22316.1; Max Score, 113) or the closely related AAV serotypes AAV6 (e.g., AF028704.1; Max Score, 113), AAV7 (e.g., NC_006260; Max Score, 89.7), AAV1 (e.g., NC_002077; Max Score, 87.9), AAV3 (e.g., NC_001729; Max Score, 80.6), and AAV4 (e.g., NC_001829; Max Score, 73.4). The sequences of AAV5 (e.g., NC_006152), AAV8 (e.g., NC_006261), and AAV9 (e.g., AX753250.1) do not contain highly similar sequences.

For assessing the cross-reactivity of the AAV2 ITR qPCR with other AAV serotypes, alignments between the AAV2 amplicon sequence and the respective sequences of other known AAV serotypes were generated with the HUSAR software tool (http:/ /genius.embnet.dkfz-heidelberg.de/menu /w2h/w2hdkfz/). The result of the respective alignments, including the number of mismatches between the individual sequences, is depicted in Fig. 5A. To practically study any potential cross-reactivity, all ortholog amplicon sequences were synthesized as 62-mer oligonucleotides and subjected to AAV2 ITR qPCR analysis, using 5 104 molecules per qPCR. Because the amplicon sequence of AAV6 is completely identical to that of AAV2, cross-reactivity was anticipated and not investigated. The results of this study revealed that the AAV2 ITR qPCR cross-reacts with the ortholog (amplicon) sequences of AAV1, AAV3, and AAV7 to low extents but not with AAV4, AAV5, AAV8, or AAV9 (Fig. 5B).

Discussion Current methods for the quantication of rAAV2 vector genome copies are based on either dot-blot analyses

AAV2 ITR SEQUENCE-SPECIFIC qPCR

FIG. 5. Cross-reactivity of AAV2 ITR qPCR with other AAV serotypes. (A) Alignments of the AAV2 amplicon sequence with respective sequences of other known AAV serotypes. The number of mismatches between the individual sequences and the 62-bp AAV2 amplicon is depicted in the right-hand column. The localization of the RBE and TRS within the amplicon sequences is indicated below. (B) Amplication plot of cross-reactivity analyses. All ortholog amplicon sequences were synthesized as 62-mer oligonucleotides and subjected to AAV2 ITR qPCR analysis, using 5 104 molecules per qPCR. The AAV2 ITR qPCR cross-reacts with the ortholog (amplicon) sequences of AAV1, AAV3 (AAV6), and AAV7, but not with AAV4, AAV5, AAV8, or AAV9. RBE, Rep-binding element; TRS, terminal resolution site.

(Grieger et al., 2006) or qPCR-based approaches that target transgene-specic genome sequences (Rohr et al., 2002). The quantication of vector genomes by dot-blot analyses is a rather labor-intensive and time-consuming process. In contrast, current qPCR-based methods are limited by the fact that primers and probes used for real-time PCR amplication of transgene or promoter sequence need to be newly designed for each rAAV2 vector carrying a different transgene expression cassette. Therefore, a universal qPCRbased method enabling the detection and quantication of rAAV2 vector genomes independent of their respective transgene expression cassette would be highly desirable. Such a method should target the viral ITRs as the lowest common denominator of all AAV2-derived vector genomes. However, up to now no qPCR-based method for the detection and quantication of AAV2 ITRs could be established because of their extensive secondary hairpin structure formation. Herein, we describe the rst qPCR-based method for the detection and quantication of AAV2 ITR sequences, targeting a 62-nucleotide sequence within the ITR stem region (Fig. 1). We could demonstrate that our AAV2 ITR qPCR qualies for the detection and quantication of both single-

stranded DNA derived from AAV2 vector particles and double-stranded DNA derived from vector plasmids. Our AAV2 ITR qPCR exhibits a linear dynamic range within 102 to 107 assayed pZACmSB plasmid copies and a limit of detection of approximately 50 copies. The intraassay variance (repeatability), as expressed by the standard deviation of the median of quantication cycle values, could be determined as 0.7% and the interassay variance (reproducibility), calculated as the coefcient of variation for the copy numbers, could be dened as 14.8%. We could use our method for the quantication of vector genomes within two different AAV2 vector preparations. Respective AAV genome titers were calculated as the median titer of all 10 assayed vector dilutions (Fig. 4 and Table 1). In both cases, the qPCR-based quantication exhibited slightly higher vector genome titers (2.9 1010/ll for AAV2.8 mTRIF and 2.08 1010/ll for AAV2.8 EGFP) as compared with the respective titers determined by dot-blot analyses (5.6 109/ll for AAV2.8 mTRIF and 6.7 109/ll for AAV2.8 EGFP). Nevertheless, both qPCR-based titers appeared within the same order of magnitude. It should be mentioned that the assessment of 10 different AAV vector dilutions (each extracted in triplicates and each subjected to

10 qPCR and duplicates) was performed only to determine the linear correlation coefcient, slope, and efciency of the AAV2 ITR qPCR when using extracted vector genomes as a template. As a practical method for the fast and easy quantication of vector genomes within AAV vector preparations we recommend the use of 5 100 and 5 103 ll of the original sample for a single extraction followed by qPCR analysis in duplicates. Future studies should address whether this method is suitable in the context of genomic DNA. For instance, when determining the infectious units of rAAV vectors a complex infectious center assay can be omitted by simply quantifying the transduced genomes, using our novel qPCR shortly after infection. In a separate study to assess the cross-reactivity of this AAV2 ITR qPCR with other AAV serotypes, cross-reactivity could be shown for AAV1, AAV3, AAV6, and AAV7, but not for AAV4, AAV5, AAV8, or AAV9. In summary, we have developed the rst qPCR system enabling the detection and quantication of AAV2 ITR sequences. Because this method can be used for all AAV2 genome-based vectors in a one-for-all based manner, it will signicantly simplify rAAV2 vector titrations in the future. Acknowledgments Financial support by the Deutsche Forschungsgemeinschaft (DFG SPP1230 priority program Mechanisms of Gene Vector Entry and Persistence) and the Bayerisches r Umwelt und Gesundheit is gratefully Staatsministerium fu acknowledged. Author Disclosure Statement No competing nancial interests exist. References
Asokan, A. (2010). Reengineered AAV vectors: Old dog, new tricks. Discov. Med. 9, 399403. Astell, C.R., Thomson, M., Merchlinsky, M., and Ward, D.C. (1983). The complete DNA sequence of minute virus of mice, an autonomous parvovirus. Nucleic Acids Res. 11, 9991018. Atchison, R.W., Casto, B.C., and Hammon, W.M. (1965). Adenovirus-associated defective virus particles. Science 149, 754756. Bish, L.T., Morine, K., Sleeper, M.M., et al. (2008). Adenoassociated virus (AAV) serotype 9 provides global cardiac gene transfer superior to AAV1, AAV6, AAV7, and AAV8 in the mouse and rat. Hum. Gene Ther. 19, 13591368. Bustin, S.A., Benes, V., Garson, J.A., et al. (2009). The MIQE guidelines: Minimum information for publication of quantitative real-time PCR experiments. Clin. Chem. 55, 611622. Carter, B.J. (2005). Adeno-associated virus vectors in clinical trials. Hum. Gene Ther. 16, 541550. Choi, V.W., Samulski, R.J., and McCarty, D.M. (2005). Effects of adeno-associated virus DNA hairpin structure on recombination. J. Virol. 79, 68016807. Daya, S., and Berns, K.I. (2008). Gene therapy using adenoassociated virus vectors. Clin. Microbiol. Rev. 21, 583593. Gao, G., Lu, Y., Calcedo, R., et al. (2006). Biology of AAV serotype vectors in liver-directed gene transfer to nonhuman primates. Mol. Ther. 13, 7787.

AURNHAMMER ET AL.
Goncalves, M.A. (2005). Adeno-associated virus: From defective virus to effective vector. Virol. J. 2, 43. Grieger, J.C., Choi, V.W., and Samulski, R.J. (2006). Production and characterization of adeno-associated viral vectors. Nat. Protocols 1, 14121428. Grimm, D., Kern, A., Pawlita, M., et al. (1999). Titration of AAV2 particles via a novel capsid ELISA: Packaging of genomes can limit production of recombinant AAV-2. Gene Ther. 6, 13221330. Grimm, D., Pandey, K., and Kay, M.A. (2005). Adeno-associated virus vectors for short hairpin RNA expression. Methods Enzymol. 392, 381405. Heilbronn, R., Burkle, A., Stephan, S., and Zur Hausen, H. (1990). The adeno-associated virus rep gene suppresses herpes simplex virus-induced 1DNA amplication. J. Virol. 64, 3012 3018. Knipe, D.M., and Howley, P.M. (2001). Fundamental Virology. (Lippincott Williams & Wilkins, Philadelphia, PA). Laughlin, C.A., Tratschin, J.D., Coon, H., and Carter, B.J. (1983). Cloning of infectious adeno-associated virus genomes in bacterial plasmids. Gene 23, 6573. Lusby, E., Fife, K.H., and Berns, K.I. (1980). Nucleotide sequence of the inverted terminal repetition in adeno-associated virus DNA. J. Virol. 34, 402409. Mayor, H.D., Torikai, K., Melnick, J.L., and Mandel, M. (1969). Plus and minus single-stranded DNA separately encapsidated in adeno-associated satellite virions. Science 166, 12801282. McAlister, V.J., and Owens, R.A. (2007). Preferential integration of adeno-associated virus type 2 into a polypyrimidine/ polypurine-rich region within AAVS1. J. Virol. 81, 9718 9726. McCarty, D.M., Fu, H., Monahan, P.E., et al. (2003). Adenoassociated virus terminal repeat (TR) mutant generates selfcomplementary vectors to overcome the rate-limiting step to transduction in vivo. Gene Ther. 10, 21122118. Mitchell, A.M., Nicolson, S.C., Warischalk, J.K., and Samulski, R.J. (2010). AAVs anatomy: Roadmap for optimizing vectors for translational success. Curr. Gene Ther. 10, 319340. Monahan, P.E., and Samulski, R.J. (2000). Adeno-associated virus vectors for gene therapy: More pros than cons? Mol. Med. Today 6, 433440. Monahan, P.E., Jooss, K., and Sands, M.S. (2002). Safety of adeno-associated virus gene therapy vectors: A current evaluation. Expert Opin. Drug Safety 1, 7991. Rohr, U.P., Wulf, M.A., Stahn, S., et al. (2002). Fast and reliable titration of recombinant adeno-associated virus type-2 using quantitative real-time PCR. J. Virol. Methods 106, 8188. Samulski, R.J., Chang, L.S., and Shenk, T. (1987). A recombinant plasmid from which an infectious adeno-associated virus genome can be excised in vitro and its use to study viral replication. J. Virol. 61, 30963101. Sonntag, F., Schmidt, K., and Kleinschmidt, J.A. (2010). A viral assembly factor promotes AAV2 capsid formation in the nucleolus. Proc. Natl. Acad. Sci. U.S.A. 107, 10220 10225. Srivastava, A., Lusby, E.W., and Berns, K.I. (1983). Nucleotide sequence and organization of the adeno-associated virus 2 genome. J. Virol. 45, 555564. Thomas, C.E., Storm, T.A., Huang, Z., and Kay, M.A. (2004). Rapid uncoating of vector genomes is the key to efcient liver transduction with pseudotyped adeno-associated virus vectors. J. Virol. 78, 31103122.

AAV2 ITR SEQUENCE-SPECIFIC qPCR

Wang, J., Faust, S.M., and Rabinowitz, J.E. (2011). The next step in gene delivery: Molecular engineering of adeno-associated virus serotypes. J. Mol. Cell. Cardiol. 50, 793802. Wang, Z., Ma, H.I., Li, J., et al. (2003). Rapid and highly efcient transduction by double-stranded adeno-associated virus vectors in vitro and in vivo. Gene Ther. 10, 21052111. Wobus, C.E., Hugle-Dorr, B., Girod, A., et al. (2000). Monoclonal antibodies against the adeno-associated virus type 2 (AAV-2) capsid: Epitope mapping and identication of capsid domains involved in AAV-2cell interaction and neutralization of AAV-2 infection. J. Virol. 74, 92819293. Yan, Z., Zak, R., Zhang, Y., and Engelhardt, J.F. (2005). Inverted terminal repeat sequences are important for intermolecular recombination and circularization of adeno-associated virus genomes. J. Virol. 79, 364379. Yang, C.C., Xiao, X., Zhu, X., et al. (1997). Cellular recombination pathways and viral terminal repeat hairpin structures are sufcient for adeno-associated virus integration in vivo and in vitro. J. Virol. 71, 92319247.

11 Address correspondence to: Dr. Armin Baiker Bavarian Health and Food Safety Authority 85764 Oberschleissheim Germany E-mail: armin.baiker@lgl.bayern.de or Dr. Anja Ehrhardt Max von Pettenkofer Institute Department of Virology 80336 Munich Germany E-mail: ehrhardt@mvp.uni-muenchen.de Received for publication March 3, 2011; accepted after revision June 3, 2011. Published online: June 6, 2011.