Polymers for Improving the In Vivo Transduction Efficiency of AAV2 Vectors

Background Adeno-associated virus has attracted great attention as vehicle for body-wide gene delivery. However, for the successful treatment of a disease such as Duchenne muscular dystrophy infusion of very large amounts of vectors is required. This not only raises questions about the technical feasibility of the large scale production but also about the overall safety of the approach. One way to overcome these problems would be to find strategies able to increase the in vivo efficiency. Methodology Here, we investigated whether polymers can act as adjuvants to increase the in vivo efficiency of AAV2. Our strategy consisted in the pre-injection of polymers before intravenous administration of mice with AAV2 encoding a murine secreted alkaline phosphatase (mSeAP). The transgene expression, vector biodistribution and tissue transduction were studied by quantification of the mSeAP protein and real time PCR. The injection of polyinosinic acid and polylysine resulted in an increase of plasmatic mSeAP of 2- and 12-fold, respectively. Interestingly, polyinosinic acid pre-injection significantly reduced the neutralizing antibody titer raised against AAV2. Conclusions Our results show that the pre-injection of polymers can improve the overall transduction efficiency of systemically administered AAV2 and reduce the humoral response against the capsid proteins.


Introduction
Adeno-associated virus mediated gene transfer has a great potential due to its capacity to transduce a wide spectrum of both dividing and non-dividing cell types. For example, widespread transduction of skeletal muscles in adult mice and hamster after a single intravenous administration of recombinant adeno-associated virus has been described [1,2]. These results open new hopes for the treatment by gene therapy of neuromuscular diseases such as Duchenne muscular dystrophy (DMD). However, the broad tropism of the AAV vectors presents also a drawback since it precludes specific in vivo targeting. Another consequence when considering a gene therapy approach for a disorder such as DMD is that because a majority of the diseased cells have to be transduced in order to have a therapeutic benefit, whole-body AAV transduction of muscle requires the injection of very large amounts of vectors. For human application, this raises not only safety concerns but it also represents a hurdle at the level of large scale production of rAAVs. One way to overcome both problems would be to find strategies able to increase the transduction efficiency in vivo, without increasing the safety risks. Attempts to increase the transduction efficiency or to modify the tropism have been done by using genetic capsid modification approaches. These include for example site-directed mutagenesis of surface-exposed tyrosine residues [3][4][5] and incorporation of targeting peptides selected by phage display on the surface of AAV capsids [6,7].
Alternatively, to engineer gene vectors that target a given tissue, random libraries of adeno-associated virus (AAV) can be generated by shuffling the capsid genes [8].
A few non-genetic strategies for enhancing the efficiency of recombinant AAV have also been developed. Walters and coworkers for example have shown that incorporation of AAV2 in a calcium phosphate co-precipitate improves gene transfer to airway epithelia after intranasal instillation [9]. More recently, it was shown that addition of cationic lipids to AAV2/9 increased the transduction efficiency of the lung [10]. Covalent modification of the capsid by either polyethyleneglycol chains [11] or by HPMA copolymers have also been evaluated [12].
In the present work, we investigated two original non-genetic strategies to enhance the in vivo efficiency of AAV. For our study, we focused on AAV2 because among the currently used AAV serotypes, it is the most studied and best characterized serotype, as well as the most used in clinical trials. The strategies that we have studied here consisted in the pre-injection of cationic or anionic polymers a few minutes before intravenous infusion of the AAV2 viral particles. The anionic polymer which we focused on was polyinosinic acid (pI). This compound was chosen since it has been previously reported to bind scavenger receptors and because it has been shown that an injection of pI prior to adenovirus administration results in enhanced transgene expression [13]. It was therefore of interest to test whether the pre-injection of pI could prevent infection by AAV2 of Kupffer cells or other antigen presenting cells, and thus, not only enhance the transduction of other tissues but also reduce the humoral immune response against AAV2 capsid proteins.
We have also evaluated the effect of pre-or co-injection of a series of cationic polymers, namely polylysines (pLys) with different degrees of polymerization (dp). The rationale was that AAV2 uses cell membrane-associated heparan sulfate proteoglycan (HSPG) as its primary binding receptor [14,15]. This binding is driven by electrostatic interaction between negative HSPG and five positively charged amino acids [16]. Notably, the insertion of negatively charged peptides in this HSPG binding motif was shown to ablate HSPG binding, whereas positively charged peptides could restore the interactions [17]. Since polylysine is able to bind through electrostatic interactions to HSPG [18], we hypothesized that this polymer could alter the AAV biodistribution and transduction efficiency.
The experiments were conducted using a murine secreted alkaline phosphatase (mSeAP), whose production can be quantified in blood samples as well as on tissue sections. Vector biodistribution was further evaluated in different tissues by real time PCR of the recombinant viral genomes. Our results show that the pre-injection of polylysine resulted in a 5 to 12-fold increase of plasmatic mSeAP levels while the injection of polyinosinic acid resulted in a 2-fold increase of mSeAP expression. Interestingly, the polyinosinic acid pre-injection also resulted in a significant reduction of the neutralizing antibody titer raised against the AAV2 capsid.

Ethics Statement
Care and manipulation of mice were performed in accordance with national and European legislations on animal experimentation. The experiments were approved by the veterinary office of Essonne, France (Agreement number 91-89).

Materials
Polyinosinic acid potassium salt, Poly-L-arginine hydrochloride (pArg) and Poly-L-lysine hydrobromide (pLys) of different molecular weights were obtained from Sigma (France). Polymer solutions for pre-injection were prepared extemporaneously by solubilization of the desired amount in 5% glucose.
Recombinant AAV2 vector production AAV2 vectors were produced in an adenovirus-free system by triple transfection using polyethylenimine (PEI 25 kDa; Sigma-Aldrich, France) as described previously [19]. Briefly, human embryonic kidney 293 cells were transfected in 15 cm plates with the trans-complementing adenovirus helper plasmid pXX6 [20], the pRepCap4 packaging plasmid expressing AAV2 rep and cap genes, and a cis acting AAV vector plasmid (pXL3937-CMV-mSeAP, pXL3937-MCK-mSeAP, pGG2-CMV-Luc or pGG2-CMV-LacZ). The AAV vector plasmids pXL3937-CMV-mSeAP, pGG2-CMV-Luc and pGG2-CMV-LacZ contain, respectively, the cDNA of a murine secreted alkaline phosphatase (mSeAP) [21], the luciferase gene and the cytosolic LacZ bacterial gene subcloned into an AAV plasmid backbone which has the ITRs of AAV2. The resulting AAV vector plasmids are under the transcriptional control of the cytomegalovirus immediate-early (CMV IE) promoter and the SV40 polyA sequence. The CMV promoter was replaced with a functional truncated muscle creatine kinase (MCK) promoter (CK6; [22]) in pXL3937-CMV-mSeAP to obtain pXL3937-MCK-mSeAP. After 72 h of transfection the cells underwent 4 cycles of freeze/thaw, the crude lysate was then treated with 25 U/ml benzonase, and AAV vectors were precipitated with cold saturated ammonium sulfate before purification by double CsCl 2 ultracentrifugation gradient. This step was followed by extensive dialysis against sterile phosphatebuffered saline containing Ca 2+ and Mg 2+ . The concentration of encapsidated viral genomes was determined by real-time quantitative PCR against a standard plasmid range and titers were expressed as viral genomes per ml (vg/ml). Titers were 7.3610 11 vg/ml for AAV2-CMV-mSeAP, 3610 11 vg/ml for AAV2-MCK-mSeAP, 2.4610 11 vg/ml for AAV2-Luc and 4.5610 11 vg/ml for AAV2-LacZ.

In vivo experiments
Eight to ten week-old female Balb/C mice (Charles River, France) were injected in the tail vein with 200 ml of polymer solution 5 minutes before a second tail vein injection of the vector solution containing 1610 11 vg (unless otherwise stated) of the AAV in a final volume of 450 ml of sterile phosphate-buffered saline supplemented with Ca 2+ and Mg 2+ . Co-injection consisted in a single tail vein injection of 650 ml of the mix of the polymer and AAV solutions. Retro-orbital plexus blood samples were obtained from anaesthetized mice the day before the injection and thereafter every week using heparinized capillary tubes. Plasma was obtained by centrifugation at 4000 g for 15 min and either analysed immediately or stored at 220uC. Mice were sacrificed by cervical dislocation. Several tissues were collected and frozen in liquid nitrogen-cooled isopentane.

Histological analysis
In situ stainings were performed on 8 mm muscle transversal cryosections. The mSeAP and b-galactosidase detections on muscle sections were performed on slices fixed with 0.5% glutaraldehyde. For mSeAP detection, fixed slices were washed twice with PBS and endogenous alkaline phosphatase was heatinactivated for 30 min at 65uC before overnight incubation at 37uC in 0.165 mg/ml 5-bromo-4-chloro-3-indolylphosphate (BCIP) and 0.33 mg/ml of nitroblue tetrazolium in 100 mM Tris-HCl, 100 mM NaCl and 50 mM MgCl 2 . For b-galactosidase detection fixed slices were washed twice with PBS containing 1 mM MgCl 2 and incubated in 1 mg X-Gal/ml (5-bromo-4chloro-3-indolyl-b-D-galactopyranoside), 5 mM K 3 Fe(CN) 6 , 5 mM K 4 Fe(CN) 6 , and 1 mM MgCl 2 in PBS for 5 hours. mSeAP or b-galactosidase revealed sections were finally counterstained with nuclear fast red, mounted and analysed by light microscopy. Non-overlapping tiled images of complete transverse muscle sections were obtained using Cartograph software (Microvision, France) and a Nikon ECLIPSE E600 microscope with a Sony 3CCD DSP camera system using a 46 magnification objective.
Blood and tissue analysis mSeAP levels in plasma and in tissue extracts were quantified by chemiluminescent detection of the enzyme activity. Briefly, endogenous alkaline phosphatase was heat-inactivated for 5 or 30 min at 65uC (for plasma and tissue, respectively) and the heatresistant mSeAP was quantified by addition of the reaction buffer and CSPD chemiluminescent substrate, according to the manufacturer's instructions (Phosphalight kit TROPIX, Applera). Chemiluminescence was measured in 96-well plate format with a luminometer (Perkin Elmer, Victor 2 1420 Multilabel counter). Expression levels were determined using a standard curve of purified human placental alkaline phosphatase and expressed as ng of mSeAP per ml of plasma. Tissues were quantified for protein content by fluorimetry using NanoOrange kit (Molecular Probes) to normalize mSeAP expression, and results were expressed as ng of mSeAP per mg of total protein in the extract.

DNA isolation and real-time PCR
DNA isolation from tissues was performed using the Wizard Genomic DNA Purification Kit (Promega, France) in accordance with the manufacturer's protocol. Total DNA concentration was determined using a Nanodrop ND-8000 spectrophotometer (Nanodrop Technologies, France), and 70 ng of DNA of each sample was used as the template material for real-time-PCR. Taqman real-time PCR was performed on each sample for both the CMV promoter in order to determine copies of the viral genome, and the mouse titin gene, to standardize for number of mouse genomes present in each sample. Primers and probe used for CMV amplification were: 59-CATCAATGGGCGTGGA-TAGC-39 (forward), 59-GGAGTTGTTACGACATTTTGGA-AA-39 (reverse) and 59-ATTTCCAAGTCTCCACCC-39 (probe). Primers and probe used for titin were: 59-AAAACGAGCAGT-GACGTGAGC-39 (forward), 59-TTCAGTCATGCTGCTAG-CGC-39 (reverse), and 59-TGCACGGAAGCGTCTCGTCT-CAGTC-39 (probe). The PCR amplifications were performed using 70 ng of DNA diluted in Absolute QPCR ROX Mix (Thermo Fischer scientific, France), 0.1 mM of Taqman probes, and 0.2 mM primers (forward and reverse) in a final volume of 18 ml. Cycling conditions consisted of a Thermo-Start DNA Polymerase activation step at 95uC for 15 min followed by 40 cycles of two steps, 15 s of denaturation at 95uC and 60 s of annealing and extension at 60uC. The PCR was performed on a 7900 HT thermocycler (Applied Biosystem, France). A standard dilution range of a plasmid containing CMV and titin sequences was used on each real-time-PCR plate as copy number control. All samples and controls were run in duplicate. Data are expressed as the number of viral genome copies per diploid genome (copies/nucleus).

Neutralization assay
On day 1, 48-well plates were seeded with 5610 4 Hela cells/well and incubated for 24 h. On day 2, recombinant AAV2-CMV-Luc was diluted in DMEM (Invitrogen Life Technology, Auckland, CA, USA) supplemented with 10% fetal calf serum and incubated with a 10-fold dilution, then 2-fold serial dilutions (1:20 to 1:12800) of heat inactivated (at 56uC for 30 min) plasma samples for 1 h at 37uC. Subsequently, the plasma-AAV2-CMV-Luc vector mixtures corresponding to 5610 3 viral genome/cell, were added to cells plated on day 1 and incubated in DMEM +10% FCS for 48 h at 37uC and 5% CO 2 . Each mix was performed in duplicate. Cells were then washed in PBS and lysed for 10 minutes in 0.2% Triton Lysis Buffer at 4uC. The lysate was transferred to 96-well plates then the luciferase activity was read on a luminometer (VICTOR 2 1420 multilabel counter, Perkin Elmer/Life Sciences). Transduction efficiency was measured as relative light units (RLU), per second per well and normalized per amount of protein per well expressed as optical density (RLU/sec/w/OD). The percentage of transduction inhibition is then calculated for each plasma dilution relatively to the maximal transduction efficiency which is determined by a control AAV infection in presence of PBS. We added sera from 5 mice before injection as negative control and we checked the absence of pI interference in the neutralizing assay using sera from mice injected with pI alone at 200 mg/mouse. The neutralizing antibody factors (NAF) were defined as the reciprocal plasma dilution inhibiting maximal transduction by 50%.

Statistical analysis
All data are expressed as means +/2 standard deviation (SD). Differences between two groups were tested by using the unpaired T test with Welch correction as SDs could be different between the two populations tested. Statistical significance was defined by a two-tail p value below 0.05.

Results
Effect of pre-injection of cationic polymers on the AAV2 transduction efficiency To monitor gene expression, we used as reporter gene a murine secreted alkaline phosphatase (mSeAP). mSeAP expression results in the production of a non immunogenic protein which allows specific detection in blood and tissues. Notably, the background is very low due to a higher thermostability of mSeAP as compared to the endogenous phosphatases [21].
As mentioned in the introduction, we hypothesized that preinjection of cationic polymers could increase the transduction efficiency of AAV2. Since the size of the polymer may play a role, we evaluated four different poly-L-lysines (pLys) with increasing degrees of polymerization (dp): pLys12, pLys45, pLys107, and pLys238. The polymers diluted in 200 ml of a 5% glucose solution were injected 5 minutes prior administration of 450 ml containing 1610 11 vg of AAV2. The kinetics of the secretion of the reporter protein was followed over a period of 28 days using the blood samples which were taken once per week. The results indicate a critical effect of the size of the polymer. Indeed, while pLys12 did not alter the mSeAP levels at any time point as compared to the non pre-treated AAV2 group, the 3 pLys with higher degrees of polymerization enhanced the expression of reporter gene at all time points ( Figure 1A). In this first experiment, the pLys of dp 238 displayed some toxicity since one of the 2 mice died after the 150 mg injection. Further experiments indicated that pLys238 was lethal in 50% of cases after injection of 175 to 200 mg. However, at the doses used to perform the study (i.e. 150 mg) mortality never exceeded 20% of injected mice (not shown).
As shown in Figure 1A, the pre-injection of pLys238 resulted in a higher transgene secretion than when using for example the pLys with dp of 45. To confirm this point, we repeated the experiment with these two polymers. Figure 1B shows that there is a significant difference between both polymers. The enhancement factor of mSeAP secretion was of 4 and 7 at day 42 after pre-injection of pLys45 and pLys238, respectively (statistical difference was achieved between pLys45 and pLys238 at every mean point of the kinetic). Taking into account this result, we used pLys238 in the subsequent experiments.
In order to check whether this enhancement effect on AAV in vivo efficiency is restricted to polylysines, we pre-injected another cationic polymer before the infusion of AAV2: a poly-arginine with a dp of 64 (pArg64). As shown in Figure 1C, the pretreatment with pArg64 allowed for an increase of the secretion of mSeAP of the same order than the polylysines. This suggests that the pre-injection of a cationic polymer displays a general enhancing effect on AAV-mediated gene delivery.
Next, we tested the influence of the time-lapse between cationic polymer pre-injection and AAV administration. The results indicated that while co-injection of both, polymer and AAV, resulted in similar effects than with a 5 minute pre-injection, we also found that longer periods between the two injections reduced the effect (data not shown).
In order to check whether the enhancement mediated by pLys is AAV2 dose-dependent, we pre-injected 150 mg of pLys238 before infusion of either 2.5610 10 or 4610 11 vg. As shown in Figure 2, the pre-injection of the cationic polymer resulted in an increase of the mSeAP levels in blood of about a factor 12. These results clearly show that the pLys effect is AAV2-dose independent.
The mice that received 4610 11 vg of AAV2 with or without a pLys pre-injection were analyzed in greater details. First, by using quantitative real-time PCR, we determined the copy number of viral genomes present in different tissues. As revealed by the results ( Figure 3A) the pre-injection of the cationic polymer increased the copy number in all analyzed tissues, the greatest increase being found in the heart, diaphragm and liver (30-, 15-and 5-fold increase, respectively as compared to the AAV2 treated group). To check whether there is a correlation between copy number of viral genomes and the expression level we quantified the amount of mSeAP present in the tissue lysate. The mSeAP quantification in the various tissues showed a good correlation with the Q-PCR results, except for the liver where the expression levels were particularly low considering that it was the tissue with the highest vg copy number ( Figure 3B). This can be explained by the welldocumented fact that liver is prone to CMV promoter extinction [23,24]. Since mSeAP can also be revealed on tissue sections, we looked whether and to which extent skeletal muscle was transduced. As shown in Figure 3C, while no positive fibers could be seen in the quadriceps of mice from the AAV2 group, a significant amount of fibers were mSeAP positive in the AAV2 treated mice pre-injected with pLys238. Altogether, the data from Q-PCR, mSeAP quantification from tissue lysate and histochemistry of tissue section are in good agreement and show that pLys increases the overall transduction of all tested tissues.
Notably, in vitro data from a transfected cell culture showed that about 90% of mSeAP is secreted, while the remaining 10% are found in the cell lysate [21]. As most of the mSeAP is secreted, we probably underestimate the amount of transduced cells in tissues, especially when using low vector doses. To circumvent this problem, we made use of an AAV2 vector encoding LacZ. Due to the fact that we used a non-secreted cytoplasmic form of this reporter gene, we expected a higher sensitivity as compared to mSeAP. The results showed again a major improvement of heart transduction as seen in Figure 4. Indeed, very few cardiomyocytes expressed b-galactosidase in control mice injected with AAV2 alone whereas pre-injection of pLys238 enabled AAV2 to transduce a large amount of cardiomyocytes.
It has been previously shown that tail vein injection in mice of an AAV vector coding for a reporter gene under the control of the muscle creatine kinase (MCK) promoter results in strong expression in muscles while expression in nonmuscle tissues remains #4% of the highest activity seen in most skeletal muscles [22]. We used the same MCK promoter (AAV2-MCK-mSeAP) in order to follow more specifically the effect of polylysine preinjection on the transduction of skeletal muscles. Our results show that while in the absence of injection of pLys the plasma level of mSeAP was extremely low (11 pg/ml on day 34), a pre-treatment with either pLys45 or pLys238 dramatically increased the reporter gene expression ( Figure 5). This experiment clearly demonstrates that pre-injection of a cationic polymer improves muscle transduction.

Effect of pre-injection of polyinosinic acid on AAV2 mediated in vivo transduction
Based on the results obtained by other groups with recombinant adenoviruses [13], we asked whether the pre-injection of an anionic polymer such as polyinosinic acid (pI) could increase the transduction efficiency of AAV2. A putative mechanism could be, for example, a polyinosinic acid saturation of the scavenger receptors of Kupffer cells, and this in turn could lead to a decreased viral uptake. We injected 1610 11 vg of AAV2-CMV-mSeAP 5 minutes after infusion of 200 mg of pI. The plasma level of mSeAP was determined at different time points. As shown in Figure 6A, the pre-treatment led to a low but significant increase of the mSeAP plasma level.
Since pI may have reduced, or inhibited, the transduction of antigen presenting cells (APCs) by binding to scavenger receptors, we asked whether such a treatment could reduce the humoral response directed against AAV2 capsid proteins. We quantified the neutralizing factors (NAF) in mice 28 days after injection. Neutralizing antibodies from injected mice were titrated in vitro by a virus neutralization assay, using the plasma from the animals ( Figure 6B). Results indicate that the pI pre-treatment resulted, in average, in a significant 4-fold lower NAF titer than the control group (1/1800 vs 1/7400 mean titer). This result clearly suggests that intravenous injection of pI before infusion of AAV2 particles results in a reduced uptake of the virus by APCs, which in turn leads to a reduced production of NAF.
Notably, we also quantified the NAF in 150 mg pLys238-treated mice. The results showed that this latter treatment did not change the NAF titer as compared to the AAV2 injected group ( Figure   S1). This suggests that pLys238 does not increase the transduction of professional antigen presenting cells.

Discussion
The success of gene therapy for the treatment of genetic diseases such as muscular dystrophies requires widespread and stable gene delivery. Adeno-associated viruses (AAV) are, to date, the only vectors that enable body-wide gene delivery. However, this can only be achieved by injecting extremely high amounts of viral particles. For example, using AAV-6, a serotype which efficiently transduces muscle, 10 12 vg (about 5610 13 vg/kg) have to be administered for widespread transduction of both cardiac and skeletal muscles in adult mice [1]. Based on this and other results, it is hypothesized that $10 15 vg will have to be administered intravenously to obtain similar results in humans. Besides raising safety questions, the feasibility of producing such large amounts of AAVs in GMP conditions remains to be shown. Indeed, with the routine method consisting in the transfection of HEK293 cells cultured in 15 cm diameter plates, production of 10 15 particles would require 5700 plates [25].
Here, we investigated whether polylysine or polyinosinic acid could act as adjuvants and improve the in vivo transduction efficiency. The results obtained with cationic polymers, and in particular polylysines, showed that co-or pre-injection a few minutes before infusion of AAV2 particles significantly increases the overall transduction efficiency. This was shown by quantitative PCR, as well as by the mSeAP expression in plasma and in different tissues. However, it is to be noted that the increase was not similar in all tissues: the greatest effects were obtained in the heart and diaphragm. The magnitude of the improved transduction was, depending on the experiment, 5 to 12-fold more mSeAP than in the control group. Notably, this effect was not AAV2 dose-dependent. Interestingly, the enhancement of the mSeAP secretion was correlated to the polymerization degree of the pLys. This suggests that a minimal size of the polymer is required for boosting the  transduction. The results obtained after the injection of pArg64 ( Figure 1C) show that it is also possible to increase the transduction efficiency of AAV2 by using other cationic polymers than polylysines. This finding is important since it means that alternative compounds which may have less toxic effects than pLys238 exist. The exact reasons for the lethality provoked by injection of high doses of pLys238 are unknown. Of note however, our results are in agreement with the study of Arnold and colleagues which showed that in vivo toxicity of polylysines decreases with the molecular weight of the polymer [26]. These authors suggest that cytotoxicity of polylysines results from membrane perturbations which, for example, affect membrane permeability to small molecules. Besides the intrinsic toxicity of high doses of pLys238, one could mention two other parameters that may have played a role in the toxic effects: firstly, the purity of the polymer. Indeed, pLys238 is a mixture of polylysines (the molecular weight ranges between 30,000 and 70,000) and we can not exclude the presence of cytotoxic contaminants. Secondly, the counterion of pLys238 is bromide, a ion that has been shown to be toxic for cells [27]. Taken together, high molecular weight polylysines are cytotoxic at high doses but it may be possible to reduce this toxicity by using a highly purified polylysine and by exchanging the bromide ion by chloride for example. Alternatively, in view of clinical applications one could use polylysines of lower molecular weight which are less toxic or use other cationic polymers.
How exactly the polymers act to increase the transduction remains unknown. One possibility could be that the effect results from the modification of the surface charge of the AAV particles. Indeed, it has been reported that the surface charge (zetapotential) of AAV2 is negative and that addition of a cationic compound such as protamine modifies its surface charge [28].
The results obtained with AAV2 encoding for mSeAP under the control of the muscle specific promoter MCK are interesting, since the increase after pLys treatment of the circulating mSeAP levels is impressive. It has been reported that the synthesis of heparan sulfate proteoglycan is down-regulated during murine skeletal muscle maturation, and this has been proposed to be responsible for the loss of HSV infectivity [29] (heparan sulfate acts as a coreceptor for attachment of HSV to cells). Based on this report, we propose the hypothesis that due to low levels of heparan sulfate proteoglycan in muscle, cationic polymers improve the transduction by facilitating the primary interaction between virus and the muscle fibers. Co-receptors would subsequently intervene for virus internalization. Alternatively, pLys could neutralize the anionic charges present on the vascular endothelium and this, in turn, could favour AAV2 tissue penetration.
Based on results obtained with adenoviruses [13], we evaluated a second strategy which consisted in pre-injecting the anionic polymer polyinosinic acid. In fact, we asked whether an injection of pI could prevent infection by AAV2 of Kupffer cells or other antigen presenting cells and thus, not only enhance the transduction of other tissues but also reduce the humoral immune response against AAV2 capsid proteins. The results showed that a pI treatment resulted in a two-fold increase of the circulating mSeAP levels. A second, interesting effect of the pre-treatment was the fact that it reduced by 4-fold the levels of neutralizing factors directed against the capsid proteins. We attribute this latter effect to an inhibition of AAV2 uptake by antigen presenting cells, due to the inhibition of scavenger receptors by pI. This is of particular importance when considering the fact that the immune response against AAV vectors after a first injection hinders future readministrations. Although the present reduction in NAF may not be sufficient to allow a re-administration, it opens new perspectives for developing strategies that will be more efficient than a pI preadministration.
In conclusion, our results show that pre-injection of polymers, especially cationic ones, have the potential to significantly increase the transduction efficiency of various tissues, including skeletal muscle. This strategy potentially allows a reduction of the injected vector dose as compared to an injection of AAV2 alone. This is not only a benefit in terms of safety (a lower viral load reduces potential for immune response to the capsid proteins) but it is also of real interest from a GMP production perspective. Indeed, if one can divide by several-fold the production scale this will represent a very significant reduction of the costs.

Supporting Information
Figure S1 Impact of pre-injection of pLys238 on the antibody response against AAV2. Neutralizing factors (NAF) generated against the capsid proteins of the vector in AAV2 and pLys238/AAV2 treated animals were titrated in Figure 6. Impact of polyinosinic acid pre-injection on the transduction efficiency and antibody response against AAV2. (A) The mSeAP secretion in blood was measured after the pre-injection of 200 mg pI followed five minutes later by an IV injection of 1610 11 vg of AAV2-CMV-mSeAP. Plasma concentration of mSeAP was monitored every week for 4 weeks. Results are expressed as the average value +/2 SD, with n = 9 for mice injected with the vector alone (solid line), and n = 5 for mice treated with pI (dotted line). (B) Inhibition of neutralizing factors (NAF) generated against AAV2 with polyinosinic acid preinjection. NAF from plasma collected at day 28 were titrated. Serial dilutions of plasma from mice injected with AAV2 with or without a preinjection of 200 mg pI were incubated with the identical serotype-based AAV2-Luc encoding the luciferase gene. Residual AAV infectivity was then measured on HeLa cells. Each curve represents a mouse: dotted curves represent pI treated mice and solid curves represent control mice injected with the vector alone. doi:10.1371/journal.pone.0015576.g006 plasma collected at day 28. Serial dilutions of plasma from mice injected with 4610 11 vg of AAV2 (experiment Figure 2) with or without a pre-injection of 150 mg pLys238 were incubated with AAV2-CMV-Luc encoding the luciferase gene. Residual AAV2 infectivity was then measured on HeLa cells. Each curve represents a mouse: dotted curves represent pLys238 treated mice (n = 3) and solid curves represent control mice injected with the vector alone (n = 3). (TIF)