AAV-Mediated Gene Targeting Is Significantly Enhanced by Transient Inhibition of Nonhomologous End Joining or the Proteasome In Vivo

Abstract

Recombinant adeno-associated virus (rAAV) vectors have clear potential for use in gene targeting but low correction efficiencies remain the primary drawback. One approach to enhancing efficiency is a block of undesired repair pathways like nonhomologous end joining (NHEJ) to promote the use of homologous recombination. The natural product vanillin acts as a potent inhibitor of NHEJ by inhibiting DNA-dependent protein kinase (DNA-PK). Using a homology containing rAAV vector, we previously demonstrated in vivo gene repair frequencies of up to 0.1% in a model of liver disease hereditary tyrosinemia type I. To increase targeting frequencies, we administered vanillin in combination with rAAV. Gene targeting frequencies increased up to 10-fold over AAV alone, approaching 1%. Fah ^−/− Ku70 ^−/− double knockout mice also had increased gene repair frequencies, genetically confirming the beneficial effects of blocking NHEJ. A second strategy, transient proteasomal inhibition, also increased gene-targeting frequencies but was not additive to NHEJ inhibition. This study establishes the benefit of transient NHEJ inhibition with vanillin, or proteasome blockage with bortezomib, for increasing hepatic gene targeting with rAAV. Functional metabolic correction of a clinically relevant disease model was demonstrated and provided evidence for the feasibility of gene targeting as a therapeutic strategy.

Introduction

A deno-associated virus (AAV) is one of the safest viral vectors for use in human gene therapy (Buning et al., 2003). The field of AAV gene therapy research continues to be dominated by gene addition approaches using cDNA cassettes driven by heterologous promoter and enhancer sequences to augment expression of a mutated disease gene. Yet, drawbacks to this methodology abound including unregulated and transient episomal expression (Garrick et al., 1998; Russell and Kay, 1999), random integration (Hendrie et al., 2003), increased oncogenic risks (Donsante et al., 2007), and the inability to treat genetically dominant disorders arising from misfolded or abnormally functioning proteins.

AAV-based gene-targeting paradigms differ in that patient genomes are manipulated in vivo using site-specific recombination to correct underlying mutations. In this way, gene function is restored within the context of endogenous regulatory elements, thus eliminating problems with inadequate or inappropriate expression. AAV gene-targeting vectors can correct different genomic loci, show both targeted and stable expression through integration, and cover a greater number of applicable human diseases (Hendrie and Russell, 2005). Single-stranded AAV genomes utilize homologous recombination machinery to correct mutations through targeted site-specific integration (Parekh-Olmedo et al., 2005).

Our lab has demonstrated the utility of recombinant adeno-associated virus (rAAV) for performing gene targeting in an ENU-generated point mutation mouse model of hereditary tyrosinemia type I (HTI) (Paulk et al., 2010), a fatal disease arising from a breakdown in tyrosine catabolism at fumarylacetoacetate hydrolase (FAH). In both HTI patients and Fah^−/− mice, hepatotoxic metabolites accumulate, causing death in a cell-autonomous manner. Treatment with 2-(2-nitro-4-trifluoromethylbenzoyl)-1,3-cyclohexanedione (NTBC), an inhibitor that blocks upstream of FAH, prevents metabolite accumulation and rescues the phenotype (Al-Dhalimy et al., 2002). FAH+ cells have a strong selective advantage in the HTI mouse liver (Overturf et al., 1996) that can be exploited for selection of repaired hepatocytes in vivo. One advantage of the Fah model for gene-targeting studies is that the readout is functional phenotype restoration, not merely correction of an artificial reporter system with no bearing on physiology. When rAAV-Fah vectors are administered to Fah ^−/− mice, only corrected hepatocytes survive and repopulate the liver. Our repair vectors contain a 4.5-kb genomic Fah fragment centered on the nucleotide needed for correction in the Fah^−/− mouse. Hepatic correction frequencies of up to 0.1% were previously achieved with this model, a level that was subtherapeutic without selection (Paulk et al., 2010).

To improve upon earlier results and further increase correction to more therapeutic levels, we sought to transiently inhibit unwanted repair pathways such as nonhomologous end joining (NHEJ) to promote the sole use of homologous recombination (HR) machinery. NHEJ is the primary pathway of double-strand break (DSB) repair in postembryonic human cells (van Gent et al., 2001). AAV vectors do not contain an endonuclease to create DNA DSBs and free chromosomal ends, thus naturally occurring breaks must be used when integrating. If NHEJ utilization is prevented, theoretically HR proteins should take over to repair the break with the homologous AAV repair template. Vanillin (3-methoxy-4-hydroxybenzaldehyde), a natural phytochemical, acts as a potent inhibitor of NHEJ by binding to and inhibiting DNA-dependent protein kinase (DNA-PK) (Durant and Karran, 2003). Vanillin is highly selective for DNA-PK and has been shown to have no measurable effect on other steps of NHEJ or on various other unrelated protein kinases (Durant and Karran, 2003).

Through work using proteasomal inhibitors, it was shown that the majority of AAV particles do not deliver their genomic payload to the nucleus (Johnson and Samulski, 2009). Indeed, several labs have shown that proteasomal inhibitors are capable of enhancing AAV transduction both in vitro and in vivo (Johnson and Samulski, 2009; Finn et al., 2010; Monahan et al., 2010), but the mechanism by which this occurs remains unknown. Several hypotheses have been put forward including that they increase AAV trafficking efficiency, block cytoplasmic or nuclear capsid degradation indirectly improve genome stability; decrease T-cell activation/proliferation; or decrease capsid antigen presentation. The only proteasomal inhibitor currently approved for clinical use is bortezomib (Velcade^™), and we therefore sought to uncover whether this compound would work alongside vanillin in our AAV gene-targeting studies by improving AAV transduction.

Materials and Methods

Mouse strains and animal husbandry

The ENU-generated Fah^5981SB mouse (herein described as Fah^−/− ) models human HTI with a single point mutation in exon 8 of Fah (Culiat et al., 2005). Neonates die from acute liver failure if NTBC is not continually administered. NTBC treatment at 4 mg/L in drinking water prevents hepatorenal injury and rescues the phenotype. NTBC withdrawal allows modeling of HTI as mice rapidly develop hepatorenal disease over the course of several weeks and die of end-stage liver disease in 6–8 weeks (Grompe et al., 1995). These mice have been backcrossed 10 generations onto a C57BL/6J background. The various Fah/Ku70 mice were generated by crossing Fah ^−/− mice with Ku70^+/− mice (Gu et al., 1997) to establish Fah^−/−Ku70^+/− breeders. All mice described are maintained on irradiated high-fat, low-protein mouse chow (Lab Diet, Brentwood, MO, cat. no. Picolab 5LJ5) ad libitum to decrease flux through the tyrosine pathway. The Institutional Animal Care & Use Committee of Oregon Health & Science University approved all mouse procedures.

Plasmid vectors and recombinant AAV production

Plasmid construction and design strategy were described previously (Paulk et al., 2010). Standard transfection and viral isolation protocols described previously (Grimm, 2002), were used with the following modifications. Unpurified virus was sedimented and resuspended in benzonase buffer for freezing at −80°C. After three freeze–thaw cycles in a dry ice/ethanol bath, the suspension was benzonase-treated (EMD Chemicals, San Diego, CA). Virus was pelleted and precipitated with 1 M CaCl₂, followed by 40% PEG-8000 treatment and resuspension in a sodium/HEPES buffer and rotated overnight at 4°C. After resuspension, purification by sequential cesium chloride density centrifugation was performed as described (Choi et al., 2007) with the following modifications. Peak fractions from both gradients were determined by optical refractometry. Positive final fractions were extensively dialyzed against phosphate-buffered saline lacking Ca²⁺ or MgCl₂ (Gibco, Grand Island, NY) and supplemented with 5% (w/v) D-sorbitol. Vector titering was performed by dot blot.

Neonatal vector administration

For comparative treatment studies, P1 Fah ^−/− neonates were injected with 1×10¹¹ vector genomes (vg) per mouse (in 10-μl volume) of rAAV2/8-Fah by intravenous (IV) injection into the superficial temporal vein. All mice were maintained on NTBC throughout. Livers were harvested at weaning, 21 days post-treatment. For random integration studies, P1 Fah ^−/− neonates received 4×10¹⁰ vg/mouse of both rAAV2/8-Fah (Paulk et al., 2010) and rAAV2/8-LSP-eGFP (Wang et al., 1999) in a 20-μL volume by IV injection into the superficial temporal vein. Neonates were treated with NTBC until weaning and then withdrawn from it to select for corrected hepatocytes. Serum and liver tissue were collected after serial transplantation.

Adult vector administration

Adult Fah ^−/− mice (age ≥13 weeks) were administered 2×10¹¹ vg/mouse of rAAV2/8-Fah in 100-μl volume by IV lateral tail vein injection. NTBC treatment was continued for 1 week following vector administration during capsid uncoating, and then withdrawn for 3 weeks up to 3 months depending on the study. Liver tissue and serum were collected at the end of each experiment.

Vanillin administration

For each experiment, a fresh vanillin (Alfa Aesar, Ward Hill, MA) stock solution of 10 mg/ml solubilized in dH₂O was prepared, sterile filtered, and kept at 55°C to keep vanillin in solution. This solution was then administered at 100 mg/kg/d by intraperitoneal (IP) injection for 7 days starting on the day of vector administration.

Bortezomib administration

For each experiment, a fresh bortezomib (Melone Pharmaceutical, Edison, NJ) stock solution of 5 mg/ml solubilized in dimethyl sulfoxide (DMSO) was prepared, sterile filtered and diluted in 0.9% sterile saline for co-administration with AAV at 0.5-mg/kg by IV lateral tail vein injection (100-μl total volume).

4,5-Dimethoxy-2-nitrobenzaldehyde administration

For each experiment, a fresh 4,5-dimethoxy-2-nitrobenzaldehyde (DMNB; Sigma, Milwaukee, WI) stock solution of 10 mg/ml solubilized in DMSO was prepared, sterile filtered, and diluted in 0.9% sterile saline. This solution was then administered at 100 mg/kg on days 2 and 5 after vector administration by IP injection.

Transplantation

For serial transplantations, livers of corrected donor mice were perfused as described (Grompe et al., 1992) and 4.5×10⁵ random hepatocytes were transplanted into six recipients, each by intrasplenic injection as described (Ponder et al., 1991).

Liver immunohistochemistry

In all experimental mice, a minimum of four liver sections from varying depths and randomly selected liver lobes were analyzed for the number of FAH+ nodules. Since nodules represent the clonal expansion of a single corrected hepatocyte, nodule frequency was corrected for nodule size. Briefly, the surface area of each liver section was measured by scanning the glass slide along with a size standard using an Epson Perfection V700 scanner at 400-dpi resolution. Adobe Photoshop 7.0 software was then used to select and count pixels of each liver section. Pixel counts were then converted to numbers of hepatocytes based on the diameter of a mouse hepatocyte (∼25 μm). Correction factors were then applied to frequencies of FAH+ nodules in a section containing x hepatocytes based on the following assumptions, as described previously (Wang et al., 2002): 1) hepatocyte nodules are spherical; 2) all nodules in a given sample are approximately the same size; 3) the number of hepatocytes in the largest clone represents the middle of that nodule. Immunohistochemical fixation and staining protocols for FAH and hematoxylin and eosin (H&E) were completed as described (Wang et al., 2002). Two separate, blinded investigators performed quantitation. Microscopy was performed on a DM IL LED microscope (Leica, Buffalo Grove, IL) using Leica LAS Image Analysis Software. Blood for serology was collected by terminal cardiac puncture and biochemical measurements were completed as described (Grompe et al., 1995).

Statistical analysis

Statistical analyses were conducted with GraphPad Prism software v.4.0 (GraphPad, San Diego, CA). Experimental differences were evaluated by student two-tailed t-test assuming equal variance. P values <0.05 were considered statistically significant.

eGFP copy number quantitative PCR

Total DNA was isolated from randomly dissected liver tissue with a MasterPure DNA Purification kit (Epicentre Biotechnologies, Madison, WI). Genomic DNA (75 ng) was subjected to a two-step PCR amplification under the following conditions: one cycle at 95°C for 3 min; 45 cycles at 95°C for 15 sec and 68°C for 40 sec; and primers, eGFP F: 5′-ACTTCAAGATCCGCCACAAC-3′; eGFP R: 5′-GAACTCCAGCAGGACCATGT-3′; Actb F: 5′-CCACCCCAGCAAGGACACTG-3′; Actb R: 5′-GCTCCCTAGGCCCCTCCTGT-3′. Dilutions of eGFP plasmid into mouse genomic DNA were used to generate copy number standards. Results were normalized to β-actin expression. PCR was performed on an iQ5 Multicolor Real-Time PCR (Bio-Rad, Hercules, CA), using iQ5 Standard Edition Software, v.2.0.

Results

Transient inhibition of either NHEJ or the proteasome increases gene targeting

To examine the effects of pharmacological inhibition of NHEJ or the proteasome on gene targeting, 13-week-old Fah^−/− adult mice were subjected to one of four treatments: AAV alone, AAV and vanillin, AAV and bortezomib, or AAV and both vanillin and bortezomib. rAAV2/8-Fah was given at a dose of 2×10¹¹ vg/mouse by IV tail vein injection. Based on pharmacokinetic dosage data on vanillin in a model of neuropathic pain (Beaudry et al., 2010), we administered vanillin at 100 mg/kg/d by IP injection for 7 days starting on the day of vector administration. Using published dosages of bortezomib from promising studies in hemophilic mice (Monahan et al., 2010), we co-injected bortezomib with rAAV at 0.5 mg/kg by IV tail vein injection. Mice were maintained on NTBC for 1 week following vector administration, and then it was withdrawn for 3 weeks. Livers were harvested from mice and scored for FAH+ nodules (Fig. 1). Nodules represent the clonal expansion of a single corrected hepatocyte, thus nodule frequency was corrected for nodule size (Wang et al., 2002).

FIG. 1.

Transient non-homologous end joining (NHEJ) or proteasome inhibition enhances gene targeting. Frequencies of corrected nodules were quantified by counting the number of FAH+ clones per x hepatocytes (1/x) from adult mice harvested 4 weeks posttreatment. Mean and SD are shown, with the number of independent animals analyzed above each bar. Black, adeno-associated virus (AAV); white, AAV + vanillin; gray, AAV + bortezomib; lines, AAV + vanillin + bortezomib.

Figure 2 shows representative FAH immunohistochemistry from control (Fig. 2a, b) and rAAV treated mice (Fig. 2c–f). All cotreatment regimens had significantly higher gene-targeting frequencies than AAV alone. H&E staining on control mice treated only with vanillin (Fig. 2g) or only with bortezomib (Fig. 2h) showed a lack of overall toxicity.

FIG. 2.

Liver immunohistochemistry. (a) Fumarylacetoacetate hydrolase positive (FAH+) staining from adult Fah^+/+ control mouse; (b) FAH− staining from adult Fah^−/− control mouse; (c) FAH stain on adult Fah^−/− mouse treated with AAV, (d) AAV + vanillin, (e) AAV + bortezomib, or (f) AAV + vanillin + bortezomib; (g) hematoxylin and eosin (H&E) stain on adult Fah^−/− mouse treated with vanillin; (h) H&E stain on adult Fah^−/− mouse treated with bortezomib; (i) FAH stain on adult Fah^−/− serial transplant recipients; (j) FAH stain on Fah^−/−Ku70^+/+ neonate treated with AAV; (k) FAH stain on Fah^−/−Ku70^+/− neonate treated with AAV; (l) FAH stain on Fah^−/−Ku70^−/− neonate treated with AAV. Scale bar=10 μm.

Given that bortezomib was no better at enhancing gene targeting than vanillin, we decided to focus exclusively on vanillin cotreatment for the duration of the studies. Of note, attempts to replicate the adult cotreatment studies in neonates were not possible because bortezomib treatment was lethal during this developmental window (data not shown).

Gene targeting frequencies are sexually dimorphic

To examine effects of sex and age on gene targeting, P1 Fah^−/− neonates and 13-week-old Fah^−/− adult mice of different sexes were treated with either rAAV or rAAV and vanillin. rAAV2/8-Fah was given at 2×10¹¹ vg/mouse by IV tail vein injection to adults, and 1×10¹¹ vg/mouse by IV facial vein to neonates. Vanillin was administered at 100 mg/kg/d by IP injection for 7 days starting on the day of vector administration. Neonates were maintained on NTBC until harvest at weaning to prevent metabolic selection of FAH+ cells. Adults were maintained on NTBC for 1 week following vector administration and then it was withdrawn for 3 weeks. Livers were harvested from mice and scored for FAH+ nodules (Fig. 3). In both ages of mice, males consistently showed significantly higher gene targeting than female littermates when treated with AAV alone. In neonates, a significant difference was seen in both genders with vanillin cotreatment, compared to AAV alone. Even with cotreatment, male neonates had significantly more gene targeting than female littermates. This trend continued in adult mice (Fig. 3), wherein males always showed significantly greater gene targeting than females, no matter the treatment cohort. Interestingly, adult females showed no significant difference between AAV treatment and AAV and vanillin combination therapy.

FIG. 3.

Sexually dimorphic responses to AAV and vanillin. Frequencies of corrected nodules were quantified by counting the number of FAH+ clones per x hepatocytes (1/x) from neonates harvested at 3 weeks and adults harvested 4 weeks posttreatment. Mean and SD are shown, with the number of independent animals analyzed above each bar. Black, AAV; white, AAV + vanillin; M, male; F, female.

Toxicity profile of vanillin in vivo

To investigate whether the significant increases in gene targeting with vanillin treatment was the result of NHEJ inhibition or simply a toxic response to vanillin causing hepatocellular turnover, we performed in vivo toxicity studies on mice treated only with vanillin. Fah^−/− adult mice were treated with vanillin alone at 100 mg/kg/d by IP injection for 7 days. Mice were maintained on NTBC during the experiment to mimic the experimental conditions used during vector combination therapy studies. Livers were immediately harvested after the final day of vanillin treatment and examined for signs of injury by immunohistochemistry and serology. Results by H&E showed only a slight loss of hepatic glycogen retention but no signs of injury (Fig. 2g). Serology to examine potential hepatic toxicity showed normal bilirubin and transaminase levels (Table 1, column d), further demonstrating the safety and nontoxic attributes of vanillin.

Table 1.

Serological Measures of Functional Correction

Mice treatment	a Fah^−/− off NTBC	b Fah^−/− on NTBC	c Wild-type	d Fah^−/− Van	e Fah^−/− AAV	f Fah^−/− AAV + Van	g sTx recipients AAV + Van
Bili (mg/dl)	3.4±1.3	0.2±0	0.0±< 0.1	<0.1±0	<0.1±0	<0.1±0	0.2±0.1
AST (U/L)	353±86	121±47	91±27	98±26	86±8	91±32	116±13
ALT (U/L)	291±51	59±56	36±5	38±4	45±1	53±12	92±10
n	6	5	5	3	3	7	6

AAV, adeno-associated virus; AST, aspartate aminotransferase; ALT, alanine aminotransferase; Bili, bilirubin; Fah, fumarylacetoacetate hydrolase; n, number of animals; NTBC, 2-(2-nitro-4-trifluoromethylbenzoyl-1,3-cyclohexanedione; sTx, serial transplant; U/L, units per liter; Van, vanillin. From left to right, values represent mean±SD from untreated Fah^−/− mice off NTBC, untreated Fah^−/− mice on NTBC, untreated wild-type mice, Fah^−/− mice treated with vanillin alone for 7 days on NTBC, Fah^−/− mice treated with AAV alone and selected for 3 months off NTBC, Fah^−/− mice treated with AAV and vanillin and selected for 3 months off NTBC, and serially transplanted Fah^−/− mice selected for 3 months off NTBC, respectively. All values in columns (b–g) are significantly different from corresponding values in column (a) (p<0.05).

Genetic mutations in Ku70 mimic chemical affects of vanillin in vivo

Having established that vanillin treatment is not toxic, we sought to confirm the effects of blocking NHEJ with a genetic approach. Ku70^−/− mice, which genetically lack the ability to perform NHEJ through loss of functional Ku70, were bred with Fah^−/− mice to create double mutants. P1 Fah ^−/− Ku70 ^−/− neonates were treated with 1×10¹¹ vg/mouse of rAAV2/8-Fah by IV facial vein and compared to identically treated Fah^−/−Ku70^+/+ and Fah^−/−Ku70^+/− littermates (Fig. 4). If NHEJ loss was truly responsible for the increase in gene targeting seen with vanillin cotreatment, then AAV treatment alone in mice genetically incapable of NHEJ should show similar gene correction frequencies. As hypothesized, Fah^−/−Ku70^+/+ and Fah^−/−Ku70^+/− neonates treated with AAV had normal levels (∼0.1%) of gene correction (Fig. 2j,k). Only the Fah^−/−Ku70^−/− littermates (Fig. 2l) had significantly (p<0.03) higher gene correction frequencies at the levels seen in Fah^−/− mice treated with AAV and vanillin (∼1%).

FIG. 4.

NHEJ inhibition is responsible for increases in gene targeting seen with vanillin. Frequencies of corrected nodules were quantified by counting the number of FAH+ clones per x hepatocytes (1/x) from neonates harvested at 3 weeks posttreatment. Mean and SD are shown, with the number of independent animals analyzed above each bar. Black, Fah ^−/− Ku70 ^+/+; white, Fah ^−/− Ku70 ^+/−; gray, Fah ^−/− Ku70 ^−/−.

Random integration frequencies remain unaffected by transient NHEJ inhibition

Although phenotypic reversion indicated successful site-specific gene targeting, random integration of AAV genomes could also occur. Current data in the field are conflicting as to whether inhibition of NHEJ would worsen (Song et al., 2004) or improve (Fattah et al., 2008) random integration frequencies. Since the majority of postembryonic DNA damage is repaired by NHEJ, it is plausible that a transient block to NHEJ would decrease the frequency of random integration. To assess whether random integration frequencies change with vanillin treatment, P1 Fah ^−/− neonates were co-injected with 4×10¹⁰ vg/mouse of both rAAV2/8-Fah and rAAV2/8-LSP-eGFP (nonselectable serotype-matched control vector). Concomitantly, neonates were treated by IP injection with vanillin at 100 mg/kg for 7 days following viral administration. Postweaning, mice were subjected to NTBC withdrawal to select for corrected hepatocytes. To ensure no episomes remained and to select for integrants, 5×10⁵ random hepatocytes were then serially transplanted into six secondary Fah ^−/− recipients as described (Overturf et al., 1997). After >12 weeks off NTBC, serum and liver tissue of the secondary transplant recipients were collected.

Liver function tests for bilirubin and transaminases demonstrated complete correction in all treatment cohorts when compared to untreated controls off NTBC (Table 1). Liver FAH immunohistochemistry illustrated the extensive repopulation in serial transplant recipients (Fig. 2i). Quantitative PCR was used to determine eGFP copy numbers in each of these highly repopulated mice (Table 2). The average copy number of irrelevant vector corrected for repopulation efficiency indicated that ∼0.8% of gene-targeted hepatocytes also had a random rAAV2/8-LSP-eGFP integration event. This number was well within the range of reported random AAV integration estimates from the literature between 0.1% and 10% (Porteus et al., 2003; Miller et al., 2005; Nakai et al., 2005; Vasileva et al., 2006). As a result, it can be concluded that vanillin treatment did not significantly increase or decrease AAV random integration frequencies.

Table 2.

eGFP Copy Numbers in Recipient Mice

Donor	sTx recipient	# copies/dGE	Average
1	1a	0.01	0.01
	1b	0.01
	1c	0.01
2	2a	0.00	0.006
	2b	0.01
	2c	0.01
Total			0.008
Random integration			0.008=0.8%

dGE, diploid genome equivalent; sTx, serial transplant. After neonatal co-injection of 1×10¹⁰ vg of rAAV2/8-Fah and rAAV2/8-LSP-eGFP and 7 days of vanillin treatment, FAH+ hepatocytes were selected for and then serially transplanted to remove episomes. Data represent the number of copies of eGFP per diploid genome equivalent in serial transplant recipients. Roughly 50% of the hepatocytes in repopulated recipients were donor derived at the time of harvest, so the frequency was corrected by a factor of 2 to give an estimated random integration frequency of 0.8%.

Discussion

In spite of advances made in gene therapy research, the disadvantages of traditional AAV gene addition methodologies remain: episomal expression will always be inherently unregulated, expression will be transient in cells that turn over, the risk of eventual transgene silencing will always be present, integration events will be random because they are not based on homology, random integration will produce increased mutagenic and oncogenic risks, and disorders caused by aberrantly functioning or misfolded proteins are untreatable by this approach. Our interest in AAV-mediated gene targeting stems from its ability to elegantly bypass these disadvantages through simple site-directed gene repair.

The precise nature of the interactions between NHEJ machinery and AAV are not clear and thus is was unknown whether inhibition of NHEJ would exacerbate (Song et al., 2004) or improve (Fattah et al., 2008) both random and targeted integration frequencies. We pursued both pharmacological and genetic approaches in various ages and genders of mice with a clinically relevant disease model to resolve the differences in the current literature. Our results show that transient NHEJ inhibition with vanillin both increased targeted AAV integration and surprisingly, had no significant effect on random AAV integration.

Our results with vanillin combination therapy are all the more promising when other attributes of vanillin are taken into consideration: vanillin is a known anticarcinogen (Akagi et al., 1995), antioxidant (Burri et al., 1989), anticlastogen (Sasaki et al., 1990), antimutagen (Keshava et al., 1998), antimicrobial (Davidson and Naidu, 2000), and analgesic (Beaudry et al., 2010), and it has antineoplastic activity (McCann et al., 2007). Moreover, vanillin is already a U.S. Food and Drug Administration (FDA)-approved food additive, thus an off-label use for vanillin in gene therapy seems feasible. Attempts to further improve gene targeting with vanillin analogs were unfruitful. DMNB, a known vanillin analog and even more potent disruptor of DNA-PK (Durant and Karran, 2003), was unsuccessful at improving gene targeting at doses tested in a 2-day time course (Supplementary Fig. S1; Supplementary Data are available online at www.libertpub.com/hum) and was not well tolerated in mice. Both 5- and 7-day courses of DMNB proved lethal.

In addition to transient NHEJ inhibition, transient proteasomal inhibition is a promising avenue in gene therapy research given the results seen in mouse and dog models of hemophilia treated with AAV and bortezomib (Monahan et al., 2010). Our data with bortezomib highlights its potential as a second parallel approach to vanillin treatment, wherein bortezomib combination therapy improved gene-targeting frequencies in a safe and efficacious manner. Having already gained FDA approval for human multiple myeloma and mantle cell lymphoma studies, bortezomib combination therapies with AAV have potential for rapid approval and the benefit of pre-existing toxicity studies in humans (Richardson et al., 2005).

As in other fields, gene therapists are beginning to appreciate sex differences in response to AAV therapies (Davidoff et al., 2003; Ogura et al., 2006; Ho et al., 2008; Rebuffat et al., 2010). For example, in hepatocytes there are more than 1000 genes that are differentially expressed between males and females (Rinn and Snyder, 2005; Clodfelter et al., 2006). However, no data exist to date for AAV gene-targeting vectors in animal models of any disease in which sex was controlled. Here we present the first data demonstrating that at all life stages, male mice displayed significantly more AAV-mediated gene targeting than their female counterparts. The detailed mechanisms responsible for sex differences in AAV liver transduction are still unclear but have at least been shown to involve androgen-dependent pathways (Davidoff et al., 2003). The implications of our findings could affect the design and dosing strategies of future clinical trials.

Obstacles to AAV-mediated gene targeting remain. Notably, therapeutic levels of gene correction have proven elusive in disease models. However, our current work highlights the capacity of AAV-mediated gene targeting in a suitable selection-based disease. While disorders characterized by selection for gene-corrected cells overall are rare, there are other examples of this phenomenon: Fanconi's anemia (Battaile et al., 1999), the copper storage disorder Wilson's disease (Allen et al., 2004), bile-acid transporter defects (De Vree et al., 2000), and junctional epidermolysis bullosa (Ortiz-Urda et al., 2003) to name a few. If correction frequencies achieved here were increased just fivefold, they would become clinically relevant even for diseases that do not display a selective advantage. For example, it has been predicted that gene-targeting frequencies of 5% would have therapeutic benefit for patients with hemophilia A (Kay and High, 1999), and 10% for phenylketonuria patients (Fang et al., 1994). Patients with diseases arising from point mutations are the most appropriate recipients for gene-targeting therapies. Historically, point mutations have been the most frequent genetic abnormality and source of acquired genetic disease (Bergeron, 2004).

This study establishes the utility of transient NHEJ inhibition with vanillin, or proteasome inhibition with bortezomib, for increasing hepatic gene targeting with AAV in all ages and sexes of mice. The drug doses used could be applied in humans; hence, these approaches may have human clinical relevance.

Footnotes

Acknowledgments

We thank Angela Major (Texas Children's Hospital) for histology support, the Molecular Morphology Core (Texas Gulf Coast Digestive Diseases Center) for serological analysis, Fred Alt and Yuko Fujiwara (Harvard University) for the Ku70^+/− mice, and Aurelie Snyder (Oregon Health & Science University) for microscopy support. This work was supported by grants from the National Institute of Diabetes, Digestive & Kidney Diseases to M.G. (R01-DK48252) and M.J.F. (P30-DK56338) and the National Cancer Institute to N.K.P. (F31-CA130116). The funding organizations played no role in experimental design, data analysis, or manuscript preparation.

Author Disclosure Statement

No competing financial interests exist for any of the authors of this study.

References

Akagi

, Hirose

, Hoshiya

et al. 1995. Modulating effects of ellagic acid, vanillin and quercetin in a rat medium term multi-organ carcinogenesis model. Cancer Lett., 94:113–121.

Al-Dhalimy

, Overturf

, Finegold

, Grompe

2002. Long-term therapy with NTBC and tyrosine-restricted diet in a murine model of hereditary tyrosinemia type I. Mol. Genet. Metab., 75:38–45.

Allen

K.J.

, Cheah

D.M.

, Wright

P.F.

et al. 2004. Liver cell transplantation leads to repopulation and functional correction in a mouse model of Wilson's disease. J. Gastroenterol. Hepatol., 19:1283–1290.

Battaile

K.P.

, Bateman

R.L.

, Mortimer

et al. 1999. In vivo selection of wild-type hematopoietic stem cells in a murine model of Fanconi anemia. Blood, 94:2151–2158.

Beaudry

, Ross

, Lema

P.P.

, Vachon

2010. Pharmacokinetics of vanillin and its effects on mechanical hypersensitivity in a rat model of neuropathic pain. Phytother. Res., 24:525–530.

Bergeron

2004. Case Studies in Genes and Disease. American College of Physicians Press: Philadelphia, PA.

Buning

, Nicklin

S.A.

, Perabo

et al. 2003. AAV-based gene transfer. Curr. Opin. Mol. Ther., 5:367–375.

Burri

, Graf

, Lambelet

, Loliger

1989. Vanillin: more than a flavouring agent—a potent antioxidant. J. Sci. Food Agric., 48:49–56.

Choi

V.W.

, Asokan

, Haberman

R.A.

, Samulski

R.J.

2007. Production of recombinant adeno-associated viral vectors for in vitro and in vivo use. Curr. Protoc. Mol. Biol.Chapter 16, Unit 16 25.

10.

Clodfelter

K.H.

, Holloway

M.G.

, Hodor

et al. 2006. Sex-dependent liver gene expression is extensive and largely dependent upon signal transducer and activator of transcription 5b (STAT5b): STAT5b-dependent activation of male genes and repression of female genes revealed by microarray analysis. Mol. Endocrinol., 20:1333–1351.

11.

Culiat

C.T.

, Klebig

M.L.

, Liu

et al. 2005. Identification of mutations from phenotype-driven ENU mutagenesis in mouse chromosome 7. Mamm. Genome, 16:555–566.

12.

Davidoff

A.M.

, Ng

C.Y.

, Zhou

et al. 2003. Sex significantly influences transduction of murine liver by recombinant adeno-associated viral vectors through an androgen-dependent pathway. Blood, 102:480–488.

13.

Davidson

P.M.

, Naidu

A.S.

2000. Phyto-phenols. Natural Food Antimicrobial Systems. Naidu

A.S.

CRC Press: Washington, DC, 265–294.

14.

De Vree

J.M.

, Ottenhoff

, Bosma

P.J.

et al. 2000. Correction of liver disease by hepatocyte transplantation in a mouse model of progressive familial intrahepatic cholestasis. Gastroenterology, 119:1720–1730.

15.

Donsante

, Miller

D.G.

, Li

et al. 2007. AAV vector integration sites in mouse hepatocellular carcinoma. Science, 317:477.

16.

Durant

, Karran

2003. Vanillins—a novel family of DNA-PK inhibitors. Nucleic Acids Res., 31:5501–5512.

17.

Fang

, Eisensmith

R.C.

, Li

X.H.

et al. 1994. Gene therapy for phenylketonuria: phenotypic correction in a genetically deficient mouse model by adenovirus-mediated hepatic gene transfer. Gene Ther., 1:247–254.

18.

Fattah

F.J.

, Lichter

N.F.

, Fattah

K.R.

et al. 2008. Ku70, an essential gene, modulates the frequency of rAAV-mediated gene targeting in human somatic cells. Proc. Natl. Acad. Sci. U. S. A., 105:8703–8708.

19.

Finn

J.D.

, Hui

, Downey

H.D.

et al. 2010. Proteasome inhibitors decrease AAV2 capsid derived peptide epitope presentation on MHC class I following transduction. Mol. Ther., 18:135–142.

20.

Garrick

, Fiering

, Martin

D.I.

, Whitelaw

1998. Repeat-induced gene silencing in mammals. Nat. Genet., 18:56–59.

21.

Grimm

2002. Production methods for gene transfer vectors based on adeno-associated virus serotypes. Methods, 28:146–157.

22.

Grompe

, Jones

S.N.

, Loulseged

, Caskey

C.T.

1992. Retroviral-mediated gene transfer of human ornithine transcarbamylase into primary hepatocytes of spf and spf-ash mice. Hum. Gene Ther., 3:35–44.

23.

Grompe

, Lindstedt

, Al-Dhalimy

et al. 1995. Pharmacological correction of neonatal lethal hepatic dysfunction in a murine model of hereditary tyrosinaemia type I. Nat. Genet., 10:453–460.

24.

, Seidl

K.J.

, Rathbun

G.A.

et al. 1997. Growth retardation and leaky SCID phenotype of Ku70-deficient mice. Immunity, 7:653–665.

25.

Hendrie

P.C.

, Russell

D.W.

2005. Gene targeting with viral vectors. Mol. Ther., 12:9–17.

26.

Hendrie

P.C.

, Hirata

R.K.

, Russell

D.W.

2003. Chromosomal integration and homologous gene targeting by replication-incompetent vectors based on the autonomous parvovirus minute virus of mice. J. Virol., 77:13136–13145.

27.

K.J.

, Bass

C.E.

, Kroemer

A.H.

et al. 2008. Optimized adeno-associated virus 8 produces hepatocyte-specific Cre-mediated recombination without toxicity or affecting liver regeneration. Am. J. Physiol. Gastrointest. Liver Physiol., 295:G412–419.

28.

Johnson

J.S.

, Samulski

R.J.

2009. Enhancement of adeno-associated virus infection by mobilizing capsids into and out of the nucleolus. J. Virol., 83:2632–2644.

29.

Kay

M.A.

, High

1999. Gene therapy for the hemophilias. Proc. Natl. Acad. Sci. U. S. A., 96:9973–9975.

30.

Keshava

, Keshava

, Ong

T.M.

, Nath

1998. Protective effect of vanillin on radiation-induced micronuclei and chromosomal aberrations in V79 cells. Mutat. Res., 397:149–159.

31.

McCann

M.J.

, Gill

C.I.

, O'Brien

et al. 2007. Anti-cancer properties of phenolics from apple waste on colon carcinogenesis in vitro. Food Chem. Toxicol., 45:1224–1230.

32.

Miller

D.G.

, Trobridge

G.D.

, Petek

L.M.

et al. 2005. Large-scale analysis of adeno-associated virus vector integration sites in normal human cells. J. Virol., 79:11434–11442.

33.

Monahan

P.E.

, Lothrop

C.D.

, Sun

et al. 2010. Proteasome inhibitors enhance gene delivery by AAV virus vectors expressing large genomes in hemophilia mouse and dog models: a strategy for broad clinical application. Mol. Ther., 18:1907–1916.

34.

Nakai

, Wu

, Fuess

et al. 2005. Large-scale molecular characterization of adeno-associated virus vector integration in mouse liver. J. Virol., 79:3606–3614.

35.

Ogura

, Mizukami

, Mimuro

et al. 2006. Utility of intraperitoneal administration as a route of AAV serotype 5 vector-mediated neonatal gene transfer. J. Gene Med., 8:990–997.

36.

Ortiz-Urda

, Lin

, Yant

S.R.

et al. 2003. Sustainable correction of junctional epidermolysis bullosa via transposon-mediated nonviral gene transfer. Gene Ther., 10:1099–1104.

37.

Overturf

, Al-Dhalimy

, Tanguay

et al. 1996. Hepatocytes corrected by gene therapy are selected in vivo in a murine model of hereditary tyrosinaemia type I. Nat. Genet., 12:266–273.

38.

Overturf

, Al-Dhalimy

, Ou

C.N.

et al. 1997. Serial transplantation reveals the stem-cell-like regenerative potential of adult mouse hepatocytes. Am. J. Pathol., 151:1273–1280.

39.

Parekh-Olmedo

, Ferrara

, Brachman

, Kmiec

E.B.

2005. Gene therapy progress and prospects: targeted gene repair. Gene Ther., 12:639–646.

40.

Paulk

N.K.

, Wursthorn

, Wang

et al. 2010. Adeno-associated virus gene repair corrects a mouse model of hereditary tyrosinemia in vivo. Hepatology, 51:1200–1208.

41.

Ponder

K.P.

, Gupta

, Leland

et al. 1991. Mouse hepatocytes migrate to liver parenchyma and function indefinitely after intrasplenic transplantation. Proc. Natl. Acad. Sci. U. S. A., 88:1217–1221.

42.

Porteus

M.H.

, Cathomen

, Weitzman

M.D.

, Baltimore

2003. Efficient gene targeting mediated by adeno-associated virus and DNA double-strand breaks. Mol. Cell Biol., 23:3558–3565.

43.

Rebuffat

, Harding

C.O.

, Ding

, Thony

2010. Comparison of adeno-associated virus pseudotype 1, 2, and 8 vectors administered by intramuscular injection in the treatment of murine phenylketonuria. Hum. Gene Ther., 21:463–477.

44.

Richardson

P.G.

, Sonneveld

, Schuster

M.W.

et al. 2005. Bortezomib or high-dose dexamethasone for relapsed multiple myeloma. N. Engl. J. Med., 352:2487–2498.

45.

Rinn

J.L.

, Snyder

2005. Sexual dimorphism in mammalian gene expression. Trends Genet., 21:298–305.

46.

Russell

D.W.

, Kay

M.A.

1999. Adeno-associated virus vectors and hematology. Blood, 94:864–874.

47.

Sasaki

Y.F.

, Ohta

, Imanishi

et al. 1990. Suppressing effects of vanillin, cinnamaldehyde, and anisaldehyde on chromosome aberrations induced by X-rays in mice. Mutat. Res., 243:299–302.

48.

Song

, Lu

, Choi

Y.K.

et al. 2004. DNA-dependent PK inhibits adeno-associated virus DNA integration. Proc. Natl. Acad. Sci. U. S. A., 101:2112–2116.

49.

van Gent

D.C.

, Hoeijmakers

J.H.

, Kanaar

2001. Chromosomal stability and the DNA double-stranded break connection. Nat. Rev. Genet., 2:196–206.

50.

Vasileva

, Linden

R.M.

, Jessberger

2006. Homologous recombination is required for AAV-mediated gene targeting. Nucleic Acids Res., 34:3345–3360.

51.

Wang

, Takabe

, Bidlingmaier

S.M.

et al. 1999. Sustained correction of bleeding disorder in hemophilia B mice by gene therapy. Proc. Natl. Acad. Sci. U. S. A., 96:3906–3910.

52.

Wang

, Montini

, Al-Dhalimy

et al. 2002. Kinetics of liver repopulation after bone marrow transplantation. Am. J. Pathol., 161:565–574.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.04 MB