Localized Gene Expression Following Administration of Adeno-associated Viral Vectors via Pancreatic Ducts

ARTICLE

doi:10.1016/j.ymthe.2005.04.017

Localized Gene Expression Following Administration of Adeno-associated Viral Vectors via Pancreatic Ducts Scott A. Loiler,1,2,3 Qiushi Tang,1,2,3 Tracy Clarke,4 Martha L. Campbell-Thompson,4 Vince Chiodo,5,6 William Hauswirth,1,2,5,6 Pedro Cruz,1,2,3 Marcel Perret-Gentil,7 Mark A. Atkinson,4 Vijayakumar K. Ramiya,3 and Terence R. Flotte1,2,3,* 1

Powell Gene Therapy Center, 2Genetics Institute, 3Department of Pediatrics, 4Department of Pathology, 5Department of Molecular Genetics, 6 Department of Ophthalmology, and 7College of Medicine, University of Florida, Gainesville, FL 32610, USA *To whom correspondence and reprint requests should be addressed at the University of Florida, Box 100266 JHMHSC, Gainesville, FL 32610-0266, USA. Fax: +1 352 392 3047. E-mail: [email protected].

Available online 23 June 2005

Gene transfer into pancreatic cells in vivo could be of immense therapeutic benefit in cases of type 1 diabetes (T1D) through the production of molecules capable of interrupting the progression of autoimmunity or promoting regeneration of insulin-secreting B cells. We adapted a clinically relevant surgical technique (endoscopic retrograde cholangiopancreatography) to deliver rAAV encoding human A1-antitrypsin (approved gene symbol SERPINA1) to the pancreas of 3-week-old Fisher 344 rats and C57BL/6 mice. We compared natural as well as bioengineered serotypes of rAAV (rAAV1, rAAV2/Apo (Burkhardt, B. R., et al. (2003). Ann. N. Y. Acad. Sci. 1005: 237–241) [1], rAAV8) as well as different promoters (chicken B-actin, human insulin) for their expression in vivo. Rats injected with rAAV1 showed the highest hAAT expression (week 2, rAAV1/CB-AT, 579 F 457 ng/ml). In mice, rAAV8 vector delivered the highest serum concentration of hAAT (week 2, rAAV8/CB-AT, 19 F 6 Mg/ml). The chicken B-actin promoter provided the highest expression in both rodent experiments. Immunohistochemical staining indicated transduction primarily of pancreatic acinar cells with either the rAAV1/CB-AT vector in the rat or the rAAV8/CB-AT vector in the mouse. This study demonstrates that rAAV vectors can be designed to deliver therapeutic genes efficiently to the pancreas and achieve high levels of gene expression and may be useful in treating pancreatic disorders, including T1D. Key Words: gene therapy, adeno-associated virus, targeting, pancreas, A1-antitrypsin

INTRODUCTION The pancreas is a vital endocrine–exocrine organ that is the site of primary pathology in many diseases [2–4]. Pancreatic cancer is responsible for 200,000 deaths per year and is the fourth leading cause of cancer death in the United States [5,6]. It is estimated that there are over 880,000 people living in the United States with type 1 diabetes diagnosed before 30 years of age [7]. Pancreatic insufficiency is a common complication in patients with cystic fibrosis [8] that requires exogenous pancreatic enzyme replacement therapy with meals. These and many other diseases may be candidates for genetic therapies that target the expression of therapeutic molecules within the pancreas. Unfortunately, there are few effective ways to deliver exogenous genes efficiently to the pancreas without causing significant unwanted side effects, the most noteworthy being pancreatitis. Adenoviral vectors

MOLECULAR THERAPY Vol. 12, No. 3, September 2005 Copyright C The American Society of Gene Therapy 1525-0016/$30.00

have been successfully used for this purpose in studies of animal models of pancreatic delivery; however, this mode of delivery often resulted in strong immune response to the vector and short transgene expression [9–11]. Adenoassociated viruses (AAV) are a relatively nonimmunogenic virus group that has been shown to infect pancreatic islets both in vitro and in vivo [1,10,12,13]. Recent work by Wang et al. indicated that rAAV8 vectors could efficiently infect cells of the mouse pancreas after a direct pancreatic injection. Wang et al. compared rAAV serotypes 2, 5, and 8 and adenovirus vector transduction of mouse pancreas in vivo and showed that direct injection of rAAV8 vector resulted in stable expression in both acinar and islet cells, whereas adenovirus expression was transient. Despite that progress, several issues still remain regarding the direct injection approach to deliver genes

519

ARTICLE

to the pancreas. One concern is the size and shape of the pancreas. The pancreas is a long and relatively narrow lobed structure, hence delivery of genes to the entire pancreas would require multiple sites of injection and, in addition, may not reach the deepest regions. Multiple injections may also increase the damage to the pancreas and the chance of inducing pancreatitis. Arterial or venous delivery to the pancreas is another option that has been tried and an approach that has seen some success [9]. A limitation of this approach is that systemic delivery of vector leads to infection of other organs along with the pancreas. Pancreatic duct (PD) injection provides a means of delivering genes efficiently throughout the pancreas. PD has successfully been used to deliver genes to the pancreas through a variety of carrier systems (e.g., cationic lipids, adenovirus) [14–16]. In this report we investigated the effectiveness of gene delivery to the pancreas through the use of PD combined with the safety of using rAAV vectors. Using pseudotyped packaging we compared three different serotypes of rAAV (rAAV1, rAAV8, rAAV2/Apo) as well as two different promoters (chicken h-actin, human insulin) to deliver and produce a

doi:10.1016/j.ymthe.2005.04.017

secreted gene product into the bloodstream of rats and mice.

RESULTS Transduction of Mouse Pancreas by rAAV In this study we used a pancreatic duct injection method to compare the transduction efficiency of various serotypes of rAAV with two different promoters in rats and mice. We injected 5  1011 particles of rAAV8/CB-ATWPRE, rAAV2Apo/CB-AT-WPRE, rAAV1/CB-AT-WPRE, or rAAV1/Ins-AT into the pancreatic duct of 4-week-old C57BL/6J mice. We confirmed pancreatic administration of vector by the flow of Evans blue dye carried with the saline/vector solution into the duct and the surrounding pancreatic tissue. We sacrificed the mice at 6 weeks postinjection (n = 3–5 per group) and evaluated major organs by immunohistochemistry. We detected serum concentrations of human a1antitrypsin (hAAT; approved gene symbol SERPINA1) as early as 1 week after injection with rAAV8/CB-AT-WPRE virus. Mean values of 7955 F 2444 ng/ml hAAT were detected at week 1, peaked at weak 2 with mean values of

FIG. 1. Mouse serum hAAT levels following pancreatic injections. The CB-AT-WPRE vector depicted in Supplemental Fig. 1 was packaged into each of three different rAAV serotype capsids (rAAV8, rAAV1, and rAAV2/ApoE), whereas the Ins-AT vector was packaged in the rAAV1 capsid only. The packaged vectors were injected into the pancreatic duct of 4-week-old C57BL/6J mice (5  1011 particles/mouse). The level of expression of hAAT as determined by ELISA on serum samples is shown on the y axis. The mean and standard deviation of each group (n = 3 to 5 animals per group) are shown at weekly intervals.

520

MOLECULAR THERAPY Vol. 12, No. 3, September 2005 Copyright C The American Society of Gene Therapy

doi:10.1016/j.ymthe.2005.04.017

19,420 F 6519 ng/ml, and plateaued by week 4 at 11,625 F 1689 ng/ml serum hAAT (Fig. 1). In contrast, both rAAV1 virus groups (CB-AT-WPRE, Ins-AT) as well as the rAAV2Apo (CB-AT-WPRE) virus group showed no significant hAAT expression in the serum. We did not anticipate the low levels of rAAV2Apo vector transduction since our earlier in vitro experiments [13] demonstrated that rAAV2Apo vectors give rise to a 220fold enhancement of vector transduction of islet cells over rAAV2 alone. Immunohistochemical staining for hAAT showed 7.3 F 6.4% of cells transduced in the pancreas of rAAV8/CB-AT-WPRE-injected mice (Fig. 2). The majority of the transduced cells appear to be acinar cells with less than 5% of ductal cells being transduced and a minimal amount of h cells transduced (Figs. 2B and 2F). The distribution of vector-transduced cells was relatively uniform throughout the pancreas (Fig. 2E and data not shown). There was no evidence of a cellular immune response observed in any of the immunohistochemical

ARTICLE

sections (Fig. 2 and data not shown). It is clear from these data that rAAV8-pseudotyped vectors were the only virus group to transduce mouse pancreas tissue efficiently, including islets and ductal cells. In addition, rAAV8transduced cells gave rise to significant levels of hAAT protein in the serum even after 6 weeks of expression. Transduction of Rat Pancreas by rAAV We also examined the ability to transduce rat pancreas via the pancreatic duct and compared different rAAV serotypes and promoters as in the mouse experiments. We injected 4  109 particles of rAAV8/CB-AT-WPRE, rAAV2Apo/CB-AT-WPRE, rAAV1/CB-AT-WPRE, or rAAV1/ Ins-AT into the pancreatic duct of 4-week-old Fischer rats. Pancreatic administration of vector was confirmed by the flow of Evans blue dye carried with the saline/ vector solution into the duct and the surrounding pancreatic tissue. We sacrificed the rats at 3 weeks postinjection (n = 3–5 per group) and evaluated most major organs by immunohistochemistry.

FIG. 2. Immunohistochemistry staining for insulin and hAAT in mouse pancreas. Sections of mouse pancreas injected with rAAV8/CB-AT-WPRE or saline at 6 weeks postinjection showing the transduced cell types. (A and B) rAAV8/CB-AT-WPRE-injected pancreas, (C and D) saline-injected pancreas. Insulin-specific staining (A and C), human a1-antitrypsin staining (B and D). Immunohistochemistry using alkaline phosphatase–vector blue with nuclear fast red counterstain. (E) Transduced acinar cells and (F) transduced islets.

MOLECULAR THERAPY Vol. 12, No. 3, September 2005 Copyright C The American Society of Gene Therapy

521

ARTICLE

doi:10.1016/j.ymthe.2005.04.017

FIG. 3. Rat serum hAAT levels and anti-hAAT levels following pancreatic injection. The CB-AT-WPRE vector depicted in Supplementary Fig. 1 was packaged into each of three different rAAV serotype capsids (rAAV8, rAAV1, and rAAV2/ApoE), whereas the Ins-AT vector was packaged in the rAAV1 capsid only. The packaged vectors were injected into the pancreatic duct of 3-week-old Fisher 344 rats (4  109 particles/rat). (A) The level of expression of hAAT as determined by ELISA on serum collected during weekly blood samples is shown on the y axis. The mean and standard deviation of each group (n = 3 to 5 animals per group) are shown at weekly intervals. (B) The level of anti-hAAT antibody expression as determined by ELISA on serum samples.

522

MOLECULAR THERAPY Vol. 12, No. 3, September 2005 Copyright C The American Society of Gene Therapy

doi:10.1016/j.ymthe.2005.04.017

Serum concentrations of hAAT were detected as early as 1 week following injection with rAAV1/CB-AT-WPRE virus. Peak values were observed at week 2, with mean values of 579 F 457 ng/ml hAAT (Fig. 3A). rAAV1/Ins-AT virus showed delayed expression kinetics compared to rAAV1/ CB-AT-WPRE virus. The peak expression for rAAV1/Ins-AT virus occurred at week 3 (254 F 199 ng/ml). In contrast, rAAV8/CB-AT-WPRE and rAAV2Apo/CB-AT-WPRE virus groups showed no significant hAAT expression in the serum. In these experiments, we sacrificed the rats at week 3 because of concerns that an immune response to hAAT would prevent the detection of transduced cells by either in situ or immunohistochemical methods. We observed anti-hAAT antibodies for rAAV8/CB-AT-WPRE, rAAV2Apo/CB-AT-WPRE, and rAAV1/CB-AT-WPRE groups (Fig. 3B). We observed no evidence of a cellular immune response in any of the immunohistochemical sections (Fig. 4 and data not shown). Immunohistochemical staining for hAAT showed limited numbers of

ARTICLE

transduced cells in the pancreas of rAAV1/CB-AT-WPREinjected rats (Fig. 4B). The majority of the transduced cells appeared to be acinar cells with a minimal amount of h cells transduced (Fig. 4B). The distribution of vector-transduced cells was relatively uniform throughout the pancreas (Fig. 4B and data not shown). Transcription from the human insulin promoter may not be limited to rat h cells, as we would have expected, as evidenced by the positive gene expression seen in some acinar cells of rAAV1/Ins-AT-injected animals (Fig. 4D), although we do see many examples of single insulinpositive cells in what appears to be acinar tissue of the rat pancreas by immunohistochemical staining (Figs. 4E and 4F). There may be some transcription from the human insulin promoter vector in non-insulin-producing cells. In addition, immunohistochemical staining for hAAT expression in rAAV1/Ins-AT-injected rats pancreas showed a unique diffuse expression of hAAT protein around the peripheral cells of most islets that was not found in the other groups (Fig. 4). These data

FIG. 4. Immunohistochemistry staining for human a1antitrypsin or insulin in rat pancreas. Sections of rat pancreas injected with (A) saline, (B) rAAV1/CB-ATWPRE, or (C and D) rAAV1/Ins-AT 3 weeks postinjection showing the different transduced cells. (E and F) Lowpower images of insulin staining of rat pancreas sections showing a number of single insulin-positive cells.

MOLECULAR THERAPY Vol. 12, No. 3, September 2005 Copyright C The American Society of Gene Therapy

523

ARTICLE

show that rAAV1-pseudotyped vectors transduce rat pancreas more efficiently than the other tested serotypes and produce measurable levels of hAAT in the serum. Evaluation of Different rAAV Serotypes We observed significant differences in the ability of distinct rAAV serotypes to transduce pancreatic tissues from the two rodent species tested. rAAV8 was very efficient at transducing mouse pancreatic cells (Fig. 2E), whereas it was very poor at transducing rat pancreatic cells (Fig. 3A). Conversely, rAAV1 was moderately effective at transducing rat pancreatic cells and very poor at transducing mouse pancreatic cells. The rAAV2/Apo-serotype virus, which is an artificially engineered virus designed to increase the transduction of pancreatic tissue, showed poor transduction efficiency in vivo in both mice and rats. This may indicate that the rAAV2Apo vectors are not in close enough proximity to transduce islets efficiently via pancreatic duct delivery. Evaluation of Different Promoters The cytomegalovirus chicken h-actin (CB) promoter exhibited much higher expression of hAAT in the serum than the insulin promoter. This is most likely due to the fact that the majority of the transduced cell types were acinar cells, in which the insulin promoter is not very active [17,18]. Very few h cells were transduced by any of the rAAV serotypes in our study and consequently little hAAT expression was observed. The insulin promoter was not tested in combination with the rAAV8 serotype in mice, in which it would be expected to give low but measurable levels of hAAT expression as evidenced by the observation from Wang et al., showing up to 4.9% of h cells could be transduced with rAAV8 vectors [10]. In the rat pancreas injections, the insulin promoter showed several differences in gene expression compared to the CB promoter. The serum levels of hAAT expression with the insulin promoter showed a delay in gene expression compared to the CB promoter (Fig. 3A). The antibody responses to hAAT were much higher with the CB promoter than with the insulin promoter (Fig. 3B).

doi:10.1016/j.ymthe.2005.04.017

Although no cellular infiltration was observed at the time the rats were sacrificed, the immune response to CB-driven hAAT-expressing cells probably led to the observed reduction in hAAT-positive staining in the immunohistochemical sections done at week 3 (Figs. 4B and 4D). Tissue Distribution of Vector Gene Expression RT-PCR analysis of vector-specific hAAT mRNA expression demonstrated that the majority of expression was contained within the pancreas (Fig. 5). There was little expression in other organs showing that the clamps prevented systemic distribution of vector, thereby containing the delivery within the pancreas. One animal in the rAAV8/CB-AT-WPRE group showed a low level of expression in the liver and in the duodenum. The spread of vector outside of the pancreas in this animal may have been due to incomplete closure of the ducts, or leakage of vector after the clamps were removed, or spread of vector through the circulatory system resulting in transduction of other organs after systemic delivery. Lack of Cellular Immune Response to Vector or Transgene Expression Immunohistochemical staining of mouse pancreas injected with either rAAV8/CB-AT-WPRE or saline showed no significant T, B, or macrophage cellular infiltration. The rAAV8/CB-AT-WPRE-injected pancreas showed equivalent levels of T, B, and macrophage cells compared to the saline-injected control. Positive control sections of spleen were included for reference.

DISCUSSION Our observations demonstrate that rAAV-mediated gene transfer to the pancreas can be efficiently achieved. Using a surgical procedure that that can be easily translated into the less invasive and commonly utilized clinical procedure endoscopic retrograde cholangiopancreatography (ERCP) in humans, we delivered rAAV vectors to the pancreas of rats and mice. In this procedure we are able to minimize the exposure of other organs to the vector. The RT-PCR data showed a minimal amount of transduction of liver

FIG. 5. RT-PCR for hAAT mRNA expression in different organs. One microgram of total RNA from the indicated tissues was subjected to RT-PCR with primers for either hAAT or GAPDH. The AAT PCR product is 152 bp in length, while the GAPDH PCR product is approximately 120 bp. Animals 38 and 46 were injected with rAAV8/CB-AT-WPRE. Animal 67 was injected with saline. Li, liver; Pa, pancreas; Sp, spleen; He, heart; Du, duodenum.

524

MOLECULAR THERAPY Vol. 12, No. 3, September 2005 Copyright C The American Society of Gene Therapy

doi:10.1016/j.ymthe.2005.04.017

and duodenum, which most likely occurred after the clamps were released, and residual virus entered those organs (Fig. 5). We also showed in rats that the insulin promoter could be successfully used when transduced gene expression in islet h cells is desired. The highest levels of hAAT expression in the mice were observed with rAAV8/CB-AT-WPRE vector. Previous work in our lab has shown that C57BL/6J mice do not mount an immune response to the hAAT transgene and are capable of sustained gene expression for more than 1 year [19,20]. We observed stable hAAT gene expression in the mice out to 6 weeks postinjection along with no immune response. Wang et al. saw a drop in LacZ gene expression sometime between 4 and 8 weeks postinjection and attributed the drop to either an immune response or acinar and h cell turnover [10]. We suspect the drop in LacZ gene expression seen by Wang et al. was due to an immune response that developed between 4 and 8 weeks postinjection [10]. We are continuing several longer studies with other transgenes and hope to have a definitive answer to the duration of acinar and h cell gene expression in the near future. It was not surprising that the transduced cells appeared to be primarily the acinar cells, with very few h cells showing hAAT expression by immunohistochemistry. Vector administration through the pancreatic duct favors exocrine cell distribution to the acinar cells with very little translocation of vector to the endocrine compartments. Because of our concerns about eliciting pancreatitis, we did not administer the vector under significant pressure, which might have aided in distribution to the endocrine cells. The minimal number of h cells that were transduced might have occurred as a result of vascular distribution of the vector after injection. Transduction of the rat pancreas was best achieved with the rAAV1/CB-AT-WPRE vector. The rAAV1/CB combination resulted in the highest early levels of hAAT expression in the serum but was short-lived, apparently due to the immune response to hAAT. Surprisingly, rAAV1-pseudotyped vector with the insulin promoter driving expression resulted in delayed expression compared to the CB promoter vector and also avoided much of the immune response that was seen with the other vectors. This is somewhat reminiscent of the findings by the Sweeney laboratory that show that intramuscular injection of vector elicits less immune response to the transgene when a muscle-specific promoter is used [21]. The rAAV/Ins-AT vector also gave a unique diffuse peripheral staining of islet cells that was not seen with any of the other vectors. The recently published work by Wang et al. on rAAV8 gene delivery utilized a direct injection approach to the pancreas [10]. We chose a more clinically relevant ERCP procedure that is commonly used to image the pancreas in human patients. One potential hazard to ERCP delivery is that the vector must overcome the harsh environment presented in the pancreatic duct. The exocrine pancreas contains high concentrations of bicar-

MOLECULAR THERAPY Vol. 12, No. 3, September 2005 Copyright C The American Society of Gene Therapy

ARTICLE

bonate ions as well as a variety of digestive enzymes like proteases and nucleases that could degrade the vector before it has a chance to reach its target [2]. Fortunately, rAAV vectors have a tightly structured outer capsid that is resistant to degradation [22,23]. Our data show that rAAV8 vectors are resistant to degradation in the harsh environment of the exocrine pancreas. Our data also show that ERCP delivery can limit the distribution of rAAV vector to the pancreas, although ERCP does not limit the biodistribution of a secreted gene product as evidenced by the significant amount of hAAT found in the serum. Another observation by Wang et al. was that rAAV8 vector gene expression was significantly reduced (~90%) by 8 weeks postinjection [10]. They suggest that pancreatic cell turnover is the most likely explanation, although one would expect to see less than a 50% reduction of transduced acinar cells and slightly more than a 50% reduction in transduced h cells at 8 weeks postinjection. We did not observe a similar reduction in gene expression at 6 weeks postinjection. Several explanations for these apparent differences may be that we simply did not monitor the experiment for the full 8week duration and that a precipitous drop in gene expression occurs after 6 weeks. A second explanation might be the development of an immune response to the h-gal transgene that was used to monitor transduced cells. We did not see any significant cellular infiltration by T, B, or macrophage cells to the a1-antitrypsin transgene in our vector-injected animals. Another explanation may be that we are transducing a different cell population that has a longer half-life when we deliver vector through the exocrine system. The therapeutic possibilities for localized gene delivery and expression in the pancreas are abundant. Pancreatic cancer is the fourth leading cause of cancer death in the United States [4]. Approximately 95% of pancreatic cancers are adenocarcinomas of the exocrine pancreas. One can envision the delivery of a rAAV vector that produces siRNA molecules that target the K-ras oncogene that is found activated in nearly all cases of pancreatic cancer. Another therapeutic approach to targeting pancreatic cancer would be the delivery of antiangiogenic proteins that limit the growth of tumors. Experiments utilizing this approach have already been tested using rAAV2 vectors with moderate success [24]. Another obvious pancreatic disease candidate for rAAV8-mediated therapeutic intervention is diabetes. Gene products that either reduce the immune response to h cells (e.g., interleukin-10) [25] or promote the growth (insulin-like growth factor-1) [26] and regeneration (glucagon-like peptide-1) [27] of h cells would be of great clinical benefit. The delivery of these and other therapeutic genes via rAAV8 vectors to the pancreas may hold great promise toward ameliorating diabetes development as well as many other diseases. A number of other issues must be addressed before this technique could be applied to human gene therapy. First

525

ARTICLE

doi:10.1016/j.ymthe.2005.04.017

among these is the issue of safety and feasibility in a larger animal model. Second, there is a significant motivation to improve delivery to the islets (e.g., to introduce antiapoptotic genes), which might require a higher infusion pressure. Nonetheless, the ability to deliver genes to the exocrine pancreas could still be of substantial therapeutic importance in promoting islet regeneration and other processes, which involves paracrine interaction and so may be worth pursuing in future preclinical studies.

MATERIALS

AND

METHODS

Animals. Four-week-old Fischer 344 rats (Harlan, Inc., Indianapolis, IN, USA) and 4-week-old C57BL/6J female mice (The Jackson Laboratory, Bar Harbor, ME, USA) were used in these experiments. Animal care and experimental procedures were approved by the Institutional Animal Care and Use Committee of the University of Florida (Gainesville, FL, USA). Surgical delivery of vector to the pancreas. A 3- to 4-cm ventral midline incision was made starting just caudal to the xyphoid process and continued on caudally. The abdomen was entered through another ventral midline through the linea alba. The duodenum was isolated with the common bile duct attached. A 29-gauge needle was used to make an opening in the duodenum and to perform a sphincterotomy on the sphincter of Oddi. A 30-gauge catheter was inserted through the duodenal opening and advanced retrogradely through the sphincter of Oddi and into the common bile duct. The catheter was secured in place with a microclamp around the bile duct and catheter to avoid leakage of the vector. Another microclamp was placed on the bile duct just caudal to the liver to prevent vector distribution to the liver. Through the catheter, either Evans blue dye or rAAV was slowly injected into the pancreatic duct with PBS in a total volume of 200 Al for the rat and 100 Al for the mouse. After approximately 2 min the microclamps and catheter were removed, and the abdomen was rinsed with warm sterile saline solution and closed with a two-layer approach. Plasmid constructs and rAAV packaging. The rAAV serotype 2 vector plasmids used for these experiments are depicted diagrammatically (Supplemental Fig. 1). Briefly, the CB promoter drives expression of the human a1-antitrypsin gene followed by the woodchuck hepatitis virus posttranscriptional regulatory element with the entire construct flanked by AAV2 inverted terminal repeats (ITRs) as previously described [28]. The human insulin promoter from plasmid pIns-IL-4 (kindly provided by Dr. Nora Sarvetnick, Department of Immunology, The Scripps Research Institute, La Jolla, CA, USA) [18] was subcloned in front of the hAAT gene with the entire construct flanked by AAV2 ITRs. The rAAV-ApoE construct was made by inserting an oligonucleotide that coded for the human ApoE amino acids LRKLRKRLLR and DWLKAFYDKVAEDLDEAF, which code for the hApoE LDL-receptor ligand and the lipid-associated peptide, respectively, immediately after amino acid 138 of the VP1 coding sequence as described previously [13], generating the plasmid pIM45Apo138. The entire cap gene of pIM45Apo138 was subcloned into the plasmid pDG8 (kindly provided by Dr. Kleinschmidt, Department of Applied Tumor Virology, German Cancer Research Center, Heidelberg, Germany) [29] generating the plasmid pDG-Apo138. rAAV1 vectors were generated and purified by previously described methods [30,31]. rAAV8 vectors were generated by subcloning the entire cap gene from the plasmid p5E18VO 2/8 (kindly provided by Dr. James Wilson, Gene Therapy Program, Division of Medical Genetics, The University of Pennsylvania School of Medicine, Philadelphia, PA, USA) [32] into the plasmid pDG, replacing the AAV2 cap gene sequence and generating the plasmid pDG8. rAAV production was performed as previously described [31]. The method involves cotransfection with two plasmids by calcium phosphate coprecipitation of a permissive human cell line (HEK293). HEK293 cells were grown as monolayers (initially seeded with 6  108 cells per Nunc

526

cell factory) in Dulbecco’s modified minimal essential medium containing 10% fetal bovine serum (378C, 5% CO2). After 18 h, the cells were transfected with different pairs of plasmids. The first nonrescuable helper plasmid (pDG or the like) contained the rAAV2 complementing functions, rep and cap, as well as the Ad helper genes (E2a, VA RNA, E4) required for helper function. The second vector contained a eukaryotic expression cassette and flanking ITRs. Transfected cells were maintained at 378C in culture (5% CO2) for 60 h before harvest. Cells were then dissociated by treatment with EDTA, pelleted, resuspended in lysis buffer (20 mmol/L Tris, pH 8.0; 150 mmol/L NaCl; 5% deoxycholate) containing benzonase (Nycomed Pharma) and incubated for 30 min (378C, 5% CO2). Crude lysates were clarified by centrifugation with virus-containing supernatant purified by iodixanol density gradient centrifugation, followed by anion-exchange (Q-Sepharose) chromatography and concentration as described previously [31]. Purity of preparations was determined by subjection of a sample to silver-stained SDS–PAGE. Dot-blot assays were performed to quantify the titer of the rAAV physical particles and then confirmed by real-time PCR. a1-Antitrypsin measurement in serum. The level of hAAT was determined by enzyme-linked immunoassay (ELISA). Microtiter plates (Immulon 4; Dynex Technologies, Chantilly, VA, USA) were coated with 100 Al of goat anti-hAAT (1:200 diluted; Sigma Immunochemical Co., St. Louis, MO, USA) in Voller’s buffer overnight at 48C. Duplicate standard curves (hAAT; Sigma Immunochemical) and serially diluted unknown samples were incubated in the plate at 378C for 1 h. After blocking with 3% bovine serum albumin (BSA), a second antibody, rabbit anti-hAAT (1:1000 diluted, Roche Molecular Biochemicals, Indianapolis, IN, USA), was reacted with the captured antigen at 378C for 1 h. A third antibody, goat anti-rabbit IgG conjugated with peroxidase (1:800 diluted; Roche Molecular Biochemicals), was incubated at 378C for 1 h. The plate was washed with PBS–Tween 20 between reactions. After reaction with the substrate (o-phenylenediamine dihydrochloride; Sigma Immunochemical), plates were read at 490 nm on an MRX microplate reader (Dynex Technologies). a1-Antitrypsin mRNA expression in tissues. The levels of hAAT mRNA were measured by reverse transcription-PCR using primers that flanked an intron/exon boundary to determine levels of spliced hAAT mRNA. The primers used for RT-PCR were AATforward, 5V-TCTGACTGACCGCGTTACTCCCACAG; AATreverse, 5V-AGAAGACGGCATTGTCGATT; GAPDHforward, 5V-CCACTACATGGTCTACATGTTCCAG; GAPDHreverse, 5VCTCAGCCTTGACCGTGCC. The reactions were analyzed by agarose gel electrophoresis and scanned by a Bio-Rad Quantity-One (v. 4.3.1) gel documentation system. Anti-human-a1-antitrypsin antibody detection. The level of anti-hAAT was determined by ELISA. Microtiter plates (Immulon 4; Dynex Technologies) were coated with 1 Ag/well of hAAT (Sigma Immunochemical) in Voller’s buffer overnight at 48C. Duplicate standard curves (mouse antihAAT; ICN) and serially diluted unknown samples were incubated in the plate at 378C for 1 h. After blocking with 3% BSA, a second antibody, goat anti-rat IgG (H+L) conjugated with peroxidase (1:5000 diluted, Roche Molecular Biochemicals), was reacted with the captured antigen at 378C for 1 h. The plate was washed with PBS–Tween 20 between reactions. After reaction with the substrate (o-phenylenediamine dihydrochloride, Sigma Immunochemical), plates were read at 490 nm on an MRX microplate reader (Dynex Technologies). Statistical analysis. ELISA data from the transduction experiments are represented as means F SD. ANOVA was used to compare the means in the different groups and Student–Newman–Keuls multiple comparisons test was performed. Data were considered significant at P b 0.05.

ACKNOWLEDGMENTS These studies were supported by grants from the Juvenile Diabetes Research Foundation (to M.A. and T.R.F.), by the NIH (DK58327, HL51811, and HL59412), and by generous support from the Powell Endowment and Shands Hospital. Some of the authors (S.L., M.A., T.R.F.) may be entitled to patent royalties for technology described herein.

MOLECULAR THERAPY Vol. 12, No. 3, September 2005 Copyright C The American Society of Gene Therapy

doi:10.1016/j.ymthe.2005.04.017

RECEIVED FOR PUBLICATION NOVEMBER 11, 2004; ACCEPTED APRIL 12, 2005.

APPENDIX A. SUPPLEMENTARY DATA Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ymthe.2005. 04.017. REFERENCES 1. Burkhardt, B. R., et al. (2003). Glucose-responsive expression of the human insulin promoter in HepG2 human hepatoma cells. Ann. N.Y. Acad. Sci. 1005: 237 – 241. 2. Edlund, H. (2002). Pancreatic organogenesis—developmental mechanisms and implications for therapy. Nat. Rev. Genet. 3: 524 – 532. 3. Homo-Delarche, F., and Drexhage, H. A. (2004). Immune cells, pancreas development, regeneration and type 1 diabetes. Trends Immunol. 25: 222 – 229. 4. Johansson, K. A., and Grapin-Botton, A. (2002). Development and diseases of the pancreas. Clin. Genet. 62: 14 – 23. 5. Lowenfels, A. B., and Maisonneuve, P. (2004). Epidemiology and prevention of pancreatic cancer. Jpn. J. Clin. Oncol. 34: 238 – 244. 6. Michaud, D. S. (2004). Epidemiology of pancreatic cancer. Minerva Chir. 59: 99 – 111. 7. Roy, M. S., Klein, R., O’Colmain, B. J., Klein, B. E., Moss, S. E., and Kempen, J. H. (2004). The prevalence of diabetic retinopathy among adult type 1 diabetic persons in the United States. Arch. Ophthalmol. 122: 546 – 551. 8. Virgilis, D. [Cystic fibrosis from the exocrine pancreatic point of view]. Harefuah 143: 67–72, 83. 9. Ayuso, E., et al. (2004). In vivo gene transfer to pancreatic beta cells by systemic delivery of adenoviral vectors. Hum. Gene Ther. 15: 805 – 812. 10. Wang, A. Y., Peng, P. D., Ehrhardt, A., Storm, T. A., and Kay, M. A. (2004). Comparison of adenoviral and adeno-associated viral vectors for pancreatic gene delivery in vivo. Hum. Gene Ther. 15: 405 – 413. 11. Taniguchi, H., Yamato, E., Tashiro, F., Ikegami, H., Ogihara, T., and Miyazaki, J. (2003). Beta-cell neogenesis induced by adenovirus-mediated gene delivery of transcription factor pdx-1 into mouse pancreas. Gene Ther. 10: 15 – 23. 12. Flotte, T., et al. (2001). Efficient ex vivo transduction of pancreatic islet cells with recombinant adeno-associated virus vectors. Diabetes 50: 515 – 520. 13. Loiler, S. A., et al. (2003). Targeting recombinant adeno-associated virus vectors to enhance gene transfer to pancreatic islets and liver. Gene Ther. 10: 1551 – 1558. 14. Schmid, R. M., et al. (1994). Liposome mediated in vivo gene transfer into different tissues of the gastrointestinal tract. Z. Gastroenterol. 32: 665 – 670. 15. Yang, Y., Raper, S. E., Cohn, J. A., Engelhardt, J. F., and Wilson, J. M. (1993). An approach for treating the hepatobiliary disease of cystic fibrosis by somatic gene transfer. Proc. Natl. Acad. Sci. USA 90: 4601 – 4605.

MOLECULAR THERAPY Vol. 12, No. 3, September 2005 Copyright C The American Society of Gene Therapy

View publication stats

ARTICLE

16. Raper, S. E., and DeMatteo, R. P. (1996). Adenovirus-mediated in vivo gene transfer and expression in normal rat pancreas. Pancreas 12: 401 – 410. 17. Walker, M. D., Park, C. W., Rosen, A., and Aronheim, A. (1990). A cDNA from a mouse pancreatic beta cell encoding a putative transcription factor of the insulin gene. Nucleic Acids Res. 18: 1159 – 1166. 18. Mueller, R., Krahl, T., and Sarvetnick, N. (1996). Pancreatic expression of interleukin-4 abrogates insulitis and autoimmune diabetes in nonobese diabetic (NOD) mice. J. Exp. Med. 184: 1093 – 1099. 19. Song, S., et al. (1998). Sustained secretion of human alpha-1-antitrypsin from murine muscle transduced with adeno-associated virus vectors. Proc. Natl. Acad. Sci. USA 95: 14384 – 14388. 20. Xu, L., et al. (2001). CMV–beta-actin promoter directs higher expression from an adeno-associated viral vector in the liver than the cytomegalovirus or elongation factor 1 alpha promoter and results in therapeutic levels of human factor X in mice. Hum. Gene Ther. 12: 563 – 573. 21. Cordier, L., et al. (2001). Muscle-specific promoters may be necessary for adenoassociated virus-mediated gene transfer in the treatment of muscular dystrophies. Hum. Gene Ther. 12: 205 – 215. 22. Zolotarskaia, E. E., Dukhovnaia, E. M., and Dreizin, R. S. (1975). Study of the possibilities of inactivating adeno-associated virus type 4. Vopr. Virusol. 3: 341 – 344. 23. Johnson, F. B., and Bodily, A. S. (1975). Effect of environmental pH on adenovirusassociated virus. Proc. Soc. Exp. Biol. Med. 150: 585 – 590. 24. Noro, T., et al. (2004). Adeno-associated viral vector-mediated expression of endostatin inhibits tumor growth and metastasis in an orthotropic pancreatic cancer model in hamsters. Cancer Res. 64: 7486 – 7490. 25. Goudy, K., et al. (2001). Adeno-associated virus vector-mediated IL-10 gene delivery prevents type 1 diabetes in NOD mice. Proc. Natl. Acad. Sci. USA 98: 13913 – 13918. 26. George, M., Ayuso, E., Casellas, A., Costa, C., Devedjian, J. C., and Bosch, F. (2002). Beta cell expression of IGF-I leads to recovery from type 1 diabetes. J. Clin. Invest. 109: 1153 – 1163. 27. Brubaker, P. L., and Drucker, D. J. (2004). Glucagon-like peptides regulate cell proliferation and apoptosis in the pancreas, gut, and central nervous system. Endocrinology 145: 2653 – 2659. 28. Loiler, S. A., and Flotte, T. (2003). Alternative capsid strategies for targeting recombinant AAV gene therapy. Gene Ther. Regul. 2: 139 – 148. 29. Grimm, D., Kern, A., Rittner, K., and Kleinschmidt, J. A. (1998). Novel tools for production and purification of recombinant adenoassociated virus vectors. Hum. Gene Ther. 9: 2745 – 2760. 30. Zolotukhin, S., et al. (1999). Recombinant adeno-associated virus purification using novel methods improves infectious titer and yield. Gene Ther. 6: 973 – 985. 31. Zolotukhin, S., et al. (2002). Production and purification of serotype 1, 2, and 5 recombinant adeno-associated viral vectors. Methods 28: 158 – 167. 32. Gao, G. P., Alvira, M. R., Wang, L., Calcedo, R., Johnston, J., and Wilson, J. M. (2002). Novel adeno-associated viruses from rhesus monkeys as vectors for human gene therapy. Proc. Natl. Acad. Sci. USA 99: 11854 – 11859.

527