Journal Logo

BRIEF COMMUNICATIONS

DOUBLE GENETIC MODIFICATION OF ADENOVIRUS FIBER WITH RGD POLYLYSINE MOTIFS SIGNIFICANTLY ENHANCES GENE TRANSFER TO ISOLATED HUMAN PANCREATIC ISLETS1

Contreras, Juan L.2,4; Wu, Hongju3; Smyth, Cheryl A.2; Eckstein, Christopher P.2; Young, Carlton J.2; Seki, Toshiro3; Bilbao, Guadalupe2; Curiel, David T.3; Eckhoff, Devin E.2

Author Information
doi: 10.1097/01.TP.0000066361.02042.CA
  • Free

Abstract

Pancreatic islet transplantation (PIT) has been validated as a treatment for human type 1 diabetes mellitus (1). However, in the most successful clinical experience in PIT, two or more islet infusions were necessary to achieve sufficient engrafted islet mass for establishing normoglycemia without exogenous insulin therapy (2). Thus, new strategies for improving durable functional islet mass will be instrumental in facilitating PIT as a cure for type 1 diabetes.

The ability to transfer immunoregulatory, cytoprotective, or antiapoptotic genes into pancreatic islets may enhance posttransplant survival of islet allografts or xenografts. In this regard, gene transfer has been successfully achieved in isolated pancreatic islets with nonviral and viral vectors (3–15). Nonviral strategies, including lipofection, electroporation, and biolistic particles, lead to low transfection efficiency and transgene expression. Alternatively, viral vectors often mediate highly efficient gene transfer. Adenoviral vectors (Ad5) are the most efficient vectors currently available for transducing nondividing cells. The epichromosomal location diminishes the risk of insertional mutagenesis, and they can be produced in high titers for clinical studies. Limitations associated with the use of Ad5 for gene therapy are related to the reliance of the virus on the presence of its primary receptor, the transient nature of the transgene expression, and the immediate inflammatory and immune response elicited by the infection to the host cell (16).

Advances in the tropism modification of adenoviral vectors are based on the understanding of adenoviral infection biology. Primary attachment of the virus is mediated by the knob region of the fiber binding to the coxsackievirus and adenovirus (CAR) receptor (17), whereas internalization is mediated by a secondary interaction between integrins on the target cell and an arginine-glycine-aspartame (RGD) motif located on the viral penton base (18). Low expression of the CAR receptor has been demonstrated in several cells including endothelial cells. Previous reports have shown that tropism of the adenovirus can be altered using retargeting strategies (19–23). To this end, incorporation of peptide motifs into the knob domain of fiber enhances gene delivery by Ad5 vectors. Two locales in this region, the HI loop and the Cterminal end, have been shown to be feasible for incorporating extra cell-binding motifs to achieve CAR-independent gene delivery. Among the motifs that have been incorporated into these locales are RGD and polylysine (pK7) peptides (23). These two motifs are particularly interesting because their receptors, integrins for RGD peptide and cell-surface proteins containing polyanion motifs such as heparan sulfate for the positively charged pK7 peptide, are widely expressed in different cell types. Therefore, retargeting adenovirus by RGD and pK7 may have potential therapeutic advantages.

Because RGD and pK7 have different receptors and can independently mediate gene delivery (23), we hypothesized that they could act additively to improve infectivity of Ad5 and reduce toxicity to isolated human pancreatic islets. To test this hypothesis, we constructed a double-modified Ad5 vector containing an RGD motif in the HI loop and a pK7 motif at the C-terminus of fiber (Ad5RGDpK7), and characterized its transfection efficiency and toxicity on isolated human pancreatic islets (IHPI). In this report, we discovered that, compared with unmodified and singly-modified Ad5 vectors, Ad5RGDpK7 exhibited higher transfection efficiency and reduced toxicity. Furthermore, we found that reduction in the viral dose required to infect more than 80% of the islet cells significantly reduced the immediate inflammatory and immune response related to Ad5 infection. This strategy promises to overcome toxicity and inflammation of Ad5 vectors to successfully modify isolated pancreatic islets.

MATERIALS AND METHODS

Islet Isolation

Pancreata were recovered from cadaveric donors with informed consent after in situ vascular perfusion using University of Wisconsin (UW) solution at 4°C (39.2°F) as part of a multiorgan procurement. Pancreata were immediately transported to the islet isolation laboratory for processing. Islets were isolated by a semiautomated method and purified using the Cobe 2991 (Cobe Blood Component Technology, Inc., Lakewood, CO) as described (2). The number of islets within each size class was converted to the standard number of islets of 150 μm diameter equal in volume to the sample. Purity was assessed by comparing the relative quantity of dithizone-stained (Sigma, St. Louis, MO) endocrine tissue with unstained exocrine tissue. Only islet isolations with greater than 90% viability and greater than 40% purity were used. All the experiments were performed on the islets derived from a single donor in triplicate and repeated on at least four occasions on islets from subsequent donors. For immunologic studies, islets were isolated from normal male C57BL/6 mice (The Jackson Laboratory, Bar Harbor, ME) as previously described (24). Islets were separated with density gradients in Histopaque (Sigma). After isolation, human and murine islets were hand-picked using a siliconized pipette. Human islets were placed in tissue culture plates containing CMRL 1066 (Gibco, Grand Island, NY) containing 10% fetal bovine serum (FBS) (Gibco), 25 mM of HEPES (Gibco), and 100 U/mL of penicillin and 100 mg/mL of streptomycin (Gibco) at 22°C (71.6°F) in a humidified atmosphere of 95% air and 5% carbon dioxide. Murine islets were placed in tissue culture plates containing RPMI 1640 containing 10% FBS, 100 U/mL of penicillin, and 100 mg/ml of streptomycin (Gibco) at 37°C (98.6°F) in a humidified atmosphere of 95% air and 5% carbon dioxide.

Animals

Human islet recipients were 9- to 12-week-old, male, nonobese diabetic (NOD), severe combined immunodeficiency (SCID) mice (The Jackson Laboratory). To study the immunogenicity of the Ad5 vectors, a series of islet autotransplants were performed in C57BL/6 mice. All animals were fed a standard laboratory diet with water and food ad libitum. Surgical and nonsurgical procedures were carried out in accordance with the National Institutes of Health Guide forthe Care and Use of Laboratory Animals under supervision of the Institutional Animal Care and Use Committee. Type 1 diabetes mellitus was induced in NOD-SCID mice or C57BL/6 by a single intraperitoneal injection of freshly reconstituted streptozotocin (STZ) (Sigma) 200 mg/kg. Blood glucose levels were obtained from tail vein samples by commercial strips (Roche Accu-Check III, Roche Diagnostics, Basel, Switzerland). Type 1 diabetes mellitus was defined as a nonfasting glucose level greater than 300 mg/dL for at least 3 consecutive days. Euglycemia was defined as nonfasting blood glucose less than 200 mg/dL.

Generation of Recombinant Adenovirus Vectors

The following vectors were generated in our laboratory: an Ad5 vector containing a heterologous RGD motif in the HI loop of the fiber knob (Ad5RGD), an Ad5 vector containing a pK7 motif at the Cterminus of fiber (Ad5pK7), and a double-modified Ad5 vector containing RGD and pK7 motifs (ad5RGDpK7). The RGD peptide CDCRGDCFC that we chose is known to bind several types of mammalian integrins with high affinity. The pK7 motif contains five Gly-Ser repeats that act as a spacer to ensure the right positioning of the lysine residues so that they will have access to cell surfaces (23). All vectors contain two reporter genes, green fluorescent protein (GFP) and firefly Luciferase (Luc) gene, in the E1 region of the Ad5. All viruses were rescued in 293 cells that express the complementary E1 region for adenovirus growth and purified with two rounds of equilibrium centrifugation in CsCl2 gradients. The virus particle titers were determined using a conversion factor of 1.1×1012 viral particles (VP) per absorbance unit at 260 nm. The titers of these viruses are all in the range of 1.5 to 3.0×1012 VP/mL (23).

Gene Transfer

Twelve hours after isolation, pancreatic islets were washed twice in CMRL 1066 supplemented with less than 2% of FBS. Groups of 100 or 200 islet equivalents (IEQ) were infected in 24-well tissue plates at different VP in a minimum volume of 0.5 mL at 37°C. After 2 hr, CMRL 1066 supplemented with 10% FBS was added and cultured overnight at 26°C (78.8°F) and 5% carbon dioxide. Twelve hours after the infection, the islets were washed (three times) with CMRL 1066 medium and cultured for an additional 12 hr before being tested for further experiments. Uninfected (mock) islets were processed in parallel.

Transfection Efficiency

Transfection efficiency was analyzed by GFP detection after islet infection with Ad5, Ad5RGD, Ad5pK7, and Ad5RGDpK7. Twenty-four hours after gene transfer, islets were dissociated into single cells by incubation with 0.25% EDTA-trypsin (Sigma) for 17 min followed by syringe injection through progressively narrower gauge needles sized from 16 to 22 as described (25). A sample was taken and GFP fluorescence was visualized microscopically under standard excitation-emission parameters. Approximately 2 to 5×104 cells were counted per experiment and the results expressed as a mean±SD of the percentage of positive cells.

Analysis of the Transgene Expression

Analysis of Luc expression after infection with Ad5, Ad5RGD, Ad5pK7, or Ad5RGDpK7 was performed using a kit according to the manufacturer's directions (Promega Luciferase Assay System; Promega Corporation, Madison, WI). Briefly, islets were lysed with 400 μL of Promega 1X cell lysis reagent. Samples were incubated on ice for 1 hr, after which lysates were collected and centrifuged at 14,000 rpm for 5 min at 4°C. Next, 20 μL of supernatant was added to 100 μL of Promega Luciferase Substrate and the emitted light was analyzed in a Lumat LB 9510 luminometer (Berthold Detection Systems, Gaithersburg, MD). Results are expressed as mean±SD of reading of light units (RLU) per milligram of total protein per second.

Islet Apoptosis and Caspase 3 Activation

Twenty-four hours after infection, the percentage of apoptotic islet cells was assessed using a quantitative assay (TiterTACS, Trevigen, Gaithersburg, MD) following the manufacturer's instructions. The results are expressed as mean±SD of the percentage of apoptotic cells. Similar results were obtained by annexin V binding (data not shown). The colorimetric assay ApoAlert Caspase 3 (Clontech Laboratories, Inc., Palo Alto, CA) was used to measure caspase 3 activities following the manufacturer's instructions. The protein content was determined using Bio-Rad Protein Assay kit I (Bio-Rad, Hercules, CA). Results are expressed as mean±SD of the relative units of caspase 3 activity per milligram of total protein.

Islet Transplantation and Metabolic Studies

After genetic modification, 2,000 human IEQ were transplanted per mouse into the portal vein as previously described (4). One week after the transplant, a glucose disposal rate was calculated from an intraperitoneal glucose tolerance test as described. Briefly, after an overnight fast, mice were injected intraperitoneally with glucose (1.0 g/kg body weight). Blood samples were taken at various time points (0-60 min). Glucose disposal rate was calculated as the slope of the natural log of glucose values between 10 and 30 min after injection. For immunologic studies, 300 IEQ genetically modified murine islets were infused into the portal vein of syngeneic STZ-induced diabetic recipients. Nonfasting glucose levels less than 200 mg/dL confirmed the functionality of the islets after the transplant.

Nuclear Factor-κB Nuclear Translocation

To examine the effects of adenoviral infection on nuclear factor (NF)-κB nuclear translocation, intact human islets were transiently transfected 12 hr after the isolation with a luciferase reporter gene fused to five tandemly arrayed NF-κB consensus binding sites (Stratagene, La Jolla, CA) using LipofectAMINE (Life Technologies, Inc., Gaithersburg, MD) following the manufacturer's instructions. The plasmid pFC-MEKK-Luc was used as positive control and pGAS-Luc as negative control (Strategene). Then, the islets were infected with Ad5, Ad5RGD, Ad5pK7, or Ad5RGDpK7 as previously described. Twenty-four hours after the infection, islets were assessed for Luc expression using a commercial kit (Promega). Results are expressed as mean±SD of RLU per milligram of total protein per second.

Induction of Chemokines after Islet Infection with Adenoviral Vectors

Twenty-four hours after islet infection, culture supernates were collected and stored at −70°C (−158°F). Human RANTES (regulated upon activation, normal T-cell expressed and presumably secreted) was determined by enzyme-linked immunosorbent assay (ELISA) (Quantikine Human RANTES, R&D Systems, Inc., Minneapolis, MN). Results are expressed as mean±SD of the concentration of human RANTES (in picograms per milliliter).

Adenovirus-Driven Specific T-Helper Type 1 Response

Islets were isolated from normal C57BL/6 mice and infected with adenoviral vectors as previously described. Twenty-four hours after infection, 300 IEQ were infused into the portal vein of STZ-induced diabetic C57BL/6 mice. Twenty-one days after the transplant, spleens were collected, processed into single cell suspension, and cultured at a concentration of 1×106/mL, in 200-μL culture volumes with either Ad5 or Ad5RGDpK7 at 500 VP per cell as described (26). Culture supernatants were collected after 72 hr and stored at −70°C. Interferon (IFN)-γ was determined by ELISA (Quantikine M, mouse IFN-γ Immunoassay, R&D Systems). Results are expressed as mean±SD of the concentration of IFN-γ (in picograms per milliliter).

Adenovirus-Specific Antibody Response

Antibody responses against adenovirus vectors after islet transplantation were measured in the sera of mice by ELISA as described (26). Briefly, 109 VP of Ad5, Ad5RGD, Ad5pK7, and Ad5RGDpK7 were immobilized on wells of a 96-well plate (Nunc Maxisorp, Nalgene Nunc International, Rochester, NY) by overnight incubation at 4°C. After extensive washes with phosphate-buffered saline and blocking with 2% bovine serum albumin and 0.05% Tween-20 for 2 hr, wells were incubated for 2 hr with dilutions of serum samples obtained from the islet autotransplant recipients described above. Plates were washed and incubated with biotinylated, isotype-specific sheep antibodies against mouse (The Binding Site, San Diego, CA) for 1 hr. Wells then were washed and incubated with peroxidaseconjugated avidin (Sigma) for 30 min and developed with 2,2′-azinto-bis(3-ethylbenzthiazoline-6-sulfonic acid. Absorbance was determined at 405 nm.

Statistical Analysis

A double-sided t test was used for statistical analysis. The level of significance was at P<0.05.

RESULTS

Double Modification of Adenovirus Fiber with RGD and Polylysine Improved Transfection Efficiency to Intact Isolated Human Pancreatic Islets

Initial studies reported by our group using retargeted adenovirus demonstrated enhanced transfection efficiency, compared with standard Ad5 vectors in nonhuman primate islets (25). Following this strategy to alter the tropism of Ad5 vectors, we constructed an Ad5 vector with an RGD motif in the HI loop and a polylysine (pK7) motif at the C-terminal end of fiber by incorporating the oligonucleotides encoding the two motifs into the corresponding regions of the fiber gene (Fig. 1). The ability of Ad5 vectors to infect IHPI, assessed by the expression of GFP and Luc, is demonstrated in Figure 2. At 0.1 VP/cell, greater than 80% of the islet cells were infected with Ad5RGDpK7. In contrast, only 5% to 20% of the cells were transfected with standard Ad5, single-modified Ad5RGD, or Ad5pK7 (Fig. 2A). These results were confirmed with the analysis of GFP expression in intact islets after the infection with Ad5 or Ad5RGDpK7 (Fig. 2B). Higher Luc expression was demonstrated after islet infection with Ad5RGDpk7 at all viral concentrations (Fig. 2C). These results demonstrate that double modification of the Ad5 fiber with RGD and pK7 significantly increases the transfection efficiency of Ad5 vectors into intact human islets.

F1-44
Figure 1. Fiber proteins in single- and double-modified Ad5 vectors. An Ad5 vector containing a heterologous RGD motif in the HI loop of the fiber knob (Ad5RGD), an Ad5 vector containing a pK7 motif at the C-terminus of fiber (Ad5pK7), and a double-modified Ad5 vector containing RGD and pK7 motifs (ad5RGDpK7) were developed in our laboratory. The RGD peptide CDCRGDCFC that we chose is known to bind several types of mammalian integrins with high affinity. The pK7 motif contains five Gly-Ser repeats that act as a spacer to ensure the proper positioning of the lysine residues for access to cell surfaces. All these vectors contain two reporter genes,:
GFP and firefly Luc gene, in the E1 region of the Ad5.
F2-44
Figure 2. Analysis of the transfection efficiency of Ad5 vectors in intact, isolated human pancreatic islets. Gene transfer of pancreatic islets was performed (see:
Materials and Methods ). Twenty-four hours after the infection, islets were dissociated into single cells and visualized microscopically for GFP fluorescence under standard excitationemission parameters (A). Approximately 2 to 5 × 104 cells were counted for experiment, and the results are expressed as mean ±SD of the percentage of GFP -positive cells. (B) Microscopic visualization of intact islets 24 hr after infection with Ad5 or AdRGDpK7. (C) Luc expression 24 hr after infection with Ad5 vectors using a commercial kit (see Materials and Methods ). Results are expressed as a mean±SD of RLU per milligram of total protein per second.

Assessment of Islet Cell Apoptosis after Infection with Adenoviral Vectors

Previous studies demonstrated that the infection of isolated pancreatic islets with a high adenoviral dose significantly increased islet apoptosis (9). In this regard, IHPI were infected with Ad5, Ad5RGD, Ad5pK7, and Ad5RGDpK7 at different concentrations (Fig. 3A). A dose-dependent increase in islet apoptosis was demonstrated after the infection with all of the vectors tested. However, a higher incidence of apoptosis was observed in islets infected with Ad5RGDpK7 at 10, 100, and 500 VP/cell. Because Ad5RGDpK7 significantly increased the percentage of infected islet cells compared with the other vectors (Fig. 2), we analyzed the incidence of apoptosis using the minimal viral concentration required to infect more than 80% of the islet cells (i.e., Ad5, 500 VP/cell; Ad5RGD and Ad5pK7, 10 VP/cell; Ad5RGDpK7, 0.1 VP/cell). A higher percentage of apoptotic islet cells was demonstrated after the infection with Ad5 (16.2±1.8) compared with Ad5RGDpK7 (7.82±0.81, P<0.05) (Fig. 3B). Similar results were observed after the evaluation of caspase 3 activity after infection with adenoviral vectors (Fig. 3C and D). Overall, these results demonstrate that reducing the viral dose required to infect the majority of the islet cells with the use of a double-modified Ad5 vector was associated with a reduction in the incidence of apoptosis after gene transfer.

F3-44
Figure 3. Analysis of activation of caspase 3 and apoptosis after infection with Ad5 vectors. (A) Intact human islets were infected (see:
Materials and Methods ). Twenty-four hours after infection, the percentage of apoptotic islet cells was assessed using a quantitative assay (TiterTACS) following the manufacturer's instructions. The results are expressed as mean±SD of the percentage of apoptotic cells. (B) Analysis of the percentage of apoptotic islet cells after infection with Ad5, Ad5RGD, Ad5pK7, or Ad5RGDpK7 at a dose required to infect greater than 80% of the islet cells (i.e., Ad5, 500 VP/cell; Ad5RGD and Ad5pK7, 10 VP/cell; Ad5RGDpK7, 0.1 VP/cell). (C) The colorimetric assay ApoAlert Caspase 3 was used to measure caspase 3 activities after infection with Ad5, Ad5RGD, Ad5pK7, or Ad5RGDpK7 at different viral concentrations. Results are expressed as mean±SD of the relative units of caspase 3 activity per milligram of total protein. (D) Caspase 3 activation after infection of 80% of the islet cells using a minimal viral dose (i.e., Ad5, 500 VP/cell; Ad5RGD and Ad5pK7, 10 VP/cell; Ad5RGDpK7, 0.1 VP/cell).

Functional Analysis of Genetically Modified Human Islets In Vivo after Transplantation

The demonstration of normal cell function after genetic modification is a prime requisite for a successful gene therapy. In this regard, IHPI were infected with modified and unmodified adenoviral vectors at different concentrations and then transplanted into STZ-induced NOD-SCID mice (Fig. 4A). After intraportal infusion of 2,000 IEQ/mouse, all the animals achieved nonfasting glucose levels less than 200 mg/dL (data not shown). However, a dose-dependent reduction in glucose disposal rate (Kg) was demonstrated after the infection with adenoviral vectors. Lower Kg was observed in animals infected with Ad5RGDpK7. However, after the analysis of the Kg in all recipients after the infection of greater than 80% of the islet cells using a minimal viral dose, we observed significantly higher Kg in animals that received islets infected with Ad5RGDpK7, compared with other vectors (Fig. 4B). These results demonstrate that targeting the adenoviral vector with RGD and pK7 allows a significant reduction in viral dose required to infect greater than 80% of the islet cells, an effect associated with better functionality after transplantation.

F4-44
Figure 4. Functional analysis of genetically modified human islets in vivo after transplantation. (A) Twenty-four hours after the infection with Ad5, Ad5RGD, Ad5pK7, or Ad5RGDpK7 at different viral concentrations, each STZ-induced NOD-SCID mouse received 2,000 human IEQ into the portal vein. Eight days after the transplant, glucose disposal rate (Kg) was calculated from intraperitoneal glucose tolerance testing (see:
Materials and Methods ). Results are expressed as mean±SD. (B) Analysis of the Kg in recipients that received IHPI infected with the minimal adenoviral dose required to infect greater than 80% of the islet cell (i.e., Ad5, 500 VP/cell; Ad5RGD and Ad5pK7, 10 VP/cell; Ad5RGDpK7, 0.1 VP/cell).

Analysis of Adenovirus Vector-Induced Acute Inflammation in Transduced Human Islets

Adenovirus vectors are increasingly being recognized as inducers of acute inflammation of transduced tissues. In contrast to cell-mediated responses, acute inflammation occurs within 24 hr of infection. Bowen et al. found that the C-C chemokine RANTES is rapidly induced after Ad5 infection and is dependent on NF-κB activation (27). This chemokine not only contributes to the acute inflammatory process but also represents an important step in the development of host antiviral immunity (28, 29). To determine the activation of inflammation after infection with Ad5 vectors, we infected IHPI with different doses of Ad5, Ad5RGD, Ad5Pk7, and Ad5RGDpK7. A concentration-dependent increase in NF-κB nuclear translocation (Fig. 5A) and RANTES production (Fig. 5C) was observed after infection with all adenoviral vectors, being higher after the infection with Ad5RGDpK7. However, a significant reduction in nuclear translocation of NF-κB was demonstrated in human islets infected with Ad5RGDpK7, compared with single-modified Ad5 vectors (Ad5RGD or Ad5pK7) and nonmodified Ad5, when greater than 80% of the cells were infected with a minimal viral concentration (Fig. 5B). In parallel with a reduction in NF-κB activation, reduction of viral dose was associated with lower concentration of RANTES in culture supernatants after infection with the double-modified Ad5 (Fig. 5D).

F5-44
Figure 5. Analysis of NF-κB nuclear translocation and RANTES expression after gene transfer to human islets with adenoviral vectors. To examine the effects of adenoviral infection on NF-κB nuclear translocation (A and B), intact human islets were transiently transfected 12 hr after the isolation with a luciferase reporter gene fused to five tandemly arrayed NF-κB consensus-binding sites using LipofectAMINE following the manufacturer's instructions. The plasmid pFC-MEKK-Luc was used as positive control and pGAS-Luc as negative control. Then, the islets were infected with adenoviral vectors at different doses (A) or at the dose required to infect greater than 80% of the islet cells (i.e., Ad5, 500 VP/cell; Ad5RGD and Ad5pK7, 10 VP/cell; Ad5RGDpK7, 0.1 VP/cell). (B) Twenty-four hours after the infection, islets were assessed for:
Luc expression using a commercial kit. Results are expressed as mean±SD of RLU per milligram of total protein per second. Human RANTES was determined by ELISA in islet culture supernates 24 hr after infection with Ad5 or Ad5RGDpK7 at different doses (C) or at a dose required to infect 80% of the cells (i.e., Ad5, 500 VP/cell; Ad5RGDpK7, 0.1 VP/cell) (D). Results are expressed as mean±SD of the concentration of human RANTES.

Next, we tested the immune response to nonmodified and modified Ad5 vectors after islet autotransplantation in STZ-induced diabetic C57BL/6 mice. Preliminary experiments demonstrated that the adenoviral vector transfection efficiency in mouse islets using Ad5, Ad5RGD, Ad5pK7, and Ad5RGDpK7 was similar to the results obtained in human islets (data not shown). The viability and functionality of the islets infected with different Ad5 vectors was confirmed in all recipients after transplantation (nonfasting glucose <200 mg/dL, data not shown). The use of syngeneic murine transplants allowed assessment of the immune response to Ad5 vectors without the complication of a xenoantigen-driven response. We assessed the adenovirus-specific T-helper cell type 1 (Th1) response (Fig. 6A) after infection of murine islets at different viral doses. A dose-dependent increase in Th1 response was observed after infection with Ad5 or Ad5RGDpK7. Interestingly, a lower response was observed after transplantation of adenovirus-infected islets, compared with in vivo delivery of Ad5. Twenty-one days after the transplant, a higher Th1 response was observed after the infection with Ad5RGDpK7 (Fig. 6A), an effect probably related to a higher percentage of infected cells and a higher number of viral copies per cell (Fig. 2). A significant reduction in Th1 response was observed in recipients of Ad5RGDpK7-infected islets compared with Ad5-infected islets after a decrease in the viral dose required to infect greater than 80% of the islet cells (Fig. 6B). Similar results were demonstrated after analysis of adenovirus-specific antibody response (Fig. 6C). Higher levels of immunoglobulin (Ig) G1 were demonstrated in animals that received an intravenous dose of Ad5 compared with animals that received islets infected ex vivo with Ad5. Moreover, a significant reduction in the humoral immune response was observed in recipients of islets infected ex vivo with a reduced dose of Ad5RGDpK7, compared with higher doses of Ad5, Ad5RGD, or Ad5pK7 required to infect greater than 80% of the islet cells. Similar results, but with lower titers, were demonstrated after the analysis of IgG2a and IgM (data not shown). Overall, these results indicate that reduction in the viral dose was associated with a reduced inflammatory and immune response after infection of pancreatic islets.

F6-44
Figure 6. Analysis of adenovirus vector-induced immune response in transduced islets. Islets were isolated from normal C57BL/6 and infected with Ad5 or Ad5RGDpK7 at different concentrations (A) or at a minimal concentration required to infect greater than 80% of the islet cells (B) (Ad5, 500 VP/cell; Ad5RGDpK7, 0.1 VP/cell). Twenty-four hours after infection, 300 IEQ were infused into the portal vein in STZ-induced diabetic C57BL/6. Twenty-one days after the transplant, adenovirus-driven specific Th1 response was analyzed (A and B). To this end, spleens were collected, processed into single cell suspensions, and cultured at a concentration of 1×106/mL in 200-μL culture volumes with either Ad5 or Ad5RGDpK7 at 500 VP/cell (see:
Materials and Methods ). Culture supernatants were collected after 72 hr and stored at −70°C. IFN-γ was determined by ELISA. Results are expressed as mean±SD of the concentration of IFN-γ. Antibody responses (IgG1) against adenovirus vectors after transplantation of genetically modified islets (80% transfection efficiency: Ad5, 500 VP/cell; Ad5RGD and Ad5pK7, 10 VP/cell; Ad5RGDpK7, 0.1 VP/cell) were measured in the sera of mice by ELISA (see Materials and Methods ). Results are expressed as mean±SD of the absorbances obtained at 405 nm.

DISCUSSION

Previous studies have demonstrated the feasibility of ex vivo genetic modification of isolated pancreatic islets using viral and nonviral vectors (3–15, 25). However, for some strategies involving gene therapy for islet transplantation, such as cytoprotection, most of the cells need to express the therapeutic gene. To this end, alternative strategies for highly efficient gene transfer with minimal toxicity need to be addressed. In the present study, we tested the possibility of retargeting the adenovirus vector by incorporation of the RGD and pK7 motifs in the fiber knob to bind novel cell receptors. Double-genetic modification of Ad5 resulted in an adenoviral vector with a CAR-independent mechanism for cell infection (23). This novel, infectivity-enhanced gene transfer vector allowed us to significantly reduce the viral dose required to infect the majority of the islet cells.

Weber et al. (9) reported a significant increase in apoptosis when pancreatic islets were transfected with standard adenoviral vectors at high concentrations. Islet loss by apoptotic mechanisms appears to be limited to the first days after the infection and does not appear to influence the glucosestimulated insulin release or the capacity to restore euglycemia of the surviving islets. However, in this study, only 43% of the cells expressed the transgene at a multiplicity of infection (MOI) of 1:1,000. Previous studies in nonhuman primate islets demonstrated a direct relationship between the viral dose and the reduced capacity to release insulin after high glucose stimulation (25). In vitro studies performed in our laboratory have shown that human islets infected with Ad5RGDpK7 have a higher capacity to release insulin after a glucose challenge compared with islets infected with a singlemodified vector (Ad5RGD or AdpK7) or nonmodified Ad5 vectors (data not shown). Moreover, we demonstrated higher functional islet mass after transplantation, as evaluated by intraperitoneal glucose tolerance testing in animals that received human islets infected with Ad5RGDpK7 versus control vectors (Fig. 4).

Potent host inflammatory and immune responses limit the effective application of adenovirus vectors for gene therapy in humans (16, 27–29). Within 1 week of administration, firstgeneration adenovirus vectors are known to induce major histocompatibility complex class I-restricted CD8+ cytotoxic T lymphocytes directed against adenovirus proteins and a Th1 dominant immune response dependent on IFN-γ. Furthermore, acute inflammation and injury to the host cell occurs within 24 hr of adenoviral infection. For these reasons, newer generations of adenoviral vectors are being developed. Previous studies demonstrated that part of the inflammatory response after Ad5 infection depends on the C-C chemokine RANTES expression induced through NF-κB nuclear translocation. NF-κB is a ubiquitous transcription factor that is activated by a variety of stimuli including adenoviral infection (29, 30). NF-κB is also involved in the apoptotic pathways leading to elimination of virally infected cells, a process that is independent of an immune response. In the present study, we addressed the inflammatory and immune-mediated effects elicited by adenoviral-infected isolated islets. We observed that after the reduction in viral dose required to infect the majority of the islet cells, islets infected with Ad5RGDpK7 presented a significant reduction in NF-κB nuclear translocation after the infection versus control vectors (Fig. 5). Reduction in NF-κB activation correlated with a reduction in the concentration of RANTES in islet culture supernates. Moreover, we demonstrated, in a syngeneic islet transplant model in immunocompetent mice, a significant reduction in the Th1 and humoral-specific immune response to the adenoviral vector when islets were infected with the double-modified Ad5 (Ad5RGDpK7). The fact that the islets can be genetically modified ex vivo represents an important advantage for gene therapy purposes compared with in vivo gene therapy approaches. Chan et al. analyzed the potential therapeutic value of adenovirus-mediated gene transfer in cardiac transplantation and demonstrated that the immunologic limitations of these vectors are not universal for all tissue types (26). Indeed, we observed in the present study that the immune response to Ad5 vectors is reduced when the genetic modification is performed ex vivo, compared with intravenous administration of the vector.

Despite the advantages offered by the targeted adenoviral vectors, the temporal transgene expression is a limiting factor. However, for some therapeutic strategies, such as islet cytoprotection early after transplantation, temporal transgene expression is probably recommended because the factors related to early islet injury, such as islet isolation, proinflammatory cytokine release after transplantation, initial poor vascularization, hyperglycemia, and so forth, usually play a minor role in islet injury weeks after transplantation (1). At this time, islets are engrafted and vascularized. In contrast, for immunomodulatory strategies, indefinite transgene expression is required and therefore viral vectors with integrating capabilities such as adenoassociated virus (12) or lentivirus (13) might be more appropriate for the success of the gene therapy. Overall, the double-modified Ad5 vector that we generated in this study, Ad5RGDpK7, is a novel, infectivity-enhanced, gene transfer vector with direct therapeutic potential in islet transplantation.

Acknowledgments.

The authors thank Stacie M. Jenkins, B.S., and Barry Grace, B.S., C.P.T.C., for their technical assistance during the human islet isolations.

REFERENCES

1. Hering B, Ricordi C. Results, research priorities, and reason for optimism: Islet transplantation for patients with type I diabetes. Graft 2000; 2: 12.
2. Shapiro AM, Lakey JR, Ryan EA, et al. Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N Engl J Med 2000; 343: 230.
3. Benhamou PY, Moriscot C, Badet L, et al. Strategies for graft immunomodulation in islet transplantation. Diabetes Metab 1998; 24: 215.
4. Contreras JL, Bilbao G, Smyth CA, et al. Cytoprotection of pancreatic islets before and early after transplantation by gene transfer of the anti-apoptotic Bcl-2 gene. Transplantation 2001; 71: 1015.
5. Giannoukakis N, Rudert WA, Ghivizzani SC, et al. Adenoviral gene transfer of the interleukin-1 receptor antagonist protein to human islets prevents IL-1beta-induced beta-cell impairment and activation of islet cell apoptosis in vitro. Diabetes 1999; 48: 1730.
6. Grey ST, Arvelo MB, Hasenkamp W, et al. A20 inhibits cytokine-induced apoptosis and nuclear factor kappaB-dependent gene activation in islets. J Exp Med 1999; 190: 1135.
7. Rodriguez Rilo HL, Paljug WR, Lakey JR, et al. Biolistic bioengineering of pancreatic beta-cells with fluorescent green protein. Transplant Proc 1998; 30: 465.
8. Saldeen J, Curiel DT, Eizirik DL, et al. Efficient gene transfer to dispersed human pancreatic islet cells in vitro using adenovirus-polylysine/DNA complexes or polycationic liposomes. Diabetes 1996; 45: 1197.
9. Weber M, Deng S, Kucher T, et al. Adenoviral transfection of isolated pancreatic islets: A study of programmed cell death (apoptosis) and islet function. J Surg Res 1997; 69: 23.
10. Csete ME, Benhamou PY, Drazan KE, et al. Efficient gene transfer to pancreatic islets mediated by adenoviral vectors. Transplantation 1995; 59: 263.
11. Leibowitz G, Beattie GM, Kafri T, et al. Gene transfer to human pancreatic endocrine cells using viral vectors. Diabetes 1999; 48: 745.
12. Flotte T, Agarwal A, Wang J, et al. Efficient ex vivo transduction of pancreatic islets cells with recombinant adeno-associated virus vectors. Diabetes 2001; 3: 515.
13. Giannoukakis N, Mi Z, Gambotto A, et al. Infection of intact human islets by a lentiviral vector. Gene Ther 1999; 6: 1545-1551.
14. Curran MA, Ochoa MS, Molano RD, et al. Efficient transduction of pancreatic islets by feline immunodeficiency virus vectors 1. Transplantation 2002; 74: 299.
15. Embury J, Klein D, Pileggi A, et al. Proteins linked to a protein transduction domain efficiently transduce pancreatic islets. Diabetes 2001; 50: 1706.
16. Glorioso JC, Naldini L, Kay MA. Viral vectors for gene therapy: The art of turning infectious agents into vehicles of therapeutics. Nat Med 2001; 7: 33.
17. Bergelson JM, Cunningham JA, Droguett G, et al. Isolation of a common receptor for Coxsackie B viruses and adenoviruses 2 and 5. Science 1997; 275: 1320.
18. Louis N, Fender P, Barge A, et al. Cell-binding domain of adenovirus serotype 2 fiber. J Virol 1994; 68: 4104.
19. Dmitriev I, Krasnykh V, Miller CR, et al. An adenovirus vector with genetically modified fibers demonstrates expanded tropism via utilization of a coxsackievirus and adenovirus receptor-independent cell entry mechanism. J Virol 1998; 72: 9706.
20. Dmitriev I, Kashentseva E, Rogers BE, et al. Ectodomain of coxsackievirus and adenovirus receptor genetically fused to epidermal growth factor mediates adenovirus targeting to epidermal growth factor receptorpositive cells. J Virol 2000; 74: 6875.
21. Reynolds P, Dmitriev I, Curiel D. Insertion of an RGD motif into the HI loop of adenovirus fiber protein alters the distribution of transgene expression of the systemically administered vector. Gene Ther 1999; 6: 1336.
22. Krasnykh V, Dmitriev I, Mikheeva G, et al. Characterization of an adenovirus vector containing a heterologous peptide epitope in the HI loop of the fiber knob. J Virol 1998; 72: 1844.
23. Wu H, Seki T, Dmitriev I, et al. Double modification of adenovirus fiber with RDG and polylysine motifs improves CAR-independent gene transfer efficiency. Hum Gene Ther 2002; 13: 1647.
24. Gotoh M, Maki T, Kiyoizumi T, et al. An improved method for isolation of mouse pancreatic islets. Transplantation 1985; 40: 437.
25. Bilbao G, Contreras JL, Dmitriev I, et al. Genetically modified adenovirus vector containing an RGD peptide in the HI loop of the fiber knob improves gene transfer to nonhuman primate isolated pancreatic islets. Am J Transplant 2002; 2: 237.
26. Chan SY, Li K, Piccotti JR, et al. Tissue-specific consequences of the anti-adenoviral immune response: Implications for cardiac transplants. Nat Med 1999; 5: 1143.
27. Bowen GP, Borgland SL, Lam M, et al. Adenovirus vector-induced inflammation: Capsid-dependent induction of the C-C chemokine RANTES requires NF-κB. Hum Gene Ther 2002; 13: 367.
28. Muruve DA, Barnes MJ, Stillman IE, et al. Adenoviral gene therapy leads to rapid induction of multiple chemokines and acute neutrophil-dependent hepatic injury in vivo. Hum Gene Ther 1999; 10: 965.
29. Lieber A, He CY, Meuse L, et al. Inhibition of NF-κB activation in combination with Bcl-2 expression allows for persistence of first-generation adenovirus vectors in the mouse liver. J Virol 1998; 72: 9267.
30. Richmond A. NF-κB, chemokine gene transcription and tumor growth. Nat Rev Immunol 2002; 2: 664.

1 Supported by research grants from the Diabetes Trust Fund and the Protective Life Clinical Initiative Award, University of Alabama at Birmingham (J.L.C. and D.E.), and National Institutes of Health grants R01 HL67962, P50 CA89019, R01 CA86881, and U19 DK57958 (D.T.C.).

© 2003 Lippincott Williams & Wilkins, Inc.