Introduction

Cystic fibrosis (CF) is the most common lethally inherited single gene disorder in Caucasians. This autosomal recessive monogenetic disorder most commonly results from the F508 mutation, in 70% of patients, but can also result from over 1000 unique mutations in the CFTR gene¹,². A dysfunctional protein results in defective cAMP-mediated airway epithelial cell chloride secretion³. This deficiency leads to hyperviscous exocrine secretions with resultant impaired mucociliary clearance, initiating a cycle of obstruction, infection, and inflammation, the hallmark of the pulmonary CF phenotype⁴. Currently approved therapies are primarily restricted to the consequences of end-stage CF. Rapid advances in the molecular understanding of the determinants of disease theoretically make CF an ideal disease for targeted gene therapy.

Adeno-associated virus (AAV), a naturally replication-deficient single-stranded DNA parvovirus, has potential as a therapeutic for pulmonary gene therapy because of its ability to infect airway epithelial cells⁵ and to persist. Notably, longterm in vivo expression of 1 year has been achieved after a single intramuscular injection⁶,⁷. This long-term persistence has been primarily attributed to the generation of stable episomal forms, such as concatemers produced by ligation of monomers of double-stranded recombinant AAV (rAAV) DNA⁸. Successful in vivo airway gene transfer with rAAV vectors carrying the CFTR transgene has been demonstrated in multiple species, including rabbits⁹,¹⁰, mice¹¹, monkeys¹²,¹³,¹⁴, and human CF patients¹⁵. Recombinant CFTR transcripts as well as protein expression have been detected in rabbits and macaques for at least 6 months following endobronchial delivery¹⁶,¹³, exceeding the expected life span of an airway epithelial cell. The clinical trials that document in vivo gene transfer and functional evidence of CFTR reconstitution are further testament to the promise of rAAV gene therapy. A phase I clinical trial demonstrated that airway delivery of aerosolized rAAV-CFTR vector can be accomplished safely by nebulization to CF patients¹⁷. A phase I/II clinical trial further demonstrated that rAAV-CFTR delivery to the maxillary sinuses of CF patients resulted in dose-dependent gene transfer for 10 weeks and a functional correction of CFTR¹⁵. An early loss of expression was noted by day 28¹⁵ and illustrates the necessity for repeated administrations for sufficient CFTR expression. In fact, extrapolation from in vitro studies do suggest that only a small 5–10% correction is required to overcome the chloride ion transport defect in cultured CF airway epithelium¹⁸; thus, even low-efficiency gene therapy has the potential to become a practical intervention for CF, if vector-derived transgene expression is maintained over time.

Bronchoscopic delivery is an attractive strategy to optimize deposition of high doses of vector in situ to the target organ. We recently reported that bronchoscopic microspraying of rAAV vector to rhesus macaques delivers a higher dose to the lung than standard endotracheal nebulization¹². In fact, these studies showed that high-dose aerosolized delivery of rAAV2-GFP resulted in dose-dependent rAAV-mediated expression¹². These results are promising because the more efficient delivery of higher titer vector directly to the target organ did translate into enhanced gene transfer and expression.

Even with enhanced delivery of high-dose vector, maintenance of sufficient CFTR expression over a lifetime would require repeated vector administrations, due to the limited life span of most transduced cells. Serial dosing strategies may potentially elicit an escalating humoral immune response to the recombinant vector. The presence of an immune response to either the transgene or the viral capsid protein could limit the success of repeated delivery of rAAV vectors. Preexisting immunity indicated by seropositivity toward a specific serotype may also inhibit gene transduction. The aim of this study is to address whether repeated administrations of rAAV2 in nonhuman primates can elicit escalating vector-specific immune responses inhibitory to gene transfer and/or expression to assess the potential for clinical applicability. We will utilize our model of bronchoscopic microspraying to establish a strategy of repeated administration to induce a serotype-specific anti-AAV2 immune response and subsequently measure rAAV2-mediated transfer and expression in the setting of an escalated humoral response.

Results

Deposition of rAAV2-GFP vector

Using aerosolized bronchoscopic delivery via a MicroSprayer¹², we dosed macaques repeatedly with rAAV2 vectors. We hypothesized that serial exposure to rAAV2 with the first two aerosolized doses may elicit an immune response primarily to the vector serotype, rAAV2. Thus the first two doses contained CFTR, an endogenous protein. The third vector contained GFP to track transgene expression resulting specifically from the final dose of vector. The rationale for tracking GFP was to identify the efficiency of gene transfer subsequent to serial rAAV2 exposure and specifically attributable to the final aerosolized dose. Fig. 1A illustrates the dosing schedule: rAAV2-CFTR was administered initially and at week 2 and rAAV2-GFP at week 5.

Figure 1.

(A) Study design of the experimental protocol to establish serial rAAV2 exposure. (B) Schematic representation of the lung regions. A nine-region grid was first delineated on a ¹³³Xe gas image to determine functional lung borders and was then superimposed on the ^99mTc-rAAV-GFP aerosol image to quantitate regional deposition.

Full figure and legend (66K)

Deposition to the right lung measured by gamma scintigraphy averaged 50.3 23.6% of the initial dose or a total dose of 4.53 10¹¹ 2.1 10¹¹ infectious units (iu). As shown in Fig. 1B, we quantified right lung deposition further by region. Despite the aerosol being delivered with the MicroSprayer centrally to the bronchi, deposition was achieved throughout the lung including the peripheral regions (Fig. 2A). Average deposition, expressed as a percentage of total right lung deposition, was 0.17 0.09, 2.14 1.64, 2.84 2.05, 2.94 2.69, 12.61 3.61, 14.76 7.02, 2.21 2.20, 7.93 6.75, and 4.71 1.54% for regions 1–9, respectively. At autopsy, the weights of specific regions were factored into regional deposition to normalize for regional differences (Fig. 2A). The greatest variance in isotope deposition resulted in up to a 1000-fold difference peripherally (region 1) versus centrally (5 and 6). However, region 1, as the smallest region, with consequently the smallest fractional dose, is insignificant relative to overall regional dose deposition. The pattern of deposition was consistent between macaques. Multiple comparisons subsequent to a Kruskal–Wallis rank sum test indicated that deposition, when normalized for weight, was similar and did not statistically differ in regions 2 to 9.

Figure 2.

(A) Deposition of aerosolized rAAV2-GFP in the nine right lung regions. The average regional dose of infectious units of ^99mTc-rAAV2-GFP admixture per gram of tissue per region is shown. Error bars represent standard deviation (n = 4 for each region). (B) Distribution of vector DNA transfer detected by PCR. Average copy number of rAAV2-GFP DNA per gram of tissue per region is shown. Error bars represent standard deviation (n = 4 for each region).

Full figure and legend (67K)

Raav2-mediated DNA transfer

The design to quantify GFP transfer following serial doses of rAAV2-CFTR was to detect gene transfer attributable to only the final dose of vector. We detected GFP DNA in all lung regions by real-time PCR and it averaged 1.43 10⁹ 4.25 10⁹ iu/g tissue (n = 36) in experimental macaques (Fig. 2B), whereas gene transfer was absent in the control. A Kruskal–Wallis rank sum test confirmed uniform regional vector transfer and no statistical difference in GFP transfer to any of the nine lung regions.

rAAV2-mediated RNA expression

To evaluate whether repeated vector exposure impacted on gene transduction, we determined the presence of mRNA expression by RT-PCR qualitatively (Fig. 3). Of note, the level of mRNA expression when present was not within the limits of detection for quantitative analysis. Of the 36 possible regions analyzed (9 lung regions per macaque), 18 were positive for mRNA expression. We performed a regional analysis of vector dose, DNA transfer, and presence of mRNA expression per host (Figs. 3A and 3B). Analysis according to each macaque demonstrated that two animals had fewer regions of positive mRNA expression (2 of 9 regions positive) in contrast to the other two, with most regions positive (7 or 9 of a total of 9). The two macaques with the most regions positive for transduction did not vary in terms of sex, age, or initial dose from those with only 1 or 2 regions positive.

Figure 3.

Regional analysis of GFP-specific RNA expression. (A) Regional dose deposition per animal. Squares depict dose deposited in each of the nine lung regions per animal. Black squares () specifically depict regions positive for GFP mRNA expression, whereas white squares () depict regions where mRNA for GFP was not detectable based on nested RT-PCR results. (B) Average GFP-specific DNA copy number per gram of lung tissue per animal. Squares indicate the average copy number per gram of tissue in each of the nine lung regions per animal. Again, GFP mRNA expression is indicated as either present () or absent (). (C) RT+ assay of GFP present in experimental animals (1–12) and not present in control animal (13). (D) RT+ assay of serial dilutions of GFP plasmids, 10^-9, 10^-12, 10^-15, 10^-16, 10^-17, and 10^-18g DNA (lanes 1–6, respectively) and blank (lane 7). (E) -Actin RT- (lanes 1, 2, 4–6) and RT+ (lanes 3, 7–11) assay of experimental and control animals. Lane 11 is the positive control and lane 12 is the negative control.

Full figure and legend (85K)

rAAV2-mediated protein expression

Fluorescence from GFP was detected in epithelial cells of the small and large airways of all experimental animals by confocal microscopy (Figs. 4A and 4B). GFP fluorescence was visualized in the cytoplasm of ciliated epithelial cells along bronchial airways and submucosal glands in all experimental animals, whereas GFP was absent in the control (Fig. 4C). Immunohistochemistry using an affinity-purified polyclonal rabbit anti-GFP antibody and a Cy3-conjugated secondary antibody further confirmed the presence of GFP in airway epithelium in vector-treated animals (Fig. 4D). In contrast, GFP antibody staining was absent in the control animal (Fig. 4E). The absence of nonspecific background staining from the secondary antibody is illustrated in Figs. 4F and 4G. Western blot analysis further verified the presence of rAAV2-mediated GFP protein expression in the vector-treated animals (Fig. 4H). Gene transfer, transduction, and rAAV2-mediated expression were accomplished despite serial vector exposure.

Figure 4.

Protein expression. (A–C) Fluorescent GFP expression in bronchial epithelium sectioned at the level of segmental branching and analyzed for GFP-specific expression by confocal microscopy (400 original magnification). (A and B) GFP expression in the cytoplasm of ciliated epithelial cells in AAV2-GFP-treated macaques. (C) The absence of GFP-specific expression in pulmonary sections from a control macaque. (D–G) Immunohistochemistry of sagittal bronchopulmonary sections to demonstrate GFP expression by an affinity-purified anti-GFP antibody (AB6556). The sections presented are from (D and F) rAAV2-GFP-treated macaques and (E and G) a control macaque. Representative secondary antibody (Cy3) counterstaining of an anti-GFP antibody applied to (D) the experimental and (E) the control is shown. (F and G) Absence of nonspecific staining with the Cy3 secondary antibody alone. Confocal microscopy original magnification 400. (H) Western blot analysis of GFP expression. Top demonstrates GFP expression (27-kDa product) and the bottom demonstrates -actin expression (43-kDa product) for an internal control. The positive control is rAAV2-GFP-transduced human embryonic kidney cells, 293 cells (lane 1). Lanes 2 and 9–12 are samples from the control demonstrating the absence of the GFP product. Lanes 3–8 are samples from all experimental macaques and demonstrate the GFP product.

Full figure and legend (322K)

Evidence of minimal inflammation

All macaques tolerated repeated bronchoscopies and vector aerosols without adverse sequelae. All macaques remained healthy with vigorous appetites, normal behavior, and no signs of respiratory distress or acute upper respiratory infections. Upon comprehensive review of the pathology, there was no acute or chronic inflammation along the airways, interstitium, or alveolar spaces specific to the experimental animals compared to the control by a pathologist blinded to the experiment. A review of hematoxylin and eosin lung sections from each region revealed no identifiable differences between the experimental and the control animals (Fig. 5). Specifically, there was no evidence of destructive acute pulmonary changes such as eosinophilia, enlarged bronchial lymphoid aggregates, or inflammatory infiltrates or chronic changes such as chronic bronchiectasis obliterans or bronchiectasis. Mild to moderate peribronchial/bronchiolar lymphocyte or eosinophilic infiltration in localized areas was present in most airways. Three of four experimental macaques demonstrated bronchial lymphoid aggregates around peripheral airways (Fig. 5B), noted even in the normal macaque lung (Fig. 5D). Since host immune responses to the vector and viral capsid proteins may play a role in triggering inflammation and cellular recruitment to the lung, we analyzed bronchoalveolar lavage (BAL) fluid counts. There was no increase in BAL cellularity after any dose of vector (Fig. 6) and no evidence of a recruited inflammatory infiltrate.

Figure 5.

Hematoxylin and eosin-stained lung sections of (A and B) experimental and (C and D) control animals, which demonstrate the range from absence of pathologic inflammation (A, C) to mild localized peribronchial lymphoid aggregates (B, D).

Full figure and legend (342K)

Figure 6.

Total cellular BAL counts. Total cell counts from cytospins of BAL fluid prior to each dose of rAAV2 vector administered are shown. The circles () demonstrate the averaged cell count and the error bars demonstrate SEM.

Full figure and legend (57K)

Presence of anti-AAV2 neutralizing activity

To determine whether the doses of rAAV2 vector elicited a significant humoral immune response, we analyzed sera at each time point for anti-AAV2 neutralizing antibodies (Figs. 7A–7F). Anti-AAV2 neutralizing activity is a functional measure of sera capable of blocking reinfection following reexposure to rAAV2. There was a substantial rise in the antibody response after the first dose for three animals as the animals transitioned from a preimmune state to post-vector exposure. A markedly elevated anti-AAV2 titer was established in all vector-treated recipients by the second time point and persisted until autopsy 3 weeks later (Fig. 7M). By the third dose all had escalating titers in excess of a fourfold increase from preimmune samples consistent with seroconversion. In summary, serial reexposures to vector induced anti-AAV2 antibodies capable of blocking vector transfer in vitro.

Figure 7.

Anti-AAV neutralizing antibody activity. (A–F) Fluorescence micrographs of Ad-infected 293 cells coinfected with a rAAV-GFP vector preincubated with sera from an experimental animal either (A) on the day of infection or (D) prior to autopsy. The sera from (C) the control monkey demonstrates the same GFP fluorescence as (B) an irrelevant serum in the positive control or (A) the preimmune sera of the experimental monkeys. (E) The negative control contains no vector and (F) the bright-field image demonstrates the underlying cellular microarchitecture, excluding a cytopathic effect from the helper adenovirus. (G–L) Fluorescence micrographs of Ad-infected 293 cells coinfected with the rAAV2-GFP vector preincubated with BAL as above. BAL from the experimental animal (G) at the initial time point or (J) prior to autopsy and from the control animal. (H) The positive control and (K) the negative control with (L) corresponding bright-field image. (M) Table indicates the titer of anti-AAV2 neutralizing antibody detected in the serum and BAL capable of inhibiting gene transfer in vitro at all time points. Titers are expressed as the reciprocal dilution of serum required for 10-fold neutralization and documented per animal.

Full figure and legend (249K)

We also detected immune responses in BAL: anti-AAV2 neutralizing activity absent in preimmune BAL was detected upon vector exposure (Figs. 7G–7L). A higher titer of BAL was required to demonstrate an anti-AAV2 neutralizing activity equivalent to that in sera (1:1000 vs 1:10,000, respectively). The higher titer required for equivalent inhibition may reflect either the dilution inherent in sampling BAL or the overall lower concentration of immunoglobulins in BAL as opposed to sera¹⁹.

We performed an ELISA to detect whether a humoral response specific to the foreign transgene, GFP, was induced. Anti-GFP antibodies were not detectable in either pre- or postimmune sera (data not shown). The absence of anti-GFP antibodies may reflect the low expression of GFP upon a single exposure to GFP or the delayed onset of rAAV2-mediated GFP expression.

Discussion

Strategies for redosing recombinant vectors are essential for long-term transgene expression and to make gene therapy a therapeutic option for pulmonary diseases such as CF. Advances in the field have mirrored the numerous challenges to gene therapy with the production of higher titer vectors, stronger promoter-driven expression, tissue-directed tropism, and suppression of immunologic barriers to sustain transgene expression²⁰. Repeated administrations of vector will probably be necessary to maintain transgene expression long term due to episomal persistence as opposed to the integration of rAAV genomes into the host genome²¹. One of the ongoing current goals for successful CFTR gene transfer is to evaluate whether repeated airway delivery of high doses of recombinant vector translates into transgene expression in the milieu of active immunity to a given serotype. These experiments document the feasibility of high-titer rAAV2 vector delivery to the pulmonary epithelium with ensuing gene transfer and expression despite eliciting potential immunologic interference.

In CF, the application of airway delivery has an obvious utility for directing highly concentrated vector in situ to the end organ. Our previous bronchoscopic MicroSprayer model advanced the efficiency of airway delivery¹². This mode of delivery produced the most uniform and consistent deposition to the pulmonary epithelium without losses associated with nebulization and permitted a 50- to 100-fold improvement in vector delivery comparatively. The experiments presented here confirm our earlier bronchoscopic results, since high doses of vector (9 10¹¹ iu) were delivered to the main-stem bronchi and, on average, 50.3% or 4.53 10¹¹ iu of vector was deposited per lung. Initial deposition of rAAV2-GFP, as quantified by gamma scintigraphy, was delivered to all lung regions with relative uniformity. Despite delivery to the central lung regions, deposition was distributed throughout all regions secondary to alveolar activity and pulmonary dynamics. Gamma scintigraphy permitted a comparison of vector deposition with subsequent gene transfer. GFP transfer was detected in all regions, with an average DNA transfer of 1.43 10⁹ copies per gram of lung tissue. Regions 5, 6, 8, and 9 did have the highest average dose. However, regional dose when normalized for tissue weight in regions 2 through 9 did not differ statistically. The regional variability may reflect that gamma scintigraphy measures initial deposition but does not reflect the multiple pulmonary factors that potentially affect subsequent vector redistribution, cellular uptake, and level of gene transfer by the time of autopsy. In contrast, the single-dose monkeys demonstrated an average of 7.54 10⁶ copies/g tissue of GFP. The aim of the single-dose study was to determine the naïve transduction efficiency of AAV-GFP alone and thus these experiments were not duplicated for the repeated-dose study. In fact the same window of transduction after GFP administration, a 3-week period, was identical in both studies to mirror the same period of transduction prior to measuring GFP. Overall, average gene transfer increased with repeated administration, in part due to individual monkeys and some regional variability reflecting higher gene transfer.

Most importantly, the repeated dosing strategy did elicit immune responses but did not inhibit gene transfer. Furthermore, mRNA expression detected by RT-PCR was present in each macaque, although it precluded quantitative analysis. The products of RT-PCR were detectable upon second-round PCR analysis. Overall detection of mRNA expression per lung region was 50%, with 18/36 regions positive for mRNA. Two subsets emerged—one with almost all regions positive for gene transduction and the other with only 1 to 2 regions positive. This difference was probably due to host variability and did not stratify according to differences in size, sex, or age. The higher expressing subset demonstrated mRNA expression of 88%, which exceeded the average transduction previously reported in the single-dose experiments as 81% (22/27 regions positive)¹². In those experiments, doses exceeding 3 10⁹ iu resulted in a statistically significant increase in vector transduction. The efficiency of delivery in the current experiments resulted in the overwhelming majority of regional doses exceeding the prior threshold for transduction. The overall incidence of transduction after repeated doses suggests that host-specific factors due to an immune response or a complexity of host cellular factors influence but do not inhibit gene transfer and expression.

Despite lower levels of mRNA expression, there was clear evidence of protein expression. GFP expression was detected in airway epithelium by not only fluorescence microscopy but also immunohistochemistry. The site of GFP expression along the airway epithelium was also identical with both methodologies. Detection of irrelevant crossreacting proteins with the affinity-purified anti-GFP antibody is highly unlikely since the control macaque had comparatively very low background staining. Western blot analysis further documented the presence of GFP in experimental macaques and its absence in the control.

The data did not demonstrate a significant immune response with repeated doses of rAAV2. The anti-AAV2 antibody response escalated in the sera but did not prevent reinfection with rAAV2 vector and gene transfer. Likewise the anti-AAV2 neutralizing response in the BAL increased after the third dose and remained less robust than sera. Although we did not measure the constitutive differences in types and concentrations of immunoglobulins in BAL and sera directly, the ability to reinfect the macaques clearly demonstrates that the anti-neutralizing activity is not entirely inhibitory, permitting repeated exposures of the same serotype. Furthermore, these results are consistent with epidemiologic studies demonstrating that natural AAV infection is common in humans, with over 50–80% being seropositive for AAV2²⁰,²². Such prevalence of preexisting neutralizing antibodies does not seem to prevent a reinfection by the wild-type virus²³. Reexposure to AAV infection also does not result in clinically significant inflammation or protective immunity, supporting the feasibility of serial readministration of rAAV-CFTR vectors to humans. The rhesus macaque model as a nonhuman primate model parallels the inflammation inducible by rAAV in humans. Other studies have shown decreased transduction efficiency following vector readministration⁹. The decrease was attributed to contamination of vector preparations with wild-type virus and necessitated immunosuppression for gene transfer after repeated vector exposures¹¹,¹¹. Other studies have attributed longterm expression of rAAV transgenes directly to the low immunogenicity of rAAV. T cells are activated after AAV exposure only if capsid proteins are released from transduced cells to be cross presented by local antigen-presenting cells for subsequent T cell activation²⁴, whereas adenoviral exposure elicits a robust T cell activation both by direct transduction of dendritic cells and by cross presentation of antigens by local antigen-presenting cells. The pathologic analysis presented here did not demonstrate any evidence of cellular inflammation, airway cell destruction, bronchitis, bronchiectasis, or bronchiolitis. In fact the absence of an inflammatory infiltrate in the BAL and the indistinguishable pathologic readings between the experimental and the control macaques argue against a significant inflammatory response.

In summary, repeated bronchoscopic delivery results in successful gene transfer, transduction, and protein expression without an escalating inhibitory inflammatory response prohibitive to gene transfer. These studies demonstrate that redosing with one serotype is clinically applicable. Redosing strategies with one vector serotype, as well as alternating serotypes, may broaden the clinical applicability of gene therapy delivery and make CFTR gene replacement therapy a closer reality.

Material and methods

Study design

Juvenile rhesus macaques with an average weight of 13.1 lb were housed according to Animal Care and Use Committee guidelines. All macaques tested negative for simian immunodeficiency virus and herpesvirus simiae. The experimental design, outlined in Fig. 1A, was to deliver aerosolized rAAV2 vector serially. Lung regions were defined as shown in Fig. 1B. The initial doses of rAAV2-CFTR were given at time 0 and 2 weeks. At week 5, the third and final dose (rAAV2-GFP) was given. The final dose also contained technetium (^99mTc) with diethylene triamine penta acetic acid (DTPA) for scintigraphy and, thus, contained an admixture of ^99mTc–DTPA–saline with 9 10¹¹ iu of rAAV2-GFP. The control animal received aerosolized saline. Immediately following the third dose, deposition was quantified by gamma scintigraphy. The animals were autopsied 3 weeks later for detection of GFP DNA, RNA, and protein expression as well as pathologic analysis. Samples of sera were taken prior to each vector administration. BAL was taken prior to the initial dose and at autopsy.

Vectors

rAAV-GFP and rAAV-CFTR vectors, serotype 2, were produced as previously described²⁵,²⁶. Vector preparations were from a single pooled stock with particle-to-infectious unit ratios of <1:100.Vector stocks were free of replication-competent AAV (<1 iu/10⁶ rAAV iu). Vector dose was 9 10¹¹ iu diluted in normal saline for a total volume of 3 ml.

Radiolabeled saline and vector admixed with radiolabel

Three hundred microcuries of ^99mTc (Syncor, Baltimore, MD) were chelated to DTPA and admixed with 3 ml of either saline (control) or rAAV-GFP vector in saline (experimental animals) as previously described¹². We assumed that during the few minutes that were needed to administer the aerosol, the isotope and vector were codistributed. Thus, the isotope marked the initial deposition of the rAAV2-GFP. The vector particles were significantly smaller (20 nm diameter)²⁷ than the saline–isotope droplets generated by the MicroSprayer (22 m diameter) and they would have been suspended in the saline–isotope solution and codistributed with the radioisotope.

Aerosol delivery by MicroSprayer

Direct delivery to the tracheobronchial tree was accomplished using a MicroSprayer (PennCentury, Philadelphia, PA) inserted through a 3.5-mm flexible fiber optic bronchoscope (Olympus, Melville, NY). Macaques were sedated with ketamine and isoflurane, intubated with a 5.0-mm ETT, and given supplemental oxygen. Lidocaine (1%) was instilled onto the carina to minimize coughing. The MicroSprayer was advanced approximately 3 mm beyond the tip of the bronchoscope to visualize aerosolization (10 sprays of approximately 150 l each) of rAAV2-CFTR, ^99mTc–DTPA–saline, or ^99mTc–DTPA–saline–rAAV2-GFP into the main-stem bronchi as previously described¹².

Regional deposition in the right lung

The lung border (Fig. 1B) was first drawn on a ventilation image acquired with inhaled xenon-133. That image was superimposed on the subsequent ^99mTc–aerosol image. Right lung regions are analyzed by a nine-region grid. Total deposition was calculated as previously described¹². The left lung was excluded from our analyses because the overlying heart impaired precise identification of the lung borders.

DNA and RNA analysis

At week 8, the lungs and heart were removed. Lungs were infused with 60 cc of saline and divided into a nine-region grid as previously described¹². Approximately 90% of each lung region sample was flash frozen and the remaining 10% was fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned onto slides for histological analysis. Lung regions were individually weighed, homogenized, and divided for RNA, DNA, and protein extraction. Quantitative PCR with LightCycler (Roche Molecular Biochemicals, Indianapolis, IN) assessed GFP-specific DNA. The PCR product was 205 bp, amplified from 125 ng of genomic DNA as previously described¹².

Total RNA was DNase treated (Life Technologies, Rockville, MD) and then underwent RT-PCR as previously described¹². Qualitative analysis of the RNA by nested RT-PCR was performed.

Protein analysis

Lung samples positive for RNA expression were analyzed by Western blot²⁸. Proteins were separated by 12% SDS–PAGE and transferred to PVDF membranes. Membranes were blocked, washed, and incubated with the same primary antibody, Ab6556 (Abcam, Cambridge, UK), and secondary antibodies (donkey anti-rabbit HRP conjugate, Amersham Life Science, Arlington Heights, IL) as previously described¹².

Multiple transverse 5-m sections were made through each region perpendicular to visible bronchi or bronchioles when possible. Unstained sections were examined for GFP green fluorescence using confocal microscopy (Perkin–Elmer UltraView LCI (Live Cell Imaging) System; Perkin–Elmer Life Sciences, Inc., Boston, MA) and processed using a Dell Optiplex 400 computer, UltraView software (Perkin–Elmer Life Sciences).

The same samples were also tested for immunohistochemistry using a purified rabbit polyclonal antibody to GFP, Ab6556, as used previously²⁹. The secondary antibody was a goat anti-rabbit IgG antibody conjugated to Cy3 (red) (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) and visualized using confocal microscopy.

Neutralizing antibody experiments

Serial dilutions of serum and BAL fluid were analyzed for the presence of anti-AAV2 neutralizing antibodies as previously described³⁰. Serial dilutions of sera or BAL at different time points were incubated with 10⁵ iu of rAAV-GFP for 1 h and added to 293 cells. Adenovirus (2 10⁵ PFU) was likewise added to the cells and medium (Dulbecco's modified Eagle medium supplemented with 20% fetal bovine serum and 200 U of penicillin/streptomycin per milliliter of medium). After 24-hour incubation, the cells were visualized with a fluorescence microscope to detect GFP-fluorescing cells. The end point was defined as the dilution of serum required for 10-fold inhibition of transgene fluorescence. The inverse of the dilution of serum required for 10-fold neutralization was tabulated. If documented as >10,000, then the dilution required for the inhibitory effect exceeded 1:10,000. The serial dilutions were done in triplicate per time point.

Statistical analysis

Data are reported as means standard deviation. A Kruskal–Wallis rank sum test was used to assess the statistical significance of regional deposition fraction and vector transfer (copy number per region). A Fisher exact test was used to compare the success of transduction between single-dosing experiments and repeated-dosing experiments. Nonparametric tests were used because the data were not normally distributed. P values were deemed significant at 0.05.

References

1. Hodson, M. E. and Geddes, D. M. (1994). Cystic Fibrosis. 1st ed., Arnold: London.
2. (2002). The Cystic Fibrosis Genetic Analysis Consortium. CFTR Mutation Table. Online.
3. Schwiebert, E. M., et al. (1995). CFTR regulates outwardly rectifying chloride channels through an autocrine mechanism involving ATP. Cell 81: 1063–1073. | Article | PubMed | ISI | ChemPort |
4. Rosenstein, B. J. (1998). What is a cystic fibrosis diagnosis? Clin. Chest Med. 19: 423–441. | Article | PubMed | ISI | ChemPort |
5. Carter, B. J. (1992). Adeno-associated virus vectors. Curr. Opin. Biotechnol. 3: 533–539. | Article | PubMed | ChemPort |
6. Xiao, X., Li, J. and Samulski, R. J. (1996). Efficient long-term gene transfer into muscle tissue of immunocompetent mice by adeno-associated virus vector. J. Virol. 70: 8098–8108. | PubMed | ISI | ChemPort |
7. Kessler, P. D., et al. (1996). Gene delivery to skeletal muscle results in sustained expression and systemic delivery of a therapeutic protein. Proc. Natl. Acad. Sci. USA 93: 14082–14087. | Article | PubMed | ChemPort |
8. Song, S., Laipis, P. J., Berns, K. I. and Flotte, T. R. (2001). Effect of DNA-dependent protein kinase on the molecular fate of the rAAV2 genome in skeletal muscle. Proc. Natl. Acad. Sci. USA 98: 4084–4088. | Article | PubMed | ChemPort |
9. Halbert, C. L., et al. (1997). Transduction by adeno-associated virus vectors in the rabbit airway: efficiency, persistence, and readministration. J. Virol. 71: 5932–5941. | PubMed | ISI | ChemPort |
10. Beck, S. E., et al. (1999). Repeated delivery of adeno-associated virus vectors to the rabbit airway. J. Virol. 73: 9446–9455. | PubMed | ISI | ChemPort |
11. Halbert, C. L., Standaert, T. A., Wilson, C. B. and Miller, A. D. (1998). Successful readministration of adeno-associated virus vectors to the mouse lung requires transient immunosuppression during the initial exposure. J. Virol. 72: 9795–9805. | PubMed | ISI | ChemPort |
12. Beck, S. E., et al. (2002). Deposition and expression of aerosolized rAAV vectors in the lungs of rhesus macaques. Mol. Ther. 6: 546–554. | Article | PubMed | ChemPort |
13. Conrad, C. K., et al. (1996). Safety of single-dose administration of an adeno-associated virus (AAV)-CFTR vector in the primate lung. Gene Ther. 3: 658–668. | PubMed | ISI | ChemPort |
14. Afione, S. A., et al. (1996). In vivo model of adeno-associated virus vector persistence and rescue. J. Virol. 70: 3235–3241. | PubMed | ISI | ChemPort |
15. Wagner, J. A., et al. (1998). A phase I/II study of tgAAV-CF for the treatment of chronic sinusitis in patients with cystic fibrosis. Hum. Gene Ther. 9: 889–909. | PubMed | ChemPort |
16. Flotte, T. R., et al. (1993). Stable in vivo expression of the cystic fibrosis transmembrane conductance regulator with an adeno-associated virus vector. Proc. Natl. Acad. Sci. USA 90: 10613–10617. | Article | PubMed | ChemPort |
17. Aitken, M. L., et al. (2001). A phase I study of aerosolized administration of tgAAVCF to cystic fibrosis subjects with mild lung disease. Hum. Gene Ther. 12: 1907–1916. | Article | PubMed | ChemPort |
18. Egan, M., et al. (1992). Defective regulation of outwardly rectifying Cl^- channels by protein kinase A corrected by insertion of CFTR. Nature 358: 581–584. | Article | PubMed | ISI | ChemPort |
19. Kaplan, J. M., et al. (1996). Humoral and cellular immune responses of nonhuman primates to long-term repeated lung exposure to Ad2/CFTR-2. Gene Ther. 3: 117–127. | PubMed | ChemPort |
20. Monahan, P. E. and Samulski, R. J. (2000). AAV vectors: is clinical success on the horizon? Gene Ther. 7: 24–30. | Article | PubMed | ISI | ChemPort |
21. Kearns, W. G., et al. (1996). Recombinant adeno-associated virus (AAV-CFTR) vectors do not integrate in a site-specific fashion in an immortalized epithelial cell line. Gene Ther. 3: 748–755. | PubMed | ISI | ChemPort |
22. Chirmule, N., et al. (1999). Immune responses to adenovirus and adeno-associated virus in humans. Gene Ther. 6: 1574–1583. | Article | PubMed | ChemPort |
23. Blacklow, N. R., Hoggan, M. D., Kapikian, A. Z., Austin, J. B. and Rowe, W. P. (1968). Epidemiology of adenovirus-associated virus infection in a nursery population. Am. J. Epidemiol. 88: 368–378. | PubMed | ChemPort |
24. Jooss, K., Yang, Y., Fisher, K. J. and Wilson, J. M. (1998). Transduction of dendritic cells by DNA viral vectors directs the immune response to transgene products in muscle fibers. J. Virol. 72: 4212–4223. | PubMed | ISI | ChemPort |
25. Hauswirth, W. W., Lewin, A. S., Zolotukhin, S. and Muzyczka, N. (2000). Production and purification of recombinant adeno-associated virus. Methods Enzymol. 316: 743–761. | PubMed | ISI | ChemPort |
26. Zolotukhin, S., et al. (1999). Recombinant adeno-associated virus purification using novel methods improves infectious titer and yield. Gene Ther. 6: 973–985. | Article | PubMed | ISI | ChemPort |
27. Berns (1990). Parvoviridae and their replication. In. Fields Virology. 2nd ed., Raven Press, New York 1743–1763.
28. Wang, X. T., et al. (2000). The mRNA of L-type calcium channel elevated in colon cancer: protein distribution in normal and cancerous colon. Am. J. Pathol. 157: 1549–1562. | PubMed | ISI | ChemPort |
29. Wang, X. T., et al. (2000). Cardiac L-type calcium channel 1-subunit is increased by cyclic adenosine monophosphate: messenger RNA and protein expression in intact bone. J. Bone Miner. Res. 15: 1275–1285. | Article | PubMed | ISI | ChemPort |
30. Hernandez, Y. J., et al. (1999). Latent adeno-associated virus infection elicits humoral but not cell-mediated immune responses in a nonhuman primate model. J. Virol. 73: 8549–8558. | PubMed | ISI | ChemPort |

Acknowledgements

We thank the Cystic Fibrosis Foundation (S.E.B., W.B.G.), American College of Surgeons (A.C.F.), and the National Institutes of Health (Grants NIH PO1 HL51811-06 (W.B.G.) and NHLBI P01, HL51811, NIDDK R01 DK51809 (T.R.F.)), Jerry Wright, Ph.D., of Johns Hopkins Medical Institutions for statistical analyses and expertise, and Deborah Van Kempen of Johns Hopkins, Department of Comparative Medicine, for her veterinary expertise. Dr. Flotte is an inventor of two patented technologies regarding the use of AAV in cystic fibrosis patients. Dr. Guggino is an inventor of one patented technology regarding the use of AAV in cystic fibrosis patients. These technologies have been licensed to Targeted Genetics Corp. Drs. Flotte and Guggino, the National Institutes of Health, and The Johns Hopkins University (the holders of the two patents) conceivably could benefit monetarily from royalties paid by Targeted Genetics Corp. if this gene therapy treatment proves beneficial in human patients suffering from cystic fibrosis.