Main

Alpha-1-antitrypsin (A1AT) deficiency is a recessive condition characterized by low plasma levels of A1AT.1 The unstable variant of A1AT that is expressed in patients with this condition is retained in the endoplasmic reticulum of the hepatocytes rather than released into circulation. An adequate amount of circulating A1AT is essential to protect elastin fibers, and other connective tissue components of the alveolar wall, from degradation by neutrophil elastase.1 A small amount of A1AT is also synthesized by alveolar macrophages in the lungs. In individuals with A1AT deficiency, low plasma A1AT concentrations are associated with the early onset of panlobular basal emphysema. Tobacco exposure also plays an important role in the development of emphysema.1,2 Current clinical therapy for A1AT deficiency includes avoidance of exposure to cigarette smoke and weekly i.v. infusions of blood-derived human A1AT (hA1AT) protein.3,4

Since A1AT deficiency involves a single protein abnormality, the condition is a good candidate for gene therapy. Gene transfer studies utilizing various nonviral vectors and viral vectors, including adeno-associated virus (AAV), have successfully delivered A1AT gene to target organs and produced hA1AT in vivo in animal models.5,6,7,8,9,10,11,12 Currently, AAV are popular vectors because they are capable of stable in vivo expression and are less likely than other viral vectors to induce an immune response.13,14,15 CFTR gene transfer into airway cells with AAV vectors is currently being tested for the treatment of cystic fibrosis patients in two phase I trials.16 AAV vectors have been used in mice for in vivo transfer of hA1AT to both muscle cells and hepatocytes, resulting in stable expression of the protein in the serum.17,18

The simplest approach to increase A1AT levels in the lung is to enhance local production of A1AT in lung cells. Transfer of the A1AT gene to airway epithelial cells in vivo is possible with various vectors,19 but is limited by the same problems that hamper gene transfer of CFTR, including low-efficiency, short-lived gene expression and potentially an adverse immune response.16,20 Recently, Gill et al20 demonstrated persistent transgene expression in the airway using nonviral vectors by modifying their promotor sequences. In our study, we have used an alternative approach whereby the A1AT gene was transferred to macrophages in vitro by use of a recombinant adeno-associated virus (rAAV) vector system, and the transduced cells were delivered to the lung via the airway as previously described.21 This approach is a novel method to enhance levels of A1AT antiprotease activity in the lungs that could potentially be used to prevent the development of emphysema in A1AT-deficient patients.

We transferred hA1AT gene into J774A.1 cells, a macrophage cell line, using the hA1AT gene-containing recombinant AAV plasmid pAAV-CB-A1AT (a gift from Dr JM Wilson, University of Pennsylvania, Philadelphia, PA, USA).22 This AAV construct was packaged into infectious virus in 293 cells by cotransfection with a helper plasmid pSH3.23 When J774A.1 macrophages were infected with the packaged pAAV-CB-A1AT virus, hA1AT was detected in the medium by ELISA as early as 24 h after transfection and reached levels of 50–70 mg/l (Figure 1). The presence of hA1AT mRNA in the gene-transfected macrophages was also confirmed by RT-PCR (Figure 2, lane 2).

Figure 1
figure 1

Production of hA1AT in J774A.1 macrophages after AAV-mediated gene transfer. J774A.1 macrophages were transfected with the packaged pAAV-CB-A1AT virus. The culture medium was changed every 24 h after gene transfection, and hA1AT levels in the medium were quantified by ELISA.

Full size image
Figure 2
figure 2

hA1AT mRNA expression in J774A.1 macrophages. Total RNA was isolated from gene-transfected J774A.1 macrophages or BAL cells using an RNeasy kit (Qiagen, Valencia, CA, USA). The oligonucleotides used for PCR amplification of the hA1AT gene were (forward) 5′-GTG CAT AAG GCT GTG CTG AC-3′ and (reverse) 5′-GCT GGG ATT CAC CAC TTT TC-3′. hA1AT expression was detected in cultured J774A.1 macrophages following transfection with recombinant AAV (lane 2). At 24 h after airway delivery of the transfected J774A.1 macrophages, mRNA expression was detected in BAL cells (lane 3).

Full size image

To increase hA1AT expression in the lower respiratory tract, we delivered 5 × 105 or 1 × 106 of the gene-transferred J774A.1 macrophages in a volume of 50 μl intratracheally into mechanically ventilated C57BL/6J female mice as previously described.21 The serum level of A1AT in C57BL/6J mice is relatively low compared to other strains.24 We confirmed that genetically engineered macrophages continue to express hA1AT mRNA and protein.

At 24 h post macrophage delivery, the cell differential of the bronchoalveolar lavage (BAL) was 90±3% macrophages, 4±2% lymphocytes, 6±2% neutrophils and 1±2% eosinophils; and at day 7, the BAL revealed 94±4% macrophages, 3±2% lymphocytes, 3±3% neutrophils and 1±2% eosinophils. hA1AT mRNA transcripts in BAL cells were detectable by RT-PCR (Figure 2, lane 3). A total of 10±4% of BAL cells from mice that received 1 × 106 gene-transferred J774A.1 macrophages were positive for hA1AT by immunohistochemistry (Figure 3a and b). No expression of hA1AT was detected in BAL cells from mice that received saline or wild-type J774A.1 macrophages (Figure 3c).

Figure 3
figure 3

hA1AT-expressing macrophages in BAL cells. Macrophages expressing hA1AT were detected among BAL cells by immunohistochemical staining 24 h after airway delivery of gene-transferred J774A.1 macrophages (a). The BAL consisted of about 10% hA1AT-producing macrophages after delivery of infected macrophages (b) (n=5), which was significantly more (P<0.01) than in the BAL of mice receiving uninfected macrophages (c) (n=5).

Full size image

Following airway delivery of the macrophages transfected with the hA1AT gene, hA1AT protein was easily detected in the BAL by ELISA (American Laboratory Products Company, Windham, NH, USA) in a recovery volume of 5 ml (Figure 4). The concentration of hA1AT protein in the BAL was 1.5±0.8 mg/l for mice receiving 5 × 105 macrophages and 2.5±0.92 mg/l for mice receiving 1 × 106 macrophages at 24 h. At 1, 3 and 7 days after intratracheal delivery of transfected macrophages, hA1AT protein in BAL from C57BL/6J mice increased from undetectable levels to 2.5±0.9, 2.6±1.1 and 2.2±0.8 mg/l, respectively. The hA1AT levels did not vary significantly up to 7 days following macrophage instillation (P=0.89). The volume of the epithelial lining fluid (ELF) in C57BL/6J mice was 28±6 μl, which was determined using the urea dilution method.25 For example, at day 1 following delivery of 1 × 106 macrophages, the estimated hA1AT concentration in the ELF was between 260 and 440 mg/l, which is higher than the theoretical protective A1AT level in the ELF of 100 mg/l.26 There was no detectable hA1AT in BAL from mice that received saline or control J774A.1 macrophages (Figure 4).

Figure 4
figure 4

Expression of hA1AT in BAL assayed by ELISA. hA1AT appears in BAL in recipient mice as early as 24 h and persists at least for 7 days post airway delivery of 5 × 105 or 1 × 106 gene-transfected macrophages. It is undetectable in saline and nongene-transfected macrophage-treated mice.

Full size image

The lungs were also examined for the presence of hA1AT-expressing macrophages 24 h after airway delivery. While most of the cells in the alveolar spaces were not stained with hA1AT antibody, protein expression was easily detected in macrophages in the alveolar spaces using a low-powered field (× 200 magnification, Figure 5). Expression of the hA1AT was not detected in alveolar epithelial cells, vascular endothelial cells or interstitial cells.

Figure 5
figure 5

Evidence for hA1AT-expressing macrophages in mouse lungs. At 24 h after airway delivery of transfected J774A.1 macrophages, lung sections were evaluated by immunohistochemical staining for hA1AT. A1AT-expressing macrophages (see arrows) were easily detected (× 200 magnification; inset is × 400 magnification).

Full size image

Our study demonstrates a potential method to increase A1AT levels in the alveolar space in human patients deficient in A1AT. Other methods that augment A1AT levels include direct administration of the protein,3,4 drug therapies that induce endogenous A1AT synthesis and secretion27,28 and in more recent efforts, gene therapy to target the respiratory epithelium or nonrespiratory organ systems such as hepatocytes and skeletal muscle.6,7,8,9,11,12,17,22

Intravenous augmentation therapy using pooled human plasma A1AT infusion has been available since 1989. However, since only 2–3% of plasma A1AT can be detected in the lungs after intravenous administration, a high dosage is necessary to maintain adequate serum levels.3,4 Danazol, an androgenic drug, increased serum A1AT concentration by 37–52% in subjects with A1AT deficiency; however, A1AT levels did not reach therapeutic levels.3 In addition, adverse effects such as weight gain, water retention and acne were common.4,29

In previous gene-therapy studies targeting the respiratory system, the A1AT gene was successfully transferred to the airway epithelium, as demonstrated by mRNA expression, but A1AT protein levels in the lung tissue were either very low or not documented.11,12 Gene therapy has also been directed to hepatocytes or skeletal muscles using an AAV vector system resulting in high levels of serum A1AT,8,17,22 but the A1AT levels in the lung tissue were not determined. This may be important as A1AT can be secreted by alveolar macrophages30 and alveolar epithelial cells,31,32 and it has been speculated that high intra-alveolar level of A1AT may potentially dampen an acute inflammatory response in the lungs.31

It is unclear how plasma A1AT diffuses across the alveolar capillary endothelial barrier into the interstitium and across the epithelial barrier into the alveolar spaces. In humans, A1AT levels are 20–53 μM in the serum, 10–40 μM in the interstitial fluid and 2–5 μM in the alveolar spaces.33 Thus, therapeutic strategies that increase serum levels are limited by the small fraction of serum A1AT that crosses to the pulmonary interstitium and alveolar spaces.33 In contrast, when A1AT was administered by inhalation of A1AT aerosol in a study with healthy volunteers, the half-life of A1AT protein in the lungs and antiprotease activity were approximately double the baseline values.34 In another recent study in patients with A1AT deficiency, A1AT was easily detected in the periphery of the lungs after inhalation of an A1AT aerosol preparation.35 This study did not address the legitimate concern of whether A1AT delivered via the airway will effectively cross the epithelial barrier to reach the interstitium,33 but a transgenic mouse study showed that A1AT secreted in situ by pulmonary epithelium resulted in significant transport from alveolar spaces to the pulmonary interstitium.36

Airway delivery of normal alveolar macrophages or genetically engineered macrophages into the lungs of immunodeficient mice potentially restores immune function in the lungs.21,37 One advantage to the use of gene transfer in vitro with subsequent airway delivery of transduced cells is to minimize some of the adverse immune reactions associated with in vivo gene transfer with viral vector. In vivo gene transfer using nonviral system may also avoid such concomitant expression of potential immunogenic antigens.5,20 Our data demonstrate that in vitro transfer of the hA1AT gene results in significantly and persistently elevated hA1AT within the lung with only minimal inflammatory response. This novel method for gene transfer using transfected macrophages may be useful as a therapeutic approach for A1AT deficiency in human subjects. Additional studies are required to determine if this method can protect alveolar structures from elastolytic damage.