This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chen, J.
Right arrow Articles by An, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chen, J.
Right arrow Articles by An, Y.

 Previous Article  |  Next Article 

Antimicrobial Agents and Chemotherapy, February 2007, p. 724-731, Vol. 51, No. 2
0066-4804/07/$08.00+0     doi:10.1128/AAC.00360-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Protection of Mice from Lethal Escherichia coli Infection by Chimeric Human Bactericidal/Permeability-Increasing Protein and Immunoglobulin G1 Fc Gene Delivery{triangledown}

Jindong Chen,1,2 Chengyao Li,3 Yuanzhi Guan,4 Qingli Kong,1 Chen Li,1 Xianghua Guo,1 Qinghua Chen,1 Xuefang Jing,1 Zhe Lv,1 and Yunqing An1*

Department of Microbiology and Immunology, Capital University of Medical Sciences, Beijing, China,1 Department of Immunology, Cancer Institute and Cancer Hospital of CAMS and PUMC, Beijing, China,2 Division of Transfusion Medicine, National Blood Service and University of Cambridge, Cambridge CB2 2PT, United Kingdom,3 Department of Microbiology, Institute of Basic Medical Sciences of CAMS and PUMC, Beijing, China4

Received 13 March 2006/ Returned for modification 6 August 2006/ Accepted 22 November 2006


arrow
ABSTRACT
 
To evaluate the potentiality of applying gene therapy to bacterial infections, especially for preventing infection in high-risk patients, we investigated protection of mice from challenge with lethal Escherichia coli infection by adeno-associated virus serotype 2 (AAV2)-mediated gene transfer of a chimeric BPI23-Fc{gamma}1 gene, which consisted of human bactericidal/permeability-increasing protein (BPI) gene encoding the functional N terminus (amino acid residues 1 to 199) of human BPI and an Fc{gamma}1 gene encoding the Fc segment of human immunoglobulin G1. Here we show that the target protein that was expressed and secreted into the serum of the gene-transferred mice demonstrated the activity of a neutralizing endotoxin, killing E. coli and mediating opsonization. After lethal E. coli infection, the count of bacteria and the levels of endotoxin and proinflammatory cytokines in the gene-transferred mice were decreased. The survival rate of BPI23-Fc{gamma}1 gene-transferred mice markedly increased, especially in conjunction with antibiotics. Our data suggest that AAV2-mediated chimeric BPI23-Fc{gamma}1 gene delivery could potentially be used clinically for the protection and treatment of infection with gram-negative bacteria in high-risk individuals.


arrow
INTRODUCTION
 
Infections by gram-negative bacteria (GNB) are common in clinics, and cases of infection that develop into sepsis and subsequent endotoxic shock, with high mortality (over 100,000 deaths in the United States annually), have increased in recent decades. The increase can be attributed to an escalating number of patients who are undergoing cancer therapy, immunosuppressive therapy, invasive surgical procedures, and human immunodeficiency virus infection, who are at high risk for developing sepsis (5, 18, 23). In addition, GNB lysis may excessively release endotoxin and cause endotoxemia during antibiotic therapy. Various novel sepsis therapies currently under development and evaluation in clinical trials, including anticoagulant therapy, neutralization of lipopolysaccharide (LPS), and cytokine therapies, have not progressed essentially (10, 19, 26, 30).

Bactericidal/permeability-increasing protein (BPI) is a 55- to 60-kDa human neutrophil granule-associated defense molecule specific for gram-negative bacteria, which was found in 1978 (13, 33, 35). BPI has the specific effect of neutralizing endotoxin and directly killing GNB, but has no adverse effect on eukaryotic cells. It was demonstrated that the N terminus of BPI is identical to natural BPI in its effect on LPS and GNB (12, 15, 25, 34). Recent studies with animal models of sepsis and endotoxemia and clinical trials treating septic patients suggested that the recombinant N terminus of BPI (rBPI21) was a promising therapeutic agent. However, rBPI21 has relatively low efficacy and short half-life in vivo, administration of rBPI21 in a large dosage is very expensive, and it is difficult to maintain an optimal therapeutic level (6, 11). In addition, recombinant BPI21 and conventional antibiotics are principally suited to the treatment of existing bacterial infection rather than prevention of high-risk patients from developing sepsis.

In order to prolong and improve the activity of recombinant BPI21 for clinical therapy of GNB infection, we applied a strategy (CAP18-immunoglobulin [Ig] fusion protein) similar to that of Warren and colleagues (32) to design and express a recombinant chimeric BPI23-Fc{gamma}1 protein that consisted of the functional N terminus (amino acid residues 1 to 199) of human BPI and the Fc segment of human IgG1. It has been demonstrated that the chimeric BPI23-Fc{gamma}1 protein has the effect of neutralizing endotoxin, directly killing GNB (including drug-resistant GNB), as well as mediating opsonization (2). Based on our preliminary work, we have sought to develop a BPI23-Fc{gamma}1 transgene-based modality and to evaluate its potential in preventing GNB infection of clinical high-risk patients and, accordingly, in reducing the mortality of sepsis caused by GNB. In this study, the chimeric BPI23-Fc{gamma}1 gene was reconstructed within a recombinant adeno-associated virus serotype 2 (rAAV2) vector as rAAV2-BPI23-Fc{gamma}1, and subsequently delivered and expressed both in vitro and in vivo. The protective efficacy of chimeric BPI23-Fc{gamma}1 gene delivery mediated by AAV2 against lethal Escherichia coli infection in the gene-transferred mice was fully characterized.


arrow
MATERIALS AND METHODS
 
Construction and production of rAAV2-BPI23-Fc{gamma}1. The BPI gene fragment encoding the signal peptide and the functional N terminus (amino acid residues 1 to 199) of human BPI, named BPI23, was generated by reverse transcription-PCR (RT-PCR) using the primers P1 (5'-CTGGTACCATGAGAGAGAACATGGCCA-3') and P2 (5'-GCAAGCTTCTATTTTGGTCATTACTGGCAG-3') from the mRNA of HL-60 cell line (ATCC CCL-240; ATCC, Manassas, VA). The Fc{gamma}1 gene fragment encoding the Fc fragment of human immunoglobulin G1 was generated by RT-PCR using the primers P3 (5'-GTAAGCTTCTACATGCCCACCGTGCCCAG-3') and P4 (5'-TCGTCGACGGATCCTTATTTACCCGGAGACAGGGAG-3') from the mRNA of human peripheral blood lymphocytes derived from a healthy volunteer. BPI23 and Fc{gamma}1 DNA fragments were then digested with KpnI/HindIII and HindIII/SalI, respectively, and were then coligated into the KpnI/SalI sites of the pSNAV vector (AGTC Gene Technology Co. Ltd., Beijing, China) (38), designated as the pSNAV-BPI23-Fc{gamma}1 expression vector, in which the chimeric BPI23-Fc{gamma}1 gene controlled by the cytomegalovirus (CMV) promoter and simian virus 40 (SV40) poly(A) was flanked by AAV serotype 2 inverted terminal repeats.

rAAV2-BPI23-Fc{gamma}1 viruses were prepared by AGTC Gene Technology Co. Ltd., complying with the guidelines of SFDA and GMP facilities, according to the protocols described previously (36, 37). Briefly, BHK-21 cells (ATCC CCL-10) were transfected with the pSNAV-BPI23-Fc{gamma}1 plasmid DNA using Metafectene (Biontex Laboratories GmbH, Munich, Germany) and selected by G418. rAAV2-BPI23-Fc{gamma}1 viruses were rescued and produced by infecting the G418-resistant BHK-21 clones containing a BPI23-Fc{gamma}1 gene with recombinant HSV1-rc/{Delta}UL2 helper viruses (AGTC Gene Technology Co. Ltd.) (37). The rAAV2-BPI23-Fc{gamma}1 viruses were purified and diluted to the concentration of 1 x 1012 vector genomes (v.g.)/ml used for the study.

Verification and expression of rAAV2-BPI23-Fc{gamma}1 in CHO cells. rAAV2-BPI23-Fc{gamma}1 virus at multiplicities of infection (MOI) of 5 x 104, 1 x 105, and 5 x 105 v.g./cell was used to infect CHO-K1 cells (ATCC CCL-61). The infected cells were incubated for 48 to 72 h in serum-free Dulbecco's modified Eagle's medium-F-12 medium at 37°C and with 5% CO2. The supernatants were analyzed for the presence of secreted BPI23-Fc{gamma}1 protein by dot blotting and Western blot analysis using horseradish peroxidase (HRP)-conjugated goat anti-human IgG antibody (Sigma, St. Louis, MO) and chemiluminescent substrate (Pierce Biotech Inc., Rockford, IL).

Mouse models of gene transfer. Five- to 6-week-old female BALB/c mice (provided by the Laboratory Animal Centre of The Academy of Military Medical Sciences, Beijing, China) were used to develop mouse models of gene transfer by AAV2. rAAV2-BPI23-Fc{gamma}1 gene-transferred mice were administered a 100-µl injection containing 1 x 1011 v.g. of rAAV2-BPI23-Fc{gamma}1 through the quadriceps muscles of the right hind leg; rAAV2-EGFP gene-transferred mice were administered a 100-µl injection containing 1 x 1011 v.g. of rAAV2-enhanced green fluorescent protein (EGFP)-expressing virus (AGTC Gene Technology Co. Ltd.); phosphate-buffered saline (PBS)-treated control mice were administered a 100-µl injection of PBS. All experiments with gene-transferred mice described below were performed at an interval of 2 weeks after these injections were administered.

MLD of endotoxin or E. coli to BALB/c mice. LPS (Sigma) was diluted to 9.0 µg/ml, 7.5 µg/ml, 6.0 µg/ml, and 4.5 µg/ml with PBS containing 60 mg/ml D-galactosamine (Sigma). One hundred microliters of the selected dose of LPS was intraperitioneally injected into four separate groups of mice. The minimal dose that caused mortality of 90 to 100% of mice within 48 h was determined as the minimal lethal dose (MLD) of LPS for BALB/c mice.

E. coli O111:B4 [CMCC (B) no. 44101-9; CMCC, Beijing, China] was diluted to 2.5 x 105, 5 x 104, 2.5 x 104, and 5 x 103 CFU/0.5 ml with autoclaved PBS buffer containing 5% (wt/vol) dried yeast. A 0.5-ml portion of the selected dose of E. coli for each mouse was intraperitoneally injected into four separate groups of mice. The minimal dose that caused 90 to 100% mortality within 48 h was determined as the MLD of E. coli O111:B4 for BALB/c mice.

RT-PCR. mRNA was extracted from rAAV2-BPI23-Fc{gamma}1-injected mouse muscles with the Oligotex direct mRNA kit (QIAGEN, Hilden, Germany). RT-PCR was performed according to the manufacturer's instruction for the Access RT-PCR system (Promega, Madison, WI) in order to detect BPI23-Fc{gamma}1 gene expression at the mRNA level. The specific primers used in RT-PCR are P1 and P4, as described above.

Immunohistochemical and histopathological observation. The paraffin-embedded sections of the injected mouse muscles were prepared and analyzed by standard immunohistochemical staining with HRP-conjugated mouse anti-human IgG Fc (Zymed Laboratories, Inc., San Francisco, CA) and diaminobenzidine (DAB) (Boster Biotech, Wuhan, China).

The paraffin-embedded sections of the main tissues of liver, small intestine, spleen, and kidney at 24 h after lethal E. coli challenge were prepared and examined by standard hematoxylin and eosin (H&E) staining.

ELISA. A modified enzyme-linked immunosorbent assay (ELISA) was performed as follows to examine the secreted BPI23-Fc{gamma}1 protein in mouse sera. One hundred microliters of serum from rAVV2-BPI23-Fc{gamma}1 gene-transferred mice was fully absorbed by nitrocellulose membrane for 15 min, while 100 µl of serum from rAVV2-EGFP gene-transferred mice and 100 µl of serum from PBS control mice were used as controls. The nitrocellulose membranes were dried in air for 20 min and laid on naked microtiter plates; these were tested with a biotinylated antibody against human BPI according to the protocol of a human BPI ELISA kit (HyCult Biotechnology b.v., Uden, The Netherlands).

Proinflammatory cytokines, interleukin-1ß (IL-1ß) and tumor necrosis factor alpha (TNF-{alpha}) in serum samples were detected by ELISA according to the instructions of a kit from R&D Systems Inc., Minneapolis, MN.

LAL assay. Serum samples were diluted in pyrogen-free water and then tested according to the instructions of the Limulus amebocyte lysate (LAL) kit (Shyihua Corp., Shanghai, China).

Bacterial count assay. For the in vitro bacterial count assay, 100 µl of serum or anticoagulated blood from rAAV2-BPI23-Fc{gamma}1 gene-transferred mice or rAAV2-EGFP gene-transferred mice was mixed with 100 µl of E. coli (1 x 103 CFU/ml), incubated at room temperature for 15 min, and then plated on two LB agar plates. Bacterial counts were performed after overnight incubation at 37°C from three individual experiments.

For the in vivo bacterial count assay, after E. coli attack, serum samples and homogenated samples from spleens and livers were serially diluted for the bacterial count and then dilution samples were inoculated onto two LB agar plates and incubated at 37°C for 24 h. Clones were counted, and the average from two plates was calculated in order to determine the bacterial count.

Statistical analysis. Data are presented as means ± standard deviation (SD). A chi-square test was performed for the survival rate comparison. Differences among groups were analyzed by an independent-samples t test. ({alpha} = 0.05, two-sided). P values of <0.05 were considered statistically significant.


arrow
RESULTS
 
Construction and verification of rAAV2-BPI23-Fc{gamma}1. We constructed the pSNAV-BPI23-Fc{gamma}1 expression vector and successfully produced recombinant AAV2-BPI23-Fc{gamma}1 virus as mentioned in Materials and Methods. The expression cassette of chimeric BPI23-Fc{gamma}1 gene controlled by the CMV promoter and SV40 poly(A) is flanked by AAV2 inverted terminal repeats (Fig. 1). The rAAV2-BPI23-Fc{gamma}1 virus was purified and diluted to 1 x 1012 v.g./ml as used for the study.


Figure 1
View larger version (11K):
[in this window]
[in a new window]

  		FIG. 1. Gene structure of rAAV2-BPI23-Fc{gamma}1. ITR, inverted terminal repeat.

First, we verified the integrity of our vectors in vitro. About 60% confluent CHO-K1 (ATCC CCL-61) cells were infected with rAAV2-BPI23-Fc{gamma}1 virus at various MOI and then cultured in serum-free medium for 48 h. Dot blot analysis demonstrated the presence of secreted BPI23-Fc{gamma}1 protein in the conditioned medium in a virus load-dependent manner (Fig. 2A). Western blot analysis demonstrated the presence of a 48-kDa band in dithiothreitol (DTT)-deoxidized medium and a 96-kDa band in nondeoxidized medium, which matches the expected size of BPI23-Fc{gamma}1 protein (Fig. 2B). It also was found that the secreted BPI23-Fc{gamma}1 protein in the conditioned medium of CHO-K1 cells infected by rAAV2-BPI23-Fc{gamma}1 virus had high efficacy of killing GNB (including drug-resistant GNB), neutralizing endotoxin, and mediating opsonization in vitro as well as of protecting mice from lethal E. coli infection in vivo (data not shown).


Figure 2
View larger version (46K):
[in this window]
[in a new window]

  		FIG. 2. Expression of chimeric BPI23-Fc{gamma}1 protein in CHO cells infected with rAAV2-BPI23-Fc{gamma}1 virus. Shown are the results of dot and Western blot analysis of the conditioned medium of CHO-K1 infected by rAAV2-BPI23-Fc{gamma}1 virus at various MOI. (A) Three-microliter medium samples were dotted. The supernatants of CHO-K1 cells without infection were dotted as a negative control (dot 1). The MOI were 5 x 104 (dot 2), 1 x 105 (dot 3), and 5 x 105 (dot 4) v.g./cell, respectively. Human IgG1 at 0.1 µg was dotted as a positive control (dot 5). (B) Western blot analysis of the 10-fold-concentrated medium of CHO-K1 infected with rAAV2-BPI23-Fc{gamma}1 virus at 5 x 105 v.g./cell (MOI). Lane 1, prestained protein molecular mass (M) marker; lane 2, DTT treated (reduced); and lane 3, non-DTT treated (nonreduced).

BPI23-Fc{gamma}1 expression in vivo. The injected muscles were examined by RT-PCR and immunohistochemical staining to identify target gene expression at mRNA and protein levels in rAAV2-BPI23-Fc{gamma}1 gene-transferred mice 2 weeks after intramuscular administration. An expected 1.4-kb-size band was found by RT-PCR using the primers 5'-CTGGTACCATGAGAGAGAACATGGCCA-3'and 5'-TCGTCGACGGATCCTTATTTACCCGGAGACAGGGAG-3'for rAAV2-BPI23-Fc{gamma}1 gene-transferred mice.

DAB-positive staining was observed in rAAV2-BPI23-Fc{gamma}1 gene-transferred mice, but not in rAAV2-EGFP gene-transferred mice and PBS control mice (Fig. 3). Green fluorescence was observed in the sections of the injected muscle of rAAV2-EGFP gene-transferred control mice. The results indicated that the target gene was successfully expressed in the injected muscles of rAAV2-BPI23-Fc{gamma}1 gene-transferred mice 2 weeks after injection.


Figure 3
View larger version (45K):
[in this window]
[in a new window]

  		FIG. 3. In situ expression of BPI23-Fc{gamma}1 protein detected by immunohistochemical staining. (A to C) Immunohistochemical staining of the injected muscles from PBS control mice (A), rAAV2-EGFP gene-transferred mice (B), and rAAV2-BPI23-Fc{gamma}1 gene-transferred mice (C). (D) Green fluorescence in the section of the injected muscles from rAAV2-EGFP gene-transferred mice under epifluorescent microscopy. DAB staining; original magnification, x200.

Detection of secreted BPI23-Fc{gamma}1 in serum. Serum samples were collected 2 weeks after administration of vector genomes. The following experiments were performed at the same time.

First, the serum was analyzed by a modified ELISA to detect the secreted BPI23-Fc{gamma}1 protein in serum. The optical density at 450 nm (OD450) of the serum from rAAV2-BPI23-Fc{gamma}1 gene-transferred mice was 0.849±0.164 (n = 3), while those for the serum from rAAV2-EGFP gene-transferred control mice and PBS control mice were 0.283 ± 0.026 (n = 3) and 0.290 ± 0.020 (n = 3), respectively. There was a statistically significant difference between rAAV2-BPI23-Fc{gamma}1 gene-transferred mice and the control mice (P < 0.05) but not between rAAV2-EGFP gene-transferred control mice and PBS control mice (t = 0.23, P = 0.832). It was proven that there was secreted BPI23-Fc{gamma}1 protein in the serum of rAAV2-BPI23-Fc{gamma}1 gene-transferred mice.

The mouse sera were precipitated with ammonium sulfate solution. The intactness of BPI23-Fc{gamma}1 protein in the precipitated sera was confirmed by Western blotting. The result is shown in Fig. 4. The explanations are as follows. (i) An expected 96-kDa band was shown on the film in the rAAV2-BPI23-Fc{gamma}1-transferred mice, but not in the rAAV2-EGFP control mice. (ii) A heavily stained 140-kDa band was shown on the film both in rAAV2-BPI23-Fc{gamma}1-transferred mice and in the control mice, which proved that anti-human IgG antibody labeled by the HRP does cross-react with the mouse IgG.


Figure 4
View larger version (75K):
[in this window]
[in a new window]

  		FIG. 4. Intactness of BPI23-Fc{gamma}1 protein in the murine sera as identified by Western blotting. Lane 1, serum from rAAV-BPI23-Fc{gamma}1 mice; lane 2, serum from rAAV-EGFP control mice.

Detection of BPI23-Fc{gamma}1 protein activity in vitro. To assess the endotoxin-neutralizing activity of secreted BPI23-Fc{gamma}1 protein in serum, 50 µl serum was incubated with 50 µl endotoxin solution (0.25 endotoxin units [EU]/ml) diluted with pyrogen-free water for 30 min at 37°C, and then the mixture was tested with the LAL assay. LPS standard in pyrogen-free water (0.125 EU/ml LPS standard) had an OD545 of 0.320 ± 0.03 (n = 3). However, the same quantity of LPS in the dilution containing 50% rAAV2-EGFP control mouse serum had an OD545 of 0.287 ± 0.021 (n = 3), which is lower than that of the standard. The result proved, as reported in the literature (16, 26, 28), that some components (high-density lipoprotein; cathelicidin) in mouse serum are able to neutralize LPS. Nevertheless, the LPS sample containing serum from rAAV2-BPI23-Fc{gamma}1 gene-transferred mice had an OD545 of 0.173 ± 0.021(n = 3), which is significantly lower than the value of 0.287 ± 0.021(n = 3) from the control group containing normal mouse serum. Thus, we believe that the target gene expression creates BPI23-Fc{gamma}1 protein to neutralize LPS, although there are some nonspecific LPS-neutralizing activities in normal mouse serum.

Meanwhile, we also verified protection of mice from lethal endotoxin (LPS) attack as follows. First, we titrated the MLD of LPS (600 ng/mouse) for BALB/c mice. Then, the survival rate was observed within 48 h after lethal LPS challenge. The survival rate of rAAV2-BPI23-Fc{gamma}1 gene-transferred mice (40%; n = 20) was significantly higher than those of rAAV2-EGFP gene-transferred control mice (5%; n = 20) and PBS control mice (0%; n = 20) (20). The results also suggested that the target product in rAAV2-BPI23-Fc{gamma}1 gene-transferred mice could neutralize endotoxin (LPS) and protect mice from the challenge of lethal endotoxemia.

We further assessed the bactericidal activity and the opsonization effect of secreted BPI23-Fc{gamma}1 in serum on E. coli O111:B4 [CMCC (B) no. 44101-9]. As shown in Table 1, the bacterial counts from rAAV2-BPI23-Fc{gamma}1 gene-transferred mice (21.33 ± 2.08 and 13.67±3.06) were significantly lower than those from rAAV2-EGFP gene-transferred mice (33.33 ±4.93 and 34.67 ± 5.51) either in serum (P = 0.018) or in anticoagulated blood (P = 0.002). The bacterial count from anticoagulated blood in rAAV2-BPI23-Fc{gamma}1 gene-transferred mice (13.67 ± 3.06) was significantly lower than that from serum in rAAV2-BPI23-Fc{gamma}1 gene-transferred mice (21.33 ± 2.08) (P = 0.037). However, there was no difference between bacterial counts from serum and anticoagulated blood in the rAAV2-EGFP gene-transferred control mice (P = 0.770). These results suggested that while the secreted chimeric BPI23-Fc{gamma}1 protein alone could kill E. coli, the capacity to kill E. coli could be improved significantly in the presence of phagocytes; in contrast, phagocytes alone were unable to kill E. coli under the experimental conditions.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Comparison of E. coli counts in serum or anticoagulated blood between rAAV2-BPI23-Fc{gamma}1 and rAAV2-EGFP gene-transferred mice

Protection of mice from lethal E. coli infection. To titrate the MLD of E. coli O111:B4 for BALB/c mice, different doses of E. coli O111:B4 were administered intraperitoneally to four groups of BALB/c mice (n = 10). The infected mice began to die 18 h after infection, and the mortality rates of the group injected with the doses of 2.5 x 105, 5 x 104, 2.5 x 104, and 5 x 103 CFU/0.5 ml E. coli O111:B4 within 48 h were 100% (10/10), 100% (10/10), 92.5% ± 5% (37/40), and 60%(10/10), respectively. Hereby, 2.5 x 104 CFU/0.5 ml was determined to be the MLD of E. coli O111:B4 for BALB/c mice.

To prove protection of mice from challenge of lethal E. coli infection, BALB/c mice were attacked with an MLD of E. coli O111:B4 intraperitoneally 2 weeks after transfer with a 100-µl injection containing 1 x 1011 v.g. of rAAV2-BPI23-Fc{gamma}1 virus intramuscularly, while other mice were gene transferred with a 100-µl injection containing 1 x 1011 v.g./100 µl of rAAV2-EGFP virus or injected with 100 µl PBS as controls. Then, the survival rates were observed within 48 h after infection with the MLD of E. coli, and reproducible results were obtained (Table 2). The survival rate of rAAV2-BPI23-Fc{gamma}1 gene-transferred mice (37.50%) was significantly higher than those of rAAV2-EGFP gene-transferred mice (2.50%) and PBS control mice (4.17%), while there was no statistically significant difference between the survival rates of rAAV2-EGFP gene-transferred control mice and PBS control mice. The results demonstrated that rAAV2-mediated BPI23-Fc{gamma}1 gene transfer protected mice from the challenge of lethal E. coli O111:B4 infection.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Protection of rAAV2-BPI23-Fc{gamma}1 gene-transferred mice from lethal E. coli challenge

Furthermore, data in Table 2 also show that the survival ratio of rAAV2-BPI23-Fc{gamma}1 gene-transferred mice administered the minimal dose of antibiotic (25 µg cefuroxime per mouse) was markedly increased to 67.50%. The survival ratios of the cefuroxime group and rAAV2-EGFP-cefuroxime group were 5.00% and 7.50%, respectively. This indicates the potential for cooperative use of rAAV2-mediated BPI23-Fc{gamma}1 gene delivery and the minimal dose of antibiotics in the clinical scenario to improve the therapy of GNB infections.

Biological functions of secreted BPI23-Fc{gamma}1 and resulting changes in vivo. To further evaluate the biological activity of secreted BPI23-Fc{gamma}1 protein in rAAV2-BPI23-Fc{gamma}1 gene-transferred mice, the levels of bacterial count, endotoxin, and proinflammatory cytokines and histological alterations in vivo were measured after lethal E. coli challenge. First, blood samples from orbital bulb and main viscus were simultaneously collected at 6, 9, 12, and 24 h after the MLD of E. coli O111:B4 infection, and then the serum samples from the collected blood samples and the homogenated samples from the collected integrated spleens and livers were prepared. Each serum sample and homogenated sample was detected by bacterium-counting assay. It was shown that the counts of bacteria in the serum, spleen, and liver of rAAV2-BPI23-Fc{gamma}1 gene-transferred mice were markedly lower than those for rAAV2-EGFP gene-transferred control mice (Fig. 5A, B, and C). The results demonstrated that the target product in rAAV2-BPI23-Fc{gamma}1 gene-transferred mice could kill E. coli and improve the resistance of mice against lethal E. coli infection.


Figure 5
View larger version (13K):
[in this window]
[in a new window]

  		FIG. 5. Bacterial counts in rAAV2-BPI23-Fc{gamma}1 and rAAV2-EGFP gene-transferred mice with lethal E. coli infection. According to the time point, bacterial counts (CFU) were performed for serum (A), homogenated spleen (B), and liver (C) samples from two groups of mice after injection of lethal E. coli. •, rAAV2-BPI23-Fc{gamma}1 gene-transferred mice; {circ}, rAAV2-EGFP control mice.

Second, blood samples were collected from orbital bulb at 6, 12, 18, and 24 h after the MLD of E. coli O111:B4 infection and serum samples of three mice from the same group were prepared and mixed together. Each 100 µl of the mixed serum sample was detected by LAL assay for endotoxin and by ELISA for proinflammatory cytokines. The level of endotoxin in the serum of rAAV2-BPI23-Fc{gamma}1 gene-transferred mice reached its peak at 12 h and was significantly lower than that of rAAV2-EGFP gene-transferred control mice, with a peak at 18 h (Fig. 6A). Correspondingly, the levels of IL-1ß and TNF-{alpha} in the serum of rAAV2-BPI23-Fc{gamma}1 gene-transferred mice reached a peak at 12 h and was significantly lower than that of rAAV2-EGFP gene-transferred control mice, with a peak at 18 h after lethal E. coli infection (Fig. 6B and C). It was obvious that the markedly increasing levels of endotoxin and proinflammatory cytokines in serum of rAAV2-EGFP-transferred control mice with a peak at 18 h were responsible for the death of the animals (mortality up to 92.5%) occurring 18 h after E. coli infection. The results suggested that rAAV2-BPI23-Fc{gamma}1 gene-transferred mice could resist the endotoxic shock caused by lethal E. coli infection through killing E. coli, neutralizing endotoxin, and decreasing the level of proinflammatory cytokines.


Figure 6
View larger version (13K):
[in this window]
[in a new window]

  		FIG. 6. Levels of endotoxin and proinflammatory cytokines in rAAV2-BPI23-Fc{gamma}1 and rAAV2-EGFP gene-transferred mice after lethal E. coli infection. The quantities of endotoxin (A), IL-1ß (B), and TNF-{alpha} (C) in sera were measured from the time points shown by LAL or ELISA. •, rAAV2-BPI23-Fc{gamma}1 gene-transferred mice; {circ}, rAAV2-EGFP control mice.

In addition, the main viscus of the experimental mice, including liver, small intestine, spleen, and kidney was examined by standard H&E staining 24 h after challenge with lethal E. coli infection. In comparison, the main viscus of the surviving mice protected by rAAV2-BPI23-Fc{gamma}1 gene transfer showed only a slight congestion, while the main viscus of the agonal mice from rAAV2-EGFP-transferred control mice showed significant pathological alterations, such as capillary dilatation and congestion, which were consistent with what endotoxic shock should show.


arrow
DISCUSSION
 
The viral delivery system has been widely used in gene therapy protocols for its high efficiency (3). A "gutless" viral vector is safer because there is less oncogenicity and less immunogenicity (7, 17) and is suitable for gene therapy of bacterial infections because it does not exaggerate the inflammatory reaction caused by infection. AAV vectors, a kind of gutless vector, are based upon a class of viruses that commonly inhabit a human host without causing any detectable pathology. In particular, AAV2 has been widely used as a gene delivery vehicle in preclinical studies and its use in early-phase clinical trials has been reported (8, 9). AAV2-mediated gene delivery has a slow but long-term gene expression which reaches the peak at the 2nd to 3rd week and then persists more than several months after delivery (9, 14). In this study, we produced rAAV2-BPI23-Fc{gamma}1 virus with a high viral load that successfully mediated BPI23-Fc{gamma}1 gene transfer and expression in mouse muscle cells (Fig. 3): it was further secreted into blood circulation, suggesting that the AAV2-mediated BPI23-Fc{gamma}1 gene delivery system may be suitable for administration to patients at high risk of infection.

By 2004, over 700 gene therapy/transfer protocols, which cover cancer, monogenic diseases, and viral infections, primarily peripheral and coronary artery diseases, but not bacterial infections, had been initiated worldwide (29). The latest studies have shown that adenovirus-mediated full-length BPI gene transfer could protect mice from endotoxemia but not from lethal E. coli infection (1). The half-life of recombinant BPI in vivo was determined to be less 45 min; however, BPI itself required 3 h or more to kill bacteria, which might explain why BPI alone was not able to protect BPI gene-transferred mice from lethal E. coli challenge. In comparison with BPI, chimeric BPI23-Fc{gamma}1 protein not only had a longer half-life, but also accumulated in blood as it acted like immunoglobulin. The secreted BPI23-Fc{gamma}1 protein in the serum of rAAV2-BPI23-Fc{gamma}1 gene-transferred mice displayed the effects of neutralizing endotoxin, killing E. coli and mediating opsonization (Table 1). After challenge with lethal E. coli O111:B4 infection, the count of bacteria in serum and in main viscus, as well as the level of endotoxin and proinflammatory cytokines in serum of rAAV2-BPI23-Fc{gamma}1 gene-transferred mice decreased (Fig. 5 and 6), and the survival rate of rAAV2-BPI23-Fc{gamma}1 gene-transferred mice increased markedly, especially when combined with antibiotics (Table 2), which was similar to the findings in previous studies on synergism of recombinant BPI and antibiotics (6, 11), implying that the efficacy of anti-GNB infection could be maximized by delivering the BPI23-Fc{gamma}1 gene and administering minimal amounts of antibiotics in patients. In addition, corresponding to the markedly increased levels of circulating endotoxin and proinflammatory cytokines, the rAAV2-EGFP control group mice had high mortality (above 90%) during lethal E. coli infection, and the agonal mice in the rAAV2-EGFP control group showed significant histological alterations, such as capillary dilation and congestion in the main viscus, which were consistent with the clinical finding of endotoxic shock.

The results of the experiment indicate AAV2-mediated BPI23-Fc{gamma}1 gene delivery has potential for preventing clinical high-risk patients from being infected by GNB. In general, before reaching the MLD, the invading GNB in high-risk patients may be effectively eliminated by rAAV2-mediated BPI23-Fc{gamma}1 gene transfer. In addition, compared with traditional antibiotics, BPI23-Fc{gamma}1 protein has the advantages of neutralizing endotoxin that can protect host from endotoxemia and endotoxic shock and of killing drug-resistant GNB with dual pathways by BPI's direct killing and Fc{gamma}1-mediated opsonization (11, 22). These findings highlight how chimeric BPI23-Fc{gamma}1 protein can induce LPS-anchored phagocytosis by phagocytes in order to kill GNB in vivo. With the success of the rAAV2-BPI23-Fc{gamma}1 gene transfer modality against GNB infection in mice model, we believe that rAAV2-BPI23-Fc{gamma}1 gene transfer can protect high-risk patients from serious GNB infection and sepsis. We also consider that BPI23-Fc{gamma}1 gene delivery by double-stranded AAV vector or mini-adenovirus vector (4, 21, 24, 27, 31), as well as other kinds of gutless vectors developed recently, can enable the therapeutic gene to be expressed more quickly and strongly than the single-stranded AAV vector, so it will have quick and strong effects against GNB infection and will show special potential for treatment and prophylaxis.


arrow
ACKNOWLEDGMENTS
 
This study was supported by the Beijing Municipality Natural Science Foundation.

We thank Daniel Candotti and Lara Compston, National Blood Service and Cambridge University, Cambridge, England, for constructive suggestions and revision of the English in the manuscript.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Department of Microbiology and Immunology, Capital University of Medical Sciences, Beijing 100069, China. Phone: 86-10-83911439. Fax: 011-86-10-83911439. E-mail: anyunq{at}ccmu.edu.cn. Back

{triangledown} Published ahead of print on 4 December 2006. Back


arrow
REFERENCES
 
    1
  1. Alexander, S., J. Bramson, R. Foley, and Z. Xing. 2004. Protection from endotoxemia by adenoviral-mediated gene transfer of human bactericidal/permeability-increasing protein.Blood 103:93-99.[Abstract/Free Full Text]
    2. 2
  2. An, Y. Q., Y. Z. Guan, Y. Ke, and G. Z. Yang. 2002. Construction, expression and biological assessment of BPI23-Fc{gamma}1 recombinant protein prokaryotic expression vector. Chin. Med. Sci. J. 17:140-147.[Medline]
    3. 3
  3. Anderson, W. F. 1998. Human gene therapy.Nature 392:25-30.[CrossRef][Medline]
    4. 4
  4. Balague, C., J. Zhou, Y. Dai, R. Alemany, S. F. Josephs, G. Andreason, M. Hariharan, E. Sethi, E. Prokopenko, H. Y. Jan, Y. C. Lou, D. Hubert-Leslie, L. Ruiz, and W. W. Zhang. 2000. Sustained high-level expression of full-length human factor VIII and restoration of clotting activity in hemophilic mice using a minimal adenovirus vector. Blood 95:820-828.[Abstract/Free Full Text]
    5. 5
  5. Balk, R. A. 2000. Severe sepsis and septic shock. Definitions, epidemiology, and clinical manifestations. Crit. Care Clin. 16:179-192.[CrossRef][Medline]
    6. 6
  6. Beamer, L. 2002. Human BPI: one protein's journey from lab into clinical trials. ASM News 68:543-548.
    7. 7
  7. Bessis, N., F. J. Garcia Cozar, and M. C. Boissier.2004 . Immune responses to gene therapy vectors: influence on vector function and effectors mechanisms. Gene Ther. 11(Suppl. 1):S10-S17.
    8. 8
  8. Buning, H., F. M. Braun-Falco, and M. Hallek. 2004. Progress in the use of adeno-associated viral vectors for gene therapy.Cells Tissues Organs 177:139-150.[CrossRef][Medline]
    9. 9
  9. Buning, H., S. A. Nicklin, L. Perabo, M. Hallek, and A. H. Baker. 2003. AAV-based gene transfer. Curr. Opin. Mol. Ther. 5:367-375.[Medline]
    10. 10
  10. Dahlberg, P. S., R. D. Acton, R. J. Battafarano, M. E. Uknis, C. A. Ratz, J. W. Johnston, J. R. Haseman, B. H. Gray, and D. L. Dunn. 1996. A novel endotoxin antagonist attenuates tumor necrosis factor-alpha secretion. J. Surg. Res. 63:44-48.[CrossRef][Medline]
    11. 11
  11. Elsbach, P., and J. Weiss. 1998. Role of the bactericidal/permeability-increasing protein in host defence.Curr. Opin. Immunol. 10:45-49.[CrossRef][Medline]
    12. 12
  12. Gazzano-Santoro, H., J. B. Parent, L. Grinna, A. Horwitz, T. Parsons, G. Theofan, P. Elsbach, J. Weiss, and P. J. Conlon.1992 . High-affinity binding of the bactericidal/permeability-increasing protein and a recombinant amino-terminal fragment to the lipid A region of lipopolysaccharide.Infect. Immun. 60:4754-4761.[Abstract/Free Full Text]
    13. 13
  13. Gray, P. W., G. Flaggs, S. R. Leong, R. J. Gumina, J. Weiss, C. E. Ooi, and P. Elsbach.1989 . Cloning of the cDNA of a human neutrophil bactericidal protein. Structural and functional correlations.J. Biol. Chem. 264:9505-9509.[Abstract/Free Full Text]
    14. 14
  14. Hildinger, M., A. Auricchio, G. Gao, L. Wang, N. Chirmule, and J. M. Wilson. 2001. Hybrid vectors based on adeno-associated virus serotypes 2 and 5 for muscle-directed gene transfer.J. Virol. 75:6199-6203.[Abstract/Free Full Text]
    15. 15
  15. Horwitz, A. H., S. D. Leigh, S. Abrahamson, H. Gazzano-Santoro, P. S. Liu, R. E. Williams, S. F. Carroll, and G. Theofan. 1996. Expression and characterization of cysteine-modified variants of an amino-terminal fragment of bactericidal/permeability-increasing protein. Protein Expr. Purif. 8:28-40.[CrossRef][Medline]
    16. 16
  16. Hubsch, A. P., A. T. Casas, and J. E. Doran.1995 . Protective effects of reconstituted high-density lipoprotein in rabbit gram-negative bacteraemia models.J. Lab. Clin. Med. 126:548-558.[Medline]
    17. 17
  17. Kochanek, S., G. Schiedner, and C. Volpers. 2001. High-capacity ‘gutless’ adenoviral vectors. Curr. Opin. Mol. Ther. 3:454-463.[Medline]
    18. 18
  18. Lazaron, V., and R. A. Barke. 1999. Gram-negative bacterial sepsis and the sepsis syndrome. Urol. Clin. N. Am. 26:687-699.[CrossRef][Medline]
    19. 19
  19. Levy, S. B., and B. Marshall. 2004. Antibacterial resistance worldwide: causes, challenges and responses. Nat. Med. 10:s122-s129.
    20. 20
  20. Li, C., J. Li, Z. Lv, X. H. Guo, Q. H. Chen, Q. L. Kong, and Y. Q. An. 2006. Protection of mice from lethal endotoxemia by chimeric human BPI-Fc{gamma}1 gene delivery. Cell. Mol. Immunol. 3:221-225.[Medline]
    21. 21
  21. Lieber, A., D. S. Steinwaerder, C. A Carlson, and M. A. Kay. 1999. Integrating adenovirus-adeno-associated virus hybrid vectors devoid of all viral genes. J. Virol. 73:9314-9324.[Abstract/Free Full Text]
    22. 22
  22. Mannion, B. A., J. Weiss, and P. Elsbach. 1990. Separation of sublethal and lethal effects of the bactericidal/permeability increasing protein on Escherichia coli. J. Clin. Investig. 85:853-860.[Medline]
    23. 23
  23. Martin, G. S., D. M. Mannino, S. Eaton, and M. Moss.2003 . The epidemiology of sepsis in the United States from 1979 through 2000. N. Engl. J. Med. 348:1546-1554.[Abstract/Free Full Text]
    24. 24
  24. Mitani, K., and S. Kubo. 2002. Adenovirus as an integrating vector. Curr. Gene Ther. 2:135-144.[CrossRef][Medline]
    25. 25
  25. Ooi, C. E., J. Weiss, M. E. Doerfler, and P. Elsbach. 1991. Endotoxin-neutralizing properties of the 25 kD N-terminal fragment and a newly isolated 30 kD C-terminal fragment of the 55-60 kD bactericidal/permeability increasing protein of human neutrophils. J. Exp. Med. 174:649-655.[Abstract/Free Full Text]
    26. 26
  26. Riedemann, N. C., R. F. Guo, and P. A. Ward.2003 . Novel strategies for the treatment of sepsis.Nat. Med. 9:517-524.[CrossRef][Medline]
    27. 27
  27. Sakhuja, K., P. S. Reddy, S. Ganesh, F. Cantaniag, S. Pattison, P. Limbach, D. B. Kayda, M. J. Kadan, M. Kaleko, and S. Connelly. 2003. Optimization of the generation and propagation of gutless adenoviral vectors. Hum. Gene Ther. 14:243-254.[CrossRef][Medline]
    28. 28
  28. Sorensen, O., T. Bratt, A. H. Johnsen, M. T. Madsen, and N. Borregaard. 1999. The human antibacterial cathelicidin hCAP-18 is bound to lipoproteins in plasma. J. Biol. Chem. 274:22445-22451.[Abstract/Free Full Text]
    29. 29
  29. Spink, J., and D. Geddes. 2004. Gene therapy progress and prospects: bringing gene therapy into medical practice: the evolution of international ethics and the regulatory environment. Gene Ther. 11:1611-1616.[CrossRef][Medline]
    30. 30
  30. Vincent, J. L., Q. Sun, and M. J. Dubois.2002 . Clinical trials of immunomodulatory therapies in severe sepsis and septic shock. Clin. Infect. Dis. 34:1084-1093.[CrossRef][Medline]
    31. 31
  31. Wang, Z., H. I. Ma, J. Li, L. Sun, J. Zhang, and X. Xiao.2003 . Rapid and highly efficient transduction by double-stranded adeno-associated virus vectors in vitro and in vivo.Gene Ther. 10:2105-2111.[CrossRef][Medline]
    32. 32
  32. Warren, H. S., R. Matyal, J. E. Allaire, D. Yarmush, P. Loiselle, J. Hellman, B. G. Paton, and M. P. Fink. 2003. Protective efficacy of CAP18106-138-immunoglobulin G in sepsis.J. Infect. Dis. 188:1382-1393.[CrossRef][Medline]
    33. 33
  33. Weiss, J., and I. Olsson. 1987. Cellular and subcellular localization of the bactericidal/permeability-increasing protein of neutrophils. Blood 69:652-659.[Abstract/Free Full Text]
    34. 34
  34. Weiss, J., P. Elsbach, C. Shu, J. Castillo, L. Grinna, A. Horwitz, and G. Theofan. 1992. Human bactericidal/permeability-increasing protein and a recombinant NH2-terminal fragment cause killing of serum-resistant gram-negative bacteria in whole blood and inhibit tumor necrosis factor release induced by the bacteria. J. Clin. Investig. 90:1122-1130.[Medline]
    35. 35
  35. Weiss, J., P. Elsbach, I. Olsson, and H. Odeberg. 1978. Purification and characterization of a potent bactericidal and membrane active protein from the granules of human polymorphonuclear leukocytes.J. Biol. Chem. 253:2664-2672.[Free Full Text]
    36. 36
  36. Wu, X. B., X. Y. Dong, and Z. J. Wu.2001 . A novel method for purification of recombinant adeno-associated virus vectors in large-scale. Chin. Sci. Bull. 46:485-489.
    37. 37
  37. Wu, Z. J., X. B. Wu, H. Cao, X. Y. Dong, H. Wang, and Y. D. Hou. 2001. A novel and high efficient production system for recombinant adeno-associated virus vector. Sci. China Ser. C 31:423-430.
    38. 38
  38. Wu, Z. J., X. B. Wu, and Y. D. Hou.2000 . Construction of a series of adeno-associated virus vectors and their expression of b-galactosidase gene. Chin. J. Virol. 16:1-6.

Antimicrobial Agents and Chemotherapy, February 2007, p. 724-731, Vol. 51, No. 2
0066-4804/07/$08.00+0     doi:10.1128/AAC.00360-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chen, J.
Right arrow Articles by An, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chen, J.
Right arrow Articles by An, Y.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»