- Email: [email protected]
Recombinant adeno-associated virus vector: Is it ideal for gene delivery in liver transplantation?
Recombinant Adeno-Associated Virus Vector: Is it Ideal for Gene Delivery in Liver Transplantation? Zhen-Fan Yang,* Xiao-Bing Wu,† Tung-Yu Tsui,* Yun-De Hou,† John M. Luk,* and Sheung-Tat Fan* Recombinant adeno-associated virus vector (rAAV) is an effective and safe gene-delivery tool. However, its application in solid-organ transplantation has not been addressed. The present study is designed to introduce human cytotoxic T-lymphocyte–associated antigen 4 immunoglobulin G (hCTLA4Ig) by rAAV (rAAV-hCTLA4Ig) into rat liver grafts to analyze the effects of virus titer, exposure time, and route of administration on transgene expression and possible side effects caused by the gene-delivery approach. Different rAAV-hCTLA4Ig titers were introduced into liver grafts through back-table portal vein perfusion and preserved for a certain time. rAAV-hCTLA4Ig also was administered by intravenous and intramuscular injection. Transgene expression in grafts and plasma was detected by immunohistochemistry and enzyme-linked immunosorbent assay. Intragraft cytokine level was detected by reverse-transcriptase polymerase chain reaction. Anti-hCTLA4Ig antibodies in plasma were detected by flow cytometry. A higher virus titer (1 ⴛ 1012 viral genomes/animal) introduced through backtable portal vein perfusion and a longer preservation time (3 hours) achieved a greater level of transgene expression until day 180. Back-table portal vein perfusion induced a greater level of hCTLA4 expression in plasma than intramuscular or intravenous injection. Increased interleukin-2 and interferon-␥ messenger RNA levels were detected in grafts with rAAV-hCTLA4Ig gene transfer compared with those without virus delivery, but the response was minor. Such a cellular immune response could be suppressed by low-dose FK506 administration during the first 3 postoperative days. Anti-hCTLA4Ig antibodies could be detected in long-term surviving animals, but the extent of humoral response was not severe. This study shows that rAAV can be an effective and safe vector for gene delivery in liver transplantation. (Liver Transpl 2003;9:411-420.)
W
ith the development of the gene-delivery system, researchers recently have focused on adeno-associated virus (AAV), a promising vector derived from a contaminant in adenovirus culture. Attractions of this replication-defective viral vector include its nonpathogenicity, induction of undetectable or mild immune responses,1,2 and a broad range of infectivity involving both dividing and nondividing cells. Transduction of recombinant AAV vector (rAAV) into skeletal muscle, central nervous system, gastrointestinal tract, and respiratory epithelia was successful because sufficient therapeutic protein was produced to correct various kinds of genetic disorders and autoimmune diseases.3-9 However, there are only a few reports on the application of rAAVs in transplantation, limited mostly to cell transplantation.
Heparan sulfate proteoglycan,10 a receptor of rAAV, is distributed widely in hepatocytes. Therefore, most AAVs are extracted by the liver, even by intravenous injection.11,12 In addition, the presence of much more porous vasculature in the liver and mini AAV particles favors the infection of hepatocytes. Based on the specificity of the liver to AAVs, we consider that rAAV may be an ideal vector in liver transplantation.13 The study by Halbert et al14 showed that efficient rAAV transduction had an area-limiting potency in bronchial epithelium in which a balloon catheter was lodged, indicating possible roles of trauma in rAAVmediated delivery.15-17 In organ transplantation, unavoidable ischemia-reperfusion injury results in damage to cells and subsequent repair of injured structures. Whether this process has an impact on rAAVmediated gene delivery remains to be determined. In addition, the role of virus titer, exposure time, and route of rAAV delivery in transduction efficiency and possible cellular and humoral immune responses induced by gene delivery are not known. In this study, we investigate the potential value of rAAV as a vehicle of gene delivery in a rat liver transplantation model.
Materials and Methods Vector Construction rAAV pSNAV1 was constructed by replacing the viral replicable genes rep and cap with a human cytomegalovirus promoter/enhancer, a multiple-cloning site, a neomycin-resistant
From the *Centre for the Study of Liver Disease and Department of Surgery, University of Hong Kong Medical Centre, Queen Mary Hospital, Hong Kong; and the †National Laboratory of Molecular Virology and Genetic Engineering, Chinese Academy of Preventive Medicine, Beijing, China. X.-B.W. contributed equally to this study. Supported in part by a grant from the Distinguished Research Achievement Award of The University of Hong Kong. Address reprint requests to Sheung-Tat Fan, MS, MD, PhD, FRCS (Edin & Glasg), FACS, Department of Surgery, The University of Hong Kong, Queen Mary Hospital, 102 Pokfulam Rd, Hong Kong, China. Telephone: 852-2855-4703; Fax: 852-2818-4407; E-mail: hrmsfst@ hkucc.hku.hk Copyright © 2003 by the American Association for the Study of Liver Diseases 1527-6465/03/0904-0015$30.00/0 doi:10.1053/jlts.2003.50058
Liver Transplantation, Vol 9, No 4 (April), 2003: pp 411-420
411
412
Yang et al
gene, a simian virus 40 promoter, and a polyadenylation signal, retaining the two inverted terminal repeats intact. A human cytotoxic T-lymphocyte–associated antigen 4 immunoglobulin G (hCTLA4Ig) fusion protein gene was inserted into the site between SalI and BglII in the multiple-cloning site. After selection by neomycin and with the help of mutated herpes simplex virus 1 (HSV-1), large-scale rAAVs were produced and purified by affinity chromatography and ultracentrifugation in cesium chloride (Calbiochen-Novabrochem Corp, San Diego, CA) gradients. Virus titer was determined by dot blot hybridization. The helper virus HSV was heat inactivated at 56°C for 30 minutes, and contamination of HSV was found to be less than 1 in 1010 viral genomes (vgs).
Animals and Experimental Groups Male Lewis (LEW, RT-1l) rats weighing 200 to 300 g were purchased from the Animal Resources Centre (Murdoch, Western Australia). They were maintained under standard conditions and cared for according to institutional guidelines for animal care. They were used as donors and recipients. An orthotopic liver transplantation model was created according to the protocol of Kamada and Calne.18 Virus was diluted in 0.5 mL of Ringer’s lactate solution and injected into the liver graft through the portal vein within 2 minutes. Before implantation, the graft was flushed with another 1 mL of Ringer’s lactate solution to remove any free virus. Experimental groups included: group a: no gene transfer (grafts were perfused with Ringer’s lactate solution and preserved for 3 hours; n ⫽ 6); group b: rAAV-hCTLA4Ig, 1 ⫻ 1011 vg/animal, administered through back-table portal vein perfusion and preserved in 4°C Ringer’s lactate solution for 1 hour (n ⫽ 15); group c: rAAV-hCTLA4Ig, 1 ⫻ 1011 vg/animal, administered through back-table portal vein perfusion, 4°C for 3 hours (n ⫽ 15); group d: rAAV-hCTLA4Ig, 1 ⫻ 1012 vg/animal, administered through back-table portal vein perfusion, 4°C for 1 hour (n ⫽ 15); group e: rAAV-hCTLA4Ig, 1 ⫻ 1012 vg/animal, administered through back-table portal vein perfusion, 4°C for 3 hours (n ⫽ 15); group f: rAAV-hCTLA4Ig, 1 ⫻ 1012 vg/animal, administered through back-table portal vein perfusion, 4°C for 3 hours, and combined with FK506 intramuscular injection from days 0 to 3 [low-dose FK506 was administered to suppress the possible minor cellular immune responses initiated by the gene-delivery approach] (n ⫽ 15); group g: rAAV-hCTLA4Ig, 1 ⫻ 1012 vg/animal, administered by intramuscular injection (n ⫽ 6); and group h: rAAV-hCTLA4Ig, 1 ⫻ 1012 vg/animal, administered by intravenous injection (n ⫽ 6). Plasma samples were collected from recipients days 3, 7, 14, 30, 60, 90, 120, 150, and 180 after the operation, intramuscular injection, or intravenous injection. Tissue samples were collected days 3, 7, 14, 30, and 180 after the operation in groups b, c, d, e, and f.
Expression of hCTLA4 in Grafts and Plasma When animals were killed, half the liver grafts were snapfrozen in liquid nitrogen and stored at ⫺80°C for intragraft hCTLA4 protein level analysis by enzyme-linked immunosorbent assay (ELISA). Total protein was extracted from 100 mg of graft tissue by Tris-buffered saline (USB Corp, Cleveland, OH) and 1% Triton (USB Corp) at 4°C overnight. Total protein, 5 mg, was added to each well of the ELISA plate, which had been precoated by purified antihCTLA4 antibody at 4°C overnight and incubated at room temperature for 4 hours, followed by biotinylated antibody incubation for 1 hour at room temperature. Color was developed by 2,2⬘-azino-bis (3-ethylbenothiazoline sulfonic acid) substrate solution at room temperature for 10 minutes (purified and biotinylated goat anti-hCTLA4 antibodies; R&D Systems Inc, Minneapolis, MN). Optical density was measured at 405 nm in a Vmax kinetic microplate reader (Molecular Devices Corp, Sunnyvale, CA). Concentrations of samples were calculated under the standard curve setup by serial dilutions of hCTLA4/Fc Chimera. Soluble hCTLA4 levels in plasma were measured by means of ELISA. Another half of graft tissues was fixed in 10% buffered formalin and embedded in paraffin for histological study by hematoxylin-eosin staining and detection of transduced hCTLA4 protein expression by immunohistochemistry (antibody 1, goat anti-hCTLA4 polyclonal antibody; R&D Systems Inc; antibody 2, horseradish peroxidase– conjugated rabbit antigoat polyclonal antibody; Zymed Laboratories Inc, S San Francisco, CA). Paraffin-embedded tissue was cut into 5-m thick sections and incubated with antibody 1 for 2 hours at room temperature, followed by a 1-hour incubation of antibody 2. Color development was performed in substrate solution with diaminobenzidine (Dako Corp, Carpinteria, CA) for 5 minutes. Slides then were counterstained with hematoxylin (Vector Laboratories Inc, Austin, TX). Five fields were randomly chosen on each slide, and positively stained cells were counted using MetaMorph software (Universal Imaging Corp, Downingtown, PA) at original magnification ⫻400. Relative hCTLA4 expression was described by a positive rate (number of hCTLA4-positive cells versus total number of cells, in percent).
Intragraft Cytokine Messenger RNA Expression Snap-frozen graft tissue was homogenized, and messenger RNA (mRNA) was extracted for analysis of interleukin-2 (IL-2), IL-4, IL-10, and interferon-␥ (IFN-␥) expression by reverse-transcriptase polymerase chain reaction (RT-PCR). RT-PCR reagents for mRNA extraction and first-strand synthesis were purchased from Amersham Pharmacia Biotech Inc (Buckinghamshire, UK). Five micrograms of mRNA was used as template for the first-strand DNA reaction. Primers specific for different cytokines were designed according to Kita et al.19 PCR reactions were performed through reversetranscription incubation at 94°C for 2 minutes and 40 cycles of 85°C for 1 minute, 60°C for 1 minute, and 72°C for 1 minute. PCR products were analyzed with electrophoresis in
rAAV in Liver Transplantation
2% agarose gel stained with ethidium bromide. Relative cytokine expression was evaluated by the optical density ratio between samples and -actin.
Phenotypes of Lymphocyte Infiltration in Grafts Snap-frozen tissue was cut into 5-m thick sections and fixed in acetone for 10 minutes. Monoclonal mouse antirat CD4, CD8, CD25, B, T, and natural killer (NK) antibodies (BD Pharmingen, San Diego, CA) were added to sections and incubated at room temperature for 2 hours, followed by another 1-hour incubation in horseradish peroxidase– conjugated goat antimouse IgG polyclonal antibody (Zymed Laboratories Inc). Color was developed by diaminobenzidine incubation for 5 minutes. Cells were counted and analyzed by MetaMorph software at original magnification ⫻400. Five fields were chosen randomly from each section, and a positive rate was described by the number of positive cells over the total number of lymphocytes in each field (in percent).
Anti-hCTLA4 Antibodies in Plasma Antibodies against hCTLA4Ig were detected indirectly through the induction of hCTLA4Ig expression in the thirdparty cells, 293 cell lines. If anti-hCTLA4Ig antibodies existed in the recipient’s plasma, they would bind to the hCTLA4Ig expressed in these cells, and the antibody titer could be determined by flow cytometry. Two hundred ninety-three cell lines were maintained as monolayer culture in Dulbecco’s modified Eagle medium (DMEM) with 10% fatal bovine serum (FBS) and 1% penicillin (Life Technologies, Carlsbad, CA) at 37°C in a humidified atmosphere of 5% carbon dioxide in air. Cells were inoculated in 60-mm culture dishes until 70% to 80% confluent. Medium was removed and replaced by FBS-free DMEM with rAAV-hCTLA4Ig, 1 ⫻ 108 vg/dish, and cultured for 1 hour. Five percent FBS-DMEM was fed again into each dish and maintained for 7 days. Cells were harvested and washed twice with FACS (Becton Dickinson Immunocytometry Systems, San Jose, CA) medium (1% bovine serum albumin in phosphate-buffered saline with 0.1% sodium azide and 0.2% formaldehyde) and inoculated into a 96-well culture plate (5 ⫻ 105 cells/well). At the same time, plasma from long-term surviving recipients (day 180) was heat inactivated in a 56°C water bath for 30 minutes and diluted with FACS medium (1:20; plasma from group a was used as control). Fifty microliters of diluted plasma was added to each well and incubated at 4°C for 30 minutes, followed by secondary antibodies (biotinylated mouse antirat IgG1, IgG2a, IgG2b, IgG2c, and IgM monoclonal antibodies; BD Pharmingen) in 1:100 dilution and incubated for another 30 minutes at 4°C. Positive cells were labeled by Streptavidin PE (BD Pharmingen) and detected in FACS Calibur (Becton Dickinson Immunocytometry Systems). Appropriate isotypes of irrelevant monoclonal antibodies were used as controls.
413
Liver Function Tests Plasma samples were collected days 3, 7, 14, 30, 60, 90, 120, 150, and 180 after transplantation (three samples for each time in each group). Liver function was evaluated in the clinical biochemistry laboratory (Queen Mary Hospital, The University of Hong Kong) by testing total bilirubin, glutamic-pyruvic transaminase (GPT), and glutamic-oxaloacetic transaminase (GOT) levels in plasma.
Statistical Analysis All comparative analyses were performed by means of Student’s t-test using SPSS statistical software (version 10.0 for Windows; SPSS Inc, Chicago, IL). P less than 0.05 is considered statistically significant. Values in all figures are expressed as mean ⫾ SD.
Results In Vivo hCTLA4 Expression hCTLA4 expression in grafts. Postoperative day 3, no hCTLA4-positive cell in the graft was detected in any experimental group, whereas postoperative day 7, a certain number of hCTLA4-positive cells was found in groups b, c, d, e, and f (Fig. 1A). When quantified by ELISA on the same day after delivery, the greatest hCTLA4 level always was detected in grafts administered rAAV-hCTLA4Ig, 1 ⫻ 1012 vg/animal, that had been preserved for 3 hours, except day 3. Low-dose FK506 administration in the early postoperative period did not alter hCTLA4 levels in grafts significantly (Fig. 1B). Soluble hCTLA4 in plasma. Plasma levels of soluble hCTLA4 also showed a dose- and time-dependent manner. In groups b, c, d, and e, soluble hCTLA4 in plasma remained detectable days 7 to 180, with peaks at days 60 to 90. The highest plasma hCTLA4 levels were achieved in group e at all times (Fig. 2). Role of delivery routes. Similar to back-table perfusion, intramuscular and intravenous injection of rAAVhCTLA4Ig also could induce sustained hCTLA4 expression in plasma, with peaks at postoperative day 60 (intramuscular injection, 7.02 ⫾ 0.84 ng/mL; intravenous injection, 8.55 ⫾ 0.67 ng/mL), and maintain a relative stable level until day 180 (Fig. 3). Greatest hCTLA4 plasma levels were achieved by back-table perfusion at all times. Cellular Immune Responses Intragraft cytokine mRNA expression by RT-PCR. Three types of cytokines, including T helper subtype 1 (TH1) cytokines IL-2 and IFN-␥, TH2 cytokine IL-4, and
414
Yang et al
Figure 1. hCTLA4 expression in the back-table–perfused graft by immunohistochemical staining and ELISA. (A) Various cell types, including hepatocytes, endothelial cells, and Kupffer cells, were infected by rAAV-hCTLA4Ig and expressed hCTLA4. The arrow points at one hCTLA4-positive cell. (B) Greatest intragraft hCTLA4 expression was induced by the highest titer and longest preservation time (*P < .05, group e compared with groups b, c, and d). Administration of low-dose FK506 did not significantly increase hCTLA4 levels in grafts (group f).
TH3 cytokine IL-10, were measured by RT-PCR. Postoperative day 3, IL-2 and IFN-␥ were detected in grafts with rAAV-hCTLA4Ig gene delivery (Fig. 4A, lane 2) and even upregulated postoperative day 7 (Fig. 4A, lane
3), whereas none was detectable in grafts without gene delivery (Fig. 4A, lane 1). When low-dose FK506 was administered days 0 to 3, IL-2 and IFN-␥ became undetectable postoperative day 7 (Fig. 4A, lane 4). IL-4
rAAV in Liver Transplantation
415
Figure 2. Plasma levels of hCTLA4 demonstrated in a dose- and time-dependent manner. The highest hCTLA4 level in plasma was induced by 1 ⴛ 1012 vg/animal back-table perfusion and preservation for 3 hours (*P < .05, group e compared with groups b, c, and d). FK506 (group f) did not further enhance the plasma level of soluble hCTLA4.
and IL-10 remained undetectable in all treatment groups. Relative cytokine mRNA to -actin is shown in Figure 4B. Intragraft lymphocyte infiltration. Few lymphocytes were found in periportal area of grafts without gene delivery postoperative day 7, whereas an increased number of cell influx was observed in grafts with rAAV-
hCTLA4Ig gene delivery (70 ⫾ 7 v 48 ⫾ 3.6; Fig. 5A). The greatest proportion of cells were CD4⫹ cells in groups a, e, and f. However, in grafts with virus delivery, increased numbers of T, B, and NK cells were detected, whereas low-dose FK506 administration significantly decreased the proportion of T cells only (Fig. 5B).
Figure 3. Different routes of virus administration affected hCTLA4 expression in plasma. The highest level was achieved by back-table perfusion and 3-hour preservation. *P < .05, group e compared with groups g and h.
416
Yang et al
short-term and low-dose FK506 decreased the number of cell influx (Fig. 7C). Liver Function Tests Total plasma bilirubin levels increased during the first week after the operation in both groups a and e, but gradually returned to normal levels with time. GPT and GOT levels increased during the first 2 weeks after the operation and also decreased to normal with time. There were no statistically significant differences in total bilirubin, GPT, and GOT levels between the group without gene delivery (group a) and that with rAAV-hCTLA4Ig gene transfer (group e; Fig. 8A, B, and C).
Discussion
Figure 4. Intragraft cytokine mRNA expression by RTPCR. IL-2 and IFN-␥ mRNA were not detected in grafts without gene delivery postoperative day 7 (group a; lane 1). Virus administration induced IL-2 and IFN-␥ mRNA expression in grafts (lane 2, rAAV-hCTLA4Ig day 3; lane 3, rAAV-hCTLA4Ig day 7), whereas low-dose FK506 (group f) eliminated these two cytokines day 7 (lane 4). *P < .05 compared with grafts without gene delivery (group a). #P < .05 compared with gene delivery (group e) day 7.
Antibody Production Against hCTLA4Ig After 293 cells were incubated by rAAV-hCTLA4Ig and adenovirus 5 for 48 hours, plasma from long-term survivors (180 days after rAAV-hCTLA4Ig transfer) was introduced to these 293 cells. An increased mean channel of fluorescence was detected in IgG2a, IgG2b, IgG2c, and IgM isotypes in plasma of long-term survivors with rAAV-hCTLA4Ig gene transfer compared with those without gene delivery (Fig. 6). Survival Time and Histological Characteristics of Long-Term Grafts All recipients survived indefinitely, with no evidence of infection. When no virus was administered, long-term isografts showed normal liver architecture with proliferation of small bile ducts (Fig. 7A). However, in grafts treated with rAAV-hCTLA4Ig back-table perfusion, there was a slight increase in number of lymphocytes in the periportal area (Fig. 7B), whereas to some extent,
As a promising vehicle in the gene-delivery system, rAAV can induce sustained transgene expression. In this study, we show that an increase in virus titer and extension of exposure time of host cells to virus can achieve greater target protein expression. These two approaches enhance the chance to form more concatemers, which is the necessary step for AAV transcription and subsequent protein expression.20 However, in the situation of organ transplantation, at the same time, prolonged preservation time can aggravate preservation and ischemia-reperfusion damage to grafts. Our data show that back-table perfusion through the portal vein with 1 ⫻ 1012 of rAAV and preservation of grafts for 3 hours might be an optimal protocol to induce sufficient protein expression. In this study, we observed a greater plasma level of hCTLA4 in the back-table perfusion group than the intramuscular- or intravenous-injection groups. It may be caused by the unavoidable ischemia-reperfusion injury that stimulates DNA repair enzyme activation and subsequently helps second-stranded synthesis and transcription of rAAV. Another reason may be that when viruses are administered intravenously, phagocytes may engulf some of them in the host circulation before they reach the liver. Finally, when rAAVs are delivered through intramuscular injection, most of them are trapped in the site of injection. Contrary to most reports showing that rAAVs did not elicit detectable immune responses, our data indicate that rAAVs could stimulate measurable immunologic changes in both cellular and humoral processes. RT-PCR data show detectable levels of IL-2 and IFN-␥ mRNA in grafts with rAAV-hCTLA4Ig gene transfer postoperative days 3 and 7. Moreover, when target pro-
Figure 5. Phenotypes of infiltrated lymphocytes postoperative day 7. (A) An obvious increased number of T, B, and NK cells in grafts with rAAV-hCTLA4Ig gene delivery. (Original magnification ⴛ200). Arrows point to B, T, and NK cells. (B) The number of T, B, and NK cells in grafts with gene delivery showed a significant increase compared with those without virus administration, whereas low-dose FK506 only decreased the proportion of T cells. *P < .05, group e compared with group a. #P < .05, group f compared with group e.
418
Yang et al
Figure 6. Detection of anti-rAAV-hCTLA4Ig antibodies in plasma of long-term survivors by flow cytometry. Significant increases in IgG2a, IgG2b, IgG2c, and IgM titers were detected in plasma of long-term survivors with rAAV-hCTLA4Ig delivery (group e). *P < .05, group e compared with group a. (MCF, mean channel of fluorescence.)
tein emerged in grafts, greater IL-2 and IFN-␥ mRNA levels were detected postoperative day 7 rather than day 3, suggesting that cellular immune responses might be stimulated by both rAAV-transduced cells and transgene products. Low-dose FK506 could subdue IL-2 and IFN-␥ mRNA expression in grafts, yielding further cellular responses against rAAV-mediated gene delivery, although reactions appeared minor. In addition to cellular immune parameters, our data also show the presence of antibodies against hCTLA4Ig in plasma of long-term surviving animals. These antibodies seemed neutralizing and might account for the gradual decline in plasma hCTLA4 levels starting day 60. Because the 293 cell lines were from a human source, the existence of some natural antibodies against humans in rat plasma gave rise to some fluorescent signals. However, when we administered rAAV-encoding green fluorescent protein into the rat liver and observed it for the same period as in rAAV-hCTLA4Ig delivery (data not shown), we also detected an increased level of fluorescence, further proving the existence of antibodies. In addition, these antibodies might not be against transgene products only, but also the virus itself. Another possible reason that may contribute to the decline in hCTLA4 expression is that not all viruses have integrated into host chromosomes, and some of them may be eliminated with the death of host cells. The increased titer of antibodies may have some negative impact on long-term target protein production. However, the extent of humoral response is not severe, and because transgene expression by rAAV is a continuous process, therapeutic effects may have been
Figure 7. Histological characteristics of isografts postoperative day 180. A slightly increased number of infiltrated lymphocytes was detected in the periportal area of the graft with rAAV-hCTLA4Ig gene delivery. (A) Graft without gene delivery. (B) Graft with rAAV-hCTLA4Ig transfer. (C) Graft with rAAV-hCTLA4Ig plus FK506. (Original magnification ⴛ100.)
rAAV in Liver Transplantation
Figure 8. Detection of liver biochemistry in plasma. There was no significant difference in total bilirubin, GTP, and GOT levels between the groups with and without gene delivery. Although both groups demonstrated elevation of GPT and GOT levels within 2 weeks after transplantation, they could return back to normal with time.
achieved before antibodies are produced. Further studies are needed to analyze the correlation of antibodies and long-term therapeutic effect of transgene. However, it seemed that short-term low-dose FK506 application could only reduce cellular responses, but had no effect on antibody production. The possible reason is that when antibodies were produced, FK506 had already been eliminated from the circulation of animals.
419
Conversely, studies of some groups showed that such transient immunosuppression as anti-CD4 or antiCD40L antibodies during initial exposure might achieve successful readministration of the virus.2,14 Therefore, application of short-term FK506 in our study also may help a repeated dose of virus administration. Our study shows the possible application of rAAV in future clinical transplantation settings. Viruses can be introduced into liver grafts right after harvesting and during ischemic time (ranging from 1 to several hours), and the majority of them may enter into host cells. Based on the virus titer in our study, 3 ⫻ 1012 to 1 ⫻ 1013 vg/kg may be sufficient to achieve therapeutic purposes. However, large-animal experiments should be performed to address this issue. Some viral vectors, such as retrovirus vector, may induce oncogene activation and tumorgenicity because of the randomness of viral integration into host chromosomes. However, rAAV seems to be an exception. Because more than 60% of healthy adults are AAV seropositive and no report shows the pathogenicity of this virus, rAAV may be a safe vehicle for clinical use. In addition, our study shows that rAAV-mediated transduction does not affect function of the target organ (shown by liver function tests), further indicating its safety. Nevertheless, it is possible that some untransduced virus may enter other tissues with revascularization. We performed some experiments to detect hCTLAIg expression in organs other than liver, but did not find hCTLA4-positive cells (data not shown). As a result, we consider that even if there are some virus entries into other tissues, the number should be minor. In addition, it also is our concern whether long-term transgene expression, such as CTLA4Ig, may affect the host immune system against infectious agents, although from our serial observations, we did not find abnormalities of rAAV-transduced animals. More studies may be needed to further define the safety of this approach. In conclusion, rAAV vector can be a safe and effective vehicle in liver transplantation, especially when long-term therapeutic effects are required. However, detectable cellular and immune responses can influence the persistence of transgene products, and low-dose and short-term immunosuppressive drug administration can minimize cellular response and may help early protein expression.
References 1. Sarukhan A, Camugli S, Gjata B, von Boehmer H, Danos O, Jooss K. Successful interference with cellular immune responses
420
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Yang et al
to immunogenic proteins encoded by recombinant viral vectors. J Virol 2001;75:269-277. Manning WC, Zhou S, Bland MP, Escobedo JA, Dwarki V. Transient immunosuppression allows transgene expression following readministration of adeno-associated viral vectors. Hum Gene Ther 1998;9:477-485. During MJ, Xu R, Young D, Kaplitt MG, Sherwin RS, Leone P. Peroral gene therapy of lactose intolerance using an adeno-associated virus vector. Nat Med 1998;4:1131-1135. Greelish JP, Su LT, Lankford EB, Burkman JM, Chen H, Konig SK, et al. Stable restoration of the sarcoglycan complex in dystrophic muscle perfused with histamine and a recombinant adeno-associated viral vector. Nat Med 1999;5:439-443. During MJ, Symes CW, Lawlor PA, Lin J, Dunning J, Fitzsimons HL, et al. An oral vaccine against NMDAR1 with efficacy in experimental stroke and epilepsy. Science 2000;287:14531460. Lee HC, Kim SJ, Kim KS, Shin HC, Yoon JW. Remission in models of type 1 diabetes by gene therapy using a single-chain insulin analogue. Nature 2000;408:483-488. Mastakov MY, Baer K, Xu R, Fitzsimons H, During MJ. Combined injection of rAAV with mannitol enhances gene expression in the rat brain. Mol Ther 2001;3:225-232. Kay MA, Manno CS, Ragni MV, Larson PJ, Couto LB, McClelland A, et al. Evidence for gene transfer and expression of factor IX in haemophilia B patients treated with an AAV vector. Nat Genet 2000;24:257-261. Herzog RW, Yang EY, Couto LB, Hagstrom JN, Elwell D, Fields PA, et al. Long-term correction of canine hemophilia B by gene transfer of blood coagulation factor IX mediated by adenoassociated viral vector. Nat Med 1999;5:56-63. Summerford C, Samulski RJ. Membrane-associated heparan sulfate proteoglycan is a receptor for adeno-associated virus type 2 virions. J Virol 1998;72:1438-1445. Ponnazhagan S, Mukherjee P, Yoder MC, Wang XS, Zhou SZ, Kaplan J, et al. Adeno-associated virus 2-mediated gene transfer
12.
13.
14.
15.
16.
17.
18.
19.
20.
in vivo: Organ-tropism and expression of transduced sequences in mice. Gene 1997;190:203-210. Koeberl DD, Bonham L, Halbert CL, Allen JM, Birkebak T, Miller AD. Persistent, therapeutically relevant levels of human granulocyte colony-stimulating factor in mice after systemic delivery of adeno-associated virus vectors. Hum Gene Ther 1999;10:2133-2140. Xiao W, Berta SC, Lu MM, Moscioni AD, Tazelaar J, Wilson JM. Adeno-associated virus as a vector for liver-directed gene therapy. J Virol 1998;72:10222-10226. Halbert CL, Standaert TA, Wilson CB, Miller AD. Successful readministration of adeno-associated virus vectors to the mouse lung requires transient immunosuppression during the initial exposure. J Virol 1998;72:9795-9805. Russell DW, Alexander IE, Miller AD. DNA synthesis and topoisomerase inhibitors increase transduction by adeno-associated virus vectors. Proc Natl Acad Sci U S A 1995;92:5719-5723. Alexander IE, Russell DW, Miller AD. DNA-damaging agents greatly increase the transduction of nondividing cells by adenoassociated virus vectors. J Virol 1994;68:8282-8287. Alexander IE, Russell DW, Spence AM, Miller AD. Effects of gamma irradiation on the transduction of dividing and nondividing cells in brain and muscle of rats by adeno-associated virus vectors. Hum Gene Ther 1996;7:841-850. Kamada N, Calne RY. Orthotopic liver transplantation in the rat. Technique using cuff for portal vein anastomosis and biliary drainage. Transplantation 1979;28:47-50. Kita Y, Li XK, Ohba M, Funeshima N, Enosawa S, Tamura A, et al. Prolonged cardiac allograft survival in rats systemically injected adenoviral vectors containing CTLA4Ig-gene. Transplantation 1999;68:758-766. Miao CH, Nakai H, Thompson AR, Storm TA, Chiu W, Snyder RO, et al. Nonrandom transduction of recombinant adenoassociated virus vectors in mouse hepatocytes in vivo: Cell cycling does not influence hepatocyte transduction. J Virol 2000; 74:3793-3803.
W
ith the development of the gene-delivery system, researchers recently have focused on adeno-associated virus (AAV), a promising vector derived from a contaminant in adenovirus culture. Attractions of this replication-defective viral vector include its nonpathogenicity, induction of undetectable or mild immune responses,1,2 and a broad range of infectivity involving both dividing and nondividing cells. Transduction of recombinant AAV vector (rAAV) into skeletal muscle, central nervous system, gastrointestinal tract, and respiratory epithelia was successful because sufficient therapeutic protein was produced to correct various kinds of genetic disorders and autoimmune diseases.3-9 However, there are only a few reports on the application of rAAVs in transplantation, limited mostly to cell transplantation.
Heparan sulfate proteoglycan,10 a receptor of rAAV, is distributed widely in hepatocytes. Therefore, most AAVs are extracted by the liver, even by intravenous injection.11,12 In addition, the presence of much more porous vasculature in the liver and mini AAV particles favors the infection of hepatocytes. Based on the specificity of the liver to AAVs, we consider that rAAV may be an ideal vector in liver transplantation.13 The study by Halbert et al14 showed that efficient rAAV transduction had an area-limiting potency in bronchial epithelium in which a balloon catheter was lodged, indicating possible roles of trauma in rAAVmediated delivery.15-17 In organ transplantation, unavoidable ischemia-reperfusion injury results in damage to cells and subsequent repair of injured structures. Whether this process has an impact on rAAVmediated gene delivery remains to be determined. In addition, the role of virus titer, exposure time, and route of rAAV delivery in transduction efficiency and possible cellular and humoral immune responses induced by gene delivery are not known. In this study, we investigate the potential value of rAAV as a vehicle of gene delivery in a rat liver transplantation model.
Materials and Methods Vector Construction rAAV pSNAV1 was constructed by replacing the viral replicable genes rep and cap with a human cytomegalovirus promoter/enhancer, a multiple-cloning site, a neomycin-resistant
From the *Centre for the Study of Liver Disease and Department of Surgery, University of Hong Kong Medical Centre, Queen Mary Hospital, Hong Kong; and the †National Laboratory of Molecular Virology and Genetic Engineering, Chinese Academy of Preventive Medicine, Beijing, China. X.-B.W. contributed equally to this study. Supported in part by a grant from the Distinguished Research Achievement Award of The University of Hong Kong. Address reprint requests to Sheung-Tat Fan, MS, MD, PhD, FRCS (Edin & Glasg), FACS, Department of Surgery, The University of Hong Kong, Queen Mary Hospital, 102 Pokfulam Rd, Hong Kong, China. Telephone: 852-2855-4703; Fax: 852-2818-4407; E-mail: hrmsfst@ hkucc.hku.hk Copyright © 2003 by the American Association for the Study of Liver Diseases 1527-6465/03/0904-0015$30.00/0 doi:10.1053/jlts.2003.50058
Liver Transplantation, Vol 9, No 4 (April), 2003: pp 411-420
411
412
Yang et al
gene, a simian virus 40 promoter, and a polyadenylation signal, retaining the two inverted terminal repeats intact. A human cytotoxic T-lymphocyte–associated antigen 4 immunoglobulin G (hCTLA4Ig) fusion protein gene was inserted into the site between SalI and BglII in the multiple-cloning site. After selection by neomycin and with the help of mutated herpes simplex virus 1 (HSV-1), large-scale rAAVs were produced and purified by affinity chromatography and ultracentrifugation in cesium chloride (Calbiochen-Novabrochem Corp, San Diego, CA) gradients. Virus titer was determined by dot blot hybridization. The helper virus HSV was heat inactivated at 56°C for 30 minutes, and contamination of HSV was found to be less than 1 in 1010 viral genomes (vgs).
Animals and Experimental Groups Male Lewis (LEW, RT-1l) rats weighing 200 to 300 g were purchased from the Animal Resources Centre (Murdoch, Western Australia). They were maintained under standard conditions and cared for according to institutional guidelines for animal care. They were used as donors and recipients. An orthotopic liver transplantation model was created according to the protocol of Kamada and Calne.18 Virus was diluted in 0.5 mL of Ringer’s lactate solution and injected into the liver graft through the portal vein within 2 minutes. Before implantation, the graft was flushed with another 1 mL of Ringer’s lactate solution to remove any free virus. Experimental groups included: group a: no gene transfer (grafts were perfused with Ringer’s lactate solution and preserved for 3 hours; n ⫽ 6); group b: rAAV-hCTLA4Ig, 1 ⫻ 1011 vg/animal, administered through back-table portal vein perfusion and preserved in 4°C Ringer’s lactate solution for 1 hour (n ⫽ 15); group c: rAAV-hCTLA4Ig, 1 ⫻ 1011 vg/animal, administered through back-table portal vein perfusion, 4°C for 3 hours (n ⫽ 15); group d: rAAV-hCTLA4Ig, 1 ⫻ 1012 vg/animal, administered through back-table portal vein perfusion, 4°C for 1 hour (n ⫽ 15); group e: rAAV-hCTLA4Ig, 1 ⫻ 1012 vg/animal, administered through back-table portal vein perfusion, 4°C for 3 hours (n ⫽ 15); group f: rAAV-hCTLA4Ig, 1 ⫻ 1012 vg/animal, administered through back-table portal vein perfusion, 4°C for 3 hours, and combined with FK506 intramuscular injection from days 0 to 3 [low-dose FK506 was administered to suppress the possible minor cellular immune responses initiated by the gene-delivery approach] (n ⫽ 15); group g: rAAV-hCTLA4Ig, 1 ⫻ 1012 vg/animal, administered by intramuscular injection (n ⫽ 6); and group h: rAAV-hCTLA4Ig, 1 ⫻ 1012 vg/animal, administered by intravenous injection (n ⫽ 6). Plasma samples were collected from recipients days 3, 7, 14, 30, 60, 90, 120, 150, and 180 after the operation, intramuscular injection, or intravenous injection. Tissue samples were collected days 3, 7, 14, 30, and 180 after the operation in groups b, c, d, e, and f.
Expression of hCTLA4 in Grafts and Plasma When animals were killed, half the liver grafts were snapfrozen in liquid nitrogen and stored at ⫺80°C for intragraft hCTLA4 protein level analysis by enzyme-linked immunosorbent assay (ELISA). Total protein was extracted from 100 mg of graft tissue by Tris-buffered saline (USB Corp, Cleveland, OH) and 1% Triton (USB Corp) at 4°C overnight. Total protein, 5 mg, was added to each well of the ELISA plate, which had been precoated by purified antihCTLA4 antibody at 4°C overnight and incubated at room temperature for 4 hours, followed by biotinylated antibody incubation for 1 hour at room temperature. Color was developed by 2,2⬘-azino-bis (3-ethylbenothiazoline sulfonic acid) substrate solution at room temperature for 10 minutes (purified and biotinylated goat anti-hCTLA4 antibodies; R&D Systems Inc, Minneapolis, MN). Optical density was measured at 405 nm in a Vmax kinetic microplate reader (Molecular Devices Corp, Sunnyvale, CA). Concentrations of samples were calculated under the standard curve setup by serial dilutions of hCTLA4/Fc Chimera. Soluble hCTLA4 levels in plasma were measured by means of ELISA. Another half of graft tissues was fixed in 10% buffered formalin and embedded in paraffin for histological study by hematoxylin-eosin staining and detection of transduced hCTLA4 protein expression by immunohistochemistry (antibody 1, goat anti-hCTLA4 polyclonal antibody; R&D Systems Inc; antibody 2, horseradish peroxidase– conjugated rabbit antigoat polyclonal antibody; Zymed Laboratories Inc, S San Francisco, CA). Paraffin-embedded tissue was cut into 5-m thick sections and incubated with antibody 1 for 2 hours at room temperature, followed by a 1-hour incubation of antibody 2. Color development was performed in substrate solution with diaminobenzidine (Dako Corp, Carpinteria, CA) for 5 minutes. Slides then were counterstained with hematoxylin (Vector Laboratories Inc, Austin, TX). Five fields were randomly chosen on each slide, and positively stained cells were counted using MetaMorph software (Universal Imaging Corp, Downingtown, PA) at original magnification ⫻400. Relative hCTLA4 expression was described by a positive rate (number of hCTLA4-positive cells versus total number of cells, in percent).
Intragraft Cytokine Messenger RNA Expression Snap-frozen graft tissue was homogenized, and messenger RNA (mRNA) was extracted for analysis of interleukin-2 (IL-2), IL-4, IL-10, and interferon-␥ (IFN-␥) expression by reverse-transcriptase polymerase chain reaction (RT-PCR). RT-PCR reagents for mRNA extraction and first-strand synthesis were purchased from Amersham Pharmacia Biotech Inc (Buckinghamshire, UK). Five micrograms of mRNA was used as template for the first-strand DNA reaction. Primers specific for different cytokines were designed according to Kita et al.19 PCR reactions were performed through reversetranscription incubation at 94°C for 2 minutes and 40 cycles of 85°C for 1 minute, 60°C for 1 minute, and 72°C for 1 minute. PCR products were analyzed with electrophoresis in
rAAV in Liver Transplantation
2% agarose gel stained with ethidium bromide. Relative cytokine expression was evaluated by the optical density ratio between samples and -actin.
Phenotypes of Lymphocyte Infiltration in Grafts Snap-frozen tissue was cut into 5-m thick sections and fixed in acetone for 10 minutes. Monoclonal mouse antirat CD4, CD8, CD25, B, T, and natural killer (NK) antibodies (BD Pharmingen, San Diego, CA) were added to sections and incubated at room temperature for 2 hours, followed by another 1-hour incubation in horseradish peroxidase– conjugated goat antimouse IgG polyclonal antibody (Zymed Laboratories Inc). Color was developed by diaminobenzidine incubation for 5 minutes. Cells were counted and analyzed by MetaMorph software at original magnification ⫻400. Five fields were chosen randomly from each section, and a positive rate was described by the number of positive cells over the total number of lymphocytes in each field (in percent).
Anti-hCTLA4 Antibodies in Plasma Antibodies against hCTLA4Ig were detected indirectly through the induction of hCTLA4Ig expression in the thirdparty cells, 293 cell lines. If anti-hCTLA4Ig antibodies existed in the recipient’s plasma, they would bind to the hCTLA4Ig expressed in these cells, and the antibody titer could be determined by flow cytometry. Two hundred ninety-three cell lines were maintained as monolayer culture in Dulbecco’s modified Eagle medium (DMEM) with 10% fatal bovine serum (FBS) and 1% penicillin (Life Technologies, Carlsbad, CA) at 37°C in a humidified atmosphere of 5% carbon dioxide in air. Cells were inoculated in 60-mm culture dishes until 70% to 80% confluent. Medium was removed and replaced by FBS-free DMEM with rAAV-hCTLA4Ig, 1 ⫻ 108 vg/dish, and cultured for 1 hour. Five percent FBS-DMEM was fed again into each dish and maintained for 7 days. Cells were harvested and washed twice with FACS (Becton Dickinson Immunocytometry Systems, San Jose, CA) medium (1% bovine serum albumin in phosphate-buffered saline with 0.1% sodium azide and 0.2% formaldehyde) and inoculated into a 96-well culture plate (5 ⫻ 105 cells/well). At the same time, plasma from long-term surviving recipients (day 180) was heat inactivated in a 56°C water bath for 30 minutes and diluted with FACS medium (1:20; plasma from group a was used as control). Fifty microliters of diluted plasma was added to each well and incubated at 4°C for 30 minutes, followed by secondary antibodies (biotinylated mouse antirat IgG1, IgG2a, IgG2b, IgG2c, and IgM monoclonal antibodies; BD Pharmingen) in 1:100 dilution and incubated for another 30 minutes at 4°C. Positive cells were labeled by Streptavidin PE (BD Pharmingen) and detected in FACS Calibur (Becton Dickinson Immunocytometry Systems). Appropriate isotypes of irrelevant monoclonal antibodies were used as controls.
413
Liver Function Tests Plasma samples were collected days 3, 7, 14, 30, 60, 90, 120, 150, and 180 after transplantation (three samples for each time in each group). Liver function was evaluated in the clinical biochemistry laboratory (Queen Mary Hospital, The University of Hong Kong) by testing total bilirubin, glutamic-pyruvic transaminase (GPT), and glutamic-oxaloacetic transaminase (GOT) levels in plasma.
Statistical Analysis All comparative analyses were performed by means of Student’s t-test using SPSS statistical software (version 10.0 for Windows; SPSS Inc, Chicago, IL). P less than 0.05 is considered statistically significant. Values in all figures are expressed as mean ⫾ SD.
Results In Vivo hCTLA4 Expression hCTLA4 expression in grafts. Postoperative day 3, no hCTLA4-positive cell in the graft was detected in any experimental group, whereas postoperative day 7, a certain number of hCTLA4-positive cells was found in groups b, c, d, e, and f (Fig. 1A). When quantified by ELISA on the same day after delivery, the greatest hCTLA4 level always was detected in grafts administered rAAV-hCTLA4Ig, 1 ⫻ 1012 vg/animal, that had been preserved for 3 hours, except day 3. Low-dose FK506 administration in the early postoperative period did not alter hCTLA4 levels in grafts significantly (Fig. 1B). Soluble hCTLA4 in plasma. Plasma levels of soluble hCTLA4 also showed a dose- and time-dependent manner. In groups b, c, d, and e, soluble hCTLA4 in plasma remained detectable days 7 to 180, with peaks at days 60 to 90. The highest plasma hCTLA4 levels were achieved in group e at all times (Fig. 2). Role of delivery routes. Similar to back-table perfusion, intramuscular and intravenous injection of rAAVhCTLA4Ig also could induce sustained hCTLA4 expression in plasma, with peaks at postoperative day 60 (intramuscular injection, 7.02 ⫾ 0.84 ng/mL; intravenous injection, 8.55 ⫾ 0.67 ng/mL), and maintain a relative stable level until day 180 (Fig. 3). Greatest hCTLA4 plasma levels were achieved by back-table perfusion at all times. Cellular Immune Responses Intragraft cytokine mRNA expression by RT-PCR. Three types of cytokines, including T helper subtype 1 (TH1) cytokines IL-2 and IFN-␥, TH2 cytokine IL-4, and
414
Yang et al
Figure 1. hCTLA4 expression in the back-table–perfused graft by immunohistochemical staining and ELISA. (A) Various cell types, including hepatocytes, endothelial cells, and Kupffer cells, were infected by rAAV-hCTLA4Ig and expressed hCTLA4. The arrow points at one hCTLA4-positive cell. (B) Greatest intragraft hCTLA4 expression was induced by the highest titer and longest preservation time (*P < .05, group e compared with groups b, c, and d). Administration of low-dose FK506 did not significantly increase hCTLA4 levels in grafts (group f).
TH3 cytokine IL-10, were measured by RT-PCR. Postoperative day 3, IL-2 and IFN-␥ were detected in grafts with rAAV-hCTLA4Ig gene delivery (Fig. 4A, lane 2) and even upregulated postoperative day 7 (Fig. 4A, lane
3), whereas none was detectable in grafts without gene delivery (Fig. 4A, lane 1). When low-dose FK506 was administered days 0 to 3, IL-2 and IFN-␥ became undetectable postoperative day 7 (Fig. 4A, lane 4). IL-4
rAAV in Liver Transplantation
415
Figure 2. Plasma levels of hCTLA4 demonstrated in a dose- and time-dependent manner. The highest hCTLA4 level in plasma was induced by 1 ⴛ 1012 vg/animal back-table perfusion and preservation for 3 hours (*P < .05, group e compared with groups b, c, and d). FK506 (group f) did not further enhance the plasma level of soluble hCTLA4.
and IL-10 remained undetectable in all treatment groups. Relative cytokine mRNA to -actin is shown in Figure 4B. Intragraft lymphocyte infiltration. Few lymphocytes were found in periportal area of grafts without gene delivery postoperative day 7, whereas an increased number of cell influx was observed in grafts with rAAV-
hCTLA4Ig gene delivery (70 ⫾ 7 v 48 ⫾ 3.6; Fig. 5A). The greatest proportion of cells were CD4⫹ cells in groups a, e, and f. However, in grafts with virus delivery, increased numbers of T, B, and NK cells were detected, whereas low-dose FK506 administration significantly decreased the proportion of T cells only (Fig. 5B).
Figure 3. Different routes of virus administration affected hCTLA4 expression in plasma. The highest level was achieved by back-table perfusion and 3-hour preservation. *P < .05, group e compared with groups g and h.
416
Yang et al
short-term and low-dose FK506 decreased the number of cell influx (Fig. 7C). Liver Function Tests Total plasma bilirubin levels increased during the first week after the operation in both groups a and e, but gradually returned to normal levels with time. GPT and GOT levels increased during the first 2 weeks after the operation and also decreased to normal with time. There were no statistically significant differences in total bilirubin, GPT, and GOT levels between the group without gene delivery (group a) and that with rAAV-hCTLA4Ig gene transfer (group e; Fig. 8A, B, and C).
Discussion
Figure 4. Intragraft cytokine mRNA expression by RTPCR. IL-2 and IFN-␥ mRNA were not detected in grafts without gene delivery postoperative day 7 (group a; lane 1). Virus administration induced IL-2 and IFN-␥ mRNA expression in grafts (lane 2, rAAV-hCTLA4Ig day 3; lane 3, rAAV-hCTLA4Ig day 7), whereas low-dose FK506 (group f) eliminated these two cytokines day 7 (lane 4). *P < .05 compared with grafts without gene delivery (group a). #P < .05 compared with gene delivery (group e) day 7.
Antibody Production Against hCTLA4Ig After 293 cells were incubated by rAAV-hCTLA4Ig and adenovirus 5 for 48 hours, plasma from long-term survivors (180 days after rAAV-hCTLA4Ig transfer) was introduced to these 293 cells. An increased mean channel of fluorescence was detected in IgG2a, IgG2b, IgG2c, and IgM isotypes in plasma of long-term survivors with rAAV-hCTLA4Ig gene transfer compared with those without gene delivery (Fig. 6). Survival Time and Histological Characteristics of Long-Term Grafts All recipients survived indefinitely, with no evidence of infection. When no virus was administered, long-term isografts showed normal liver architecture with proliferation of small bile ducts (Fig. 7A). However, in grafts treated with rAAV-hCTLA4Ig back-table perfusion, there was a slight increase in number of lymphocytes in the periportal area (Fig. 7B), whereas to some extent,
As a promising vehicle in the gene-delivery system, rAAV can induce sustained transgene expression. In this study, we show that an increase in virus titer and extension of exposure time of host cells to virus can achieve greater target protein expression. These two approaches enhance the chance to form more concatemers, which is the necessary step for AAV transcription and subsequent protein expression.20 However, in the situation of organ transplantation, at the same time, prolonged preservation time can aggravate preservation and ischemia-reperfusion damage to grafts. Our data show that back-table perfusion through the portal vein with 1 ⫻ 1012 of rAAV and preservation of grafts for 3 hours might be an optimal protocol to induce sufficient protein expression. In this study, we observed a greater plasma level of hCTLA4 in the back-table perfusion group than the intramuscular- or intravenous-injection groups. It may be caused by the unavoidable ischemia-reperfusion injury that stimulates DNA repair enzyme activation and subsequently helps second-stranded synthesis and transcription of rAAV. Another reason may be that when viruses are administered intravenously, phagocytes may engulf some of them in the host circulation before they reach the liver. Finally, when rAAVs are delivered through intramuscular injection, most of them are trapped in the site of injection. Contrary to most reports showing that rAAVs did not elicit detectable immune responses, our data indicate that rAAVs could stimulate measurable immunologic changes in both cellular and humoral processes. RT-PCR data show detectable levels of IL-2 and IFN-␥ mRNA in grafts with rAAV-hCTLA4Ig gene transfer postoperative days 3 and 7. Moreover, when target pro-
Figure 5. Phenotypes of infiltrated lymphocytes postoperative day 7. (A) An obvious increased number of T, B, and NK cells in grafts with rAAV-hCTLA4Ig gene delivery. (Original magnification ⴛ200). Arrows point to B, T, and NK cells. (B) The number of T, B, and NK cells in grafts with gene delivery showed a significant increase compared with those without virus administration, whereas low-dose FK506 only decreased the proportion of T cells. *P < .05, group e compared with group a. #P < .05, group f compared with group e.
418
Yang et al
Figure 6. Detection of anti-rAAV-hCTLA4Ig antibodies in plasma of long-term survivors by flow cytometry. Significant increases in IgG2a, IgG2b, IgG2c, and IgM titers were detected in plasma of long-term survivors with rAAV-hCTLA4Ig delivery (group e). *P < .05, group e compared with group a. (MCF, mean channel of fluorescence.)
tein emerged in grafts, greater IL-2 and IFN-␥ mRNA levels were detected postoperative day 7 rather than day 3, suggesting that cellular immune responses might be stimulated by both rAAV-transduced cells and transgene products. Low-dose FK506 could subdue IL-2 and IFN-␥ mRNA expression in grafts, yielding further cellular responses against rAAV-mediated gene delivery, although reactions appeared minor. In addition to cellular immune parameters, our data also show the presence of antibodies against hCTLA4Ig in plasma of long-term surviving animals. These antibodies seemed neutralizing and might account for the gradual decline in plasma hCTLA4 levels starting day 60. Because the 293 cell lines were from a human source, the existence of some natural antibodies against humans in rat plasma gave rise to some fluorescent signals. However, when we administered rAAV-encoding green fluorescent protein into the rat liver and observed it for the same period as in rAAV-hCTLA4Ig delivery (data not shown), we also detected an increased level of fluorescence, further proving the existence of antibodies. In addition, these antibodies might not be against transgene products only, but also the virus itself. Another possible reason that may contribute to the decline in hCTLA4 expression is that not all viruses have integrated into host chromosomes, and some of them may be eliminated with the death of host cells. The increased titer of antibodies may have some negative impact on long-term target protein production. However, the extent of humoral response is not severe, and because transgene expression by rAAV is a continuous process, therapeutic effects may have been
Figure 7. Histological characteristics of isografts postoperative day 180. A slightly increased number of infiltrated lymphocytes was detected in the periportal area of the graft with rAAV-hCTLA4Ig gene delivery. (A) Graft without gene delivery. (B) Graft with rAAV-hCTLA4Ig transfer. (C) Graft with rAAV-hCTLA4Ig plus FK506. (Original magnification ⴛ100.)
rAAV in Liver Transplantation
Figure 8. Detection of liver biochemistry in plasma. There was no significant difference in total bilirubin, GTP, and GOT levels between the groups with and without gene delivery. Although both groups demonstrated elevation of GPT and GOT levels within 2 weeks after transplantation, they could return back to normal with time.
achieved before antibodies are produced. Further studies are needed to analyze the correlation of antibodies and long-term therapeutic effect of transgene. However, it seemed that short-term low-dose FK506 application could only reduce cellular responses, but had no effect on antibody production. The possible reason is that when antibodies were produced, FK506 had already been eliminated from the circulation of animals.
419
Conversely, studies of some groups showed that such transient immunosuppression as anti-CD4 or antiCD40L antibodies during initial exposure might achieve successful readministration of the virus.2,14 Therefore, application of short-term FK506 in our study also may help a repeated dose of virus administration. Our study shows the possible application of rAAV in future clinical transplantation settings. Viruses can be introduced into liver grafts right after harvesting and during ischemic time (ranging from 1 to several hours), and the majority of them may enter into host cells. Based on the virus titer in our study, 3 ⫻ 1012 to 1 ⫻ 1013 vg/kg may be sufficient to achieve therapeutic purposes. However, large-animal experiments should be performed to address this issue. Some viral vectors, such as retrovirus vector, may induce oncogene activation and tumorgenicity because of the randomness of viral integration into host chromosomes. However, rAAV seems to be an exception. Because more than 60% of healthy adults are AAV seropositive and no report shows the pathogenicity of this virus, rAAV may be a safe vehicle for clinical use. In addition, our study shows that rAAV-mediated transduction does not affect function of the target organ (shown by liver function tests), further indicating its safety. Nevertheless, it is possible that some untransduced virus may enter other tissues with revascularization. We performed some experiments to detect hCTLAIg expression in organs other than liver, but did not find hCTLA4-positive cells (data not shown). As a result, we consider that even if there are some virus entries into other tissues, the number should be minor. In addition, it also is our concern whether long-term transgene expression, such as CTLA4Ig, may affect the host immune system against infectious agents, although from our serial observations, we did not find abnormalities of rAAV-transduced animals. More studies may be needed to further define the safety of this approach. In conclusion, rAAV vector can be a safe and effective vehicle in liver transplantation, especially when long-term therapeutic effects are required. However, detectable cellular and immune responses can influence the persistence of transgene products, and low-dose and short-term immunosuppressive drug administration can minimize cellular response and may help early protein expression.
References 1. Sarukhan A, Camugli S, Gjata B, von Boehmer H, Danos O, Jooss K. Successful interference with cellular immune responses
420
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Yang et al
to immunogenic proteins encoded by recombinant viral vectors. J Virol 2001;75:269-277. Manning WC, Zhou S, Bland MP, Escobedo JA, Dwarki V. Transient immunosuppression allows transgene expression following readministration of adeno-associated viral vectors. Hum Gene Ther 1998;9:477-485. During MJ, Xu R, Young D, Kaplitt MG, Sherwin RS, Leone P. Peroral gene therapy of lactose intolerance using an adeno-associated virus vector. Nat Med 1998;4:1131-1135. Greelish JP, Su LT, Lankford EB, Burkman JM, Chen H, Konig SK, et al. Stable restoration of the sarcoglycan complex in dystrophic muscle perfused with histamine and a recombinant adeno-associated viral vector. Nat Med 1999;5:439-443. During MJ, Symes CW, Lawlor PA, Lin J, Dunning J, Fitzsimons HL, et al. An oral vaccine against NMDAR1 with efficacy in experimental stroke and epilepsy. Science 2000;287:14531460. Lee HC, Kim SJ, Kim KS, Shin HC, Yoon JW. Remission in models of type 1 diabetes by gene therapy using a single-chain insulin analogue. Nature 2000;408:483-488. Mastakov MY, Baer K, Xu R, Fitzsimons H, During MJ. Combined injection of rAAV with mannitol enhances gene expression in the rat brain. Mol Ther 2001;3:225-232. Kay MA, Manno CS, Ragni MV, Larson PJ, Couto LB, McClelland A, et al. Evidence for gene transfer and expression of factor IX in haemophilia B patients treated with an AAV vector. Nat Genet 2000;24:257-261. Herzog RW, Yang EY, Couto LB, Hagstrom JN, Elwell D, Fields PA, et al. Long-term correction of canine hemophilia B by gene transfer of blood coagulation factor IX mediated by adenoassociated viral vector. Nat Med 1999;5:56-63. Summerford C, Samulski RJ. Membrane-associated heparan sulfate proteoglycan is a receptor for adeno-associated virus type 2 virions. J Virol 1998;72:1438-1445. Ponnazhagan S, Mukherjee P, Yoder MC, Wang XS, Zhou SZ, Kaplan J, et al. Adeno-associated virus 2-mediated gene transfer
12.
13.
14.
15.
16.
17.
18.
19.
20.
in vivo: Organ-tropism and expression of transduced sequences in mice. Gene 1997;190:203-210. Koeberl DD, Bonham L, Halbert CL, Allen JM, Birkebak T, Miller AD. Persistent, therapeutically relevant levels of human granulocyte colony-stimulating factor in mice after systemic delivery of adeno-associated virus vectors. Hum Gene Ther 1999;10:2133-2140. Xiao W, Berta SC, Lu MM, Moscioni AD, Tazelaar J, Wilson JM. Adeno-associated virus as a vector for liver-directed gene therapy. J Virol 1998;72:10222-10226. Halbert CL, Standaert TA, Wilson CB, Miller AD. Successful readministration of adeno-associated virus vectors to the mouse lung requires transient immunosuppression during the initial exposure. J Virol 1998;72:9795-9805. Russell DW, Alexander IE, Miller AD. DNA synthesis and topoisomerase inhibitors increase transduction by adeno-associated virus vectors. Proc Natl Acad Sci U S A 1995;92:5719-5723. Alexander IE, Russell DW, Miller AD. DNA-damaging agents greatly increase the transduction of nondividing cells by adenoassociated virus vectors. J Virol 1994;68:8282-8287. Alexander IE, Russell DW, Spence AM, Miller AD. Effects of gamma irradiation on the transduction of dividing and nondividing cells in brain and muscle of rats by adeno-associated virus vectors. Hum Gene Ther 1996;7:841-850. Kamada N, Calne RY. Orthotopic liver transplantation in the rat. Technique using cuff for portal vein anastomosis and biliary drainage. Transplantation 1979;28:47-50. Kita Y, Li XK, Ohba M, Funeshima N, Enosawa S, Tamura A, et al. Prolonged cardiac allograft survival in rats systemically injected adenoviral vectors containing CTLA4Ig-gene. Transplantation 1999;68:758-766. Miao CH, Nakai H, Thompson AR, Storm TA, Chiu W, Snyder RO, et al. Nonrandom transduction of recombinant adenoassociated virus vectors in mouse hepatocytes in vivo: Cell cycling does not influence hepatocyte transduction. J Virol 2000; 74:3793-3803.