Intravenous administration of an E1/E3-deleted adenoviral ... - Nature

Gene Therapy (2001) 8, 354–361  2001 Nature Publishing Group All rights reserved 0969-7128/01 $15.00 www.nature.com/gt

RESEARCH ARTICLE

Intravenous administration of an E1/E3-deleted adenoviral vector induces tolerance to factor IX in C57BL/6 mice PA Fields1, E Armstrong1, JN Hagstrom1, VR Arruda1, ML Murphy2, JP Farrell2, KA High1 and RW Herzog1 1

Departments of Pediatrics and Pathology, University of Pennsylvania Medical Center and The Children’s Hospital of Philadelphia, and 2School of Veterinary Medicine, University of Pennsylvania, Philadelphia, PA, USA

Inbred immunocompetent C57BL/6 mice have been a favored strain to study transgene expression of human blood coagulation factor IX (hF.IX) from viral vectors because systemic expression of the secreted protein is not limited by antibody responses following intravenous (i.v.) injection of vector. For example, i.v. injection of an adenoviral (Ad) vector results in sustained expression of hF.IX in normal or hemophilic C57BL/6 mice, while anti-hF.IX antibodies rapidly emerge in other strains (Gene Therapy 4: 473; Blood 91: 784). To investigate these observations further, we injected naive C57BL/6 mice and C57BL/6 mice with pre-existing anti-hF.IX with Ad-hF.IX vector via peripheral vein. All mice expressed hF.IX antigen without detectable anti-hF.IX, even when challenged with hF.IX in different immunogenic settings at later time points. Moreover, in mice with pre-existing

immunity, anti-hF.IX titers diminished to undetectable levels after i.v. administration of Ad-hF.IX. Lymphocytes from mice that had received Ad-hF.IX i.v. failed to proliferate when stimulated with hF.IX in vitro after the animals had been repeatedly challenged with hF.IX protein formulated in complete Freund’s adjuvant. Thus, absence of anti-hF.IX in C57BL/6 mice after i.v. injection of Ad vector is not due to ignorance to the foreign transgene product. Similar experiments in other strains showed that immune tolerance to hF.IX does not correlate with the strain haplotype or expression of IL-10 cytokine. Given the well-documented immunogenicity of the first-generation adenoviral vector, data from C57BL/6 mice may therefore grossly underestimate immunological consequences in certain gene therapy protocols. Gene Therapy (2001) 8, 354–361.

Keywords: hemophilia B; factor IX; adenovirus; C57BL/6 mice

Introduction The severe bleeding disorder hemophilia is an X-linked genetic disease caused by absence of functional blood coagulation factor VIII (F.VIII, hemophilia A) or factor IX (F.IX, hemophilia B). Conventional treatment of hemophilia is based on intravenous infusion of coagulation factor concentrate (plasma-derived or recombinant), which, if performed in response to a bleed instead of prophylactically, is suboptimal due to the relatively short half-life of the clotting factor in the circulation. One of the most serious risks of any treatment based on factor replacement therapy is an immune response against the coagulation factor in the form of neutralizing antibody formation.1,2 As is evident from the recent initiation of several clinical trials, hemophilia has been a promising experimental model for treatment of genetic diseases by systemic delivery of a secreted transgene product.3–5 There are several large and small animal models available, and several cell types can produce biologically active coagulation factors. Furthermore, sustained expression of ⬎1% Correspondence: KA High, The Children’s Hospital of Philadelphia, Abramson Research Center, Room 310, 34th Street and Civic Center, Philadelphia, PA 19104, USA Received 21 August 2000; accepted 30 November 2000

of normal clotting factor levels would ameliorate the phenotype of a patient with severe disease, and tight regulation of expression is not required. F.VIII and F.IX circulate as inactive zymogens until the coagulation cascade is initiated, so that over-expression above normal levels may be safe. These features led to the development of adenoviral (ad) vectors for high levels of primarily liver-derived expression of F.VIII and F.IX following intravenous (i.v.) infusion of vector. Expression in mice, hemophilic dogs, and nonhuman primates was generally short-lived due to acute phase and adoptive inflammatory immune responses, induction of adenovirus-specific cytotoxic T lymphocytes (CTL), antibody formation against the transgene product, and (dose-limiting) hepatotoxicity, which make the vector unappealing for clinical trials.6–10 Additionally, repeat administration of the vector is prevented by formation of neutralizing antibodies against the viral capsid.11 Interestingly, it has been found that reduced vector doses not only reduce hepatotoxicity, but also result in sustained expression of high levels of human F.VIII or F.IX in C57BL/6 mice.12,13 Consequently, this strain of immunocompetent mice was extensively used for efficacy studies, particularly because the mice did not develop an antibody response against human coagulation factors, while antibody formation was often prevalent in other

Tolerance to factor IX induced in C57BL/6 mice PA Fields et al

strains. Moreover, i.v. injection of adenoviral vectors was successful in conferring sustained complete phenotypic correction of the hemophilic phenotype in hemophilia A and B mice on a C57BL/6 genetic background.12,13 It is puzzling that introduction of a human transgene product results in long-term expression in normal and in knock out mice deficient for F.VIII or F.IX using a vector that is well-characterized for its immunogenicity.14,15 Recent studies have shown that hemophilic C57BL/6 mice are not tolerant to human coagulation factors, since they rapidly develop high titer antibodies against human F.VIII or F.IX when these proteins are introduced as intravenously infused protein concentrates or when the vector is injected intramuscularly (i.m.).16,17 Additionally, data with a ␤-galactosidase transgene product showed that cellular immune responses against adenoviral vector backbone proteins may be sufficient to eliminate transgene expression within 1 month in mice transgenic, and therefore tolerant to ␤-galactosidase.15 It is thus unclear how long-term expression of coagulation factor IX is achieved in C57BL/6 following systemic delivery of adenoviral vector. In this study, we demonstrate that i.v. injection of a first generation adenoviral vector results in a state of tolerance to human F.IX in C57BL/6 mice. Tolerized animals can then be challenged with F.IX in different immunogenic settings without eliciting an anti-hF.IX response. These results are not reproducible in other strains (or other species), and do not appear to be linked to the haplotype or expression of the potentially tolerogenic cytokine IL-10. Thus, C57BL/6 mice have a unique tolerance mechanism that limits this strain as a model for development of pre-clinical studies of immune responses for intravenous administration of adenoviral, and possibly other vectors.

Results Lack of anti-human F.IX in C57BL/6 mice that received i.v. injection of adenoviral vector As demonstrated earlier by our laboratory,12 C57BL/6 mice that received i.v. injection of a first generation adenoviral vector for expression of human F.IX (hF.IX) from the CMV IE enhancer/promoter gave high levels of systemic F.IX expression without detectable antibody response against human F.IX (Figure 1a and b). Antigen levels in the circulation declined over 6–8 months from 1–1.5 ␮g/ml to 100–200 ng/ml (Figure 1 and data not shown) without detectable anti-hF.IX. When C57BL/6 mice are injected i.m. with AAV or adenoviral vector, the animals without exception developed anti-hF.IX by day 14 after vector administration.17 However, when the mice that had received adenoviral vector i.v. were challenged by an i.m. injection of AAV vector encoding hF.IX at week 8 of the experiment, there was no reduction in hF.IX antigen levels, nor were antibodies against hF.IX detected when assayed at weeks 12, 16 or later time-points (Figure 1a and b, and data not shown). Humoral immune responses against human F.IX are overcome by i.v. injection of adenoviral vector in C57BL/6 mice Naive C57BL/6 injected i.m. with AAV vector developed anti-hF.IX as shown previously17,18 and in Figure 1.

Because of this humoral immune response, the mice do not show systemic expression of hF.IX, while transduced muscle fibers are not eliminated due to a lack of a cellular immune response.17,18 As shown previously, the mice developed high titer IgG1 and to a lesser extent IgG2c and IgG2b responses against hF.IX (Figure 1d and data not shown). When these mice received adenoviral vector i.v. (week 8), high levels of hF.IX antigen were measured at weeks 12 and 16 comparable to those observed for naive mice that had received the adenoviral vector (Figure 1c). Anti-hF.IX levels were undetectable at week 12 and later time-points, while control mice that had received the i.m. injection of AAV vector, but not the i.v. injection of adenovirus, continued to produce anti-hF.IX (Figure 1d). Figure 1 shows elimination of IgG1 antihF.IX after i.v. administration of the adenoviral vector. Results were identical for the other IgG subclasses (data not shown).

355

Absence of anti-hF.IX is not linked to MHC haplotype C57BL/6 mice (haplotype H-2b) lack MHC class II allele IE, and may therefore have some deficiency in humoral immune responses. However, when mice of another strain (FBV) with the same haplotype, and therefore the same lack of the IE allele, were injected i.v. with the adenoviral vector, they developed IgG1, IgG2a and IgG2b anti-hF.IX by day 14 resulting in a rapid decline of systemic hF.IX expression (Figure 2 and data not shown). Formation or absence of anti-hF.IX in BALB/c or C57BL/6 mice does not correlate with expression of IL10 The cytokine IL-10 has been implicated in down-regulation of immune responses and induction of T cell anergy leading to antigen-specific tolerance. To test whether differences in levels or regulation of expression of IL-10 in different strains of mice may be implicated in induction of tolerance to hF.IX, we performed i.v. injections of the adenoviral vector in C57BL/6 and BALB/c mice which were either wild-type or IL-10 deficient.19 BALB/c mice (haplotype H-2K) had been shown by others to produce anti-hF.IX following i.v. injection of adenoviral vector.20 As shown in Figure 3, both normal and IL-10-deficient C57BL/6 mice produce high systemic levels of hF.IX without detectable antibody formation (Figure 3c and data not shown). In contrast, normal and IL-10 deficient BALB/c mice synthesized IgG1, IgG2a, and IgG2b anti-hF.IX by day 14 after vector administration, thereby blocking systemic expression efficiently (Figure 3d and data not shown). Thus, there is no evidence of involvement of IL-10 expression in tolerance to hF.IX. Lack of hF.IX-specific T cell proliferation in C57BL/6 mice that had received i.v. injection of adenoviral vector C57BL/6 mice (n = 3) that had received adenoviral vector via the i.v. route were challenged twice by subcutaneous injection of hF.IX protein formulated in complete Freund’s adjuvant (cFA) 5–6 months after adenoviral vector administration. The injections were 1 month apart, and the animals did not show anti-hF.IX as measure by ELISA up to 2 weeks after the last boost with hF.IX/cFA (data not shown). Subsequently, the mice were killed and in vitro proliferation of lymphocytes in response to hF.IX antigen was assessed. C57BL/6 that had received i.m. Gene Therapy


356

Figure 1 Systemic levels of hF.IX expression (a, b) and production of IgG1 anti-hF.IX (c, d) in C57BL/6 mice as a function of time. Mice were injected i.v. with ad-hF.IX vector at day 0 followed by i.m. injection of AAV-hF.IX during week 8 (a and c, n = 3) or injected i.m. with AAV-hF.IX vector at day 0 followed by i.v. injection of ad-hF.IX during week 8 (b and d, n = 4). Human F.IX antigen levels were determined by ELISA, and anti-hF.IX concentrations were measured by IgG subclass-specific ELISA. Each line represents an individual animal. All injections were performed using 4 × 1010 viral particles per mouse. The two control mice shown in graph d received AAVhF.IX at day 0 without subsequent i.v. injection of adenoviral vector.

injection of the adenoviral vector (but not cFA) were analyzed in parallel. As documented in Table 1, no evidence for T lymphocyte proliferation was obtained for mice that were i.v. injected with adenoviral vector, despite the repeated boost with hF.IX antigen in cFA, while i.m. injected mice showed mild proliferation with a stimulation index of 2–3.5.

Discussion Previously published data from our and other laboratories have shown that i.v. infusion of a first generation ad vector leads to sustained, high levels of systemic expression of human F.IX in C57BL/6 mice.12,20 Gene transfer occurs primarily to the liver, but also to other tissues such as spleen, lung, and heart muscle.11,21 Michou et al20 have shown that F.IX expression is prolonged in C56BL/6 mice despite activation of adenovirus-specific CTLs, which are measurable in vitro. The authors demonstrated absence of anti-hF.IX in this strain, whereas other strains of mice rapidly developed antihF.IX. Similar results were seen in F.IX-deficient (hemophilia B) C57BL/6 mice.12 This lack of humoral immune response against F.IX can therefore not be explained by sequence similarity of human and murine F.IX. Hemophilic C57BL/6 mice, however, show strong, T cell-dependent anti-hF.IX responses when challenged with i.v. injection of F.IX protein (the conventional treatment for hemophilia patients) or i.m. injection of the vector (Fields et al17 and this study). The latter causes hF.IXGene Therapy

specific CTL and inflammatory responses that eliminate transgene expression in muscle.17 Our results demonstrate that C57BL/6 mice are not tolerant to human F.IX, but rather are rendered tolerant following i.v. injection of ad-hF.IX vector. Since tolerized mice can be repeatedly challenged with F.IX in different immunogenic settings, and since mice with pre-existing antibodies against F.IX become tolerant after i.v. injection of ad-hF.IX, tolerance to the transgene product cannot be explained by ignorance, ie lack of reactive T cells or inadequate presentation of the F.IX antigen. Rather, tolerance may be achieved by an anergy or clonal deletion mechanism. This interpretation is supported by lack of T cell proliferation in tolerized mice that were challenged with F.IX formulated in complete FA. To our surprise, infusion of the adenoviral vector in mice with anti-hF.IX did not cause a boost of the antibody titer or anaphylactic reactions, but on the contrary down-regulated the immune response to undetectable levels of anti-hF.IX within 2 weeks. We cannot rule out the presence of antibody–antigen complexes in these mice that masked the detection of anti-hF.IX following adenovirus-derived expression of F.IX. However, antibody responses were still undetectable several months after ad vector administration, when F.IX antigen levels had declined (data not shown). Work by other investigators has shown that i.v. infusion of an adenoviral vector expressing ␤-galactosidase in mice transgenic and therefore tolerant to ␤galactosidase results in only transient expression of the


Figure 2 Systemic levels of hF.IX expression (a) and production of IgG1 anti-hF.IX (b) in FBV mice as a function of time after i.v. administration of 4 × 1010 ad-hF.IX particles per mouse (n = 4). Human F.IX antigen levels were determined by ELISA, and anti-hF.IX concentrations were measured by IgG subclass-specific ELISA. Each line represents an individual animal.

transgene in hepatocytes due to CTL responses against adenoviral proteins encoded by the vector. However, depending on the dose of vector and the mouse strain, immunogenicity of the transgene product may be an important determinant for the severity of cellular immune responses.13,20,22 A recent study with an adenoviral vector expressing a human ␣1-antitrypsin transgene has demonstrated that expression can be prolonged when a hepatocyte-specific promoter is chosen, presumably because expression in antigen presenting cells is avoided.23 Interestingly, we achieved tolerance to hF.IX in C57BL/6 mice by expression from the CMV IE enhancer/promoter, which expresses well in dendritic cells.24 C57BL/6 is an inbred strain of mice (haplotype H-2b) that is deficient in the MHC class II allele IE, but can present peptides by MHC class II molecules encoded by the IA allele. Experiments by our laboratory and others have clearly demonstrated efficient T helper cell-dependent production of antibodies against coagulation factors

by C57BL/6 mice after i.v. infusion of protein or following i.m. injection of vector.16,17 Moreover, experiments in mice with the same haplotype (FBV mice) show that presence of the IA allele is sufficient for formation of antibodies against F.IX in the context of i.v. infusion of the adenoviral vector. An alternative explanation for tolerance induction in the C57BL/6 may be a unique pattern of cytokine expression following systemic delivery of the viral vector. In particular, IL-10 has been documented to be involved in antigen-specific tolerance by its ability to down-regulate immune responses, including down-regulation of MHC class II expression and inhibition of production of inflammatory cytokines.25 IL-10 is implicated in the negative regulation of T cells and induction of tolerance in CD4+ T cells.26,27 Expression of IL-10 in Kupffer cells, professional APCs in the liver, may serve as an autoregulatory mechanism for antigen presentation by liver antigen presenting cells.28 It has been shown that Kupffer cells secrete IL-10 in response to endotoxin challenge and regulate local immune and inflammatory reactions in the liver sinusoid. We therefore hypothesized that regulation of IL-10 expression in the liver, the main target of gene transfer following i.v. injection of adenoviral vector, may play a role in tolerance or immunity against F.IX antigen. Our results, however, do not support involvement of IL10 expression in tolerance to F.IX. IL-10-deficient C57BL/6 mice did not develop anti-hF.IX while IL-10deficient BALB/c mice showed anti-hF.IX IgG by day 14 after i.v. infusion of adenoviral vector. These results were identical to those obtained with IL-10 expressing mice, and thus indicate that that expression of IL-10 is unlikely to be a requirement for F.IX-specific T cell anergy. As is evident from Figures 1–3, systemic levels of expression of hF.IX were measured during the first week after i.v. administration of adenoviral vector in FBV and BALB/c mice, although at lower levels than in C57BL/6 mice. This may be due to differences in the severity of cellular immune responses or it may reflect differences in the efficiency of gene transfer or expression in these strains of mice. Additionally, transgene expression from the CMV promoter in hepatocytes transduced by adenoviral vectors may be influenced by cytokine expression.29 From the experiments presented here, we cannot rule out that the high level of transgene expression in C57BL/6 mice is important for tolerance to hF.IX in these animals. However, experiments performed by Michou et al20 showed similar levels of hF.IX expression in C57BL/6 and other strains such as BALB/c, yet absence of anti-hF.IX was only observed in C57BL/6 mice. Genetic mapping based on hybrid mice may be a useful tool for identification of genes involved in tolerance in C57BL/6 versus immunity in BALB/c mice. Introduction of a foreign antigen by implantation of hepatocytes expressing hepatitis B surface antigen into the spleen of C57BL/6 mice has been reported to result in translocation of these cells to the liver and induction of tolerance to this antigen.30 Presentation of antigen expressed in the liver by hepatocytes may therefore be tolerogenic. However, this interpretation is likely not sufficient to explain the results obtained with the adenoviral vector in the study presented here. Besides the fact that the result is dramatically different in other strains of mice, the vector not only transduces other tissues, but moreover contains the CMV IE enhancer/promoter

357

Gene Therapy


358

Figure 3 Systemic levels of hF.IX expression (a, b) and production of IgG1 anti-hF.IX (c, d) in C57BL/6 (a, c) or BALB/c (b, d) mice as a function of time after i.v. administration of 4 × 1010 ad-hF.IX particles per mouse. Each line represents an individual animal. Open symbols indicate normal mice (n = 3 per strain), solid symbols indicate IL-10 deficient mice (n = 3 per strain). Human F.IX antigen levels were determined by ELISA, and antihF.IX was measured by IgG subclass-specific ELISA.

Table 1 Lymphocyte proliferation following in vitro stimulation with human F.IX protein Mouse

SI

Ad-hF.IX i.m. Ad-hF.IX i.m. AAV-hFIX IM + Ad-hF.IX i.v. + hF.IX/cFA Ad-hF.IX IV + AAV-hF.IX i.m. + hF.IX.cFa Ad-hF.IX IV + AAV-hF.IX i.m. + hF.IX/cFa

2.1 3.5 ⬍1 ⬍1 1.6

Shown are the stimulation indexes (SI) for individual C57BL/6 mice that had received ad-hF.IX vector i.m. (first two rows) or i.v. (last three rows). Mice injected i.v. with Ad-hF.IX vector had received i.m. infusions of AAV-hF.IX 8 weeks before or after i.v. administration of the adenoviral vector, and were boosted twice in a 1month interval with hF.IX protein formulated in complete Freund’s adjuvant (cFA) and infused subcutaneously. SI was calculated based on the hF.IX concentration that gave the highest proliferative response, which was generally 5–10 ␮g hF.IX/ml medium.

which allows efficient transgene expression in dendritic cells.24 The role of expression in APCs in the tolerance phenomenon described here remains unclear. Long-term expression of human F.IX has also been reported in normal and hemophilia B C57BL/6 mice after infusion of adeno-associated viral (AAV) vector into the portal vein for liver-directed expression.31–33 Although F.IX expression in the AAV vectors was also achieved by a viral or a ubiquitous mammalian promoter, expression Gene Therapy

was likely more hepatocyte-restricted since dissemination of vector was more confined to the liver than with systemic delivery of adenoviral vector. Additionally, AAV is comparatively inefficient in transducing professional APCs.17,24 Snyder et al33 showed the presence of antihuman F.IX antibodies in hemophilia B mice on a C57BL/6 background after portal vein infusion of AAV vector. These antibodies were detectable by Western blot, but did not neutralize systemic expression or activity of F.IX. The immune response against F.IX in liver-directed gene therapy with an AAV vector remains to be investigated for other strains of mice. Recent data show absence of humoral immunity against human F.IX in BALB/c mice following i.v. infusion of AAV vector.34 Together, the data discussed here suggest that i.v. delivery of viral vectors resulting in gene expression confined to hepatocytes may be a more generally applicable approach for avoiding humoral immunity against a secreted transgene product. Additional experiments will have to address whether indeed immune tolerance was achieved in these experiments. Interestingly, with AAV, lentiviral, and gutted adenoviral vectors neutralizing anti-human F.VIII antibodies have been described following liver-directed gene transfer in C57BL/6 mice.35–37 In parallel experiments with an identical lentiviral vector expressing hF.IX, no anti-hF.IX response was observed.37 These data underscore the complexity of immune responses following viral gene transfer, which are likely influenced by the vector, route of administration/target tissue, promoter, and the particular antigen expressed. Follow-up studies may also reveal the influence of


inflammation caused by potentially hepatotoxic vectors such as adenovirus and lentivirus on immune responses, an issue that may have to be addressed in models other than C57BL/6 mice.38,39 While i.v. infusion of the identical adenoviral vector used in this study results in long-term expression of human F.IX in hemophilia B mice on a C57BL/6 background,12 i.v. infusion of hF.IX protein in the form of recombinant F.IX (used routinely for treatment of hemophilia B patients) efficiently induces antibody formation in the same strain of mice (data not shown). The infused mice developed high titer IgG1 anti-human F.IX after only two i.v. infusions of the protein. Therefore, in C57BL/6 mice i.v. administration of the highly immunogenic adenoviral vector appears less immunogenic than conventional, nongene-based treatment. Toxicity as a consequence of adenoviral gene transfer is linked to immune responses against adenoviral antigens, in particular inflammatory responses. An additional concern for treatment of hemophilia is the development of antibodies against the transgene-encoded coagulation factor, the most serious complication of the current protein treatment of hemophilia. Results obtained from canine models of hemophilia illustrate limitations in the extrapolation from results obtained with strains of inbred mice. Studies on adenoviral gene transfer of F.VIII or F.IX to the livers of nonhuman primates or hemophilic dogs demonstrated hepatotoxicity, inflammatory responses and humoral immune responses against the human or canine transgene products.6,7,38 These studies show that the data presented here and in other studies based on C57BL/6 mice often cannot be extrapolated to other species. The mechanism of tolerance to F.IX by i.v. injection of adenoviral vector in C57BL/6 mice remains elusive, but illustrates the difficulty of extrapolation of results obtained in inbred strains of mice and highlights the importance of studies in other animal models.

Materials and methods Viral vectors E1/E3-deleted adenoviral vector ad-hF.IX expressing human F.IX from the CMV enhancer/promoter was constructed as described.11 The vector was replicated in HEK-293 cells and purified from cell lysate by two rounds of CsCl density gradient centrifugation using standard protocols. The vector has a total particles:p.f.u. ratio of 25:1 as determined by plaque assay.11 AAV-CMVhF.IX-3 vector was manufactured by Avigen (Alameda, CA, USA). This vector was produced by triple transfection of HEK-293 cells in an adenovirus-free system, purified by two rounds of CsCl density gradient centrifugation, and titered by quantitative dot blot hybridization as described.40 The three plasmids used in this production system include the AAV vector plasmid that contains the F.IX expression cassette flanked by AAV-2 145bp ITRs; an adenoviral helper plasmid (expressing E2A, E4orf6, and VA genes) for replication of AAV in the absence of helper virus; and a third plasmid containing AAV-2 rep and cap functions. This plasmid has been engineered to eliminate the potential for formation of pseudo wild-type AAV, which is undetectable in vector preparations by infectious center assay with a limit of detection of 1 in 109 viral particles.41 Purity of AAV vector

preparations was addressed by silver staining and Western blotting of viral proteins separated by SDS-PAGE as described.40 The human F.IX transgene in the AAV vector is also under transcriptional control of the CMV enhancer/promoter, and contains a portion of intron I of the human F.IX gene for increased levels of expression.18,13,42

359

Animal experiments Male inbred C57BL/6 and BALB/c mice were purchased from Jackson Laboratories (Bar Harbor, ME, USA). Male FBV mice were obtained from Charles River Breeding Laboratories (Wilmington, MA, USA). IL-10-deficient mice on a C57BL/6 or BALB/c background (4-week-old male mice) were maintained at the University of Pennsylvania School of Veterinary Medicine.19 Normal C57BL/6 mice (6 weeks old, 20–25 g) were injected with either adhFIX (i.v., tail vein injections) or AAV-CMVhFIX-3 (i.m.) at a dose of 4 × 1010 vector particles per animal using techniques described earlier.11,18 i.m. injections were at two sites of one hind limb. After 8 weeks, the vectorinjected mice (n = 3–4 per cohort) then received a second injection of the other vector via a different route of administration, ie ad-hFIX i.v. injected mice received AAV-CMV-hFIX-3 i.m., and AAV-CMV-hF.IX-3 i.m. injected mice received ad-hFIX i.v. Mice were subsequently bled retro-orbitally on a biweekly schedule using micro-capillary tubes, in order to obtain plasma samples for measurement of human F.IX antigen and anti-hF.IX antibodies. FBV (n = 4), and IL-10-deficient mice (n = 3 per genetic background) received ad-hF.IX vector i.v. (4 × 1010 particles per mouse). These mice were bled at days 0, 7 and 14 after injection, and biweekly thereafter. Normal BALB/c and C57BL/6 mice (n = 3 per background) were injected in identical fashion as controls. Hemophilia B mice with a large gene deletion (resulting in absence of F.IX transcript and antigen) on a C57BL/6 background12 (n = 3) received two i.v. injections (2 weeks apart, via tail vein) of 2 ␮g recombinant human F.IX (Benefix; Genetics Institute, Cambridge, MA, USA). Measurement of human F.IX antigen and anti-hF.IX Human F.IX antigen in plasma samples of mice was measured by ELISA as described.11 This assay does not detect murine F.IX. Antibodies against human F.IX were detected in plasma samples (1:32 dilution) by IgG subclass-specific ELISA (Boehringer Mannheim, Indianapolis, IN, USA), and quantitated using purified mouse IgG1, IgG2a, and IgG2b (Sigma Chemical Company, St Louis, MO, USA) as described.17 C57BL/6 mice express IgG2c instead of the homologous IgG2a immunoglobulin. Lymphocyte proliferation assay C57BL/6 mice that had received i.v. injections of ad-hF.IX were challenged twice by subcutaneous injection of 2 ␮g of recombinant human F.IX formulated in complete Freund’s adjuvant (FA) 4 and 5 months after adenoviral vector administration. Two weeks later, mice were killed, and their spleens isolated for bulk culture of splenocytes as described.17 Additionally, naive C57BL/6 mice were i.m. injected with 4 × 1010 particles of ad-hF.IX as described.17 These mice were killed 8 months after vector administration. Cells (triplicates of 105 cells per well in 96-well plates) were incubated with two-fold serial dilutions of human F.IX starting from 10 ␮g/ml. Controls Gene Therapy


360

were mock-stimulated cells (media only) and cells nonspecifically stimulated with concanavalin A. Lymphocyte culture media and conditions were as described.17 After 5 days of culture, lymphocytes were labeled with 3H-thymidine for 8 h. Thymidine up-take was measured with a scintillation counter. The stimulation index (SI) was calculated as the ratio of thymidine up-take in F.IXstimulated versus mock-stimulated lymphocytes.

Acknowledgements This work was supported by NIH grant R01 HL61921 to KAH and NIH grant AI-27828 to JPF. PAF was supported by grants from the Medical Research Council (MRC, UK) and the Katherine Dormandy Trust for Hemophilia. RWH is supported by a Career Development Award from the National Hemophilia Foundation. The authors thank Dr Robert Coffman for providing the BALB/c IL10 knock-out mice and Avigen, Inc., a company in which KAH holds equity, for providing AAV vector.

References 1 Hoyer LW. The incidence of factor VIII inhibitors in patients with severe hemophilia A. Adv Exp Med Biol 1995; 386: 35–46. 2 High KA. Factor IX: molecular structure, epitopes, and mutations associated with inhibitor formation. Adv Exp Med Biol 1995; 386: 79–86. 3 Herzog RW, High KA. AAV-mediated gene transfer of factor IX for treatment of hemophilia B by gene therapy. Thr Haemost 1999; 82: 540–546. 4 Herzog RW, High KA. Problems and prospects in gene therapy for hemophilia. Curr Opin Hematol 1998; 5: 321–326. 5 Kay MA et al. Evidence for gene transfer and expression of blood coagulation factor IX in patients with severe hemophilia B treated with an AAV vector. Nat Genet 2000; 24: 257–261. 6 Kay MA et al. In vivo hepatic gene therapy: complete albeit transient correction of factor IX deficiency in hemophilia B dogs. Proc Natl Acad Sci USA 1994; 91: 2353–2357. 7 Lozier JN, Metzger ME, Donahue RE, Morgan RA. Adenovirusmediated expression of human coagulation factor IX in the rhesus macaque is associated with dose limiting toxicity. Blood 1999; 94: 3968–3975. 8 Smith TAG et al. Adenovirus-mediated expression of therapeutic plasma levels of human factor IX in mice. Nat Genet 1994; 5: 397–402. 9 Fang B et al. Lack of persistence of E1− recombinant adenoviral vectors containing a temperature-sensitive E2A mutation in immunocompetent mice and hemophilia B dog. Gene Therapy 1996; 3: 217–222. 10 Connelly S et al. Complete short-term correction of canine hemophilia A by in vivo gene therapy. Blood 1996; 88: 3846–3853. 11 Walter J et al. Successful expression of human factor IX following repeat administration of an adenoviral vector in mice. Proc Natl Acad Sci USA 1996; 93: 3056–3061. 12 Kung J et al. Human F.IX corrects the bleeding diathesis of mice with hemophilia B. Blood 1998; 91: 784–790. 13 Connelly S et al. Sustained phenotypic correction of murine hemophilia A by in vivo gene therapy. Blood 1998; 91: 3273–3281. 14 Yang Y, Xiang Z, Ertl HCJ, Wilson JM. Upregulation of class I major histocompatibility complex antigens by interferon ␥ is necessary for T-cell-mediated elimination of recombinant adenovirus-infected hepatocytes in vivo. Proc Natl Acad Sci USA 1995; 92: 7257–7261. 15 Yang Y et al. Immune response to viral antigens versus transgene product in the elimination of recombinant adenovirus-infected hepatocytes in vivo. Gene Therapy 1996; 3: 137–144. 16 Qian J, Bi L, Kazazian HH, Hoyer LW. Inhibitor development and T cell response to human factor VIII in murine hemophilia A. Blood 1996; 88 (Suppl. 1): 656a. Gene Therapy

17 Fields PA et al. Choice of vector determines T cell subsets involved in immune responses against the secreted transgene product factor IX. Molec Ther 2000; 1:225–235. 18 Herzog RW et al. Stable gene transfer and expression of human blood coagulation factor IX after intramuscular injection of recombinant adeno-associated virus. Proc Natl Acad Sci USA 1997; 94: 5804–5809. 19 Murphy ML, Wille U, Hunter CA, Farrell JP. IL-10 deficient mice are resistant to Leishmania donovani infection. FASEB J 1999; 13 (Suppl. S): A655. 20 Michou A et al. Adenovirus-mediated gene transfer: influence of transgene, mouse strain and type of immune response on persistence of transgene expression. Gene Therapy 1997; 4: 473–482. 21 Vrancken Peeters M-JTFD, Lieber A, Perkins J, Kay MA. Method for multiple portal vein infusions in mice: quantitation of adenovirus-mediated hepatic gene transfer. BioTechniques 1996; 20: 278–285. 22 Barr D et al. Strain related variations in adenovirally mediated transgene expression from mouse hepatocytes in vivo: comparisons between immunocompetent and immunodeficient inbred strains. Gene Therapy 1995; 2: 151–155. 23 Pastore L et al. Use of liver-specific promoter reduces immune response to the transgene in adenoviral vectors. Hum Gene Ther 1999; 10: 1773–1781. 24 Jooss K, Yang Y, Fisher KJ, Wilson JM. Transduction of dendritic cells by DNA viral vectors directs the immune response to transgene products in muscle fibers. J Virol 1998; 72: 4212–4223. 25 Waal Malefyt RD et al. Interleukin-10 (IL-10) inhibits cytokine synthesis by human monocytes: an autoregulatory role of IL-10 produced by monocytes. J Exp Med 1991; 174: 1209–1220. 26 Groux H, Bigler M, Roncaralo MG. Interleukin 10 induces a long-term antigen-specific anergic state in human CD4+ T cells. J Exp Med 1996; 184: 19–29. 27 Buer J et al. Interleukin-10 secretion and impaired effector function of major histocompatibility complex class II-restricted T cells anergized in vivo. J Exp Med 1998; 187: 177–183. 28 Knolle PA et al. Interleukin-10 expression is autoregulated at the transcriptional level in human and murine Kupffer cells. Hepatology 1998; 27: 93–99. 29 Loeser P, Jennings GS, Strauss M, Sandig V. Reactivation of the previously silenced cytomegalovirus major immediate–early promoter in the mouse liver: involvement of NF␬B. J Virol 1998; 72: 180–190. 30 Vemuru RP et al. Immune tolerance to a defined heterologous antigen after intrasplenic hepatocyte transplantation – implications for gene therapy. FASEB J 1992; 6: 2836–2842. 31 Nakai H et al. AAV-mediated gene transfer of human blood coagulation factor IX into mouse liver. Blood 1998; 91: 4600–4607. 32 Snyder RO et al. Persistent and therapeutic concentrations of human factor IX in mice after hepatic gene transfer of recombinant AAV vectors. Nat Genet 1997; 16: 270–276. 33 Snyder RO et al. Correction of hemophilia B in canine and murine models using recombinant adeno-associated viral vectors. Nat Med 1999; 5: 64–70. 34 Nathwani A, Davidoff AM, Hanawa H, Nienhuis AW. The site of rAAV-mediated F.IX expression can significantly influence the level of expression and immunological response to the transgene product. Mol Ther 2000; 1: S287. 35 Chao H, Mao L, Bruce AT, Walsh CE. Sustained expression of human factor VIII in mice using a parvovirus-based vector. Blood 2000; 95: 1594–1599. 36 Balague C et al. Sustained high-level expression of full-length human factor VIII and restoration of clotting activity in hemophilic mice using a minimal adenovirus vector. Blood 2000; 95: 820–828. 37 Park F, Ohashi K, Kay MA. Therapeutic levels of human factor VIII and IX using HIV-based lentiviral vectors in mouse liver. Blood 2000; 96: 1173–1176. 38 Brann T et al. Adenoviral vector-mediated expression of physiological levels of human factor VIII in non-human primates. Hum Gene Ther 1999; 10: 2999–3011.


39 Park F et al. Efficient lentiviral transduction of liver requires cell cycling in vivo. Nat Genet 2000; 24: 49–52. 40 Matsushita T. Adeno-associated virus vector can be efficiently produced without helper virus. Gene Therapy 1998; 5: 938–945. 41 Matsushita T, Godwin S, Surosky R, Colosi P. Improvements in

AAV vector production: elimination of pseudo-wild-type AAV. Proc 2nd Ann Am Soc Gene Therapy. Washington, DC, 1999, p 2a. 42 Hagstrom JN et al. Enhanced muscle-derived expression of coagulation factor IX from a skeletal actin/CMV hybrid promoter. Blood 2000; 95: 2536–2542.

361

Gene Therapy