Antibody to CD40 ligand inhibits both humoral and cellular immune ...

2 downloads 0 Views 294KB Size Report
able of inhibiting humoral immunity, might enhance the elimination of Ad vector-transduced cells by CTLs. Activated T cells play a critical role in the generation of.
Gene Therapy (1997) 4, 611–617  1997 Stockton Press All rights reserved 0969-7128/97 $12.00

Antibody to CD40 ligand inhibits both humoral and cellular immune responses to adenoviral vectors and facilitates repeated administration to mouse airway A Scaria1, JA St George1, RJ Gregory1, RJ Noelle2, SC Wadsworth1, AE Smith1 and JM Kaplan1 1

Genzyme Corporation, Framingham, MA; 2Department of Microbiology, Dartmouth Medical School, Lebanon, NH, USA

Adenoviral vectors have been used successfully to transfer the human CFTR cDNA to respiratory epithelium in animal models and to CF patients in vivo. However, studies done primarily in mice, indicate that present vector systems have limitations. Among other things, transgene expression in the lung is transient and the production of neutralizing antibodies against adenovirus correlates with a reduced ability to readminister a vector of the same serotype. Here we demonstrate that in mice, a transient blockade of costimulation between activated T cells and B cells/antigen presenting cells using a monoclonal antibody (MR1)

against murine CD40 ligand inhibits the development of neutralizing antibodies to adenoviral (Ad) vector. MR1 also decreased the cellular immune response to Ad vector and allowed an increase in persistence of transgene expression. Furthermore, when administered with a second dose of Ad vector to mice preimmunized against vector, MR1 was able to interfere with the development of a secondary antibody response and allowed for high levels of transgene expression upon a third administration of vector to the airway.

Keywords: adenovirus vectors; immune response; anti-CD40 ligand; repeat administration

Introduction E1-deleted replication-defective adenoviral (Ad) vectors are attractive vehicles for gene transfer because of their ability to transduce a wide variety of dividing and nondividing cells in vivo.1–5 Such vectors have been used for gene transfer to the respiratory epithelium of experimental animals and of individuals with cystic fibrosis (CF).2–5 Studies from several laboratories have suggested that administration of high doses of first generation Ad vector results in only transient gene expression in the lung due, at least in part, to destruction of vector-transduced cells by host cellular immune responses (predominantly CD8+ cytotoxic T cells) directed against viral proteins and/or immunogenic transgene products.6–9 Reduction of this response has been reported following the use of second generation vectors with decreased viral gene expression10,11 and with transgenes encoding self rather than foreign proteins.8 The treatment of chronic diseases like CF with Ad vectors will likely require repeated administrations throughout the lifetime of the patient. However, another limitation of current vectors is the difficulty in obtaining successful readministration using a vector of the same Ad serotype. Several groups have now demonstrated that a strong dose-dependent humoral immune response is induced by Ad vectors leading to the development of neutralizing antibodies to adenoviruses.6,7,12–15 Studies

Correspondence: A Scaria, Genzyme Corporation, One Mountain Road, Framingham, MA 01701, USA Received 30 October 1996; accepted 12 February 1997

using immunodeficient mice have shown that this process is dependent on MHC class II presentation of the input viral proteins and activation of CD4+ T cells and can be induced by inactive as well as active viral particles,6 (Kaplan et al, unpublished data). Several investigators have tested the use of broad immunosuppressants11 and cytoablative agents16 to overcome the immune response of the host to first generation Ad vectors. Kay et al17 have shown that transient coadministration of soluble CTLA4-Ig along with an intravenous injection of Ad vector expressing a nonimmunogenic transgene product (human a-1 anti-trypsin) leads to persistent transgene expression from mouse liver. CTLA4-Ig blocks the B7-CD28 pathway of T cell costimulation.18 Although Ad-specific antibody levels were reduced in CTLA4-Ig-treated mice, the inhibition was not sufficient to allow secondary gene transfer under the conditions tested.17 Yang et al19 demonstrated that coadministration of interferon-g (INF-g) or interleukin-12 (IL-12) with recombinant Ad vector diminished the formation of neutralizing antibodies and allowed readministration of vector to mouse airways. However, IL-12 is a potent mediator which affects Th1 type CD4+ T cell responses and is involved in stimulating natural killer cells and promoting the differentiation of cytotoxic T cells.20–22 INF-g is known to up-regulate MHC class I on antigen presenting cells.23 Thus, both INF-g and IL-12, while capable of inhibiting humoral immunity, might enhance the elimination of Ad vector-transduced cells by CTLs. Activated T cells play a critical role in the generation of humoral and cellular immune responses. The interaction between the T cell receptor (TCR) and antigen-major histocompatibility complex (MHC) expressed on the surface

Repeat administration of adenoviral vector A Scaria et al

612

of an antigen presenting cell (APC) is necessary but not sufficient for the optimal activation of T cells which also requires additional costimulatory signals provided by several receptor–ligand pairs including B7-CD28 and CD40-CD40 ligand (CD40L).18 CD40L is expressed transiently at high levels on activated CD4+ T cells.24,25 The costimulation provided by CD40L on T cells interacting with CD40 on B cells and other APCs seems to be essential for thymus-dependent humoral immunity25–28 and may also play an important role in the generation of cellular immune responses through the production of helper cytokines.28–30 In this article, we describe experiments aimed at blocking this costimulation pathway transiently using a monoclonal antibody against CD40L. The results of these experiments show that, in the context of administration of an Ad vector to mouse airways, antiCD40L inhibited both humoral and cellular immune responses.

Results Anti-CD40 ligand suppresses development of Ad-specific antibodies To explore the role of the CD40–CD40L interaction in the generation of antibodies to Ad vectors, BALB/c mice were injected intraperitoneally with anti-CD40L monoclonal antibody (MR1, 200 mg/injection per mouse on days −2, +2, +6 and +10). Ad2/CFTR2 vector (109 IU) was instilled intranasally on day 0. Analysis of serum from these animals by ELISA showed a marked decrease in anti-Ad (IgG + IgM + IgA) titers in MR1-treated mice for up to 41 days (Figure 1a). Analysis of bronchoalveolar lavage (BAL) fluid revealed a parallel drop in Ad-specific IgA levels in MR1-treated mice (Figure 1b). On day 38 after the first administration of Ad2/CFTR2, a vector of the same serotype, Ad2/bGal2, was administered to the different groups of mice. Expression of bgalactosidase was measured by a quantitative assay on day 41. As might have been predicted from the reduced levels of pulmonary anti-Ad IgA, b-galactosidase levels were elevated in MR1-treated mice compared with control mice (Figure 2). Decreased CTL response against Ad vector, increase in persistence and decreased inflammation BALB/c mice were instilled intranasally with 109 IU of Ad2/bGal-4 on day 0 and injected with MR1 on days −2, +2, +5 and +8. As shown in Figure 3a, spleen cells from MR1-treated mice showed decreased yet measurable levels of CTL activity compared with spleen cells from untreated control mice, albeit using an assay that is not strictly quantitative. Since CTLs have been implicated in the loss of transgene expression, we tested whether administration of MR1 with an Ad vector would give rise to prolonged transgene expression. As shown in Figure 3b, transgene expression measured by a quantitative galactosidase assay, declined to background levels by day 21 in the untreated controls. By comparison, in MR1-treated mice, transgene expression also declined but remained consistently higher than in untreated mice. Lung tissue from the mice was examined for evidence of histopathological changes in the peribronchial, perivascular and alveolar regions. Lung inflammation

Figure 1 (a) Adenovirus-specific serum antibodies. Ad2/CFTR2 vector (109 IU) was instilled on day 0 and MR1 (200 mg per mouse per injection) was given IP on days −2, +2, +6 and +10. Serial two-fold dilutions of serum collected at different time-points were analyzed by ELISA. Data are presented as the mean titer of four individual animals ± standard error of the mean (s.e.m.). *P , 0.001 using Student’s t test, for all time-points comparing ± MR1. (b) Adenovirus-specific IgA levels on day 40. BAL samples obtained from mice on day 40 following administration of vector were analyzed by ELISA. Results are expressed as the mean IgA concentration from three individual mice ± s.e.m. *P , 0.05 using Student’s t test.

characterized by inflammatory cell infiltrates was present on day 5 and was not yet resolved at day 21. On day 5, there were no differences noted between the lungs of mice treated with the vector either with or without MR1 antibody treatment (data not shown). However, on day 21, there were fewer inflammatory changes in all regions of the lungs from mice treated with the antibody (Figure 4). The inflammatory cell infiltrate was markedly reduced in the peribronchial/peribronchiolar and perivascular regions.

Secondary antibody response in a preimmunized host MR1 was tested for its ability to interfere with the secondary antibody response in mice that had been preimmunized with Ad vector. Mice were instilled with 108 IU of Ad2/CFTR2 on day 0 and readministered with 108 IU

Repeat administration of adenoviral vector A Scaria et al

613

Figure 2 Efficient re-administration of Ad vector. Ad2/CFTR2 (10 9 IU) and MR1 administrations were as described for Figure 1. Ad2/bGal2 virus was administered to mice on day 38 and the lungs assayed for bgalactosidase levels on day 41. Data are presented as means ± s.e.m. (n = 4). *P , 0.05 using Student’s t test, comparing ± MR1.

of Ad2/CFTR2 on day 50. We had previously determined that an intranasal instillation of 108 IU of Ad vector elicits both humoral and cellular immune responses to the vector in several strains of mice including BALB/c (data not shown). MR1 injections were given around the time of the second virus administration on days 44, 48, 52 and 56. Sera from the different groups of animals were analyzed both for anti-Ad ELISA titers (IgG + IgM + IgA) and Ad neutralizing titers at various time-points. The secondary antibody response measured by ELISA was only slightly decreased in the MR1-treated mice but was less sustained (Figure 5a). The antibody titers declined more rapidly and by the time of the third vector administration on day 99, both the ELISA titers and Ad neutralizing titers were clearly decreased in the mice that had received MR1 along with the second administration of Ad vector (Figure 5a and b). The b-galactosidase encoding Ad2/bGal-2 vector was given as the third administration on day 99 and bgalactosidase levels in the lung were measured 3 days later on day 102. Control mice that had received the first (day 0) and second dose (day 50) of Ad2/CFTR2 showed only 6% of the b-galactosidase activity in lung homogenates compared with naive mice that received a single intranasal administration of Ad2/bGal-2 (Figure 6). Mice that received MR1 around the time of the second administration of Ad2/CFTR2 showed 10-fold higher levels of transgene expression after the third administration compared with untreated control mice (Figure 6). The effectiveness of the third vector administration in the MR1treated animals was comparable to that obtained in naive mice.

Discussion Ad-based gene therapy for genetic diseases such as cystic fibrosis will likely require repeated administrations due to the transient nature of transgene expression from current vectors. While improvements in vector design should increase persistence of expression to some degree, the barrier of humoral immunity against Ad also needs to be overcome before Ad-based gene therapy becomes

Figure 3 (a) Decreased CTL response against Ad vector. Mice were instilled intranasally with 109 IU of Ad2/bGal4 vector on day 0 and injected with MR1 (250 mg per mouse per injection) on days −2, +2, +5 and +8. On day 21, spleen cells were collected, restimulated in vitro with infected syngeneic fibroblasts and tested for cytolytic activity. Results shown are mean percentage lysis ± s.e.m. from triplicate wells at various effector:target ratios. (b) Increased persistence of transgene expression. Lungs from parallel groups of mice were assayed for b-galactosidase levels at various time-points after Ad2/bGal4 administration. Data are presented as means ± s.e.m. (n = 4).

a clinical reality. We and others6,7,13,14 have shown that administration of Ad vectors to the airways leads to the activation of Ad-specific T and B cells. The cellular immune response includes CD8+ T cells which, in many animal models, appear to be involved in the rapid decline

Repeat administration of adenoviral vector A Scaria et al

614

Figure 4 Decreased lung inflammation in MR1-treated mice. Mice were instilled with 109 IU of Ad2/CFTR2 on day 0 and given MR1 (200 mg per mouse per injection) on days −2, +2, +6 and +10. Lungs were fixed and examined for histopathological changes. Results shown are for day 21.

in transgene expression.6–9 The humoral immune response, which is dependent on MHC class II-mediated antigen presentation to CD4+ T cells,6 promotes the generation of Ad-specific neutralizing antibodies of both IgA and IgG isotypes.14 This serotype-specific antibody response correlates with the reduced effectiveness of repeated administrations of a vector of the same serotype.15 In this article, we demonstrate that a transient blockade of costimulation between T cells and B cells and other APCs using a monoclonal antibody (MR1) against CD40 ligand suppresses the development of antibodies against Ad and allows for efficient readministration of vector. MR1 treatment partially decreased the cellular immune response to Ad vector and this correlated with an increase in persistence of transgene expression. As expected, MR1 treatment did not affect the nonspecific acute phase of inflammation induced by intranasal administration of Ad vector (day 5, not shown), but the level of pathology observed during the chronic T celldependent phase of inflammation6 (day 21) was markedly reduced in MR1-treated mice. Most CF patients will have been exposed to human adenoviruses at some point in their lives and probably will have immunological memory against the serotype(s) of adenovirus to which they have been exposed. It is possible that the pre-existing anti-Ad humoral immunity in these individuals might interfere with a single or repeated Ad vector administration. To test this directly, we studied gene expression in mice that were already immunized against Ad vector. We found that coadministration of MR1 inhibited the development and persistence of a secondary antibody response and resulted in high levels of transgene expression upon a third administration of vector to the airway. This finding is consistent with reports that CD40L is critical for formation of germinal centers as well as their maintenance.31,32 Recent studies in animal models have shown that treatment with anti-CD40L in vivo can block the development of collagen-induced arthritis33 and graft-versus-host disease (GVHD)34 and prevent rejection of pancreatic islet

Figure 5 (a) Effect of MR1 on secondary antibody response. Mice were instilled with 108 IU of Ad2/CFTR2 on days 0 and 50. MR1 injections were given on days 44, 48, 52 and 56. Serial two-fold dilutions of sera collected at different time-points were assayed for Ad-specific antibodies by ELISA. Data are presented as the mean titer of four individual animals ± s.e.m.. *P , 0.05 and **P , 0.02 using Student’s t test. (b) Neutralizing antibody titers at the time of readministration. Sera from day 94 samples were also analyzed for anti-Ad neutralizing titers as described in Materials and methods. Data are presented as means ± s.e.m. (n = 4). *P , 0.05 using Student’s t test.

allografts.35 In models of acute GVHD, it has been shown that anti-CD40L can completely block the generation of anti-host CTL responses. It has been proposed that this is due to the ability of anti-CD40L to interfere with the maturation of professional APC so that they do not acquire appropriate costimulatory molecules optimally to present antigen.36 Recent studies confirm that CD40 may play an important role in dendritic cell function 37 and therefore impact on the priming of T cells.38 In this study, we have demonstrated the inhibitory effect of MR1 on humoral and cellular immune responses against Ad vectors for several weeks after the last administration of

Repeat administration of adenoviral vector A Scaria et al

615

Figure 6 Third administration of Ad vector. Ad2/CFTR2 (108 IU) and MR1 administrations were as described for Figure 5. Mice were then instilled with 2 × 109 IU of Ad2/bGal2 on day 99 and lungs were assayed for b-galactosidase levels on day 102. Data are presented as means ± s.e.m. (n = 4). *P , 0.05 using Student’s t test, comparing group 1 versus 2 and group 3 versus 4.

MR1, at which time less than 2% of biologically active MR1 should still be present in the serum.25 It therefore appears possible to readminister an Ad vector using a transient blockade of T cell costimulation without complete immunosuppression of the host for prolonged periods of time. Since it is known that several interactions including both B7–CD28 and CD40–CD40L are essential for optimal costimulation between T cells and APC,18,28,39 a combined therapy with CTLA4Ig and MR1 might be even more effective. Although the practicability of coadministration of viral vectors and humanized recombinant antibodies to CF patients is questionable, such studies have and will reveal more details about the process of immune recognition of Ad vectors and may suggest simpler strategies to block that process.

Materials and methods Antibody injections MR1, a hamster anti-mouse CD40 ligand monoclonal antibody was produced in ascites and purified by ion exchange HPLC as previously described.24 BALB/c mice were purchased from Taconic (Germantown, NY, USA) and were typically injected intraperitoneally (IP) with a total of four injections of 200 or 250 mg of MR1/mouse starting on day −2 relative to the time of administration of the Ad vector.

Adenoviral vectors The construction of Ad2/CFTR2 and Ad2/bGal-2 has been described previously.7 They are both Ad2-based vectors with most of the E1 region deleted and replaced with the transgene and the E4 region replaced with the open reading frame 6 (ORF6) of E4. Ad2/bGal-4 is identical to Ad2/bGal-2 except that it contains the complete wild-type E4 region. Histopathology On the day of death, mice were killed with an IP injection of Somlethal (Euthasol; King Pharmaceuticals, Bristol, TN, USA). The lungs were cleared of blood by vascular perfusion with phosphate-buffered saline (PBS). The trachea was cannulated and the lungs and trachea removed. The lungs were fixed by inflation with 2% paraformaldehyde with 0.2% glutaraldehyde in PBS, pH 7.4, at a pressure of 30 cm of H2 O. Following overnight fixation, portions of the left lung were embedded in glycomethacrylate, sectioned and stained with hematoxylin and eosin. These sections were evaluated by light microscopy for the presence and distribution of lung inflammation without previous knowledge of treatment.40 The lung sections were subjectively assessed for morphologic alterations on a scale of 0–4: 0 = no lesion, 1 = minimal, 2 = mild, 3 = moderate, 4 = severe. Adenovirus-specific antibodies Titers of Ad-specific serum antibodies were evaluated by ELISA. Serial two-fold dilutions of sample were added to

Repeat administration of adenoviral vector A Scaria et al

616

the wells of a 96-well plate coated with photochemically inactivated Ad2 (Lee Biomolecular Research, San Diego, CA, USA). Bound virus-specific antibodies were detected by the addition of horseradish peroxidase (HRP)-conjugated goat anti-mouse Ig (IgG, IgM, IgA-specific; Jackson Immunoresearch Laboratories, West Grove, PA, USA). The titer was defined as the reciprocal of the highest dilution of sample which produced an OD490 greater than 0.1. To evaluate levels of Ad-specific IgA in BAL, samples were diluted two-fold and added to Ad2-coated plates followed by the addition of HRP-conjugated goat antimouse IgA (a chain-specific; Cappel, Durham, NC, USA). For quantification, a standard curve was constructed using a monoclonal antibody against mouse IgA (Harlan Sera-Lab, Sussex, UK) to coat ELISA plates and capture known amounts of purified mouse IgA (Cappel). The OD490 values obtained following the addition of HRPconjugated goat anti-mouse IgA were plotted against the amounts of IgA standard (ng/ml) added to the wells. The concentrations of Ad-specific IgA present in BAL samples were then derived from the standard curve by linear regression analysis.

Cytotoxic T cell assay To evaluate cytotoxic T lymphocyte (CTL) activity, spleen cells from animals in the same group were pooled and stimulated in vitro with mitomycin-C-inactivated, syngeneic fibroblasts infected with Ad2/bGal-4 at a multiplicity of infection (MOI) of 50 for 24 h. Cells were cultured in 24-well plates containing 5 × 106 spleen cells and 6 × 104 stimulator fibroblasts per well in a 2 ml volume. The culture medium consisted of RPMI-1640 medium (Gibco, Grand Island, NY, USA) supplemented with 100 U/ml penicillin, 100 mg/ml streptomycin, 2 mm glutamine, 5 × 10−5 m 2-mercaptoethanol, 20 mm Hepes buffer and 10% heat-inactivated fetal calf serum (Hyclone, Logan, UT, USA). Cytolytic activity was assayed after 5– 7 days of culture. Target fibroblasts were infected with vector at an MOI of 100 for 48 h and were treated with 100 U/ml recombinant mouse g-interferon (Genzyme, Cambridge, MA, USA) for approximately 24 h before use to enhance MHC class I expression and antigen presentation to effector CTLs. The fibroblasts were labeled with 51 chromium (51Cr; NEN) overnight (50 mCi/105 cells) and added to the wells of a round-bottom 96-well plate in a 100 ml volume (5 × 103 fibroblasts per well). Effector cells were added in a 100 ml volume at various effector:target cell ratios in triplicate. After 5 h of incubation at 37°C/5% CO2 , 100 ml of cell-free supernatant was collected from each well and counted in a Packard (Downers Grove, IL, USA) Multi-Prias gamma counter. The amount of 51Cr spontaneously released was obtained by incubating target fibroblasts in medium alone and the total amount of 51 Cr incorporated was determined by adding 1% Triton X-100 in distilled water. The percentage lysis was calculated as follows: % Lysis =

(sample c.p.m.) − (spontaneous c.p.m.) × 100 (total c.p.m.) − (spontaneous c.p.m.)

b-Galactosidase expression For quantification of b-gal expression, lungs from individual mice were homogenized and assayed using the AMPGD kit obtained from Tropix, Bedford, MA, USA.

The protein concentration in an individual sample was determined using the BioRad DC reagent (BioRad, Hercules, CA, USA) and the results are expressed as relative light units (RLU)/mg protein.

Neutralizing antibody titers To determine neutralizing antibody titers, serial two-fold dilutions of sample were incubated with live Ad2/CFTR2 for 1 h at 37°C/5% CO2 in the wells of flat-bottom 96well plates. At the end of the incubation period, permissive 293 cells were added to the wells and the plates were incubated at 37°C/5% CO2 for 72–96 h. The assay was read when control 293 cells incubated alone reached >90% confluency. The neutralizing antibody titer was defined as the reciprocal of the highest dilution of sample that showed any detectable protection of 293 cells from cytopathic effects when compared with cells incubated with untreated virus or virus incubated with seronegative serum.

Acknowledgements We wish to thank the Virus Production Unit at Genzyme Corporation for preparation of the Ad vectors used in these studies, Wiley Smith and Carol Sacks for technical help with the animal studies, Sarah Pennington, Kristen Couture and Margaret Nichols for their participation in the immunological and b-galactosidase assays.

References 1 Stratford-Perricaudet L et al. Evaluation of the transfer and expression in mice of an enzyme-encoding gene using a human adenovirus vector. Hum Gene Ther 1990; 1: 241–256. 2 Rosenfeld MA et al. In vivo transfer of the human cystic fibrosis transmembrane conductance regulator gene to the airway epithelium. Cell 1992; 68: 143–155. 3 Zabner J et al. Adenovirus-mediated gene transfer transiently corrects the chloride transport defect in nasal epithelium of patients with cystic fibrosis. Cell 1993; 75: 207–216. 4 Crystal RG et al. Administration of an adenovirus containing the human CFTR cDNA to the respiratory tract of individuals with cystic fibrosis. Nat Genet 1994; 8: 42–51. 5 Zabner J et al. Safety and efficacy of repetitive adenovirusmediated transfer of CFTR cDNA to airway epithelia of primates and cotton rats. Nat Genet 1994; 6: 75–83. 6 Yang Y, Li Q, Ertl HCJ, Wilson JM Cellular and humoral immune responses to viral antigens create barriers to lungdirected gene therapy with recombinant adenoviruses. J Virol 1995; 69: 2004–2015. 7 Kaplan JM et al. Humoral and cellular immune responses of non-human primates to long-term repeated lung exposure to Ad2/CFTR-2. Gene Therapy 1996; 3: 117–127. 8 Tripathy SK, Black HB, Goldwasser E, Leiden JM. Immune responses to transgene-encoded proteins limit the stability of gene expression after injection of replication-defective adenovirus. Nature Med 1996; 2: 545–550. 9 Yang Y et al. Immune responses to viral antigens versus transgene product in the elimination of recombinant adenovirusinfected hepatocytes in vivo. Gene Therapy 1996; 3: 137–144. 10 Yang Y et al. Inactivation of E2a in recombinant adenoviruses improves the prospect for gene therapy in cystic fibrosis. Nature Genet 1994; 7: 362–369. 11 Engelhardt JF, Ye X, Doranz B, Wilson JM. Ablation of E2A in recombinant adenoviruses improves transgene persistence and decreases inflammatory response in mouse liver. Proc Natl Acad Sci USA 1994; 91: 6196–6200. 12 Smith TAG et al. Adenovirus mediated expression of therapeutic plasma levels of human factor IX in mice. Gene Therapy 1993; 5: 397–402.

Repeat administration of adenoviral vector A Scaria et al

13 Yei S et al. Adenovirus mediated gene transfer for cystic fibrosis: quantitative evaluation of repeated in vivo vector administration to the lung. Gene Therapy 1994; 1: 192–200. 14 Van Ginkel FW et al. Intratracheal gene delivery with adenoviral vector induces elevated systemic IgG and mucosal IgA antibodies to adenovirus and b-galactosidase. Hum Gene Ther 1995; 6: 895–903. 15 Mastrangeli A et al. ‘Sero-swith’ adenovirus-mediated in vivo gene transfer: circumvention of anti-adenovirus humoral immune defences against repeat adenovirus vector administration by changing the adenovirus serotype. Hum Gene Ther 1996; 7: 79–87. 16 Dai Y et al. Cellular and humoral responses to adenoviral vectors containing factor IX gene: tolerization of factor IX and vector antigens allows for long term expression. Proc Natl Acad Sci USA 1995; 92: 1401–1405. 17 Kay M et al. Long-term hepatic adenovirus-mediated expression in mice following CTLA4Ig administration. Nat Genet 1995; 11: 191–197. 18 Lenschow DJ et al. CD28/B7 system of T cell costimulation. Ann Rev Immunol 1996; 14: 233–258. 19 Yang Y, Trinchieri G, Wilson JM. Recombinant IL-12 prevents formation of blocking IgA antibodies to recombinant adenovirus and allows repeated gene therapy to mouse lung. Nature Med 1995; 1: 890–893. 20 Paul WE, Seder RA. Lymphocyte responses and cytokines. Cell 1994; 76: 241–251. 21 Trinchieri G. Interleukin 12: a cytokine produced by antigenpresenting cells with immunoregulatory functions in the generation of T-helper cells type 1 and cytotoxic lymphocytes. Blood 1994; 84: 4008–4027. 22 Bliss J et al. IL-12 as an adjuvant, promotes a T helper 1 cell, but does not suppress a T helper 2 cell recall response. J Immunol 1996; 156: 887–894. 23 Yang Y, Xiang Z, Ertl HCJ, Wilson JM. Upregulation of class I major histocompatibility complex antigens by interferon g is necessary for T-cell-mediated elimination of recombinant adenovirus-infected hepatocytes in vivo. Proc Natl Acad Sci USA 1995; 92: 7257–7261. 24 Noelle RJ et al. A 39-kDa protein on activated helper T cells binds CD40 and transduces the signal for cognate activation of B cells. Proc Natl Acad Sci USA 1992; 89: 6550–6554. 25 Foy TM et al. In vivo CD40-gp39 interactions are essential for thymus-dependent humoral immunity. II. Prolonged suppression of the humoral immune response by an antibody to the ligand for CD40, gp39. J Exp Med 1993; 178: 1567–1575.

26 Kawabe T et al. The immune responses in CD40-deficient mice: impaired immunoglobin class switching and germinal center formation. Immunity 1994; 1: 167–178. 27 Xu J et al. Mice deficient for CD40 ligand. Immunity 1994; 1: 423–431. 28 Foy TM et al. Immune regulation by CD40 and its ligand gp39. Ann Rev Immunol 1996; 14: 591–617. 29 Stuber E, Strober W, Neurath M. Blocking of CD40–CD40L interaction in vivo specifically prevents priming of T helper 1 cells through the inhibition of IL-12 secretion. J Exp Med 1996; 183: 693–698. 30 Stout RD et al. Impaired T cell-mediated macrophage activation in CD40L-deficient mice. J Immunol 1996; 156: 8–11. 31 Foy TM et al. gp39–CD40 interactions are essential for germinal center formation and the development of B cell memory. J Exp Med 1994; 180: 157–163. 32 Han S et al. Cellular interactions in germinal centers. Roles of CD40 ligand and B7–2 in established germinal centers. J Immunol 1995; 155: 556–567. 33 Durie FH et al. Prevention of collagen-induced arthritis with an antibody to gp39, the ligand for CD40. Science 1993; 261: 1328–1330. 34 Durie FH et al. Antibody to the ligand of CD40, gp39, blocks the occurrence of the acute and chronic forms of graft-vs-host disease. J Clin Invest 1994; 94: 1333–1338. 35 Parker DC et al. Survival of mouse pancreatic islet allografts in recipients treated with allogeneic small lymphocytes and antibody to CD40 ligand. Proc Natl Acad Sci USA 1995; 92: 9560– 9564. 36 Buhlmann JE et al. In the absence of a CD40 signal, B cells are tolerogenic. Immunity 1995; 2: 645–653. 37 Grouard G, de Bouteiller O, Banchereau J, Liu Y. Human follicular dendritic cells enhance cytokine-dependent growth and differentiation of CD40-activated B cells. J Immunol 1995; 155: 3345–3352. 38 Grewal IS, Xu J, Flavell RA. Impairment of antigen-specific Tcell priming in mice lacking CD40 ligand. Nature 1995; 378: 617–620. 39 Larsen CP et al. Long-term acceptance of skin and cardiac allografts after blocking CD40 and CD28 pathways. Nature 1996; 381: 434–438. 40 Ginsberg HS et al. A mouse model for investigating the molecular pathogenesis of adenovirus pneumonia. Proc Natl Acad Sci USA 1991; 88: 1651–1655.

617