Long term transgene expression by hepatocytes transduced with retroviral vectors requires induction of immune tolerance to the transgene

Journal of Hepatology 41 (2004) 222–228 www.elsevier.com/locate/jhep

Long term transgene expression by hepatocytes transduced with retroviral vectors requires induction of immune tolerance to the transgene Juliana Puppi1, Carole Guillonneau2, Virginie Pichard1, Marta Bellodi-Privato1, Maria Cristina Cuturi2, Ignacio Anegon2, Nicolas Ferry1,* 1

Biothe´rapie He´patique, CIC-INSERM 00-04, CHU Hoˆtel Dieu, 44093 Nantes cedex 01, France 2 INSERM Unite´ 643, CHU Hotel-Dieu, 44035 Nantes cedex, France

Background/Aims: Gene therapy for inherited liver diseases requires permanent expression of the therapeutic gene. However, in vivo liver transduction with retroviral vectors triggers an immune elimination of transduced hepatocytes. Here we investigated whether immune response could be prevented by treatment with compounds known to induce tolerance in organ transplantation: CTLA4Ig and LF-15-0195. Methods: CTLA4Ig was administered either via i.p. injection of the drug or by i.m. injection of recombinant adenoviruses encoding CTLA4Ig. LF-15-0195 was administered i.p. All animals were subjected to partial hepatectomy and received b-galactosidase retroviral vectors intravenously. Appearance of anti-b-galactosidase antibodies was monitored and the number of positive hepatocytes was assessed at day 7 and at sacrifice. Results: No b-galactosidase antibodies were detected as long as CTLA4Ig was detectable in serum. Short-term treatment with CTLA4Ig induced tolerance in a significant proportion of animals only at high dose (1 mg/kg). Administration of CTLA4Ig adenovectors resulted in prolonged secretion of CTLA4Ig and permanent absence of anti-b-galactosidase antibodies. LF-15-0915 administration achieved tolerance in some animals. Conclusions: In conclusion, manipulation of the immune system at the time of virus delivery using clinically relevant tolerance-inducing protocols is a promising approach to achieve long term expression after retrovirus-mediated gene transfer to the liver. q 2004 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. Keywords: Gene therapy; Immune response; Retrovirus; Hepatocyte; b-galactosidase

1. Introduction The ultimate goal of gene therapy for liver inherited diseases is to ensure a permanent and sufficient expression of a therapeutic gene to reverse the disease phenotype. During the past years, many efforts have been devoted to the improvement of gene transfer vectors, either from viral origin or not [1,2]. Recently, long-term correction of a liver genetic disease has been obtained in large animal models using recombinant viral vectors, thus highlighting the validity of this approach. Complete and permanent correction Received 29 January 2003; received in revised form 22 March 2004; accepted 8 April 2004; available online 5 May 2004 * Corresponding author. Tel.: þ33-2-40-08-74-88; fax: þ33-2-40-08-7506. E-mail address: [email protected] (N. Ferry).

of hemophilia B was obtained in dog using recombinant AAV encoding factor IX delivered to the muscle or the liver [3,4]. This successful experimental study led to clinical trials using AAV vectors delivered to the muscle or the liver [5,6]. Similarly, sustained correction of type VII mucopolysaccharidosis in b-glucuronidase deficient dogs was achieved after administration of recombinant retroviral vectors in newborn animals [7]. However, in both models, the genetic defect resulted from a missense mutation that affected only one amino acid in the corresponding protein. Consequently, the AAV factor IX clinical trials only enrolled patients with missense mutations [6]. Since many genetic defects result from large genetic deletions as well as point mutation leading to a premature stop codon, gene transfer may result in immune response against the therapeutic protein which is recognized as non self. Regarding AAV vectors, experiments in hemophilia dogs

0168-8278/$30.00 q 2004 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.jhep.2004.04.015

J. Puppi et al. / Journal of Hepatology 41 (2004) 222–228

with a factor IX null mutation as well as in mice bearing a large factor IX deletion revealed that AAV vectors delivered to the liver may induce tolerance in some but not all animals [8,9]. However, the mechanism of tolerance induction is not clear and depends on numerous parameters, mainly the animal strain. In contrast to AAV, we previously demonstrated that gene transfer to the regenerating liver using recombinant murine retroviral vectors resulted in strong cytotoxic as well as humoral response directed against the transgene product and precluded sustained correction in a rat model of Crigler –Najjar disease [10,11]. Induction of the immune response likely results from transduction of antigen presenting cells that are susceptible to infection with recombinant retroviral vectors [12,13]. We therefore reasoned that compounds susceptible to induce immune tolerance to the transgene product would allow persistent transgene expression after liver gene delivery using retroviral vectors. To this end we used the CTLA4Ig molecule (a fusion protein between the extracellular portion of CTLA4 and the constant portion of an immunoglobulin) that is able to inhibit immune responses by blockade of the CD80/CD28 costimulation signal or by induction of tryptophane catabolism in dendritic cells [14]. We and others have shown that CTLA4Ig treatment could induce transplantation tolerising mechanisms in vivo [15,16]. Previous studies have shown that treatment with CTLA4Ig increased the duration of transgene expression after hepatic delivery using adenoviral vectors [17,18]. We sought to determine whether short-term or sustained administration of CTLA4Ig could similarly impair immune response against the b-galactosidase protein expressed by hepatocytes after retroviral delivery. We also tested in the same experimental model a new analog of deoxyspergualin, LF 15-0195, which we and others have shown to induce donor specific tolerance to cardiac allograft after short term (20 days) administration through the induction of tolerogenic T and dendritic cells [19 –21].

223

weighing 150– 200 g, were used. All animal experiments were performed in accordance with the guidelines of French Ministe`re de l’Agriculture. The surgical procedure was conducted on deeply anesthetized animals using isoflurane inhalation (3% v/v in air). The two-third partial hepatectomy was performed according to the procedure of Higgins and Anderson [23]. Retroviral injections were carried out 24 h after partial hepatectomy by bolus injection of 3 ml of retrovirus-containing medium in the dorsal penile vein. Liver biopsies were harvested 7 days after virus infusion from the right lobe and fixed in formalin. Animals were killed at day 45, day 60 or day 127 as indicated in the tables. CTLA4Ig (kindly provided by Bristol Myers Squibb, Princeton, NJ, USA) and LF 15-0195 (kindly provided by Laboratoires Fournier, Dijon, France) were administered by intra peritoneal injections starting the day of virus infusion. CTLA4Ig was given at various doses (20 mg/kg, 100 mg/kg and 1 mg/kg for 8 days every other day and LF 15-0195 was administered at 3 mg/kg every day for 20 days). Recombinant CTLA4Ig adenoviral vectors were injected in the tibialis anterior at a dose of 1011 infectious particles/animal immediately after injection of retroviral vectors.

2.3. Detection of circulating CTLA4Ig Serum CTLA4Ig was quantitated by sandwich ELISA [15]. Plates were coated with a hamster anti-murine CTLA4 mAb (kindly provided by Dr Bluestone, Chicago, IL, USA). After blocking with PBS containing Tween 20 (0.1% v/v) and BSA (1% v/v) and rinsing, serial dilutions of serum were applied for 2 h at 37 8C. After washing, peroxidase conjugated goat anti human IgG was added and the reaction was developed using ABTS and absorbance was read at 405 nm. The detection limit of this ELISA test is 10 ng/ml.

2.4. b-galactosidase immunohistochemistry The presence of b-galactosidase was assessed by immunohistochemistry on formalin-fixed/paraffin-embedded sections (5 mm) as previously described [11] using polyclonal primary rabbit anti-b gal antibody (Chemicon International Inc. Temecula, CA, USA) at 1:2000 dilution. Positive cells were visualized with biotinylated goat anti-mouse immunoglobulin and streptavidin-peroxidase using amino ethyl carbazol (AEC) as a chromogenic substrate. The number of positive hepatocytes present in the whole liver was recorded after microscopic examination of at least 10 fields at £ 40 magnification, corresponding approximately to at least 3000 hepatocytes in hematoxylin counterstained sections.

2. Material and methods 2.1. Viral vectors Recombinant retroviruses containing the b-galactosidase gene coupled to a nuclear localization signal and under transcriptional control of the viral LTR were produced as previously described [22]. Before use, 8 mg/ml polybrezne (hexamethedine bromide) was added to the supernatant. Titers were determined by endpoint dilution using Te671 target cells and were routinely 5 £ 107 TU/ml. CTLA4-Ig recombinant adenoviral vectors were obtained as previously described [16]. They harbor the sequence encoding the extracellular portion of the murine CTLA4 fused to human IgG1 Fc fragment and placed under the transcriptional control of the CMV promoter.

2.2. Animal experiments Adult male rats from the Sprague Dawley strain, aged 5–7 weeks and

2.5. Antibody detection The presence of antibodies directed against b-galactosidase in rat serum was detected by direct ELISA as described [11]. Briefly, 96-well dishes were coated overnight at 4 8C with b-galactosidase at 10 mg/ml. After rinsing serial dilutions of the serum were incubated for 1.5 h at 37 8C. The presence of antibodies was revealed using biotinylated anti-rat IgG immunoglobulin and streptavidin/peroxidase followed by incubation with ABTS substrate.

2.6. Statistical analysis ANOVA or Student’s t test were used to compare quantitative data between groups. A P value of less than 0.05 was considered statistically significant. Two tailed Fisher exact test was used for categorical data.

224

J. Puppi et al. / Journal of Hepatology 41 (2004) 222–228

3.2. Treatment with CTLA4Ig

group 1 (i.e. injected with CTLA4Ig at 20 mg/kg) had detectable antibodies 3 weeks after gene transfer. The last animal in this group developed antibodies at day 45 after gene transfer. The animals were then sacrificed and liver immunohistochemistry revealed no b-galactosidase expressing hepatocytes indicating that elimination of transduced cells had taken place. In group 2, low levels of CTLA4Ig were detectable in the serum at day 2 and 8 but were below detection thereafter. No animals had anti-b gal antibodies at day 21 (Table 2). Three out of 6 animals were positive at day 28. One animal became positive after day 35. One animal (rat #65) died at day 45 but was still negative at this time. Finally, only one animal did not raised anti-b-galactosidase antibodies until day 120. After sacrifice, liver histology revealed that 5.6% of hepatocytes still expressed b-galactosidase in this animal. All other animals were negative for hepatocyte b-galactosidase expression at sacrifice. The proportion of tolerant animals in this group, either at day 45 when one animal died (2 out of 6) or at day 127 (1 out of 6) was not statistically different from the one observed in the control group (1 out of 22) using Fisher exact test ðP . 0:1Þ: In the third group that received the highest dose (1 mg/kg) the circulating levels of CTLA4Ig were positive until day 35 but became undetectable thereafter (Table 3). No animals had detectable anti-b-galactosidase antibody until day 60. Two animals became positive after day 60. At the time of sacrifice (day 89) the two last animals were still negative for b-galactosidase antibodies and liver immunohistochemistry revealed 7.8% b-galactosidase positive hepatocytes in both rats, whereas no positive b-galactosidase hepatocytes were present in the two other animals. The proportion of tolerant animals was significantly different from the one observed in the control series, both at day 60

We first assessed the potency of recombinant CTLA4Ig administered at the time of virus delivery to impair immune response. To this end, rats were divided in three groups and received five injections of CTLA4Ig (at the day of virus injection and every other day until day 8) at escalating dosage: group 1: 20 mg/kg, group 2: 100 mg/kg and group 3: 1 mg/kg. At day 7, a liver biopsy was harvested to check for the number of transduced hepatocytes in the liver (Fig. 1). The results obtained for each group are reported in Tables 1– 3. There was no significant differences between the three groups for the initial number of b-galactosidase positive hepatocytes present at day 7 (P ¼ 0:1192 using ANOVA). The presence of anti-b gal antibodies in the serum of treated animals was then monitored by ELISA. The presence of antibodies at a dilution of 1:1000 or higher was considered as a positive result. We also monitored by ELISA the amount of circulating CTLA4Ig in the serum of animals. The results for group 1 are shown in Table 1. In this series, we never detected circulating CTLA4Ig in the serum. Serum analysis demonstrated that 4 out of 5 animals in

Fig. 1. Detection of b-galactosidase positive hepatocytes. The presence of b-galactosidase positive hepatocytes was detected by immunohistochemistry. Because of the presence of a nuclear localisation signal, b-galactosidase is concentrated in the nucleus and positive cells appear with a brown nucleus. Hematoxylin counterstained. Original magnification 3220. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article).

3. Results 3.1. Experimental model Our experimental model is aimed at achieving permanent expression of an exogenous gene in hepatocytes using retroviral vectors. Since murine retroviral vectors are only able to infect dividing cells, administration of viruses takes place 24 h after partial hepatectomy to induce hepatocyte division. Viruses were injected intravenously at a dose of 1.5 £ 108 infectious particles and the hepatic expression of the transgene was monitored during the time. Using b-galactosidase as a marker, we previously documented efficient gene expression from transduced hepatocytes early after gene transfer using this approach [24]. However, we also reported elimination of transduced hepatocytes by a cytotoxic immune response as early as 3 weeks after gene transfer [10,11]. This immune response also resulted in appearance of anti-b-galactosidase antibodies and the detection of antibodies paralleled the cytotoxic elimination of transduced hepatocytes. In a series of 22 animals, the mean number of b-galactosidase positive hepatocytes was 5.8 ^ 2.3% at day 7. However, in 21 animals we observed the disappearance of transduced hepatocytes concomitant with the appearance of anti-b-galactosidase antibodies at high titer (1:5000 dilution or less) in less than 30 days. Only one animal did not develop antibodies and had prolonged expression of b-galactosidase in the liver until sacrifice at day 289. Therefore, we sought to determine the ability of drugs known to induce immune tolerance in transplantation models to make long term expression of the b-galactosidase in our model possible.

J. Puppi et al. / Journal of Hepatology 41 (2004) 222–228

225

Table 1 Treatment with CTLA4Ig 20 mg/kg % bgal þ cell at day 7

Rat number

43 72 73 74 75

Serum CTLA4 Ig (mg/ml) day 2

,0.01 ,0.01 ,0.01 ,0.01 ,0.01

10.2 15.9 14.1 11.2 14.7

% bgal cell at sacrifice (day 45)

Anti-b-gal antibodies Day 21

Day 28

Day 35

Day 45

1:5000 1:5000 1:1000 1:5000 neg

N.D. N.D. N.D. N.D. neg

N.D. N.D. N.D. N.D. neg

N.D. N.D. N.D. N.D. 1:5000

0 0 0 0 0

For anti-b-galactosidase antibodies, the highest dilution yielding a positive signal is indicated. N.D., not determined; neg, negative at 1:50 dilution.

Table 2 Treatment with CTLA4Ig at 100 mg/kg Rat number

65 66 67 68 69 70

% bgal þ cell at day 7

12.3 13.3 9.2 9.4 4.5 9.1

% bgal þ cell at sacrifice (day 127)

Serum CTLA4 Ig (mg/ml)

Anti-b-gal antibodies

Day 2

Day 8

Day 21

Day 28

Day 35

Day 45

Day 60

Day 90

Day 127

0.266 0.256 0.243 0.240 0.360 0.203

0.284 0.371 0.414 0.200 0.365 0.322

neg neg neg neg neg neg

neg 1:5000 neg 1:1000 neg 1:5000

neg N.D. neg N.D. neg 1:5000

neg N.D. neg N.D. 1:5000 N.D.

– N.D. neg N.D. N.D. N.D.

– N.D. neg N.D. N.D. N.D.

– N.D. neg N.D. N.D. N.D.

N.D. 0 5.6 0 0 0

For anti-b-galactosidase antibodies, the highest dilution yielding a positive signal is indicated. N.D., not determined; neg, negative at 1:50 dilution; Rat #65 died at day 45.

Table 3 Treatment with CTLA4Ig at 1 mg/kg Rat number

36 37 38 39

% bgal þ cell at day 7

12.3 13.2 9.2 9.5

Serum CTLA4 Ig (mg/ml)

% bgal cell at sacrifice (day 89)

Anti-b-gal antibodies

Day 8

Day 14

Day 21

Day 35

Day 45

Day 21

Day 28

Day 35

Day 45

Day 60

Day 89

9.5 10.6 11.8 15.4

2.2 1.3 3.3 8.4

1.06 0.4 1.3 4.5

0.43 0.04 0.28 0.86

,0.01 ,0.01 ,0.01 ,0.01

neg neg neg neg

neg neg neg neg

neg neg neg neg

neg neg neg neg

neg neg neg neg

neg 1:5000 neg 1:5000

7.8 0 7.8 0

For anti-b-galactosidase antibodies, the highest dilution yielding a positive signal is indicated. neg, negative at 1:50 dilution.

and at day 89 (P ¼ 0:0002 and 0.05, respectively, using Fisher exact test). 3.3. Treatment with CTLA4Ig adenovirus vector We next assessed whether sustained expression of CTLA4Ig would allow stable expression of the transgene. To this end, CTLA4Ig was delivered by using intra muscular injection of a recombinant adenoviral vector encoding the CTLA4Ig cDNA. In a series of rats ðn ¼ 9Þ intravenous infusion of recombinant b-galactosidase retroviruses was carried out one day after partial hepatectomy to transduce regenerating hepatocytes. Immediately after b-galactosidase retrovirus administration, CTLA4Ig

adenoviruses were injected in the tibialis anterior muscle. As shown in Table 4, in all but one animal CTLA4Ig was consistently present at moderate levels until day 14 and was still detected at later time points although at lower level. (60 days follow-up for rats 177, 178 and 180 –182 and 127 days for rats 78– 81). We never observed the presence of anti-bgalactosidase antibodies in rats that had detectable circulating levels of CTLA4Ig. In contrast, in the animal that lost CTLA4Ig expression after day 14 (rat 178), anti-bgalactosidase antibodies were present from day 35. Finally, immunohistochemistry at the time of sacrifice revealed that b-galactosidase positive hepatocytes were still present in the liver in animals that did not mount an immune response. The mean number of b-galactosidase positive hepatocytes

4.6 4.9 5.8 7.3 0.9 0 1.4 2.5 5.6 neg neg neg neg – – – – – neg neg neg neg neg N.D. neg neg neg 6.21 2.46 4.08 5.46 2.18 N.D. 3.7 N.D. 0.07 5.2 6.5 8.6 10.6 1.5 2.7 2.8 4.1 6.3 78 79 80 81 177 178 180 181 182

2.67 2.81 2.01 7.03 1.74 1.67 2.63 4.19 2.58

3.91 1.89 5.81 3.5 1.79 0 3.33 3.81 0.56

7.87 1.8 4.05 5.45 2.07 0 2.08 5.04 0.04

3.72 2.32 4.39 6.64 1.86 0 2.54 6.6 0.05

3.31 0.61 0.61 0.37 – – – – –

neg neg neg neg neg neg neg neg neg

neg neg neg neg neg neg neg neg neg

neg neg neg neg neg 1:5000 neg neg neg

neg neg neg neg neg N.D. neg neg neg

Day 127 Day 60 Day 45 Day 35 Day 28 Day 21 Day 60 Day 14

Day 21

Day 35

Day 45

Day 127

Anti-b-gal antibodies Serum CTLA4 Ig (mg/ml) % bgal cell at day 7 Rat number

Table 4 Treatment with adeno CTLA4Ig

For anti-b-galactosidase antibodies, the highest dilution yielding a positive signal is indicated. neg, negative at 1:50 dilution; N.D., not determined. – , data not available for rat sacrificed at day 60.

J. Puppi et al. / Journal of Hepatology 41 (2004) 222–228 % bgal cell at sacrifice (day 60 or 127)

226

in the liver at day 7 (5.4 ^ 3%) was lower in this group as compared to the mean from all animals in previous groups (11.2 ^ 2.9%, P , 0:001 using the Student’s t test). This may be due to a lower titer of the virus used in this group or to a deleterious effect of AdCTLA4Ig on the kinetics of liver regeneration. However, the mean number of b-galactosidase positive hepatocytes at sacrifice (4.1 ^ 2.2%) was similar to the one observed in liver biopsies performed at day 7 (P . 0:4 using the Student’s t test) demonstrating the absence of selective elimination of transduced hepatocytes. The proportion of tolerant animals at day 60 was highly different from the one in the control group (P , 0:0001 using Fisher exact test). 3.4. Treatment with LF 15-0195 Finally we assessed whether LF 15-0195, a new analog of deoxyspergualine previously used in solid organ transplantation could induce tolerance to b-galactosidase. A series of rat ðn ¼ 5Þ received LF 15-015 by intraperitoneal injection for 20 days starting the day of virus administration. This schedule was in accordance with previous studies using the same compound in the setting of heart transplantation [20]. The results obtained are shown in Table 5. All animals were negative for b-galactosidase antibodies until day 35. Two animals had detectable anti-bgalactosidase antibodies from day 45 and one from day 60. Once again, liver immunohistochemistry at sacrifice revealed the absence of positive cells in those animals that developed an immune response. The difference in the proportion of tolerant animals at day 60 as compared to the control group almost reach statistical significance (P ¼ 0:079 using the two tailed Fisher exact test). However, at day 45 the difference was highly significant ðP ¼ 0:01Þ thus indicating a significant delay in the immune response.

4. Discussion In the present study we addressed the question as to whether interference with the function of antigen presenting cells could suppress immune response to the transgene product after in vivo gene transfer to the liver using recombinant retroviral vectors. Using CTLA4Ig, we observed that detectable circulating levels of the molecule consistently resulted in suppression of the anti-transgene immune response. No animals had anti-b-galactosidase antibodies at the time they had detectable circulating levels of CTLA4Ig. When CTLA4Ig secretion was made permanent using recombinant adenoviral vectors, transduced hepatocytes were permanently protected against immune response. In contrast, when CTLA4Ig was administered for short periods of time and at low doses (20 or 100 mg/kg), tolerance was not significantly induced in experimental animals and b-galactosidase specific immune response occurred when the circulating

J. Puppi et al. / Journal of Hepatology 41 (2004) 222–228

227

Table 5 Treatment with LF 15-0195 Rat number

LF 1 LF 2 LF 3 LF 4 LF 5

% bgal þ cell at day 7

14.6 14.7 16.2 15.3 17.1

% bgal cell at sacrifice (day 60)

Anti-b-gal antibodies Day 21

Day 28

Day 35

Day 45

Day 60

neg neg neg neg neg

neg neg neg neg neg

neg neg neg neg neg

neg 1:5000 neg 1:5000 neg

1:5000 N.D. neg 1:5000 neg

0 0 12 0 13.1

For anti-b-galactosidase antibodies, the highest dilution yielding a positive signal is indicated. neg, negative at 1:50 dilution; N.D., not determined.

level of CTLA4Ig dropped. However, when using larger doses of CTLA4Ig (1 mg/kg) the proportion of animals that did not mount an immune response was significantly increased when compared with control animals. These data demonstrated that CTLA4Ig used alone is able to block the activation of immune response after retrovirus injection as long as it is present in the serum. These findings are in agreement with previous studies aimed at circumventing immune response after adenoviral vectors mediated transduction of the liver. Indeed, permanent tolerance to the transgene product could be achieved using CTLA4Ig alone only when permanent secretion of the molecule was obtained via CTLA4Ig gene transfer [18,25,26]. On the contrast, short-term administration of CTLA4Ig resulted in modest increase of transgene expression and modest delay of the immune response [17,27]. Long term expression of the transgene without immune response against the transgene product or the adenoviral proteins using a short treatment was successfully achieved only when CTLA4Ig and anti CD40 ligand antibodies were used in combination [27,28]. The same hold true for allogenic solid organ transplantation and long term graft acceptance could only been achieved when using concomitant blocking of CD28 and CD40L interactions [29] or long term CTLA4Ig treatment using adenoviral vectors [15]. Therefore in the setting of retrovirus mediated gene transfer to the liver, combination of other immune modulators with drugs interfering with CD28 signaling or acting on tryptophane metabolism could enhance the proportion of tolerant animals obtained using CTLA4Ig alone. We also demonstrated that a deoxyspergualine analog, LF 15-0195, was able to delay appearance of immune response. In addition, tolerance was achieved in a fraction of the animals, as was the case with CTLA4Ig. LF 15-0195 has already been shown to induce tolerance to cardiac allografts after transient administration [20]. It was subsequently demonstrated that this effect relied on induction of CD4þ CD25þ regulatory T cells although tolerogenic dendritic cells were also involved [19,21]. Indeed, LF 15-0195 has been shown to reduce the activation of antigen presenting cells by suppression of NF-KB signaling [30]. Since these effects were obtained using LF 15-0195 alone, we reasoned that we could also obtained long term tolerance in our

model. Our failure to get tolerance in all animals indicates that the immune response induced by retroviral gene transfer is robust and is difficult to antagonize using agents that suppress allogenic immune response. We hypothesize that direct transduction of antigen presenting cells with retroviral vectors activates these cells at sufficient level to reduce the effect of immunomodulatory drugs acting on these already activated antigen presenting cells. Pretreatment with LF 15-0195 before virus delivery could improve tolerance induction. Altogether, our present data point to immune response as a major hurdle to the development of liver directed gene therapy for genetic diseases in which a large part of the therapeutic protein is not expressed.

Acknowledgements This work was supported by grants from Association Franc¸aise contre les Myopathies (AFM). Adenoviral vectors were kindly provided by the Gene Vector Production Network from AFM.

References [1] Ferry N, Heard JM. Liver-directed gene transfer vectors. Hum Gene Ther 1998;9:1975 –1981. [2] Kay MA, Glorioso JC, Naldini L. Viral vectors for gene therapy: the art of turning infectious agents into vehicles of therapeutics. Nat Med 2001;7:33–40. [3] 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 adeno-associated viral vector. Nat Med 1999;5:56–63. [4] Snyder RO, Miao C, Meuse L, Tubb J, Donahue BA, Lin HF, et al. Correction of hemophilia B in canine and murine models using recombinant adeno-associated viral vectors. Nat Med 1999;5:64– 70. [5] 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. [6] Manno CS, Chew AJ, Hutchison S, Larson PJ, Herzog RW, Arruda VR, et al. AAV-mediated factor IX gene transfer to skeletal muscle in patients with severe hemophilia B. Blood 2003;101:2963–2972. [7] Ponder KP, Melniczek JR, Xu L, Weil MA, O’Malley TM, O’Donnell

228

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

J. Puppi et al. / Journal of Hepatology 41 (2004) 222–228 PA, et al. Therapeutic neonatal hepatic gene therapy in mucopolysaccharidosis VII dogs. Proc Natl Acad Sci USA 2002;99: 13102– 13107. Mingozzi F, Liu YL, Dobrzynski E, Kaufhold A, Liu JH, Wang Y, et al. Induction of immune tolerance to coagulation factor IX antigen by in vivo hepatic gene transfer. J Clin Invest 2003;111:1347–1356. Mount JD, Herzog RW, Tillson DM, Goodman SA, Robinson N, McCleland ML, et al. Sustained phenotypic correction of hemophilia B dogs with a factor IX null mutation by liver-directed gene therapy. Blood 2002;99:2670–2676. Izembart A, Aguado E, Gauthier O, Aubert D, Moullier P, Ferry N. In vivo retrovirus-mediated gene transfer to the liver of dogs results in transient expression and induction of a cytotoxic immune response. Hum Gene Ther 1999;10:2917 –2925. Aubert D, Menoret S, Chiari E, Pichard V, Durand S, Tesson L, et al. Cytotoxic immune response blunts long-term transgene expression after efficient retroviral-mediated hepatic gene transfer in rat. Mol Ther 2002;5:388– 396. McCormack JE, Edwards W, Sensintaffer J, Lillegren L, Kozloski M, Brumm D, et al. Factors affecting long-term expression of a secreted transgene product after intravenous administration of a retroviral vector. Mol Ther 2001;3:516– 525. Irwin MJ, Laube LS, Lee V, Austin M, Chada S, Anderson CG, et al. Direct injection of a recombinant retroviral vector induces human immunodeficiency virus-specific immune responses in mice and nonhuman primates. J Virol 1994;68:5036 –5044. Grohmann U, Orabona C, Fallarino F, Vacca C, Calcinaro F, Falorni A, et al. CTLA-4-Ig regulates tryptophan catabolism in vivo. Nat Immunol 2002;3:1097–1101. Guillot C, Mathieu P, Coathalem H, Le Mauff B, Castro MG, Tesson L, et al. Tolerance to cardiac allografts via local and systemic mechanisms after adenovirus-mediated CTLA4Ig expression. J Immunol 2000;164:5258–5268. Guillot C, Menoret S, Guillonneau C, Braudeau C, Castro MG, Lowenstein P, et al. Active suppression of allogeneic proliferative responses by dendritic cells after induction of long-term allograft survival by CTLA4Ig. Blood 2003;101:3325–3333. Jooss K, Turka LA, Wilson JM. Blunting of immune responses to adenoviral vectors in mouse liver and lung with CTLA4Ig. Gene Ther 1998;5:309 –319. Thummala NR, Ghosh SS, Lee SW, Reddy B, Davidson A, Horwitz MS, et al. A non-immunogenic adenoviral vector, coexpressing CTLA4Ig and bilirubin –uridine – diphosphoglucuronateglucuronosyltransferase permits long-term, repeatable transgene expression in the Gunn rat model of Crigler–Najjar syndrome. Gene Ther 2002;9: 981–990. Chiffoleau E, Beriou G, Dutartre P, Usal C, Soulillou JP, Cuturi MC.

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

Role for thymic and splenic regulatory CD4 þ T cells induced by donor dendritic cells in allograft tolerance by LF 15-0195 treatment. J Immunol 2002;168:5058–5069. Chiffoleau E, Beriou G, Dutartre P, Usal C, Soulillou JP, Cuturi MC. Induction of donor-specific allograft tolerance by short-term treatment with LF 15-0195 after transplantation. Evidence for a direct effect on T-cell differentiation. Am J Transplant 2002;2:745–757. Min WP, Zhou D, Ichim TE, Strejan GH, Xia X, Yang J, et al. Inhibitory feedback loop between tolerogenic dendritic cells and regulatory T cells in transplant tolerance. J Immunol 2003;170: 1304–1312. Aubert D, Pichard V, Durand S, Moullier P, Ferry N. Cytotoxic immune response after retroviral-mediated hepatic gene transfer in rat does not preclude expression from adeno-associated virus 1 transduced muscles. Hum Gene Ther 2003;14:473–481. Higgins GM, Anderson RM. Experimental pathology of the liver I. Restoration of the liver of the white rat following partial surgical removal. Arch Pathol 1931;12:186– 202. Kitten O, Cosset FL, Ferry N. Highly efficient retrovirus-mediated gene transfer into rat hepatocytes in vivo. Hum Gene Ther 1997;8: 1491–1494. Schowalter DB, Meuse L, Wilson CB, Linsley PS, Kay MA. Constitutive expression of murine CTLA4Ig from a recombinant adenovirus vector results in prolonged transgene expression. Gene Ther 1997;4:853–860. Nakagawa I, Murakami M, Ijima K, Chikuma S, Saito I, Kanegae Y, et al. Persistent and secondary adenovirus-mediated hepatic gene expression using adenovirus vector containing CTLA4IgG. Hum Gene Ther 1998;9:1739– 1745. Kay MA, Meuse L, Gown AM, Linsley P, Hollenbaugh D, Aruffo A, et al. Transient immunomodulation with anti-CD40 ligand antibody and CTLA4Ig enhances persistence and secondary adenovirusmediated gene transfer into mouse liver. Proc Natl Acad Sci USA 1997;94:4686–4691. Jiang ZL, Reay D, Kreppel F, Gambotto A, Feingold E, Kochanek S, et al. Local high-capacity adenovirus-mediated mCTLA4Ig and mCD40Ig expression prolongs recombinant gene expression in skeletal muscle. Mol Ther 2001;3:892– 900. Larsen CP, Elwood ET, Alexander DZ, Ritchie SC, Hendrix R, Tucker-Burden C, et al. Long-term acceptance of skin and cardiac allografts after blocking CD40 and CD28 pathways. Nature 1996;381: 434 –438. Yang J, Bernier SM, Ichim TE, Li M, Xia X, Zhou D, et al. LF15-0195 generates tolerogenic dendritic cells by suppression of NF-kappaB signaling through inhibition of IKK activity. J Leukoc Biol 2003;74: 438 –447.