Abstract

Gene therapy for the X-linked bleeding disorder hemophilia B may be limited by immune responses to the factor IX (F.IX) gene product. Hepatic adeno-associated virus (AAV) gene transfer can induce immune tolerance to F.IX (

JCI
111
:
1347
,
PNAS
103
:
4592
). Tolerance is associated with activation of regulatory cells that suppress antibody formation to F.IX. In order to identify these regulatory cells, splenocytes of C57BL/6 mice tolerized to human F.IX (hF.IX) by heptic gene transfer (portal vein injection of 1x1011 AAV vector genomes) were adoptively transferred to naive mice of the same strain. Recipient mice were immunized with hF.IX in adjuvant on the next day. Compared to cells transferred from control animals (no gene transfer), total splenocytes, CD4+ cells, or CD4+CD25+ cells were equally efficient in suppression of anti-hF.IX formation (n=7–8 per experimental group, P<0.02 for comparison to total splenocytes, CD4+ cells, or CD4+CD25- cells of controls), while CD4- cells failed to suppress, and CD4+CD25- cells were inefficient. CD4+CD25+ from naive control mice, which contain regulatory T cells but lack specificity for hF.IX, gave highly variable results and on average failed to suppress. When tolerized C57BL/6 mice were challenged with hF.IX/adjuvant, the animals lacked antibody formation to hF.IX and in vitro cytokine release and showed an ~2-fold increase in FoxP3 message in splenic CD4+ cells in vivo. Taken together, these data indicate that induction of regulatory CD4+CD25+ T cells is part of the tolerance mechanism. However, the significance of this finding was unclear. In the next experiment, C57BL/6 mice received hepatic AAV-hF.IX gene transfer and were additionally injected with rat anti-mouse CD25 or with isotype control rat IgG (ip injections at days 0, 14, 28, and 42, n=5 per group). Analysis of peripheral blood cells by flow cytometry showed presence of CD4+CD25+ cells at a frequency of 8–10% in controls and undetectable levels in anti-CD25 treated mice. By day 49, 4/5 anti-CD25 treated mice had a low-titer, but detectable antibody (IgG1) to hF.IX. Subsequent challenge with hF.IX/cF.IX caused a rise in anti-hF.IX to 0.5–2 μg/ml in 3/5 anti-CD25 treated mice within 3 weeks. None of the mice treated with control IgG (0/5) developed a detectable antibody to hF.IX even after challenge. These data demonstrate that CD4+CD25+ regulatory T cells are required for tolerance induction to F.IX. Thus far, we failed to break tolerance by depletion of CD25+ cells at later time points (i.e. during the maintenance phase of tolerance, when other mechanisms such as T cell anergy and deletion may become more prevalent). To obtain definitive evidence for induction of CD4+CD25+ Treg, hepatic AAV-ova gene transfer was performed in DO11.10-tg Rag-2 −/− BALB/c mice, which are deficient in Treg. The DO11.10 T cell receptor is specific for ova peptide 323–339/MHC class II I-Ad complex. Within 2 weeks after gene transfer, CD4+CD25+GITR+ cells emerged in the thymus and in secondary lymphoid organs. Frequency of these cells increased to 2–4% by 2 months and subsequently remained at that level. These cells also expressed CTLA-4 and FoxP3 (>100-fold increase in FoxP3 message compared to CD4+ cells from naive mice or compared to CD4+CD25- cells of AAV-ova transduced mice), and efficiently suppressed CD4+CD25- cells in vitro. In summary, hepatic AAV gene transfer induces transgene product-specific CD4+CD25+ Treg, which suppress antibody formation to the transgene product and are required for tolerance induction. These results should have broad implications for in vivo gene transfer.

Disclosures: RW Herzog has received royalties from Genzyme Corp. for AAV-F.IX technology.

Author notes

*

Corresponding author