David J. Rawlings

CD19+CD24hiCD38hi B Cells Exhibit Regulatory Capacity in Healthy Individuals but Are Functionally Impaired in Systemic Lupus Erythematosus Patients

The immunosuppressive function of regulatory B cells has been shown in several murine models of chronic inflammation, including collagen-induced arthritis, inflammatory bowel disease, and experimental autoimmune encephalomyelitis. Despite interest in these cells, their relevance to the maintenance of peripheral tolerance in humans remains elusive. Here, we demonstrate that human CD19(+)CD24(hi)CD38(hi) B cells possessed regulatory capacity. After CD40 stimulation, CD19(+)CD24(hi)CD38(hi) B cells suppressed the differentiation of T helper 1 cells, partially via the provision of interleukin-10 (IL-10), but not transforming growth factor-beta (TGF-beta), and their suppressive capacity was reversed by the addition of CD80 and CD86 mAbs. In addition, CD19(+)CD24(hi)CD38(hi) SLE B cells isolated from the peripheral blood of systemic lupus erythematosus (SLE) patients were refractory to further CD40 stimulation, produced less IL-10, and lacked the suppressive capacity of their healthy counterparts. Altered cellular function within this compartment may impact effector immune responses in SLE and other autoimmune disorders.

Anti-CD3 antibodies modulate anti-factor VIII immune responses in hemophilia A mice after factor VIII plasmid-mediated gene therapy

One major obstacle in gene therapy is the generation of immune responses directed against transgene product. Five consecutive anti-CD3 treatments concomitant with factor VIII (FVIII) plasmid injection prevented the formation of inhibitory antibodies against FVIII and achieved persistent, therapeutic levels of FVIII gene expression in treated hemophilia A mice. Repeated plasmid gene transfer is applicable in tolerized mice without eliciting immune responses. Anti-CD3 treatment significantly depleted both CD4+ and CD8+ T cells, whereas increased transforming growth factor-beta levels in plasma and the frequency of both CD4+CD25+FoxP3+ and CD4+CD25-Foxp3+ regulatory T cells in the initial few weeks after treatment. Although prior depletion of CD4+CD25+ cells did not abrogate tolerance induction, adoptive transfer of CD4+ cells from tolerized mice at 6 weeks after treatment protected recipient mice from anti-FVIII immune responses. Anti-CD3-treated mice mounted immune responses against both T-dependent and T-independent neo-antigens, indicating that anti-CD3 did not hamper the immune systems in the long term. Concomitant FVIII plasmid + anti-CD3 treatment induced long-term tolerance specific to FVIII via a mechanism involving the increase in transforming growth factor-beta levels and the generation of adaptive FVIII-specific CD4+Foxp3+ regulatory T cells at the periphery. Furthermore, anti-CD3 can reduce the titers of preexisting anti-FVIII inhibitory antibodies in hemophilia A mice.

Novel fluorescent genome editing reporters for monitoring DNA repair pathway utilization at endonuclease-induced breaks

The creation of a DNA break at a specific locus by a designer endonuclease can be harnessed to edit a genome. However, DNA breaks may engage one of several competing repair pathways that lead to distinct types of genomic alterations. Therefore, understanding the contribution of different repair pathways following the introduction of a targeted DNA break is essential to further advance the safety and efficiency of nuclease-induced genome modification. To gain insight into the role of different DNA repair pathways in resolving nuclease-induced DNA breaks into genome editing outcomes, we previously developed a fluorescent-based reporter system, designated the Traffic Light Reporter, which provides a readout of gene targeting and gene disruption downstream of a targeted DNA double-strand break. Here we describe two related but novel reporters that extend this technology: one that allows monitoring of the transcriptional activity at the reporter locus, and thus can be applied to interrogate break resolution at active and repressed loci; and a second that reads out single-strand annealing in addition to gene targeting and gene disruption. Application of these reporters to assess repair pathway usage in several common gene editing contexts confirms the importance that chromatin status and initiation of end resection have on the resolution of nuclease-induced breaks.

FcγRIIb and BAFF Differentially Regulate Peritoneal B1 Cell Survival

B1 cells produce most natural Abs in unimmunized mice and play a key role in the response to thymus-independent Ags and microbial infection. Enlargement of B1 cell number in mice is often associated with autoimmunity. However, the factors that control peripheral B1 cell survival remain poorly characterized. Mice lacking the inhibitory receptor FcγRIIb exhibit a massive expansion in peritoneal B1 cells, implicating this receptor in B1 cell homeostasis. In this study, we show that peritoneal B1 cells express the highest levels of FcγRIIb among B cell subsets and are highly susceptible to FcγRIIb-mediated apoptosis. B1 cells upregulate FcγRIIb in response to innate signals, including CpG, and the B cell homeostatic cytokine BAFF efficiently protects activated B1 cells from FcγRIIb-mediated apoptosis via receptor downregulation. BAFF-transgenic mice manifest an expansion of peritoneal B1 cells that express lower levels of FcγRIIb and exhibit reduced susceptibility to apoptosis. Whereas both peritoneal B1 cells from wild-type and BAFF-transgenic mice immunized with CpG exhibit an increase in FcγRIIb levels, this change is blunted in BAFF-transgenic animals. Our combined results demonstrate that FcγRIIb controls peritoneal B1 cell survival and this program can be modulated by the BAFF signaling axis.

284. Long-Term Therapeutic Immune Reconstitution in XSCID Canine Model via In Vivo Foamy Virus Delivery of Common Gamma Chain

X-linked combined immunodeficiency disease (XSCID) is caused by mutation in the common gamma chain, γC (interleukin-2 receptor subunit gamma, IL2RG) in both humans and canines. It is characterized by the inability of T-cell development leading to absence of T-cells in peripheral blood, lack of T-cell mediated immune response, low IgA and IgG levels, and early infant mortality. In the 1990s, human XSCID clinical trials utilizing gamma-retroviral vectors to deliver the IL2RG gene caused leukemia in 5 out of 20 patients due to vector integration in or near proto-oncogenes. Recent studies showed Foamy virus based vectors as an excellent alternative for in vivo gene-therapy because it is non-pathogenic in humans while exhibiting increased serum stability and favorable integration pattern. Previously, we have demonstrated CD3+ T-cell reconstitution in the canine model via intravenous injection of foamy virus expressing human elongation factor-1 alpha promoter (Ef1α)-yC. Unfortunately, the treated animals contained a low number of gene corrected progenitors at a sub-therapeutic level. Here, we achieved long-term therapeutic immune-reconstitution by intravenous delivery of a human phosphoglycerate kinase promoter (Pgk)-mediated γC foamy viral vector into XSCID neonatal canines. Long-term (2 years) post-injection follow-up demonstrated therapeutic levels of CD3+ T-cell expansion. Within the T-cell population, gene correction with Pgk-γC stabilized at ~80%. We validated T-cell functionality by using spectratyping analysis, which exhibited a diverse repertoire of receptor gene rearrangement. Retroviral integration site analysis (RIS) indicated polyclonal contribution to the reconstituted T-cells. Immunoglobulin ELISA assays showed that IgA and IgG levels in peripheral blood are comparable to normal healthy controls. We immunized the gene-corrected canine recipients with bacteriophage ϕx174 and confirmed production of specific IgG antibodies, showing the ability for isotype switching in B-lymphocytes. Currently, the gene-corrected canines exhibit comparable health and physical attributes to normal controls. Furthermore, semen from the gene-corrected male canine was used via artificial insemination to produce a litter of viable offsprings. In summary, our data demonstrate that Pgk-γC foamy viral vector delivered long-term therapeutic gene correction in a large-animal model for XSCID gene therapy. Most importantly, these results indicate that in vivo Pgk-γC foamy vector administration is a viable option for long-term immune reconstitution in future XSCID human clinical trials.

569. Precision Editing of the WAS Locus via Homologous Recombination in Primary Human Hematopoietic Cells Mediated by Either TALEN or CRISPR/Cas Nucleases

Wiskott-Aldrich Syndrome (WAS) is an inherited primary immunodeficiency caused by mutations in the WAS gene which encodes a protein (WASp) that regulates the actin cytoskeleton in multiple hematopoietic cell lineages. Currently, allogeneic stem cell transplantation constitutes the only available cure for WAS, although phase I/II clinical trials using lentiviral gene therapy of autologous stem cells are currently underway and have shown improvement in immune system defects. Although effective, the gene therapy approach carries a risk of insertional mutagenesis, and unpredictable expression due to promoter choice as well as influences of the random integration site, including epigenetic status and the presence of neighboring transcription regulatory elements. As a refinement, our goal is to develop gene editing methodologies that would allow the specific targeting of a WASp cDNA to a position within the WAS locus allowing transcriptional regulation by the endogenous WAS promoter. To this end, we first designed and tested the cleavage efficiency of several guide RNAs (delivered as self-complementary AAV along with Cas9, delivered as mRNA), as well as candidate TALEN pairs (delivered as mRNAs), in primary human T cells. Using the T7 endonuclease assay, we identified candidate nucleases from both platforms (CRISPR/Cas9 and TALEN) that achieved a high Indel frequency: 73 and 85%, respectively. We next created a synthetic AAV6 donor template for homology-directed repair (HDR) that contained an MND promoter driven GFP cDNA with 1kb of WAS homology arms flanking it. When delivered with the TALEN or CRISPR/Cas9 nucleases, we observed stable integration of the GFP reporter within >25% of primary human T cells. Subsequently, we utilzed the identical reagents to target integration of the reporter cDNA into the WAS locus in adult human mobilized CD34+ cells, albeit at lower efficiencies. The off-target cleavage sites for TALENs identified using the Prognos software were amplified and sequenced, with no evidence of off-target cleavage observed at any of the predicted loci. Thus, we have generated genome editing tools that possess a high degree of specificity, providing the foundation for site-specific modification of the WAS locus as a therapeutic option.

44. Novel Combination of megaTAL Nuclease-Driven Genome Engineering with a Drug Selection Cassette Increases Efficiency of HIV Gene Therapy

Human Immunodeficiency Virus (HIV) infection remains a substantial health problem worldwide. The C-C chemokine receptor 5 (CCR5) serves as a co-receptor for HIV entry into CD4+ T cells and represent an alternative therapeutic target. Early clinical trials using CCR5-targeting zinc finger nucleases demonstrated transient control of HIV infection in the course of antiretroviral treatment interruption (Tebas, NEJM, 2014). Our current work improves on these advances by combining high level of CCR5 gene disruption with preferential selection of gene modified cells. The CCR5-targeting megaTAL combines a LAGLIDADG homing endonuclease scaffold with an eleven repeat transcription activator-like (TAL) effector array to achieve site specific DNA cleavage. This nuclease produces highly efficient CCR5 targeting in primary human CD4+ T cells in vitro (70-90% disruption). To test the protective effects of megaTAL treatment, primary human CD4+ T cells treated with CCR5-megaTAL were transplanted into NOD/SCID/γc-null (NSG) ‘humanized’ mice and challenged with HIV-1. We observed a 100-fold increase of megaTAL-treated cells compared to untreated controls during an active in vivo infection demonstrating the functionality of this approach. Based on the decline of CCR5 modified cells in the clinical trials to date, we hypothesized that we could improve maintenance of HIV resistant cells by expanding them either ex vivo or in vivo. We propose that coupling megaTAL nuclease treatment with drug selection will help us achieve therapeutically relevant levels of HIV-protected cells by enabling efficient selection only of CCR5-modified T-cells. The mutant human dihydrofolate reductase (mDHFR) chemoselection system has been used to render cells resistant to lymphotoxic concentrations of the drug methotrexate (MTX). We tested our experimental approach by transducing cells with lentiviral vector encoding a mDHFR cassette followed by chemoselection in MTX at 0.02uM. This approach resulted in a six fold enrichment of gene modified primary CD4+ T cells ex vivo. Previously we have shown that combining megaTAL treatment with adeno-associated virus (AAV) transduction produces very high rates of homology-driven repair (HDR) in primary human T cells. Hence, we combined megaTAL/AAV treatment to integrate the mutant DHFR into the CCR5 locus, producing a population of MTX-resistant CD4+ cells that also lack CCR5. Primary human CD4+ T cells were transfected with CCR5-megaTAL mRNA and transduced with AAV6 containing a mutant DHFR donor template flanked by 0.6kb CCR5 homology arms. They demonstrated a greater than five-fold enrichment in MTX compared to untreated controls ex vivo. Next, we have transplanted NSG mice with 1×106 gene modified cells/mouse to assess the therapeutic potential of our approach. Mice that engraft effectively will be treated with daily injections of 0, 0.5 and 2 mg/kg of MTX to monitor preferential selection and enrichment of our target cell population. Subsequent studies will assess the long term control of viremia in these mice following HIV challenge. In conclusion, the CCR5-megaTAL nuclease platform produces very high levels of gene-modified CD4+ T-cells and protects these cells from subsequent HIV infection in vivo. Furthermore, combining targeted integration and chemical selection results in the specific selection of gene modified primary human T cells. To our knowledge we are the first group to report MTX-mediated chemoselection and expansion of CD4+ T cells following targeted integration at the CCR5 locus.

761. Targeted Killing of HIV Infected Cells Using CCR5-Disrupted Anti-HIV-CAR T Cells

A cure for HIV remains an important treatment goal for 30 million HIV-infected individuals worldwide. Long term control of HIV following a single treatment will require a mechanism to eradicate the latent reservoir of HIV infected cells that are capable of reactivating. A previous phase II clinical trial using an anti-HIV chimeric antigen receptor (αHIV-CAR) targeting the CD4-binding site on HIV envelope was partially effective. To optimize this approach we have developed a series of CARs based on the scFV of broadly neutralizing HIV antibodies targeting four different structural regions of the HIV envelope: the V1/V2 loop, the V3 loop, the CD4 binding site, and the membrane-proximal external region. αHIV-CAR T cells targeting different epitopes were compared and were able to kill > 80% of HIV-infected cells grown in the presence of ART. However, a limitation of αHIV-CAR T cells is that the αHIV-CAR also serves as a receptor for HIV and allows HIV infection of αHIV-CAR T cells. Therefore we disrupted the major co-receptor required for HIV cell entry, CCR5, using a megaTAL nuclease, as a means of protecting the CAR-expressing cells from HIV infection. We used two strategies for achieving these dual modifications in primary human T cells. For both strategies, the CCR5 megaTAL nuclease was delivered by mRNA electroporation, which has previously been shown to induce a high rate of bi-allelic NHEJ-mediated gene disruption. αHIV-CAR expression cassettes were delivered into the host genome via lentiviral vectors (LV), or were targeted to the megaTAL nuclease cleavage site in CCR5 using an adeno-associated virus (AAV) that included CCR5 homology arms. Both strategies resulted in stable expression of the CAR construct and specific activation of CAR+ T cells in the presence of an HIV+ cell line. In the presence of actively replicating virus, CCR5-megaTAL treated CAR+ T cells out-performed CAR+ T cells generated by LV delivery alone as measured by reduction in HIV capsid protein. To enable testing in non-human primates (NHP), we re-optimized the cell-editing protocol with NHP lymphocytes. Primary T cells from pigtail macaques were successfully transfected with CCR5 megaTAL mRNA and achieved a CCR5 disruption rate of 50%. Cells were subsequently transduced with αHIV-CAR LV resulting in ~70% of manipulated cells with αHIV-CAR. Expansion of the cells in vitro resulted in 60-fold expansion over 8 days. CAR+ NHP cells specifically killed HIV infected human cells demonstrating that NHP-derived αHIV-CAR+ T cells retained killing function. In conclusion, it is feasible to construct αHIV-CAR+ T cells that are protected from HIV infection in human and NHP cells, and warrants further study in vivo.

532. Foamy Viral Vector Integration Sites in SCID-Repopulating Cells After MGMTP140K-Mediated In Vivo Selection

Foamy virus (FV) vectors are promising for hematopoietic stem cell (HSC) gene therapy but preclinical data on the clonal composition of FV vector transduced human repopulating cells is needed. Human CD34+ human cord blood cells were transduced with an FV vector encoding a methylguanine methyltransferase (MGMT) P140K transgene, transplanted into immunodeficient NOD/SCID IL2Rgnull (NSG) mice, and selected in vivo for gene-modified cells. The retroviral insertion site (RIS) profile of repopulating clones was examined using modified genomic sequencing PCR (MGS-PCR). We observed polyclonal repopulation with no evidence of clonal dominance even with the use of a strong internal spleen focus forming virus (SFFV) promoter known to be genotoxic. However we did observe that highly captured sites were found more often near proto-oncogenes than less frequently captured sites (Figure 1Figure 1). Our data supports the use of FV vectors with MGMTP140K for HSC gene therapy, but also suggests additional safety features should be developed and evaluated.Figure 1(a) Proximity of RIS to TSS of proto-oncogenes in Network of Cancer Genes 4.0 (NCG 4.0) cancer gene list (n = 2048). (b) Proportion of RIS within NCG 4.0 proto-oncogene transcripts or within 50 kb of a proto-oncogene TSS. In vivo low and in vivo high represent the lowest and highest thirds of RIS ranked by recovery frequency. Abbreviations: ns, not significant; RIS, retroviral insertion site; TSS, transcription start site.View Large Image | Download PowerPoint Slide