In vivo HSC gene therapy enables sustained eCD4-Ig expression for SIV prevention

Figures

• 

  Figure 1. In vitro characterization of HDAd-eCD4Ig-Emm06 (A) Left: Schematic of HDAd structure. HDAd6/3+ vectors are derived from adenovirus serotype 6. These vectors incorporate the Ad6 fiber shaft and the DSG2 affinity-enhanced Ad3 fiber knob for liver detargeting and HSC targeting, respectively. (B–D) In vitro transduction of rhesus CD34+ cells (n = 3 donors) with HDAd6/35++ and HDAd6/3+ vectors expressing GFP. Infection was performed at an MOI of 4,000 viral particles (vp) per cell. (B) GFP expression measured in total CD34+ cells or in CD34+/CD45RA-/CD90+ cells at day 3 post-transduction. (C) Cell viability assessed by Annexin V and 7-AAD staining 24 h post-transduction using flow cytometry. (D) Reactive oxygen species (ROS) median fluorescence intensity (MFI) measured 24 h post-transduction. (E) HDAd vector structure. HDAd-eCD4Ig-Emm06: eCD4Ig-Emm06 contains the rhesus CD4 leader sequence, rhesus CD4 domains 1 and 2, the rhesus IgG2 hinge and Fc regions, and the CCR5 mimetic peptide (CCR5mim). The eCD4Ig-Emm06 construct is under the control of the ubiquitously active EF1α promoter. Additional functional cassettes are included for the expression of i) rhesus mgmtP140K (to confer…

• 

  Figure 2. In vivo safety and efficacy studies in CD46-transgenic mice with HDAd-eCD4Ig-Emm06 Because DSG-transgenic mice do not adequately express the HDAd6/3+ receptor, DSG2, the eCD4Ig-Emm-6 viral genome was packaged into HDAd5/35++ capsids and tested in transgenic mice that express the HDAd6/35 receptor, CD46. (A) Schematic of the experiment. CD46-transgenic mice were mobilized with G-CSF/AMD3100 and, at the peak of mobilization, intravenously injected with HDAd5/35-eCD4Ig-Emm06 and HDAd5/35-SB at a 1:1 ratio (total dose: 1.6 × 1010 vp/kg). Mice were then subjected to four cycles of in vivo selection with O6BG and increasing doses of BCNU (5, 7.5, 10, 10 mg/kg). At week 16, in vivo transduced mice were sacrificed, and bone marrow lineage-negative (Lin−) cells were transplanted into lethally irradiated C57Bl/6 mice, which were followed for another 16 weeks (secondary recipients). IS, immunosuppression. (B) Serum eCD4Ig-Emm06 levels in in vivo-transduced mice. Each symbol represents an individual mouse. (C) Vector copy number (VCN) per cell in tissues of primary mice at week 16 after HDAd transduction. (D) Serum eCD4Ig-Emm06 levels measured in secondary mice, with each symbol representing…

• 

  Figure 3. In vivo HSC transduction study in rhesus macaques (A) Schematic of the experiment created in https://BioRender.com. Animals were mobilized with G-CSF/AMD3100 and intravenously injected with either HDAd-eCD4Ig-Emm06 + HDAd-SB or HDAd-control + HDAd-SB. HSC mobilization involved two waves, with HDAd vectors intravenously injected at peak mobilization (8 h after AMD3100). The time of HAD injection is indicated as “0 h,” correspondingly at day −1 and day 0. Animals were pretreated with dexamethasone, tocilizumab, and anakinra. At weeks 4 and 8, O6BG/BCNU was administered for selection of modified cells. The first cycle of selection consisted of O6BG 120 mg/m2 and BCNU 10 mg/m2; the second cycle used the same O6BG dose, and BCNU was increased to 20 mg/m2. After at least 12 weeks from the final selection cycle, animals were challenged intravenously with SIVmac239 every 4 weeks using escalating doses until infection was established. (B–D) HDAd-eCDIg-Emm06 injected animals. Shown are the serum concentrations of eCD4Ig-Emm06 (red lines) and anti-eCD4Ig-Emm06 antibodies (anti-drug antibodies, ADA) (black line), measured in duplicate with SEM error bars. Treatment with immunosuppressants is…

• 

  Figure 4. In vitro inhibition of SIV by serum eCD4Ig-Emm06 SIVmac239 was incubated in duplicate with TZM-bl cells and serial dilutions of recombinant eCD4-Ig or serum. Luciferase activity, as an indicator of SIV infection, was measured 2 days post-infection. IC50 was calculated as the eCD4Ig-Emm06 concentration or serum dilution required to reduce the maximum relative light units (RLU) by 50%. Asterisks (∗) indicate sera collected post-SIV infection. Schematic was created in https://BioRender.com.

• 

  Figure 5. Tissue VCN Rhesus macaques treated with HDAd-eCD4Ig-Emm06 (n = 3, red) or HDAd-control (n = 3, gray) were infected with SIVmac239, and systemic tissues were analyzed at necropsy for VCN, measured by quantifying the MGMT transgene sequence. LN, lymph node; Mes, mesenteric; SM, submandibular. No statistically significant differences were observed between groups by two-way ANOVA with Šídák’s multiple comparisons correction.

• 

  Figure 6. eCD4Ig-Emm06 expression from peripheral blood and tissue immune cells (A) Intracellular flow cytometry gating strategy to detect eCD4Ig-Emm06 expression from T cells, B cells, and monocytes from NHPs treated with HDAd-eCD4Ig-Emm06 or HDAd-control. Immune cell subsets were gated as previously described.23 Since eCD4Ig-Emm06 lacks a specific detection antibody, surrogate marker combinations were used. For CD3+CD20- T cells, eCD4Ig-Emm06 expression was defined as cells co-expressing CD4 and IgG2, leveraging the fact that T cells naturally express CD4 but do not endogenously express IgG2. For CD3−CD20+ B cells, a subset of B cells naturally express IgG2, and CD4 can also be expressed via trogocytosis,35 so intracellular eCD4Ig-Emm06 was identified as cells that express the D1 domain but not the D3 domain of CD4, reflecting the D1 and D2-only structure of eCD4Ig-Emm06. CD11b+CD14+ monocytes can express CD4, and autofluorescence limits IgG2-based detection; therefore, eCD4Ig-Emm06 expression was identified by the presence of CD4 D1 and absence of CD4 D3. (B) eCD4Ig-Emm06 expression in T cell, B cell, and monocytes from peripheral blood and tissues at necropsy. Samples with <2,000…

• 

  Figure 7. Viral load and survival curves Indian-origin rhesus macaques received either HDAd-eCD4Ig-Emm06 (n = 3, red) or HDAd-control (n = 3, gray) and were intravenously challenged every 4 weeks with escalating doses of SIVmac239 until infection was established. (A) Survival curve indicating the SIVmac239 dose at which infection occurred, as well as the corresponding animal ID and serum eCD4Ig-Emm06 level at the time of infection. No statistically significant differences were observed by the log-rank and Gehan-Breslow-Wilcoxon tests. (B) SIVmac239 barcodes measured during peak viremia to quantify founder viruses responsible for infection, with each barcode representing a distinct infectious event. (C) Longitudinal viral loads from the time of infection. (D) Peak and (E) setpoint viral loads compared between the HDAd-eCD4Ig-Emm06 and HDAd-control groups, with setpoint viral load defined as the average viral load after 5 weeks of infection. No statistically significant differences were observed by the unpaired t test. (F) Anti-SIV IgG responses measured by ELISA using an anti-gp41 format to avoid cross-reactivity with eCD4Ig-Emm06, which binds gp120.

• 

  Figure 8. Tissue SIV viral loads Rhesus macaques were treated with HDAd-eCD4Ig-Emm06 (n = 3, red) or HDAd-control (n = 3, gray) and infected with SIVmac239. Systemic tissues were analyzed at necropsy for SIV viral loads, normalized to total RNA. HDAd-eCD4Ig-Emm06-treated animals showed significantly reduced SIV viral loads, as determined by two-way ANOVA with Šídák’s multiple comparisons correction (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001). LN, lymph node; Mes, mesenteric; SM, submandibular.