Platelet gene therapy corrects the hemophilic phenotype in immunocompromised hemophilia A mice transplanted with genetically manipulated human cord blood stem cells

Abstract

Our previous studies have demonstrated that platelet FVIII (2bF8) gene therapy can improve hemostasis in hemophilia A mice, even in the presence of inhibitory antibodies, but none of our studies has targeted human cells. Here, we evaluated the feasibility for lentivirus (LV)-mediated human platelet gene therapy of hemophilia A. Human platelet FVIII expression was introduced by 2bF8LV-mediated transduction of human cord blood (hCB) CD34(+) cells followed by xenotransplantation into immunocompromised NSG mice or NSG mice in an FVIII(null) background (NSGF8KO). Platelet FVIII was detected in all recipients that received 2bF8LV-transduced hCB cells as long as human platelet chimerism persisted. All NSGF8KO recipients (n = 7) that received 2bF8LV-transduced hCB cells survived tail clipping if animals had greater than 2% of platelets derived from 2bF8LV-transduced hCB cells, whereas 5 of 7 survived when human platelets were 0.3% to 2%. Whole blood clotting time analysis confirmed that hemostasis was improved in NSGF8KO mice that received 2bF8LV-transduced hCB cells. We demonstrate, for the first time, the feasibility of 2bF8LV gene delivery to human hematopoietic stem cells to introduce FVIII expression in human platelets and that human platelet-derived FVIII can improve hemostasis in hemophilia A.

Figure 1

Human cell chimerism in NSG and NSGF8KO mice that received 2bF8LV-transduced hCB CD34⁺ cells. Blood samples were collected from tail bleeds. Human platelets were identified by a mouse anti-human glycoprotein Ib alpha chain (GPIbα) antibody AP1. Mouse platelets were stained with a rat anti-mouse CD41 (GPIIb) antibody. Human leukocytes were identified by mouse anti-human CD45 antibody, and mouse leukocytes were identified by rat anti-mouse CD45 antibody. (A) Representative flow cytometric analysis of platelets in peripheral blood of NSG and NSGF8KO recipients at 3 weeks after transplantation. (B) Representative flow cytometric analysis of leukocytes in peripheral blood of NSG and NSGF8KO recipients between 5 and 7 weeks after transplantation. (C) Human platelet chimerism in NSG and NSGF8KO recipients. Samples from NSG mice and normal human individuals were used as controls. Data are summarized from 10 trials of xenotransplants. (D) Human leukocyte chimerism in NSG and NSGF8KO recipients. Samples from NSG mice and normal human individuals were used as controls. Data are summarized from 10 trials of xenotransplants. (E) Time course of human platelet production in NSG recipients. Data are summarized from 2 trials of xenotransplants (n = 4 to 9 mice for each group). (F) Time course of human leukocyte production in NSG recipients. Data are summarized from 2 trials of xenotransplants (n = 4 to 7 mice for each group). w, weeks.

Figure 2

Genetic analysis of 2bF8LV-transduced hCB transplantation recipients. (A) PCR analysis of the 2bF8 transgene. Genomic DNA was purified from peripheral leukocytes. Leukocytes from NSG mice and a normal human individual were used as controls. 2bF8 transgenic mouse DNA was used as a positive control for the 2bF8 transgene. Shown is one representative experiment that was performed 3 times. Wild-type (WT) mouse FVIII and human glyceraldehyde 3-phosphate dehydrogenase (hGAPDH) were used as internal controls. (B) qPCR quantified the average copy number of the 2bF8 transgene per cell in 2bF8LV-transduced hCB transplantation recipients. Peripheral leukocyte–derived genomic DNA was used to analyze 2bF8 proviral DNA with normalization to hGAPDH. Bars represent mean ± SD. For individual mice analyzed more than once over the course of study, the average copy number was calculated. Genomic DNA from NSG mice was used as a control. Data are summarized from 2 trials of xenotransplants.

Figure 3

Analysis of platelet FVIII expression in 2bF8LV-transduced xenotransplanted recipients. Platelets were lysed in 0.5% 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS). The levels of platelet-derived FVIII expression were determined by chromogenic FVIII activity (FVIII:C) assay of platelet lysates from recipients at least 3 weeks after transplantation. Bars represent mean ± SD. The average FVIII level was calculated for individual mice analyzed more than once over the course of study. Platelets from NSG mice, C57BL/6 mice, and normal human individuals were used as controls. (A) Quantitative evaluation of FVIII activity levels in platelet lysates from 2bF8LV-transduced hCBT NSG or NSGF8KO recipients. (B) Quantitative evaluation of FVIII activity levels in hCB-derived human platelets based on human platelet numbers as determined by flow cytometry. Data are summarized from 10 trials of xenotransplants. NA, not applicable.

Figure 4

Electron microscopy determines cellular location of 2bF8 transgene protein. Isolated platelets from (A-C) 2bF8LV-transduced xenotransplanted NSG recipients, (D-F) normal human individuals, and (G-I) untransplanted NSG control mice were immunostained for hFVIII and either endogenous human or mouse VWF. hFVIII was stained with mouse anti-human FVIII monoclonal antibody 103.3 and probed with goat anti-mouse colloidal gold probe (6 nm; some indicated by arrows). VWF was stained with rabbit anti-human VWF polyclonal antibody (Dako), which cross-reacts with mouse VWF, and were probed with goat anti-rabbit colloidal gold probe (10 nm; some indicated by arrowheads). The results show that hFVIII is stored together with human VWF in platelet alpha-granules of 2bF8LV-transduced human platelets (A-C).

Figure 5

Phenotypic correction assessment. (A) Tail clip survival test assessing the phenotypic correction in 2bF8LV-transduced humanized NSGF8KO mice. The tail clipping test was performed at least 4 weeks after transplantation. Mice surviving beyond 24 hours were considered to have achieved phenotypic correction. Untransduced hCB-transplanted and untransplanted NSGF8KO recipients were tested as controls. Data are summarized from 10 trials of xenotransplants. (B) ROTEM analysis. The ROTEM analysis was performed between 6 and 8 weeks after transplantation when human platelet chimerism was 0.4% to 5.5%. Representative TEMograms (thromboelastometry graphs) from untransplanted NSG, untransplanted NSGF8KO, and 2bF8LV-transduced humanized NSGF8KO (with 2% human platelet chimerism) mice are shown. (C) Whole blood CT determined by ROTEM analysis. Untransplanted NSG and NSGF8KO mice were used as controls. Data are summarized from 5 trials of xenotransplants. N/A, not applicable.