Restoration of FVIII Function and Phenotypic Rescue in Hemophilia A Mice by Transplantation of MSCs Derived From F8-Modified iPSCs

Hemophilia A (HA), an X-linked recessive congenital bleeding disorder, affects 80%–85% of patients with hemophilia. Nearly half of severe cases of hemophilia are caused by a 0.6-Mb genomic inversion (Inv22) that disrupts F8. Although viral-based gene therapy has shown therapeutic effects for hemophilia B (HB), this promising approach is not applicable for HA at the present stage; this limitation is mainly due to the large size of F8 cDNA, which far exceeds the adeno-associated virus (AAV) packaging capacity. We previously reported an in situ genetic correction of Inv22 in HA patient-specific induced pluripotent stem cells (HA-iPSCs) by using TALENs. We also investigated an alternative strategy for targeted gene addition, in which cDNA of the B-domain deleted F8 (BDDF8) was targeted at the rDNA locus of HA-iPSCs using TALENickases to restore FVIII function. Mesenchymal stem cells (MSCs) have low immunogenicity and can secrete FVIII under physiological conditions; in this study, MSCs were differentiated from F8-corrected iPSCs, BDDF8-iPSCs, and HA-iPSCs. Differentiated MSCs were characterized, and FVIII expression efficacy in MSCs was verified in vitro. The three types of MSCs were introduced into HA mice via intravenous injection. Long-term engraftment with restoration of FVIII function and phenotypic rescue was observed in HA mice transplanted with F8-corrected iMSCs and BDDF8-iMSCs. Our findings suggest that ex vivo gene therapy using iMSCs derived from F8-modified iPSCs can be feasible, effective, and promising for the clinical translation of therapeutic gene editing of HA and other genetic birth defects, particularly those that involve large sequence variants.


INTRODUCTION
Hemophilia A (HA) is an X-linked recessive congenital bleeding disorder caused by mutations in the factor VIII (FVIII) gene (F8) that leads to deficient blood coagulation. Spontaneous or traumatic bleeding in daily life or during surgery is the main symptom in HA patients, and the incidence is one in 5,000 male births (Nienhuis et al., 2017). Severe HA patients have only 1% or less of normal plasma FVIII activity. Currently, the main clinical treatment of HA is preventive or on-demand intravenous injections of FVIII concentrates from normal human plasma or recombinant FVIII protein. However, due to the short half-life of FVIII, patients require repeated injections, which increase the risks of infection and induce FVIII inhibitors (Berntorp and Shapiro, 2012;Weyand and Pipe, 2019).
Hemophilia is a monogenic disease considered suitable for gene therapy because a slight increase in plasma coagulation factor activity can lessen the bleeding phenotype (Graw et al., 2005;Perrin et al., 2019). Gene therapy of hemophilia based on viral vectors has been extensively investigated. Adeno-associated virus (AAV)-mediated gene therapy of hemophilia B (HB) produced promising progress in HB animal models (Crudele et al., 2015) and clinical trials (George et al., 2017;Miesbach and Sawyer, 2018). However, certain challenges of AAV-based gene therapy for HA remain mainly because the size of the FVIII coding sequence is too large (7 kb) to be packed into the AAV vector. Therefore, a smaller B-domain-deleted human F8 (BDDF8) cDNA of 4.4 kb was incorporated in the AAV vector to obtain therapeutic effects in animal models (Sabatino et al., 2011;McIntosh et al., 2013) and in clinical trials (Rangarajan et al., 2017;Batty and Lillicrap, 2019;Pasi et al., 2020).
Cell-based gene therapy has been actively studied as a potential alternative treatment for HA. Transplanted cells can provide sustained production of FVIII in vivo. Since endothelial cells (ECs) (Kumaran et al., 2005;Follenzi et al., 2008) and mesenchymal stem cells (MSCs) (Follenzi et al., 2012;Sanada et al., 2013) can physiologically secrete functional FVIII protein, these two cell types are generally the preferred target cells for HA gene therapy. Several studies have demonstrated that transplanted liver sinusoidal endothelial cells (LSECs) or ECs can synthesize FVIII in the liver and correct the bleeding phenotype in HA mice (Follenzi et al., 2008;Shi et al., 2010;Yadav et al., 2012;Fomin et al., 2013). However, LSECs and ECs are not readily available and have poor proliferation ability. MSCs have low immunogenicity and strong renewal ability and are easy to separate and propagate in vitro. Various sources of MSCs were modified using viral vectors in vitro and then transplanted into HA mice to restore the FVIII function. Some studies have used bone marrow-derived MSCs to achieve phenotypic correction in mice (Follenzi et al., 2012). Other studies have focused on longterm therapeutic plasma levels of FVIII in HA mice by systemic delivery of MSCs derived from hFVIIIBD transgenic mice (Wang et al., 2013) or by transplantation of virus-modified MSCs in utero (Kumar et al., 2018).
Due to the self-renewal potential and multidirectional differentiation, human induced pluripotent stem cells (iPSCs) can be used as an applicable autologous cell source for ex vivo gene therapy for HA (Takahashi and Yamanaka, 2006;Wong and Chiu, 2011). Non-viral genome-editing approaches in iPSCs include actively developing in situ gene correction at the F8 locus (Park et al., 2015;Wu et al., 2016; and ectopic integration of BDDF8 at the safe harbor loci, such as the ribosomal DNA (rDNA) locus  and the H11 locus (Park et al., 2019). Differentiation of iPSCs into MSCs can be an efficient means to generate many batches of MSCs. On the one hand, iPSC-derived MSCs and bone marrowderived MSCs have similar cell characteristics and differentiation potential (Diederichs and Tuan, 2014). On the other hand, iPSCs can be genetically modified and then continuously differentiated into MSCs to maintain long-term therapeutic gene expression for autologous gene therapy.
In our previous studies, cells from urine of a patient with severe HA with F8 intron 22 inversion (Inv22) were collected, reprogrammed into HA patient-specific iPSCs (HA-iPSCs), and used to generate F8-corrected iPSCs via in situ gene correction using TALENs ; BDDF8-iPSCs were generated by targeted integration of BDDF8 at the rDNA locus using TALENickases . In the present study, F8-corrected iPSCs, BDDF8-iPSCs, and HA-iPSCs were differentiated into iMSCs (iPSCs-derived MSCs), which were characterized and transplanted into HA mice by intravenous injection. Functional restoration of FVIII and phenotypic rescue were achieved in HA mice transplanted with iMSCs derived from F8-corrected iPSCs and BDDF8-iPSCs. Our data demonstrate that ex vivo therapeutic gene editing via F8-modified iPSCderived iMSCs is feasible, effective, and promising for the clinical translation of HA gene therapy.

Derivation of iMSCs From iPSCs
A STEMdiff TM mesenchymal progenitor kit (STEMCELL Technologies) was used to differentiate normal human iPSCs (hiPSCs), HA-iPSCs, 17-9-iPSCs, and T-7-iPSCs into hiMSCs, HA-iMSCs, 17-9-iMSCs, and T-7-iMSCs, respectively. According to the manufacturer's protocol, iPSCs were seeded on a Matrigelcoated 12-well plate at 5 × 10 4 cells/cm 2 in mTesR1 containing 10 µM Y27632. The cells were cultured in mTesR1 for 2 days and in STEMdiff TM -ACF mesenchymal induction medium for 4 days. The cells were then cultured in STEMdiff TM -ACF medium for additional 2 days and passaged on a 6-well plate precoated with STEMdiff TM -ACF attachment substrate. The cells were passaged every 3 days. After three passages, the cells were seeded onto a 0.1% gelatin-coated 10-cm dish in MSC medium. The medium was changed every day during differentiation.

Identification of Differentiation Potential of iMSCs
The differentiation potential of iMSCs was identified by osteogenesis (Gibco), adipogenesis (STEMCELL Technologies), and chondrogenesis (STEMCELL Technologies) differentiation kits according to the manufacturer's protocols.

RT-PCR and qRT-PCR
Total RNA was extracted using TRIzol reagent (Sigma-Aldrich, United States), and RNA was treated with gDNA wiper mix (Vazyme, China) to eliminate genomic DNA. Then, a 500 ng RNA sample was reverse transcribed using Hiscrip R II Q RT Supermix (Vazyme). Primers based on exons 19 and 23 were used to detect the transcript of F8. Primers based on exons 23 and 26 were used to detect relative F8 (exons 23-26) mRNA levels. The glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene was used as an endogenous control.

FVIII Assay of Culture Supernatants and Cell Lysates
The culture supernatants were harvested in triplicate from 6well plates 24 h after the medium was changed. The cells were dissociated with TrypLE TM select enzyme (Gibco) and counted. After washing with DPBS, cell pellets were resuspended in 500 µL of sample diluent for ELISA (Cedarlane, Burlington, ON, Canada) and lysed by three freeze-thaw cycles. ELISA was performed using Paired Antibodies for ELISA-factor VIII:C (Cedarlane) according to the manufacturer's instructions.

Western Blot
Cell lysates of iPSCs and iMSCs were prepared using RIPA lysis buffer (Beyotime, China) supplemented with 1 × phenylmethanesulfonyl fluoride (PMSF, 1 mM) and Protease Inhibitor Cocktail (Sigma-Aldrich). Protein in the samples was measured by a Pierce TM BCA protein assay kit (Thermo Fisher Scientific, United States). Protein samples (15 µg) were loaded on an electrophoresis gel that was run for 2 h at 120 V, and the proteins were transferred to PVDF membranes (Millipore, United States). After blocking with 5% non-fat milk in 0.1% TBST (TBS containing 0.1% Tween-20), the membrane was incubated overnight with anti-LMAN1 antibodies (Abcam, England) at 4 • C. The membrane was then incubated with HRP-labeled anti-rabbit IgG (Sigma-Aldrich) for 1 h at room temperature, and the signal was visualized using an ECL detection kit (Thermo Fisher Scientific). Then, the membrane was washed with stripping buffer and 0.1% TBST. After blocking, the membrane was incubated overnight with anti-β-actin antibodies (Sigma-Aldrich) at 4 • C. The membrane was incubated with anti-mouse IgG (Sigma-Aldrich) for 1 h at room temperature, and the signal was visualized using an ECL detection kit (Thermo Fisher Scientific).

Transplantation of iMSCs Into HA Mice
The use and care of animals complied with the guidelines of the Ethics Committee of the School of Life Sciences of Central South University.
HA mice (strain: B6; 129S4-F8 tm1Kaz/J; Jackson Laboratory) at 6-8 weeks of age were used for cell transplantation. C57BL/6 mice at 6-8 weeks of age were used as wild-type controls. Each mouse was anesthetized with Avertin (Sigma-Aldrich) and then infused with 2 × 10 6 CM-Dil-labeled iMSCs derived from HA-iPSCs, 17-9-iPSCs, T-7-iPSCs, or hiPSCs. Mouse plasma was collected for the assay of FVIII activity at four time points: 1, 2, 3, and 4 weeks after transplantation. HA mice without transplanted iMSCs served as a negative control. In brief, the mice were anesthetized, and then retro-orbital blood samples were collected to isolate plasma. FVIII activity was evaluated using calibration plasma for coagulation tests (Stago, France), and Compact Max (Stago, France) was used to determine activated partial thromboplastin time (aPTT). The main organs (heart, liver, spleen, lung, and kidney) from the treated and HA mice were harvested to detect CM-Dil-positive cells by immunofluorescence analysis.

Tail-Bleeding Assay
A tail-clip challenge was carried out 1 week after transplantation. Mice were anesthetized, and the tail-bleeding assays were performed as described previously (Yadav et al., 2012;Chen et al., 2017;Neumeyer et al., 2019). Briefly, the distal part of the mouse tail with a diameter of 1.5 mm was sheared. The tail was then immediately immersed in a 50 mL Falcon tube containing isotonic saline prewarmed in a water bath to 37 • C. Each mouse was monitored for 20 min to record the bleeding time. Firm pressure on the tail was applied for 1 min, and the survival rates for each group of mice were recorded in 2 days.

Statistical Analysis
GraphPad Prism 5.0 was used for data analysis. Data were analyzed using ANOVA for more than two groups. The survival curves were analyzed using the log-rank test. All values are presented as the mean ± SEM.
To evaluate the FVIII expression in iPSCs and iMSCs, the FVIII antigen in the cell lysate and supernatants of iPSCs and iMSCs was detected by ELISA. FVIII expression was detected in all iPSC cell lysates but was not detected in the supernatants, indicating that FVIII was not secreted at the iPSC stage ( Figure 2C). In the iMSC stage, ELISA results showed that the FVIII protein levels were 0.59 ± 0.11 ng/(10 6 at 24 h) and 0.68 ± 0.14 ng/(10 6 at 24 h) in the supernatant of 17-9-iMSCs and T-7-iMSCs, respectively, but were not detected in the samples of HA-iMSCs. Several studies have demonstrated that the complex formed by lectin mannosebinding 1 (LMAN1) and multiple coagulation factor deficiency 2 (MCFD2) serves as a cargo receptor for efficient transportation of FVIII from the endoplasmic reticulum to the Golgi (Zheng et al., 2010;Zhu et al., 2018). Detection of LMAN1 by western blot indicated that LMAN1 was expressed in HA-iMSCs, 17-9-iMSCs, T-7-iMSCs, and hiMSCs but was barely GAPDH was used as the internal reference. Bars represent the mean ± SEM (n = 3, each group); ***p < 0.001, and **p < 0.01 compared with HA-iPSCs, ns, not significant, and ***p < 0.001 compared with hiPSCs; ***p < 0.001 compared with HA-iMSCs; ns, not significant; and ***p < 0.001 compared with hiMSCs (one-way ANOVA). (C) FVIII antigen in iPSCs and iMSCs was detected by ELISA. Bars represent the mean ± SEM (n = 3, independent cultures). (D) Western blot analysis of LMAN1 protein in iPSCs and iMSCs. β-Actin was used as the internal reference.

Restoration of FVIII Function in HA Mice by Transplantation of F8-Modified iPSC-Derived iMSCs
To test whether iMSCs derived from F8-modified iPSCs restore FVIII function in vivo, we carried out animal studies in HA mice with a disrupted F8 gene. Initially, we labeled iMSCs using CellTracker TM CM-Dil, a fluorescent dye (Supplementary Figure 2A). Then, CM-Dil-labeled HA-iMSCs, 17-9-iMSCs, T-7-iMSCs, and hiMSCs were infused into HA mice via the orbital vein. We collected mouse plasma and assayed FVIII activity at four time points: 1, 2, 3, and 4 weeks after transplantation. The FVIII activity was recalculated as a percentage of the activity in wild-type mice. As shown in Figure 3A, 1 week after transplantation, the FVIII activity in HA mice treated with 17-9-iMSCs, T-7-iMSCs, and hiMSCs reached the highest levels of up to 10.32, 12.91, and 9.96%, respectively, which were significantly higher than that in HA mice (4.78%). From 2 to 3 weeks after transplantation, the FVIII activity in HA mice treated with 17-9-iMSCs, T-7-iMSCs, and hiMSCs was gradually decreased but remained significantly higher than that in HA mice. The changes in FVIII activity during the 4-week observation period are shown in Figure 3B. The data indicate that the FVIII activity in HA mice treated with 17-9-iMSCs, T-7-iMSCs, and hiMSCs was the lowest 4 weeks after transplantation, which was similar to that in HA mice.
Major organs (heart, liver, spleen, lung, and kidney) from treated and HA mice were sectioned to detect CM-Dil-labeled cells by immunofluorescence analysis. CM-Dil-positive cells were observed in the liver ( Figure 3C) and lung sections ( Figure 3D) of mice treated with iMSCs 1, 2, and 3 weeks after transplantation but were not detected in treated mice 4 weeks after transplantation or in HA mice. CM-Dil-positive cells were not observed in other organs (heart, spleen, and kidney) of treated and HA mice (Supplementary Figure 2B). The data show that iMSCs underwent sustained implantation in some organs, such as the liver and lungs, during the 3-week observation period, indicating that F8-modified iPSC-derived iMSCs are likely to restore FVIII function in vivo.

Phenotypic Rescue in HA Mice 1 Week After Transplantation of F8-Modified iPSC-Derived iMSCs
A tail-bleeding assay was performed to determine the phenotypic rescue in HA mice 1 week after transplantation of iMSCs derived from F8-modified iPSCs. Figure 4A shows representative images of the tails after 20 min of the bleeding assay in each group. At the recorded bleeding time, some HA mice engrafted with 17-9-iMSCs (n = 8), T-7-iMSCs (n = 6), and hiMSCs (n = 7) had already developed coagulation, while all mice of the HA group (n = 6) and animals treated with HA-iMSCs (n = 6) were still bleeding ( Figure 4B). We examined the survival rates of each group 48 h after the tail-clip challenge experiment. The data indicated that none of the HA mice or animals receiving HA-iMSCs survived more than 20 h after the FIGURE 3 | Restoration of FVIII function in HA mice using F8-modified iPSC-derived iMSCs. (A) Relative FVIII activity was detected 1-4 weeks after transplantation. HA mice, untreated HA mice; HA-iMSCs, HA mice transplanted with HA-iMSCs; 17-9-iMSCs, HA mice transplanted with 17-9-iMSCs; T-7-iMSCs, HA mice transplanted with T-7-iMSCs; hiMSCs, HA mice transplanted with hiMSCs. Bars represent the mean ± SEM, n = 5-6; ns, not significant, ***p < 0.001, **p < 0.01, and *p < 0.05 compared with HA mice (one-way ANOVA). (B) The data in (A) are shown as a line graph to describe the changes in FVIII activity during the 4-week observation period. (C) Liver and (D) lung tissue sections of iMSC-treated HA mice were analyzed using immunofluorescence; red fluorescence represents CM-Dil-positive cells, and DAPI was used for nuclear staining.
tail-clip challenge. Notably, two of the eight HA mice treated with 17-9-iMSCs, one of the six HA mice treated with T-7-iMSCs, and one of the seven HA mice treated with hiMSCs were alive 2 days after the tail-clip challenge, which was the endpoint of this experiment ( Figure 4C). These results suggest that transplantation of F8-modified iMSCs can partially rescue the bleeding phenotype of HA mice.

DISCUSSION
Mesenchymal stem cells -based therapy for various diseases has been in progress in clinical trials, including cardiovascular diseases (Zhao et al., 2015;Bartolucci et al., 2017;Butler et al., 2017;Kim et al., 2018), neurodegenerative diseases (Syková et al., 2017;Oh et al., 2018), and autoimmune diseases (Ghoryani et al., 2019), which emphasizes the therapeutic effects of MSCs. Additionally, MSCs have been applied for tissue repair and regeneration (Tan et al., 2016;Gjerde et al., 2018;Sha et al., 2019). Therefore, MSC-based therapy has been proven to be a safe and effective strategy. The immunosuppressive effects with low immunogenicity and the ability to produce endogenous FVIII (Sanada et al., 2013) suggest that MSCs are suitable for HA therapy.
Currently, MSCs used in preclinical studies and clinical trials are mainly isolated from bone marrow and adipose tissue; these sources may be limited in a single donor and may cause heterogeneity (Phinney et al., 1999;Pevsner-Fischer et al., 2011;Zaim et al., 2012;McLeod and Mauck, 2017;Huang et al., 2019). Moreover, the MSC isolation procedure is traumatic for the patients. Therefore, considering that invasive biopsy should be avoided for bleeding diseases, iPSCs generated from urine cells were selected as the source of iMSCs of HA patients. On the one hand, the source of iMSCs can be infinite because iPSCs possess unlimited proliferative capacity (Takahashi et al., 2007;Yu et al., 2007). On the other hand, iMSCs derived from iPSCs can maintain uniform quantity and quality thus avoiding heterogeneity. At present, genetic modification of MSCs is mainly performed via viral vector transduction; hence, immunogenicity of the viral capsid and the risk of random insertional mutations has to be addressed (Kotterman et al., 2015;Evens et al., 2018;Milone and O'Doherty, 2018;Shirley et al., 2020). Non-viral genetic modification by targeted integration can be efficiently achieved at the iPSC stage for sustainable expression of the target gene, representing an applicable strategy of ex vivo gene therapy for HA. In our study, 17-9-iPSCs and T-7-iPSCs with highly effective FVIII expression and no detectable off-target effects were obtained by gene editing and were selected in vitro. This screening strategy based on non-viral transduction may improve safety of subsequent in vivo experiments. To our knowledge, this is the first report that uses iMSCs derived from F8-modified iPSCs for the treatment of HA mice.
In the present study, the therapeutic level of plasma FVIII activity was sustained for 3 weeks in HA mice transplanted with 17-9-iMSCs, T-7-iMSCs, and hiMSCs. Then, FVIII activity was decreased down to the lowest value at week 4 after transplantation. This decrease may be due to the development of immunogenicity against human cells because we used immunocompetent HA mice that were not treated with immunosuppressive drugs. The SCID-HA mice transplanted FIGURE 4 | Tail-bleeding assay. (A) Representative images of the tails after 20 min of the bleeding assay for each group. (B) Bleeding time. WT, C57BL/6 mice. ns, not significant, ***p < 0.001 compared with HA mice (one-way ANOVA). (C) Proportions of surviving mice after the tail-clip challenge. HA mice, untreated HA mice (n = 6); HA-iMSCs, HA mice transplanted with HA-iMSCs (n = 6); 17-9-iMSCs, HA mice transplanted with 17-9-iMSCs (n = 8); T-7-iMSCs, HA mice transplanted with T-7-iMSCs (n = 6); hiMSCs, HA mice transplanted with hiMSCs (n = 7). **p < 0.01 compared with HA mice group (log-rank test).
with human MSCs presented restored hemostasis 1 week after the injection (Follenzi et al., 2012;Neumeyer et al., 2019). Gao et al. (2019) reported that subcutaneous injection of endothelial colony-forming cells and placenta-derived mesenchymal stromal cells transduced with a BDD-FVIII-expressing lentiviral vector achieved long-term engraftment and FVIII expression for 2 weeks in HA mice treated with cyclosporine A. However, side effects of immunosuppressive drugs have been described (Dobbels et al., 2008;Schmidt et al., 2019;Brami et al., 2020). Wang et al. (2013) implanted hFVIIIBD-MSCs isolated from hFVIIIBD-transgenic mice into HA mice by intravenous injection and achieved a therapeutic effect that lasted for 5 weeks. This effect may be attributed to reduced immune rejection between allogeneic inbred mice. In the present study, to some extent, a decrease in CM-Dil-positive iMSCs in the liver and lung sections can explain a decline in plasma FVIII activity in the treated mice, which might have produced hFVIII inhibitors; however, additional verification of this hypothesis is needed. Future studies will involve SCID-HA mice to reduce the immune response and provide a more accurate model of the effect of autologous treatment.
In summary, transplantation of iMSCs derived from F8corrected patient iPSCs into HA mice via the orbital vein restored FVIII function and rescued the HA phenotype. Our results provide proof-of-concept for cell-based gene therapy via targeted genetic correction of FVIII in autologous iPSCs. In future studies, iMSCs will be transplanted into an HA mouse model with immunodeficiency to test long-term efficacy of the treatment. This work also suggests a feasible strategy of therapeutic gene editing for HA and other genetic birth defects involving large sequence variants.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

ETHICS STATEMENT
The animal study was reviewed and approved by School of Life Sciences of Central South University.

AUTHOR CONTRIBUTIONS
LW and ZH designed the research and edited the manuscript. LQ and MX performed the experiments. LQ prepared the figures and drafted the manuscript. MZ edited the manuscript. XL assisted in manuscript preparation. All authors read and approved the final manuscript.