Ectopic Expression of FVIII in HPCs and MSCs Derived from hiPSCs with Site-Specific Integration of ITGA2B Promoter-Driven BDDF8 Gene in Hemophilia A

Hemophilia A (HA) is caused by mutations in the coagulation factor VIII (FVIII) gene (F8). Gene therapy is a hopeful cure for HA; however, FVIII inhibitors formation hinders its clinical application. Given that platelets promote coagulation via locally releasing α-granule, FVIII ectopically expressed in platelets has been attempted, with promising results for HA treatment. The B-domain-deleted F8 (BDDF8), driven by a truncated ITGA2B promoter, was targeted at the ribosomal DNA (rDNA) locus of HA patient-specific induced pluripotent stem cells (HA-iPSCs). The F8-modified, human induced pluripotent stem cells (2bF8-iPSCs) were differentiated into induced hematopoietic progenitor cells (iHPCs), induced megakaryocytes (iMKs), and mesenchymal stem cells (iMSCs), and the FVIII expression was detected. The ITGA2B promoter-driven BDDF8 was site-specifically integrated into the rDNA locus of HA-iPSCs. The 2bF8-iPSCs were efficiently differentiated into 2bF8-iHPCs, 2bF8-iMKs, and 2bF8-iMSCs. FVIII was 10.31 ng/106 cells in lysates of 2bF8-iHPCs, compared to 1.56 ng/106 cells in HA-iHPCs, and FVIII was 3.64 ng/106 cells in 2bF8-iMSCs lysates, while 1.31 ng/106 cells in iMSCs with CMV-driven BDDF8. Our results demonstrated a high expression of FVIII in iHPCs and iMSCs derived from hiPSCs with site-specific integration of ITGA2B promoter-driven BDDF8, indicating potential clinical prospects of this platelet-targeted strategy for HA gene therapy.


Introduction
Hemophilia A (HA) is an X-linked recessive genetic disorder that affects approximately 1 in 5000 newborn boys. The clinical symptoms of HA patients are mainly spontaneous joint and muscle bleeding, or traumatic bleeding, leading to disability; it can even be lifethreatening due to intracranial hemorrhage [1]. At present, there is no cure, and the main clinical treatment is replacement therapy with intravenous injections of plasma-derived or recombinant coagulation factor VIII (FVIII) protein. However, as the half-life of FVIII in the plasma is only 12 h [2], patients require repeated infusions, which makes this treatment very costly and brings a heavy economic and psychological burden to them and their families. Currently, the novel EHL-factors and the non-factor therapies are reducing the burden of the life-long treatment without impacting the long-term therapeutic solution of gene therapy in hemophilia [3]. A major complication is the formation of FVIII inhibitor, which affects approximately 25% to 30% of HA patients, decreasing its coagulation efficacy [4,5].
Gene therapy provides a promising option for HA [6]. As a slight increase in the coagulation activity of plasma FVIII can significantly alleviate the clinical bleeding symptoms, HA has been considered as a suitable disease for gene therapy [7]. Currently, gene therapy for hemophilia based on adeno-associated virus (AAV) vectors has been advanced to clinical trials [8,9]. However, there are still some challenges for the AAV-based gene therapy. About 30-60% of the population have antibodies against different serotypes of AAV capsids, and different serotypes have cross-immunity, which greatly limits its wide application [10]. Moreover, it is worth noting that there is a risk of insertion mutation in the use of viral elements. Recently, in the long-term follow-up study of HA gene therapy based on AAV, the integration of AAV vectors near the genes that control cell growth in HA model dog hepatocytes was found, which has a potential risk of inducing tumors [11].
To circumvent some concerns, reduce the potential risks, and provide a choice for the HA patients that have antibodies against AAV capsids, an ex vivo gene therapy strategy has been actively studied [12,13]. This strategy can select cells without off-target mutations via whole-genome sequencing analysis and provide a reliable cell source, possessing higher safety. Induced pluripotent stem cells (iPSCs) have become ideal targets for ex vivo gene therapy due to their unlimited proliferation ability and multidirectional differentiation potential.
Immune response has always been a key problem in HA gene therapy. 25-30% of HA patients produce FVIII inhibitory antibodies, resulting in ineffective therapy. The ectopic expression of FVIII in platelets is an effective strategy that reduces the immune response against FVIII [14,15]. Once blood vessels are damaged, platelets will immediately gather at the damaged site and secrete FVIII to coagulate the damaged vessels. The genetically modified FVIII function acts directly from platelets to the bleeding site without entering the circulatory system, thus reducing the risk of producing FVIII inhibitors. Some studies have used bone marrow-derived hematopoietic stem cells (HSCs) modified using lentivirus to achieve the restoration of the FVIII function [16]. However, the HSCs isolation procedure is traumatic for patients and there is an insertional risk for lentivirus. In recent decades, there have been reports of hematopoietic progenitor cells (HPCs) generated from human embryonic stem cells and human iPSCs (hiPSCs), and harvested megakaryocytes (MKs) and platelets by further differentiation in vitro [17,18]. Further, some studies have shown that lipogenic mesenchymal stem cells can differentiate into MKs in vitro [19]. Differentiation of hiPSCs into HPCs and mesenchymal stem cells (MSCs) may be an efficient method to generate an infinite source of genetically modified targeted cells with uniform quantity and quality.
In this study, the truncated ITGA2B promoter-driven B-domain-deleted F8 (BDDF8) named 2bF8 was targeted at the ribosomal DNA (rDNA) locus of HA patient-specific iPSCs (HA-iPSCs), which were subsequently differentiated into HPCs and MSCs. The ectopically expressed FVIII was validated in HPCs and MSCs. This strategy may provide an innovative approach for platelet-targeted gene therapy for HA.

Construction of Nonviral Targeting Vector minipHrn-2bF8
In this study, the donor plasmid minipHrn-2bF8 was constructed based on the plasmid minipHrneo described previously. Between the 5 long homologous arm (935 bp) and the 3 short homologous arm (591 bp), the 889 bp megakaryocyte-specific promoter of ITGA2B gene driving the BDDF8 and a promoterless neomycin resistance (Neo) cassette flanked by locus of X (cross)-overinP1 (LoxP) sites were contained ( Figure 1A). The Neo cassette contained an encephalomyocarditis virus internal ribosomal entry site (EMCV-IRES), which enabled Neo gene expression under the control of the endogenous RNA polymerase I (Pol I) promoter after homologous recombination, so that the gene-targeted hiPSCs could be enriched by this promoter trapping. The donor plasmid was identified via restriction endonuclease digestion ( Figure 1B) and Sanger sequencing ( Figure 1C).

Gene Targeting of F8 into the rDNA Locus of HA-iPSCs Using TALENickases
The donor vectors minipHrn-2bF8 and TALENickases were nucleofected into the rDNA locus of HA-iPSCs ( Figure 2A). After selection by G418 for 7 days, 18 clones were picked up and expanded. PCR screening with primers F2/R2 and F3/R3 showed that 12 out of 18 clones (66.67%) were positive for the expected band of 1652 bp and 1279 bp ( Figure 2B). The PCR products were then verified by Sanger sequencing, which indicated the junction between the exogenous 2bF8 cassette and the endogenous rDNA sequence ( Figure 2C). In addition, eight targeted clones were confirmed by Southern blotting using a probe homologous to the Neo gene, showing the gene-targeting fragments of 4920 bp ( Figure 2D). We randomly selected one stable F8-modified hiPSCs (2bF8-iPSCs) for further research. These results indicated the site-specific integration of 2bF8 cassette into the rDNA locus. To analyze the off-target activity of TALENickases, gDNA of the HA-iPSCs and 2bF8-iPSCs were isolated and the top three potential off-target sites of TALENickases (https://tale-nt.cac.cornell.edu/) (accessed on 1 June 2021) were amplified, followed by Sanger sequencing, and no indels were observed compared to the HA-iPSCs ( Figure S1).

Verification of F8 Expression in F8-Modified hiPSCs
We designed specific primers based on the sequence crossing the exon 19-exon 23 of F8 to detect the transcription of exogenous F8 using RT-PCR. The results showed the transcripts crossing the exon 19-exon 23 of F8 could be detected in 2bF8-iPSCs, but not in HA-iPSCs, suggesting that the transcript of F8 exon 19-exon 23 in HA-iPSCs was nonexistent ( Figure 3A). The qRT-PCR showed that the expression of integrated exogenous F8 in the 2bF8-iPSCs was 172 times higher than that of endogenous F8 in normal hiPSCs, indicating that F8 mRNA could be effectively transcribed in 2bF8-iPSCs ( Figure 3B). FVIII antigen was detected by ELISA in the cell lysates of HA-iPSCs (1.82 ng/10 6 cells), 2bF8-iPSCs (2.42 ng/10 6 cells), and hiPSCs (3.35 ng/10 6 cells), while not detectable in the cell supernatant ( Figure 3C). The mRNA of F8 genes from HA-iPSCs, hiPSCs, and 2bF8-iPSCs were detected by RT-PCR. (B) Relative expression of F8 in HA-iPSCs, hiPSCs, and 2bF8-iPSCs was detected by qRT-PCR. GAPDH was used as the internal reference. ND, not detected. * p < 0.05. Bars represent the mean ± SEM (n = 3, each group). (C) ELISA detection of the FVIII antigen in the supernatant and cell lysate of HA-iPSCs, hiPSCs, and 2bF8-iPSCs. Data represent the mean ± SEM (n = 3, independent cultures).

Generation and Characterization of Induced HPCs (iHPCs) and Induced MKs (iMKs)
We differentiated HA-iPSCs, 2bF8-iPSCs, and hiPSCs into HA-iHPCs, 2bF8-iHPCs, and hiHPCs, respectively, during which the cells converted from adherent to suspension ( Figure 4A). After cell sorting with anti-CD34 beads, more than 90% of the iHPCs were double-positive for CD34 and CD43 and 90% were CD34 positive and CD38 negative, while only 47% of the unsorted iHPCs were double-positive for CD34 and CD43 and 45.5% were CD34 positive and CD38 negative. These results revealed that purer iHPCs were obtained via sorting. In addition, 1.1% of the sorted iHPCs showed as double-positive for CD34 and CD41a, indicating that sorted iHPCs have the potential to differentiate into iMKs.
Sorted iHPCs continued to differentiate into iMKs. After 5 days' culture, the percentage of CD34 decreased and CD43 was negative, while the positive rate of CD41a increased to 2.1%, indicating that iMKs were obtained ( Figure 4B).

Verification of F8 Expression in F8-Modified iHPCs
The qRT-PCR results showed that exogenous F8 could be effectively expressed in the 2bF8-iHPCs ( Figure 4C), and the F8 expression of 2bF8-iHPCs was 218 times higher than that in hiHPCs ( Figure 4D). Similar to that of hiPSCs, FVIII antigen was detected by ELISA in cell lysates, but not detected in the supernatants of iHPCs ( Figure 4E). Therefore, it is indicated that the 2bF8-iHPCs produced abundant FVIII protein (10.31 ng/10 6 cells) in cell lysates at the iHPCs stage, being prepared for the enrichment of FVIII in platelets.

Verification of FVIII Expression in F8-Modified iMSCs
The exogenous F8 expression was detected in the 2bF8-iMSCs via RT-PCR ( Figure 5D). The F8 mRNA of 2bF8-iMSCs was 13 times higher than that of hiMSCs, while the F8 mRNA of T-7-iMSCs was 17 times higher than that of hiMSCs ( Figure 5E), indicating that the universal promoter (CMV) was stronger than the megakaryocyte-specific promoter at the iMSCs stage. We evaluated the FVIII antigen in T-7-iMSCs and 2bF8-iMSCs by ELISA ( Figure 5F) and found that FVIII in the supernatant of T-7-iMSCs was 0.23 ng/10 6 cells and 0.05 ng/10 6 cells in 2bF8-iMSCs, while it was opposite in the cell lysates of T-7-iMSCs (1.31 ng/10 6 cells) and 2bF8-iMSCs (3.64 ng/10 6 cells), providing the potential of FVIII stored in platelets derived from 2bF8-iMSCs.   (B) Identification of osteogenic, chondrogenic, and adipogenic differential potential of 2bF8-iMSCs.

Discussion
Immune response which results in the development of anti-FVIII or IX inhibitors with ineffectiveness of the substitutive therapy is still a key problem in hemophilia therapy. Heterotopic storage of FVIII in platelets is an ingenious method for HA gene therapy. This strategy can not only avoid the inhibitory antibodies to FVIII in circulation, but also increase the concentration of FVIII in the bleeding sites through platelet aggregation and achieve the goal of maximizing the therapeutic effect with low-level FVIII. Du et al. used a lentiviral vector encoding the ITGA2B gene promoter-driven BDDF8 to transfect the isolated HSCs, then the cells were transplanted into a dog model and the platelet-specific human FVIII was detected, stored, and released from activated platelets [22]. In the present study, the truncated megakaryocyte-specific promoter was used to drive the expression of BDDF8 that was targeted at the rDNA locus of HA-iPSCs. We detected the specificity of the promoter and found that the endogenous ITGA2B gene could be generally transcribed in hiPSCs and iHPCs, decreasingly transcribed in iMSCs, and not transcribed in T-lymphocyte ( Figure S2), which was consistent with the results of the expression of BDDF8 driven by this truncated promoter in corresponding cells and in the previous report.
The hiPSCs have been widely used as the target cells of gene editing due to their unlimited proliferation and multi-differentiation potential ability [23,24]. Gene editing in hiPSCs has been proven to be easily and effectively operated [25]. The hematopoietic capabilities of HPCs derived from gene-edited hiPSCs provides new opportunities for cell sources of hematopoietic transplantation, which is one of the most common cell transplantation approaches [26,27]. In this study, we differentiated 2bF8 gene-modified hiPSCs into iHPCs in vitro and detected FVIII expression. However, the number of iHPCs obtained by in vitro differentiation is limited, and it is still difficult to conduct in vivo experiments. Further studies need to be performed to improve the efficiency of differentiation of hiPSCs to HPCs and overcome the obstacles to clinical application.
In addition to aiming at obtaining platelets through HPCs differentiation, we tried to explore some alternative approaches [28,29]. Megakaryocytogenesis and platelet formation are the result of the interaction between HPCs, humoral factors, and bone marrow MSCs in a physiological environment. Human adipose-derived mesenchymal stromal/stem cells (ASCs) can differentiate into megakaryocyte lineages without gene modification and, besides that, ASCs secrete endogenous thrombopoietin to promote platelet production [19]. Since ASCs are composed of heterogenous cells, studies have established ASCs lines that extend to obtain a more homogenous population [30]. For HA patients, isolation of autologous ASCs is invasive. In the present study, we generated and gene-modified hiPSCs from the urine cells that were collected from HA patients non-invasively as the source for iMSCs expressing FVIII. These iMSCs could be continuously produced due to the infinite proliferation capacity of hiPSCs. For avoiding cell heterogeneity, iMSCs derived from hiPSCs are consistent in quantity and quality. A comparative study on the proliferation and differentiation abilities between iMSCs and autologous tissue-derived MSCs showed that iMSCs have better potential in adipogenic differentiation experiments in vitro [31]. Given the above, it is feasible to generate megakaryocytes from hiPSCs-derived iMSCs.
In summary, a megakaryocyte-specific promoter-driven BDDF8 was efficiently targeted into the multi-copy rDNA locus of HA-iPSCs using a non-viral vector and TALENickases without potential off-target effects. The exogenous FVIII was stably expressed in iHPCs and iMSCs derived from F8-modified hiPSCs. These results provide a proof of concept for an efficient approach for the site-specific gene addition in hiPSCs for platelet-targeted HA gene therapy.

Plasmids and Gene Targeting
In our previous study, we constructed the non-viral human rDNA targeting vector, minipHrneo [34,35]. In this study, we used two Phanta Max Master Mixes (Vazyme, Nanjing, China) and amplified the 889 bp promoter region upstream of ITGA2B gene from human genomic DNA (gDNA) with the following pairs of primers: F1: 5 -GGATCCGTGCTCA-ATGCTGTGCCTACGTG-3 /R1: 5 -GCTAGCTAGACATATGGGGCCAGCTCCTCCTCCTT-3 . We then inserted them into the targeting vector minipHrneo after digestion by NheI endonuclease (New England BioLabs, Hitchin, England) and BamH1 endonuclease (New England BioLabs, Hitchin, England). Then the BDDF8 open reading frame was amplified from pHrnF8 vector [36] and inserted into the NheI restriction site of the targeting vector containing the 889 bp ITGA2B promoter by ClonExpress II One Step Cloning Kit (Vazyme, Nanjing, China) to generate the donor vector minipHrn-2bF8.

Southern Blotting
The gDNA was extracted and digested with XhoI endonuclease (New England Biolabs, Hitchin, England) at 37 • C overnight and 10 µg of each digested sample was loaded on 1% agarose gel, followed by electrophoresis at 150 V for 2 h. The samples were then transferred to nylon membranes and the DIG-labeled Molecular WeightII (Roche, Basel, Switzerland) was used as a DNA marker. Then the membranes were hybridized with DIG Easy Hyb Granules (Roche, Basel, Switzerland) at 42 • C overnight, washed, and detected with CDP-Star. The probes were labeled by DIG Probe Synthesis Kit with F5: 5 -GCCGAGAAAGTATCCATCA-3 /R5: 5 -CAGAGTCCCGCTCAGAAG-3 . All reagents and materials were purchased from Roche.

Immunofluorescence Staining
Cells plated on chamber slides were rinsed by DPBS (Thermo Fisher Scientific, Waltham, MA, USA), then fixed with 4% paraformaldehyde for 15 min, followed by permeabilization with PBST (DPBS containing 0.1% Triton X-100 (Sigma-Aldrich, St. Louis, MO, USA)) for 15 min. After washing with DPBS, cells were blocked in 5% BSA (R&D Systems, MN, USA) for 30 min, then incubated with the diluted primary antibodies (BD Biosciences, Franklin Lakes, NJ, USA) in 5% BSA at 4 • C overnight. After a thorough washing, the cells were blocked with 5% BSA for 30 min again and then incubated with the secondary antibodies for 1 h at room temperature in the dark. Nuclei were stained with DAPI (Sigma-Aldrich, St. Louis, MO, USA) for 5 min. Confocal (Leica, Wetzlar, Germany) photography was used to analyze the stained cells.

Differentiation of hiPSCs into iHPCs and iMKs
The hiPSCs, HA-iPSCs, and 2bF8-iPSCs were differentiated into hiHPCs, HA-iHPCs, and 2bF8-iHPCs using a STEMdiffTM hematopoietic kit (STEMCELL Technologies, Vancouver, BC, Canada) according to the manufacturer's instructions. Briefly, cells were dissociated to aggregates by 0.5 mM EDTA (Thermo Fisher Scientific, Waltham, MA, USA) and plated 40-80 aggregates/well on Matrigel-coated six-well plates in mTesR Plus. The cells were cultured in differentiation A medium on the second day, then the medium was half-refreshed. On day 3, the cells were cultured with differentiation B medium and then the medium was half-refreshed every other day, and cells were gradually suspended. The iHPCs were harvested after being cultured in differentiation B medium for 7 days.
The differentiated cells were sorted using the Direct CD34 Progenitor Cell Isolation Kit (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the manufacturer's instructions.

Characterization of iHPCs and iMKs
The iHPCs were dissociated into single cells with TrypLE™ Express and counted, and 5 × 10 4 cells were suspended in 100 µL DPBS with 5% BSA. Then the cells were incubated with BV421-conjugated anti-human CD34, FITC-conjugated CD38, BV510-conjugated CD43, and APC-conjugated CD41a (BD Biosciences, Franklin Lakes, NJ, USA) at 4 • C for 30 min in the dark. The stained cells were washed twice using DPBS and characterized by flow cytometry. The iMKs were incubated with BV421-conjugated anti-human CD34, BV510conjugated CD43, and APC-conjugated CD41a (BD Biosciences, Franklin Lakes, NJ, USA) at 4 • C for 30 min in the dark and analyzed by flow cytometry.

Differentiation of hiPSCs into iMSCs
A STEMdiffTM mesenchymal progenitor kit (STEMCELL Technologies, Vancouver, BC, Canada) was used to differentiate hiPSCs into iMSCs. According to the manufacturer's protocol, hiPSCs were seeded on a Matrigel-coated 12-well plate at 5 × 10 4 cells/cm 2 in mTesR Plus containing 10 µM Y27632. After 2 days, the medium was refreshed with STEMdiffTM-ACF mesenchymal induction medium (STEMCELL Technologies, Vancouver, BC, Canada) for 4 days. Then the cells were cultured in STEMdiffTM-ACF medium for an additional 2 days and passaged on a six-well plate precoated with STEMdiffTM-ACF attachment substrate (STEMCELL Technologies, Vancouver, BC, Canada). The cells were passaged every 3 days. After three passages, the cells were seeded onto a 0.1% gelatincoated 10-cm dish in αMEM basal medium containing 10% FBS, 2 mM GlutaMAXTM, and 0.1% bFGF. The medium was changed every day.
The differentiation potential of iMSCs was identified via osteogenesis (Gibco, New York, NY, USA), adipogenesis (STEMCELL Technologies, Vancouver, BC, Canada), and chondrogenesis (STEMCELL Technologies, Vancouver, BC, Canada) differentiation kits, according to the manufacturer's instructions. After being cultured with differentiation medium for 2 to 3 weeks, cells were stained with alizarin red, oil red O, or alcian blue dye (Cyagen Biosciences Inc., Guangzhou, China)) for 30 min separately, then the stained cells were analyzed using a light microscope.

FVIII Assay
The hiPSCs and iMSCs culture supernatants were harvested 24 h after the medium was changed. The cells were isolated by TrypLE TM Express, counted, and re-suspended in a 600 µL sample dilution (Cedarlane, Burlington, ON, Canada). The iHPCs culture supernatants and cells were harvested by centrifugation 24 h after the medium was changed. The cells were counted and re-suspended in a 600 µL sample dilution (Cedarlane, Burlington, ON, Canada). The cells were lysed by three freeze-thaw cycles and the supernatant was collected by centrifugation. ELISA was performed using paired antibodies for ELISA-factor VIII:C (Cedarlane, Burlington, ON, Canada) according to the manufacturer's instructions.

Statistical Analysis
GraphPad Prism 5.0 was used for data analysis. Data were analyzed using ANOVA for more than two groups. All values are presented as the mean ± SEM.

Conclusions
The BDDF8 gene driven by a megakaryocyte-specific promoter was efficiently targeted into the rDNA locus of HA-iPSCs and stable expression of exogenous FVIII in F8-modified iHPCs and iMSCs was verified. This innovative strategy based on platelets derived from gene-modified hiPSCs can provide a promising alternative for HA gene therapy.