Received: 18 May 2022 Revised: 13 August 2022 Accepted: 6 September 2022 DOI: 10.1002/ctm2.1063 RESEARCH ARTICLE ABO gene editing for the conversion of blood type A to universal type O in Rhnull donor-derived human-induced pluripotent stem cells Paolo Petazzi1 Laia Miquel-Serra2,3 Sergio Huertas2,3 Cecilia González2,3 Neus Boto2 Eduardo Muñiz-Diaz2,3,4 Pablo Menéndez1,5,6,7,8 Ana Sevilla9,10 Núria Nogués2,3,4 1 JosepCarreras Leukemia Research Institute, Barcelona, Spain Graphical Abstract 2 Immunohematology Laboratory, Barcelona, Spain 3 Transfusionalmedicine. Vall d’Hebron Research Institute (VHIR), Barcelona, Spain 4 Department of Medicine, Universitat Autònoma de Barcelona (UAB), Barcelona, Spain 5 Department of Biomedicine, School of Medicine, University of Barcelona, Barcelona, Spain 6 Centrode Investigación Biomédica en Red de Cáncer-CIBER-ONC, Instituto de Salud Carlos III, Barcelona, Spain 7 Red Española de Terapias Avanzadas (TERAV), Instituto de Salud Carlos III (RICORS, RD21/0017/0029) ∙ CRISPR/Cas9 mediated ABO geneedition allows the conversion of blood type 8 Institució Catalana de Recerca i Estudis A to universal type O in Rhnull donor-derived hiPSCs Avançats (ICREA), Barcelona, Spain ∙ A targeted knock-in strategyallows to reproduce the c.261delG polymorphism 9 Department of Cell Biology, Physiology and Immunology, Faculty of Biology, present in the most commoninactive ABO*O.01 allele University of Barcelona, Barcelona, Spain ∙ This work provides a newapproach potentially applicable to cell lines estab- 10 Institute of Biomedicine of the lished from rare blood donorsnot carrying blood type O University of Barcelona (IBUB), Barcelona, Spain Correspondence Pablo Menéndez, Josep Carreras Leukemia Research Institute, Barcelona, Spain. Email: pmenendez@carrerasresearch.org Clin. Transl. Med. 2022;12:e1063. wileyonlinelibrary.com/journal/ctm2 https://doi.org/10.1002/ctm2.1063 PETAZZI et al. Ana Sevilla, Department of Cell Biology, Physiology and Immunology, Faculty of Biology, University of Barcelona, Barcelona 08028, Spain. Email: asevilla@ub.edu Núria Nogués, Immunohematology Labo- ratory, Banc de Sang i Teixits, Pg. Taulat 116, Barcelona 08005, Spain. Email: nnogues@bst.cat Pablo Menéndez, Ana Sevilla and Núria Nogués are the co-last authors. Received: 18 May 2022 Revised: 13 August 2022 Accepted: 6 September 2022 DOI: 10.1002/ctm2.1063 RESEARCH ARTICLE ABO gene editing for the conversion of blood type A to universal type O in Rhnull donor-derived human-induced pluripotent stem cells Paolo Petazzi1 Laia Miquel-Serra2,3 Sergio Huertas2,3 Cecilia González2,3 Neus Boto2 Eduardo Muñiz-Diaz2,3,4 Pablo Menéndez1,5,6,7,8 Ana Sevilla9,10 Núria Nogués2,3,4 1 Josep Carreras Leukemia Research Institute, Barcelona, Spain 2 Immunohematology Laboratory, Barcelona, Spain 3 Transfusional medicine. Vall d’Hebron Research Institute (VHIR), Barcelona, Spain 4 Department of Medicine, Universitat Autònoma de Barcelona (UAB), Barcelona, Spain 5 Department of Biomedicine, School of Medicine, University of Barcelona, Barcelona, Spain 6 Centro de Investigación Biomédica en Red de Cáncer-CIBER-ONC, Instituto de Salud Carlos III, Barcelona, Spain 7 Red Española de Terapias Avanzadas (TERAV), Instituto de Salud Carlos III (RICORS, RD21/0017/0029) 8 Institució Catalana de Recerca i Estudis Avançats (ICREA), Barcelona, Spain 9 Department of Cell Biology, Physiology and Immunology, Faculty of Biology, University of Barcelona, Barcelona, Spain 10 Institute of Biomedicine of the University of Barcelona (IBUB), Barcelona, Spain Correspondence Pablo Menéndez, Josep Carreras Abstract Leukemia Research Institute, Barcelona, The limited availability of red cells with extremely rare blood group phenotypes Spain. Email: pmenendez@carrerasresearch.org is one of the global challenges in transfusion medicine that has prompted the search for alternative self-renewable pluripotent cell sources for the in vitro gen- Ana Sevilla, Department of Cell Biology, eration of red cells with rare blood group types. One such phenotype is the Physiology and Immunology, Faculty of Biology, University of Barcelona, Rhnull , which lacks all the Rh antigens on the red cell membrane and represents Barcelona 08028, Spain. one of the rarest blood types in the world with only a few active blood donors Email: asevilla@ub.edu available worldwide. Rhnull red cells are critical for the transfusion of immu- Núria Nogués, Immunohematology nized patients carrying the same phenotype, besides its utility in the diagnosis Laboratory, Banc de Sang i Teixits, Pg. of Rh alloimmunization when a high-prevalence Rh specificity is suspected in Taulat 116, Barcelona 08005, Spain. Email: nnogues@bst.cat a patient or a pregnant woman. In both scenarios, the potential use of human- induced pluripotent stem cell (hiPSC)-derived Rhnull red cells is also dependent Pablo Menéndez, Ana Sevilla and Núria on ABO compatibility. Here, we present a CRISPR/Cas9-mediated ABO gene Nogués are the co-last authors. edition strategy for the conversion of blood type A to universal type O, which we have applied to an Rhnull donor-derived hiPSC line, originally carrying blood Paolo Petazzi and Laia Miquel-serra equally contributed to this article. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2022 The Authors. Clinical and Translational Medicine published by John Wiley & Sons Australia, Ltd on behalf of Shanghai Institute of Clinical Bioinformatics. Clin. Transl. Med. 2022;12:e1063. wileyonlinelibrary.com/journal/ctm2 1 of 13 https://doi.org/10.1002/ctm2.1063 2 of 13 PETAZZI et al. Funding information Spanish Ministry of Economy and group A. This work provides a paradigmatic example of an approach potentially Competitiveness, Grant/Award Number: applicable to other hiPSC lines derived from rare blood donors not carrying blood SAF2016-80481-R; Fundación Leo Mesi; Catalan Goverment, Grant/Award type O. Numbers: PERIS 2017-2019, SGR330; European Research Council, KEYWORDS Grant/Award Number: CoG-2014-646903; ABO, CRISPR, gene edition, iPSC, rare blood groups, Rhnull , transfusion medicine Ministerio de Ciencia e Innovación, Grant/Award Numbers: PGC2018-098626-B100, RTC-2017-6367-1, RYC-2016-19962; Merck Salud Foundation; CERCA/Generalitat de Catalunya and Fundació Josep Carreras- Obra Social la Caixa 1 INTRODUCTION specific blood group genes.16 However, the potential use of hiPSC-derived red cells is also dependent on the ABO The transfusion of red blood cells (RBCs), currently type. Except for the rare Bombay phenotype, extremely obtained from volunteer blood donations, is an essen- infrequent or null blood group types are not necessarily tial therapy for patients with chronic or acute anaemia. encountered in blood type O donors. Here, we present a This form of cell-based therapy is an indispensable part of CRISPR/Cas9-mediated ABO gene edition strategy for the modern healthcare systems. However, the perspective of conversion of blood type A to universal type O, which we insufficient blood supply due to population aging, and the have applied to an Rhnull donor-derived hiPSC line, origi- potential risk of transfusion-transmitted infections remain nally carrying blood group A. This approach is potentially major concerns.1,2 In addition, the scarcity of donors with applicable to other hiPSC lines derived from rare blood rare blood types represents a global challenge when com- donors, not carrying blood type O. patible red cells with a rare blood phenotype are required for transfusion.3,4 For these reasons, the in vitro generation of RBCs, to supplement the donation system, is nowadays 2 MATERIALS AND METHODS a major focus of research in transfusion medicine. Beyond transfusion requirements, red cells with infre- 2.1 Ethics approval quent phenotypes are also necessary for diagnostic pur- poses in clinical laboratories. The identification of unusual Procedures for sample collection and hiPSC line gen- red cell antibody specificities in patient sera depends on eration were approved by the Clinical Research Ethics the availability of reagent red cells with rare phenotypes Committee (CEIC) of the Vall d’Hebron Research Institute, or infrequent antigen combinations for serological cross- the Comisión de Garantías para la Donación y Utilización matching, which is crucial to allow the accurate selection de Células y Tejidos Humanos (Spanish National Stem and effective search of compatible units for transfusion. Cell Bank, ISCIII) and the Catalan Authority for Stem During the past decade, enormous progress has been Cell Research (0336E/10472/2017). Informed consent was made in the in vitro manufacture of human RBCs from obtained from the donor. All procedures involving animals different cell sources.5–9 Among these, human-induced were approved by the Institutional Animal Ethics Board, pluripotent stem cells (hiPSCs) provide an unlimited and the protocols were approved by the Catalan Authority. source of hematopoietic progenitor cells, which can subse- quently be differentiated into erythroid cells. hiPSC lines can also be derived from easily accessible peripheral blood 2.2 Generation and culture of hiPSCs mononuclear cells (PBMCs) from selected donors10,11 and be amenable to gene editing.12,13 Overall, these features To generate Rh null donor-derived hiPSCs, PBMCs were make them a promising source for sustainable production isolated from a 20 ml-whole blood sample of the selected of customized red cells. blood donor, previously identified and characterized at Different hiPSC lines have already been obtained from the Immunohematology Reference Laboratory of the Banc existing donors with rare blood types14,15 or have been de Sang iTeixits (Barcelona). MNCs were isolated using modified using CRISPR/Cas9 gene editing approaches to standard density gradient centrifugation with SepMate™ reproduce uncommon null phenotypes by knocking-out tubes (StemCell Technologies, Canada). The PBMCs were PETAZZI et al. 3 of 13 carefully recovered from the interface and washed in PBS at https://github.com/anasevilla/ABO-gene-editing (see 1×. PBMCs were reprogrammed using the integration-free Table S2 for sequencing primers). CytoTune-iPS R 2.0 Sendai Reprogramming Kit (Ther- moFisher, USA), which contains Sendai virus particles for the expression of the four Yamanaka factors.17 2.4 T7 endonuclease assay Undifferentiated hiPSCs were maintained in mTeSR™1 (StemCell Technologies, Canada) on 3 μg/ml of HEK-293T cultures were dissociated with TrypLE express Laminin-521 (StemCell Technologies)-coated plates (ThermoFisher) and 2 × 105 cells were transfected by and expanded using EDTA passaging solution (Ther- CRISPRMAX Transfection Reagent (ThermoFisher) with moFisher). Media samples were routinely tested for the 12 pmol of each of the RNAs (gRNA and tracrRNA) and absence of mycoplasma contamination using the selective the Cas9 protein. Genomic DNA was extracted 2 days after biochemical test MycoAlert™PLUS (Lonza, Switzerland). transfection. Genomic regions flanking the CRISPR target sites were PCR amplified (Table S1). PCR products were denatured, re-annealed and subsequently treated with 5U 2.3 CRISPR/Cas9 gene editing of T7EI at 37◦ C for 15 min. For ABO gene targeting, two strategies were approached: (1) the generation of a gene knock-out (KO) and (2) 2.5 RHAG and ABO gene sequencing the insertion of the naturally occurring c.261delG single nucleotide deletion through a short sequence knock-in For RHAG gene sequence analysis, DNA was extracted (KI). For both strategies, we designed RNA guides (gRNA) from donor PBMCs or hiPSCs by an automated method using CRISPR-direct tool (https://portals.broadinstitute. using the QIAsymphony instrument (Qiagen, Germany). org/gpp/public/analysis-tools/sgrna-design). We selected Primers to amplify RHAG gene exon 6 are listed in Table in both cases the target sites with the lowest number of S2. For CRISPR edit validation, genomic DNA was isolated predicted off-targets. The gRNA sequences are depicted from each expanded clone using DNeasy Blood and Tis- in Table S1. For the KI strategy, we designed a single- sue Kit (Promega). DNA regions encompassing guide sites strand donor DNA carrying the c.261delG mutation and were amplified using specific primers (Table S2). Ampli- a mutated PAM to avoid re-cutting of the target sequence fication was performed with SequalPrep™ Long PCR Kit (Table S1). Each guide was previously transfected with with dNTPs (Applied Biosystems, USA). The PCR prod- the Alt-R R S.p. Cas9 Nuclease V3 (IDT# 1081058) and ucts were Sanger sequenced using the Big Dye Terminator the Alt-R CRISPR-Cas9 tracrRNA, ATTO™ 550, (IDT R v1.1 kit (Applied Biosystems). Sequencing primers are also #1075927) in HEK-293T cells according to IDT protocol listed in Table S2. DNA sequences were aligned with the and tested for cutting efficiency by the T7 endonuclease reference genomic sequences: NG_011704.1 for the RHAG assay (as described below). The most efficient guide was gene and NG_006669.2 for ABO gene, using the CLC nucleofected with the Cas9 protein into the parental hiPSC GenomicWorkbench 21.0.3 software (Qiagen). line hiPSC#1 using the Neon Electroporation transfection System (ThermoFisher), and cells were plated into geltrex- coated plates. Nucleofection efficiency was assessed after 2.6 Cell line identity 24 h. Forty-eight hours post nucleofection cells were plated as single cells on geltrex-coated 96-well plates for clonal To confirm the cell line identity, genomic DNA was selection in mTeSR™1 (StemCell Technologies, Canada) extracted from iPSC clones as well as donor PBMCs and supplemented with 10 μM Y-27632 and cloneR (StemCell used for short-tandem repeat (STR) marker analysis using Technologies). Single-cell clones were expanded for 2 to the Mentype R Chimerasystem R (BiotypeDiagnostic R 3 weeks and subsequently analysed by Sanger sequencing GmbH, Germany). for the target sites modified. To check for unintended targets of our sgRNAs, IDT design checker (https://eu.idtdna.com/site/order/ 2.7 Karyotype analysis designtool/index/CRISPR_SEQUENCE) was used, and the top five in silico-predicted off-targets cod- Genomic integrity of the generated hiPSCs was evaluated ing regions of ABO E3 sgRNA (CCDC78, HTR5A, by G-banded metaphase karyotype analysis (Molecular PRRG2, RHBDL2, UCKL1-AS1) and ABO E6.1 sgRNA Citogenetics Laboratory, Hospital del Mar, Barcelona). (C16orf89, LINC02794, LZTS1, NLCN, SLC8A1) were Briefly, cultures of hiPSCs (70% confluent) were treated analysed by sanger sequencing. Sequences are available with KaryoMaxcolcemid (Invitrogen), dissociated, 4 of 13 PETAZZI et al. incubated in hypotonic solution and fixed in Carnoy differentiation stage (days 7–11), the basal medium was solution (75% methanol, 25% acetic acid). Karyotyping was supplemented with 10 ng/ml SCF and 3 U/ml EPO; in the performed following standard procedures. A minimum of third differentiation stage (days 11–21), basal medium was 15 metaphases were examined. supplemented with 3 U/ml EPO and additional transferrin to a final concentration of 500 μg/ml. Cells were cultured into stationary plastic tissue culture flasks at a density of 1– 2.8 RNA isolation and quantitative 3 × 105 cells/ml at the first stage, 3 × 105 – 1 × 106 cells/ml reverse-transcription polymerase chain at the second stage and 1–2 × 106 cells/ml at the third stage reaction by full media changes every other day starting on day 3. Total RNA was isolated from hiPSCs and developing embryoid body (EB) cells using the RNeasy Micro kit 2.10 Flow cytometry analysis (Qiagen) and treated with RNase-free DNase (Qiagen). Cell surface marker staining was performed by direct Total RNA (1 μg) was reverse transcribed using a high- immunofluorescence with conjugated monoclonal anti- capacity reverse transcription kit (Applied Biosystems). All bodies listed in Table S4. Briefly, a sample of 1 × 105 quantitative PCR analyses were performed using the Fast cells was incubated with conjugated antibodies in 50 μl SYBR Green Master Mix (Applied Biosystems) following of phosphate-buffered saline (PBS) (Lonza, USA) con- the manufacturer’s protocols on the Light Cycler 480 Real- Time PCR System (Roche). Gene-specific primers used for taining 1% (v/v) bovine serum albumin (BSA) (Grifols, this study are listed in Table S3. Spain) (PBS-BSA) for 30 min at room temperature (RT) with continuous mixing in the dark. Cells were washed twice with PBS-BSA and then analysed on a MACSQuant 2.9 Hematopoietic and erythroid flow cytometer using MACSQuantify software (Miltenyi differentiation Biotech, Germany). For RhAG staining, a sample of 5 × 105 cells was incubated with the monoclonal antibody LA1818 hiPSCs were differentiated into hematopoietic progeni- (kindly provided by Prof. Ellen van der Schoot, Sanquin, tor cells (HPCs), using the STEMdiff™Hematopoietic kit Netherlands) for 30 min at RT, and then labelled with (StemCell Technologies) following the manufacturer’s rec- FITC-conjugated goat anti-mouse IgG and IgM (Becton Dickinson) for 30 min at RT in the dark. Results were ommendations. The 12-day differentiation protocol was analysed on a MACSQuant system (Miltenyi Biotech). Cell performed in two stages. First, to induce hiPSC commit- viability was assessed by 7-AAD staining. ment towards mesoderm, hiPSC aggregates were plated on laminin-521-coated plates and cultured during the first 3 days with STEMdiff™ Hematopoietic Supplement A 2.11 Preparation of cytospins and added to the basal medium. Second, for the subsequent May-Grünwald Giemsa staining 9 days, mesodermal cells were further differentiated into HPCs using basal medium supplemented with STEMdiff™ Cytospins were prepared at the Hematological Cytology Hematopoietic Supplement B performing half-medium Service (Hospital del Mar, Barcelona). Briefly, a sample changes at days 5, 7 and 10 according to the manu- of 1 × 104 cells was prepared by centrifuging onto glass facturer’s instructions. At day 12, HPCs were harvested slides at 500 rpm for 10 min in a Thermo Scientific from the culture supernatant and re-cultured in erythroid Cytospin 4 cytocentrifuge. The slides were stained with differentiation media. May-Grünwald Giemsa stain (Merck) according to the To induce erythroid differentiation, HPCs were cultured Hematology Cytology Service’s protocol. Cytospins were in basal medium: IMDM (HyClone™, GE Healthcare Life imaged at 400× using an optical microscope. Sciences, USA) containing 1 U/ml Glutamine, 3% (v/v) human AB serum (Banc de Sang i Teixits, Barcelona, Spain), 2% (v/v) foetal bovine serum (Gibco, USA), 10 2.12 Helix pomatia agglutinin (HPA) μg/ml Insulin (Sigma, USA), 3 U/ml heparin (Sigma, Lectin fluorescence staining USA), 200 μg/ml Iron-Saturated Transferrin (R&D Sys- tems, USA) and 1U/ml penicillin/streptomycin (Pen/Strep, Expression of A antigen was analysed by direct fluo- Gibco) according to the protocol described by Griffiths rescence staining with 50 μg/ml AF488-conjugated helix et al.18 In the first differentiation stage (days 0–7), basal pomatia agglutinin (HPA) lectin (ThermoFisher) in living medium was supplemented with 10 ng/ml SCF, 1 ng/ml cultured erythroid cells. As positive and negative controls, IL-3 and 3 U/ml EPO (PeproTech, USA); in the second RBCs from individuals of A and O blood group types were PETAZZI et al. 5 of 13 fixed with 4% paraformaldehyde previous direct staining. detectable numerical or structural chromosomal abnor- Nuclei were stained with DAPI. Images were taken using malities (Figure 1B). The established Rhnull hiPSC lines Zeiss Axio Observer Z1 – Apotome inverted fluorescent displayed hallmarks of pluripotency, being positive for microscope and analysed using the Image J software.19 alkaline phosphatase staining (Figure 1C) and enhanced endogenous gene expression of common pluripotency markers (Figure 1D). The stemness of hiPSCs was also 2.13 Serological detection of Rh blood verified by immunofluorescence of pluripotency markers group antigens in hiPSC colonies from passages 8 to 15 (Figure S1A). Definitive proof of a pluripotent phenotype was shown in Bio-Rad DiaClon Rh-Subgroups+K ID-Cards with mon- in vitro-directed differentiation assays towards the three oclonal typing reagents for C (RH2), c (RH4), E (RH3), germinal layers (Figure S1B) and in in vivo teratoma for- e (RH5) were used for serological detection of RhCE mation assays (Figure S1C,D). Additionally, the presence antigens. Bio-Rad DiaClon ABD-Confirmation for Donors of the RHAG gene c.836G > A homozygous mutation was ID-Cards with monoclonal typing reagents for RhD: ESD- also confirmed by gene sequencing (Figure 1E). These 1 M and 175-2, were used for serological detection of the results demonstrate that we have successfully obtained an RhD antigen. Briefly, cell suspensions prepared from 1– integration-free and feeder-free hiPSC line carrying the 2 × 106 hiPSC-derived reticulocytes were pelleted and genotype that leads to the Rhnull phenotype in derived red resuspended in 50 μl of ID-Diluent 2 (Bio-Rad Labora- cells. tories, Switzerland). Cards were centrifuged as per the manufacturer’s instructions. A monoclonal anti-k (Cellano) reagent (Pelikloon IgM 3.2 Conversion of blood type A hiPSC monoclonal Lk1) was also used to detect Cellano anti- line to type O by CRISPR/Cas9-mediated gen expression using Bio-Rad NaCl, Enzyme test and Cold gene edition Agglutinins ID-Cards (Bio-Rad Laboratories). Fifty micro- liters of prepared cell suspension were added to a column The original A blood type of the Rhnull donor was asso- followed by 25 μl of the anti-Cellano antibody. Cards were ciated with a heterozygous A2/O1 ABO genotype, which centrifuged as per the manufacturer’s instructions. was also confirmed in the resultant hiPSC line by ABO gene sequencing (Figure S2). In order to convert the Rhnull hiPSC line from blood type A to the universal type O, we 3 RESULTS designed two strategies based on CRISPR/Cas9 technol- ogy. The first approach was based on the generation of 3.1 Establishment of an Rhnull a KI, mimicking the natural (c.261delG) polymorphism, donor-derived hiPSCs present in the most common inactive ABO*O.01 (O1) allele. This deletion of guanine in exon 6 causes a frameshift To obtain an integration-free Rhnull hiPSC line, PBMCs (p.Thr88Profs*31) in a sequence that otherwise is identi- from an Rhnull female blood donor were reprogrammed. cal to the consensus A sequence (Figure 2A). The second The donor subject had been previously identified as a approach relied on the generation of a KO targeting the homozygous carrier of a single-base mutation (c.836G > A) third exon of the ABO gene, also generating a frameshift. in the RHAG gene, leading to the rare Rhnull blood type In both approaches, three guide gRNA sequences per (ISBT RHAG Blood Group Alleles Table:https://www. CRISPR site were designed, transfected and previously isbtweb.org/static/5d593bb0-02e1-47a2-9a8fe3e34df68a5e/ tested on HEK293T cells to evaluate their cutting effi- ISBT030RHAGbloodgroupallelesv6230-NOV-2021.pdf. ciency through endonuclease T7 assay. In the KI strategy, PBMCs were reprogrammed using integration-free Sendai the best-performing gRNA, which specifically targeted virus vectors expressing OCT4, SOX2, KLF4 and cMYC the exon 6 (Figure 2B), was co-transfected with the donor under serum-free and feeder-free conditions. DNA sequence containing the 261G deletion, which is Two hiPSC lines were generated from this donor and used by the DNA repair machinery as the new template representative clones, named BST PBiPS6-SV4F-9 (abbre- after the cut. For the KO generation, the Rhnull hiPSC line viated as hiPSC#1) and BST PBiPS6a-SV4F-6 (abbreviated was transfected by electroporation with the RNP complex as hiPSC#2), were fully characterized. STR analysis con- whose gRNA targeted the ABO gene exon 3 (Figure 2C). firmed the identity of both lines when compared to the Different clones were isolated and screened by Sanger original PBMCs (Figure 1A). Both hiPSC lines robustly sequencing to confirm the ABO gene CRISPR edition. For proliferated for more than 20 passages, showing a nor- the KO strategy we checked 26 clones and found five with mal diploid female [XX, 46] karyotype, without any indels producing a truncated protein (19% efficiency). For 6 of 13 PETAZZI et al. F I G U R E 1 Characterization of the Rhnull hiPSC line carrying the c.836G > A mutation in the RHAG gene. (A) Short Tandem Repeat (STR) analysis demonstrating identical allelic profiles of the two Rhnull hiPSC clones in comparison with the original allelic profile of the donor. (B) Representative karyotype images of the two Rhnull hiPSC clones. (C) Representative picture of an AP-stained hiPSC colony from each clone, captured with an optical microscope. Scale bar 100 μm. (D) Expression levels of endogenous pluripotency genes determined by qPCR, normalized to the GAPDH housekeeping gene expression levels. Two-way ANOVA statistic analysis with Bonferroni posttest, ***p-value < 0.001. (E) RHAG gene exon 6 partial sequence showing the homozygous c.836G > A mutation in genomic DNA from the two Rhnull hiPSC clones, as well as in the original donor DNA sample the KI strategy, we identified five clones carrying the 261G Importantly, cell line identity of all four clones was con- deletion out of 16 clones screened (31% efficiency) (see firmed by STR analysis (Figure S3A) and also showed data at https://github.com/anasevilla/ABO-gene-editing). normal diploid [XX, 46] karyotypes (Figure S3B). Fur- We then selected two KO (KO-C31 and KO-C52) and thermore, gene-edited hiPSC clones remained pluripotent two KI (KI-C4 and KI-C5) clones for further charac- after CRISPR/Cas9 gene editing, retained hESC-like mor- terization (Figure 2B,C). Using the IDT design checker phology and expressed similar (p > 0.05) RNA (Figure (https://eu.idtdna.com/site/order/designtool/index/ S3C) and protein (Figure S3D,E) levels of the pluripo- CRISPR_SEQUENCE) software, we analysed the top tency markers SSEA3, SSEA4, TRA-1-60 and TRA-1-81 five in silico-predicted off-targets of ABO E3 sgRNA to the parental Rhnull hiPSC line (Figure S3F). In addi- (CCDC78, HTR5A, PRRG2, RHBDL2 and UCKL1- tion, their pluripotent capacity was tested in vitro through AS1) and ABO E6.1 sgRNA (C16orf89, LINC02794, directed differentiation into cell lineages representing all LZTS1, NLCN and SLC8A1) by sanger sequencing three germ layers through the embryoid body assay (Figure and found them all consistently unaltered in the S4A,B). Although expected differences across the iPSC four selected clones, demonstrating the specificity of lines were observed for the early lineage differentiation our gene editing strategy (sequences are available at markers (SOX17, T, TUJ1) in the differentiating embry- https://github.com/anasevilla/ABO-gene-editing). oid bodies, no statistical differences were observed for PETAZZI et al. 7 of 13 F I G U R E 2 CRISPR/Cas9-mediated gene edition strategies to convert hiPSCs of blood type A to type O. (A) Structure of the ABO gene, located on chromosome 9q34.2. The O allele differs from the A allele by a single nucleotide deletion of guanine (G) at position 261(c.261delG). This deletion leads to a truncated, nonfunctional protein without transferase activity. Adapted from Hosoi et al., 200820 . (B) KI approach. ABO exon 6 partial sequence confirming homozygosity for the c.261delG deletion in C4 and C5 KI clones. (C) KO approach. Partial sequence showing the region of the ABO gene exon 3 containing a 13 bp deletion (c.135-147del) in KO clone 31 and a 20 bp deletion (c.135-154del) in KO clone 52. The region matching the RNA guide used to induce the knock-out is coloured in blue the expression of the pluripotency markers OCT4, SOX2 towards HPCs with the STEMDiff™ Hematopoietic Kit and NANOG at the pluripotency state between gene-edited (Figure S5A). At day 12 of the differentiation protocol, iPSC lines and the parental lines (Figure S4C). Thus, we HPCs released from hematopoietic clusters were harvested have established two CRISPR/Cas9-mediated gene edition from the culture supernatant (Figure S5B). This popu- strategies to convert blood type A hiPSC lines to type O lation contained around 90% CD34+ cells, and around with no impact on the stemness potential. 60% of these cells were CD45low/+ (Figure S5C). To fur- ther characterize the CD34+ CD45low/+ fraction, we also analysed the expression of the erythroid lineage surface 3.3 Morphological changes and markers CD71, CD235a, CD49d and CD233. We confirmed immunophenotype confirm consistent the presence of a variable range (36–75%) of early ery- erythroid differentiation in ABO-edited throblasts CD71+ CD235a+ CD49d+ CD233− (Figure S5C). hiPSC lines The collected cells, containing erythroid-committed HPCs, were further cultured in erythroid differentiation medium The potential of the ABO-edited hiPSC lines to differ- according to the three-step protocol described in Materials entiate towards the erythroid lineage was evaluated in and Methods (Figure 3A). The follow-up of cell viabil- parallel with the parental Rhnull hiPSCs in three inde- ity and expansion throughout the culture showed no pendent experiments. The parental Rhnull hiPSCs and the statistically significant differences between the edited and KI-C5 and KO-C52-edited clones were first differentiated the parental cell lines regarding the survival/proliferative 8 of 13 PETAZZI et al. F I G U R E 3 Differentiation of hiPSC-derived CD34+ cells to the erythroid lineage. (A) Schematic representation of the erythroid differentiation protocol. (B) Representative images of different erythroblast maturation steps from cytospins prepared at days 0, 7, 14 and 21, using May-Grunwald Giemsa staining. Scale bar: 10 μm. (C) Percentage representation of the different stages of erythroid maturation (pro-, basophilic, polychromatic and orthrochromatic erythroblasts and reticulocytes) on different days of the erythroid culture (days 0, 7, 14 and 21). (n = 3, mean ± SD, Two-way ANOVA statistics with Bonferroni post-test, P-value > .05) (D) Flow cytometry analysis of the cell surface markers: CD44, CD49d, CD71, CD235a, CD233 and CD238, during erythroid differentiation at day 0 (grey), day 7 (pink), day 14 (dark pink) and day 21 (red). The ordinate represents the number of cells displaying the fluorescence intensity given by the abscissa Abbreviations: EPO, erythropoietin; ERY, erythroblasts; IL3, interleukin-3; SCF, stem cell factor. PETAZZI et al. 9 of 13 capacity of hiPSC-derived erythroid progenitors (Figure ducing the Rhnull phenotype of the original donor subject, S6A,B). Distinct stages of erythroid maturation were which has neither been affected by the reprogramming of assessed morphologically at four time points (d0, d7, the donor’s PBMCs, nor by the CRISPR edition of the ABO d14 and d21) (Figure 3B). The erythroid cells progressed gene. through distinct erythroid stages with orthochromatic erythroblasts already appearing at day 7, showing no statis- tically significant differences between the parental Rhnull 3.5 ABO blood group conversion in hiPSC and the edited clones, KI-C5 and KO-C52, across erythroid cells differentiated from the 21-day differentiation period (Figure S6C). At day 21, ABO-edited iPSC lines orthochromatic erythroblasts were the predominant cells (approximately 80%), with a very low proportion of enu- To analyse A antigen expression in differentiated erythroid cleated cells (6–8%), in both the parental Rhnull hiPSC cells from parental Rhnull hiPSCs and edited clones, fluo- line and the edited clones (Figure 3C). Moreover, ery- rescence labelling was performed using HPA, a lectin that throid differentiation was also assessed by flow cytometry has anti-A human blood group specificity. The erythroid immunophenotyping of erythroid surface markers: CD44, cells derived from ABO-edited clones were negative for A CD49d (α4-integrin), CD71, CD235a (glycophorin A, GPA), antigen expression in flow cytometry studies (Figure 4C). CD233 (Band3) and CD238 (KEL). The observed dynamic Similarly, no differences in H antigen expression were changes of expression revealed an analogous progression detected between the parental Rhnull hiPSCs, express- through erythroid differentiation in parental Rhnull hiPSC ing blood type A2 , and the edited lines (blood type O) line and the edited clones. In brief, we observed two dis- (Figure 4D). This result indicates that H antigen, which tinct patterns of expression (Figure 3D). The adhesion is the precursor for A and B antigen synthesis, is likewise molecules CD44 and CD49d, as well as the transferrin expressed in erythroid cells differentiated from both the receptor CD71 presented a pattern with high levels of Rhnull and edited hiPSCs. expression in early-stage erythroblasts and a progressive In the parental Rhnull hiPSC line, we observed early decrease in late-stage erythroblasts. In contrast, a dif- A antigen expression at day 7 of erythroid differentiation ferent pattern was observed for the CD235a, CD233 and and its maintenance until day 21 (Figure 4E). As expected, CD238 markers, which displayed low levels of expression though, cultured red cells showed weak A antigen expres- in early-erythroblasts with a progressive increase in late- sion, in agreement with the original A2 subgroup of the stage erythroblasts. Our results concur with the expected Rhnull donor (Figure S7).24 In contrast, we could not detect progression of erythroid cell differentiation cultures from A antigen-labelled cells in those cultures differentiated hiPSC-derived CD34+ HPCs,21–23 with no significant dif- from any of the edited clones. These results confirm the ferences between the parental Rhnull hiPSCs and the edited successful conversion of blood type A to blood type O using clones (Figure S6D). CRISPR-Cas9 editing strategies in hiPSCs carrying the rare Rhnull blood group. 3.4 The Rhnull phenotype is retained in erythroid cells differentiated from parental 4 DISCUSSION and ABO-edited Rhnull hiPSCs Red cells with rare blood types are currently in limited To confirm the Rhnull phenotype, we first assessed RhAG supply due to the scarce representation of these pheno- expression by flow cytometry, in cells differentiated from types in the global population. The provision of compatible both the parental and ABO-edited hiPSCs, using the red cells for the transfusion of immunized patients car- LA1818 anti-RhAG monoclonal antibody. No RhAG expres- rying rare blood types is one of the first potential target sion was observed on the membrane of erythroid cells applications of human red cells manufactured in vitro. On derived either from the parental Rhnull hiPSCs or from the the other hand, the diagnosis of red cell alloimmuniza- ABO-edited clones KI-C5 and KO-C52 (Figure 4A). As the tion, crucial to ensure the safe transfusion of immunized RhAG glycoprotein is essential for the Rh complex forma- patients, relies on using carefully selected reagent red cells tion, we next assessed the expression of the Rh antigens with well characterized phenotypes. These red cells are (D, C, c, E and e) by agglutination tests using gel card tech- also obtained from blood donors, so the availability of nology, and no agglutination was observed with any of the infrequent phenotypes needed to properly identify rare anti-Rh typing reagents (Figure 4B), further confirming antibody specificities is likewise very limited. In this sense, the Rhnull phenotype. These data confirm the successful having an alternative (unlimited) cell source to derive red generation of in vitro differentiated erythroid cells repro- cells with rare blood types in vitro, could potentially over- 10 of 13 PETAZZI et al. F I G U R E 4 Analysis of the Rh and ABO blood group phenotype. (A) Flow cytometry analysis of RhAG expression in cultured red cells from the parental (hiPSC#1) and edited (KI-C5 and KO-C52) lines, at day 21 of erythroid differentiation. RhD+ RBCs were used as a positive control. Mouse IgG1, κ antibody (BD Pharmingen, USA) was used as isotype control. (B) Gel card Rh typing results of cultured red cells from KO-C52 line showing no agglutination with monoclonal typing reagents for the D (RH1), C (RH2), c (RH4), E (RH3), e (RH5) antigens. Cellano antigen (k) expression is detected in a parallel sample of cultured red cells from the same cell line. (C) Flow cytometry analysis of HPA-FITC at day 21 of erythroid differentiation. Blood group O RBCs (RBC-O) and blood group A2 RBCs (RBC-A2) were used as reference. (D) Flow cytometry analysis of H antigen (CD173) at day 21 of erythroid differentiation. Blood group A2 RBCs were used as control. (E) Optical microscope images showing specific binding of HPA-FITC to the cell membrane of differentiated erythroid cells from the parental hiPSC#1 cell line, whereas no labelling was observed in erythroid cells from the edited KI-C5 and KO-C52 lines. Nuclei were stained with DAPI. Scale bars: 10 μm (day 7) and 5 μm (day 21) PETAZZI et al. 11 of 13 come the current rare blood limitations in both transfusion tional sugar residue (GalNAc or Gal, respectively) on the and diagnostics. precursor H-antigen found on type O RBCs. The concept of One of the approaches that have been considered to removing these immunogenic sugars by specific enzymes address this issue, is the use of hiPSCs obtained from exist- (glycosidases) from blood type A or B red cells, was first ing donors or patients with rare blood types.25,26 Of course, proposed and demonstrated by Goldstein (1982).33 The such donors are not easily available in practice, as rare first attempts required massive amounts of enzyme but phenotypes are usually found in less than 1 per 1,000 in novel α-galactosidases and α-N-acetylgalactosaminidases the general population, and donors with exceptional ‘null’ have been shown to improve the conversion efficiency.34 phenotypes are even much less represented. One such However, this technology has not yet moved into clinical example is the H deficiency, also known as the Bombay practice, as there are hold-ups pending to be solved.32 (Oh ) phenotype, which is found in 1 in 10,000 individu- Alternatively, we addressed the conversion of the blood als in India. The generation of hiPSCs from the dermal type A Rhnull hiPSC line into universal type O using fibroblasts of a Bombay blood-type individual14 provided CRISPR/Cas9-mediated gene editing technology, which the first proof of concept for this approach. More recently, allows the precise, robust and efficient edition of genes of hiPSC lines have been obtained by reprogramming ery- interest.13 With the aim to abrogate the expression of the α- throid progenitors from peripheral blood of individuals 1,3-N-acetylgalactosamine transferase (A-transferase), we with the Jr(a−) and D− rare blood types,15 demonstrat- designed two different ABO gene edition approaches. The ing the feasibility of producing autologous hiPSC-derived first approach was based on the generation of a KI, mimick- red cells for the transfusion of patients with rare blood ing the c.261delG single nucleotide deletion, present in the groups. However, the potential utility of hiPSC-derived red most common inactive ABO*O.01 (O1) allele.35 The specific cells with null phenotypes or infrequent antigen combina- and precise incorporation of this c.261delG polymorphism tions, extends beyond an autologous use. Such hiPSC lines within the ABO gene has been attempted in the present could provide cultured red cells with difficult-to-supply work aiming to reproduce the genetic basis naturally blood types for the transfusion of certain groups of immu- associated to blood type O. Exploiting the CRISPR/Cas9- nized patients (e.g., sickle cell disease patients). Likewise, targeted integration to correct genetic defects has led to hiPSC lines could solve the limited availability of reagent- a number of proof-of-principle works in patient-derived red cells with rare phenotypes, which are also necessary for hiPSCs, in which the mutations responsible for cystic fibro- the identification of rare RBC antibody specificities. sis, haemophilia A and β-thalassemia were successfully In this study, we present the generation of a hiPSC line corrected, although with a limited efficiency.36–38 Thus, derived from an Rhnull donor with blood type A. The donor a second approach by gene KO was undertaken in paral- subject had been previously identified as a homozygous lel to maximize the possibilities to achieve our final goal, carrier of the c.836G > A single-base mutation in the RHAG which was to obtain an Rhnull hiPSC line converted to gene, leading to the rare Rhnull blood type.27–29 The Rh- universal blood type O. Indeed, both the KI and the KO blood group deficiency, or Rhnull , is an extremely rare strategies successfully rendered hiPSC-edited clones with phenotype which lacks all the Rh antigens on the red cell the intended ABO gene modifications, as demonstrated by membrane. Such valuable blood is necessary for the trans- ABO sequencing analysis. Moreover, the established ABO- fusion support of Rh immunized patients, not only those edited hiPSCs lines maintained the Rhnull -related RHAG with Rhnull phenotype but also patients with antibodies gene mutation as well. against any high-prevalence Rh specificity, for whom com- The results obtained in the characterization of the KI patible blood is always difficult to procure.3 Nonetheless, and KO-edited hiPSC lines showed no changes on their the potential use of hiPSC-derived red cells for both, trans- stemness potential. Likewise, these lines have been suc- fusion and diagnostics, is also dependent on ABO compat- cessfully differentiated into HPCs and, subsequently, to ibility. Extremely infrequent or null blood group types are the erythroid lineage. No remarkable differences have not necessarily encountered in blood type O donors, as it been observed between the parental Rhnull hiPSCs line is the case in this blood type A Rhnull donor. This circum- and the edited clones in erythroid differentiation experi- stance limits the potential use of the hiPSC-derived red ments, with overall results concording with the expected cells due to the naturally occurring ABO hemagglutinins. progression of erythroid cell differentiation cultures from To overcome this limitation, we considered the conversion hiPSC-derived CD34+ progenitor cells.21–23 of blood type A to universal type O. It is worth noting that we have been able to produce The conversion of blood group types A and B to uni- differentiated erythroid cells reproducing the Rhnull phe- versal type O has been pursued for a long time through notype, proving no alteration of the original donor’s rare approaches based on enzymatic treatment.30–32 Blood phenotype due to the PBMCs reprograming or to the sub- types A and B differ from type O in the presence of an addi- sequent hiPSCs CRISPR ABO-gene edition. Remarkably, 12 of 13 PETAZZI et al. the results obtained from both KI and KO gene edition ORCID strategies provide the first demonstration of blood type Ana Sevilla https://orcid.org/0000-0002-9251-4759 A conversion to the universal type O using CRISPR/Cas9 technology. The knock-out of specific blood group genes, REFERENCES other than ABO, in pre-existing hiPSC lines has been 1. Greinacher A, Fendrich K, Brzenska R, Kiefel V, Hoffmann recently reported as an strategy to reproduce uncommon W. Implications of demographics on future blood supply: a null phenotypes.16 Here, we demonstrate the feasibility population-based cross-sectional study. Transfusion 2011; 51(4): of robust and sustainable ABO blood type conversion 702-709. doi: 10.1111/j.1537-2995.2010.02882.x 2. Williamson LM, Devine DV. Challenges in the management of using these newly designed CRISPR/Cas9 gene editing the blood supply. Lancet 2013; 381(9880): 1866-1875. doi: 10.1016/ approaches, allowing the production of cultured red cells S0140-6736(13)60631-5 with improved ABO compatibility. The potential applica- 3. Nance S, Scharberg EA, Thornton N, Yahalom V, Sareneva I, tion of these approaches is not restricted to hiPSC lines, Lomas-Francis C. International rare donor panels: a review. Vox since they can also be applied to other cell lines of inter- Sang 2016; 110 (3): 209-218. doi: 10.1111/vox.12357 est for cultured red cells production, such as immortalized 4. Muñiz-Diaz E, Castro A, Flores E, et al. The Spanish program for human erythroblast cell lines,5,39 derived from individuals rare blood donors. Immunohematology 2016; 32(2): 59-61. https:// www.ncbi.nlm.nih.gov/pubmed/27901570 not carrying blood type O. 5. Trakarnsanga K, Griffiths RE, Wilson MC, et al. An immor- During the past decade, significant advances have been talized adult human erythroid line facilitates sustainable and made in the production of manufactured red cells from scalable generation of functional red cells. Nat Commun. 2017; different cell sources.7,9,40,41 Despite the known limita- 8: 14750. doi: 10.1038/ncomms14750 tions that still need to be overcome (e.g., low enucleation 6. Heshusius S, Heideveld E, Burger P, et al. Large-scale in vitro rate and cost-efficient scaling), the deeper knowledge of production of red blood cells from human peripheral blood the regulatory pathways involved in terminal erythroid mononuclear cells. Blood Adv. 2019; 3(21): 3337-3350. doi: 10.1182/ differentiation, together with the continuous progress in bloodadvances.2019000689 7. Seo Y, Shin KH, Kim HH, Kim HS. Current advances in red scaled-up protocols and technological achievements, allow blood cell generation using stem cells from diverse sources. Stem to anticipate that in vitro production of RBCs will be Cells Int. 2019; 2019: 9281329. doi: 10.1155/2019/9281329 possible in the near future. In this context, CRISPR/Cas9- 8. Pellegrin S, Severn CE, Toye AM. Towards manufactured red mediated blood group gene edition will certainly play blood cells for the treatment of inherited anemia. Haematologica an important role as a tool to improve blood group 2021;106(9): 2304-2311. doi: 10.3324/haematol.2020.268847 compatibility, like this work demonstrates. 9. Sivalingam J, SuE Y, Lim ZR, et al. A scalable suspension plat- form for generating high-density cultures of universal red blood cells from human induced pluripotent stem cells. Stem Cell Rep. AC K N OW L E D G E M E N T S 2021; 16(1): 182-197. doi: 10.1016/j.stemcr.2020.11.008 This research was funded by the Spanish Ministry of Sci- 10. Bueno C, Sardina JL, Di Stefano B, et al. Reprogramming human ence, Innovation and Universities (RETOS-Colaboración, B cells into induced pluripotent stem cells and its enhancement Ref: RTC-2017-6367-1). We thank CERCA/Generalitat de by C/EBPα. Leukemia 2016; 30(3): 674-682. doi: 10.1038/leu.2015. Catalunya and Fundació Josep Carreras-Obra Social la 294 Caixa for their institutional support. Financial support 11. Gu H, Huang X, Xu J, et al. Optimizing the method for gen- eration of integration-free induced pluripotent stem cells from for this work was obtained from the Catalan Goverment human peripheral blood. Stem Cell Res Ther 2018; 9(1): 163. doi: (SGR180 and PERIS 2017–2019). AS was supported by the 10.1186/s13287-018-0908-z project (PGC2018-098626-B100) funded by the MCIN/AEI 12. Romero-Moya D, Santos-Ocaña C, Castaño J, et al. Genetic res- /10.13039/501100011033/ ERDF ‘A way of making Europe’ cue of mitochondrial and skeletal muscle impairment in an and the Merck Salud Foundation. AS is a recipient of a induced pluripotent stem cells model of coenzyme Q. Stem Cells. Ramón y Cajal contract (RYC-2016-19962) funded by the 2017; 35(7): 1687-1703. doi: 10.1002/stem.2634 MCIN/AEI /10.13039/501100011033 and ‘ESF Investing in 13. Petazzi P, Menéndez P, Sevilla A. CRISPR/Cas9-mediated gene knockout and knockin human iPSCs. Methods Mol Biol. 2022; your Future’. 2454:559-574. doi: 10.1007/7651_2020_337 14. Seifinejad A, Taei A, Totonchi M, et al. Generation of human CONFLICTS OF INTEREST induced pluripotent stem cells from a Bombay individual: The authors declare no competing interests. moving towards “universal-donor red blood cells.” Biochem Biophys Res Commun 2010; 391(1): 329-334. doi: 10.1016/j.bbrc. 2009.11.058 D A T A AVA I L A B I L I T Y S A T A T E M E N T 15. Park YJ, Jeon SH, Kim HK„ et al. Human induced pluripotent The data that support the findings of this study are stem cell line banking for the production of rare blood type ery- openly available at https://github.com/anasevilla/ABO- throcytes. J Transl Med. 2020; 18(1): 236. doi: 10.1186/s12967-020- gene-editing and in the Supplementary Files. 02403-y PETAZZI et al. 13 of 13 16. Pandey P, Zhang N, Curtis BR, Newman PJ, Denomme GA. Gen- 31. Olsson ML, Clausen H. Modifying the red cell surface: towards eration of ‘designer erythroblasts’ lacking one or more blood an ABO-universal blood supply. Br J Haematol 2008; 140(1): 3-12. group systems from CRISPR/Cas9 gene-edited human-induced doi: 10.1111/j.1365-2141.2007.06839.x pluripotent stem cells. J Cell Mol Med. 2021; 25(19):9340-9349. 32. Rahfeld P, Withers SG. Toward universal donor blood: enzy- doi: 10.1111/jcmm.16872 matic conversion of A and B to O type. J Biol Chem 2020;295(2): 17. Takahashi K, Okita K, Nakagawa M, Yamanaka S. Induction of 325-334. doi: 10.1074/jbc.REV119.008164 pluripotent stem cells from fibroblast cultures. Nat Protoc 2007; 33. Goldstein J, Siviglia G, Hurst R, Lenny L., Reich L. Group B ery- 2(12): 3081-3089. doi: 10.1038/nprot.2007.418 throcytes enzymatically converted to group O survive normally 18. Griffiths RE, Kupzig S, Cogan N, et al. Maturing reticulocytes in A, B, and O individuals. Science. 1982;215(4529):168-170. doi: internalize plasma membrane in glycophorin A-containing vesi- 10.1126/science.6274021 cles that fuse with autophagosomes before exocytosis. Blood 34. Liu QP, Sulzenbacher G, Yuan H, et al. Bacterial glycosidases for 2012; 119(26): 6296-6306. doi: 10.1182/blood-2011-09-376475 the production of universal red blood cells. Nat Biotechnol 2007; 19. Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25(4): 454-464. doi: 10.1038/nbt1298 25 years of image analysis. Nat Methods 2012; 9(7): 671-675. doi: 35. Yamamoto F, Clausen H, White T, Marken J, Hakomori S. 10.1038/nmeth.2089 Molecular genetic basis of the histo-blood group ABO system. 20. Hosoi E. Biological and clinical aspects of ABO blood group Nature 1990; 345(6272): 229-233. doi: 10.1038/345229a0 system. J Med Invest. 2008;55(3-4):174-182. doi: 10.2152/jmi.55.174 36. Firth AL, Menon T, Parker GS, et al. Functional gene correction 21. Chen K, Liu J, Heck S, Chasis JA, An X, Mohandas N. Resolving for cystic fibrosis in lung epithelial cells generated from patient the distinct stages in erythroid differentiation based on dynamic iPSCs. Cell Rep. 2015; 12(9): 1385-1390. doi: 10.1016/j.celrep.2015. changes in membrane protein expression during erythropoiesis. 07.062 Proc Natl Acad Sci U S A 2009; 106 (41): 17413-17418. doi: 10.1073/ 37. Park CY, Kim DH, Son JS, et al. Functional correction of pnas.0909296106 large factor VIII gene chromosomal inversions in hemophilia a 22. Ackermann M, Liebhaber S, Klusmann JH, Lachmann N. Lost in patient-derived iPSCs using CRISPR-Cas9. Cell Stem Cell 2015; translation: pluripotent stem cell-derived hematopoiesis. EMBO 17(2): 213-220. doi: 10.1016/j.stem.2015.07.001 Mol Med 2015; 7(11): 1388-1402. doi: 10.15252/emmm.201505301 38. Liu Y, Yang Y, Kang X, et al. One-step biallelic and scarless 23. Bernecker C, Ackermann M, Lachmann N, et al. Enhanced ex correction of a β-thalassemia mutation in patient-specific iPSCs vivo generation of erythroid cells from human induced pluripo- without drug selection. Mol Ther Nucleic Acids 2017; 6: 57-67. doi: tent stem cells in a simplified cell culture system with low 10.1016/j.omtn.2016.11.010 cytokine support. Stem Cells Dev. 2019;28(23): 1540-1551. doi: 10. 39. Hawksworth J, Satchwell TJ, Meinders M, et al. Enhancement of 1089/scd.2019.0132 red blood cell transfusion compatibility using CRISPR-mediated 24. Tulliez M, Villeval JL, Lejeune F, et al. Expression of blood group erythroblast gene editing. EMBO Mol Med. 2018; 10(6): e8454. A antigen during erythroid differentiation in A1 and A2 sub- doi: 10.15252/emmm.201708454. https://www.ncbi.nlm.nih.gov/ jects. Leukemia 1987; 1(1): 44-51. https://www.ncbi.nlm.nih.gov/ pubmed/29700043 pubmed/3669734 40. Singh VK, Saini A, Tsuji K, Sharma PB, Chandra R. Manu- 25. Peyrard T, Bardiaux L, Krause C, et al. Banking of pluripotent facturing blood ex vivo: a futuristic approach to deal with the adult stem cells as an unlimited source for red blood cell pro- supply and safety concerns. Front Cell Dev Biol 2014;2: 26. doi: duction: potential applications for alloimmunized patients and 10.3389/fcell.2014.00026 rare blood challenges. Transfus Med Rev. 2011; 25 (3): 206-216. 41. Christaki EE, Politou M, Antonelou M, et al. Ex vivo generation doi: 10.1016/j.tmrv.2011.01.002 of transfusable red blood cells from various stem cell sources: a 26. Manian KV, Bharathan SP, Maddali M, Srivastava VM, concise revisit of where we are now. Transfus Apher Sci. 2019; Srivastava A, Velayudhan SR. Generation of an integration-free 58(1): 108-112. doi: 10.1016/j.transci.2018.12.015 iPSC line (CSCRi005-A) from erythroid progenitor cells of a healthy Indian male individual. Stem Cell Res. 2018; 29: 148-151. doi: 10.1016/j.scr.2018.03.020 S U P P O RT I N G I N F O R M AT I O N 27. Vos GH, Vos D, Kirk RL, Sanger R. A sample of blood with no Additional supporting information can be found online detectable Rh antigens. Lancet 1961;1(7167): 14-15. doi: 10.1016/ s0140-6736(61)92183-3 in the Supporting Information section at the end of this 28. Huang CH, Liu Z, Cheng G, Chen Y. Rh50 glycoprotein article. gene and rhnull disease: a silent splice donor is trans to a Gly279→Glu missense mutation in the conserved transmem- brane segment. Blood 1998;92(5): 1776-1784. https://www.ncbi. nlm.nih.gov/pubmed/9716608 How to cite this article: Petazzi P, Miquel-Serra 29. Hyland CA, Chérif-Zahar B, Cowley N, et al. A novel single mis- L, Huertas S, et al. ABO gene editing for the sense mutation identified along the RH50 gene in a composite conversion of blood type A to universal type O in heterozygous Rhnull blood donor of the regulator type. Blood Rhnull donor-derived human-induced pluripotent 1998; 91(4): 1458-1463. stem cells. Clin Transl Med. 2022;12:e1063. 30. Goldstein J. Conversion of ABO blood groups. Transfus Med Rev. https://doi.org/10.1002/ctm2.1063 1989; 3(3): 206-212. doi: 10.1016/s0887-7963(89)70080-8