HLA-homozygous hiPSC haplobank for the Spanish population: generation, GMP expansion and population coverage (Kuebler et al., 2023)
Background
Stem cell therapies using induced pluripotent stem cells (iPSCs) offer a route to regenerative medicine and have potential for broad application. A key cost driver is generating patient-specific iPSCs; an alternative is allogenic, homozygous HLA iPSC lines to create a clinical-grade haplobank that covers a large fraction of the population.
The study reports the generation of the first clinical-grade, HLA-homozygous iPSC bank (haplobank) in Spain using cord blood units (CBUs) that are homozygous for the most common HLA-A, HLA-B and HLA-DRB1 haplotypes.
Source material: CD34+ hematopoietic progenitors isolated from frozen CBUs; these CD34+ cells were reprogrammed using Sendai virus vectors to iPSCs under good manufacturing practice (GMP) conditions.
Aim: to create master cell banks (MCBs) and working cell banks (WCBs) of seven haplotyped iPSC lines, expanded and banked under GMP, to support advanced therapy medicinal product (ATMP) development and off-the-shelf cell therapies.
This haplobank is intended as starting material for ATMP development and as a foundation for European/international haplobanks through shared haplotype resources.
Licensing and ethics: All donor material was used under project-specific informed consent, with ethics approval and adherence to regulatory guidelines.
Key concepts and definitions
iPSC: human-induced pluripotent stem cells; somatic cells reprogrammed to a pluripotent state.
Haplobank: a bank of iPSC lines that are homozygous for common HLA haplotypes to maximize population coverage and minimize immune rejection in allogeneic therapies.
HLA haplotypes: combinations of alleles at HLA-A, HLA-B, and HLA-DRB1 (and related loci) that determine immune compatibility.
CD34+ cells: hematopoietic progenitors used as starting material for reprogramming to hiPSCs.
GMP (Good Manufacturing Practice): regulatory framework ensuring quality and safety in manufacturing biologics and cell therapies.
ATMP (Advanced Therapy Medicinal Product): regulatory category for cell- and gene-based therapies.
Sendai virus (SeV) reprogramming: non-integrating viral method used to reprogram somatic cells to iPSCs; transient and eventually cleared from cells.
MCB/WCB: Master Cell Bank and Working Cell Bank; layers of cell banking used for consistent manufacturing.
QC (Quality Control): a set of assays to ensure sterility, purity, identity, genetic stability, and absence of contaminants.
STR (Short Tandem Repeat) profiling: genetic fingerprinting to confirm cell line identity.
EB (Embryoid Body) formation: in vitro assay to test differentiation potential into three germ layers.
0MM, 1MM, 2MM: zero, one, or two mismatches tolerated in matching at the HLA loci during coverage calculation.
Haplobank design and population coverage
Donor source and haplotype selection:
Cord blood units were screened from the Spanish Stem Cell Transplant Registry for homozygosity at the most common HLA haplotypes (HLA-A, HLA-B, HLA-DRB1).
Seven donors were selected to maximize population coverage of Spain, achieving 21.37% coverage under zero-mismatch (0MM) criteria.
The haplotypes provide the widest possible coverage of the Spanish population based on the dataset used.
Donors’ cord blood units were quality verified and donors provided informed consent for the use of CB units to generate hiPSC haplolines.
Population coverage calculation:
Population cohort: combined data from adult bone marrow donors and cord blood donors in REDMO, at four-digit resolution for 56{,}798 individuals.
Method: iterative, in-house algorithm in Microsoft Access. Brief steps:
Step 1: Check the first haplotype against the whole cohort and extract all matched individuals.
Step 2: Recompute coverage for the next haplotype using the remaining data.
Step 3: Repeat across haplotypes, accumulating matches.
Step 4: To compute “beneficial match” (allowing 1 or 2 mismatches in any locus), repeat the same strategy but tolerate one or two mismatches.
Coverage metric:
For any given level of mismatch allowance, cumulative coverage = (matched individuals in all iterations) / N × 100, where N = 56{,}798.
Reported coverage for the seven haplotypes (Spain):
0MM (no mismatches): 21.37 ext{%}
1MM (allowing one mismatch): 50.83 ext{%}
2MM (allowing two mismatches): 92.46 ext{%}
Haplotypes and donor details (Table 1 overview):
Seven haplo-homozygous haplotype lines are designated as CBiPS1–CBiPS8, each associated with a donor and specific HLA-A, HLA-B, HLA-C, HLA-DRB1, HLA-DQB1, HLA-DPB1 alleles; ABO blood group and donor sex are recorded.
Examples of haplotype lines and their donor metadata were presented (precise allele values per line are provided in the study’s Table 1).
Rationale and context:
Cord blood was chosen because it is routinely collected, HLA-typed, has low risk to mother/child at collection, and harbors hematopoietic progenitors suitable for iPSC generation.
CB banks often retain surplus units suitable for iPSC derivation; cost and time reductions are expected with haplobanks.
Ethical/regulatory context:
CBUs were provided from the clinical inventory with reconsenting per IPS-PANIA project; ethics committee approval (ID ethics committee: PR(AG)428/2018).
The haplobank is intended to serve as starting materials for ATMP development under GMP, ensuring traceability and regulatory compliance.
Donor selection and cord blood processing
Donor and sample selection:
CB donors homozygous for the most frequent HLA-A, HLA-B, and HLA-DRB1 haplotypes were chosen to maximize coverage of the Spanish population.
Donors signed consent for use of donated CB for hiPSC haplotype derivation.
Cord blood processing:
CB units thawed, osmotically equilibrated, and mononuclear fraction isolated by Ficoll density gradient in closed Sepax2 system.
CD34+ cells isolated using clinical-grade anti-CD34 magnetic beads and magnetic columns (Miltenyi Biotec).
Reprogramming and hiPSC generation
CD34+ cell expansion:
Expanded in StemPro 34 Serum Free Medium with cytokines for 4 days: 50 ng/mL SCF, 50 ng/mL FLT3L, 10 ng/mL TPO, 10 ng/mL IL-6; 37°C, 5% CO2; medium changed every other day.
Reprogramming with Sendai virus kits:
Transduction with CTS CytoTune-iPS 2.1 Sendai Reprogramming Kit at 1×10^4 cells per transduction in SP34 SFM with cytokines and polybrene.
Residual CD34+ cells cryopreserved; next day, infected cells seeded on Biolaminin-521 CTG-coated dishes in SP34 SFM with cytokines.
Day 6: switch to Essential 8 Flex Medium; monitor for colonies; emergence of reprogrammed colonies typically around 16–18 days after infection.
Colonies picked and seeded on CTG laminin-coated dishes in Essential 8 Flex (+ RevitaCell) for expansion.
Upon colony rise and single-clone picking, Sendai virus-infected cells were frozen for later testing.
Absence of Sendai virus confirmed by RT-PCR: RNA extracted, cDNA synthesized, PCR performed to detect Sendai genome; positive control included.
Genomic integrity check by G-banding karyotype performed on two clones; one virus-free clone with normal karyotype selected per line for expansion.
Objective: produce clones free of Sendai virus and transgenes with euploid karyotype.
Culture, cryopreservation, and clone selection
hiPSC culture and cryopreservation:
Clones passaged with CTS DPBS (no Ca/Mg) and CTS Versene; seeded on CTG LN-521-coated plates in Essential E8 Flex.
Cryopreservation: cells pelleted, resuspended in CTS PSC Cryo-medium with RevitaCell, frozen at −80°C, then transferred to liquid nitrogen.
Clone selection criteria:
Absence of Sendai virus genome and transgenes confirmed by RT-PCR around passages 7–8 with 15 days at 39°C.
Normal karyotype confirmed by G-banding.
One clone per line selected for full characterization and banking after screening 2 virus-free clones.
GMP generation of MCBs and WCBs; downstream QC
GMP expansion to create MCBs and WCBs:
MCBs produced under GMP in classified facilities in a clean room with GMP-grade reagents.
Previous steps were assessed for sterility (BacTAlert), mycoplasma (qPCR), and adventitious viruses.
For banking: two vials per line, thawed and expanded to 70–80% confluence, then split into 6–8 plates per line; after expansion, cells cryopreserved in 40–50 vials with ≥1×10^6 viable cells per vial.
After 24 h at −80°C, cryovials moved to LN2 storage.
WCB production:
Two MCB vials thawed, expanded, and frozen following the same procedure as MCB.
Characterization workflow (applied to MCB and WCB):
Immunocytochemistry (ICC) for pluripotency markers (Nanog, OCT4, SOX2, TRA-1-81, TRA-1-60, SSEA3, SSEA4).
Alkaline phosphatase (AP) staining.
In vitro differentiation via embryoid body (EB) formation and ICC for endoderm, ectoderm, and mesoderm markers.
Short tandem repeats (STR) analysis for identity.
Mycoplasma testing at early passages and before banking; karyotyping repeated after banking.
MCB/WCB quality controls:
Sterility (Ph. Eur. 2.6.27), mycoplasma (2.6.7), endotoxin (2.6.14), adventitious virus testing.
Viability before freezing and recovery after thawing (7AAD viability assay and colony-forming ability after thaw).
Acceptance criteria included viable cell numbers, recovery, and confluence metrics as per Table 3 (details below).
Characterization and identity testing
Immunocytochemistry for pluripotency:
Markers tested: Nanog, OCT4, SOX2, TRA-1-81, TRA-1-60, SSEA3, SSEA4.
Differentiation capacity:
EB differentiation to three germ layer markers: endoderm (AFP, FOXA2), ectoderm (TUJ1, GFAP), mesoderm (ASMA, GATA4).
Identity and genetic stability:
STR analysis showed iPSC line STRs identical to the parental CB CD34+ cells.
Karyotyping confirmed euploid status; lines 3×46,XX and 4×46,XY across lines.
Registration and traceability:
All seven iPSC lines registered with the Spanish National Stem Cell Bank and the European human pluripotent stem cell registry (hPSCReg).
Supplementary data:
Supplementary information contains additional figures validating Sendai virus absence (Fig. S1) and other QC data (Figs. S2–S3, Tables S2–S3).
Detailed results for the seven haplotype lines
Recurrent data points per line (Table 2 overview):
For each donor-derived hiPSC line, expansion of CD34+ cells from the CB unit was quantified at day 1 and day 4, showing expansion in the range from ~0.23×10^6 to ~3.46×10^6 cells and day-4 expansions up to ~3.46×10^6 cells.
Expansion factor (day4/day1) ranged from ~1.45 to ~3.87 depending on line.
Reprogramming efficiency per line ranged roughly from ~0.45% to ~1.8% (values shown in Table 2).
From each line:
Between 10 and 14 clones were initially picked and expanded; five clones per line were selected for further analysis.
From these five clones, RNA was extracted and tested to confirm absence of Sendai virus signals; two virus-free clones per line were subjected to karyotype analysis.
Finally, one virus-free clone per line with normal karyotype was chosen for full characterization and banking.
Table 1 (haplotype details): seven haplotypes (CBiPS1–CBiPS8) each mapped to specific HLA-A, HLA-B, HLA-C, HLA-DRB1, HLA-DQB1, HLA-DPB1 alleles; ABO type and sex recorded.
Example haplotype entries include (illustrative formatting):
CBiPS1-Sv4F-B8: HLA-A 29:02, HLA-B 44:03, HLA-C 16:01, HLA-DRB1 07:01, HLA-DQB1 02:02, HLA-DPB1 11:01; ABO O+, Sex XY.
CBiPS2-Sv4F-D10: HLA-A 30:02, HLA-B 18:01, HLA-C 05:01, HLA-DRB1 03:01, HLA-DQB1 02:01, HLA-DPB1 02:02/04:01; ABO O+, Sex XY.
CBiPS3-Sv4F-E9: HLA-A 03:01, HLA-B 07:02, HLA-C 07:02, HLA-DRB1 15:01, HLA-DQB1 06:02, HLA-DPB1 02:01/04:01; ABO O+, Sex XY.
CBiPS4-Sv4F-F6: HLA-A 01:01, HLA-B 08:01, HLA-C 07:01, HLA-DRB1 03:01, HLA-DQB1 02:01, HLA-DPB1 01:01/03:01; ABO O-, Sex XX.
CBiPS6-Sv4F-H6: HLA-A 33:01, HLA-B 14:02, HLA-C 08:02, HLA-DRB1 01:02, HLA-DQB1 06:02, HLA-DPB1 02:01P; ABO O-, Sex XX.
CBiPS7-Sv4F-I12: HLA-A 24:02, HLA-B 07:02, HLA-C 07:02, HLA-DRB1 15:01, HLA-DQB1 06:02, HLA-DPB1 04:01P; ABO O+, Sex XY.
CBiPS8-Sv4F-J1: HLA-A 11:01, HLA-B 27:05, HLA-C 01:02, HLA-DRB1 01:01, HLA-DQB1 05:01, HLA-DPB1 04:01/11:01; ABO O+, Sex XX.
Karyotype and pluripotency:
All seven haplolines were euploid (3 lines 46,XX; 4 lines 46,XY) at clone selection (passages 9–16) and banking (passages 13–17).
All seven lines stained positive for AP and expressed pluripotency markers (Nanog, OCT4, SOX2, TRA-1-81, TRA-1-60, SSEA3, SSEA4).
In vitro differentiation demonstrated capacity for endoderm (AFP, FOXA2), ectoderm (TUJ1, GFAP), and mesoderm (ASMA, GATA4).
STR analysis confirmed identity with the corresponding CB CD34+ source.
Banking status:
All seven haplolines were registered with the Spanish National Stem Cell Bank and with the European registry (hPSCReg).
Figure and table references (in the study):
Fig. 1 shows cumulative haplotype coverage of the Spanish population with 0MM, 1MM, and 2MM scenarios.
Fig. 2 and Fig. 3 depict morphology and immuno-detection of pluripotency and differentiation markers.
Table 3 presents the quality control assays and acceptance criteria for GMP-MCB and GMP-WCB.
Supplementary figures (S1–S3) provide additional QC data, including proof of absence of Sendai virus.
Comparative context and discussion
Context of haplobanking in other populations:
Prior work in Japanese, Korean and European populations shows that relatively small numbers of haplotypes can cover substantial portions of populations, with larger haplotypes banks needed for higher coverage.
Nakatsuji et al. and Okita et al. demonstrated that a targeted set of homozygous haplotypes can cover large fractions of the population, but scale and donor availability vary by population.
CiRA (Japan) and Korea’s KHIB have produced substantial haplobanks, with several dozen lines covering a portion of their populations.
Population coverage implications:
For Spain, the seven haplo-lines achieve 21.37% coverage with 0MM, 50.83% with 1MM, and 92.46% with 2MM, highlighting that increasing the mismatch tolerance dramatically increases population coverage.
In some clinical scenarios, less stringent HLA matching may be acceptable with immunosuppression, enabling broader coverage.
Global/regulatory perspective:
Haplobanks aim to reduce manufacturing costs and immune rejection risk by providing off-the-shelf iPSC-derived products for multiple indications.
The regulatory pathway classifies iPSC derivatives as ATMPs; GMP-compliant production and traceability are essential.
GMP adaptation: some haplotype iPSC lines are generated under GMP or are adapted to GMP after initial generation; many current clinical trials use non-GMP lines subsequently qualified for GMP.
Ethical and governance considerations:
Donor consent, re-consenting for research use, and ethics approvals are essential; ongoing donor engagement may be required.
International coordination (e.g., COST Action HAPLO-iPS) aims to standardize methodologies, data sharing, and training, and to integrate haplobanks into registries such as hPSCReg.
Practical implications for therapy development:
Haplobanks provide a scalable, cost-reducing platform for generating derivatives of a wide array of tissues (retina, heart, neural tissue, blood cells, etc.).
The study emphasizes that iPSCs should be used as an intermediate production step under strict QC to ensure comparability across lines and facilities.
Limitations and future directions:
The approach relies on donor availability and continued validation of genetic stability across passaging and banking.
Further standardization of QC assays and regulatory guidance is necessary to harmonize cross-border use of haplobanks.
Regulatory and scientific takeaway:
The haplobank represents a strategic approach to off-the-shelf cell therapies, balancing immunogenetics, manufacturing feasibility, and clinical utility within a European context.
Conclusions and implications for ATMP development
The haplobank described here constitutes a valuable intermediate cell bank capable of covering >20% of the Spanish population for 0MM HLA matching, with substantial additional coverage when mismatches are permitted (up to ~92% for 2MM).
These haplolines, generated from seven CBUs homozygous for common HLA haplotypes, provide a scalable starting material for ATMP development and may be shared across Europe and beyond due to haplotype sharing.
The work highlights the feasibility of producing GMP-compliant MCBs/WCBs from CB-derived hiPSCs, with rigorous QC, donor consent, and traceability to support clinical applications.
The initiative supports a coordinated European framework (e.g., COST action, hPSCReg) and aligns with global efforts to establish national haplobanks for safer, faster, and more cost-effective cell therapies.
Practical details and takeaways for exam preparation
Key workflow sequence:
Identify homozygous CBUs with common Spanish haplotypes → isolate CD34+ progenitors → expand in defined cytokine-supplemented medium → reprogram with Sendai CytoTune kit → select virus-free iPSC clones with normal karyotype → expand under GMP → bank MCBs and WCBs → perform QC and characterize lines → register lines.
Major outcomes to remember:
7 haplo-homozygous CBUs yielded 7 hiPSC lines (CBiPS1–CBiPS8) with euploid karyotypes and pluripotency markers.
Population coverage of Spain: 0MM = 21.37%, 1MM = 50.83%, 2MM = 92.46%.
All lines were banked under GMP with MCBs and WCBs tested for sterility, mycoplasma, endotoxin, and adventitious viruses; identity confirmed by STR; regulatory registration in national and European registries.
Important regulatory concepts:
iPSC derivatives used for cell therapy are considered ATMPs and require GMP manufacturing, traceability, and robust QC.
The study shows how to balance donor selection, haplotype coverage, and GMP requirements to build a clinically useful haplobank.
Abbreviations
ATMP: Advanced Therapy Medicinal Product
BMI: (not used here; see context) – ignore if not relevant
BAC: (not used here; see context) – ignore if not relevant
A list of abbreviations used in the paper includes:
ATMP, BM, CB, CBU, CTS, CTG, EMA, EB, FDA, GMP, hiPSC, HLA, ICC, iPSC, MCB, QC, REDMO, STR, WCB, WMDA, etc.
Supplementary information and references
The article provides extensive supplementary data including details on: absence of Sendai virus in clones (Additional Fig. S1), additional QC data (Figs. S2–S3, Table S2–S3).
Funding and ethics: Project IPS-PANIA funded by MCIN; ethics committee approval; donor informed consent. The COST Action and hPSCReg integration are highlighted for future European-scale haplobank collaboration.
Summary statement
This study documents the generation of the first clinical-grade, HLA-homozygous iPSC haplobank in Spain, derived from seven cord blood units homozygous for common haplotypes, with GMP-compliant expansion and banking, comprehensive characterization, and robust QC. The resulting haplolines provide a foundation for off-the-shelf cell therapies in Spain and potentially across Europe, illustrating a scalable path toward broader ATMP accessibility while addressing immunological compatibility and regulatory requirements.