Abstract
Introduction
Paroxysmal nocturnal hemoglobinuria (PNH) is a rare, life-threatening hematologic disorder caused by clonal expansion of hematopoietic stem cells (HSCs) with somatic PIGA mutations. These mutations lead to the loss of GPI-anchored proteins, such as CD55 and CD59, on red blood cells, resulting in complement-mediated hemolysis. Although complement inhibitors like eculizumab provide symptomatic relief, a substantial proportion of patients, particularly in East Asia (including Japan), exhibit a limited therapeutic response due to bone marrow failure (BMF), either present at diagnosis or developing later, such as aplastic anemia (AA) or myelodysplastic syndromes (MDS). This highlights the need for curative approaches.
While PIGA-mutated (PNH-type) and wild-type HSCs coexist, the former preferentially expand. Immune escape due to GPI-AP deficiency has been proposed, yet mechanisms driving clonal dominance and ineffective hematopoiesis remain unclear (Murakami et al., Blood 2002; Hanaoka et al., Blood 2006). Although PNH-specific mutations have been hypothesized, only alterations in genes like TET2, ASXL1, and JAK2—commonly found in AA, MDS, or clonal hematopoiesis—have been reported (Shen et al., J Clin Invest 2014), and no PNH-specific mutations have been conclusively identified.
Induced pluripotent stem cells (iPSCs) derived from patients offer a useful platform for studying rare diseases with limited sample availability, allowing functional gene analysis and disease modeling. We therefore generated iPSCs from PNH patients to explore gene alterations involved in PNH-associated BMF (PNH-BMF).
Methods
Five PNH patients with BMF were enrolled. PNH-iPSCs and matched wild-type N-iPSCs were established by reprogramming mononuclear hematopoietic cells from each patient. These iPSCs were differentiated into hematopoietic stem/progenitor cells (HSPCs), and their differentiation and proliferation were evaluated. PIGA mutations in PNH-iPSCs were corrected using CRISPR/Cas9. Whole-genome sequencing (WGS) was performed on PNH-iPSCs, N-iPSCs, and normal somatic cells from the same patients. PNH-specific variants were defined based on stopgain, frameshift, and splice-site mutations, as well as nonsynonymous SNVs annotated as pathogenic or likely pathogenic in AlphaMissense and ClinVar. Candidate variants were cross-referenced with curated databases of hematopoiesis-associated genes potentially linked to hematopoietic dysfunction.
Results
Among the five patients, one had hypoplastic marrow and four had normoplastic marrow. Median neutrophil counts, hemoglobin, and platelet counts were 1.6 × 10⁹/L, 8.4 g/dL, and 97 × 10⁹/L, respectively. Median proportions of PNH-type erythrocytes and granulocytes were 24.3% and 67.2%, suggesting an underlying BMF tendency.
Hematopoietic dysfunction was reproduced in the PNH-iPSC model. Correction of PIGA mutations failed to fully restore normal hematopoiesis, suggesting the involvement of additional pathogenic factors.
WGS analysis was successfully completed in four patients. Multiple PNH clone-specific somatic mutations were identified in all cases. In two patients, variants involved known hematopoiesis-related genes such as DNMT3A and ASXL1. One patient progressed to MDS during follow-up, indicating a possible genotype–phenotype correlation. The remaining two patients harbored novel mutations not previously reported to associate with hematopoietic dysfunction. Although not known to directly promote proliferation or survival, database analyses suggested these mutations may alter HSC biology—through impaired quiescence, lineage skewing, or disrupted interactions with the bone marrow niche. No shared mutations were observed across patients, supporting the notion that the pathogenesis of PNH-associated BMF is genetically heterogeneous.
Conclusions
We successfully established matched PNH- and N-iPSC lines from four patients and modeled hematopoietic abnormalities in vitro. Limited hematopoietic recovery following PIGA correction suggests the involvement of additional genetic factors. WGS identified PNH clone–specific somatic mutations in all cases, involving both known and novel genes potentially affecting hematopoiesis. These findings support the genetic heterogeneity of PNH-BMF and highlight the potential value of precision medicine in future treatment strategies. Ongoing CRISPR/Cas9-mediated knock-in experiments targeting selected genes in N-iPSCs are underway to assess their pathogenic relevance.
