SF3B1 Mutant Clones From Patients With Refractory Anaemia With Ringed Sideroblasts (RARS) Originate From The Early Haematopoietic Stem Cells and Maintain Their Engraftment Potential

Defects in pre-mRNA splicing have recently been implicated in the pathogenesis of myelodysplastic syndrome, particularly RARS where SF3B1, a key component of the spliceosome machinery is mutated in up to 83% of the cases. Moreover, as homozygosity is the exception to the rule for these mutations and SF3B1 homozygous systems have been previously reported as lethal, then such aberrations are most certainly not benign bystanders and are likely to play a significant role in pathogenesis of the disease. Besides the strong correlation of SF3B1 mutations and the presence of ringed sideroblasts, the mechanism through which these mutations affect their target cells, distort the RNA splicing process and subsequently lead to disease still remains unclear.

In order to gain insight into the effects of SF3B1 mutations on the engraftment kinetics of haemopoietic stem cells, bone marrow CD34+ cells from RARS patients (n=4) with SF3B1 mutations (with mutant allele burden 20% to 48%) were transplanted in NOD/SCID/interleukin-2 receptor γ chain–null (NOD/SCID/IL2rγ^−/−) mice (1-3 mice/patient), and were also plated in committed progenitor (colony-forming cell, or CFC) and primitive progenitor (long-term bone marrow culture-initiating cell, or LTC-IC) cultures. Following on, whole-exome sequencing (Illumina) was performed on pre-transplant (bone marrow CD34+/TNC), post-transplant week-24 (human CD45⁺CD33⁺) and LTC-IC week-5 samples. In addition, we also performed whole-exome sequencing on FACS-purified HSCs (Lin- CD45+CD34+CD38- CD45RA-CD90+CD49f+), MPPs (Lin- CD45+CD34+CD38-CD45RA-CD90-CD49f-), GMPs (Lin- CD45+CD34+CD38+CD135+ CD45RA+) and MEPs (Lin- CD45+CD34+CD38+CD135-CD45RA-) from 3 of these patients.

Our data demonstrates that the SF3B1 mutations arise from the early HSCs and are propagated without diminution in the mutant allele burden from HSC to more committed myeloid progenitors (i.e. MPPs, GMPs and MEPs). We also show, that the transplantation of bone marrow CD34+ cells derived from SF3B1 mutant RARS patients into mice leads to persistent long-term engraftment (0.01% - 4%, up to 24 weeks) of only myeloid cells but not lymphoid cells, irrespective of the type of SF3B1 mutation. The RARS-originated cells recovered from the human cell-engrafted mice maintained its genotypic characteristics, as demonstrated by the presence of SF3B1 mutations which remained largely unchanged between the pre- and post-transplant samples. This phenomenon was also observed in LTC-IC and CFC culture experiments where the SF3B1 mutant allele burden was maintained at the same level. In all analysed samples, there was no difference in the SF3B1 mutation burden between the LTC-IC with cytokines or without cytokines, and CFC under normoxic or hypoxic conditions. Whole-exome sequencing also revealed additional 20-60 known and/or novel candidate somatic mutations which were present concurrently in mice, LTC-IC and/or CD34+ (pre-transplant) cells for each RARS experiment. These include genes involved in epigenetic modifications, cell signalling/transcription regulation, DNA maintenance and cytoskeletal organization. These candidate mutations were either maintained at the same mutant allele burden or reduced/increased between the pre-transplant, post-transplant and LTC-IC, and are likely to be the sub-clones of SF3B1 clone. For example, in one RARS case JAK2 V617F mutation changed from 20% at pre-transplant to 9% (average mutant allele burden in mice, n=3) at post-transplant and 10% in LTC-IC. However, DNMT3A W581S mutation present in another RARS case was largely maintained i.e. 44% at pre-transplant, 38% at post-transplant (n=1) and 43% in LTC-IC.

These results demonstrate that SF3B1 mutations arise from the early HSCs, persist in the MEPs/GMPs, and retain their engraftment capacity in-vivo as well as their long-term proliferation potential in-vitro. Our murine xenograft RARS model comprised of human myeloid progenitor and mature cells, the majority of these cells, if not all, originated from the patients original clonal HSC pool as exhibited by its aberrant genotypic characteristics. Furthermore, this study also suggests that the cells carrying the SF3B1 mutations have a clonal growth advantage as compared to normal cells present in these RARS patients, and is therefore likely to contribute to the RARS disease phenotype.

Bonnet:Janssen and Janssen: Collaborative Project and Consultant Other; Genetech: Collaborative project, Collaborative project Other.