Genetic variation at PPM1H predicts mobilization efficiency in stem cell donors Open Access

TO THE EDITOR:

Hematopoietic stem cell transplantation is a cornerstone in the treatment of blood disorders.^1,2 CD34⁺ hematopoietic stem cells are commonly collected through leukapheresis, which requires the mobilization of CD34⁺ cells from the bone marrow into the blood with medications such as granulocyte colony-stimulating factors (G-CSF), cyclophosphamide, and CXCR4 inhibitors.³ However, in ∼5% of healthy allogeneic donors and 15% of autologous donors, CD34⁺ cells remain difficult to mobilize.^4,5 Identifying genetic factors influencing blood CD34⁺ cell levels could reveal novel drug targets to facilitate stem cell mobilization and genetic biomarkers to predict mobilization efficiency.

Recently, we reported, to our knowledge, the first large-scale genome-wide association study on blood CD34⁺ cell levels. Across 13 167 adult blood donors and primary care patients, we identified 9 significant and 2 suggestive associations, accounted for by 8 loci (PPM1H, CXCR4, ENO1-RERE, ITGA9, ARHGAP45, CEBPA, TERT, and MYC).^6,7 The most significant association maps to the PPM1H gene (protein phosphatase, Mg²⁺/Mn²⁺ dependent 1H), not previously implicated in hematopoiesis or stem cell biology.⁸ 

Compared to other blood cell types, PPM1H is preferentially expressed in CD34⁺ cells, and the allele that confers higher blood CD34⁺ cell levels downregulates PPM1H. Through functional studies, we found that this downregulation is caused by the single-nucleotide polymorphism rs772557[A>G].⁶ The rs772557 variant is polymorphic in all geographic ancestries, and the minor allele frequency varies from 3.5% to 47%.⁹ The rs772557-A allele abrogates a MYB transcription factor binding site in PPM1H intron 1, which is active in several CD34⁺ cell subpopulations and interacts with the PPM1H promoter by chromatin looping. Furthermore, PPM1H knockdown increased the proportion of CD34⁺ cells in cord blood assays.⁶ These data suggest that PPM1H could be used as a novel inhibition target to facilitate stem cell mobilization.

To further explore the relevance of PPM1H as a target for stem cell mobilization, we sought to explore the effects of rs772557 on CD34⁺ cell mobilization in stem cell donors. We hypothesized that donors who carry the rs772557-A allele should respond better to mobilization therapy than donors carrying the rs772557-G allele. The rationale for this hypothesis is that rs772557-A associates with low PPM1H expression and higher basal CD34⁺ cell levels, whereas rs772557-G associates with high PPM1H expression and low basal CD34⁺ cell levels.⁶ Accordingly, the PPM1H activity in rs772557-A/A homozygotes and rs772557-A/G heterozygotes should be endogenously inhibited relative to rs772557-G/G homozygotes. To test this hypothesis, we prospectively analyzed the association between rs772557 genotype and mobilization efficiency in unselected stem cell donors.

We analyzed blood CD34⁺ cell counts in routine adult stem cell donors at the Stem Cell Transplantation Unit at the Division of Hematology at Skåne University Hospital, Sweden (Table 1). All donors were first mobilized solely with G-CSF (filgrastim 10 μg/kg per day). One peripheral blood sample (4 mL) was collected before mobilization during donor evaluation for baseline CD34⁺ levels and 1 in the morning on day 4 (autologous donors and related allogeneic donors) or day 5 (unrelated registry allogeneic donors) after mobilization (supplemental Methods). All donors gave written informed consent, and the study was approved by the Swedish Ethical Review Authority (approval number 2022-02329-01). Donors were recruited consecutively, for a period of ∼18 months from 2023 to 2025.

Exactly 2 mL of blood was used for flow cytometry analysis and volumetric counting. We used standardized methods for erythrocyte lysis, washing, and staining (supplemental Methods). To define CD34⁺ cell percentage, we first gated singlet cells based on forward scatter (FSC) area and FSC height. We then gated peripheral blood mononuclear cells from singlets based on FSC area and side scatter area. Finally, from peripheral blood mononuclear cells, we defined CD34⁺ cells as CD34⁺45^low. The absolute CD34⁺ count was calculated by multiplying the total number of events per microliter, obtained from the volumetric counting, by the percentage of CD34⁺CD45^low cells among total events. Genotyping was performed from the remaining blood (supplemental Methods).

To calculate the mobilization efficiency, we subtracted the absolute count of baseline CD34⁺ (before mobilization) from the absolute CD34⁺ count after mobilization and divided this by the absolute count of baseline CD34⁺ (ie, mobilization efficiency = [CD34⁺_{count_mobilized} – CD34⁺_{count_baseline}]/CD34⁺_{count_baseline}). Because of the limited sample size, we pooled rs772557-A/A and rs772557-A/G donors into 1 group, representing donors with varying degrees of endogenous, transcriptional PPM1H inhibition. We contrasted this group with rs772577-G/G homozygotes, representing donors with full PPM1H expression.

We recruited 87 routine adult stem cell donors, including 31 healthy allogeneic (35%) and 56 autologous donors (65%; Table 1). We analyzed the 2 groups separately because blood CD34⁺ levels in autologous donors could be influenced by previous bone marrow–suppressive treatments, primary disease, and aging.^10,11 To reduce the possibility of observer bias, the flow cytometry analysis and gating were done before genotyping.

In each donor, we quantified mobilization efficiency as the increase in blood CD34⁺ cell concentration after mobilization relative to their baseline concentration before mobilization. We reasoned that this quantification is more likely to be influenced by genetic variation than the total CD34⁺ cell yield, which in practice is affected by several nongenetic factors, including collection time, volume of blood processed, and the target CD34⁺ cell yield.^12-14 

We asked whether the genetic variant rs772557, which modulates basal CD34⁺ levels by regulating the expression of PPM1H, influences stem cell mobilization efficiency. Strikingly, in allogeneic donors, we observed a significant difference in mobilization efficiency between rs772557-G/G homozygotes and the combined group of rs772557-A/A homozygotes and rs772557-A/G heterozygotes (8-fold vs 19-fold increase; 1-sided Mann-Whitney test P = .0214; 95% confidence interval, 0.064-24.7; Figure 1A). Baseline characteristics did not differ significantly between the 3 variant groups (Kruskal-Wallis: P = .999 for age; P = .956 for age; P = .71 for body mass index). Furthermore, 3 of the 7 allogeneic donors (43%) who were rs772557-G/G homozygous required additional mobilization therapy to reach the target CD34⁺ cell yield, and the only allogeneic donor who required plerixafor (CXCR4 inhibitor) was 1 of these (Table 1). In comparison, 20% of rs772557-A/G heterozygotes (n = 15) and 13% of rs772557-A/A homozygotes (n = 8) required additional G-CSF. Although these observations were not significant (Fisher exact test; P = .3 [for G/G vs A/A]; P = .3 [for G/G vs A/G]), and further studies in a larger cohort are needed to assess any potential difference between the groups. Finally, within the autologous group (n = 56), there was no significant difference in mobilization effect (P = .108; Figure 1B). This could be explained by aforementioned covariates, which are known to affect mobilization and thus is likely to reduce the effect size of inborn genetic variants.^10,11,15 

In conclusion, we have previously reported an inverse relationship between PPM1H activity and blood CD34⁺ cell levels, supported by a highly significant genetic association, gene expression data in CD34⁺ cells, and ex vivo knockdown studies in primary human CD34⁺ cells. Expanding on these results, we now find an association between the PPM1H–low-expressing rs772557-A allele and a greater increase in blood CD34⁺ cell levels upon G-CSF mobilization therapy in a prospective series of consecutively recruited allogeneic stem cell donors. Our results provide additional genetic support for PPM1H as a potential inhibition target to facilitate CD34⁺ cell mobilization.

Acknowledgments: The authors thank the nurses and study coordinators at the Hematology Clinic at Skåne University Hospital in Lund for their assistance. The authors are indebted to the patients and allogeneic stem cell donors who participated in the study.

This work was supported by grants from the European Research Council (CoG-770992), the Swedish Research Council (2017-02023 and 2018-00424), the Swedish Cancer Society (23-2851 Pj), the Swedish Children’s Cancer Fund (PR2023-0067), ARMEC Lindeberg’s Foundation, Inga-Britt and Arne Lundberg’s Foundation (2017-0055), and ALF grants from Region Skåne.

Contribution: B.N., Y.P.L., and A.L.d.L.P. designed the study with input from S.W. and C.C.; A.L.d.L.P. and Y.P.L. developed flow cytometry protocols; Y.P.L., D.T.D.B., M.L.M.E., C.C., and A.L.d.L.P. performed experiments; Y.P.L., A.L.d.L.P., and B.N. performed statistical analyses; Y.P.L. and B.N. drafted the manuscript; and all authors contributed to the final manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Björn Nilsson, Hematology and Clinical Immunology, Department of Laboratory Medicine, Lund University, BMC B13, 22184 Lund, Sweden; email: bjorn.nilsson@med.lu.se.