Mutant Samd9l expression impairs hematopoiesis and induces bone marrow failure in mice

Abstract

SAMD9 and SAMD9L germline mutations have recently emerged as a new class of predispositions to pediatric myeloid neoplasms. Patients commonly have impaired hematopoiesis, hypocellular marrows, and a greater risk of developing clonal chromosome 7 deletions leading to MDS and AML. We recently demonstrated that expressing SAMD9 or SAMD9L mutations in hematopoietic cells suppresses their proliferation and induces cell death. Here, we generated a mouse model that conditionally expresses mutant Samd9l to assess the in vivo impact on hematopoiesis. Using a range of in vivo and ex vivo assays, we showed that cells with heterozygous Samd9l mutations have impaired stemness relative to wild-type counterparts, which was exacerbated by inflammatory stimuli, and ultimately led to bone marrow hypocellularity. Genomic and phenotypic analyses recapitulated many of the hematopoietic cellular phenotypes observed in patients with SAMD9 or SAMD9L mutations, including lymphopenia, and pinpointed TGF-β as a potential targetable pathway. Further, we observed nonrandom genetic deletion of the mutant Samd9l locus on mouse chromosome 6, mimicking chromosome 7 deletions observed in patients. Collectively, our study has enhanced our understanding of mutant Samd9l hematopoietic phenotypes, emphasized the synergistic role of inflammation in exaggerating the associated hematopoietic defects, and provided insights into potential therapeutic options for patients.

Keywords: Hematology; Hematopoietic stem cells; Mouse models.

Figure 1. Samd9l-Mut mouse has altered hematopoiesis.

(A) Model for the generation of conditional knockin Samd9l-W1171R mutation (Samd9l cKI^+/+) crossed with the hematopoietic cell–specific Vav1-Cre mouse (Vav1-Cre^+/Tg) to produce offspring with heterozygous Samd9l mutations (Vav1-Cre^+/Tg Samd9l cKI^+/–). (B) PCR analysis verifying the genotypes of the mice. The left gel shows the PCR result for the knockin insertion to have a 1,270-bp product if the cassette is present, 514-bp product if not, and both for heterozygosity. The right gel shows the PCR products for the Vav-1 amplicon using iCre primers (390 bp) and internal positive control (324 bp). (C) Complete blood count (CBC) of Samd9l-KO (Samd9l^–/–, blue, n = 6), Samd9l-WT (Samd9l^cKI+/–, black, n = 11), and Samd9l-Mut (Vav1-Cre^+/Tg Samd9l^cKI+/–, red, n = 14) mice at 3 months: white blood cells (WBC, left), red blood cells (RBC, middle), and platelets (PLT, right). (D and E) Flow cytometric analysis of C57BL/6 (gray, n = 4), Samd9l-KO (blue, n = 8), Samd9l-WT (black, n = 8), and Samd9l-Mut (red, n = 8) mice assessing the BM compartment for (D) Lineage^– (Lin^–), common lymphoid progenitors (CLPs), myeloid progenitors (MPs), common myeloid progenitors (CMPs), granulocyte-macrophage progenitors (GMPs), and megakaryocyte/erythroid progenitors (MEPs), and (E) KSL (Lin^–cKit⁺Sca-1⁺), multipotent progenitors (MPP2 and MPP3/4), and short-term and long-term HSCs (ST-HSCs and LT-HSCs). (F and G) Percentage of mature cells in (F) peripheral blood (PB) or (G) BM cells of the C57BL/6, Samd9l-KO, Samd9l-WT, and Samd9l-Mut mice assessed by flow cytometry. For panels C–G, groups were initially compared by Kruskal-Wallis test. Significant Kruskal-Wallis results were followed by pairwise comparisons with Wilcoxon’s rank-sum test. *P < 0.05; **P < 0.01; ***P < 0.001. Error bars indicate the SEM for biological replicates.

Figure 2. Mutant Samd9l expression skews murine hematopoietic lineage profiles.

(A and B) scRNA-seq of Samd9l-KO (n = 1,495), Samd9l-WT (n = 1,498), and Samd9l-Mut (n = 1,436) mice from a single experiment. (A) Uniform manifold approximation and projection (UMAP) plots of scRNA-seq data. Eleven clusters were identified according to established marker expression and consolidated into 5 main populations as annotated. (B) A circular plot showing the proportion of the identified 11 clusters in each group relative to Samd9l-WT control. (C) Myeloid cell differentiation trajectories based on reported markers, including S100a6, Retnig, Camp, Psap, and others on the scRNA-seq UMAP plot (top panel) and the percentage of cells in each cluster (middle and bottom panels) from Samd9l-KO, Samd9l-WT, and Samd9l-Mut mice. The colors of each dot represent the normalized expression level of genes indicated above. (D) B cell differentiation trajectory based on established markers, including Cd19, Il7r, and Ms4a1 (encodes CD20 protein) (top panel) and the percentage of cells in each cluster (middle and bottom panels) from Samd9l-KO, Samd9l-WT, and Samd9l-Mut mice. (E) CBC of Samd9l-KO (n = 6), Samd9l-WT (n = 11), and Samd9l-Mut (n = 14) mice at 3 months of age showing neutrophils (NEs), monocytes (MOs), eosinophils (EOs), basophils (BAs), and lymphocytes (LYs). (F and G) Histological assessment of BM (F, ×40 magnification) and spleen (G, ×4 magnification) sections from 3-month-old Samd9l-WT or Samd9l-Mut. Sections were stained with modified Romanowsky stain for morphological assessment and IHC labeling was done using anti–PAX-5 (B cells) or anti-MPO (myeloid cells). For panel E, groups were initially compared by Kruskal-Wallis test. Significant Kruskal-Wallis results were followed by pairwise comparisons with Wilcoxon’s rank-sum test. *P < 0.05, **P < 0.01, ***P < 0.001. Error bars indicate the SEM for biological replicates.

Figure 3. Impaired ex vivo and in vivo fitness of Samd9l-mutant cells.

(A and B) Serial colony-forming unit cell (CFU-C) analysis of Lin^– cells from C57BL/6, Samd9l-KO, Samd9l-WT, and Samd9l-Mut mice. (A) Cells were plated (3,000 cells/plate) in methylcellulose media. One week later, colonies were counted and serially replated (10,000 cells/plate) for a total of 3 rounds. (B) The number of colonies in the weekly replates. (C) Schematic illustration of the competitive transplantation model. Lin^– cells from CD45.2 (C57BL/6, Samd9l-KO, Samd9l-WT, or Samd9l-Mut cells) and CD45.1 (wild-type competitor) were injected i.v. at a 1:1 ratio via tail vein into lethally irradiated CD45.1/CD45.2 mice. (D) CD45.2 chimerism in the PB of recipient animals (n ≥ 12 per group) from 0 to 8 weeks. (E) Intrafemoral (i.f.) competitive transplantation scheme. Lin^– cells from CD45.2 (Samd9l-WT or Samd9l-Mut cells) and CD45.1 (wild-type competitor) were injected at a 1:1 ratio into the femurs of lethally irradiated CD45.1/CD45.2 mice. (F) CD45.2 chimerism in the PB of recipient animals in E (n ≥ 8 per group). Time is denoted in weeks after injections. A longitudinal mixed-effects regression model was used for statistical analysis. Initially, a global test of whether all 4 groups had the same longitudinal trend was performed. A significant result from this global test was followed by pairwise tests evaluating the equality of trends over time for the 2 groups. ***P < 0.001. Error bars indicate the SEM for biological replicates. Gray, C57BL/6; blue, Samd9l-KO, black, Samd9l-WT; red, Samd9l-Mut.

Figure 4. Inflammation promotes Samd9l-mutant cell death and further impairs repopulation potential.

(A–C) SAMD9L protein expression (A), annexin V⁺ cell percentage (B), and CFU-C (C) of BM harvested from indicated mice and treated twice with IFN-α (1,000 U) or vehicle for 48 hours ex vivo. For CFU-C, BM cells (10,000) were cultured in methylcellulose media, and colonies were counted after 1 week. (D–F) SAMD9L protein expression (D), annexin V⁺ cell percentage (E), and CFU-C (F) of BM harvested from indicated mice treated with pI:pC (5 mg/kg) or vehicle twice a week for 4 weeks. (G) Weekly colony counts of vehicle- or IFN-α–treated (1,000 U) BM cells derived from a patient harboring the SAMD9L-S626L mutation normalized to healthy donor cord blood. (H) Model for Samd9l-WT and Samd9l-Mut pI:pC treatment followed by 5:1 competitive intrafemoral BM transplants of CD45.2 (Samd9l-WT or Samd9l-Mut treated with pI:pC or vehicle) versus CD45.1. (I–K) CD45.2 chimerism in BM at 6-week endpoint (I) and PB biweekly (J). The comparison was done between Samd9l-Mut vehicle (n = 9) vs. Samd9l-Mut pI:pC (n = 9) or Samd9l-WT vehicle (n = 3) vs. Samd9l-WT pI:pC (n = 4). (K) Relative levels of annexin V⁺ cells within total BM, B, T, or myeloid cells of the donor cells from each group (ratio of annexin V⁺ CD45.2 divided by annexin V⁺ CD45.1 cells. For panels G and J, a longitudinal mixed-effects regression model was used for statistical analysis followed by pairwise Tukey-adjusted tests. For panel K, the data were not normally distributed; therefore, the Kruskal-Wallis test was initially performed and followed by Wilcoxon’s rank-sum test for pairwise comparisons. For all other panels, 2-way ANOVA was used to evaluate the main effects and genotype/treatment interactions and followed by Tukey’s pairwise comparisons. Error bars indicate the SEM for biological replicates. Blue, Samd9l-KO, black, Samd9l-WT; red, Samd9l-Mut. *P < 0.05; **P < 0.01; ***P < 0.001 compared with vehicle-treated groups. ^#P < 0.05 compared with pI:pC-treated groups. Color indicates the comparison group.

Figure 5. The lack of fitness in Samd9l-mutant cells is partly via TGF-β activation.

(A) Model of pI:pC treatment regimen followed by sorting the Lin^–cKit⁺ (LK) population from Samd9l-WT and Samd9l-Mut mice to perform RNA-seq. (B) Heatmap showing the up- and downregulated pathways in Samd9l-WT and Samd9l-Mut treated with vehicle or pI:pC. (C) A plot of pathway enrichments of DEGs downregulated in pI:pC-treated Samd9l-Mut mice relative to vehicle-treated. The size of the circles represents gene counts, and the significance was determined by FDR. The color is dependent on the fold of enrichment. (D) Gene set enrichment analysis (GSEA) showing TGF-β pathway activation in pI:pC-treated Samd9l-Mut mice relative to vehicle-treated Samd9l-Mut mice. Normalized enrichment score (NES), FDR, and P value are indicated. (E) Phospho-SMAD2/3 expression in Samd9l-KO, Samd9l-WT, and Samd9l-Mut BM cells treated with IFN-α or vehicle (n = 4 per group). (F) Representative histograms of phospho-SMAD2/3 expression in B cells of Samd9l-WT and Samd9l-Mut cells after IFN-α or vehicle. (G and H) Serial CFU-C replating of Samd9l-WT and Samd9l-Mut cells (G) or human cells from a patient with SAMD9L-S626L mutation or control (H) with or without TGF-β inhibitor (SD-208). Data show at least 3 independent experiments. For panel E, Kruskal-Wallis test was used to perform an initial comparison across all groups and followed by pairwise comparisons with Wilcoxon’s rank-sum test. For panel G, for each genotype/time point, we used Wilcoxon’s rank-sum test to compare across the 2 treatments because data were not normally distributed. For panel H, a longitudinal mixed-effects regression model was used for statistical analysis followed by pairwise Tukey-adjusted tests evaluating the equality of means across each pair of groups at each time point. *P < 0.05, **P < 0.01, ***P < 0.001 compared with vehicle-treated groups. ^#P < 0.05 compared with pI:pC-treated groups. Error bars indicate the SEM for biological replicates. Blue, Samd9l-KO; black, Samd9l-WT; red, Samd9l-Mut (red); stripes, IFN-α or pI:pC; solid, vehicle. Color indicates the comparison group.

Figure 6. Inflammation worsens cytopenias in Samd9l-Mut mouse.

(A) Illustration for treatment of Samd9l-WT and Samd9l-Mut mice with pI:pC or vehicle. (B) CBC analysis showing WBC, lymphoid cells, and monocytes/neutrophils/basophils (myeloid) for the treated mice (n = 8). (C and D) BM (C) and spleen (D) were stained with anti–PAX-5, anti-CD3, and anti-MPO to assess B, T, and myeloid cells, respectively (n = 2 per group). For panel B, the Kruskal-Wallis test was used to perform an initial comparison across all groups, and followed by pairwise comparisons with Wilcoxon’s rank-sum test. *P < 0.05, **P < 0.01, ***P < 0.001 compared with vehicle-treated groups. ^#P < 0.05, ^###P < 0.001 compared with pI:pC-treated groups. Color indicates the comparison group. Error bars indicate the SEM for biological replicates. Black, Samd9l-WT, red, Samd9l-Mut; stripes, pI:pC; solid, vehicle.

Figure 7. Inflammation impairs cell maturation in Samd9l-Mut mouse.

(A–D) scRNA-seq of WBM and Lin^–cKit⁺ (LK) sorted cells from Samd9l-WT or Samd9l-Mut mice treated with either pI:pC or vehicle. (A) UMAP plots of scRNA-seq data showing 5 main populations. (B) A circular plot showing the proportion of the identified 17 clusters from each WBM sample relative to WBM from Samd9l-WT control. (C) A circular plot showing 8 clusters from LK sorted samples. (D) A heatmap of the proportion of cells in each erythroid maturation stage is denoted as megakaryocyte/erythroid progenitor (MEP), early erythroid (Ery 1), and late erythroid (Ery 2). Among the compared groups, red indicates higher levels and blue indicates lower levels, and values represent the percentage of total cells. (E) BM cross sections (magnification, ×40) from Samd9l-WT or Samd9l-Mut mice treated with pI:pC or vehicle stained with anti-GATA1 antibody.

Figure 8. Samd9l mutation induces bone marrow hypocellularity and nonrandom chromosome 6 loss in vivo.

(A) Survival analysis of Samd9l-WT (n = 12) and Samd9l-Mut mice (n = 25) over 300 days. Median survival was not reached by either group. Kaplan-Meier estimates were used, and significance was evaluated using the log-rank test. (B–D) Pathological assessment of Samd9l-Mut mice (died at 6 months) versus age-matched Samd9l-WT mice. (B) BM sections of Samd9l-Mut or Samd9l-WT mice (H&E; magnification, ×10 and ×40). (C) Peripheral blood WBC (n = 3). (D) Cell counts of viable BM cells per leg (n = 3). Wilcoxon’s rank-sum pairwise comparisons test was used for statistical analysis. *P < 0.05. Data shown as mean ± SEM. (E) Schematic of the affected region in chromosome 6 (chr6, qA1–qA3.1) in pI:pC-treated Samd9l-Mut mice relative to vehicle-treated Samd9l-Mut mice or Samd9l-WT mice treated with vehicle or pI:pC. The illustration shows the end of chr5, the entire chr6, and the start of chr7. The y axis shows the gene expression, and the x axis shows a genomic position index where each dot represents a gene and corresponds to its location. (F) Integrative Genomics Viewer (IGV; https://software.broadinstitute.org/software/igv/home) plot from RNA-seq analysis showing RNA transcripts of Samd9l wild-type allele (green) and Samd9l W1171R (A>G) heterozygous mutation (orange) in both Samd9l-Mut treated with vehicle (upper track) and Samd9l-Mut treated with pI:pC (lower track). (G) Illustration of mouse chr6 showing the affected region (red stripes) and the positions of the custom-designed FISH probes to detect the deleted region (proximal to Samd9l locus) or control nondeleted region (distal from Samd9l locus). (H) FISH analysis of spleens from Samd9l-Mut or Samd9l-WT mice treated with vehicle or pI:pC. FISH probes: proximal probe at chr6:3,496,083–3,687,193 (red) and distal probe at chr6:28,129,437–28,303,622 (green). Nuclei were stained with DAPI and are outlined by white dashed lines. Images were acquired on a Nikon C2 laser scanning confocal microscope (60×). Scale bar: 10 μm.