Mutant TP53 allelic states in human hematopoietic stem and progenitor cells

In human hematopoietic stem and progenitor cells (HSPCs), TP53 mutations characterize clonal hematopoiesis of indeterminate potential (CHIP) and are recognized as a high-risk factor for the development of non-neoplastic and neoplastic disorders, particularly myelodysplastic syndromes (MDS) and acute myeloid leukemia [ 1 , 2 ]. During leukemic transformation, the remaining TP53 wild-type (WT) allele is frequently affected, most commonly through loss-of-heterozygosity (LOH), copy-neutral LOH, or a second concurrent mutation. The TP53 allelic state—reflecting whether one or both alleles are affected—remains a subject of debate regarding its prognostic relevance in patients with MDS [ 3 ]. Recently, the biological consequences of monoallelic versus biallelic aberrations were investigated using a Trp53 mouse model [ 4 ]. Here, we address this question in human HSPCs by introducing defined TP53 aberrations via CRISPR/Cas9-mediated genome editing. Material and methods are described in detail in the Supplementary information . Enriched, umbilical cord blood (UCB) derived CD34 + HSPCs were targeted at exon 2 of the TP53 gene using a CRISPR/Cas9 and recombinant adeno-associated virus serotype 6 mediated knock-in strategy (Fig. 1A and Supplementary Figs. S1A, B and S2 ) [ 5 ]. By targeting exon 2, we were able to generate WT and aberrant TP53 alleles within the same genomic context while maintaining the endogenous regulatory control. Integration of the donor template at exon 2 disrupts the endogenous TP53 gene, ensuring that only the donor-encoded cDNA sequences are expressed from the endogenous TP53 promoter. Mutant TP53 cDNA—carrying either the variant of interest or a stop codon—or WT cDNA, each linked to a fluorescent reporter (FR) expression cassette, was introduced via homology-directed repair to generate the following monoallelic and biallelic genotypes: TP53 R175H/WT , TP53 R273H/WT , TP53 KO/WT and TP53 R175H/KO , TP53 R273H/KO , TP53 KO/KO . The p.R175H and p.R273H aberrations are TP53 hot-spot mutations with reported gain-of-function and dominant-negative/loss-of-function properties, respectively (Supplementary Table S1 ) [ 6 , 7 , 8 ]. FR inserted at the A AVS1 safe-harbor locus served as TP53 WT/WT control. Stable, biallelic integration of TP53 cDNAs was achieved in 0.5–10.0% of HSPCs (mean, 2.23 ± 1.76 SD) and enriched by flow cytometry (Supplementary Fig. 1C ). Site-specific integration was confirmed by PCR and each engineered TP53 genotype was validated by Sanger sequencing of both genomic DNA and complementary DNA (Fig. 1B and Supplementary Fig. 3A, B ), p53 protein expression was confirmed by immunocytochemistry (Supplementary Fig. 3C ). Functionally, the introduced TP53 aberrations resulted in reduced or absent expression of the downstream targets p21 and MDM2 (Supplementary Fig. S4 ). Fig. 1: The TP53 allelic state differentially modulates human HSPC growth and self-renewal. Full size image A CRISPR-mediated knock-in of wild-type (WT) or mutant (MUT) TP53 cDNA into cord blood-derived CD34⁺ HSPCs. B Representative genomic DNA chromatograms displaying monoallelic (R175H/WT, R273H/WT, KO/WT) and biallelic (R175H/KO, R273H/KO, KO/KO) TP53 genotypes. C CFUs from different genotypes 14 days after seeding into methylcellulose. D Morphological classification of colonies as BFU-E (red), monocyte (M, gray) or granulocyte/macrophage (GM, white) from primary plating after 14 days. E CFUs from different genotypes upon replating. C – E Data are presented as mean ± s.e.m. of two pooled samples plated in triplicates. Statistical comparisons to AAVS1 control by one-way ( C ) or two-way ( D ) ANOVA. ns not significant; * P < 0.05; ** P < 0.01; *** P < 0.001; **** P < 0.0001. HSPCs hematopoietic stem and progenitor cells, CFUs colony-forming units, BFU-E burst-forming unit erythroid. We initially investigated whether distinct TP53 mutant allelic states affect colony growth and self-renewal using methylcellulose-based colony-forming unit (CFU) assays. Reporter-positive, genome-edited HSPCs with biallelic TP53 aberrations derived from pooled UCB samples generated significantly more colonies than cells with monoallelic aberrations and AAVS1 controls (Fig. 1C, D ). Both CFU-GM and BFU-E colony numbers increased, with erythroid colonies exhibiting a particularly pronounced rise. PCR and Sanger sequencing of colonies showed complete retention of the initially introduced mutations in HSPCs (Supplementary Fig. S5 ). Serial replating CFU assays demonstrated that HSPCs with WT TP53 or monoallelic aberrations lost their colony-forming capacity after three passages, whereas cells with biallelic aberrations displayed elevated clonogenic potential (Fig. 1E ). Cell cycle analysis revealed a non-significant increase in the proportion of cells with biallelic TP53 aberrations in the S-phase of the cell cycle (Supplementary Fig. S6 ). TP53 aberrations are closely linked to structural chromosomal abnormalities (CAs). Using single-cell template strand sequencing of genome-edited HSPCs derived from single UCB donors, we analyzed CAs induced by TP53 aberrations [ 9 ]. Cells were treated with either doxorubicin, a potent cellular stressor, or DMSO as control (Supplementary Figs. S7 , S8 ). Upon doxorubicin treatment, cells with TP53 monoallelic aberrations already exhibited a significantly higher frequency of CAs compared to AAVS1 -targeted controls. In addition, a spontaneous and significant increase in CAs was observed in HSPCs carrying the TP53 R175H mutation (Fig. 2A , Supplementary Fig. S9 ). In HSPCs harboring the TP53 R175H and R273H mutations, chromosomes 3, 5, 7 and 17 were among the most frequently affected. Aberrations of chromosomes 5 and 7 are also common in patients with TP53 mutated myeloid neoplasms. Consistent with the clinical situation, we observed terminal losses (64% of all CAs affecting chromosomes 5 and 7), terminal gains (14%), arm gains (7%), whole chromosome gains (7%) and complex aberrations (8%) in our experimental setting (Fig. 2B , Supplementary Fig. S10 ). Fig. 2: TP53 mutations in human HSPCs drive chromosomal aberrations (CAs) and activate inflammatory transcriptional programs. Full size image A Frequency of CAs in TP53 mutant HSPCs treated with DMSO or doxorubicin; shaded areas of the bars indicate multiple CAs per single nucleus. Statistical comparisons to AAVS1 controls were performed using two-sided Fisher’s exact test. Data were generated from one biological donor per genotype. B Representative CA plots for chromosome 5 loss, chromosome 7 gain and complex aberrations on chromosome 7 with sequencing reads aligned to the positive (Watson, orange) or negative (Crick, green) strands. C Volcano plots depicting differentially expressed genes (DEGs) in monoallelic and biallelic TP53 mutants versus AAVS1 controls. D DeepVenn diagrams showing DEG overlaps. (E) Gene set enrichment analysis (MSigDB hallmark gene sets) highlighting relevant enriched pathways (FDR < 0.25); positive normalized enrichment scores (NES) in red, negative NES in blue. C – E Data were generated from three independent biological donor-derived HSPC pools. To delineate the transcriptional consequences of monoallelic and biallelic TP53 aberrations, we performed RNA sequencing on reporter-positive, genome-edited cells from pooled UCB samples. Compared with AAVS1 controls, the majority of differentially expressed genes (DEGs) were shared between both TP53 aberrant groups (Fig. 2C, D ). Among uniquely expressed genes, monoallelic TP53 aberrant HSPCs—in particular those with the R175H and R273H genotypes—showed upregulation of PD-L1 (CD274) which was confirmed by flow cytometry (Supplementary Fig. S11 ). Biallelic TP53 aberrations were associated with overexpression of HOXA10 (Supplementary Tables S2 and S3 ). Gene Set Enrichment Analysis of shared DEGs highlighted upregulation of oxidative phosphorylation, heme metabolism and inflammatory signaling including TNFα/NF-κB, IL6-JAK-STAT3 and IL2-STAT5 (Supplementary Fig. S12 ), whereas downregulated genes were enriched for mitotic spindle assembly and G 2 M checkpoint regulation (Fig. 2E ). Furthermore, we observed negative enrichment of a set of 135 myeloid-specific genes in HSPCs carrying either monoallelic or biallelic TP53 aberrations, while 100 erythroid-associated genes were positively enriched in both groups compared to T P53 WT HSPCs. The ratio of CEBPA to erythroid-specific transcripts such as GATA1 and PLIN2 was reduced in both monoallelic and biallelic mutant HSPCs, consistent with a shift in lineage commitment as we have observed in the CFU assays (Supplementary Fig. S13 ). In this study, we established a model of mutant TP53 allelic states in human HSPC using CRISPR/Cas9 genome engineering. By employing UCB-derived cells, we were able to investigate the functional consequences of particular TP53 aberrations without confounding effects of an aging cellular microenvironment. We found that genomic aberrations can be triggered by a single TP53 aberration, consistent with dysregulation of mitotic spindle and G 2 M checkpoints, as revealed by transcriptome analysis. The spectrum of CAs mirrors that observed in TP53- mutated myeloid malignancies indicating that these lesions constitute early events in leukemogenesis. Notably, the R175H mutation led to a spontaneous rise in CAs, supporting its previously described gain-of-function activity in acute leukemia models [ 6 ]. The early onset of copy-number alterations has also been reported in patients with Li-Fraumeni syndrome carrying TP53 germline mutations [ 10 ]. Monoallelic TP53 aberrations also induced inflammatory pathways characteristic of preleukemic states. Similar pathway upregulation has been described in heterozygous TP53 mutant HSPCs from patients with myeloproliferative syndromes and was shown to confer a fitness advantage of Trp53 mutant clones in experimental in vivo models [ 11 ]. TP53 mutations have been reported to confer an immunosuppressive phenotype. We observed upregulation of the immune checkpoint ligand PD-L1 in edited HSPCs harboring monoallelic TP53 aberrations, particularly with the R175H and R273H genotypes. PD-L1 expression has also been described in HSPCs of patients with TP53 mutant MDS and AML where it correlates with poor survival outcomes [ 12 ]. However, disruption of a single TP53 allele is insufficient to drive the outgrowth of UCB-derived human HSPCs. Following inactivation of the second allele, TP53 aberrant HSPCs acquired a clonogenic growth and self-renewal advantage. The homeobox transcription factor HOXA10 was identified among upregulated genes that potentially contribute to this transition. In transgenic mice, HOXA10 has been reported to promote both proliferation and lineage commitment of hematopoietic stem cells in a concentration-dependent manner [ 13 ]. Enhanced self-renewal of biallelic mutations, considered a prerequisite for leukemic transformation, has also been described recently in a novel Trp53 mouse model [ 4 ]. These findings provide mechanistic insights into early TP53 -driven transformation of human HSPCs and highlight potential clinical implications. In particular, TP53 mutant CHIP clones expand during cytotoxic therapy, ultimately giving rise to therapy-related myeloid neoplasms that are aggressive disorders associated with a dismal outcome [ 14 ]. In a recent report, patients with solid cancers and TP53 mutant clonal hematopoiesis undergoing chemotherapy were additionally treated with the CDK4/6 inhibitors trilaciclib or palbociclib. Compared with placebo, these agents substantially mitigated the expansion of the mutant clone without causing severe side effects [ 15 ]. Given that even monoallelic TP53 aberrations induce profound functional changes, our data support the application of such strategies targeting mutant TP53 as early as possible—at the monoallelic CHIP stage.