Role of p53 in regulation of hematopoiesis in health and disease

Abstract

Purpose of review

Human aging is associated with an exponential increase in the occurrence of clonal hematopoiesis of indeterminate potential (CHIP). CHIP is associated with increased risks of de novo and therapy-related hematologic neoplasms and serves as a reservoir for leukemic relapse. Somatic mutations in the TP53 gene, which encodes the tumor suppressor protein p53, rank in the top five among genes that were mutated in CHIP. TP53 mutations in CHIP are associated with an increased incidence of myeloid neoplasms such as myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML). This review focuses on mechanisms by which mutant p53 promotes CHIP progression and drives the pathogenesis of MDS and AML. We will also discuss potential therapeutic approaches that can target mutant p53 and improve treatment outcomes of MDS and AML.

Recent findings

TP53 was frequently mutated in individuals with CHIP as well as in patients with MDS and AML. While clinical studies suggest that p53 mutant hematopoietic stem and progenitor cell (HSPC) expansion may predispose the elderly to hematologic neoplasms, the underlying mechanisms are not fully understood. Recent findings suggest that mutant p53 may utilize both cell autonomous and non-cell autonomous mechanisms to promote CHIP development. Furthermore, we and others have demonstrated that several gain of function (GOF) mutant p53 proteins have enhanced oncogenic potential beyond dominant-negative (DN) and loss of function (LOF) effects. Notably, TP53 allelic state has important implications for genome stability, clinical presentation, and outcomes in MDS. Some small molecules reactivating WT p53 tumor suppressor activity show promising effects on some human MDS and AML cells with TP53 mutations in preclinical and early phases of clinical studies.

Summary

TP53 mutations in MDS and AML are correlated with advanced disease, poor prognosis, reduced overall survival, and dismal outcomes. Deep understanding of the functions of mutant p53 proteins is essential to devise effective therapies for patients with myeloid neoplasms and other human cancers with TP53 mutations. Targeting mutant p53 directly or pathways regulated by mutant p53 holds great potential in preventing CHIP progression and treating MDS and AML patients with TP53 mutations.

INTRODUCTION

Clonal hematopoiesis of indeterminate potential (CHIP) occurs when a single mutant hematopoietic stem ad progenitor cell (HSPC) contributes to a significant, measurable clonal proportion of mature blood lineages [1, 2 , 3]. CHIP is associated with an increased risk of hematological malignancies, such as myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML), as well as a higher incidence of other age-related pathological conditions such as cardiovascular disease (CVD) [4-7]. Most of the mutations identified in CHIP are dispersed across the genome. However, five genes, including DNMT3A, TET2, ASXL1, JAK2 and TP53, have disproportionately high numbers of somatic mutations [5-7]. This review will focus on the involvement of p53 in CHIP progression and pathogenesis of myeloid neoplasms, including MDS and AML.

TP53 MUTATIONS IN CLONAL HEMATOPOIESIS OF INDETERMINATE POTENTIAL

TP53 ranks in the top five among genes that were mutated in CHIP [5-7]. The majority of TP53 mutations found in CHIP are missense mutations, whereas the remainder are nonsense, frameshift, and splice site mutations, leading to TP53 loss [5-7]. TP53 mutation spectrums in CHIP are similar to that of MDS and AML (Fig. 1) [8^■-12 ]. Given that different mutant p53 proteins have been shown to exhibit context - dependent roles in promoting cancer initiation, progression, or metastasis [13-14], we introduced several hot-spot TP53 mutations identified in CHIP, MDS, and AML, including p53^Y220C, p53^R248W, and p53^R273H, into wild-type (WT) mouse HSPCs using retrovirus-mediated transduction and performed in vitro and in vivo assays. We found that ectopic expression of these mutant p53 proteins enhance the replating potential of WT HSPCs [15^■]. Further, we discovered that these mutant p53 proteins increase the engraftment of W T HSPCs in transplantation assays, demonstrating that mutant p53 enhances HSPC self-renewal in vivo [15^■-16].

Figure 1.

Hot-spot TP53 mutations in clonal hematopoiesis of indeterminate potential (CHIP), myelodysplastic syndromes (MDS), and acute myeloid leukemia (AML).

Cell autonomous mechanisms underlying p53 mutant HSPC expansion

Given that codon 248 of the p53 protein (p53^R248) is most frequently mutated in CHIP, MDS, and AML [5-12], we focused our investigation on p53^R248W in clonal hematopoiesis. We found that p53^R248W enhances the repopulating potential of HSPCs, thereby conferring a competitive advantage to HSPCs following transplantation [15^■-16]. In addition, we showed that chemotherapy and radiation promote p53^R248W HSPC expansion in vivo [16].

To understand how mutant p53 enhances HSPC self-renewal, we performed RNA-seq studies to compare gene expression in HSPCs isolated from p53^+/+ and p53^R248W/+ mice. We then employed Gene Set Enrichment Assays (GSEA) to group potential mutant p53 target genes into specific pathways important in HSPC behavior. We found that epigenetic factor EZH2 target genes are significantly downregulated in p53 mutant HSPCs [15]. EZH2 is a key component of the Polycomb repressive complex 2 (PRC2) that catalyzes the methylation of histone H3K27me3, a repressive histone mark [17-18]. Our findings suggest that EZH2 activity may be increased in p53 mutant HSPCs. That is indeed the case. Western blot analysis showed increased levels of H3K27me3 in p53 mutant HSPCs compared to WT HSPCs. Further, we found that several mutant p53 proteins, including p53^R175H, p53^R248W and p53^R273H, show enhanced association with EZH2 compared to WT p53 [15]. Thus, we have uncovered a cell autonomous mechanism novel by which mutant p53 drives clonal hematopoiesis (Fig. 2A). Our findings also underscore the importance of dysregulated epigenetic control in CHIP development.

Figure 2.

TP53 mutations identified in CHIP utilize both cell autonomous and non-cell autonomous mechanisms to promote hematopoietic stem and progenitor cell (HSPC) expansion during aging. Expanded mutant HSPCs become myelodysplastic syndromes (MDS) stem cells after acquiring additional genetic changes. (A). Mutant p53 interacts with epigenetic regulator EZH2 and increases the levels of H3K27me3 in HSPCs, leading to downregulation of Gadd45g and Cebpα that promotes mutant HSPC expansion. (B). Mutant p53 enhances inflammatory response of HSPCs, leading to increased secretion of pro-inflammatory cytokines such as IL-1β and IL-6 that inhibit wild-type (WT) HSPCs in a paracrine manner.

Non-cell autonomous mechanisms underlying p53 mutant HSPC expansion

While we have identified a cell autonomous mechanism by which mutant p53 drives clonal hematopoiesis [15-16], recent studies indicate that mutations identified in CHIP may utilize non-cell autonomous mechanisms to promote clonal hematopoiesis [19]. During aging, chronic and low-grade inflammation - inflammaging - develops, which may contribute to the pathogenesis of age-related diseases such as CHIP and MDS [20]. People with CHIP show an aberrant systematic inflammatory milieu [21^■-22^■] and aberrant innate immune activation and pro-inflammatory signaling have been identified as key pathogenic drivers of MDS [19].

A growing body of evidence suggest that dysregulated inflammation may contribute to clonal expansion of mutant HSPCs. For example, we and others found that microbial signals drive the clonal expansion of TET2 mutant HSPCs [23-24]. Recently, chronic infection has been shown to promote Dnmt3a-null HSPC expansion, potential through IFNγ signaling pathway [25^■]. Avagyan and colleagues studied how acquired mutations affect clonal fitness in a native hematopoietic environment and discovered that resistance to inflammation underlies enhanced fitness seen in mutant HSPCs [26^■].

We examined the impact of inflammation on the function of p53 mutant HSPCs and found that chronic inflammation induced by infections promote p53 mutant HSPC expansion. Notably, p53 mutant HSPCs are resistant to inflammatory stress-induced by infections. Mechanistically, chronic inflammation activates inflammatory response in p53 mutant HSPCs, leading to increased secretion of pro-inflammatory cytokines that inhibit wild-type HSPC function in a paracrine fashion (S.B. and Y.L. unpublished data). The growth advantage seen in p53 mutant HSPCs is likely due to elevated inflammatory signaling in both mature hematopoietic cells (supplying pro-inflammatory cytokines) and HSPCs (supplying and responding to pro-inflammatory cytokines), which likely forms a feed-forward loop to promote CHIP development during aging (Fig. 2B).

TP53 MUTATIONS IN THE PATHOGENESIS OF MYELODYSPLASTIC SYNDROMES AND ACUTE MYELOID LEUKEMIA

TP53 mutations in MDS and AML

TP53 is the most frequently mutated gene in human cancer [13-14]. In patients with MDS, TP53 mutations are associated with high-risk disease, reduced overall survival, rapid transformation to AML, and dismal outcomes [8^■-10]. MDS patients have both mono- and biallelic TP53 mutations. Recently, Bernard and colleagues investigated the biological and clinical implications of TP53 allelic state in patients with MDS. One-third of TP53-mutated MDS patients have mono-allelic mutations whereas two-thirds have multiple hits (multi-hit) consistent with bi-allelic targeting [8^■]. Notably, multi-hit TP53-mutated MDS patients show complex karyotype, few co-occur mutations, high risk presentation, and poor outcomes compared to MDS patients with mono-allelic mutations [8^■]. While TP53 multi-hit state predict risk of death and leukemic transformation, monoallelic patients did not differ from TP53 wild-type patients in outcomes and response to therapy [8^■]. These studies underscore the importance of TP53 allelic state for diagnosis and prognostic precision in MDS.

In patients with AML, TP53 mutations are associated with complex karyotype, advanced disease, reduced overall survival, resistance to conventional therapies, and dismal outcomes [11-12]. Similar to MDS patients, AML patients have both mono- and biallelic TP53 mutations [11-12]. However, the biological and clinical implications of TP53 allelic state have not been fully investigated in AML.

Dominant negative (DN) and gain of function (GOF) mutant p53 proteins

Most TP53 mutations observed in human cancers abrogate or attenuate binding of p53 to its consensus DNA sequence (p53 responsive element) and impede transcriptional activation of p53 target genes, leading to partial or complete loss of tumor suppressor function (LOF). Some mutant p53 proteins have dominant-negative (DN) effects, inhibiting the remaining wild-type (WT) p53 allele [13-14]. However, mounting evidence demonstrates that some mutant p53 proteins not only lose their tumor suppressor function, but also acquire new oncogenic properties that are independent of WT p53, known as gain-of-function (GOF) properties [13-14]. To ascertain the oncogenic effects of GOF TP53 mutations, p53^R248W and p53^R273H mutant were engineered into the endogenous Trp53 locus in mice [27]. Homozygous p53^R248W/R248W and p53^R273H/R273H mice developed novel tumors compared to p53^−/− mice [27]. Thus, GOF mutant p53 proteins have enhanced oncogenic potential beyond the simple loss of p53 function in vivo. Deep understanding of the functions of mutant p53 proteins will be essential to devise effective therapies for patients with myeloid neoplasms and other human cancers with TP53 mutations [13-14].

Boettcher and colleagues utilized CRISPR-Cas9 to generate isogenic human leukemia cell lines, including K562 and Molm-13, of hot-spot TP53 mutations identified in myeloid malignancies. RNA-seq, ChIP-seq, and functional assays revealed that these mutants have lost their p53 response element binding, suggesting that they are LOF mutants [28^■]. Mutational scanning of p53 single-amino acid variants showed that mutations in the DNA binding domain of p53 appear to have DN effect over the remaining wild-type p53 allele. Boettcher and colleagues thus concluded that a DN effect drives the selection of TP53 mutations in myeloid malignancies [28^■]. As both K562 and Molm-13 have been fully transformed, there may not depend on mutant p53 for transcriptional regulation. AML is a stem cell disease that originates from a small population of leukemia-initiating cells (LICs) after stepwise acquisition of genetic and/or epigenetic alterations in a normal hematopoietic stem and progenitor cell (HSPC) [11-12]. However, the impact of these hot-spot mutant proteins on human HSPCs were not investigated in the study. Most (neomorphic) GOF properties are believed to stem from binding of mutant p53 to cellular proteins such as transcription factors and altering their activity [13-14]. However, proteins interacting with mutant p53 proteins were not examined in human leukemia cells. Given that AML is a heterogeneous blood disorders [11-12], it is possible that some mutant p53 proteins have GOF properties in pathogenesis of some AML subtypes.

We and others have demonstrated that GOF mutant p53 proteins have enhanced oncogenic potential beyond DN and LOF effects [15^■-16, 29-30^■]. Genome-wide transcriptome assays revealed that hematopoietic stem cell (HSC) and AML signatures are enriched in p53^R248W mutant HSPCs, which is different from gene expression signatures regulated by the wild-type p53 protein [15^■]. Thus, our findings provide experimental evidence that TP53 mutations identified in CHIP and MDS regulate gene expression in a distinct manner compared to WT p53. In addition, we found several mutant p53 proteins, including p53^R175H, p53^R248W and p53^R273H, show enhanced association with EZH2 compared to WT p53. Further, we found that ectopic expression of p53^R248W, but not WT p53, increases the levels of H3K27me3 in hematopoietic cells [15^■]. Notably, we found that mutant p53 enhances inflammatory response, leading to increased secretion of pro-inflammatory cytokines that inhibit WT HSPC function (S.B. and Y.L., unpublished data). Since p53 mutant HSPCs show distinct features that are absent in p53^+/− and p53^−/− HSPCs, our findings suggest that some TP53 mutations identified in CHIP and MDS have GOF properties.

Loizou and colleagues identified a GOF mutant p53 (p53^R175H) that accelerates the initiation of complex karyotype AML (CK-AML) in mice [30^■]. Notably, the GOF p53 mutant is required for disease maintenance. Similar to what we found in p53 ^R248W knock-in mice [15^■-16, 29], p53^R175H enhances the self-renewal potential of both normal HSCs and LICs. Mechanistically, they identified pluripotency factor FOXH1 as a key mediator of mutant p53 function in HSCs. In summary, Loizou and colleagues demonstrates that a GOF p53 mutant can hijack an embryonic transcription factor to promote aberrant self-renewal [29]. Given that GOF mutant p53 proteins have been commonly found in solid tumors [13-14, 27], we expect to observe more experimental and clinical evidence on GOF p53 mutants in pathogenesis of hematological malignancies in the future.

TARGETING MUTANT p53 TO PREVENT CLONAL HEMATOPOIESIS OF INDETERMINATE POTENTIAL PROGRESSION AND TREAT HEMATOLOGICAL MALIGNANCIES

While TP53 is mutated in more than 50% human cancer, no drug that abrogates the oncogenic functions of mutant p53 has yet been approved for cancer treatment [13-14]. MDS and AML patients with TP53 mutations are characterized by frequent relapses, poor or short responses, and poor survival with the currently available therapies including chemotherapy and 5-azacitidine (AZA) [8-12]. However, only 20% of MDS patients with TP53 mutations achieve complete remission (CR) with hypomethylating agents [8-10]. Thus, there is an urgent need to develop therapeutic strategies that can directly target mutant p53 or pathways regulated by mutant p53.

Small molecules reactivating WT p53 tumor suppressor activity

Eprenetapopt (APR-246) is a novel, first-in-class, small molecule that induces apoptosis in human tumor cells through restoration of the transcriptional transactivation function of mutant p53 [31]. Maslah and colleagues found that low doses of APR-246 on its own or in combination with AZA reactivate the p53 pathway and induce an apoptosis program in human MDS and AML cells [32]. In addition, low doses of APR-246 on its own or in combination with AZA show significant efficacy in patient-derived xenograft (PDX) models of AML in vivo [32]. These preclinical studies suggest that TP53-mutated MDS/AML may be targeted by the addition of APR-246 to conventional treatments.

Sallman and colleagues performed a phase Ib/II study to determine the safety, recommended phase II dose, and efficacy of eprenetapopt administered in combination with AZA in MDS or AML patients with TP53 mutations [33^■]. The overall response rate was 71% with 44% achieving CR. Of patients with MDS, 73% responded with 50% achieving CR and 58% a cytogenetic response. The overall response rate and CR rate for patients with AML was 64% and 36%, respectively [33^■]. These findings suggest that combination treatment with eprenetapopt and AZA is well-tolerated, yielding high rates of clinical response in MDS and AML patients with TP53 mutations [33^■].

While TP53 mutations are rarely found in patients with acute lymphoblastic leukemia (ALL), they are associated with poor therapy response and occur at higher frequency in relapsed ALL [34]. As asparaginase depletes extracellular asparagine in the blood, it has become an important therapy for ALL. However, resistance to asparaginase often occurs due to high expression of asparagine synthetase (ASNS) in ALL, which generates asparagine from intracellular sources [34]. Recent findings suggest that ASNS is a direct target of APR-246 and that APR-246 synergizes with asparaginase in inducing growth suppression of human ALL cells. [34]

Major challenges in drugging p53 mutations include heterogeneous mechanisms of inactivation and the absence of broadly applicable allosteric sites [13-14]. Chen and colleagues have identified arsenic trioxide (ATO), an established agent in treating acute promyelocytic leukemia (APL), as cysteine-reactive compound that rescues structural TP53 mutations [35^■]. Crystal structures of arsenic-bound p53 mutants reveal that arsenic binding stabilizes the DNA-binding loop-sheet-helix motif alongside the overall β-sandwich fold, endowing p53 mutants with transcriptional activity [35^■]. However, most hot spot p53 mutants are stabilized structurally and only some p53 mutants can be transcriptionally rescued by ATO. ATO reactivates mutant p53, leading to tumor suppression in mouse xenograft models [35]. Given that ATO is an FDA approved drug, these findings suggest that ATO may be used clinically to treat some cancer patients with TP53 mutations.

EZH2 inhibitors

We found that mutant p53 interacts with epigenetic factor EZH2 and increases the levels of H3K27me3 in HSPCs [15^■]. Further, we found that genetic and pharmacological inhibition of EZH2 decrease the repopulating potential of p53 mutant HSPCs [15^■]. As EZH2 is rarely mutated in CHIP [5-7], our findings suggest that EZH2 may be a novel target for preventing CHIP progression in aged individuals with TP53 mutations. The impact of EZH2 inhibitors on human MDS and AML cells with TP53 mutations will require further investigation.

CONCLUSION

Tumor suppressor gene TP53 was frequently mutated in individuals with CHIP as well as in patients with MDS and AML. Mutant p53 proteins appear to utilize both cell autonomous and non-cell autonomous mechanisms to promote CHIP development. Recent findings suggest that some mutant p53 proteins have enhanced oncogenic potential (i.e. gain of function) beyond dominant-negative and loss of function effects in CHIP and myeloid neoplasms. Thus, targeting mutant p53 directly or pathways regulated by mutant p53 holds great potential in preventing CHIP progression and treating MDS and AML patients with TP53 mutations.

KEY POINTS.

Tumor suppressor gene TP53 was frequently mutated in individuals with CHIP as well as in patients with MDS and AML.

Mutant p53 proteins utilize both cell autonomous and non-cell autonomous mechanisms to promote CHIP development.

Gain of function (GOF) mutant p53 proteins have enhanced oncogenic potential beyond dominant-negative (DN) and loss of function (LOF) effects in CHIP development and pathogenesis of myeloid neoplasms.

Targeting mutant p53 directly or pathways regulated by mutant p53 holds great potential in preventing CHIP progression and treating MDS and AML patients with TP53 mutations.