A combinatorial therapeutic approach to enhance FLT3-ITD AML treatment

Summary

Internal tandem duplication mutations of the FMS-like tyrosine kinase-3 (FLT3-ITDs) occur in 25%–30% of patients with acute myeloid leukemia (AML) and are associated with dismal prognosis. Although FLT3 inhibitors have demonstrated initial clinical efficacy, the overall outcome of patients with FLT3-ITD AML remains poor, highlighting the urgency to develop more effective treatment strategies. In this study, we reveal that FLT3 inhibitors reduced protein stability of the anti-cancer protein p53, resulting in drug resistance. Blocking p53 degradation with proteasome inhibitors restores intracellular p53 protein levels and, in combination with FLT3-ITD inhibitors, shows superior therapeutic effects against FLT3-ITD AML in cells, mouse models, and patients. These data suggest that this combinatorial therapeutic approach may represent a promising strategy to target FLT3-ITD AML.

Graphical abstract

Highlights

• •FLT3 inhibition downregulates p53 protein in FLT3-ITD AML to promote drug resistance
• •MYC/USP10 signaling downstream of FLT3-ITD regulates p53 stability
• •Proteasome inhibitors restore p53 protein to alleviate resistance to FLT3 inhibitors
• •Dual-drug strategy is efficacious in treating relapsed/refractory FLT3-ITD AML

Long et al. find that FLT3 inhibition induces p53 ubiquitination and proteasomal degradation through MYC/USP10 signaling, leading to drug resistance. Proteasome inhibitor restores p53 protein and corroborates with FLT3 inhibitor to eliminate FLT3-ITD AML. Salvage therapy with FLT3 plus proteasome inhibitors shows extraordinary therapeutic efficacy in patients with relapsed/refractory FLT3-ITD AML.

Introduction

Acute myeloid leukemia (AML) is a hematological malignancy characterized by abnormal differentiation and clonal proliferation of hematopoietic cells.¹ One of the most frequent mutations in AML involves the internal tandem duplication (ITD) of FMS-like tyrosine kinase-3 (FLT3) and is associated with adverse prognosis. FLT3-ITD mutation results in constitutive FLT3 phosphorylation and hyperactivation of downstream survival/proliferation signaling pathways.² Several FLT3 inhibitors have been approved for treatment of patients with AML with FLT3 mutations and have demonstrated good initial response. However, the duration of the clinical response to FLT3 inhibitors is often short lived, and most patients eventually relapse with FLT3-inhibitor-resistant disease,^3,4 highlighting the urgency to identify potential mechanisms of such drug resistance and to develop additional therapeutic strategies.

The tumor-suppressor p53 plays a key role in maintaining genome integrity and preventing tumorigenesis. TP53 aberrations are early leukemogenic-initiating driver lesions and usually predict poor responses to therapies and adverse prognosis in AML.⁵ Although TP53 mutations occur in only ∼8% of de novo AML incidences,⁶ recent studies revealed that wild-type (WT) p53 protein functional deficiency appears to be rather frequent in various AML entities.^7,8 For example, the p53 protein E3 ligase MDM2 is frequently overexpressed in AML, which retains WT TP53 alleles, mediating WT-p53 inactivation and contributing to dismal prognosis.^8,9 MDM2 inhibitors, which address dysfunctional WT-p53 protein, have demonstrated therapeutic activity against AML,^10,11 indicating that restoring the normal tumor-suppressive function of WT-p53 is beneficial for AML therapy. In AML with FLT3-ITD mutations, WT-p53 function is misregulated via multiple FLT3-ITD-related mechanisms,⁷ and MDM2 inhibition exhibited superior effects against FLT3-ITD AML than FLT3-WT subtypes in preclinical studies.¹² However, the status of p53 protein in FLT3-ITD AML upon FLT3 inhibition has not been thoroughly defined.

Here, we found that FLT3 inhibition reduced WT-p53 protein stability and inhibited p53 activity in FLT3-ITD AML, leading to drug resistance. Downregulation of p53 protein is attributed to elevated ubiquitination of p53. Blocking p53 degradation with proteasome inhibitors restored intracellular protein levels and tumor-suppressive function of p53, which subsequently enhanced the leukemia-killing effects of FLT3 inhibitors against FLT3-ITD AML in in vitro cultures and patient-derived xenograft (PDX) models, as well as in patients. Collectively, these data revealed that p53 degradation might be the major underlying mechanism for drug resistance to FLT3 inhibition in FLT3-ITD AML. Our study also postulated a feasible combinatorial therapeutic strategy to improve FLT3-ITD AML treatment.

Results

FLT3 inhibition downregulates p53 signaling in FLT3-ITD AML

To define the potential mechanisms underlying resistance to FLT3 inhibition (Figure 1A), we first examined intracellular signaling network changes in FLT3-ITD AML upon FLT3 inhibition with an antibody microarray. Quizartinib (a potent FLT3 inhibitor) treatment of FLT3-ITD-expressing MV4-11 cells inhibited many cancer-related signaling pathways (Figures S1A and S1B; Table S1). Unexpectedly, quizartinib also reduced the expression of several proteins related to the p53 signaling pathway, including the pro-apoptotic proteins p53, caspase-3, and BID (Figure 1B). Remarkably, p53 was one of the most significantly downregulated proteins (Figure 1C). Consistently, gene set enrichment analysis (GSEA) of transcriptional datasets from FLT3-ITD-expressing MV4-11 and MOLM-13 cells also revealed that p53 signaling was inhibited by quizartinib (Figures 1D, S1C, and S1D).^13,14,15

Figure 1.

FLT3 inhibition downregulates p53 signaling

(A) Schematic outline of this study. Though FLT3 inhibitors have demonstrated good clinical responses by inhibiting constitutively activated FLT3 signaling, the remaining residual leukemia cells may still develop mechanisms to survive and serve as a source for disease relapse.

(B) Expression of proteins related to p53 signaling pathway in MV4-11 cells treated with 5 nM quizartinib for 24 h based on the antibody microarray.

(C) Volcano plot of protein expression changes in MV4-11 cells post-5 nM quizartinib treatment for 24 h. MV4-11 cells post-5 nM quizartinib treatment for 24 h. Significance was defined as p <0.05 by multiple t test and ratio (quizartinib/vehicle) >1.2 or <1/1.2.

(D) GSEA plots depicting suppressed p53 signaling in MV4-11 cells upon quizartinib treatment. FDR, false discovery rate; NES, normalized enrichment score.

(E) Immunoblot of indicated proteins in MV4-11 and MOLM-13 cells treated with 5 nM quizartinib for 24 h.

(F) Immunoblot of p53 in primary blasts from twelve patients with FLT3-ITD AML treated with 20 nM quizartinib for 24 h.

(G) Quantitative real-time PCR of TP53 in MV4-11 and MOLM-13 cells treated with 5 nM quizartinib for 24 h (n = 3; mean ± SD; paired t test).

(H) Quantitative real-time PCR of TP53 in primary blasts from patients with FLT3-ITD AML (n = 12) treated with 20 nM quizartinib for 24 h (paired t test).

(I and J) MV4-11 cells and MOLM-13 cells were treated as indicated (10 nM daunorubicin [DNR]; 500 nM cytarabine [Ara-C] for MV4-11, 50 nM Ara-C for MOLM-13; 5 nM quizartinib [Quiz]) for 24 h: immunoblot of indicated proteins (I), and quantitative real-time PCR of TP53 and target genes (J). For (J), n = 3, mean ± SD, and one-way ANOVA was used. ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001 by one-way ANOVA.

To confirm that FLT3 inhibition downregulates p53 signaling, we treated MV4-11 and MOLM-13 cells (containing FLT3-ITD) with quizartinib and demonstrated that FLT3 inhibition indeed reduced p53 protein levels in both cell lines (Figure 1E). Knockdown of FLT3 with small interfering RNA (siRNA) or blockage of FLT3 activity with 2 other FLT3 inhibitors, crenolanib and gilteritinib, decreased p53 protein as well (Figures S1E and S1F). However, quizartinib did not affect the p53 protein in FLT3-WT OCI-AML2 or OCI-AML3 cells (Figure S1G). These data support that p53 downregulation is indeed due to inhibition of FLT3-ITD signaling rather than off-target effects. Treatment of primary human FLT3-ITD AML samples with quizartinib also resulted in p53 downregulation (Figure 1F). However, quizartinib did not alter TP53 mRNA levels (Figures 1G and 1H). We further examined whether FLT3 inhibition would affect transcriptional function of p53 and found that neither pharmacological inhibition nor knockdown of FLT3 altered mRNA levels of p53 target genes (Figures S1H and S1J). It is worth mentioning that in FLT3-ITD AML samples, p53 transcriptional function is already dampened by p53 post-translational modifications as well as interactions with other proteins such as BCL2 and MDM2,^7,8 FLT3 inhibition, while may alleviate some of these aversive influences on p53, triggered a reduction of p53 protein levels, which sustained p53 dysfunction. On the contrary, conventional chemotherapy drugs, such as daunorubicin (DNR) and cytarabine (Ara-C), and γ-irradiation induced p53 signaling, as indicated by elevated protein and mRNA levels of p53 as well as mRNA levels of p53 targets, while co-treatment with quizartinib significantly suppressed p53 upregulation and reduced p53 transcriptional activity (Figures 1I, 1J, S1K, and S1L).

p53 downregulation promotes FLT3-ITD AML drug resistance

We further investigated the effects of p53 downregulation on drug resistance of FLT3-ITD AML. We first analyzed the time course of p53 downregulation and apoptosis onset upon FLT3 inhibition. p53 protein levels started to decrease at 6 h post-quizartinib treatment, while PARP cleavage began 9 h post-quizartinib treatment (Figure S2A). On the other hand, annexin V/PI staining revealed that both MV4-11 and MOLM-13 cells only demonstrated mildly increased apoptotic rates at 24 h post-quizartinib treatment (Figure S2B), indicating that downregulation of p53 preceded apoptosis onset. Using the Beat AML dataset,¹⁶ we found that in AML patient samples, TP53 mRNA levels correlated positively with sensitivity to quizartinib and midostaurin (low area under curve [AUC]; Figure S2C). Short hairpin RNA (shRNA)-mediated knockdown of TP53 significantly increased the half-maximal inhibitory concentration (IC₅₀) of FLT3 inhibitors as well as chemotherapy drugs in MV4-11 and MOLM-13 cells (Figures 2A and S2D–S2G). TP53-depleted MV4-11 and MOLM-13 cells were more resistant to quizartinib- and/or chemotherapy-drug-induced apoptosis (Figures 2B and S2H), whereas ectopic expression of WT-p53, but not an acetylation-defective p53 mutant (a dysfunctional p53 mutant with 8 potential acetylation sites mutated, p53-8KR),¹⁷ re-sensitized these cells to quizartinib and chemotherapy drugs (Figures 2C and S2I). Collectively, these data support that p53 downregulation at protein levels induced by FLT3 inhibition promotes FLT3-ITD AML drug resistance.

Figure 2.

p53 downregulation promotes drug resistance

(A) MV4-11 and MOLM-13 cells were exposed to escalating doses of Quiz for 48 h. Viability was determined by CCK-8. The half-maximal inhibitory concentration (IC₅₀) was calculated by non-linear regression (n = 3; mean ± SD).

(B) MV4-11 and MOLM-13 cells transduced with shRNAs against TP53 (shTP53-1 or -2) or scramble shRNA (shSCR) were treated with 5 nM Quiz for 48 h. Cell apoptosis was analysis by annexin V/PI staining. Asterisks indicate statistical comparisons versus shSCR (n = 3; mean ± SD; unpaired t test).

(C) MV4-11 and MOLM-13 cells with TP53 knockdown were transduced with lentiviral vectors expressing wild-type p53 (p53-WT) or acetylation-defective p53 mutant (p53-8KR). The cells were treated 5 nM Quiz for 48 h. Cell apoptosis was analysis by annexin V/PI staining. Asterisks indicate statistical comparisons versus vector (n = 3; mean ± SD; unpaired t test). ns p > 0.05; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001. ns, not significant.

Blocking p53 protein degradation with proteasome inhibitors improves sensitivity to FLT3 inhibition

To investigate the mechanisms underlying FLT3-inhibition-induced downregulation of p53 protein, given that FLT3 inhibition did not alter TP53 mRNA levels, we analyzed p53 protein stability and found that p53 demonstrated an extended half-life in FLT3-ITD-expressing AML cell lines compared to cell lines with FLT3-WT (Figure S3A). Introducing FLT3-ITD into normal CD34⁺ hematopoietic stem/progenitor cells (HSPCs) led to elevated p53 protein levels and enhanced protein stability, without affecting p53 mRNA levels (Figures S3B–S3D). In contrast, cycloheximide (CHX) chase assay revealed that FLT3 inhibition remarkably reduced the stability of p53 protein (Figure 3A). Moreover, FLT3 inhibition increased total and K48-specific ubiquitination, but not sumoylation or serine/threonine phosphorylation, of p53 (Figures 3B, S3E, and S3F). Collectively, these data suggest that inhibition of FLT3 may reduce p53 protein stability through promoting its ubiquitination.

Figure 3.

Blocking p53 degradation with proteasome inhibitor enhances the elimination of FLT3-ITD AML in combination with FLT3 inhibitor

(A) MV4-11 and MOLM-13 cells were treated with vehicle or Quiz (5 nM) overnight and then were exposed to cycloheximide (CHX) for indicated minutes. The expression of p53 was analyzed by western blot. Right: the relative amount of p53 analyzed by densitometry (n = 3; mean ± SD; multiple t tests).

(B) MV4-11 and MOLM-13 cells were treated with 5 nM Quiz for 24 h. Protein lysates were immunoprecipitated with p53 antibody and then immunoblotted for K48-speicfic ubiquitin (ub) and p53.

(C) Schematic of FLT3 inhibitor and proteasome inhibitor synergy assessment.

(D) Immunoblot of indicated proteins in primary blasts from 2 patients with FLT3-ITD AML were treated as indicated for 24 h (20 nM Quiz; 10 nM bortezomib).

(E) Primary AML blasts from patients with FLT3-ITD (n = 4) were treated as indicated for 48 h (20 nM Quiz; 10 nM bortezomib). Cell survival was analyzed by annexin V/PI staining (mean ± SD; one-way ANOVA).

(F) Count of colonies formed by primary AML blasts from patients with FLT3-ITD AML (n = 4) after indicated treatment (20 nM Quiz; 10 nM bortezomib) (mean ± SD; one-way ANOVA).

(G) Immunoblot of indicated proteins in primary blasts from a patient with chemo- and sorafenib-refractory FLT3-ITD AML after treatment as indicated (0.5 μM sorafenib; 20 nM ixazomib) for 24 h.

(H) Primary AML blasts from (G) were treated as indicated (0.5 μM sorafenib; 20 nM ixazomib) for 48 h. Cell survival was analyzed by annexin V/PI staining (n = 3; mean ± SD; one-way ANOVA). ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001.

We reasoned that suppressing the ubiquitin-proteasome pathway may restore the expression and function of intracellular p53 and collaborate with FLT3 inhibitors to eliminate FLT3-ITD AML (Figure 3C). We therefore treated MV4-11 and MOLM-13 cells with quizartinib and/or bortezomib. Bortezomib effectively prevented quizartinib-induced downregulation of intracellular p53 protein levels (Figure S3G). p53 target genes were also significantly induced upon quizartinib and bortezomib combination (Figure S3H), indicating activated p53 signaling. Consistently, FLT3 inhibitors plus proteasome inhibitors substantially reduced survival of malignant cells (Figures S3I–S3K). TP53 knockdown could reverse the effect of quizartinib plus bortezomib or ixazomib on leukemia cell apoptosis (Figure S3I). Quizartinib and bortezomib co-treatment also significantly induced intracellular p53 protein as well as reduced cell survival and colony-forming capacity of primary FLT3-ITD AML cells (Figures 3D–3F). We further investigated another patient sample with chemotherapy-refractory FLT3-ITD AML. This patient also received standard sorafenib treatment, but the percentage of leukemic blasts in bone marrow fluctuated between 20% and 40%. Sorafenib plus ixazomib substantially induced p53 protein expression and reduced cell survival (Figures 3G and 3H).

Restoration of p53 protein enhances the elimination of FLT3-ITD AML in PDX model

We further tested the in vivo effects of combined FLT3 inhibitor and proteasome inhibitor on PDX models engrafted with primary human FLT3-ITD AML samples (Figure 4A). Following confirmation of human leukemia engraftment in mouse peripheral blood (>5% human CD45⁺ cells), mice were administered with vehicle, quizartinib (10 mg/kg daily, gavage), bortezomib (0.5 mg/kg twice weekly, intraperitoneal), or the combination for 4 weeks. Engrafted leukemia cells expressed human CD45 and CD33 (Figure S4A). Quizartinib or bortezomib treatment reduced leukemia burden in PDX mice, while in mice treated with both drugs, more substantial reductions were observed (Figures 4B and 4C). Quizartinib and bortezomib combination also significantly prolonged the survival period of PDX mice (Figure 4D). Notably, quizartinib plus bortezomib reduced the proportion and total numbers of human CD45⁺CD34⁺ leukemia cells more extensively (Figures 4E and S4B–S4D). Since CD34 positivity marks a subgroup of cells with leukemia-initiating capacity, we evaluated effects of combinatorial treatment by secondary transplantation. Although we observed a significant leukemia burden in mice receiving vehicle-, quizartinib-, or bortezomib-treated transplants, there were much fewer residual human CD45⁺ leukemia cells in dual-drug-treated secondary transplants (Figure S4E), indicating that the dual-drug strategy had more potent therapeutic effects against leukemia stem cells.

Figure 4.

The in vivo therapeutic effect of FLT3 and proteasome inhibitors against FLT3-ITD AML

(A) Schematic depiction of in vivo combination therapy experimental protocol using FLT3-ITD AML patient-derived xenograft (PDX) model.

(B and C) Percentage (B) and number (C) of human CD45⁺ cells in the bone marrow (BM) of treated mice (n = 6 per group; mean ± SD).

(D) Kaplan-Meier survival of mice transplanted with 2 different samples is displayed, respectively (n = 10 per group).

(E) Number of human CD45⁺CD34⁺ cells in the BM of treated mice (n = 6 per group; mean ± SD).

(F) Percentage of human CD45⁺ cells in the BM of treated mice (n = 6 per group; mean ± SEM).

(G) Kaplan-Meier survival of mice transplanted with a chemo- and sorafenib-refractory sample (n = 10 per group).

(H) Number of human CD45⁺CD34⁺ cells in the BM of treated mice (n = 6 per group; mean ± SEM).

For (B), (C), (E), (F), and (H), one-way ANOVA was used. For (D) and (G), Mantel-Cox test was used. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001.

To examine whether the dual-drug strategy would also benefit patients with chemotherapy- and FLT3-inhibition-refractory FLT3-ITD AML, using corresponding PDX mice derived from a chemo- and sorafenib-refractory patient sample, we observed a significant reduction in leukemia burden and a remarkable elongation of survival period in the sorafenib (50 mg/kg twice daily for 4 weeks, gavage) plus ixazomib (1 mg/kg twice weekly for 4 weeks, gavage)-treated group compared to groups treated with a single drug or standard chemotherapy (Ara-C 100 mg/kg daily for 5 days, intraperitoneal; DNR 2.5 mg/kg daily for 3 days, intravenous) (Figures 4F, 4G, and S4F). Sorafenib plus ixazomib also demonstrated better therapeutic effects against the CD45⁺CD34⁺ leukemia stem cells (Figures 4H and S4G). Besides, sorafenib plus ixazomib in vivo treatment significantly induced p53 protein expression in human CD45⁺ leukemia cells (Figure S4H). Taken together, these data provide a strong rationale for combining proteasome inhibition and FLT3 inhibition to treat FLT3-ITD AML.

Salvage therapy for patients with relapsed/refractory FLT3-ITD AML

We performed pilot clinical studies using dual-drug strategies to treat two patients with FLT3-ITD AML for salvage purposes (Figure 5A). The first patient suffered from relapse after allogeneic hematopoietic stem cell transplantation (HSCT). Bone marrow aspiration revealed 54% leukemia blasts. Gene mutation panel discovered a FLT3-ITD mutation with an allelic ratio of 1.53. The patient was administered decitabine, sorafenib, and venetoclax for two cycles; however, peripheral blood blasts persisted around 20%, and the ratio of bone marrow blasts increased at the end of the second cycle, implying the possibility of treatment failure (Figure 5B). Based on our compelling preclinical data, compassionate treatment of this patient was initiated with sorafenib (400 mg twice daily) plus off-label/off-trial ixazomib (4 mg on days 1, 8, and 15), in compliance with all relevant ethical regulations. The treatment resulted in a clearance of blasts from the peripheral blood and complete remission at the end of the treatment course (Figures 5C and 5D). The patient then underwent a second allogeneic HSCT and continued to be in complete remission. The second patient had undergone several chemotherapy regimens, but complete remission was never achieved (Figure 5E). The treatment options had been exhausted. Therefore, compassionate treatment of sorafenib plus ixazomib was initiated. The patient achieved complete remission after one cycle treatment with sorafenib plus ixazomib (Figure 5E) and was able to receive a subsequent allogeneic HSCT.

Figure 5.

Treatment of two patients with FLT3-ITD AML with sorafenib and ixazomib led to complete remission

(A) Schematic depiction of sorafenib and ixazomib co-treatment of two patients with FLT3-ITD AML.

(B) Changes of percentage of blasts in peripheral blood (PB) and BM during treatment course of patient UPN ∗∗∗573.

(C) Changes of hemoglobin concentration, leukocyte count, and platelet count during treatment course of patient UPN ∗∗∗573.

(D) Wright-Giemsa staining of BM smears through the treatment course of patient UPN ∗∗∗573. Magnification 400×; scale bar, 10 μm.

(E) Changes of percentage of AML blasts in BM of patient UPN ∗∗∗∗032 at diagnosis and at the end of indicated therapy regimens.

MYC/USP10 signaling regulates p53 ubiquitination and protein stability in FLT3-ITD AML

We further set out to identify the molecular pathways downstream of FLT3-ITD that regulate p53 protein ubiquitination. MDM2, the principal E3 ligase for p53, was reduced upon quizartinib treatment (Figure S5A), indicating the existence of alternative, MDM2-independent pathways for p53 ubiquitination in FLT3-ITD AML. We therefore analyzed the expression of E3 ligases and deubiquitinases¹⁸ in MV4-11 cells as well as in MLL-AF9 and FLT3-ITD co-expression-induced murine leukemia cells following FLT3 inhibition using public available datasets.¹⁵ Only 5 deubiquitinases (DUBs)—USP10, USP31, USP36, USP14, and USP1—were downregulated in all cells, whereas no E3 ligase was upregulated (Figure 6A). Among the five downregulated DUBs, only USP10 has been reported to deubiquitinate p53.¹⁹ We therefore focused on exploring the mechanisms of USP10 in regulating p53 in FLT3-ITD AML. We found that USP10 interacted with p53 (Figure S5B). Inhibition of FLT3 reduced USP10 protein and mRNA levels (Figures 6B and 6C). USP10 knockdown or overexpression decreased or increased p53 protein levels, respectively, but had no effect on TP53 mRNA levels (Figures S5C–S5F). Bortezomib partially restored p53 expression in USP10-knockdown cells (Figure 6D). USP10 overexpression in MV4-11 and MOLM-13 cells inhibited cell growth, which was partially reversed by TP53 knockdown (Figure S5G). USP10 overexpression also abrogated quizartinib-induced p53 downregulation (Figure S5H). We further evaluated the role of USP10 in regulating p53 ubiquitination. Both total and K48-specific ubiquitination levels of p53 were increased upon USP10 depletion but decreased upon USP10 overexpression (Figures 6E, 6F, S5I, and S5J). More importantly, overexpression of USP10 prolonged the half-life of p53 (Figures 6G and S5K).

Figure 6.

MYC-USP10 signaling regulates p53 ubiquitination and protein stability

(A) Venn diagram of suppressed deubiquitinases from MLL-AF9/iFLT3-ITD-OFF cells in vitro and in vivo at 48 h and human MV4-11 cells treated with 5 nM Quiz for 24 h. A p <0.01 cutoff was applied.

(B) Immunoblot of indicated proteins in MV4-11 and MOLM-13 cells treated with 5 nM Quiz for 24 h.

(C) MV4-11 and MOLM-13 cells from (B) were used to analyzed the expression of USP10 by quantitative real-time PCR (n = 3; mean ± SD).

(D) Immunoblot of USP10 and p53 in shUSP10-expressing MV4-11 and MOLM-13 cells after treatment with 50 nM bortezomib for 4 h.

(E and F) MV4-11 and MOLM-13 cells transduced with inducible shRNAs (E) or USP10-expressing lentiviral vectors (F) were treated with doxycycline to induce expression overnight. Protein lysates were immunoprecipitated with p53 antibody and then immunoblotted for ub and p53.

(G) MV4-11 and MOLM-13 cells transduced with inducible empty (vector) or USP10-expressing (USP10) lentiviral vectors were treated with doxycycline to induce expression overnight. Immunoblot of p53 after CHX treatment for indicated minutes.

(H) Quantitative real-time PCR of USP10 in MV4-11 and MOLM-13 cells transduced with inducible shRNAs or MYC-expressing lentiviral vectors after doxycycline treatment overnight (n = 3; mean ± SD; paired t test).

(I) MV4-11 and MOLM-13 cells transduced with an inducible lentiviral vector expressing MYC were co-transduced with inducible shSCR or shUSP10. Immunoblot of indicated proteins after doxycycline treatment overnight.

(J) MV4-11 and MOLM-13 cells transduced with an inducible lentiviral vector expressing shMYC were co-transduced with inducible empty (vector) or USP10-expressing (USP10) lentiviral vectors. Immunoblot of indicated proteins after doxycycline treatment overnight. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001.

We next sought to identify the key factor downstream of FLT3-ITD that regulates USP10 expression. GSEA of RNA sequencing data from MV4-11 and MOLM-13 cells treated with quizartinib revealed that MYC targets were the most significantly downregulated following FLT3 inhibition (Figure S6A). FLT3 inhibition also reduced MYC in MV4-11 and MOLM-13 cells (Figure 5B). Introducing FLT3-ITD into CD34⁺ HSPCs induced MYC and USP10 expression (Figure S6B). Motif searching revealed that the promoter of USP10 contained a canonical MYC-binding E-box motif—CACGTG (Figure S6C). Analysis of two chromatin immunoprecipitation sequencing (ChIP-seq) data proved that MYC bound to the USP10 promoter (Figure S6D).^20,21 ChIP-qPCR against MYC confirmed the binding of MYC to the USP10 promoter at basal conditions, whereas quizartinib treatment decreased the binding signal as well as the amount of acetylated H3K27 (H3K27Ac), indicative of transcriptional suppression (Figure S6E). MYC knockdown reduced USP10 expression, whereas MYC overexpression induced USP10 expression (Figures 6H and S6F), further supporting the notion that MYC is a transcriptional activator for USP10. Similar to USP10, MYC knockdown or overexpression decreased or increased p53 protein levels, respectively, but had no effect on mRNA levels (Figures S6G–S6J). Moreover, K48-specific ubiquitination levels of p53 were increased upon MYC depletion but decreased upon MYC overexpression (Figures S6K and S6L). MYC overexpression also prolonged the half-life of p53 (Figure S6M). Finally, we performed a series of epistatic analyses between MYC and USP10 and found that USP10 knockdown completely abrogated MYC-overexpression-induced p53 protein upregulation (Figure 6I), whereas USP10 overexpression still enhanced p53 expression in MYC knockdown cells (Figure 6J). Collectively, these findings indicate that MYC transactivates USP10 to deubiquitinate p53 and promote its protein stability in FLT3-ITD AML.

Discussion

FLT3-ITD AML represents a subtype of AML with poor overall prognosis. Several FLT3 inhibitors have been approved for FLT3-ITD AML therapy. These inhibitors initially yield encouraging clinical responses; however, most patients with FLT3-ITD AML relapse shortly after treatment with adaptive resistance. Therefore, it is urgent to elucidate the mechanisms underlying drug resistance to FLT3 inhibition and develop more effective strategies to eliminate FLT3-ITD AML. Through this study, we revealed FLT3-inhibition-induced p53 protein degradation as a pivotal mechanism of FLT3-ITD AML drug resistance and identified restoration of p53 protein as being a promising therapeutic strategy to treat this aggressive subtype of AML.

p53 plays a pivotal role in both normal and leukemic hematopoiesis. TP53 aberrations confer early leukemogenic capacity and represent a subgroup of AML with most dismal prognosis.^{5,22,23,24,25} Compared with other solid tumors, TP53 mutations are relatively rare in AML. However, aberrations in other genes that render p53 dysfunction can occur during leukemogenesis, which allows aggressive leukemias to arise in the presence of WT-p53 and portends a poor prognosis.^7,8 Therefore, elucidating the mechanisms underlying WT-p53 dysfunction may unveil potential AML vulnerabilities that can be therapeutically exploited. Previously, we found that FLT3 inhibition induced HDAC8 to deacetylate p53 and sustain FLT3-ITD AML. Targeting HDAC8 synergized with FLT3 inhibition to eliminate FLT3-ITD AML.²⁶ However, the lack of clinically available HDAC8 inhibitors limited the practical value of this combinatorial strategy.

In this study, we found that FLT3 inhibition leads to p53 ubiquitination and proteasomal degradation. Our data indicated a potential role for MYC/USP10 signaling in regulating p53 protein stability in FLT3-ITD AML. MYC can be upregulated through AKT, ERK, and STAT5 signaling downstream of FLT3-ITD.^27,28,29 Analysis of public datasets shows significant downregulation of MYC signaling upon FLT3 inhibition. Manipulation of MYC expression in FLT3-ITD AML cells did not change TP53 mRNA levels. Instead, MYC potently affected p53 ubiquitination, hence its protein stability. The effect of MYC on p53 ubiquitination was attributed to transactivation of the DUB USP10. Although USP10 does not belong to the MYC core signature,³⁰ we found a canonical E-box motif at the promoter of USP10, suggesting potential binding of MYC. We further demonstrated that MYC regulated USP10 expression by binding to its promoter. Consistently, Aram and colleagues described a similar phenomenon in human fibroblast cells.³¹ These studies suggest that USP10 is a bona fide target of MYC.

The USP10 DUB has been reported to complex with p53 and act as a p53 DUB.^19,32 We showed that USP10 is significantly upregulated by FLT3-ITD/MYC signaling in FLT3-ITD AML cells. This may explain why FLT3-ITD AML cells simultaneously exhibit high intracellular levels of p53 and MDM2.^33,34 Although MDM2 is highly expressed in FLT3-ITD AML cells due to hyperactivated AKT and ERK, USP10 can override MDM2-mediated p53 degradation and lead to heightened p53 protein levels. Thus, compared to MDM2, USP10 is a stronger determinant for p53 protein levels in FLT3-ITD AML. Through its action toward p53, USP10 could function as a tumor suppressor. Indeed, overexpression of USP10 in FLT3-ITD AML can inhibited cell growth, which was partially reversed by TP53 knockdown.

Accordingly, a report exhibited at the American Society of Hematology Annual Meeting showed that FLT3 inhibitors promote p53 ubiquitination through downregulation of the DUB YOD1.³⁵ Although we did not observe YOD1 downregulation in our system, another DUB, USP10, was downregulated with FLT3 inhibition, and we therefore propose that ubiquitin-proteasome-system-mediated p53 degradation serves as a critical mechanism to cause WT-p53 dysfunction and promotes FLT3-ITD AML drug resistance. Indeed, blocking p53 degradation with proteasome inhibitors restored intracellular p53 protein levels and enhanced p53 activity in combination with FLT3-ITD inhibitors. Moreover, the combination of proteasome and FLT3 inhibitors showed superior therapeutic effects against FLT3-ITD AML. The therapeutic effect of proteasome inhibitor on FLT3-ITD AML has been reported in a previous study, wherein bortezomib was showed to induce FLT3 degradation through autophagy. In our preliminary study, we had tested the effects of different doses of proteasome inhibitors on FLT3-ITD and cell viability and found that high doses of proteasome inhibitors reduced FLT3-ITD protein levels and induced significant cell death. In order to rule out these effects, we chose lower doses of proteasome inhibitors. Under such conditions, proteasome inhibitors mainly restored p53 protein without affecting FLT3 levels and only showed significant anti-leukemia effect when combined with FLT3 inhibitors.

Since both FLT3 inhibitors and proteasome inhibitors are clinically available drugs, our study has high translational value. Besides, the safety and tolerability of combined usage of FLT3 and proteasome inhibitors are highly likely to be superior over chemotherapy. Remarkably, the dual inhibition strategy resulted in complete remission in two heavily pretreated patients FLT3-ITD AML. However, since only two patients received combined treatment of sorafenib and ixazomib (off-label usage), caution must be taken in interpreting the treatment outcomes, and the potential benefits of the combinatorial strategy in patients with FLT3-ITD AML should be evaluated only in the context of adequately designed clinical trials. Still, the compelling preclinical data and remarkable outcomes of the two human patients in our study do provide convincing evidence to launch a rigorous clinical trial to investigate the combinatorial strategy to treat FLT3-ITD AML.

Limitations of the study

Due to the bone marrow suppression of patients receiving sorafenib and ixazomib co-treatment, we could not collect enough myeloblasts to evaluate whether proteasome inhibitors also stabilized p53 in patients. Besides, while our cell lines and PDX data showed that proteasome inhibitors did not reduce FLT3 protein levels in FLT3-ITD AML, we were unable to investigate whether ixazomib induced autophagic degradation of p53 in patients with FLT3-ITD AML.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
p53	Abcam	Cat# ab1101; RRID: AB_297667
p53	Abcam	Cat# ab32389; RRID: AB_776981
FLT3	Cell Signaling Technology	Cat# 3462S; RRID: AB_2107052
phospho-FLT3 (Y589/591)	Cell Signaling Technology	Cat# 3464S; RRID: AB_2107051
AKT	Cell Signaling Technology	Cat# 4685S; RRID: AB_2225340
phosphor-AKT (S473)	Cell Signaling Technology	Cat# 4060S; RRID: AB_2315049
p44/42 MAPK (ERK1/2)	Cell Signaling Technology	Cat# 4695S; RRID: AB_390779
Phospho-p44/42 MAPK (Erk1/2) (Thr202/Tyr204)	Cell Signaling Technology	Cat# 4370S; RRID: AB_2315112
MYC	Cell Signaling Technology	Cat# 18583S; RRID: AB_2895543
PARP	Cell Signaling Technology	Cat# 9532S; RRID: AB_659884
STAT5	Abcam	Cat# ab194898; RRID: AB_2924353
phosphor-STAT5 (Y694)	Abcam	Cat# ab32364; RRID: AB_778105
MYC	Abcam	Cat# ab32072; RRID: AB_731658
USP10	Abcam	Cat# ab109219; RRID: AB_2920539
Ubiquitin	Abcam	Cat# ab134953; RRID: AB_2801561
Ubiquitin (linkage-specific K48)	Abcam	Cat# ab140601; RRID: AB_2783797
SUMO1	Abcam	Cat# ab32058; RRID: AB_778173
SUMO2 + SUMO3	Abcam	Cat# ab109005; RRID: AB_10859949
phosphoserine/threonine	ECM biosciences	Cat# PP2551; RRID: AB_1184778
Histone H3 (acetyl K27)	Abcam	Cat# ab177178; RRID: AB_2828007
β-actin	Sigma-Aldrich	Cat# A5441; RRID: AB_476744
Normal mouse IgG	Santa Cruz	Cat# sc-2025; RRID: AB_737182
Rabbit IgG isotype control	Cell Signaling Technology	Cat# 3900S; RRID: AB_1550038
Brilliant Violet 510™ anti-human CD45	Biolegend	Cat# 304036; RRID: AB_2561940
PE/Cyanine7 anti-human CD34	Biolegend	Cat# 343516; RRID: AB_1877251
PE anti-human CD33	Biolegend	Cat# 303404; RRID: AB_314348
FITC anti-human CD11b	Biolegend	Cat# 301330; RRID: AB_2561703
APC anti-human CD3	Biolegend	Cat# 317318; RRID: AB_1937212
APC/Cyanine7 anti-human CD14	Biolegend	Cat# 325620; RRID: AB_830693
Brilliant Violet 421™ anti-human CD19	Biolegend	Cat# 302234; RRID: AB_11142678
Biological samples
Patient AML cells	Tongji Hospital	N/A
Chemicals, peptides, and recombinant proteins
Human IL-3	PEPROTECH	Cat# 200-03
Human IL-6	PEPROTECH	Cat# 200-06
Human SCF	PEPROTECH	Cat# 300-07
Human TPO	PEPROTECH	Cat# 300-18
Quizartinib	Selleck	Cat# S1526
Midostaurin	Selleck	Cat# S8064
Sorafenib	Selleck	Cat# S1040
Crenolanib	Selleck	Cat# S2730
Gilteritinib	Selleck	Cat# S7754
Cycloheximide	Selleck	Cat# S7418
N-ethylmaleimide	Selleck	Cat# S3692
MK2206	Selleck	Cat# S1078
AZD6244	Selleck	Cat# S1008
Pimozide	Selleck	Cat# S4358
Bortezomib	Selleck	Cat# S1013
Ixazomib	Selleck	Cat# S2180
MG132	Selleck	Cat# S2619
Protease inhibitor cocktail	Selleck	Cat# B14001
Phosphatase inhibitor cocktail	Selleck	Cat# B15001
Quizartinib	Daiichi-Sankyo	N/A
Sorafenib	Bayer	N/A
Bortezomib	Xian-Janssen	N/A
Ixazomib	Takeda	N/A
Cytarabine	Pfizer	N/A
Daunorubicin	Shenzhen Main Luck	N/A
Critical commercial assays
Cell Signaling Profiling Plus (CSP100 Plus) antibody microarray	Wayne Biotechnologies	CSP100 Plus
ChIP-IT Express kit	Active Motif	Cat# 53008
StemSpan SFEM II	Stemcell Technologies	Cat# 09605_C
MethoCult™ H4434 Classic	Stemcell Technologies	Cat# 04434
Cell Counting Kit-8	Dojindo	Cat# CK04
Cell Line Nucleofector™ Kit V	Lonza	Cat# VCA-1003
Deposited data
ChIP-seq data of MYC	Lin et al.20	GSE36354
ChIP-seq data of MYC	Liang et al.21	GSE112608
RNA-seq data of MV4-11 and MOLM-13 treated with AC220	Li et al.13	GSE126933
RNA-seq data of FLT3 inhibition	Bjelosevic et al.15	GSE163932
RNA-seq data of FLT3 inhibition	Zhang et al.14	GSE181586
Experimental models: Cell lines
MV4-11	ATCC	Cat# CRL-9591
MOLM-13	DSMZ	Cat# ACC 544
OCI-AML2	DSMZ	Cat# ACC 99
OCI-AML3	DSMZ	Cat# ACC 582
HEK-293T	ATCC	Cat# CRL-3216
GP2-293	Takara	N/A
Experimental models: Organisms/strains
NOD.Cg-PrkdcscidIl2rgtm1Sug/JicCrl mice	Vital River	Cat# 408
Oligonucleotides
FLT3 Silencer Select siRNA s5290	Thermo Fisher	Cat# 4392420
FLT3 Silencer Select siRNA s5291	Thermo Fisher	Cat# 4392420
Silencer Select Negative Control No.1	Thermo Fisher	Cat# 4390843
shRNA primers	This paper	Table S3
RT-qPCR primers	This paper	Table S4
ChIP-qPCR USP10-F: ATGAATGAACCAACGACCCC	This paper	N/A
ChIP-qPCR USP10-R: AGGGCATAAAAGCACGCGTAA	This paper	N/A
Recombinant DNA
Tet-pLKO-Puro	Addgene	Cat# 21915
psPAX2	Addgene	Cat# 12260
pMD2.G	Addgene	Cat# 12259
pCMV-VSV-G	Addgene	Cat# 8454
pMYs-IRES-GFP	Cell Biolabs	Cat# RTV-021
pLVX-TetOne-Bsd	Takara	N/A
pLVX-shRNA2	Takara	N/A
Software and algorithms
Flowjo v10	Flowjo	https://www.flowjo.com
Graphpad Prism v8	Graphpad Prism	https://www.graphpad-prism.cn/?c=i&a=prism
GSEA v4.1.0	N/A	http://software.broadinstitute.org/gsea/downloads.jsp

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Aibin Liang (lab7182@tongji.edu.cn).

Materials availability

This study did not generate new unique reagents.

Data and code availability

• •The RNA-seq data analyzed in this study were obtained from the GEO under accession numbers GSE36354, GSE112608, GSE126933, GSE163932, and GSE181586. Original western blot images and microscopy data reported in this paper will be shared by the lead contact upon request.
• •This paper does not report original code.
• •Any additional information required to reanalyze the data reported in this work paper is available from the lead contact upon request.

Experimental model and study participant details

Cell lines and cell culture

The cell lines used in the study included MV4-11 (ATCC, CRL-9591), HEK-293T (ATCC, CRL-3216), MOLM-13 (DSMZ, ACC 554), OCI-AML2 (DSMZ, ACC 99), OCI-AML3 (DSMZ, ACC 582), and GP2-293 (Takara). MV4-11, MOLM-13, OCI-AML2, and OCI-AML3 cells were cultured in RPMI 1640 (Gibco). HEK-293T and GP2-293 cells were cultured in DMEM (Gibco). All media were supplemented with 10% FBS (Gibco), 1% Glutamax (Gibco), and 1% penicillin-streptomycin (Gibco). All cell lines were grown at 37°C in a humidified atmosphere containing 5% CO₂. All cell lines were validated via short tandem repeat profiling prior to use. All cells were routinely tested for Mycoplasma (Universal Mycoplasma Detection Kit, ATCC, 30-1012K).

Human samples

Normal HSPCs were obtained from allogeneic transplant donors after mobilization. FLT3-ITD AML patient samples were obtained from AML patients with active disease at diagnosis or at relapse. All subjects signed informed consent. Protocols for sample handling and data analysis were approved by the Tongji Hospital Ethics Committee and were performed in compliance with the Declaration of Helsinki. Primary human FLT3-ITD AML blasts were culture in StemSpan SFEM II (Stemcell Technologies) supplemented with 20 ng/mL IL-3, 10 ng/mL IL-6, 100 ng/mL SCF, 50 ng/mL TPO (all from Peprotech), and 1% Glutamax (Gibco). Compassionate treatment of patients with sorafenib and ixazomib was approved by the Tongji Hospital Ethics Committee under the protocol KYSB-124, and was performed after getting written informed consent on the off-label, off-trial use following the Principles of Helsinki. Patient information is shown in Table S2.

Patient-derived xenograft (PDX) model

NOD.Cg-Prkdc^scidIl2rg^tm1Sug/JicCrl mice (NOG mice, Charles River) were used for the human AML PDX model. The mice were housed in specific pathogen free facilities. T cell–depleted human FLT3-ITD AML blasts (2 × 10⁶ cells/mouse) were injected into sub-lethally irradiated (150 cGy) 6-week-old female NOG mice via tail vein. After confirmation of human leukemia engraftment in murine peripheral blood (>5% human CD45⁺ cells), treatment was initiated as indicated (quizartinib, 10 mg/kg daily for 4 weeks, gavage; bortezomib, 0.5 mg/kg twice weekly for 4 weeks, intraperitoneal; sorafenib, 50 mg/kg twice daily for 4 weeks, gavage; ixazomib, 1 mg/kg twice weekly for 4 weeks, gavage; Ara-C, 100 mg/kg daily for 5 days, intraperitoneal; DNR, 2.5 mg/kg daily for 3 days, intravenous). After treatment, mice were euthanized, and bone marrow cells were analyzed for engraftment of human cells by labeling with anti-human CD45 antibody followed by flow cytometry analysis. Subpopulations and immunophenotype of human CD45⁺ cells were evaluated by labeling with anti-human CD33, CD34, CD11b, CD14, CD19 and CD3 antibodies (all from Biolegend). The survival of the mice was monitored. Bone marrow cells from primary recipients were used for secondary transplantation, and the bone marrow cells of secondary recipients were analyzed after 12 weeks. All mouse experiments were carried out according to the National Institutes of Health’s Public Health Service Policy on Humane Care and Use of Laboratory Animals.

Method details

FLT3-ITD AML patients pilot study

Compassionate treatment of relapsed/refractory FLT3-ITD AML patients with sorafenib and ixazomib was approved by the Tongji Hospital Ethics Committee under the protocol KYSB-124, and was performed after getting written informed consent on the off-label, off-trial use following the Principles of Helsinki. Sorafenib was administered 400 mg twice daily for 28 consecutive days. Ixazomib was administered 4mg once a week on Day 1, 8, and 15 of a 28-day treatment cycle. Hemoglobin levels, white blood cell and platelet counts, and the percentage of blasts in peripheral blood were monitored through blood routine test every 3 days. Bone marrow aspirate was performed on Day 28 to assess the disease status.

Compounds and drugs

Quizartinib, midostaurin, sorafenib, crenolanib, gilteritinib, cycloheximide, MK2206, AZD6244, pimozide, bortezomib, ixazomib, MG132, and N-ethylmaleimide were purchased from Selleck and reconstituted in dimethyl sulfoxide (DMSO) or sterile water according to the solubility for in vitro use. For in vivo animal experiments, quizartinib was purchased from Daiichi-Sankyo and reconstituted in sterile water daily. Sorafenib was purchased from Bayer and reconstituted in sterile water daily. Bortezomib was purchased from Xian-Janssen and reconstituted in normal saline. Ixazomib was purchased from Takeda and reconstituted in sterile water at the days for gavage. Cytarabine and daunorubicin was from Pfizer and Shenzhen Main Luck, respectively. Both drugs were dissolved in PBS.

siRNA and shRNA sequences and DNA constructs

FLT3 and scramble siRNA were from Thermo Fisher (FLT3 Silencer Select siRNA ID: s5920 and s5921; Control siRNA: Silencer Select Negative Control No.1).

shRNAs targeting TP53 were cloned into pLVX-shRNA2 (Takara). shRNAs targeting MYC and USP10 were cloned into Tet-pLKO-Puro (Addgene plasmid #21915; a gift from Dmitri Wiederschain). The shRNA sequences used are listed in Table S3.

Full-length FLT3-WT and FLT3-ITD were cloned into pMYs-IRES-GFP (Cell Biolabs). Full-length TP53-WT, TP53-8KR, MYC, and USP10 were cloned into pLVX-TetOne-Bsd.

Cell transfection and virus packaging and transduction

MV4-11 and MOLM-13 cells were transfected with siRNA using Cell Line Nucleofector Kit V according to the manufacturer’s instructions (Lonza). FLT3-ITD AML blasts were transfected with siRNAs using the Human CD34⁺ Cell Nucleofector Kit according to the manufacturer’s instructions (Lonza).

For retrovirus packaging, the target vector and pCMV-VSV-G (Addgene plasmid #8454; a gift from Bob Weinberg) packaging plasmid were co-transfected into GP2-293 cells using Lipofectamine 2000 (Invitrogen). For lentivirus packaging, the target vector and psPAX2 (Addgene plasmid #12260; a gift from Didier Trono) and pMD2.G (Addgene plasmid #12259; a gift from Didier Trono) packaging plasmids, were co-transfected into HEK-293T cells using Lipofectamine 2000. Retroviral or lentiviral particles were collected 48 and 72 h after transfection and filtered.

For retrovirus and lentivirus transduction, cells were cultured in medium as mentioned above, and spinoculated with virus–contaning supernatant supplemented with 8 μg/mL polybrene (Sigma) at 33°C for 2 consecutive days. The medium was then refreshed, and cells were cultured for further selection by appropriate cell sorting or antibiotics.

Antibody microarray

MV4-11 cells were cultured in the presence or absence of 5 nM quizartinib for 24 h. Whole-cell lysates were prepared with specific lysis buffer according to microarray manufacturer’s instructions. Protein phosphorylation and expression were evaluated with Cell Signaling Profiling Plus (CSP100 Plus) antibody microarray by Wayne Biotechnologies according to the manufacturer’s instructions. Protein expression was normalized by β-actin. The antibody microarray data is provided in Table S1.

Immunoprecipitation and western blot

For immunoprecipitation, cells were lysed in lysis buffer containing 50 mM Tris-HCl pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.25% sodium deoxycholate, and 10% glycerol, supplemented with protease inhibitor cocktail (Selleck) for 30 min at 4°C with constant shaking. For SUMO assay, 10 mM N-ethylmaleimide (Selleck) was contained in lysis buffer. For p53 phosphorylation assay, phosphatase inhibitor cocktail (Selleck) was contained in lysis buffer. Then protein supernatants were collected through centrifuge at 13000 g for 10 min. The protein lysates were incubated with magnetic beads (Invitrogen) pre-labelled with p53 [DO-1] antibody (Abcam, ab1101) or normal mouse IgG (Santa Cruz, sc-2025) overnight at 4°C with constant rotation. The immunoprecipitates were then washed five times with lysis buffer and used for immunoblotting.

For Western blot, cells were collected and lysed in SDS lysis buffer (containing 2% SDS) with boiling at 95°C for 10 min. Protein lysates were resolved by SDS-PAGE and subsequently transferred to PVDF membranes. The membranes were blocked in 5% non-fat milk and incubated with the following primary antibodies overnight 4°C with constant rotation: β-actin (Sigma-Aldrich, A5441, 1:10000), FLT3 (Cell Signaling Technology, 3462S, 1:1000), phospho-FLT3 (Y589/591) (Cell Signaling Technology, 3464S, 1:1000), AKT (pan) (Cell Signaling Technology, 4685S, 1:1000), phosphor-AKT (S473) (Cell Signaling Technology, 4060S, 1:1000), ERK1/2 (Cell Signaling Technology, 4695S, 1:1000), phosphor-ERK1/2 (T202/Y204) (Cell Signaling Technology, 4370S, 1:1000), STAT5 (Abcam, ab194898, 1:2000), phosphor-STAT5 (Y694) (Abcam, ab32364, 1:1000), p53 [DO-1] (Abcam, ab1101, 1:1000), p53 [E26] (Abcam, ab32389, 1:1000), MYC (Abcam, ab32072, 1:1000), USP10 (Abcam, ab109219, 1:1000). Membranes were then washed in TBS-T and incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies at room temperature for appropriate time. Signals were visualized using Immobilon Western Chemiluminescent HRP Substrate (Millipore).

Ubiquitination, sumoylation, and phosphorylation assay

For the endogenous ubiquitination assay, MV4-11 and MOLM-13 cells with indicated manipulation were pretreated with 20 μM MG132 for 2 h before collection and lysis. Immunoprecipitation was performed and immunoblots were analyzed with anti-p53 [E26] (Abcam, ab32389, 1:1000), anti-ubiquitin (Abcam, ab134953, 1:1000) and anti-ubiquitin (Linkage-specific K48) (Abcam, ab140601, 1:1000). For p53 sumoylation, immunoblots were analyzed with anti-SUMO1 (Abcam, ab32058, 1:1000) and anti-SUMO2 + SUMO3 (Abcam, ab109005, 1:1000). For p53 phosphorylation, immunoblots were analyzed with anti-phosphoserine/threonine (ECM biosciences, PP2551, 1:1000).

Quantitative real-time PCR

Total RNA was isolated using TRIzol (Thermo Fisher) following manufacturer’s instructions. First-strand cDNA was synthesized using ReverTra Ace qPCR RT Master Mix with gDNA Remover (TOYOBO). Quantitative real-time PCR was performed with cDNA using TB Green Premix Ex Taq (Takara) on the ABI7500 or Vii7 system (Thermo Fisher). For quantification, the CT values were obtained and normalized to the CT values of RPLP0 gene. Fold changes in expression were calculated by the 2^−ΔΔCT method. Primer sequences used in this study are listed in Table S4.

Chromatin immunoprecipitation quantitative PCR (ChIP-qPCR)

MV4-11 and MOLM-13 cells were cultured in the presence or absence of 5 nM quizartinib for 24 h, and then cross-linked in cell culture media containing 1% formaldehyde with gentle rotation for 10 min at room temperature. Fixation was stopped by the addition of glycine (125mM final concentration). Cells were harvested for ChIP with the ChIP-IT Express kit (Active Motif, 53008) according to the manufacturer’s instructions. The antibodies used for ChIP were anti-MYC (Cell Signaling Technology, 18583S), anti-H3K27Ac (Abcam, ab177178), or rabbit IgG isotype control (Cell Signaling Technology, 3900S). The primers used for ChIP-qPCR analysis are listed in key resource table.

Cell viability and cell apoptosis

Cell viability was measured using the Cell Counting Kit-8 (CCK-8, Dojindo) according to the manufacturer’s instructions. CCK-8 is a tetrazolium dye conversion assay similar to MTT assay, which is based on WST-8 and used to determine viable cells. In the presence of 1-methoxy PMS, WST-8 can be reduced to highly water-soluble orange-yellow formazan by intracellular dehydrogenase.

Briefly, 0.03 × 10⁶ cells/well with triplicate were seeded into 96-well clear flat bottom microplates (Corning) in their corresponding media (100 μL). Cells were treated as indicated for 48 h. Then 10 μL/well CCK-8 solution was added and incubated for an appropriate time at 37°C before measuring the absorbance at 450 nm. For each experiment, the absorbance of the blank wells (growth media) was subtracted from the values for the wells containing cells. The mean absorbance value of triplicate wells was calculated. The viability was calculated using the equation: % viable cells = mean absorbance value of dose X/mean absorbance value of vehicle × 100%. For cell apoptosis, cells with indicated treatment were stained with Annexin V and PI in binding buffer according to the manufacturer’s instructions (BD Biosciences) and then analyzed by flow cytometry (BD LSRFortessa).

Colony formation assay

C34⁺ FLT3-ITD AML cells were plated in methylcellulose medium (MethoCult H4434, Stemcell Technologies) supplemented with penicillin and streptomycin according to the manufacturer’s instructions. Colonies were evaluated and counted after 14 days incubation.

Gene set enrichment analysis

Gene set enrichment analysis was performed with the desktop client version (GSEA v4.1.0, http://software.broadinstitute.org/gsea/downloads.jsp) with default parameters. Microarray and RNA sequencing data were obtained from the Gene Expression Omnibus (GEO) datasets. Differentially expressed genes were used to run gene set enrichment analysis with gene sets obtained from MSigDB, and normalized enrichment score (NES), and false discovery rate (FDR) q values were calculated.

Quantification and statistical analysis

GraphPad Prism v8.3.0 were used for data processing, statistical analysis, and result visualization. Statistical tests performed for each experiment are highlighted in the figure legends. Unless otherwise specified, error bars are indicative of ±SD, and threshold for significance was p < 0.05. ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001; ∗∗∗∗, p < 0.0001. ns denotes not significant. No statistical methods were used to predetermine sample sizes. All data from representative experiments were repeated at least two times independently with similar results obtained.

Published: November 10, 2023