This article is mentioned in:

Introduction

Despite recent improvements in long-term survival rates for acute lymphoblastic leukaemia (ALL) (1) and acute myeloid leukaemia (AML) (2), subgroups of patients continue to experience poor long-term outcomes. TP53 alterations are a high-risk genomic feature present in 16% of newly diagnosed ALL (3,4) and 13% of newly diagnosed AML cases (4). The presence of a TP53 alteration is associated with poor long-term overall survival compared with TP53 wild-type (TP53WT) cases in both ALL (24 months vs. not reached, P=0.001) and AML (6 vs. 26 months, P<0.001) (3). To date, no targeted therapies are currently available to treat TP53-mutated (TP53MUT) acute leukaemia outside of clinical trials, and more efficacious therapeutic options are required to improve long-term outcomes.

CBL0137 is a small molecule curaxin that epigenetically modulates multiple cancer-related signalling pathways (5,6), including casein kinase 2 (CK2)-mediated activation of the TP53-encoded p53 protein through the Facilitates Chromatin Transcription (FACT) complex (7). CBL0137 is effective in pre-clinical ALL patient-derived xenograft (PDX) models, most notably in infant KMT2A-rearranged (KMT2Ar) ALL (8,9). KMT2Ar is present in 5–10% of newly-diagnosed acute leukaemia cases overall (10), including 70–80% of infant ALL and 35–50% of infant AML diagnoses (11), and TP53 mutations occur in ~16% of KMT2Ar ALL cases (4). Outcomes for KMT2Ar leukaemia are exceptionally poor across all age groups, with five-year event-free survival (EFS) rates of 30–50% (12). Evidently, advances in treatment options are required to improve patient outcomes.

In a study by Lock et al (8), CBL0137 induced complete remission in a KMT2Ar infant ALL PDX model, and partial responses were observed in several B-cell ALL (B-ALL) and T-cell ALL (T-ALL) models of unknown genomic subtype. Somers et al (9) similarly reported CBL0137 efficacy in infant KMT2Ar B-ALL PDX models, with responses ranging from delayed cancer progression to maintained complete remission. These early data suggested that the role of CBL0137 deserves further exploration. For instance, the presence of loss-of-function TP53 alterations, a reasonably common event in KMT2Ar B-ALL, may reasonably be expected to confer resistance. Furthermore, given its in vitro efficacy against KMT2Ar B-ALL, CBL0137 is also likely to be active against other subtypes of B-ALL. In the present study, it was confirmed that CBL0137 induces leukaemic cell apoptosis in a number of cell lines with various driver alterations, including those with KMT2A rearrangements, with LC50 concentrations in the range of 166 to 676 nM. Notably, it was also demonstrated that the potency of CBL0137 is attenuated in the presence of loss of function TP53 alterations.

Materials and methods

Cell line maintenance

U-937 (CRL-1593.2), MV4;11 (CRL-9591), Jurkat (TIB-152), THP-1 (TIB-202) and RS4;11 (CRL-1873) cell lines were purchased from the American Type Culture Collection (ATCC). RCH-ACV (ACC 548), REH (ACC 22) and NALM-19 (ACC 522) cell lines were purchased from DSMZ. All cell lines were maintained in culture at 37°C in RPMI-1640 media (cat. no. R0883; Sigma-Aldrich; Merck) containing foetal calf serum (FCS) (AU-FBS/SF; CellSera), 5 mM L-glutamine, 50 U/ml penicillin and 50 µg/ml streptomycin. THP-1 cells were cultured in 20% FCS, and all other cell lines were cultured in 10% FCS. Cells up to 20 passages from the original stock were used for experiments.

Inhibitor storage

CBL0137 (cat. no. S0507; Selleck Chemicals) was stored long-term at 10 mM in DMSO at −80°C and diluted in DMSO immediately prior to use.

Generation of CRISPR/Cas9 TP53 knock-out RS4;11 cell lines

Vectors FUCas9mCherry and FgH1tUTG were a gift from Professor Marco Herald (Walter and Eliza Hall Institute of Medical Research, Melbourne, Australia) (13). Two independent single guide RNA (sgRNA) targets were designed, to account for off-target effects. sgRNA target sequences were designed to generate random indel and frameshift mutations in exon 4 of TP53, to disrupt the p53 protein prior to the critical DNA-binding domain, where deleterious TP53 mutations identified in human leukaemia cell lines are located (Fig. 1A). sgRNA oligonucleotides were designed with online Benchling® Software (Benchling) and purchased from Sigma-Aldrich; Merck KGaA: TP53 Oligo 1 forward, 5′-TCCCACCAGCAGCTCCTACACCGG-3′ and reverse, 5′-AAACCCGGTGTAGGAGCTGCTGGT-3′; and TP53 Oligo 2 forward, 5′-TCCCCCATTGTTCAATATCGTCCG-3′ and reverse, 5′AAACCGGACGATATTGAACAATGG-3′.

Figure 1. CBL0137 induces apoptosis in acute leukaemia cell lines, and LC50 are higher in TP53-mutated cell lines. (A) TP53 mutations identified in leukaemia cell lines. Data was obtained from the Broad Institute Cancer Cell Line Encyclopedia portal. (B) Protein domain schematic of p53, with mutations identified in cell lines annotated. Exons are denoted by dotted lines. (C) Human acute leukaemia cell lines were exposed to increasing doses of CBL0137 in duplicate and apoptosis was measured by Annexin-V/7-AAD staining after 72 h. Concentration-response curves and logLC50 values were extrapolated using non-linear regression analysis in GraphPad Prism. Data are mean + SD for 3 biological replicates. (D) Mean CBL0137 logLC50 values for each cell line are shown. LC50 values are provided in Table SII. For statistical analysis, *P<0.05 for comparison of TP53 wild-type (TP53WT) and mutated (TP53MUT) mean logLC50, Kruskal-Wallis test with multiple comparisons. (E) Representative immunoblot of total p53 protein in DMSO vehicle-treated (−) and 1 µM CBL0137-treated (+) cell lines. AML, acute myeloid leukaemia; AMoL, acute monocytic leukaemia; B-ALL, B-lineage acute lymphoblastic leukaemia; MUT, mutated; WT, wild-type; T-ALL, T-lineage acute lymphoblastic leukaemia; TAD, transactivation domain; NLS, nuclear localisation signal; Tetra, Tetramerisation motif.

TP53 FgH1tUTG sgRNA vectors were generated by BsmBI digestion and T4 DNA ligation, incubated overnight at 4°C and transformed by 42°C heat shock for 2 min in DH5a™-T1R chemically competent E. coli (Thermo Fisher Scientific, Inc.), and selected based on ampicillin resistance (50 µg/ml) conferred by transformation of the FgH1tUTG plasmid. Successful ligation of inserts was confirmed by Sanger sequencing with the BigDye™ Terminator v3.1 Cycle Sequencing Kit (used according to manufacturer's protocol) and SeqStudio™ Genetic Analyser System (both from Thermo Fisher Scientific, Inc.). DNA chromatogram results were analysed using online Benchling® Software.

293T cells (CRL-3216; purchased from ATCC) were co-transfected with FUCas9mCherry (4.2 µg) or FgH1tUTG sgRNA (4.2 µg) vectors, and 2nd generation lentiviral packaging vectors pMD2.G (1.6 µg), pMDLg/pRRE (2.4 µg) and pRSV-Rev (1.1 µg) (Addgene, Inc.). Each transfection was prepared in 450 µl Opti-MEM™ Reduced Serum Media and 30 µl Lipofectamine™ 2000 transfection reagent (both from Thermo Fisher Scientific, Inc.), and incubated for 1 h at room temperature prior to application to 293T cells. Viral supernatant was harvested from 293T cultures 48 h post-transfection (MOI not quantified), and RS4;11 cells were transduced by spinfection in the presence of 4 mg/ml Polybrene® (Santa Cruz Biotechnology, Inc.) at 220 × g for 1 h at room temperature in six-well tissue culture plates. One week post-transduction, GFP and mCherry double-positive populations were sorted with a BD FACSFusion flow cytometer (BD Biosciences). Sorted populations were activated with doxycycline (1 µg/ml) for 3 days, and genomic DNA was extracted by phenol-chloroform to confirm induction of frameshift mutations by PCR amplification of TP53 exon 4 using Q5® High-Fidelity DNA Polymerase (cat. no. M0491; New England Biolabs, Inc.) and the following primers: TP53 intron 4 forward, 5′-TCCTCTGACTGCTCTTTTCACCCAT-3′ and reverse, 5′-AATATTCAACTTTGGGACAGGAGTCAGAGA-3′. Thermocycling conditions were as follows: 98°C for 1 min, followed by 33 cycles at 98°C for 10 sec, 64°C for 15 sec and 72°C for 2 min, followed by 1 cycle at 72°C for 10 min. Samples were then stored at 4°C until they were visualised in 2% agarose gel containing 1:10,000 GelRed (Biotium, Inc.).

Sanger sequencing was utilised to identify the profile of mutations present (Fig. S1). Sanger sequencing was performed using the BigDye™ Terminator v3.1 Cycle Sequencing Kit according to manufacturer's instructions (Thermo Fisher Scientific, Inc.) and SeqStudio™ Genetic Analyser System (Thermo Fisher Scientific, Inc.). DNA chromatogram results were analysed using online Benchling® Software. RNA was harvested after 7 and 14 days in culture, to quantify TP53 expression by quantitative reverse transcription-quantitative PCR (RT-qPCR). As a comparator, RS4;11 cells expressing Cas9 vector only were used as a TP53WT control in all relevant experiments. RT-qPCR thermocycling conditions were as follows: 10 min at 95°C for one cycle, followed by 40 cycles at 95°C for 15 sec and 60°C for 60 sec. RT-qPCR cycling reactions were performed on a QuantStudio 7 Real-Time PCR system (Thermo Fisher Scientific, Inc.).

Apoptosis detection via Annexin V/7-Aminoactinomycin D staining

Cells were seeded at 2×105 cells/ml and treated for 72 h with a range of CBL0137 doses in 0.3% DMSO, and incubated in 96-well tissue culture plates at 37°C/5% CO2 for 72 h. Treated cells were harvested at room temperature by centrifugation at 220 × g for 5 min, washed twice in flow cytometry binding buffer (Hank's Balanced Salt Solution (cat. no. H9394; Sigma-Aldrich; Merck KGaA), 10 mM HEPES, 5 mM CaCl2), and cells in each well stained with Annexin V (cat. no. 556421; BD Biosciences) and 7-Aminoactinomycin (7-AAD) (cat. no. A1310; Thermo Fisher Scientific, Inc.) according to the supplier's protocol (BD Biosciences), and incubated on ice for 30 min. Assays were analysed on a BD FACSCanto flow cytometer (BD Biosciences) and data were analysed using FlowJo version 10 software (FlowJo LLC) to determine apoptotic (Annexin V and/or 7-AAD positive) and non-apoptotic (Annexin V and 7-AAD negative) populations.

The in vitro efficacy of CBL0137 to induce apoptosis was assessed on four TP53WT (RS4;11, RCH-ACV, NALM-19, MV4;11) and four TP53MUT (REH, Jurkat, U-937 and THP-1) human acute leukaemia cell lines using Annexin-V/7-AAD. The U-937 cell line was initially classified as histiocytic lymphoma (14) but is now classified as acute monocytic leukaemia (AMoL) and is therefore referred to as such throughout the manuscript (15). TP53 mutations present in each cell line investigated are provided in Fig. 1A, obtained from the Broad Institute Cancer Cell Line Encyclopedia portal (16) (https://sites.broadinstitute.org/ccle/). All identified TP53 mutations within the cell lines utilised are predicted to be deleterious according to COSMIC (17) (https://cancer.sanger.ac.uk) or ClinVar (18) (https://www.ncbi.nlm.nih.gov/clinvar/) databases. Later, the in vitro efficacy of CBL0137 to induce apoptosis was assessed on CRISPR/Cas9 TP53 knock-out RS4;11 cell lines.

RNA analysis

RNA was extracted from 5×106−1×107 cells using TRIzol® reagent (Thermo Fisher Scientific, Inc.). RNA concentration and purity were quantified with a NanoDrop 2000 Spectrophotometer (Thermo Fisher Scientific, Inc.). The QuantiTect reverse transcription kit (Qiagen GmbH) was used to synthesise cDNA from 1 µg of RNA as per manufacturer's instructions. RT-qPCR reactions were prepared in duplicate as follows: 12.5 µl RT2 SYBR® Green ROX™ qPCR Mastermix (Qiagen GmbH), 400 nM each of forward and reverse primers, 1 µl cDNA, nuclease-free H2O to 25 µl final volume. Cycling reactions were performed on a QuantStudio 7 Real-Time PCR system (Thermo Fisher Scientific, Inc.) with the following primers: TP53 qPCR forward, 5′-GAAGGAAATTTGCGTGTGG-3′ and reverse, 5′-TGTTACACATGTAGTTGTAGTGG-3′. Values were normalised against housekeeping gene ACTB (forward, 5′-GATCATTGCTCCTCCTGAGC-3′ and reverse, 5′-TCTGCGCAAGTTAGGTTTTGTC-3′). The thermocycling conditions were as aforementioned. Relative gene expression was calculated by the ∆∆Cq method and fold-change expression (2−∆∆Cq) was calculated relative to Cas9 (TP53-WT) control (19).

Protein analysis

Cells were treated for 6 h in either 0.3% DMSO or 1 µM CBL0137 (0.3% DMSO) final concentration in RPMI-1640 + 2% FCS + 50 U/ml penicillin + 50 µg/ml streptomycin. Cells were pelleted by centrifugation at 10,000 × g for 10 min at 4°C and lysed in 60 µl 1% NP-40 (IGEPAL®) buffer (Sigma-Aldrich; Merck KGaA). Protein concentration was determined by generating a standard curve using a bicinchoninic acid (BCA) assay. Lysates were denatured and 60 µg protein was electrophoresed on 4–15% pre-cast gels (Bio-Rad Laboratories, Inc.). Proteins were transferred to polyvinylidene difluoride membranes (Bio-Rad Laboratories, Inc.) and blocked with 1X Intercept blocking buffer (LI-COR Biosciences) for 1 h at room temperature before incubation with primary antibodies (mouse anti-p53, product no. 2524; rabbit anti-p21, product no. 2947; and rabbit anti-GAPDH, product no. 2118) at 4°C for 17–48 h. All primary antibodies were purchased from Cell Signaling Technology, Inc. and diluted at 1:1,000 in 1X Intercept blocking buffer. After washing with TBS-T (containing 0.1% Tween-20) and TBS, membranes were incubated with secondary antibodies [IRDye® 800CW anti-mouse (1:10,000) or IRDye® 800CW anti-rabbit (1:10,000)] (cat nos. 926-32212 and 925-32213, respectively; LI-COR Biosciences) for 2 h in the dark at room temperature and visualised on the LI-COR Odyssey® fluorescent scanner. For calculation of p21 expression fold change, Empiria Studio® Software version 2.0 (https://www.licor.com/bio/empiria-studio/) was used to normalise protein expression to housekeeping GAPDH protein. Fold expression change was then calculated for CBL0137-treated samples, relative to untreated samples for each respective cell line.

Statistical analysis

All calculations and statistical analysis of data were performed using GraphPad Prism Version 9.2.0 for Mac OS (GraphPad Software, Inc.). Kruskal-Wallis test with Dunn's multiple comparison post hoc test or Mann-Whitney U tests were performed to compare mean logLC50 values between each cell line tested. LogLC50 and LC50 values were extrapolated using non-linear regression analysis. All data was generated from 3 independent biological replicates. P<0.05 was considered to indicate a statistically significant difference.

Results

All identified TP53 mutations within the cell lines REH, Jurkat, U-937 and THP-1 occur within a critical region of the p53 DNA binding domain (Fig. 1A and B). The presence of TP53 loss-of-function genomic alterations significantly reduced the sensitivity of cells to CBL0137, independent of other genomic lesions present in each cell line (Fig. 1C and D). The aggregate mean logLC50 was greater for TP53MUT cell lines compared with TP53WT cell lines (mean logLC50=2.38 nM vs. 2.75 nM, P=0.046; Fig. 1D), with a more than two-fold increase in LC50 (Table SI). Notably, the KMT2A WT B-ALL cell lines NALM-19 and RCH-ACV exhibited low logLC50 values (logLC50=2.39 and 2.48 nM respectively; Fig. 1C). The KMT2Ar cell line THP-1 was the most resistant cell line investigated (logLC50=2.83 nM; Fig. 1C and D). A total of four cell lines, including two TP53WT and two TP53MUT were probed for the presence of total p53 protein, demonstrating increased levels of p53 following CBL0137 treatment in TP53WT cell lines only (Fig. 1E).

To further investigate the role of p53 in CBL0137 efficacy, heterozygous TP53 loss-of-function cell lines (TP53+/−) were generated using CRISPR/Cas9 in the human KMT2A-AFF1 ALL cell line RS4;11. The RS4;11 cell line was selected as a representative of KMT2Ar cell lines as it is highly sensitive to CBL0137 and expresses KMT2A-AFF1, the most common KMT2Ar identified in B-ALL (10). Heterozygous TP53 loss-of-function was confirmed in cell lines by RT-qPCR and immunoblot analysis (Fig. 2A and B). Both TP53+/− cell lines exhibited a significant reduction in TP53 expression by RT-qPCR, compared with control TP53WT RS4;11 cells (Fig. 2A). Immunoblot analysis demonstrated that treatment with 1 µM CBL0137 stimulated p53 expression in WT RS4;11 and positive control NALM-19 cells, but this effect was abrogated in both TP53+/− RS4;11 cell lines (Fig. 2B). The p53 effector protein p21 was upregulated ~two-fold in CBL0137-treated TP53WT RS4;11 cells but not TP53MUT cells, demonstrating that heterozygous TP53 loss-of-function is sufficient to abrogate activation of downstream p53 pathways (Fig. 2C). The cell senescence effector protein p21 is a well-characterised p53 effector protein, where p21 is rapidly activated following p53 activation, to induce cellular senescence and apoptosis (20,21). Annexin-V/7-AAD staining of TP53+/− cell lines revealed a ~two-fold increase in CBL0137 logLC50 (mean logLC50 WT=2.32 nM vs. TP53+/− #1=2.59 nM, P=0.043; WT vs. TP53+/− #2=2.62 nM, P=0.021; Fig. 2D and E), indicating that heterozygous TP53 loss-of-function is sufficient to cause a significant reduction in the sensitivity of cell lines to CBL0137.

Figure 2. Heterozygous loss of TP53 (TP53+/−) promotes CBL0137 resistance in KMT2Ar ALL. (A) Reverse transcription-quantitative PCR analysis of TP53 expression in RS4;11 cell lines validated heterozygous loss of TP53 expression. Data are mean 2−∆∆Cq + SD, relative to Cas9 control RS4;11 cells. Baseline expression refers to TP53 expression in TP53WT RS4;11 cells. (B) Representative immunoblot of total p53 protein in DMSO vehicle-treated (−) and 1 µM CBL0137-treated (+) RS4;11 cell lines. (C) Representative immunoblot of total p21 protein in DMSO vehicle-treated (−) and 1 µM CBL0137-treated (+) RS4;11 cell lines. Fold change in p21 expression following CBL0137 treatment was calculated relative to untreated samples, after normalizing to GAPDH. (D and E) CBL0137 dose-response and logLC50 for TP53WT and TP53MUT RS4;11 cell lines, measured by Annexin-V/7-AAD apoptosis assay. LC50 values are provided in Table SII. Data are presented as the mean + SD. For all statistical analysis, *P<0.05 and **P<0.01, for comparison of cell lines by Kruskal-Wallis test with multiple comparisons. All data is representative of 3 independent biological replicates. KMT2Ar, KMT2A-rearranged; ALL, acute lymphoblastic leukaemia; TP53WT TP53-wild type; TP53MUT, TP53-mutated; n.s., non-significant.

Discussion

In the present study, it was demonstrated that CBL0137 has similar potency in all tested TP53WT acute leukaemia cell lines, regardless of the presence of any additional genomic lesions, including KMT2Ar. However, the presence of a TP53 loss-of-function mutation significantly increased CBL0137 LC50, with a ~2-fold increase (range 1.7-4.0-fold) compared with TP53WT cell lines, regardless of genotype. It was also demonstrated, for the first time to the best of our knowledge, that heterozygous TP53 loss-of-function is alone sufficient to cause a significant increase in the LC50 of CBL0137 in the KMT2Ar B-ALL cell line RS4;11.

These results suggested that CBL0137 is indeed a promising therapy which may be broadly applicable in acute leukaemia. These data also indicated that cytotoxicity of CBL0137 is not specific to KMT2Ar, but rather the target depends on the presence or absence of functional TP53. Reduced sensitivity is expected in the context of TP53 mutated malignancies, and further testing is warranted to understand the clinical significance of TP53 mutation status in CBL0137 efficacy. It is important to note that thus far, all ALL xenograft models tested by the Pediatric Preclinical Testing Program were TP53WT (8). An additional 6/31 paediatric solid tumour models exhibited tumour growth delay, and 5/6 of these were TP53WT (8). This should also be investigated through further pre-clinical testing, and anticipated in its clinical development program. For instance, two recently completed Phase I studies of CBL0137 reported acceptable tolerability and some clinical activity against cancers of the liver, prostate, uterus, breast and ovary (22,23). The TP53 mutational status would optimise patient selection and target those most likely to benefit from this drug for further clinical testing. It is possible that mutations in p53 downstream mediators may also influence CBL0137 potency. However, alterations within p53 downstream mediators such as CDKN1A (p21) and BCL2 (BCL-2) are rarely reported in acute leukaemia. It is also possible that alterations within CK2 or the FACT complex may influence CBL0137 potency, as CBL0137-mediated activation of p53 occurs via these pathways (7), but data on the occurrence of mutations in these pathways in acute leukaemia is currently lacking (24).

The present findings highlighted the importance of p53 activation in CBL0137 efficacy in acute leukaemia, indicating that TP53 mutation status is an important factor in the clinical application of CBL0137. There is an ongoing need to identify therapeutic strategies for patients with high-risk acute leukaemia, such as those with TP53 mutations, to improve outcomes for these patients. The data of the present study indicated that CBL0137 is a promising anticancer therapy that depends on WT p53 activity, which thus far has not been therapeutically targetable. The clinical feasibility of CBL0137 across both TP53MUT and TP53WT malignancies will become evident as further clinical trials are performed. In the present study, it was clearly demonstrated that the potency profile of CBL0137 is tightly linked to TP53 mutation status, and it was revealed for the first time that heterozygous TP53 loss-of-function alone significantly affects response to CBL0137 in vitro. These results support the need for accurate determination of TP53 mutation status for patients enrolling in future CBL0137 clinical trials.

Supplementary Material

Supporting Data

Supporting Data

Acknowledgements

The authors would like to thank Dr Randall Grose (SAHMRI, Adelaide, Australia) for his technical assistance with flow cytometric experiments.

Funding

The present study was supported in part by the National Health and Medical Research Council (NHMRC), the Bristol-Meyers Squibb company, the Tour de Cure Australia, the Leukaemia Foundation Australia and the University of Adelaide.

Availability of data and materials

The datasets used and/or analysed in the present study are available from the corresponding author on reasonable request.

Authors' contributions

MOF conceptualised the presented idea. MOF performed experiments and constructed the manuscript in consultation with BJM, ECP, DTY, LNE and DLW. MOF and BJM confirm the authenticity of all the raw data. All authors provided critical feedback and helped shape the manuscript. All authors read and approved the final manuscript.

Ethics approval and consent to participate

The present study was performed under the International Bioethics Committee (IBC) certification number BC02/2018, and all work contained within the present study was approved by the Royal Adelaide Hospital HREC committee (approval no. HREC/15/RAH/54).

Patient consent for publication

Not applicable.

Competing interests

DLW receives research support from Bristol-Meyers Squibb, and Honoraria from Bristol-Meyers Squibb and Amgen. DTY receives research support from Bristol-Meyers Squibb and Novartis, and Honoraria from Bristol-Meyers Squibb, Novartis, Pfizer and Amgen. None of these agencies have had a role in the preparation of this manuscript. All other authors declare that they have no competing interests.

Glossary

Abbreviations

Abbreviations:

References