Fig 2: The RelA T505A mutation leads to reduced survival of Eμ-Myc/RelaT505A mice, genomic instability and resistance to CHK1 inhibitors.(A) Reduced survival of Eμ-Myc/RelaT505A mice. Kaplan–Meier survival curves for Eμ-Myc (n = 27) and homozygous Eμ-Myc/RelaT505A male mice (n = 11). Overall survival is significantly shorter in Eμ-Myc/RelAT505A mice (***P = 0.0001, Mantel–Cox test) and hazard ratio (HR) analysis indicates that these mice are at 7.5 times greater risk of dying earlier due to lymphoma compared with Eμ-Myc mice. The median survival for each genotype is indicated. (B) Increased genomic instability in lymphomas from Eμ-Myc/RelaT505A mice. Flow cytometric analysis of B cells prepared from Eμ-Myc and Eμ-Myc/RelaT505A spleens and stained for γH2AX, a marker of DNA damage (**P = 0.0018, Unpaired Student's t-test). The red dot in each case illustrates the mouse shown in Supplementary Figure S3A,B. (C) Western blot analysis of RPA2 phospho-Ser33, total RPA2, total CHK1 and ACTIN in snap frozen tumour extracts prepared from three different reimplanted WT Eμ-Myc and Eμ-Myc/RelaT505A tumours mouse inguinal lymph nodes. (D) Schematic diagram illustrating the CHK1i in vivo study in Eμ-Myc and Eμ-Myc/RelaT505A mice. Six-week-old C57Bl/6 WT mice were implanted with either Eμ-Myc or Eμ-Myc/RelaT505A lymphoma cells (blue arrow) and once tumours became palpable were treated with either 100 mg/kg CCT244747 p.o or vehicle control once daily for 9 days (black arrows). Mice were euthanised 24 h after the final dose (red arrow) and tumour burden assessed. (E) Reimplanted Eμ-Myc/RelaT505A lymphoma cells are resistant to CHK1 inhibition in vivo. Line graphs showing the mean response of the five reimplanted Eμ-Myc (red) and four Eμ-Myc/RelaT505A (orange) tumours and their response to CCT244747. A response was defined as a significant reduction (or increase) in tumour burden (P < 0.05) using Unpaired Student's t-tests (*P < 0.05, **P < 0.01). (F) Table showing the response of reimplanted Eμ-My and Eμ-Myc/RelaT505A tumours to CCT244747, in all sites where lymphoid tumour burden is anticipated in this model.

Fig 3: ASNase-treated malignant B cells exhibit oxidative stress-associated DNA damages, which are prevented by PHGDH overexpression.a Total protein extracts prepared from Eµ-Myc #688 cells cultured in Asn-free medium, supplemented (+) or not (−) with Asn (0.37 mM) and ASNase (0.003 IU/ml) for 24 h, were immunoblotted for the indicated proteins. Etoposide treatment (1 µg/ml) for 3 h was used as a positive control. b Relative quantification of immunoblots presented in (a). (#506 cells, square, n = 2 independent experiments; #688 cells, triangle, n = 3 independent experiments). c As in a for the indicated proteins expressed in Eµ-Myc #506 and #688 cells. From #506 cells, the samples derive from the same experiments, but different gels for P-ATR (S428), P-Chk1 (S345), KAP1, ERK2, another for ATR, another for Chk1, P-RPA32 (S33), P-KAP1 (S824), another for RPA32, and another for γH2AX were processed in parallel. From #688 cells, the sample derived from the same experiments, but different gels for P-ATR (S428), P-Chk1 (S345), P-KAP1 (S824), ERK2, another for ATR, Chk1, KAP1, another for P-RPA32 (S33), another for RPA32, and another for γH2AX were processed in parallel. d As in (b) for immunoblots presented in (c) (#506 cells, square, n = 2 independent experiments; #688 cells, triangle, n = 3 independent experiments). e γH2AX expression levels in Eµ-Myc cells (#506 and #688) following 24 h incubation in Asn-containing medium supplemented (+) or not (−) with ASNase (0.003 IU/ml) and/or N-acetyl-L-cysteine (NAC, 10 mM). f Relative quantification of γH2AX expression in Eµ-Myc cells presented in (e) (#688 cells, triangle, n = 2 independent experiments; #506 cells, square, n = 1). g γH2AX expression levels in Eµ-Myc (#688) cells treated (+) or not (−) for 24 h with DMSO (−/−), ASNase (0.003 IU/ml) and/or BI-4916 (10 µM). h Relative quantification of γH2AX expression levels in Eµ-Myc cells presented in (g). (#688 cells, triangle, n = 2 independent experiments; #506 cells, square, n = 1). i Total protein extracts prepared from BCL harvested from C57BL/6 mice engrafted with control (CTL) or V5-tagged murine PHGDH-overexpressing (PHGDH OE) Eµ-Myc (#506) cells and treated with Vehicle or ASNase from day 7 until disease endpoint, were immunoblotted for the indicated proteins (CTL-Vehicle, n = 3; CTL-ASNase, n = 3; PHGDH OE-Vehicle, n = 4; PHGDH OE-ASNase, n = 4 mice). The samples derive from the same experiments, but different gels for γH2AX, ERK2 (upper), another for P-ChK1 (S345), ERK2 (middle), another for Chk1, another for PHGDH, ERK2 (lower), and another for V5 were processed in parallel. j Relative quantification of γH2AX and P-Chk1 (Ser345) expression levels in BCL presented in (i). V Vehicle, A ASNase. For all graphs, data are expressed as mean ± SD. P-values are from one-way Anova followed by Tukey’s test (b, d), 2-way Anova followed by Tukey’s test (f, h, j) and indicated as ns, not significant; *, p < 0.05; **, p < 0.01; ***, p < 0.001 ****, p < 0.0001. For detailed individual P-value, please refer to the Source data table.

Fig 4: Combination therapy can overcome resistance mediated by MYC overexpression and ABC transporters.A. Immunoblot of c-MYC, H3K9ac, Ɣ-H2AX, PARP protein levels in whole cell lysate of normal human pancreas cells with or without treatment. B. Immunoblot of c-MYC, H3K9ac, p-RPA2, Ɣ-H2AX, PARP protein levels in whole cell lysate of wild type (parental) and c-MYC over-expressing MIa PaCa-2 cells with or without treatment. C. Immunoblot analysis of c-MYC, H3K9ac, γ-H2AX, and cleaved PARP (c-PARP) protein levels in whole cell lysates derived from KPC mouse model exhibiting high and low c-MYC expression. Cells were treated as indicated before. D. Analysis of apoptotic cell death in normal human pancreas cells, wild type (parental), and MYC-high transfected Mia PaCa-2 cells following treatment for a total of 24 hours. Annexin V–positive cells were determined by flow cytometry. Statistical significance was determined by two-way ANOVA in the Prism software, [**], p value 0.0028, and [****], p value<0.0001. E. Left: Cytotoxicity graphs of the effect of DMPatA in HEK293 cell lines transfected with various ABC transporters: ABCB1, ABCG2, ABCC1, and pc-DNA vector following treatment. Three independent biological replicate experiments were performed. E. Right: Comparison of IC50 values in HEK293 cell lines transfected with ABC transporters (ABCB1, ABCG2, ABCC1) or pcDNA vector following treatment. Statistical significance was determined by one-way ANOVA in the Prism software. [*], p-value<0.0147. F-H. Gene expression plots from RNA-seq analysis in KLM and MIA PaCa-2 cells. black dots denote individual samples.

Fig 5: Hyperacetylation cause accumulation of co-transcriptional R-loops, replication fork stalling and increase in DNA damage.A-C The experiments were conducted in MIA PaCa-2 cells incubated for a total of 9h including 6h of 15nM romidepsin (both Romi and Romi+DMP) followed by an additional 3h in drug free medium (Romi) or 2nM DMPatA (Romi + DMP) or 9 hours in 2nM DMPatA (DMPatA). A. Left Confocal images of DNA–RNA hybrid accumulation in treated cells. Staining was carried out using antibodies against S9.6 (red) and nucleolin (green). Nuclei were stained with DAPI. As seen in the three lower photomicrographs, treatment with RNaseH1 eliminated the DNA– RNA hybrids seen with treatment. Scale bar: 20 µm. Right Quantification of fluorescence intensities of nuclear S9.6 immunofluorescence signal without and with RNaseH1. The S9.6 signal was quantified only in the nuclear regions. The median of the nuclear S9.6 signal intensity per nucleus is shown. Statistical significance was determined by one-way ANOVA using Tukey multiple comparisons test in Prism software, [****], p-value<0.0001. B. DRIP–qPCR fold increase in the percent of input values for the VGF, HBA2, and SNRPN loci. Genomic DNA was treated in vitro with or without RNaseH1 before immunoprecipitation using the S9.6 antibody. Three independent biological replicate experiments were performed, each with n=3 technical replicates. Statistical significance was determined by 3-way ANOVA [**], p-value<0.003, [***], p-value<0.001. C. ChIP-qPCR shows fold increase in the percent of input values for the VGF, HBA2, and SNRPN loci. Genomic DNA was immunoprecipitated using H3K9ac antibody. Three independent biological replicate experiments were performed, each with n=3 technical replicates. Statistical significance was determined by 3-way ANOVA. [*], p-value<0.03, [***], p-value<0.001. D, E, and G The experiments were a total of 24h including 6h of 15nM romidepsin (both Romi and Romi+DMP). D. Immunoblot of p-RPA2 and p-CHK1 protein levels from whole cell extracts following drug treatment. E. Left DNA fiber analysis of untreated control cells or treated cells. Scale bar: 0.5 µm. Right Dot plot of IdU to CldU tract length ratios for individual replication forks in the cells following treatment. To calculate the length in kilobase (Kb) the track lengths were multiplied by a factor 2.59. Three independent biological replicate experiments were performed, each with n=3 technical replicates. F. Immunoblot of H3K9ac and Ɣ-H2AX protein levels from whole cell extracts following treatment. G. Left Confocal imaging of ƔH2AX-labeled (red) and DAPI counterstain (blue) cells. Scale bar: 0.5 µm. Right Quantification of nuclear Ɣ-H2AX signal. The median value of at least 50 nuclei signal per experimental condition is indicated. Statistical significance was determined by one-way ANOVA using Tukey multiple comparisons test. [*], p-value<0.028, [**], p-value<0.007, [****], p value<0.0001. Abbreviations: DMP, DMPatA; Romi, romidepsin.