Combined HAT/EZH2 modulation leads to cancer-selective cell death

Epigenetic alterations have been associated with both pathogenesis and progression of cancer. By screening of library compounds, we identified a novel hybrid epi-drug MC2884, a HAT/EZH2 inhibitor, able to induce bona fide cancer-selective cell death in both solid and hematological cancers in vitro, ex vivo and in vivo xenograft models. Anticancer action was due to an epigenome modulation by H3K27me3, H3K27ac, H3K9/14ac decrease, and to caspase-dependent apoptosis induction. MC2884 triggered mitochondrial pathway apoptosis by up-regulation of cleaved-BID, and strong down-regulation of BCL2. Even aggressive models of cancer, such as p53–/– or TET2–/– cells, responded to MC2884, suggesting MC2884 therapeutic potential also for the therapy of TP53 or TET2-deficient human cancers. MC2884 induced massive apoptosis in ex vivo human primary leukemia blasts with poor prognosis in vivo, by targeting BCL2 expression. MC2884-treatment reduced acetylation of the BCL2 promoter at higher level than combined p300 and EZH2 inhibition. This suggests a key role for BCL-2 reduction in potentiating responsiveness, also in combination therapy with BCL2 inhibitors. Finally, we identified both the mechanism of MC2884 action as well as a potential therapeutic scheme of its use. Altogether, this provides proof of concept for the use of epi-drugs coupled with epigenome analyses to ‘personalize’ precision medicine.

Melting points were determined on a Buchi 530 melting point apparatus and are uncorrected. 1 H NMR and 13 C NMR spectra were recorded at 400 and 100 MHz, respectively, on a Bruker AC 400 spectrometer; chemical shifts are reported in δ (ppm) units relative to the internal reference tetramethylsilane (Me 4 Si). EIMS spectra were recorded with a Fisons Trio 1000 spectrometer; only molecular ions (M + ) and base peaks are given. All compounds were routinely checked by TLC and 1H NMR. TLC was performed on aluminum-backed silica gel plates (Merck DC, Alufolien Kieselgel 60 F254) with spots visualized by UV light. All solvents were reagent grade and, when necessary, were purified and dried by standard methods. Concentration of solutions after reactions and extractions involved the use of a rotary evaporator operating at reduced pressure of ca. 20 Torr. Organic solutions were dried over anhydrous sodium sulfate. Elemental analysis has been used to determine purity of the described compounds, that is, >95%. Analytical results are within ±0.40% of the theoretical values. All chemicals were purchased from Sigma-Aldrich, Milan (Italy), or from Alfa Aesar, Karlsruhe (Germany), and were of the highest purity.

Surface plasmon resonance
SPR analyses were performed on a Biacore 3000 optical biosensor equipped with research-grade CM5 sensor chips (Biacore AB). Recombinant p300/ KAT3B (Enzo Life Sciences, cat. # BML-SE451; GenBank accession no. NM_001429) HAT domain was immobilized (30 μg/mL in 10 mM sodium acetate, pH 4.5) were immobilized at a flow rate of 10 μL/min by using standard amine-coupling protocols to obtain densities of 15 kRU. Myoglobin was used as negative control and one flow cell was left empty for background subtractions. All compounds, dissolved in DMSO (100%), were diluted in HBS (10 mM HEPES pH 7.4, 0.15 M NaCl) always maintaining a final 0.1% DMSO concentration. Binding experiments were performed at 25° C, by using a flow rate of 30 μL/min, with 120 s monitoring of association and 300 s monitoring of dissociation. Regeneration of the surfaces was performed, when necessary, by a 10 s injection of 1 mM NaOH.
The simple 1:

Sar StudieS
When tested in leukemia NB4 cells at 3 μM for 30 h, MC2884 induced 52.1% cell death. Some MC2884 analogues have been tested to obtain SAR data (Supplementary Table 1). In particular, the shift of bromine atoms from meta to ortho (MC2908) and mainly to para (MC2909) position at the benzene rings reduced cell death to 30.3% and to 8.8%, respectively. Also the introduction of further bromine atoms at the meta positions of the two phenyl rings (MC2910) abated cell death induction (4.2%) of the derivative. At the N1 position, replacement of the benzyl with a methyl group (MC2914) led to a decrease of cell death induction (36%), whereas introduction of the longer phenylethyl moiety (MC3269) gave an increase of the effect (62.5% cell death). Further stretching of the N1 substituent to the phenylpropyl one (MC3272) lowered cell death induction (29.2%). Changing the N1-benzyl group of MC2884 with the N1-benzoyl group (MC3146) produced a severe loss of activity (13.2% cell death), but it was partially restored by introducing at N1 a 1-phenyl-2-ethan-1-one (MC3207, 38% cell death) or 3-phenyl-1-propan-1-one (MC3187, 28.9% cell death) unit, thus giving more flexibility to the N1 substituent. Replacement of the N1-benzyl-4-piperidone moiety with the tetrahydro-4H-pyran-4-one (MC2912) totally abated the cellular effect of the derivative (4.1% cell death), whereas the use of the isosteric tetrahydro-4H-thiopyran-4-one (MC2913) strongly restored the cell death, MC2913 showing the same potency of MC2884 (52.7% cell death).
Western blot analyses (and their quantification by ImageJ) performed on NB4 cells to detect the levels of acetyl-H3K9-14 (as a marker of HAT/anti-HAT activity) and H3K27me3/2 (as marker of PRC2/EZH2 activity) showed a decrease of histone acetylation and methylation after treatment with 3 μM compounds for 48 h, thus confirming the role of dual HAT/EZH2 inhibitors for these derivatives. In addition, the tested compounds also lowered the level of the EZH2 protein (SAR Figure 1).  Table 1). In general, the SHSY5Y cell line is less sensitive to the tested molecules than NB4 cells. Anyway, also in this cell line the bis 2-bromophenyl (MC2908), bis 4-bromophenyl (MC2909), and bis 3,5-dibromophenyl (MC2910) were less potent or not active as cell death inducers. As in NB4 cells, substitution at N1 with methyl (MC2914) or 3-phenylpropyl (MC3272) groups reduced the potency of the corresponding derivatives whereas introduction of a phenethyl group (MC3269) gave a compound with the same potency as MC2884. The presence of a carbonyl function within the N1 substituent abated the activity of the compounds, with the exception of that showing a 3-phenyl-1-propan-1-one chain at N1 (MC3187), which in this case was slightly more potent than MC2884. The tetrahydro-4H-pyran-4-one MC2912 failed in inducing cell death in SHSY5Y cells, whereas its thio-analogue MC2913 displayed the best death induction activity, similarly to what observed in NB4 cells.

Sar-figure 1: analysis of histone & non-histone target modulation by mC2884 and analogues in nB4 cells.
Interestingly, compounds inactive in inducing cell death in both cell lines such as MC2910, 2912, 2909, 2886, 2911, 3395 where also unable to modulate histone acetylation and EZH2 deregulation (SAR Figure 2). In full agreement, only MC2884 was able to down-regulate BCL2 expression levels at both RNA (histogram bars) and protein levels (western blot) suggesting that this action is crucial for the induction of apoptosis.
Finally, to causally connect the anticancer action with the dual p300 and EZH2 targeting of the drug, we evaluated MCF7 cells viability after overexpression of WT or catalytic mutants of p300 or EZH2, after treatment with MC2884 (SAR Figure  3). Interestingly, expression of p300 or EZH2 to higher levels leaded to increased cell death after MC2884, suggesting that both targets are important for the anticancer action of the drug in these settings. When catalytic mutants where used, only the mutant p300 expression fully reversed the MC2884 induction of cell death, whereas the mutant EZH2 gave rise to a partial reversion, potentially suggesting that both activities are needed but p300 inhibitory enzymatic action might be prevalent. Note that MC2884 action on EZH2 expression levels (and thus action that might be additional to the enzymatic regulation) is also potentially playing a role.

pHarmaCology pharmacokinetic analysis (mice-iV, ip pK study)
Compound was administered both intravenously and intraperitoneally to mice. Blood samples were collected at 8 time points over 24 hr and plasma analysed by LC-MS/ MS to determine the concentration of compound.
The plasma time concentration profile was calculated with the main PK parameters (Co, AUClast, t.,VD, and CL).
formulation For ip dosing. After formulation assessment, MC2884 was dissolved at 0.2 mg/mL. This provided a dose of 2 mg free base material/kg when administered ip in a 10 mL/kg dosing volume.
For iv dosing. Compound was dissolved at 0.4 mg/mL. This provided a dose of 2 mg free base material/kg when administered iv in a 5 mL/kg dosing volume.

mouse terminal ip pK study
Compound MC2884 was given ip at a dose of 2 mg/kg to a group of 24 normally fed male CD1 mice in a 10 mL/ kg injection volume. Terminal blood samples (> 230 μL) were taken under CO2 from groups of 3 mice at each of 8 time-points post dose (0.083, 0.25, 0.5, 1, 2, 4, 8 and 24 h) and placed into heparinized tubes. Samples were placed on ice for no longer than 30 min before centrifugation (10,000 rpm × 3 min). Plasma samples (>100 μL) were collected into fresh tubes and frozen on dry ice. All samples were stored at −20° C Number of plasma samples = 24.

Bioavailability mouse terminal iV pK study
Compound MC2884 was given iv via the lateral tail vein at a dose of 2 mg/kg to a group of 24 normally fed male CD1 mice in a 5 mL/kg injection volume. Terminal blood samples (> 230 μL) were taken under CO2 from groups of 3 mice at each of 8 time-points post dose (0.083, 0.25, 0.5, 1, 2, 4, 8 and 24 h) and placed into heparinized tubes. Samples were placed on ice for no longer than 30 min before centrifugation (10,000 rpm × 3 min). Plasma samples (>100 μL) were collected into fresh tubes and frozen on dry ice. Number of plasma samples = 24.
Consistent IV profiles were obtained, with little variability across the time-points (the average profile is shown in the second spreadsheet). An average half-life of 4.58 h was calculated.

Cytochrome p450 inhibition (5 isoform iC50 determination)
ADME-Tox Cytochrome P450 inhibition of MC2884 using human liver microsomes as an estimation of in vitro metabolism.
MC2884 (0.1 μM-25 μM) was incubated with human liver microsomes and NADPH in the presence of a cytochrome P450 isoform-specific probe substrate. For the CYP2C9, CYP2C19, CYP2D6 and CYP3A4 specific reactions, the metabolites were monitored by mass spectrometry. CYP1A activity was monitored by measuring the formation of a fluorescent metabolite. A decrease in the formation of the metabolite compared to the vehicle control was used to calculate an IC50 value (test compound concentration which produces 50% inhibition).

Cyp3a4 inhibition
Six test compound concentrations (0.1, 0.25, 1, 2.5, 10, 25 μM in DMSO; final DMSO concentration 0.26%) were incubated with human liver microsomes (0.1 mg/mL) and NADPH (1 mM) in the presence of the probe substrate midazolam (2.5 μM) for 5 min at 37° C. The selective CYP3A4 inhibitor, ketoconazole, was screened alongside the test compounds as a positive control. For the CYP1A incubations, the reactions were terminated by methanol, and the formation of the metabolite, resorufin, was monitored by fluorescence (excitation wavelength = 535 nm, emission 5 wavelength = 595 nm). For the CYP2C9, CYP2C19, CYP2D6 and CYP3A4 incubations, the reactions were terminated by methanol. The samples were then centrifuged, and the supernatants were combined, for the simultaneous analysis of 4-hydroxytolbutamide, 4-hydroxymephenytoin, dextrorphan, and 1-hydroxymidazolam plus internal standard by LC-MS/MS. Formic acid in deionized water containing internal standard (final formic acid concentration = 0.1%) was added to the final sample prior to analysis. A decrease in the formation of the metabolites compared to vehicle control was used to calculate an IC50 value (test compound concentration which produces 50% inhibition). a) CYP1A no inhibition, IC50 (µM) >25 b) CYP2C9 no inhibition, IC50 (µM) >25 c) CYP2C19 no inhibition, IC50 (µM) >25