ABSTRACT
Cyclin-dependent kinases are therapeutic targets frequently deregulated in various cancers. By convenient alkylation of the 5-sulfanyl group, we synthesized 3-isopropyl-7-[4-(2pyridyl)benzyl]amino-1(2)H-pyrazolo[4,3-d]pyrimidines with various substitutions at position 5 with potent antiproliferative activity in non-Hodgkin lymphoma cell lines. The most potent derivative 4.35 also displayed activities across more than sixty cancer cell lines. The kinase profiling confirmed high selectivity of 4.35 toward CDKs 2, 5 and 9, and the cocrystal with CDK2/cyclin A2 revealed its binding in the active site. Cultured lymphoma cell lines treated with 4.35 showed dephosphorylation of CDK substrates, cleavage of PARP-1, downregulation of XIAP and Mcl-1 and activation of caspases, which collectively confirmed ongoing apoptosis. Moreover, 4.35 demonstrated significant activity in various cell line xenograft and patient-derived xenograft mouse models in vivo both as a monotherapy and as a combination therapy with the BCL2-targeting venetoclax. These findings support further studies of combinatorial treatment based on CDK inhibitors.
KEYWORDS
cyclin-dependent kinase, lymphoma, pyrazolo[4,3-d]pyrimidine, venetoclax
INTRODUCTION
Non-Hodgkin lymphomas (NHLs) are the most frequent hematologic malignancies in the Western hemisphere, comprising approximately 30% of all hematologic cancers. Based on the behavior of the cancer, NHLs can be classified as indolent, aggressive and highly aggressive lymphomas. The latter two NHL types include diffuse large B-cell lymphoma, Burkitt’s lymphoma, mantle cell lymphoma and peripheral T-cell lymphomas. The deregulation of cell cycle machinery together with the disruption of programmed cell death are hallmarks of aggressive NHL. These hallmarks include mutated or overexpressed oncogenes (e.g; MYC, BCL6, MYD88, cyclin D1,BCL2, CDK4/6) or the deletion of tumor suppressors (e.g; TP53, TP63, RB1, CDKN2A, ATM). Doubleand triple-hit lymphomas characterized by the concurrent genetic aberrations of MYC, BCL-2 and/or BCL6 represent a subgroup of NHL
with an extremely dismal prognosis.1
Some of the recurrently overexpressed or mutated genes encode drug targets, such as BCL-2 protein, Bruton’s tyrosine kinase or phosphoinositide 3-kinase, which can be modulated by the small molecule inhibitors venetoclax, ibrutinib or idelalisib, respectively.2-4 Another emerging pharmacological approach targets cyclin-dependent kinases (CDK), which were first discovered for their role in regulating the cell cycle.5 Depending on their selectivity, CDK inhibitors can interfere either with cell cycle progression by blocking RB protein phosphorylation (governed by CDKs 2/4/6) or with global transcription by inactivating RNA polymerase II (governed by CDKs 7/8/9/12). However, many inhibitors have broad specificity and simultaneously obstruct both of these processes,6-8 providing the advantage of overcoming possible compensatory functions among the CDK family.9
CDK inhibitors, including clinical candidate drugs such as flavopiridol, roscovitine, dinaciclib and the recently approved drugs palbociclib and abemaciclib, have been
increasingly evaluated in various lymphomas.10-14 In NHL, similar to other cancers, the pharmacological inhibition of transcriptional CDKs leads to the downregulation of short-lived mRNAs related to cell survival, such as those of MCL-1, XIAP, MYC and cyclin D1. The downregulation of cyclin D1 and MCL-1 protein levels was observed in mantle cell lymphomas and in aggressive MYC-driven B-cell lymphoma treated with roscovitine and dinaciclib, respectively.11,12 Despite these compounds inhibit multiple CDKs, a widely accepted mechanism of action involves CDK9, which was independently supported by findings obtained with the highly CDK9-selective drug atuveciclib.15 Importantly, the loss of MCL-1 as a consequence of CDK9 or CDK7 inhibition by these drugs has been found to strongly enhance apoptosis induction in NHL by small-molecule antagonists of BCL-2.16-21
The pan-selective purine CDK inhibitor roscovitine, which was discovered in our laboratory, was among the first CDK inhibitors that entered clinical trials.22,23 The further exploration of roscovitine has been mainly oriented toward either modifications in its substitutable positions24-26 or the redistribution of nitrogen atoms of the purine scaffold; the latter approach culminated in nanomolar inhibitors based on pyrazolo[4,3-d]pyrimidines and pyrazolo[1,5-a][1,3,5]triazine.27-30
The objective of this work was to synthesize novel potent pyrazolo[4,3-d]pyrimidine CDK inhibitors, which proved to be more potent than related purines,28-30 and explore their activity toward lymphoma models. The heterocyclic core was substituted with groups that conferred high activity in purines, identified during our previous structure-activity studies24,26; i.e. isopropyl moiety at position 3, benzylamines at position 7 and alkylor cycloalkylamines at position 5.29-30 The compounds were prepared using more convenient alkylation of 5-sulfanyl derivatives, providing substantially higher yields over the 5-amino-substituted pyrazolo[4,3-d]pyrimidines described earlier.30,31 In line with our expectations, the newly prepared derivatives displayed nanomolar potency against CDKs and cancer cell lines. Our previous study demonstrated that roscovitine strongly sensitizes leukemia and lymphoma cells to TRAIL-induced apoptosis by altering the levels of BCL-2 proteins in the mitochondria32 and, together with other observations,11 this led us to explore anticancer activity of new pyrazolo[4,3-d]pyrimidines against several lymphoma models. The compounds proved to potently kill various lymphoma cell lines and also displayed significant in vivo activity both in monotherapy and in a combination with venetoclax, which corresponds to the abovementioned findings.
Results and Discussion
Design and Synthesis
We recently described 5-alkylamino-3-isopropyl-7-[4-(2-pyridyl)benzyl]amino-1(2)Hpyrazolo[4,3-d]pyrimidines that share some of the same substitutions as the purine CDK inhibitor roscovitine but that showed substantially higher potency against CDKs and cancer cell lines than did roscovitine.28,30 Due to the laborious synthesis of these derivatives, we sought to improve the synthetic accessibility of these compounds while retaining or even improving their biochemical potency. We focused on the substituted position 5 on the pyrazolo[4,3-d]pyrimidine scaffold and report here a new series of 3,5,7-trisubstituted pyrazolo[4,3-d]pyrimidines in which the alkylamino group at the 5-position of the heterocycle is replaced by a 5-alkylthio group. To verify the effect of this modification, we employed molecular modeling. We applied the computational procedure delineated in our previous work, in which we identified the pyrazolo[4,3-d]pyrimidine core as the most favorable central heterocycle of the purine bioisosteres of CDK2 inhibitors.33 Importantly, we have shown that the contributions of the central heterocyclic cores and the individual substituents, quantified and evaluated in relation to conformations of the optimized protein-inhibitor complexes, are not simply additive. Depending on the scaffold, the same substituents can be associated with different interaction “free” energies ΔG’int (in kcal mol-1). Our preliminary calculations suggested that the replacement of the alkylamino group at the 5-position with the 5-alkylthio group can increase the binding affinity to CDK2 (selected data are available in Supplementary Information).
The new compounds have been prepared from 3-isopropyl-5-sulfanyl-1(2)Hpyrazolo[4,3-d]pyrimidin-7-ol 1 (synthesized according to a described procedure28), which was first converted to 5,7-dithiol 2 and subsequently substituted at the heterocycle position 7 with 4-(2-pyridyl)benzylamine. The final compounds 4 were prepared by the alkylation of the reactive 5-sulfanyl functional group of 3 by various agents (Scheme 1); in the final step, we successfully prepared 24 alkyl halogens (yielding compounds 4.2 4.4, 4.6 4.11, 4.17, 4.18, 4.21 4.23, 4.25, 4.26, 4.28, 4.30 4.33, 4.35), 8 epoxides (yielding compounds 4.1, 4.5, 4.12, 4.16, 4.19, 4.29, 4.33, 4.34) and aziridines (yielding compounds 4.13, 4.24 and 4.35).
The reactivity of the sulfanyl functional group enables the use of different classes of alkylating agents, eventually leading to the same compounds with similar yields (4.33 and 4.35 were prepared not only by using alkyl halogens but also by using suitable epoxide or aziridine, respectively). This new synthetic approach overcomes the less convenient nucleophilic aromatic substitution (SNAr, which requires difficult reaction conditions, makes undesired by-reactions and subsequently by-products) described in our previous article examining 5-alkylamino-1(2)H-pyrazolo[4,3-d]pyrimidine derivatives.30
Structure-Activity Relationships
The implementation of the most suitable substitution at the 5 position of the pyrazolo[4,3d]pyrimidine core was mostly inspired by previous SAR studies on trisubstituted pyrazolo[4,3-d]pyrimidine CDK inhibitors,29,30 related purines26,34-37 and pyrazolo[1,5a]pyrimidines.38 The compounds we prepared included inhibitors with significant activity against CDK2/E; all compounds exhibited regular medication an IC50 below 200 nM, and approximately 50% of inhibitors blocked the activity of the simple 5-methylthio derivative 4.14 (IC50=48 nM). The substituent in position 5 is mostly in contact with the flexible parts of the enzyme surface or with the solvent (water molecules) and therefore can be more variable, whereas the side chains at positions 3 and 7 do not allow such variability. Despite low differences in the CDK2 inhibition measured with the studied compounds, we found some relationships that are associated with the Leptomycin B cell line size/length of the substituent at position 5 or its polarity. Compound 3, which contains an acidic sulfanyl functional group in position 5, was the weakest in the series, with nearly three-fold lower potency than an analogous 3,7-disubstituted compound described earlier.30 The methylation of the 5-sulfanyl functional group results in short lipophilic substituents in 4.14, increasing the potency more than ten-fold.
The weakest CDK2 inhibition (IC50 between 100 200 nM) was observed for compounds bearing rigid piperazine and morpholine moieties (4.2 and 4.1, respectively). Additional groups with weak activity against CDK2 included compounds with less polar (4.4, 4.7, 4.11, 4.14) or cycloalkyl (4.5, 4.12, 4.13) substituents at position 5 or those with tertiary amino groups (4.8, 4.9) (IC50=48 138 nM). In contrast, the most potent inhibitors of the series (IC50=2 39 nM) bear at least one polar group (mainly a hydroxy or primary amino group) placed either on a short linear (4.35, 4.33, 4.22, 4.23, 4.28) or on a small branched chain (4.34, 4.32, 4.30, 4.29, 4.27, 4.26). Further modification of 2-aminoethylthio in the most active compound 4.35 led to the impairment of CDK2 inhibition; this effect was observed when the terminal amino group was modified by methylation, acetylation and conversion to urea or guanidine (4.28, 4.15, 4.9, 4.27 and 4.20, respectively) and is probably due to steric hindrance due to possible hydrogen bonding, increased basicity of the amine (4.28 and 4.20), or a combination of both. Other attempts to modify the primary amine in 4.35 diminished its activity, such as its extension with aminomethyl (4.30) or its cyclization (4.13 a 4.17). A similar scenario is also evident for the modified hydroxyethylthio derivatives, among which only compound 4.34 reached the same potency as the analog 4.33; all other compounds (4.32, 4.29, 4.19, 4.18, 4.16, 4.12, 4.11, 4.5, 4.3, 4.2, 4.1, 4.31) have worse CDK2 inhibition.
Recent experiments have shown that pharmacological CDK inhibition is an effective and rational treatment option for aggressive MYC-driven lymphomas.11 We therefore screened all prepared compounds against the non-Hodgkin lymphoma cell lines HT, OCI-LY2 and MINO.
The antiproliferative activity of the prepared compounds was consistent with enhanced CDK2 inhibition (Table 1). The most potent compounds, 4.28, 4.29, 4.31, 4.32, 4.33, 4.34 and 4.35, had notably improved potency compared with the related purine CR8, with an average GI50 value 5-fold higher than that of CR8.
To obtain structural insight into the interaction of our inhibitors with the enzyme, we determined the crystal structure of the active Thr160-phosphorylated CDK2 with a fragment of cyclin A2 (residues 175-432) in complex with compound 4.35 at 2.15 Å resolution (PDB code: 6GVA). For data collection and refinement statistics, see Supplementary Table 2. The asymmetric unit contains one CDK2/cyclin A2 heterodimer. The residues 39-40 and Cterminal residues 291-298 of CDK2 could not be modeled into the electron density map due to disorder.
The inhibitor binds to the active site located in a narrow cleft between the Nand Cterminal domains of CDK2 in the standard binding mode typical for pyrazolo[4,3d]pyrimidine-based inhibitors (see Figure 2). Specifically, the pyrazolo[4,3-d]pyrimidine core is sandwiched between the hydrophobic side-chains of Ile10 and Leu134 and interacts via three conserved hydrogen bonds with the hinge region residues (7-amino N-H···O:Leu83, N2-H···O:Glu81, and a weak N1···HN:Leu83 bond). The R3 isopropyl group forms hydrophobic interactions with the gatekeeper residue Phe80. The terminal amino group of the R5 moiety forms a hydrogen bond with the carbonyl oxygen of Gln131 from the phosphateribose pocket. The proximal phenyl ring of the R7 moiety is sandwiched between Ile10 and the backbone of His84, close to the specificity surface. The terminal pyridine ring sticks out of the binding cleft, roughly in plane with the proximal phenyl ring. Moreover, N4 from the pyrazolo[4,3-d]pyrimidine core of the inhibitor interacts through a chain of three water molecules (W414, W435, and W436) with the side chains of Lys33 and Glu51 and with the main chain of Asp145. Another chain of three water molecules (W409, W426, and W471) provides additional hydrogen bonding of the amino group on the R5 moiety to the main chain of Glu12 and to the side chains of Asp86 and Lys89 at the edge of the binding cleft.
Kinase Selectivity
As the most potent compound, 4.35 was then assayed on a panel of related CDKs. 4.35 had low nanomolar activity not only against CDK2 but also against CDK5 and CDK9 (Table 2). The compound displayed at least 10-fold increased potency toward most of the assayed CDKs compared to the related purine CR8, with the highest increase (43-fold) toward CDK4/cyclin D1. Next, preliminary selectivity profiling of 4.35 was performed on a panel of 50 additional protein kinases, a representative sampling of the human kinome. The compound was assayed at a single concentration of 1 µM. As shown in Supplementary Table 3, significant inhibition was observed for CK1δ and CHK2; the residual activity of these kinases decreased to 11 and 18%, respectively. We then determined the IC50 values for CK1δ and CHK2 (0.045µM and 0.525 µM, respectively). Weak inhibition of PAK4, CAMK1 or RSK1 kinases was also observed; these kinases are known to be sensitive to related CDK inhibitors with pyrazolo[4,3-d]pyrimidine or purine scaffolds.29,39-41 These results confirmed the reasonable selectivity of 4.35.
We next assessed the anticancer activity of 4.35 against additional lymphoma cell lines as well as other human cancer cell lines (Table 3). 4.35 displayed a broad range of activity across all assayed cell lines, with the average concentration to reduce cell proliferation by 50% (GI50 value) of 21.6 nM. The most sensitive cell lines were the diffuse large B–cell lymphomas OCI-LY3, RIVA and HBL-1. The clinical candidate dinaciclib, assayed in parallel, displayed similar potency.
To further assess the anticancer selectivity of 4.35, we also screened the NCI60 cell line panel. Notably, 4.35 robustly inhibited proliferation across the entire cancer cell line panel with a mean GI50 of 23 nM (Supplementary Table 4). In general, 4.35 was effective against leukemia, melanoma and breast cancer cells but showed decreased activity in ovarian and
renal cells.
Mechanism of Action in Cells
Compound 4.35 emerged as the most potent CDK inhibitor in this series, and because of its superior proapoptotic activity in several lymphoma cell lines (Supplementary Figures S6 and S7), we sought to characterize its cellular effects. We therefore performed immunoblotting analyses of MINO cells exposed to 4.35 both at different doses and for different lengths of time. The treated cells contained clear markers of ongoing apoptotic cell death, including the appearance of the PARP-1 fragment (89 kDa), diminution of full-length PARP-1, activation of caspases 3 and 7 and reduction in the levels of the anti-apoptotic proteins Mcl-1 and XIAP (Figure 3A, B). On the other hand, the anti-apoptotic proteins BCL2 and BCL-XL did not show any change, probably due to their high stability. Clear markers of apoptosis were also detected in UPF1H and MAVER-1 cells (Supplementary Figure S8). These results are consistent with the known mechanisms of the pro-apoptotic action of other CDK inhibitors.11,12,24,30
The kinetics of CDK inhibition by 4.35 was studied in MINO cells over the period of 24 h (Figure 3C). The dose of 12.5 nM was selected because it induced a strong response within 24 h in previous experiments (Figure 3A, B), and it was anticipated that CDK inhibition would be evident at earlier time points. Importantly, a gradual decrease in the phosphorylation of the C-terminus of RNA polymerase II (at serines 2 and 5) and retinoblastoma protein (at serines 608 and 780) in a time-dependent manner was observed, suggesting that at least CDK2, CDK7 and CDK9 were effectively inhibited.
In Vivo Activity: Monotherapy
The in vivo efficacy of 4.35 was then assessed in murine models of aggressive lymphomas based on subcutaneous xenotransplantation of three established lymphoma cell lines (MINO, MAVER-1 and HBL-2) and one patient-derived lymphoma xenograft (PDX VFN-M1) in immunodeficient mice. Due to relatively rapid pharmacokinetics and limited oral availability (for details see Supplementary Information), compound 4.35 was administered daily by i.v. injections. Treatment with 4.35 was associated with a significant reduction in the growth of all three xenografted lymphoma cell lines compared to controls (Figure 4). Importantly, antilymphoma activity was also clearly achieved in the PDX mouse model VFN-M1, which represents a surrogate model for primary lymphoma cells from patients (Figure 4).
Ex vivo analysis of selected pharmacodynamic targets of 4.35 in subcutaneously grown lymphoma tumors 24 h after a single i.v. administration of 4.35 confirmed the induction of apoptosis and the downregulation of MCL-1 and XIAP proteins but had no impact on the expression of BCL-2 (Figure 5, Supplementary Figure S10). Notably, the changes were also evident in the PDX model VFN-M1, a reliable surrogate for the primary cells of patients (Figure 5C). In addition, reduced RB and RNA polymerase II phosphorylation at several phosphorylation sites was also observed (Supplementary Figure S11).
In Vivo Activity: A Combination with Venetoclax
We and others have previously demonstrated that the concurrent inhibition of BCL-2 and MCL-1 results in marked cytotoxic synergy in lymphoma models.42,43 A loss of the antiapoptotic protein MCL-1 observed in the cell line and xenograft lymphoma models treated with 4.35 (Figures 3 and 5) suggests that the depletion may confer the lymphoma cell dependency on other BCL2 family members. Therefore, we investigated the response of a MAVER-1 xenograft to the combination 4.35 and venetoclax (ABT199), an inhibitor of BCL2 family proteins. The individual treatment of both compounds, 4.35 and venetoclax, significantly inhibited tumor growth. Importantly, the inhibitory effect was more pronounced when the mice were treated with a combination of these compounds, and the tumors did not start to expand over the observed time period (Figure 6). The results are inline with recent observations made with the CDK inhibitor dinaciclib, which clearly sensitized MYC-BCL-2 double-hit lymphomas to the BCL-2 inhibitor venetoclax.17,19
Relatively slow progression of pan-CDK inhibitors in clinical trials, which is generally attributed to their complicated mechanisms of action, narrow therapeutic windows and the lack of robust patient selection criteria, collectively emphasize the necessity to develop and explore new compounds with altered kinase profile. Herein, we describe the synthesis and biological activity of new pyrazolo[4,3-d]pyrimidines that potentially inhibited cyclindependent kinase 2 as demonstrated by biochemical assays and crystallographic analysis. Our findings confirmed the in vitro and in vivo sensitivity of aggressive non-Hodgkin lymphomas to CDK inhibitors and provide rationale for their future clinical evaluation. In addition, the combination of CDK inhibitors selected based on biological mechanism (targeting predominantly MCL-1) and the BCL-2 targeting agent venetoclax was proven to exert synergistic effects exhibit synthetic lethalityin vivo and should be considered for combinatorial treatment approaches inpatients.
Experimental Section
NMR spectra were recorded on a JEOL ECA-500 spectrometer operating at the frequencies of 500.16 MHz (1 H) and 125.76 MHz (13C). 1 H NMR and 13C NMR chemical shifts were referenced to the solvent signals; for 1 H, δ(residual CHCl3)=7.25 ppm, δ(residual DMSO d5)=2.50 ppm, δ(residual CD3OD)=3.31 ppm; 13C: δ(CDCl3)=77.23 ppm, δ(DMSO-d6)=39.52 ppm, and δ(CD3OD)=49.15 ppm. Chemical shifts are provided in the δ scale [ppm] and the coupling constants are provided in Hz.
Melting points were determined on a Kofler block and are uncorrected. Reagents were of analytical grade from standard commercial sources or were synthesized according to the referenced procedure. Thin layer chromatography (TLC) was performed using aluminum sheets with silica gel F254 from Merck. Spots were visualized under UV light (254 nm). ESI or APCI mass spectra were determined using a Waters Micromass ZMD mass spectrometer (sample solution in MeOH, direct inlet, coin voltage was in the range of 10–30 V, trace amounts of HCOOH or NH4OH was used to influence ionization). Column chromatography was performed using Merck silica gel Kieselgel 60 (230–400 mesh). The purity of all synthesized compounds was determined by HPLC–PDA (200–500 nm) and gave satisfactory results (> 95 %). Specific optical rotation was measured on a polarimeter polAAr 3001 (wave length: 589.0 nm, tube length: 50 mm, at 23 °C). All compounds gave satisfactory elemental analyses (± 0.4%).
Protein Crystallization, Diffraction Data Collection, and Structure Determination
The protein sample for crystallization was prepared by mixing 12 mg/ml CDK2/cycA2 complex in 40 mM HEPES at pH 7.5, 200 mM NaCl, and 0.02% monothioglycerol with 100 mM 4.35 in DMSO to the final inhibitor concentration of 2 mM, followed by 30 min of incubation on ice and purification by centrifugation (16000 × g, 10 min, 4 °C). The CDK2/cycA2 crystal in complex with the inhibitor 4.35 was obtained at 18 °C using the sitting-drop vapor diffusion technique and condition No. 21 from the Morpheus kit (Molecular Dimensions, USA), containing 10% w/v PEG 20000, 20% v/v PEG MME 550, 0.03 M sodium fluoride, 0.03 M sodium bromide, 0.03 M sodium iodide, and 0.1 M bicine/Trizma base at pH 8.5. The droplet contained 100 nl of the protein sample, 170 nl of the reservoir solution, and 30 nl of a seed stock prepared by crushing crystals prepared in condition No. 93 from the Morpheus kit (Molecular Dimensions, USA) in 30 µl of the precipitant solution, which contained 10% w/v PEG 20000, 20% v/v PEG MME 550, 0.02 M 1,6-hexanediol, 0.02 M 1-butanol, 0.02 M (RS)-1,2-propanediol, 0.02 M 2-propanol, 0.02 M 1,4-butanediol, 0.02 M 1,3-propanediol, and 0.1 M bicine/Trizma base at pH 8.5. The crystal was harvested after 3 weeks and flash-cooled in liquid nitrogen without additional cryoprotection. A complete dataset at 2.15 Å resolution was collected at 100 K at the beamline MX14.1 of BESSY, Berlin, Germany.51 The dataset was processed using the program XDS52 and its graphical interface XDSGUI.53 The structure was determined by molecular replacement with the program Molrep54 using the structure of CDK2/cycA2 available in the PDB under the code 5LMK55 as a search model. Model refinement was performed using the program REFMAC 5.8.015856 from the CCP4 package57 in combination with manual adjustments with Coot software.58 MolProbity server59 was used to evaluate of the final model quality. The data collection and refinement statistics are listed in Supplementary Table 1. All the figures representing structures were created using PyMOL.60 Atomic coordinates and structural factors were deposited in the PDB under the accession code 6GVA.
Lymphoma cell lines were obtained from the German Collection of Microorganisms and Cell Cultures (DSMZ), American Tissue Culture Collection (ATCC) or European Collection of Authenticated Cell Cultures and cultured in media according to providers’ instructions. HBL2 cells were kindly provided by Prof. Martin Dreyling from the University of Munich, Germany. UPF1H and UPF7U are a mantle cell lymphoma cell lines derived by Dr. Pavel Klener (Institute of Pathological Physiology, First Faculty of Medicine, Charles University, Czech Republic). UPF1H was established from the peripheral blood of a patient with treatment-refractory mantle cell lymphoma. Whole exome sequencing confirmed the clonal identity of UPF1H cells to patient lymphoma cells. UPF7U was derived from leukemized blood of a patient with 2nd relapse of mantle cell lymphoma after failure of ibrutinib. Exome sequencing confirmed that UPF7U kept majority of somatic mutations with the primary lymphoma cells, from which it was derived.
Cell Viability Assays
For the viability assays, cells were treated in triplicate with six different doses of each compound for 72 h. After treatments, Calcein AM solution was added for 1 h, and the fluorescence of live cells was measured at 485 nm/538 nm (excitation/emission) using a Fluoroskan Ascent microplate reader (Labsystems). The GI50 value, the drug concentration lethal to 50% of the cells, was calculated from the dose response curves that resulted from the assays.
Chemicals
Dinaciclib, venetoclax and ibrutinib were purchased from MedChemExpress, and CR8 was purchased from Merck.
Immunoblotting
Cell lysates were prepared, and then proteins were separated on SDS-polyacrylamide gels and electroblotted onto nitrocellulose membranes. After blocking, overnight incubation with specific primary antibodies, and incubation with peroxidase-conjugated secondary antibodies, the peroxidase activity was Medicare prescription drug plans detected with SuperSignal West Pico reagents (Thermo Scientific) using a CCD camera LAS-4000 (Fujifilm). The specific antibodies were purchased from Cell Signaling (anti-PARP, clone 46D11; anti-caspase-7; anti-caspase-3, clone 3G2; anti-Mcl-1, clone D35A5; anti-XIAP; anti-Bcl-xl, clone 54H6; anti-Rb, clone 4H1; anti-pRb S780, clone D59B7; anti-Rb S807/811, clone D20B12; anti-cleaved caspase-9 Asp330; anti-Bax, clone D2E11), Merck (anti-Bcl-2; anti-α-tubulin, clone DM1A), Santa Cruz Biotechnology (anti-βactin, clone C4; anti-c-myc, clone 9E10), Bethyl Laboratories, USA (anti-phospho-RNA polymerase II S2 and S5), and Millipore (anti-RNA polymerase II, clone ARNA-3). The antiPCNA (clone PC-10) antibody was generously gifted by Dr. B. Vojtěšek (Masaryk Memorial Cancer Institute, Brno, Czech Republic). All primary antibodies were diluted in PBS containing 5% powdered milk and 0.1% Tween 20. Peroxidase-conjugated rabbit anti-mouse immunoglobulin or porcine anti-rabbit immunoglobulin antisera (Cell Signaling) were used as the secondary antibodies and visualized with ECL reagents (GE-Healthcare Life Sciences).
All aspects of the animal study met the accepted criteria for the care and experimental use of laboratory animals. The protocols were reviewed by the Institutional Animal Care and Use Committee and approved by the Ministry of Education, Youth and Sports of the Czech Republic under number 2844/2017-2. Immunodeficient NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ mice (Jackson Laboratory, referred to as “NOD-SCID-gamma” or “NSG” mice) were maintained in individually ventilated cages. The cells were injected subcutaneously (s.c.) (510 mil. per mouse) into 8-12 week old female NSG mice; numbers of experimental animals are indicated in the legends to Figures 4 and 6. Compound 4.35 dissolved in PBS was administered intravenously (i.v.) via the tail vein. Therapy was initiated when all animals developed palpable tumors. Tumors were measured every day in two perpendicular dimensions with a digital caliper. In vivo experiments were terminated when at least one mouse in any group developed a s.c. tumor ≥ 2 cm in any diameter. For the analysis of pharmacodynamic markers, palpable tumor-bearing mice were injected with different doses of 4.35 (5 and 10 mg/kg) and euthanized after 24 h. Subsequently, tumors were excised, filtered through 45 µm nylon mesh, and evaluated for protein expression by immunoblotting analysis.