Functional diversity of inhibitors tackling the differentiation blockage of MLL-rearranged leukemia

HL-60 cells were obtained from NCI 60-Panel. Jurkat and MV4-11 cells were obtained from ATCC. OCI-AML5, RS4;11, SEM, ML-2, MOLM-13, MOLM14, NOMO-1, OCI-AML2, KOPN-8, EOL-1, and OCI-AML3 cells were obtained from the Deutsche Sammlung von Mikroorganismen und Zellkulturen (DSMZ, Braunschweig, Germany). All used cells were cultured in the appropriate media and conditions.

All inhibitors used in this study were synthesized in-house (Bayer AG). BAY-155 was synthesized according to the methods outlined in patent application WO2017207387A1. Inhibitor concentrations for EPZ-5676, Brequinar, and OTX015 used in this in vitro study are lower as plasma concentrations measured in clinical studies [24, 26, 29]. Plasma concentrations of BAY 1251152 in humans are not yet reported.

Cells were seeded in the optimal growth medium at 4000–5000 cells/well in a 96 MTP and cultured 18–24 h before inhibitor treatment. Upon treatment with the indicated inhibitor, cells were cultured for 24 h, 96 h, and 168 h and effects on proliferation were determined using alamarBlue Cell Viability Reagent (Thermo Fisher Scientific, Waltham, MA, USA).

Four thousand cells per well were seeded 24 h before they were treated with the indicated inhibitor in a 96 MTP. After 4 or 7 days of treatment, cells were washed with PBS and stained with CD11b - APC (BioLegend, San Diego, California, USA) and DAPI (Thermo Fisher Scientific, Waltham, Massachusetts, USA) or AnnexinV – FITC (BioLegend, San Diego, California, USA) and PI solution (Sigma-Aldrich St. Louis, Missouri, USA) using the FACS Canto II (BD Biosciences, Heidelberg, Germany) and data was analyzed with FACSDiva software.

Cells were washed with PBS and fixed overnight at − 20 °C with 70% ethanol. Fixed cells were stained with PI solution (Sigma-Aldrich St. Louis, MO, USA) solution containing RNase A (Qiagen, Hilden, Germany). Fluorescence was measured with FACS Canto II (BD Biosciences, Heidelberg, Germany) flow cytometer and data was analyzed with FACSDiva software.

Approximately 10,000 of cytospin prepared cells were air dried, fixed in 100% methanol for 1 min, stained in 100% in Wright-Giemsa staining solution (Sigma-Aldrich St. Louis, Missouri, USA) for 90 s, washed two times in deionized water, and air dried.

After 7 days of treatment with the indicated inhibitor, cells were washed once with PBS and quantified. Ten thousand viable cells were resuspended in fresh media along with fluorescein-labeled heat-killed Escherichia coli BioParticles (Molecular Probes, Eugene, OR, USA) (100,000 units), incubated at 37 °C for 30 min and stained with CD11b - APC (BioLegend, San Diego, CA, USA ) and DAPI. Phagocytosis capability was measured with FACS Canto II (BD Biosciences, Heidelberg, Germany). Immunofluorescence of cytospin preparations was measured on LSM700 microscope (ZEISS, Oberkochen, Germany) using CD11b (APC), DAPI, and E.coli particles (FITC).

Total RNA was isolated using RNeasy-Plus Mini kit (Qiagen, Hilden, Germany). RNA (1 μg) was reverse transcribed using SuperScript III First-Strand Synthesis SuperMix (Life Technologies, Carlsbad, CA, USA) and obtained cDNA was used for qRT-PCR at the TaqMan 7900HT Fast Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) utilizing TaqMan Fast Advanced Master Mix (Life Technologies). Commercial primers used in this study are listed in Additional file 2: Materials and methods. RNA-seq study: cells were treated for 8 h (P-TEFb—0.05 μM, OTX015—1 μM), 24 h (BAY-155—2 μM, Brequinar—2 μM, DMSO—0.1%) and 96 h (EPZ-5676—3 μM, DMSO—0.1%) prior to RNA extraction using RNeasy-Plus Mini kit (Qiagen). Obtained RNA was used for library preparation (Illumina, San Diego, CA, USA. TruSeq Stranded mRNA Kit) and obtained libraries were sequenced (Illumina, HiSeq2500 HTv4, SR, dual-indexing, 50 cycles).

RNA-seq reads were aligned to hg38 using STAR aligner. Gene expression was quantified using RSEM. Samples with less than 10 million reads aligning to the genome were excluded; protein-coding genes with more than 10 reads in more than three samples were used for the analysis (total samples N = 305; genes N = 15,007). DESeq2 was used to find genes differentially expressed upon treatment by inhibitors in either each cell line or in the group of sensitive cell lines, while controlling for differences between the cell lines. GSEA analysis was run on the pre-ranked list based on logFC in expression for each compound. To remove cell line-specific differences in PCA, average expression in the DMSO sample was subtracted for each corresponding cell line. Top 1000 variable genes were selected based on median absolute deviation. Data is available at GEO (https://​www.​ncbi.​nlm.​nih.​gov/​geo/​) under accession number GSE125437.

Western blot analysis was performed on cell lysates from at least 100,000 cells. Forty micrograms of whole cell protein extract was separated on 4–20% Tris-Glycine gels, transferred to 0.2-μm nitrocellulose membranes, and probed with anti-HEXIM1 (Bethyl, Montgomery, TX, USA) and β-ACTIN (Cell Signaling, Beverly, MA, USA) antibodies.

As a first step to better understand the similarities and differences in the inhibition of selected MLL-fusion-associated therapeutic targets, we tested all selected inhibitors (Table 1) in cell proliferation assays in two MLL-fused (MV4-11, MOLM-13) and one non-fused AML (HL-60) cell line (Fig. 1a). We observed that OTX015, BAY 1251152, and Brequinar show strong anti-proliferation effects in all tested cell lines with IC₅₀s between 30 nM and 140 nM. BAY-155 resulted in comparable strong effects in the MLL-fused cell lines. In contrast, the non-fused HL60 cell line was only affected with the 10 μM treatment. EPZ-5676 inhibited proliferation of the MLL-fused cell lines to 40–50% with no significant effects in HL-60. To further characterize the anti-proliferation effect, we assessed apoptosis induction (Additional file 1: Figure S1) and cell cycle effects (Additional file 1: Figure S2) using flow cytometry. For all tested inhibitors, a significant increase in apoptotic cells was detected at concentrations starting around their respective IC50 values confirming that apoptosis contributes to observed proliferation effects. Furthermore, in cell cycle analysis, BAY-155, OTX015, EPZ-5676, and BAY 1251152 treatment led to a decrease of cells in S and G2/M phase with increasing concentrations. In contrast, Brequinar treatment resulted in a slight S-phase arrest at lower concentrations (Additional file 1: Figure S2). Next, we investigated the ability to overcome the differentiation blockade of the AML cell lines. We performed flow cytometry analysis of CD11b protein expression as a surrogate maker for myeloid differentiation (Fig. 1b). BAY-155, Brequinar, EPZ-5676, or OTX015 treatments increased CD11b protein level in a dose and time-dependent manner in the MLL-fused cell lines. Interestingly, BAY-155 and EPZ-5676 did not increase CD11b level in the non-fused HL-60 cell line, whereas Brequinar, OTX015, and BAY 1251152 treatment did. However, BAY 1251152 only showed induction of CD11b within a limited concentration range close to IC₉₀ after 7 days of treatment, corresponding to the very steep and concentration-dependent decrease in the proliferation rate. To examine differentiation on the morphologic level we performed Wright-Giemsa staining. We detected myeloid differentiation in a fraction of evaluated cells, which was reflected by typically associated morphology changes (decreased nuclei to cytoplasm ratio, indented/kidney shaped nuclei, and less-basophilic, vacuolated cytoplasm) (Fig. 1c). Morphological differentiation correlated with effects on CD11b induction with the exception of BAY 1251152 treatment, which did not show any significant effects on morphology. To further extend our study on morphological changes also to ALL models with or without MLL-fusion, we analyzed KOPN-8 (MLL-ENL) and Jurkat (MLL-WT) cells. Brequinar treatment also resulted in MLL-fusion independent induction of differentiation in ALL cell lines, whereas BAY-155 specifically affected differentiation of the MLL-ENL fused KOPN-8 model (Additional file 1: Figure S3). In summary, all tested inhibitors showed significant anti-proliferative effects on MLL-fused AML cell lines. However, only Brequinar, BAY-155, EPZ-5676, and partially OTX015 showed additional differentiation effects as denoted by CD11b induction and morphological changes. Furthermore, a functional impact of OTX015, Brequinar, and BAY 1251152 was also observed in HL-60 and Jurkat cells, suggesting that the molecular activities of those inhibitors are not restricted to the MLL-fusion pathway.

To further characterize the inhibitors, we performed a comprehensive gene expression analysis. We extended our cell line panel with additional 11 AML/ALL cell lines. To define appropriate treatment conditions for RNA sampling, we characterized all cell lines for proliferation effects induced by inhibitor treatment. Overall, as seen in the previous cellular experiments, BAY 1251152 and OTX015 followed by Brequinar had the strongest and most ubiquitous effects on proliferation, whereas BAY-155 and EPZ-5676 had significant (IC50 < 1 μM) proliferation effects specifically in selected MLL-fused models (Fig. 2a). Interestingly, treatment with BAY 1251152 could significantly inhibit cell proliferation of all tested cell lines already after 24 h of treatment, indicating an essential function of CDK9/PTEFb for cell viability. Based on these results, we defined the individual duration of inhibitor exposure and concentration to conditions without significant proliferation effects as we were especially interested in early and primary effects on gene expression. RNA-seq analysis showed all inhibitors affect expression of a high number of genes (log2FC > 1, FDR < 0.1), with the number depending on the cell line (Fig. 2b). In contradiction to the described functional roles of the MENIN-MLL interaction and DOT1L, BAY-155 and EPZ-5676 treatment resulted in a higher proportion of upregulated than downregulated genes. Moreover, both inhibitors had the strongest impact on gene expression in the MLL-fused models. In contrast, OTX015 and BAY 1251152 treatment led to a higher proportion of downregulated genes. Both inhibitors induced significant changes in all tested cell models irrespective of the MLL-fusion status. Treatment with Brequinar resulted in a more equal distribution of up- and downregulated genes in most cell line, while three cell lines did not respond on gene expression level, which corresponded to the matched proliferation results.

Next, we analyzed the global gene expression effects in the context of (1) the individual inhibitor effect across different cell line models and (2) similarities of the inhibitors to each other (Fig. 2c). By analyzing the individual inhibitor effects across all models (Fig. 2c—black frames) OTX015, BAY 1251152, and Brequinar showed the most pronounced positive correlation across all responding cell line models (average coefficient of log2FC correlation 0.41, 0.26, and 0.3, respectively). This suggests a more universal mode of action independent of the MLL-fusion and underlying genetic background. Comparing the effects of different inhibitors, we found positive correlation between BAY-155–Brequinar and BAY 1251152–OTX015, which was most evident in the same cell line models (average coefficient of log2FC correlation 0.37 and 0.33). In a more detailed analysis of overlaps between only up- or downregulated genes, effects between BAY 1251152 and OTX015 were especially similar for gene downregulation indicating shared general activator functionality of P-TEFb and BRD4 (Additional file 1: Figure S4). As a next step, we evaluated which biological processes can be linked to the different gene expression responses. Therefore, we performed gene set enrichment analysis (GSEA) and principal component analysis (Fig. 3a and c, respectively) to address this question. The GSEA (Fig. 3a) shows that BAY-155, EPZ-5676, and Brequinar affect similar pathways in sensitive cell lines with a significant positive normalized enrichment score (NES) for myeloid and leukocyte differentiation induction. Moreover, those inhibitors significantly regulated gene sets involved in phagocytosis, chemotaxis, and immune response. In contrast, pathways regulated by MYC, MYB, MLL-fusion, HOXA9, or MEIS1 were negatively affected by all three inhibitors. Interestingly, BAY 1251152 and OTX015 negatively regulated gene sets associated with differentiation, phagocytosis, and immune signaling indicating a different mechanistic consequence for both inhibitors. On the other hand, treatment with BAY 1251152 positively regulated gene sets involved in the nonsense-mediated decay pathway and peptide chain elongation, while these gene sets were downregulated by Brequinar. Further, we analyzed several known MLL target genes which are found elevated or repressed in AML patients (Fig. 3b). We observed a strong correlation between BAY-155, EPZ-5676, and Brequinar in regulating MEF2C, ITGAM, CRISPLD2, and CD244. Interestingly, treatment with OTX015 and BAY 1251152 expression did not revert the MLL fusion-driven gene expression pattern. To better understand the similarities and differences between inhibitors effects, we used 1000 most variable genes in a principal component analysis (PCA) across all treated models. To eliminate cell line-specific differences, we centered all data on gene expression in the respective DMSO samples. Three distinct groups of samples can be seen in PC1-PC2 scores plot (Fig. 3c), where cells treated with BAY-155, EPZ-5676, and Brequinar cluster together and OTX015 as well BAY 1251152 separately. In the corresponding loadings plot, we could identify myeloid (Fig. 3d) and lymphoid (Additional file 1: Figure S5) cell surface markers as driving the difference between the samples. For the myeloid-derived cancer cell lines, we identified specific surface markers (e.g., ITGAM, ITGAX, CD68, CD86) usually present on monocytes, neutrophils, and macrophages, positively contributing to the BAY-155, EPZ-5676, and Brequinar group. For the lymphoid-derived cancer cell lines next to the specific surface markers (e.g., CD72, LAIR) associated with T/B-cells, we identified FLT3, HOXA9, MYC, and HEXIM1 as top genes driving the difference between the samples.

Interestingly, we observed HEXIM1 upregulation in all cell lines responding to Brequinar (Additional file 1: Figure S6a). In a previous study, HEXIM1 has been linked with nucleotide starvation, which was shown to sequester P-TEFb activity in melanoma [30]. Therefore, we hypothesized a direct relationship between DHODH inhibition and the elongation complex. As HEXIM1 function was associated with cell differentiation [31], we asked if HEXIM1 influences our inhibitor-induced AML differentiation. Upon HEXIM1 knockout, we observed a significant reduction of CD11b, MNDA and CD68 mRNA, and CD11b protein level after Brequinar treatment (Additional file 1: Figure S6b–d). Interestingly, induction of MNDA, LYZ, and CD68 gene expression after OTX015 treatment were also significantly reduced. This confirms the role of HEXIM1 in differentiation effects mediated by BET or DHODH inhibition. In summary, treatment with OTX015 and Brequinar showed the most pronounced and universal effects over all tested/responding cell lines. BAY-155 was in average more active in MLL-fused models. GSEA and PCA analysis of early global gene expression effects confirmed differentiation induced by treatment with of BAY-155, Brequinar, and EPZ-5676.

Short-term treatment with BAY-155, EPZ-5676, and Brequinar was sufficient to induce expression of genes associated with differentiation. This led us to hypothesize that long-term treatment could differentiate to a more terminal stage thereby recovering normal cell function. Thus, we analyzed a number of cell surface markers and other genes linked with myeloid differentiation on the gene expression level after a prolonged exposure of 7 days of treatment (Fig. 4a). We observed that all tested inhibitors decreased the expression of markers associated with multipotent progenitors and Granulocyte Monocyte precursors (CD117, FLT3, and CD123), with BAY-155 and EPZ-5676 treatment having the strongest effect. Furthermore, both inhibitors showed upregulation of Monocyte CD11b and CD14 markers and moderate to strong upregulation of macrophage-associated marker genes. Similar effects on the differentiation marker genes were detected after Brequinar treatment. Surprisingly, also OTX015 showed after prolonged exposure significant, albeit weaker, induction of those marker genes.

In contrast to that, BAY-155 and EPZ-5676 treatment in HL60 (MLL-WT) (Additional file 1: Figure S7) did not modulate differentiation-associated marker genes. In HL60, Brequinar and OTX015 showed significant upregulation of some markers (e.g., CD11b, LYZ). BAY 1251152 treatment resulted in downregulation of majority of tested genes in MOLM-13 and HL60. Next, we were interested if the observed differentiation effects would translate into regaining functional properties of myeloid differentiated cells. For this purpose, we tested the capabilities of MOLM-13 cells to phagocyte E.coli particles. As shown in Fig. 4b, Brequinar treatment increased CD11b and phagocytosis level most effectively with 30% of CD11b positive cells showing particle uptake. Increased phagocytosis activity combined with CD11b induction were observed to a lesser extent for BAY-155 and EPZ-5676. OTX015 induced CD11b and phagocytosis activity only slightly. Altogether, we observed that prolonged treatment with Brequinar, BAY-155, and EPZ-5676 induces a number of differentiation-associated markers and a partial regain of cellular functionality in vitro.

Since all inhibitors used in this study potentially interfere at different stages with MLL-fusion proteins, they could potentially be combined to achieve superior effects. Therefore, we tested all possible combinations (10 combinations per cell line model) on cell proliferation and differentiation (Fig. 5 and Additional file 1: Figure S8 and S9) by using inhibitor-inhibitor concentration matrixes combined with IC₅₀ evaluation. We observed clear anti-proliferative synergism for BAY-155 in combination with Brequinar (combination index, 0.27–0.64) and EPZ-5676 (combination index, 0.21–0.51) as well for Brequinar combined with EPZ-5676 (combination index, 0.32–0.97) (Fig. 5a). All three combinations resulted in significant differentiation synergisms (Fig. 5b). Interestingly, Brequinar used in combination with OTX015 showed clear anti-proliferative synergism (combination index, 0.28–0.71) with antagonistic differentiation effects (Fig. 5a, b). All other tested combinations resulted in anti-proliferative synergism or additivity but no differentiation synergisms effects (Additional file 1: Figures S8 and S9). In summary, we found synergistic effects on the differentiation level when BAY-155, Brequinar, and EPZ-5676 were combined.