Skip to main content
NCBI home page
HemaSphere logoLink to HemaSphere
. 2026 Mar 24;10(3):e70343. doi: 10.1002/hem3.70343

Casein kinase 1δ/ε inhibition suppresses CLL proliferation through cell‐intrinsic and microenvironmental mechanisms

Antonia Mikulova 1, Hana Plesingerova 1,2, Michaela Chorvatova 1,3, Pavlina Kebkova 1, Jana Bartosikova 1, Petra Prochazkova 1, Jan Verner 4,5, Dominik Mulidran 1, Stepan Cada 1, Olga Vondalova‐Blanarova 1, Nela Kolcakova 1, Tomas Loja 4, Terezia Kurucova 4, Eva Hoferkova 2,4, Boris Tichy 4, Jana Kotaskova 2,4, Marek Mraz 2,4, Lukas Kubala 3, Pavlina Janovska 1,6,✉, Vitezslav Bryja 1,✉
PMCID: PMC13045430  PMID: 41939248

Abstract

Chronic lymphocytic leukemia (CLL) is a consequence of pathological B‐cell accumulation in blood and lymphoid organs. Due to high refractoriness, CLL is still incurable in many cases; therefore, there is an urgent need to develop novel therapeutic options. We have shown earlier that inhibition of casein kinase 1δ/ε (CK1δ/ε) is a promising CLL treatment strategy. Herein, we elucidate the molecular and cellular mechanisms mediating CK1δ/ε inhibition efficacy in CLL. Using an in vivo Eµ‐TCL1 adoptive transfer model, we showed that CK1δ/ε inhibition caused CLL cell accumulation at the S/G2 phase in a cell‐intrinsic mode. Furthermore, CK1δ/ε inhibition led to a T‐cell decrease in lymph nodes (LNs). Using primary CLL cells and a system mimicking the LN microenvironment in vitro, we demonstrated that CK1δ/ε inhibition interfered with multiple pro‐survival mechanisms provided by the microenvironment, most notably with the nuclear factor κ B (NFκB) pathway. NFκB acts downstream of the T‐cell‐mediated CD40L:CD40 stimulus, and indeed, CK1δ/ε inhibition efficiently blocked the proliferation of primary CLL triggered by CD40L across multiple patient groups, with lower efficacy in patients with TP53 defects. We propose that CK1δ/ε inhibitors act in the multiple‐hit mode, striking both intrinsically via direct interference with cell cycle machinery and extrinsically via inhibition of multiple pro‐proliferative stimuli.

graphic file with name HEM3-10-e70343-g002.jpg

INTRODUCTION

Chronic lymphocytic leukemia (CLL) is a lymphoproliferative disease characterized by elevated counts of monoclonal B cells (above 5 × 109 cells/l) that are dysfunctional but active in proliferation and evasion into the lymphoid organs. CLL is one of the major health issues prevalent in the adult population, 1 with a global incidence 4.6 per 100,000 people per year. 2 Due to a frequent occurrence of mutations and chromosomal aberrations (e.g., ATM at 11q22 [18%], TP53 at 17p13 [7%], trisomy 12 [16%], and the deletion of 13q [55%]), which are present in more than 80% of patients, CLL represents a heterogeneous disease that often requires personalized therapy. Current front‐line agents include anti‐CD20 antibody, rituximab, 3 inhibitor of the anti‐apoptotic BCL‐2, venetoclax, 4 phosphoinositide 3‐kinase (PI3K) inhibitor, idelalisib, 5 and Bruton tyrosine kinase (BTK) inhibitors, ibrutinib, acalabrutinib, zanubrutinib, and pirtobrutinib.6, 7, 8 Despite the overall high efficacy of the mentioned therapeutic agents, a high percentage of patients develops relapse. Especially, the increasing occurrence of cases with BTK‐ and BCL2‐resistant (termed double‐refractory) disease has become a real challenge in clinics. 2 All this evidence renders CLL an incurable condition to date and argues for the development of novel therapeutic strategies.

We have shown earlier that casein kinase 1 inhibition (CK1i) can efficiently block CLL progression in a mouse model of CLL. 9 More specifically, CK1δ/ε‐specific inhibitor, PF‐670462, could block CLL cell motility and chemotaxis. 10 CK1δ/ε kinases are the key components of non‐canonical Wnt signaling, 11 which controls cell migration. 12 Since the non‐canonical Wnt receptor ROR‐1 is upregulated in CLL 13 and the pathway ligand Wnt5a is a marker of aggressive CLL, 14 we have proposed that CK1δ/ε inhibition acts in CLL by blocking the non‐canonical Wnt pathway and cell motility. 10 However, besides the Wnt pathway, CK1δ/ε kinases regulate multiple other cellular processes, e.g., cytoskeleton dynamics, cell–cell communication, cell division, 15 p53‐dependent apoptosis, 16 or the circadian rhythm. 17

Many of these CK1‐dependent processes—especially regulation of cell proliferation and cellular communication—can potentially promote tumorigenesis. All this evidence from the literature leads to the question of whether the CK1 inhibitor mechanism of action in CLL may be broader and involve other effects on both malignant CLL cells and their microenvironment,18, 19, 20, 21 especially T cells. 22 It has been shown that the CLL clone actively shapes its microenvironment—mainly by recruitment of CD4+ T‐cell subsets that provide CLL cells with essential stimuli. 23

To study the effects of CK1δ/ε inhibition on CLL cells and their microenvironment, we have used a well‐defined fully immunocompetent Eµ‐TCL1 mouse model of CLL. In this model, the T‐cell leukemia 1 (TCL1) gene, placed under the control of a B‐cell specific VH promoter‐IGH‐Eµ enhancer, drives spontaneous development of CLL within 6–10 months. 24 This mouse model replicates multiple CLL‐related symptoms, including enlarged spleen (SPL), advanced lymphadenopathy, or increased cellularity in the peritoneal cavity of the mice. 25 The Eµ‐TCL1 mice possess primarily unmutated IGHV genes and the TP53 R172H mutation (equivalent to the human TP53 R175H mutation), which make them a suitable model organism to study the aggressive form of CLL. 26 Also, the Eµ‐TCL1 model shows multiple alterations in the T‐cell pool 27 and acquisition of dysfunctional T‐cell subsets that were observed in CLL patients. 28

Molecular and functional analyses uncovered a significant effect of CK1δ/ε inhibition on the cell cycle progression in CLL cells in vivo, an effect that could be fully replicated in the in vitro models, including primary patient CLL cells. Further, we were able to identify multiple effects of CK1δ/ε inhibition on the CLL microenvironment in lymph nodes (LNs) and the bone marrow (BM), which we could replicate in the in vitro co‐culture setup with primary patient CLL cells. Altogether, our analysis uncovers novel mechanisms that are responsible for the therapeutic efficacy of the CK1δ/ε inhibitor in preclinical models of CLL.

MATERIALS AND METHODS

Animals

All animal experiments were performed with the approval of the Ethics Committee in accordance with the international ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines. The Eµ‐TCL1 mouse model, which is further described in the Supporting Information S1: Supplementary Methods was used as a donor group for adoptive transfer (AT). C57BL/6 J mice (Charles River Laboratories) were used for AT as the recipient group, as well as for wild‐type (WT) mouse treatment experiments. The recipient and WT mice were treated either with 10% Kolliphor of ethoxylated castor oil‐purified grade diluted in phosphate‐buffered saline, further addressed to as vehicle, or PF‐670462 (DC Chemicals, #DC2086) dissolved in the vehicle. Blood was drawn from mice, and they were euthanized after the group with vehicle reached the terminal stage of CLL. Cells from the peripheral blood (PB), LN, SPL, and BM were isolated and characterized using flow cytometry. Splenic CLL cells were selected by fluorescence‐activated cell sorting and subjected to bulk RNA sequencing (RNAseq). The cell cycle phase for the splenocytes was determined using the EdU Click‐iT assay. For detailed information on AT, treatment, and subsequent methods, see the Supporting Information S1: Supplementary Methods.

CLL cell lines MEC‐1 and HG‐3 and CK1 inhibitors

CLL cell lines MEC‐1 WT (DSMZ, #ACC497) and HG‐3 WT (DSMZ, #ACC765) were treated with PF‐670462 (DC Chemicals, #DC2086), an in‐house CK1δ/ε inhibitor MU1742 29 or CK1δ/ε degrader AH078, 30 and subjected to cell proliferation tracking and cell cycle tracking via PI staining, EdU Click‐iT assays, and/or western blotting, as described in more detail in the Supporting Information S1: Supplementary Methods.

CLL patients

Samples from the PB were collected after obtaining informed consent and approval from the Ethical Committee of the University Hospital Brno in accordance with the Declaration of Helsinki. Primary CLL samples were collected from 53 CLL patients (see Supporting Information S2: Table 1 for their clinical characteristics). B cells were isolated from PB by gradient centrifugation on Ficoll‐Paque PLUS (GE Healthcare, #45‐001‐750) coupled with non‐B‐cell‐depletion (RosetteSep CD3+ Cell Depletion Cocktail, #15661, RosetteSep B Cell Enrichment Cocktail, #15064, StemCell Technologies). Separation efficiency was assessed by flow cytometry by CD5, CD19, and CD45 staining, and only samples with at least 98% B cells were processed further.

CLL cells were subjected to co‐culture with CD40L‐expressing HS5 stromal cells 19 or with a BM stromal cell line M2‐10B4 (ATCC, #CRL‐1972). CLL cells co‐cultured with CD40L‐expressing HS5 cells were simultaneously treated with CK1δ/ε inhibitors PF‐670462, MU1742 or CK1δ/ε degrader AH078, and tracked for proliferation. CLL cells co‐cultured with M2‐10B4 cells were treated with PF‐670462 and further processed for RNAseq or flow cytometric detection of nuclear factor κB (NFκB) activity. For details on CLL cells' characterization, co‐culture, treatment, and subsequent analysis, see Supporting Information S1: Supplementary Methods.

Statistics

Statistical tests were performed in R Studio or GraphPad. The standard level of statistical significance was P < 0.05. For details on the R packages and tests used, see the Supporting Information S1: Supplementary Methods.

RESULTS

Expression profile of CLL cells from CK1δ/ε‐inhibited mice suggests cell cycle arrest at the G2 phase onset

To address the mechanism of CLL suppression by CK1δ/ε inhibition, 9 we conducted a series of in vivo Eµ‐TCL1 AT experiments, as schematized in Figure 1A. Engraftment was monitored by CLL cell detection in PB of AT recipients using flow cytometry (Supporting Information S1: Figure 1A). 32 CK1i with PF‐670462 significantly counteracted CLL development in PB (Figure 1B, Supporting Information S1: Figure 1B), SPL (Figure 1C, Supporting Information S1: Figure 1C), and BM (Figure 1D, Supporting Information S1: Figure 1D), which is in line with previously reported observations. 9 To identify the molecular changes caused by CK1i, we sorted CLL cells from SPL of AT recipients (Figure 1A) and subjected them to RNAseq. Differential expression (DE) using limma identified 46 significantly upregulated genes and 22 significantly downregulated genes (|log2FC| > 1 and P‐value < 0.05; FC—fold change) after CK1i (Figure 1E, Supporting Information S2: Table 2). For the downregulated genes, little or no connection to CLL progression was found in the literature. Interestingly, the significantly upregulated transcripts (Figure 1E) have been linked to CLL in some cases. For example, Traf3, which blocks CD40L:CD40 signals provided by T cells via inhibition of the NFκB pathway, was linked to CLL patient survival. 33 This indicates that CK1δ/ε inhibition, among others, interferes with the CLL microenvironment communication.

Figure 1.

Figure 1

Open in a new tab

Casein kinase 1δ/ε (CK1δ/ε) inhibition in the Eµ‐TCL1 adoptive transfer (AT) model. (A) Experimental scheme of the AT setup, based on a peritoneal injection of splenic mononuclear cells (comprising ~90% leukemic B cells) from an Eµ‐TCL1 mouse in the terminal stage of chronic lymphocytic leukemia (CLL) to a cohort of female wild‐type (WT) C57BL/6J recipients and subsequent treatment with 50 mg/kg PF‐670462 (AT + PF‐670462) or 10% Kolliphor of ethoxylated castor oil‐purified grade (ELP) dissolved in phosphate‐buffered saline (PBS) further termed vehicle (AT CTRL) for 5–8 weeks, depending on the time needed for the control group to reach the terminal stage of the disease. (B) Counts/µL of CD19+ CD5+ CD44+ leukemic B cells in peripheral blood (PB) of all mice in the in vivo AT experiments (N (WT) = 8; N (AT CTRL) = 14; and N (AT + PF‐670462) = 8; CLL cells from 6 TCL1 donors in total) at the endpoint (tested by the Kruskal–Wallis test) with post hoc pairwise Wilcoxon rank sum tests with Benjamini–Hochberg correction. (C, D) Percentage of CD19+ CD5+ CD44+ leukemic B cells out of all CD45+ SSC‐A(low) lymphocytes in the spleen (SPL) (C) and the bone marrow (BM) (D) of all mice in the in vivo experiments (N (WT) = 31; N (PF‐670462) = 6; N (AT CTRL) = 13; and N (AT + PF‐670462) = 7) at the endpoint (in both cases tested by the Kruskal–Wallis test and post hoc pairwise Wilcoxon rank sum tests with Benjamini–Hochberg correction). (E) Volcano plot with differentially expressed genes in splenic leukemic B cells from AT recipients (N (AT CTRL) = 3; N (AT + PF‐670462) = 3): top upregulated genes in blue (out of total 46) characterized by threshold log2FC > 1 and P‐value < 0.05 and top downregulated genes labeled in red (out of total 22) characterized by threshold log2FC < −1 and P‐value < 0.05. (F, G) Barplot with the top 10 Gene Ontology (GO) terms identified in gene enrichment of the upregulated genes identified by the limma voom algorithm (log2FC > 1 and P‐value < 0.05) in the section “biological process” and “cellular component,” respectively (N (AT CTRL) = 3; N (AT + PF‐670462) = 3). (H) Scatterpolar plot with the top 20 differential expression (DE) genes (P‐value < 0.05) mapped to specific phases of cell cycle based on the dataset describing the time lapse of gene expression throughout the cell cycle 31 ; theta represents the specific time point of the cell cycle when the gene is being expressed, whereas r represents the level of association of the certain gene to the time point. Color represents whether the gene has been found to be up‐ or downregulated upon CK1δ/ε inhibition, based on |log2FC| > 1 and P‐value < 0.05 (N (AT CTRL) = 3; N (AT + PF‐670462) = 3). (I) Barplot with scores displaying the activity of transcription factors (TFs) based on DE analysis (N (AT CTRL) = 3; N (AT + PF‐670462) = 3). (J, K) Seurat cell cycle scoring output shown for the S phase and the G2/M phase, respectively (N (AT CTRL) = 3 [in total, 4.5 M cells analyzed]; N (AT + PF‐670462) = 3 [in total, 3.56 M cells analyzed]), in both cases tested by the t‐test.

Unbiased analysis of the upregulated genes using Gene Ontology (GO) and KEGG pathways showed a clear link to the cell cycle (Figure 1F,G, Supporting Information S1: Figure 1E), with the top 10 GO terms being connected to this biological process (Figure 1F, Supporting Information S1: Figure 1F). Moreover, the top GO terms in the section “cellular component” indicated a connection to the late cell cycle (specifically, G2/M phase), due to top‐scoring terms like “microtubule cytoskeleton,” “microtubule,” “spindle,” “centrosome,” “microtubule‐organizing centre/MTOC,” and so forth (Figure 1G, Supporting Information S1: Figure 1G). Therefore, we decided to map our top hits to a publicly available dataset describing the gene expression time lapse during the cell cycle. 31 Clearly, most upregulated genes were cell cycle‐regulated and specific for the late S and G2 phases (Figure 1H). Using the decoupleR package, 34 we were able to link upregulated genes with a remarkable activation of upstream transcription factors (TFs) E2f4, Foxm1, Tfdp1, E2f1, E2f2, and E2f3 (Figure 1I), which all have cell cycle‐regulated activity. TFs of the E2f family are key cell cycle regulators that mediate transcription of cell cycle‐associated genes during the S phase, 35 Foxm1 36 is regulated by E2f3, 37 and Tfdp1 is an important binding partner of E2f1. 38 Finally, in silico cell cycle phase classification using Seurat showed notable enrichment of CLL cells in the S and G2/M phases upon CK1δ/ε inhibition (Figure 1J,K). Though the Seurat cell cycle scoring data did not show a statistically significant effect of treatment with PF‐670462, in combination with other bioinformatic approaches, the transcriptomic data of murine CLL cells suggest that the dominant effect of CK1i in vivo is a cell cycle slowdown or arrest, which results in the accumulation of CLL cells in the late S and G2 phases.

CK1δ/ε inhibition blocks proliferation in the Eµ‐TCL1 AT model and CLL cell lines MEC‐1 and HG‐3

To prove the accumulation of CLL cells in the S phase directly, we used the EdU Click‐iT system for DNA synthesis labeling in vivo (at the endpoint of the standard AT experiment) and subsequent analysis by flow cytometry (Supporting Information S1: Figure 2A,B). In line with RNAseq data, we found that CK1i led to a significant enrichment of EdU+, that is, S phase cells (Figure 2A) among the splenic CLL cells. At first glance, this result is in contrast with the decrease in CLL cells in treated mice (Figure 1B–D) but in fact can be easily explained by a cell cycle slowdown at the S/G2 interface.

Figure 2.

Figure 2

Open in a new tab

Validation of the cell cycle and proliferation effects of casein kinase 1δ/ε (CK1δ/ε) inhibition in vivo and in vitro. (A) Percentages of EdU‐Alexa Fluor 647+ leukemic B cells within the spleen (SPL) of treated and control TCL1 adoptive transfer (AT) recipient mice (N (AT CTRL) = 3; N (AT + PF‐670462) = 4), tested by the t‐test. (B) Relative cell counts (% of CTRL) originating from in vitro treated MEC‐1 wild‐type (WT) cells after a 72 h treatment with PF‐670462 or MU1742 (performed on the following biological replicates: N (CTRL) = 6; N (3µM PF‐670462) = 3; N (10µM PF‐670462) = 6; N (3µM MU1742) = 3; and N (10µM MU1742) = 3), tested by the Kruskal–Wallis test with post hoc pairwise Wilcoxon rank sum tests with Benjamini–Hochberg correction. (C) Relative cell counts (% of CTRL) originating from in vitro treated HG‐3 WT cells after 72 h treatment with PF‐670462 or MU1742 (performed on the following biological replicates: N (CTRL) = 4; N (3µM PF‐670462) = 4; N (10µM PF‐670462) = 4; N (3µM MU1742) = 4; and N (10µM MU1742) = 4), tested by the Kruskal–Wallis test with post hoc pairwise Wilcoxon rank sum tests with Benjamini–Hochberg correction. (D) Cell cycle assay setup with initial CK1 inhibitor treatment and mitotic arrest with nocodazole and the representative example of cell cycle alterations between analyzed conditions in MEC‐1 and HG‐3 cell lines. (E) Cell cycle phase distribution in MEC‐1 WT cells upon 9 h pre‐treatment with 3 µM PF‐670462 and 10 µM PF‐670462 and subsequent mitotic arrest with nocodazole (performed on the following biological replicates: N (CTRL) = 11; N (NOCODAZOLE) = 11; N (3µM PF‐670462) = 7; and N (10µM PF‐670462) = 11); for all cases together, the generalized linear mixed‐effects model, followed by estimated marginal means calculation (P‐value < 0.05), was used separately for comparison of CTRL versus NOCODAZOLE and NOCODAZOLE versus PF‐670462 and corrected due to usage of 2 models. (F, G) Cell cycle phase distribution in MEC‐1 WT cells upon 9 h pre‐treatment with 3 and 10 µM concentrations of MU1742 and AH078 (respectively) and subsequent mitotic arrest with nocodazole (performed on the following biological replicates: N (CTRL) = 11; N (NOCODAZOLE) = 11; N (3µM) = 3; and N (10µM) = 3); for all cases together, the generalized linear mixed‐effects model followed by estimated marginal means calculation (P‐value < 0.05), was used separately for comparison of CTRL versus NOCODAZOLE and NOCODAZOLE versus MU1742/AH078 and corrected due to usage of 2 models. (H–J) Cell cycle phase distribution in HG‐3 WT cells upon 9 h pre‐treatment with 3 and 10 µM concentrations of PF‐670462, MU1742, and AH078 (respectively) and subsequent mitotic arrest with nocodazole (performed on the following biological replicates: N (CTRL) = 4; N (NOCODAZOLE) = 4; N (3µM) = 4; and N (10µM) = 4); for all cases together, the generalized linear mixed‐effects model, followed by estimated marginal means calculation (P‐value < 0.05), was used separately for comparison of CTRL versus NOCODAZOLE and NOCODAZOLE versus PF‐670462/MU1742/AH078 and corrected due to usage of 2 models. DMSO, dimethyl sulfoxide; PI, propidium iodide.

To confirm this possibility directly, we subsequently performed a series of in vitro experiments on CLL cell lines, MEC‐1 and HG‐3. Firstly, we observed that within 72 h, CK1δ/ε inhibition was able to dramatically reduce the proliferation of both cell lines, which was visible with both PF‐670462 and chemically different and even more selective CK1δ/ε inhibitor, MU1742 29 (Figure 2B,C, Supporting Information S2: Table 3). To analyze the cell cycle dynamics, we triggered mitotic arrest using nocodazole (Figure 2D), which synchronizes cells in mitosis. In this assay, CK1i led to a dose‐dependent slowdown of cell cycle pace in MEC‐1 cells that was visible as inefficient accumulation at the G2/M boundary accompanied by increased number of cells in the S phase (Figure 2E, Supporting Information S1: Figure 2C, Supporting Information S2: Table 4). This effect has also been observed using the more selective CK1δ/ε inhibitor, MU1742 (Figure 2F), and a highly selective CK1δ/ε degrader, AH078 (Figure 2G). Furthermore, the significant increase of cells in the S phase with all three mentioned compounds has also been observed in another CLL cell line, HG‐3 (Figure 2H–J, Supporting Information S1: Figure 2D, Supporting Information S2: Table 4). Western blots in Supporting Information S1: Figure 2E,F show the efficient CK1δ/ε inhibition or degradation in MEC‐1 and HG‐3 cells at the used compound doses. An increase in CK1δ/ε protein levels reflects their inhibition, which is also visible as an increase in total p53 39 in TP53 WT HG‐3 cells or as a decrease in the phosphorylation of VANGL2 40 in MEC‐1 cells that carry mutated TP53. The treatment with any of the inhibitors or degrader did not affect cell viability (Supporting Information S1: Figure 2G). Given the fact that both inhibitors are very selective and chemically distinct, 29 and the degrader is highly specific and profiled against the whole proteome, 30 we conclude that indeed, CK1i blocks cell proliferation via intrinsic cell cycle inhibition without inducing cell death.

CK1δ/ε inhibition restores CLL‐associated skewing of T‐cell subsets

CLL cells are largely dependent on their microenvironment,18, 19, 20, 21 the important components of which are T cells. 27 , 41 Individual T‐cell subsets differ in their role in CLL, as some provide supportive signals and some confer suppressive signals on CLL cells. Currently, it is believed that follicular helper (Tfh) cells, via CD40L, lymphocyte function‐associated antigen 1 (LFA‐1), and production of interleukin (IL)‐21/IL‐21 42 —and T helper 2 (Th2) cells (via IL‐4 and IL‐13 production), 43 promote CLL progression. On the contrary, CD8+ T cells have been linked to maintenance of anti‐tumor immunity and elimination of CLL burden. 44 Importantly, T‐cell pool composition is skewed by the presence of the CLL clone itself, with naive to effector‐memory (EM) T‐cell shift, skewing of the CD4+/CD8+ T‐cell ratio in favor of the pro‐tumorigenic CD4+ T cells, 27 and acquisition of exhausted (PD‐1+) and senescent (KLRG‐1+) phenotypes in the chronically stimulated CD8+ T cells 28 being the main CLL‐associated phenomena.

To address how CK1i alters T cells in healthy and CLL‐bearing organisms, we performed standard AT experiments (Figure 1A) along with experiments on WT non‐transplanted C57BL/6 J mice that were all followed by detection of selected immune cell types and multiparametric data analysis using FlowSOM clustering (Figure 3A, Supporting Information S1: Figure 3A). As the use of automated clustering pipelines is still a relatively novel approach in flow cytometric data analysis, we compared results from FlowSOM to manual gating in FlowJo in all analyzed tissue types (a representative example is shown in Supporting Information S1: Figure 3B). As outputs of both approaches led to highly similar results, we further present only outputs from FlowSOM.

Figure 3.

Figure 3

Open in a new tab

Overview of casein kinase 1δ/ε (CK1δ/ε) inhibition effects on immune cells in the chronic lymphocytic leukemia (CLL) microenvironment. (A) Experimental design of flow cytometric detection of selected immune cell types from peripheral blood (PB), lymph node (LN), spleen (SPL), and bone marrow (BM) of both wild‐type (WT) mice and adoptive transfer (AT) recipients and subsequent cluster analysis in FlowSOM; data were first preprocessed as described in Supporting Information S1: Figure 3A, and upon dimensional reduction and clustering, the cluster of CLL cells was always excluded from the analysis; subsequently, the main immune cell types (CD4+ and CD8+ T cells, healthy B cells, and NK cells) were quantified, which was followed by creation of subsets of (a) CD4+ T cells (with calculation of relative abundance for central memory [CM], effector‐memory [EM], naive CD4+ T cells, and Treg) and (b) CD8+ T cells (with calculation of relative abundance for CM, EM, and naive CD8+ T‐cell subsets). Changes in the abundance of these cell types have been compared (i) between healthy (WT CTRL) and diseased AT Eµ‐TCL1 mice and between CK1i‐treated and untreated (ii) WT CTRL, and (iii) leukemic Eµ‐TCL1 (AT CTRL) mice. (B) Summary heatmap of the significant changes in the measured immune cell types, compared between healthy and CLL‐bearing mice (N (WT) = 31; N (AT CTRL) = 13), tested based on the output of the Shapiro–Wilk normality test and Levene's test either with analysis of variance (ANOVA) and post hoc Tukey honest significant differences (P‐value < 0.05) in case the data followed a normal distribution, or by the Kruskal–Wallis test and post hoc pairwise Wilcoxon rank sum tests with Benjamini–Hochberg correction if the data deflected from a normal distribution (detailed information for each statistical comparison is available on GitHub). (C) Summary heatmap of the significant changes in the measured immune cell types, compared between the control group of AT recipients and the treated group of AT recipients (N (AT CTRL) = 13; N (AT + PF‐670462) = 7), tested based on the output of the Shapiro–Wilk normality test and Levene's test either by ANOVA and post hoc Tukey honest significant differences in case the data followed a normal distribution, or by the Kruskal–Wallis test and post hoc pairwise Wilcoxon rank sum tests with Benjamini–Hochberg correction if the data deflected from a normal distribution (detailed information for each statistical comparison is available on GitHub). (D, E) Barplot with percentages of Treg located in the LN and SPL between all conditions and in vivo experiments, respectively (N (WT) = 31; N (PF‐670462) = 6; N (AT CTRL) = 13; and N (AT + PF‐670462) = 7), in both cases tested by the Kruskal–Wallis test and post hoc pairwise Wilcoxon rank sum tests with Benjamini–Hochberg correction. (F, G) Barplots with percentages of CD4+ and CD8+ T cells located in the LN between all conditions and in vivo experiments, respectively (N (WT) = 31; N (PF‐670462) = 6; N (AT CTRL) = 13; and N (AT + PF‐670462) = 7), in both cases tested by the Kruskal–Wallis test and post hoc pairwise Wilcoxon rank sum tests with Benjamini–Hochberg correction.

The summary of changes in lymphocyte populations in PB, LN, SPL, and BM caused by the disease (healthy vs. CLL‐bearing mice) is shown in Figure 3B; the effects of CK1i in diseased mice are shown in Figure 3C (control vs. CK1i‐treated group). As expected, CLL progression induced a large spectrum of alterations in T‐cell subsets (Figure 3B), including an increase of CD4+ and CD8+ T cells in LN (Supporting Information S1: Figure 3C,D), a decrease of naive CD4+ T cells in BM (Supporting Information S1: Figure 3E), and a significant shift from naive and central memory (CM) CD8+ T cells (Supporting Information S1: Figure 3F,G) to EM CD8+ T cells in BM (Supporting Information S1: Figure 3H). On the contrary, CK1i exerted only a few specific effects on B‐ and T‐cell subsets that would occur both in healthy mice and in treated CLL‐bearing mice (all effects are summarized in Figure 3C–G). In the context of CLL development, we can conclude that CK1i‐associated effects were mostly balancing out the CLL‐associated changes due to disease deceleration (Figure 3B,C). However, we observed significant CK1i‐associated alterations, including the Treg increase in LN (Figure 3D) and SPL (Figure 3E) that has been already described in the literature specifically for inhibition of CK1ε. 45 Apart from the effect on Treg, CK1i triggered the biggest changes in the LN (Figure 3C), where it mediated a significant decrease of CD4+ and CD8+ T cells (Figure 3F,G). Herein, we noted that CK1i significantly counteracted the majority of observed CLL‐associated alterations in the T‐cell subset composition that were previously linked to chronic stimulation of the immune system by the CLL clone, including increase of PB Treg, decrease of PB CM CD4+ and CD8+ T cells, increase of CD4+ and CD8+ T cells in LN, decrease of CM CD4+ and CD8+ T cells in SPL, decrease of naive CD4+ T cells in BM, and shift of naive and CM CD8+ T cells to EM CD8+ T cells in BM (Figure 3B). As most of the CK1i‐associated alterations took place in the LN, we then focused on how CK1i affects the CLL cell response to stimuli that originate from the cells in the LN microenvironment.

CK1δ/ε inhibition blocks microenvironment‐induced activation of the NFκB pathway in primary CLL cells

To mimic the interaction between the LN microenvironment and primary CLL cells in vitro, we used a previously described co‐culture model based on mouse BM cell line, M2‐10B4, as described in Figure 4A. 47 We cultured primary CLL cells from 9 patients (characterized in Supporting Information S2: Table 1) in 9 independent experiments in the presence/absence of M2‐10B4 and/or PF‐670462 and subjected them to bulk RNAseq. Subsequent DE analysis revealed robust changes induced by co‐culture (Figure 4B, Supporting Information S2: Table 5) that allowed us to validate the experimental system using linear discriminant analysis (LDA, Figure 4C). We trained an LDA model on the known differences among primary human CLL cells originating from the PB, BM, and LN, 46 where the model assigned correct identity with 67% accuracy. When applied on CLL cells cultured without M2‐10B4, the model assigned “PB CLL” identity to all mono‐cultured samples and “LN CLL” identity to all co‐cultured samples. To our knowledge, this is the first experimental validation of the fact that this co‐culture setup represents a reliable system mimicking the LN microenvironment.

Figure 4.

Figure 4

Open in a new tab

Casein kinase 1δ/ε (CK1δ/ε) inhibition attenuated primary chronic lymphocytic leukemia (CLL) cell responsiveness to stromal cells. (A) Scheme of a dual species co‐culture experiment. Human CLL patient (N = 9) cells were or were not co‐cultured for 6 h with bone marrow (BM) murine cell line M2‐10B4. Cell lysates were subjected to RNA sequencing (RNAseq) and subsequently, human reads were analyzed; all data in this figure are based on this sample set, with the exception of the independent cohort of seven patients in panels (F), (G), and (K). (B) Volcano plot of gene expression detected in CLL cells cultured with and without M2‐10B4 cells. Black dots—downregulated genes (adj. P‐value < 0.05, log2FC < −1, adj, adjusted) and red dots—upregulated genes (adj. P‐value < 0.05, log2FC > −1), and 10 most upregulated and 10 most downregulated genes are labeled, tested by the limma voom algorithm with Benjamini–Hochberg correction. (C) Linear discriminant analysis trained on a previously described dataset of primary CLL samples from BM, lymph node (LN), and peripheral blood (PB) 46 (N BM = 20, N LN = 14, and N PB = 26) and tested on RNAseq data of primary CLL cells cultured with and without M2‐10B4 cells. (D) Oncology‐related pathways (PROGENy) scored according to differential expression (DE) analysis of CLL cells cultured with and without M2‐10B4 cells. (E) NFKBIA expression in CLL cells cultured with and without M2‐10B4 cells according to RNAseq, tested by the limma voom algorithm with Benjamini–Hochberg correction. (F, G) Percentage of live (F) and IκBα + CLL (G) cells cultured with and without M2‐10B4 cells according to flow cytometry (FCM), tested by Friedman's test coupled with Dunn's post hoc test. (H) Volcano plot of gene expression detected in CLL cells co‐cultured with M2‐10B4 cells with and without PF‐670462 treatment. Blue dots—downregulated genes (adj. P‐value < 0.05, log2FC < −1) and red dots—upregulated genes (adj. P‐value < 0.05, log2FC > −1), with 10 most upregulated and 10 most downregulated genes labeled, tested by the limma voom algorithm with Benjamini–Hochberg correction. (I) Oncology‐related pathways (PROGENy) scored according to DE analysis of CLL cells co‐cultured with M2‐10B4 cells with and without PF‐670462 treatment. (J) NFKBIA expression in CLL cells co‐cultured with M2‐10B4 cells with and without PF‐670462 (PF‐67) treatment according to bulk RNAseq, tested by the limma voom algorithm with Benjamini–Hochberg correction. (K) Percentage of IκBα + CLL cells co‐cultured with M2‐10B4 cells with and without PF‐670462 treatment according to flow cytometry, tested by Friedman's test coupled with Dunn's post hoc test. (L–N) Heatmaps of expression changes in genes associated with nuclear factor κ B (NFκB) (L), tumor necrosis factor α (TNFα) (M), and phosphoinositide 3‐kinase (PI3K) (N) by the PROGENy model. CLL cells cultured with vs. without M2‐10B4 and CLL cells co‐cultured with M2‐10B4 treated with vs. without PF‐670462 were compared. (O) Seurat cell cycle scoring output shown for the G2/M phase, tested by Friedman's test coupled with Dunn's post hoc test (5 M cells were analyzed per sample).

To better understand the biological significance of DE results, we have scored the activity of signaling pathways and TFs using the algorithms PROGENy 34 and DoRothEA, 48 respectively (Figure 4D, Supporting Information S1: Figure 4A–H) (for detailed information, please see the code on GitHub). Not surprisingly, the most substantial changes were connected to cell adhesion and cell signaling (Supporting Information S1: Figure 4A–D). Clearly, the most upregulated pathways involved NFκB, followed by tumor necrosis factor α (TNFα), mitogen‐activated protein kinase (MAPK), epidermal growth factor (EGFR), and transforming growth factor β (TGFβ) pathways (Figure 4D, Supporting Information S1: Figure 4E–H). To validate upregulation of the NFκB pathway, we stained IκBα (encoded by NFKBIA) in primary CLL cells co‐cultured with M2‐10B4 cells (for the gating strategy, see Supporting Information S1: Figure 4I). NFKBIA, regulated via a negative feedback loop 49 and upregulated in RNAseq data (Figure 4E), was also upregulated at the protein level (Figure 4F). The pro‐survival effect of M2‐10B4 cells was confirmed by increased viability in co‐cultured CLL cells (Figure 4G). These data suggest that the co‐culture model with M2‐10B4 dramatically enhanced the survival of primary CLL cells with the NFκB pathway as the key molecular driver.

Next, we assessed the effects of PF‐670462 treatment on CLL cells in this model (Figure 4H, Supporting Information S2: Table 6). Here, the same pro‐survival and mitogenic pathways (TNFα, PI3K, NFκB, TGFβ, or EGFR) that are induced by the co‐culture were the most inhibited (Figure 4I–L, Supporting Information S1: Figure 4K–M), which is in line with the substantial decrease in the score of NFκB and other TF activity (Supporting Information S1: Figure 4J). The decrease in NFκB pathway activity manifested by downregulated NFKBIA (Figure 4M) was further confirmed at the protein level (Figure 4N). We have not observed any effects of CK1 inhibition on the activity of the PI3K (pAKT) or MAPK (pERK1/2git) pathway in primary CLL (Supporting Information S1: Figure 5), in line with the earlier literature. 9 It is noteworthy that similar to the mouse CLL cells (Figure 1K), the G2/M phase score was significantly increased after CK1i in primary CLL cells, both in the presence and in the absence of the co‐culture (Figure 4O). These data support the hypothesis about CK1i blocking perception and/or interpretation of the microenvironmental pro‐survival stimuli that are required for CLL cell survival and expansion.

CK1δ/ε inhibition suppresses the proliferation of primary human CLL cells

Our data from both the mouse model and primary CLL suggest that CK1i can efficiently inhibit CLL cell proliferation induced by the microenvironment. To address this directly, we took advantage of the recently introduced LN microenvironment‐mimicking co‐culture system that is based on direct stimulation of primary CLL cells by CD40L‐expressing HS5 stromal cells 19 (Figure 5A). Proliferation of primary CLL samples (N = 35) (Supporting Information S2: Table 1) after treatment with two different CK1 inhibitors—PF‐670462 (N = 35) and MU1742 (N = 21)—in this co‐culture model was assessed by flow cytometry using a dye dilution assay (for the gating strategy, see Supporting Information S1: Figure 6A).

Figure 5.

Figure 5

Open in a new tab

Analysis of primary chronic lymphocytic leukemia (CLL) cell proliferation upon casein kinase 1δ/ε (CK1δ/ε) inhibition in vitro. (A) Design of primary CLL cell co‐culture with the HS5 CD40L+ stromal cells, interleukin (IL)‐4 and IL‐21, and 6‐day treatment with DMSO (CTRL) or one of the tested CK1δ/ε‐targeting compounds (comprising PF‐670462, MU1742, and AH078), followed by analysis of CLL cell proliferation; the screened cohort of CLL patients (age range = 50–83 years, CLL patients with unmutated vs. mutated IGHV = approximately 2:1, TP53 WT:TP53 MUT = approximately 2:1). (B) Representative example of primary CLL cell proliferation upon 6 days of co‐culture with CD40L+ HS5 stromal cells and treatment with PF‐670462. (C–E) Percentages of proliferating primary CLL cells (calculated as % of CTRL) upon 6 days of co‐culture with CD40L+ HS5 stromal cells and treatment with (C) DMSO, 3 µM PF‐670462 or 10 µM PF‐670462, N = 35, (D) DMSO, 3 µM MU1742 or 10 µM MU1742, N = 21 and (E) DMSO, 3 µM AH078, or 10 µM AH078, N = 12. Differences were tested by the Kruskal–Wallis test and post hoc pairwise Wilcoxon rank sum tests with Benjamini–Hochberg correction. (F) Spearman correlation between the response of primary CLL cells from individual patients to both CK1 inhibitors, displayed by relative percentage of proliferating cells in response to 10 µM PF‐670462 on the x‐axis and relative percentage of proliferating cells in response to 10 µM MU1742 on the y‐axis (N = 21). (G) Division of all screened patients to response groups and the estimated threshold for classification: sensitive classified by <75% of proliferating cells (relative to CTRL) in response to 10 µM concentration of PF‐670462 and resistant classified as ≥75% of proliferating cells (relative to CTRL) in response to 10 µM concentration of PF‐670462 (N (sensitive) = 30; N (resistant) = 5). (H) Distribution of patients with TP53 WT versus TP53 MUT (further divided into monoallelic and biallelic TP53 defects) within distinct response groups (N (sensitive & TP53 WT) = 22; N (sensitive & TP53 mono) = 4; N (sensitive & TP53 bi) = 4; N (resistant & TP53 WT) = 1; N (resistant & TP53 mono) = 1; and N (resistant & TP53 bi) = 3), tested by Fisher's exact test. (I) Distribution of patients from distinct response groups among CLL patients carrying TP53 WT or TP53 MUT (N (TP53 WT & sensitive) = 22; N (TP53 WT & resistant) = 1; N (TP53 MUT & sensitive) = 8; and N (TP53 MUT & resistant) = 4), tested by Fisher's exact test.

Herein, we show that CK1i resulted in a significant dose‐dependent proliferation decrease in the case of both CK1 inhibitors and the selective CK1δ/ε degrader, AH078 (N = 12) (Figure 5B–E, Supporting Information S2: Table 7); the effects of both inhibitors were strongly correlated (Figure 5F), suggesting the same mechanism of action. This effect was not caused by the decreased viability of CLL patient cells. This accounts for all three used compounds, as the viability of patient cells upon treatment significantly correlated with their viability in the control conditions (Supporting Information S1: Figure 6B–D).

Most, but not all, patients responded to the treatment, which allowed us to stratify the patients (Figure 5G) into sensitive and resistant response groups based on the response to 10 µM PF‐670462; PF‐670462 data were used because of the larger patient cohort. The resistance was independent of the potential of CLL cells to proliferate in co‐culture, as shown by similar degrees of proliferation in both groups (Supporting Information S1: Figure 6E). Comparison of the clinical and molecular parameters between sensitive and resistant patients did not reveal any significant differences, with one notable exception—the TP53 status. Resistant patients carried significantly more often disruption of TP53 either by deletion or mutation, referred to as TP53 MUT (Figure 5H). Still, majority of TP53 MUT patients responded to CK1i treatment (Figure 5I). Further analysis of the biallelic defects of TP53 showed that significantly larger number of patients resistant to CK1i treatment carried the biallelic defect of TP53 (Figure 5H). Furthermore, it is noteworthy that biallelic defects of TP53, defined by combinatorial effects of the missense TP53 mutation, deletion, or loss of heterozygosity that affects both alleles in the genome (Supporting Information S1: Figure 6F), have already been linked to more aggressive CLL and worse patient survival. 50

However, due to the heterogeneity of primary CLL cells that were used as the input, even the situation in our in vitro assay is more complex and thus needs further examination at the single‐cell level. To gain insight into the response of individual cells in our in vitro co‐culture system, we have examined the data in the context of the proportion of CLL cells carrying TP53 MUT in each sample. Not surprisingly, a partial positive correlation between the response to treatment and the percentage of cumulative TP53 mutation existed (Supporting Information S1: Figure 6G), but still there were sensitive patients with cumulative TP53 MUT > 40% and, vice versa, poorly responding patients with low to no cumulative TP53 MUT. This suggests that despite the observed trend, the link between CK1i resistance and TP53 MUT might not be causative.

To conclude, the co‐culture model, which mimics the complex situation in CLL patients and combines the intrinsic and extrinsic factors that regulate CLL proliferation, confirmed that CK1i counteracts CLL cell expansion with high efficacy but suggested a higher risk of resistance in TP53 MUT patients, especially those carrying biallelic defects on TP53.

DISCUSSION

In our previous studies, we were able to identify CK1 as a promising target for CLL therapy. CK1δ/ε inhibitor PF‐670462 showed therapeutic efficacy in the Eµ‐TCL1 mouse model, and in some parameters, such as post‐treatment survival, was superior to ibrutinib. 9 , 11 In this study, we provide evidence that in addition to inhibition of CLL cell migration, 9 , 10 , 14 CK1i prevents CLL development by three other mechanisms: (i) cell cycle arrest and deceleration of CLL cell proliferation, (ii) interference with stroma‐derived pro‐survival stimuli that activate the NFκB pathway, and (iii) modulation of the LN microenvironment at the level of T cells (Figure 6).

Figure 6.

Figure 6

Open in a new tab

Graphical summary with the main outputs of our study. CK1i, casein kinase 1 inhibition; CLL, chronic lymphocytic leukemia; LN, lymph node; WT, wild type.

The specific and profound effect of CK1i on cell cycle is characterized by the enrichment of cells in the late S and G2 phases. We were able to map this phenotype to the increased activity of dedicated and well‐characterized cell cycle regulators from the E2f family. 35 , 51 Interestingly, another trait that these TFs have in common is their role in the DNA damage response (DDR).37, 52, 53, 54, 55 The possible activation of DDR is in line with the observed CLL cell cycle lengthening—it was shown that in multiple in vitro models, consequences of DDR included delay of the G2 phase onset 56 , 57 and, in critical cases, even induction of G2 phase arrest and apoptosis. 54 The presence of this phenomenon is often detected at the gene expression level by the upregulation of Rad51, a key player in DDR that contributes to double‐strand break (DSB) repair by orchestration of homologous recombination. 58 It is noteworthy that Rad51 was found to be upregulated by CK1i in CLL cells in our in vivo model (Supporting Information S2: Table 2).

The second mechanism underlying the therapeutic mechanism of CK1i that was uncovered in this study is the capacity of CK1 inhibitors to interfere with the pro‐survival signals triggering NFκB response. Co‐culture of CLL cells with a stromal microenvironment boosts multiple signaling pathways (predominantly NFκB, TNFα, and MAPK) and switches on dedicated TFs (mainly NFKB1, RELA, STAT3, and STAT6), the upregulation of which has been linked to CLL.59, 60, 61, 62 This response is largely blocked by CK1i. Since NFκB has been identified as one of the key pathways stimulating CLL clone expansion and survival in the microenvironment of lymphoid organs, 46 , 59 we propose this as the second mode of action of CK1i in CLL suppression.

Although our results do not provide mechanistic insight into the regulation of NFκB by CK1, and more research on this topic is needed, they still provide several hints. There is evidence that loss of CK1α leads to inhibition of NFκB signaling in Jurkat cells (acute lymphocytic leukemia cell line) due to effects on the CARMA1/CARD11, BCL10, and MALT1 (CBM) complex, which normally functions as the NFκB activating platform. 63 However, it is not clear if CK1δ/ε can act in the same way as CK1α. Alternatively, CK1i can modulate the activity of the NFκB pathway via interaction with Traf3—a negative NFκB regulator, which was previously shown to suppress the ability of CLL cells to perceive pro‐survival signals from T cells that are transmitted via the CD40L:CD40 axis. 33 CK1 inhibition in our syngeneic mouse models led to significant upregulation of Traf3 in CLL cells, and it has been shown recently that CK1ε is able to bind and phosphorylate TRAF3 in multiple human cell lines. 64 These findings support the possibility that there is a direct link between CK1ε and TRAF3 affecting the ability of TRAF3 to negatively regulate the NFκB signaling downstream of the CD40L:CD40 axis and thus, lower the CLL cell responsiveness to extracellular stimuli.

Generally, NFκB is located downstream from multiple cell surface stimuli, including CD40 activation, 65 which links the intracellular effects of CK1i from an in vitro model to the loss of the main CD40L signal providers, T cells, from the LN microenvironment in vivo. We propose this, i.e., the ability of CK1i to interfere with T‐cell accumulation in LN, as the third mechanism of action. More specifically, we found that CK1i could revert CLL‐induced increase of CD4+ and CD8+ T cells in LN. It has been shown that the CLL clone shapes its own microenvironment, mainly by recruitment of the B‐cell‐stimulatory CD4+ T‐cell subsets. 23 , 66 It is believed that particularly Tfh cells that provide supportive stimuli via CD40L, LFA‐1, and production of IL‐21 42 and Th2 cells that stimulate B cells via IL‐4 43 promote CLL cell expansion. Currently, we are unable to confirm if T‐cell alterations observed in the lymphoid tissues are the cause of CLL suppression. Our findings from in vitro co‐cultured CLL patient cells indicate that there is a substantial contribution of the T‐cell‐mediated stimuli to CLL progression, since IL‐21 and CD40L have been identified to be among the most crucial regulators of CLL cell survival 67 and expansion. 68 However, it is likely that the cell‐intrinsic effects of CK1i on the CLL cells (Figure 1) also contribute, and we are unable to dissect the relative contribution of both mechanisms. Herein, we acknowledge that further research (e.g., on the Eµ‐TCL1 AT mouse model upon selective depletion of the T‐cell pool) might be useful to identify the primary inhibitory effects of CK1 targeting drugs on CLL.

When focusing solely on the CD40L stimulus, it is noteworthy that this signal links the T cells to the CLL cell surface, which leads to activation of NFκB signaling in the cytosol. 69 The activation of the NFκB pathway has crucial therapeutic consequences, as shown by a positive correlation between CD40 signaling (coupled with activation of the NFκB pathway) and the resistance of CLL cells against ibrutinib and venetoclax. 70 , 71 Specifically, we show that CK1i is particularly potent in the co‐culture system, which triggers primary CLL proliferation by the combination of CD40L, IL‐4, and IL‐21. 19 We propose that in this system mimicking CLL–T‐cell interactions in immune niches, CK1i acts via a combination of the cell‐intrinsic effect on the proliferation of CLL cells and interference with the pro‐survival and pro‐proliferative stimuli. This multiple‐hit mode of action provides new options in the treatment of aggressive CLL clones and represents a potentially valuable aspect of CK1i. This was also supported by the high efficacy of CK1i in the case of CLL cells originating from TP53 MUT patients, except for the high‐risk CLL group of patients carrying biallelic defects on TP53. 50

In conclusion, we elucidated the multiple‐hit mode of action of CK1i on CLL cells (Figure 6). These massive and multilayered changes in the biology of CLL cells attenuate processes that drive CLL progression and aggressiveness. Our work advocates for further preclinical development of CK1δ/ε inhibitors as innovative therapeutic agents in CLL.

AUTHOR CONTRIBUTIONS

Antonia Mikulova: Conceptualization; methodology; investigation; data curation; formal analysis; writing—original draft; writing—review and editing; visualization. Hana Plesingerova: Methodology; data curation; investigation; formal analysis; visualization; writing—original draft; writing—review and editing. Michaela Chorvatova: Investigation; formal analysis; writing—review and editing. Pavlina Kebkova: Investigation; formal analysis; writing—review and editing. Jana Bartosikova: Formal analysis; writing—review and editing; investigation. Petra Prochazkova: Investigation; formal analysis; writing—review and editing. Jan Verner: Resources; writing—review and editing; investigation. Dominik Mulidran: Investigation; formal analysis; writing—review and editing. Stepan Cada: Investigation; formal analysis; writing—review and editing. Olga Vondalova‐Blanarova: Investigation; formal analysis; writing—review and editing. Nela Kolcakova: Investigation; formal analysis; writing—review and editing. Tomas Loja: Resources; investigation; writing—review and editing. Terezia Kurucova: Resources; investigation; writing—review and editing. Eva Hoferkova: Resources; writing—review and editing. Boris Tichy: Investigation; resources; writing—review and editing. Jana Kotaskova: Resources; writing—review and editing. Marek Mraz: Writing—review and editing; resources. Lukas Kubala: Conceptualization; writing—review and editing. Pavlina Janovska: Investigation; methodology; supervision; funding acquisition; conceptualization; writing—review and editing. Vitezslav Bryja: Supervision; funding acquisition; conceptualization; writing—original draft; writing—review and editing.

CONFLICT OF INTEREST STATEMENT

P.V. and V.B. are the founders and shareholders of CasInvent Pharma a.s. that develops CK1 inhibitors for clinical use. Other authors have no conflicts of interest to declare.

ETHICS STATEMENT

CLL samples were taken obtaining informed consent and approval from the Ethical Committee of the University Hospital Brno in accordance with the Declaration of Helsinki and institutional guidelines.

FUNDING

This work was supported by the Czech Science Foundation Grant No. GA23‐05561S, by the Ministry of Health of the Czech Republic, grant no. NU23‐08‐00448, MH CZ ‐ DRO (FNBr, 65269705), and the project National Institute for Cancer Research (Programme EXCELES, ID Project No. LX22NPO5102)—Funded by the European Union – Next Generation EU.

Supporting information

Supporting Information.

HEM3-10-e70343-s002.pdf (35.9MB, pdf)

Supporting Information.

HEM3-10-e70343-s001.xlsx (3.4MB, xlsx)

ACKNOWLEDGMENTS

We acknowledge the Core Facility Genomics and the Core Facility Bioinformatics supported by the NCMG research infrastructure (LM2023067 funded by MEYS CR) for their support with obtaining scientific data presented in this paper. We also acknowledge the research group of Emma Andersson at the Karolinska Institutet in Stockholm for their support regarding the RNA sequencing of co‐cultured CLL patient samples. Furthermore, we thank Lenka Radová from the Central European Institute of Technology (CEITEC) in Brno for her support with the preprocessing of RNA sequencing data from the co‐culture model. Finally, we acknowledge the research group of Prof. Stefan Knapp at the Goethe Universität in Frankfurt for providing the CK1δ/ε degrader, AH078, for the purposes of experimental work in this publication.

Contributor Information

Pavlina Janovska, Email: janovska@sci.muni.cz.

Vitezslav Bryja, Email: bryja@sci.muni.cz.

DATA AVAILABILITY STATEMENT

All data from the flow cytometry are available at the ImmPort Database under the accession number SDY3219 for the flow cytometric data from murine in vivo experiments, SDY3220 for flow cytometric data from primary co‐culture CLL experiments aimed at the analysis of proliferation, and SDY3225 for flow cytometric data from primary CLL co‐culture experiments aimed at the analysis of NFκB. The data from bulk RNA sequencing are available at the Gene Expression Omnibus Database with distinct accession numbers for murine Eµ‐TCL1 adoptive transfer RNAseq data (GSE304722) and CLL patient RNAseq data (GSE305855). The codes for both flow cytometric and bulk RNA sequencing data analysis are available on GitHub, under doi:10.5281/zenodo.17347989.

REFERENCES

  • 1. Wierda WG, Byrd JC, Abramson JS, et al. Chronic lymphocytic leukemia/small lymphocytic lymphoma, version 4.2020, NCCN clinical practice guidelines in oncology. J Natl Compr Cancer Netw JNCCN. 2020;18(2):185‐217. [DOI] [PubMed] [Google Scholar]
  • 2. Hallek M. Chronic lymphocytic leukemia: 2025 update on the epidemiology, pathogenesis, diagnosis, and therapy. Am J Hematol. 2025;100(3):450‐480. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Brown JR, Cymbalista F, Sharman J, Jacobs I, Nava‐Parada P, Mato A. The role of rituximab in chronic lymphocytic leukemia treatment and the potential utility of biosimilars. Oncologist. 2018;23(3):288‐296. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Molica S. Venetoclax: a real game changer in treatment of chronic lymphocytic leukemia. Int J Hematol Oncol. 2020;9(4):IJH31. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Lampson BL, Kasar SN, Matos TR, et al. Idelalisib given front‐line for treatment of chronic lymphocytic leukemia causes frequent immune‐mediated hepatotoxicity. Blood. 2016;128(2):195‐203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Jain N, Keating M, Thompson P, et al. Ibrutinib and venetoclax for first‐line treatment of CLL. N Engl J Med. 2019;380(22):2095‐2103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Rogers KA, Woyach JA. The evolving frontline management of CLL: are triplets better than doublets? How will we find out? Hematology. 2024;2024(1):467‐473. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Sharman JP, Munir T, Grosicki S, et al. BRUIN CLL‐321: randomized phase III trial of pirtobrutinib versus idelalisib plus rituximab (IdelaR) or bendamustine plus rituximab (BR) in BTK inhibitor pretreated chronic lymphocytic leukemia/small lymphocytic lymphoma. Blood. 2024;144(suppl 1):886. [Google Scholar]
  • 9. Janovska P, Verner J, Kohoutek J, et al. Casein kinase 1 is a therapeutic target in chronic lymphocytic leukemia. Blood. 2018;131(11):1206‐1218. [DOI] [PubMed] [Google Scholar]
  • 10. Čada Š, Vondálová Blanářová O, Gömoryová K, et al. Role of casein kinase 1 in the amoeboid migration of B‐cell leukemic and lymphoma cells: a quantitative live imaging in the confined environment. Front Cell Dev Biol. 2022;10:911966. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Janovská P, Normant E, Miskin H, Bryja V. Targeting casein kinase 1 (CK1) in hematological cancers. Int J Mol Sci. 2020;21(23):9026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Čada Š, Bryja V. Local Wnt signalling in the asymmetric migrating vertebrate cells. Semin Cell Dev Biol. 2022;125:26‐36. [DOI] [PubMed] [Google Scholar]
  • 13. Kaucká M, Plevová K, Pavlová Š, et al. The planar cell polarity pathway drives pathogenesis of chronic lymphocytic leukemia by the regulation of B‐lymphocyte migration. Cancer Res. 2013;73(5):1491‐1501. [DOI] [PubMed] [Google Scholar]
  • 14. Janovska P, Poppova L, Plevova K, et al. Autocrine signaling by Wnt‐5a deregulates chemotaxis of leukemic cells and predicts clinical outcome in chronic lymphocytic leukemia. Clin Cancer Res. 2016;22(2):459‐469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Aquino Perez C, Burocziova M, Jenikova G, Macurek L. CK1‐mediated phosphorylation of FAM110A promotes its interaction with mitotic spindle and controls chromosomal alignment. EMBO Rep. 2021;22(7):e51847. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Schittek B, Sinnberg T. Biological functions of casein kinase 1 isoforms and putative roles in tumorigenesis. Mol Cancer. 2014;13(1):231. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Etchegaray JP, Machida KK, Noton E, et al. Casein kinase 1 delta regulates the pace of the mammalian circadian clock. Mol Cell Biol. 2009;29(14):3853‐3866. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Svanberg R, Janum S, Patten PEM, Ramsay AG, Niemann CU. Targeting the tumor microenvironment in chronic lymphocytic leukemia. Haematologica. 2021;106(9):2312‐2324. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Hoferkova E, Seda V, Kadakova S, et al. Stromal cells engineered to express T cell factors induce robust CLL cell proliferation in vitro and in PDX co‐transplantations allowing the identification of RAF inhibitors as anti‐proliferative drugs. Leukemia. 2024;38(8):1699‐1711. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Collins RJ, Verschuer LA, Harmon BV, Prentice RL, Pope JH, Kerr JFR. Spontaneous programmed death (apoptosis) of B‐chronic lymphocytic leukaemia cells following their culture in vitro. Br J Haematol. 1989;71(3):343‐350. [DOI] [PubMed] [Google Scholar]
  • 21. Tretter T, Schuler M, Schneller F, et al. Direct cellular interaction with activated CD4+T cells overcomes hyporesponsiveness of B‐cell chronic lymphocytic leukemia in vitro. Cell Immunol. 1998;189(1):41‐50. [DOI] [PubMed] [Google Scholar]
  • 22. Moore J, Prystowsky M, Hoover R, Besa E, Nowell P. Defective T cell‐mediated, isotype‐specific immunoglobulin regulation in B cell chronic lymphocytic leukemia. Blood. 1988;71(4):1012‐1020. [PubMed] [Google Scholar]
  • 23. Vaca AM, Ioannou N, Sivina M, et al. Activation and expansion of T‐follicular helper cells in chronic lymphocytic leukemia nurselike cell co‐cultures. Leukemia. 2022;36(5):1324‐1335. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Bresin A, D'Abundo L, Narducci MG, et al. TCL1 transgenic mouse model as a tool for the study of therapeutic targets and microenvironment in human B‐cell chronic lymphocytic leukemia. Cell Death Dis. 2016;7(1):e2071. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Bichi R, Shinton SA, Martin ES, et al. Human chronic lymphocytic leukemia modeled in mouse by targeted TCL1 expression. Proc Natl Acad Sci. 2002;99(10):6955‐6960. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Zaborsky N, Gassner FJ, Höpner JP, et al. Exome sequencing of the TCL1 mouse model for CLL reveals genetic heterogeneity and dynamics during disease development. Leukemia. 2019;33(4):957‐968. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Roessner PM, Seiffert M. T‐cells in chronic lymphocytic leukemia: guardians or drivers of disease? Leukemia. 2020;34(8):2012‐2024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Göthert JR, Eisele L, Klein‐Hitpass L, et al. Expanded CLL CD8+ T‐cells are driven into a senescent KLRG1+ effector memory phenotype. Blood. 2010;116(21):913. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Němec V, Khirsariya P, Janovská P, et al. Discovery of potent and exquisitely selective inhibitors of kinase CK1 with tunable isoform selectivity. Angew Chem Int Ed. 2023;62(11):e202217532. [DOI] [PubMed] [Google Scholar]
  • 30. Haag A, Němec V, Janovská P, et al. Development and discovery of a selective degrader of casein kinases 1 δ/ε. J Med Chem. 2025;68(1):506‐530. [DOI] [PubMed] [Google Scholar]
  • 31. Boström J, Sramkova Z, Salašová A, et al. Comparative cell cycle transcriptomics reveals synchronization of developmental transcription factor networks in cancer cells. PLoS One. 2017;12(12):e0188772. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Widhopf GF, Cui B, Ghia EM, et al. ROR1 can interact with TCL1 and enhance leukemogenesis in Eµ‐TCL1 transgenic mice. Proc Natl Acad Sci. 2014;111(2):793‐798. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. He JQ, Oganesyan G, Saha SK, Zarnegar B, Cheng G. TRAF3 and its biological function. In: Wu H, TNF Receptor Associated Factors (TRAFs) [Internet]. Advances in Experimental Medicine and Biology, vol. 597. Springer New York; 2007:48‐59. 10.1007/978-0-387-70630-6_4 [DOI] [PubMed] [Google Scholar]
  • 34. Badia‐i‐Mompel P, Vélez Santiago J, Braunger J, et al. decoupleR: ensemble of computational methods to infer biological activities from omics data. Bioinforma Adv. 2022;2(1):vbac016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Sala A, Nicolaides NC, Engelhard A, et al. Correlation between E2F‐1 requirement in the S phase and E2F‐1 transactivation of cell cycle‐related genes in human cells. Cancer Res. 1994;54(6):1402‐1406. [PubMed] [Google Scholar]
  • 36. Alvarez‐Fernández M, Medema RH. Novel functions of FoxM1: from molecular mechanisms to cancer therapy. Front Oncol. 2013;3:30. 10.3389/fonc.2013.00030 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Wei L, Bai Y, Na L, Sun Y, Zhao C, Wang W. E2F3 induces DNA damage repair, stem‐like properties and therapy resistance in breast cancer. Biochim Biophys Acta BBA ‐ Mol Basis Dis. 2023;1869(8):166816. [DOI] [PubMed] [Google Scholar]
  • 38. Bandara LR, Buck VM, Zamanian M, Johnston LH, La Thangue NB. Functional synergy between DP‐1 and E2F‐1 in the cell cycle‐regulating transcription factor DRTF1/E2F. EMBO J. 1993;12(11):4317‐4324. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Knippschild U, Milne DM, Campbell LE, et al. p53 is phosphorylated in vitro and in vivo by the delta and epsilon isoforms of casein kinase 1 and enhances the level of casein kinase 1 delta in response to topoisomerase‐directed drugs. Oncogene. 1997;15(14):1727‐1736. [DOI] [PubMed] [Google Scholar]
  • 40. Yang W, Garrett L, Feng D, et al. Wnt‐induced Vangl2 phosphorylation is dose‐dependently required for planar cell polarity in mammalian development. Cell Res. 2017;27(12):1466‐1484. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Vlachonikola E, Stamatopoulos K, Chatzidimitriou A. T Cells in chronic lymphocytic leukemia: a two‐edged sword. Front Immunol. 2021;11:612244. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Le Saos‐Patrinos C, Loizon S, Zouine A, et al. Elevated levels of circulatory follicular T helper cells in chronic lymphocytic leukemia contribute to B cell expansion. J Leukoc Biol. 2023;113(3):305‐314. [DOI] [PubMed] [Google Scholar]
  • 43. Puzzolo MC, Del Giudice I, Peragine N, et al. TH2/TH1 shift under ibrutinib treatment in chronic lymphocytic leukemia. Front Oncol. 2021;11:637186. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44. Pagano G, Botana IF, Wierz M, et al. Interleukin‐27 potentiates CD8+ T‐cell‐mediated antitumor immunity in chronic lymphocytic leukemia. Haematologica. 2023;108(11):3011‐3024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. Maharaj K, Powers JJ, Achille A, et al. The dual PI3Kδ/CK1ε inhibitor umbralisib exhibits unique immunomodulatory effects on CLL T cells. Blood Adv. 2020;4(13):3072‐3084. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Herishanu Y, Pérez‐Galán P, Liu D, et al. The lymph node microenvironment promotes B‐cell receptor signaling, NF‐κB activation, and tumor proliferation in chronic lymphocytic leukemia. Blood. 2011;117(2):563‐574. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47. Kurtova AV, Balakrishnan K, Chen R, et al. Diverse marrow stromal cells protect CLL cells from spontaneous and drug‐induced apoptosis: development of a reliable and reproducible system to assess stromal cell adhesion‐mediated drug resistance. Blood. 2009;114(20):4441‐4450. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48. Garcia‐Alonso L, Holland CH, Ibrahim MM, Turei D, Saez‐Rodriguez J. Benchmark and integration of resources for the estimation of human transcription factor activities. Genome Res. 2019;29(8):1363‐1375. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49. Fagerlund R, Behar M, Fortmann KT, Lin YE, Vargas JD, Hoffmann A. Anatomy of a negative feedback loop: the case of I κ B α . J R Soc Interface. 2015;12(110):20150262. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50. Malcikova J, Smardova J, Rocnova L, et al. Monoallelic and biallelic inactivation of TP53 gene in chronic lymphocytic leukemia: selection, impact on survival, and response to DNA damage. Blood. 2009;114(26):5307‐5314. [DOI] [PubMed] [Google Scholar]
  • 51. Lees E, Faha B, Dulic V, Reed SI, Harlow E. Cyclin E/cdk2 and cyclin A/cdk2 kinases associate with p107 and E2F in a temporally distinct manner. Genes Dev. 1992;6(10):1874‐1885. [DOI] [PubMed] [Google Scholar]
  • 52. Iglesias‐Ara A, Zenarruzabeitia O, Buelta L, Merino J, Zubiaga AM. E2F1 and E2F2 prevent replicative stress and subsequent p53‐dependent organ involution. Cell Death Differ. 2015;22(10):1577‐1589. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53. Castillo DS, Campalans A, Belluscio LM, et al. E2F1 and E2F2 induction in response to DNA damage preserves genomic stability in neuronal cells. Cell Cycle. 2015;14(8):1300‐1314. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54. Martinez LA, Goluszko E, Chen HZ, et al. E2F3 is a mediator of DNA damage‐induced apoptosis. Mol Cell Biol. 2010;30(2):524‐536. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55. Plesca D, Crosby ME, Gupta D, Almasan A. E2F4 function in G2 maintaining G2‐arrest to prevent mitotic entry with damaged DNA. Cell Cycle. 2007;6(10):1147‐1152. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56. Chao HX, Poovey CE, Privette AA, et al. Orchestration of DNA damage checkpoint dynamics across the human cell cycle. Cell Syst. 2017;5(5):445‐459.e5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57. Crosby ME, Jacobberger J, Gupta D, Macklis RM, Almasan A. E2F4 regulates a stable G2 arrest response to genotoxic stress in prostate carcinoma. Oncogene. 2007;26(13):1897‐1909. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58. Gachechiladze M, Škarda J, Soltermann A, Joerger M. RAD51 as a potential surrogate marker for DNA repair capacity in solid malignancies. Int J Cancer. 2017;141(7):1286‐1294. [DOI] [PubMed] [Google Scholar]
  • 59. Liu FT, Jia L, Wang P, Wang H, Farren TW, Agrawal SG. STAT3 and NF‐κB cooperatively control in vitro spontaneous apoptosis and poor chemo‐responsiveness in patients with chronic lymphocytic leukemia. Oncotarget. 2016;7(22):32031‐32045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60. Adami F, Guarini A, Pini M, et al. Serum levels of tumour necrosis factor‐α in patients with B‐cell chronic lymphocytic leukaemia. Eur J Cancer. 1994;30(9):1259‐1263. [DOI] [PubMed] [Google Scholar]
  • 61. Liao W, Jordaan G, Coriaty N, Sharma S. Amplification of B cell receptor–Erk signaling by Rasgrf‐1 overexpression in chronic lymphocytic leukemia. Leuk Lymphoma. 2014;55(12):2907‐2916. [DOI] [PubMed] [Google Scholar]
  • 62. Hewamana S, Lin TT, Rowntree C, et al. Rel A is an independent biomarker of clinical outcome in chronic lymphocytic leukemia. J Clin Oncol. 2009;27(5):763‐769. [DOI] [PubMed] [Google Scholar]
  • 63. Bidère N, Ngo VN, Lee J, et al. Casein kinase 1α governs antigen‐receptor‐induced NF‐κB activation and human lymphoma cell survival. Nature. 2009;458(7234):92‐96. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 64. Zhou Y, He C, Yan D, et al. The kinase CK1ɛ controls the antiviral immune response by phosphorylating the signaling adaptor TRAF3. Nat Immunol. 2016;17(4):397‐405. [DOI] [PubMed] [Google Scholar]
  • 65. Mineva ND, Rothstein TL, Meyers JA, Lerner A, Sonenshein GE. CD40 ligand‐mediated activation of the de novo RelB NF‐κB synthesis pathway in transformed B cells promotes rescue from apoptosis. J Biol Chem. 2007;282(24):17475‐17485. [DOI] [PubMed] [Google Scholar]
  • 66. Saulep‐Easton D, Vincent FB, Quah PS, et al. The BAFF receptor TACI controls IL‐10 production by regulatory B cells and CLL B cells. Leukemia. 2016;30(1):163‐172. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 67. Kater AP, Evers LM, Remmerswaal EBM, et al. CD40 stimulation of B‐cell chronic lymphocytic leukaemia cells enhances the anti‐apoptotic profile, but also Bid expression and cells remain susceptible to autologous cytotoxic T‐lymphocyte attack. Br J Haematol. 2004;127(4):404‐415. [DOI] [PubMed] [Google Scholar]
  • 68. Pascutti MF, Jak M, Tromp JM, et al. IL‐21 and CD40L signals from autologous T cells can induce antigen‐independent proliferation of CLL cells. Blood. 2013;122(17):3010‐3019. [DOI] [PubMed] [Google Scholar]
  • 69. O'Donnell A, Pepper C, Mitchell S, Pepper A. NF‐kB and the CLL microenvironment. Front Oncol. 2023;13:1169397. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 70. Jayappa KD, Portell CA, Gordon VL, et al. Microenvironmental agonists generate de novo phenotypic resistance to combined ibrutinib plus venetoclax in CLL and MCL. Blood Adv. 2017;1(14):933‐946. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 71. Sharma S, Pavlasova GM, Seda V, et al. miR‐29 modulates CD40 signaling in chronic lymphocytic leukemia by targeting TRAF4: an axis affected by BCR inhibitors. Blood. 2021;137(18):2481‐2494. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information.

HEM3-10-e70343-s002.pdf (35.9MB, pdf)

Supporting Information.

HEM3-10-e70343-s001.xlsx (3.4MB, xlsx)

Data Availability Statement

All data from the flow cytometry are available at the ImmPort Database under the accession number SDY3219 for the flow cytometric data from murine in vivo experiments, SDY3220 for flow cytometric data from primary co‐culture CLL experiments aimed at the analysis of proliferation, and SDY3225 for flow cytometric data from primary CLL co‐culture experiments aimed at the analysis of NFκB. The data from bulk RNA sequencing are available at the Gene Expression Omnibus Database with distinct accession numbers for murine Eµ‐TCL1 adoptive transfer RNAseq data (GSE304722) and CLL patient RNAseq data (GSE305855). The codes for both flow cytometric and bulk RNA sequencing data analysis are available on GitHub, under doi:10.5281/zenodo.17347989.

Articles from HemaSphere are provided here courtesy of Wiley

RESOURCES