ST2825 was purchased from MedChemExpress. The Bruton's tyrosine kinase (BTK) inhibitor ibrutinib, and B-cell lymphoma-2 (BCL-2) inhibitor ABT-199 were purchased from Selleck Chemicals. All drugs were dissolved in 100% dimethyl sulfoxide (DMSO). For all samples in all of the experiments, the final DMSO concentrations were diluted to 0.1% with cell culture media, including the vehicle controls.

SU-DHL-4, OCI-LY10 and TMD8 cell lines were purchased from the Cell Bank of Type Culture Collection of the Chinese Academy of Sciences. The MYD88 L265P mutation of each cell line was identified using Sanger sequencing. The cells were cultured in RPMI-1640 supplemented with 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc.). The HEK293T cell line was cultured in Dulbecco's modified Eagle's medium (DMEM; Gibco; Thermo Fisher Scientific, Inc.) with 10% FBS. All cell lines were cultured at 37°C in a 5% CO₂ incubator.

Cell viability was assessed using WST-1 reagent (Roche Diagnostics) as instructed by the manufacturer. Briefly, ~2×10⁴ cells/well were seeded into 96-well plates and treated with either the vehicle (DMSO) or ST2825 at serial concentrations for 24, 48 or 72 h. After treatment, 10 µl WST-1 reagent was added to each well, followed by a 4-h incubation at 37°C. Cell viability was calculated by measuring the absorbance at 440 nm using a 96-well plate reader, and the data were normalized to that of the vehicle-treated cells. For drug combination experiments, cell viability was determined 72 h after treatment with the indicated drugs. Flow cytometric analysis of apoptosis was performed using an Annexin V-FITC Apoptosis Detection kit (Sigma-Aldrich; Merck KGaA) 48 h after ST2825 treatment (5 or 10 µM) according to the manufacturer's protocol; the results were acquired using a flow cytometer (BD Influx; BD Biosciences) and analyzed using BD FACSuite software version 1.0.6. For drug combination experiments, the apoptotic cell populations were analyzed 48 h after treatment with the indicated drugs. The nuclear morphology of the apoptotic cells was further examined using DAPI staining. Cells were collected and coated onto slides after washing with PBS three times. The cells were then fixed in 4% paraformaldehyde at room temperature for 15 min and then permeabilized with 0.1% Triton X-100 at 37°C for 5 min. The cells were subsequently stained with 5 µg/ml DAPI in a dark room at 37°C for 15 min. After three washes with PBS, the cells were visualized using a fluorescence microscope (Eclipse 80i; Nikon Corporation) under a 40× oil immersion objective.

Western blot analysis was performed as previously described (15). Cells were lysed with RIPA lysis buffer supplemented with proteinase and phosphatase inhibitors (Cell Signaling Technology, Inc.) at 4°C for 20 min. The supernatants were collected after centrifugation at 13,000 × g for 15 min (4°C), and the protein concentration was determined using a bicinchoninic acid (BCA) protein assay kit (Thermo Fisher Scientific, Inc.). Equal amounts of protein extract (30 µg per lane) were loaded for separation onto a 10% SDS-PAGE gel and transferred to PVDF membranes (EMD Millipore), before blocking in 5% non-fat milk/TBST at room temperature for 1 h. Then, the membranes were incubated overnight at 4°C with primary antibodies (all purchased from Cell Signaling Technology, Inc., and used at a dilution of 1:1,000) against MYD88 (cat. no. 4823S), BTK (cat. no. 8547S), phospho-BTK (cat. no. 5082S), inhibitor of NF-κB (IκB; cat. no. 4812S) and phospho-IκB (cat. no. 2859S), followed by incubation with anti-rabbit horseradish peroxidase (HRP)-conjugated secondary antibody (cat. no. sc-2030; Santa Cruz Biotechnology, Inc.; 1:5,000) for 1 h at room temperature. The β-actin antibody (cat. no. ab8227; Abcam) was used at a 1:1,000 dilution. Detection was performed using an enhanced chemiluminescence substrate (ECL; EMD Millipore), and the signals were visualized and analyzed using the Bio-Rad Gel Imaging System and the Cool Imager™ workstation II (Viagene, Biotech, Inc.).

MyD88 protein aggregation in DLBCL cell lines (OCI-LY10 and TMD8) was analyzed using fractionated cell lysates. Following overnight treatment with DMSO or 10 µM ST2825, 5×10⁶ cells were incubated in each well of a 6-well plate and then collected and lysed using cell lysis buffer (Beyotime Institute of Biotechnology) supplemented with PMSF and Benzonase (Sigma-Aldrich; Merck KGaA). The protein concentration was determined using a BCA protein assay kit (Thermo Fisher Scientific, Inc.) and equal amounts of cell lysates (60 µg) were then fractionated by centrifugation (16,000 × g, 10 min); the supernatant fractions were regarded as whole cell lysates (WCL) and the pellet fractions as insoluble HMW aggregates. The pellets were then washed with PBS and dissolved by boiling in SDS sample buffer with shaking. Western blotting was conducted using the aforementioned anti-MYD88 (Cell Signaling Technology, Inc.) and anti-β-actin (Abcam) antibodies and goat anti-rabbit HRP-conjugated secondary antibody at the indicated dilutions.

Cells from each line were lysed using cell lysis buffer for western and immunoprecipitation (Beyotime Institute of Biotechnology) supplemented with 1 mM PMSF. The samples were centrifuged at 13,000 × g for 15 min, and the protein concentration was determined using a BCA protein assay kit (Thermo Fisher Scientific, Inc.). Then protein A/G coated magnetic beads (Bio-Rad Laboratories, Inc.) were incubated with 5 µg IP antibodies against MYD88 or BTK (Cell Signaling Technology, Inc.) at room temperature with rotation for 10 min. After washing with PBS-T, the antibody-conjugated beads were added into the antigen-containing lysate, and rotated for 1 h at room temperature. Immunoprecipitates were then washed three times with PBS-T, separated on a magnetic stand and eluted using Laemmli loading buffer for further immunoblotting analysis.

HEK293T cells were seeded into 12-well plates (2×10⁵ cells/well) 24 h before transfection to achieve a confluency of 50–70%. Cells were then transiently transfected with 2 µg mCitrine-TIR WT or mCitrine-TIR L265P plasmids using Lipofectamine™ 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. After 24 h, the cells were treated with DMSO or 10 µM ST2825 and incubated for a further 48 h. Following incubation, the cells were observed using a laser scanning Nikon A1R confocal microscope under a 40× oil immersion objective (numerical aperture, 1.4). An excitation of 488 and 520–570 nm emission were used for visualization. The plasmids for mCitrine-TIR WT or L265P mutant were provided by Dr Roman Jerala (13).

CMV-Renilla luciferase lentivirus (Cignal Lenti CMV Renilla Control; Qiagen, Inc.) was used as an internal control for normalization, and NF-κB reporter lentivirus with firefly luciferase expression (Cignal Lenti NF-κB Reporter; Qiagen, Inc.) was used as the experimental reporter. The two lentiviruses were used to co-transfect the OCI-LY10 and TMD8 cells. Briefly, OCI-LY10 and TMD8 cells were resuspended in 0.5 ml lentivirus-containing medium (containing 250 µl Cignal Lenti NF-κB Reporter and 250 µl Cignal Lenti CMV Renilla Control) at a concentration of 1×10⁶ cells/ml in a 24-well tissue culture plate in the presence of polybrene (8 µg/ml). The plates were centrifuged at 800 × g for 90 min at 32°C. The cells were then washed and resuspended in fresh medium for an additional 72 h. Then puromycin and hygromycin were used together for selection. For the drug experiments, cells were subsequently seeded at a density of 4×10⁵ cells/ml and treated for 12 h with the vehicle, ST2825, ibrutinib or a combination of ST2825 and ibrutinib at the indicated concentrations. Luciferase activity was measured in 96-well plates using the Dual-Luciferase Reporter Assay system (Promega Corporation) according to the manufacturer's protocol.

The two ABC DLBCL cell lines were cultured in fresh media and treated for 24 h with 5 or 20 µM ST2825 or DMSO. The concentrations of IL-10 and IFN-β in the culture supernatants were measured by ELISA (cat. no. DY217B-05 and DY814-05, respectively; R&D Systems, Inc.) according to the manufacturer's protocol. All experiments were performed in triplicate.

Synergism of the drug combinations (including ST2825 combined with ibrutinib or ABT-199) was evaluated using CalcuSyn software (Premier Biosoft International), which is based on the median-effect principle applied by Chou and Talalay (16). The combined effects and combination index (CI) for each dose combination were calculated and are presented as a heat map. Drug synergism, addition and antagonism were defined by CI values of <1.0, 1.0 and >1.0, respectively.

Statistical analysis of the experimental data was performed using GraphPad Prism software version 6 (GraphPad Software, Inc.), and the data are presented as the mean ± standard deviation, unless otherwise indicated. The unpaired t-test was used to assess the statistical significance of the differences between two groups, and one-way analysis of variance with Dunnett's or Tukey's test was used for those between multiple groups. P<0.05 was considered to indicate a statistically significant difference.

To determine whether oligomerization of the MyD88 TIR-L265P mutants could be blocked by ST2825, MyD88 TIR WT or mutant linked to the fluorescent protein mCitrine were expressed in HEK293T cells. Consistent with the results of a previous report (13), mCitrine-TIR mutants were strongly aggregated, whilst this was not the case for the WT mCitrine TIR. Furthermore, the aggregation of mCitrine-TIR mutants was markedly inhibited following treatment with ST2825 (Fig. 1A).

The HMW fraction was separated from the cellular lysates via centrifugation, and western blotting was used to determine the presence of MyD88 aggregates in the pelleted HMW fractions. MyD88 aggregation was detected in the HMW fraction in DLBCL cell lines with the L265P mutation (OCI-LY10 and TMD8). The changes in MyD88 aggregation following treatment with ST2825 were then investigated by comparing the ratios of endogenous MyD88 levels between pelleted HMW fractions and whole cell lysates. Fig. 1B and C illustrate a significantly lower pellet:lysate ratio in the ST2825-treated cells compared with the DMSO treated cells, thus indicating that the degree of MyD88 aggregation was significantly decreased following ST2825 treatment. A previous study indicated that the formation of MYD88-mutant-containing aggregates coincided with that of myddosome complexes (13). According to these data, it may be that augmented MyD88 oligomerization is blocked by ST2825 in ABC DLBCL cells, resulting in the disruption of myddosome assembly.

WST-1 assays were used to evaluate the impact of ST2825-induced myddosome assembly disruption on tumor cell survival. As presented in Fig. 2, ST2825 exerted greater cytotoxicity on MYD88-mutated (OCI-LY10 and TMD8) vs. WT (SU-DHL-4) B-cells. Compared with the DMSO control groups, the cell viability of OCI-LY10 cells treated with ST2825 for 72 h was decreased in a concentration-dependent manner, with ~50% inhibition at 5 µM ST2825, ~60% at 10 µM and ~70% at 20 µM (P<0.01; Fig. 2A). Similar results were observed in TMD8 cells (P<0.01; Fig. 2B). However, the growth inhibition effects of ST2825 on SU-DHL-4 cells were only ~20-25% at 10 or 20 µM after 72 h of treatment, though these results were still significant. Flow cytometry revealed that the percentage of apoptotic cells (including early and late apoptotic cells) was significantly increased in OCI-LY10 and TMD8 cells 48 h after treatment with ST2825 (P<0.01; Figs. 2D and S1), while no significant difference was observed in SU-DHL-4 cells. Consistent with these results, the characteristic apoptotic morphological changes (condensed and fragmented nuclear) were observed by DAPI staining in OCI-LY10 and TMD8 cells treated with ST2825, while there were no obvious morphological changes in SU-DHL-4 cells (Fig. 2E).

Since the disruption of myddosome assembly inhibited survival and induced apoptosis in MYD88-L265P DLBCL cells, the present study investigated the underlying molecular mechanism responsible for this effect. Gain-of-function driver mutation L265P engages the NF-κB pathway by facilitating the phosphorylation and degradation of IκB proteins (8). Treatment of ABC DLBCL cell lines with ST2825 resulted in decreased ratios of phosho-/total IκB in both OCI-LY10 and TMD8 cells (Fig. 3A). To further investigate NF-κB activity, NF-κB reporter assays were performed. ST2825 decreased transcription of the NF-κB-dependent luciferase reporter in OCI-LY10 and TMD8 cells (Fig. 3B). In addition to the NF-κB pathway, JAK-STAT3 signaling was also demonstrated to contribute to cell survival in ABC DLBCL. MYD88 L265P mutations promote JAK-STAT3 signaling via the increased production of IL-6 and IL-10 in ABC DLBCL tumors. Furthermore, the activation of MYD88 signaling in MYD88-L265P DLBCL cells can induce IFN-β production, which may be involved in the immune modulation of the ABC DLBCL cell microenvironment (8). The present study demonstrated that treatment with ST2825 significantly reduced the secretion of IL-10 and IFN-β (Fig. 3C and D), while such inhibition was not observed for IL-6 secretion (data not shown).

Chronically active BCR signaling, responsible for the constitutive activation of the NF-κB pathway, is important for cell survival in ABC DLBCL. The BCR signaling component BTK serves a key role in triggering NF-κB activation in the majority of ABC DLBCLs (6,17). According to the sequencing results of 155 ABC DLBCL biopsy samples, Ngo et al (8) identified that the most common mutations were found in MYD88, CD79B/A, A20 and CARD11. These mutations partially overlapped, and the overlap between the MYD88 L265P and CD79B/A mutations had the highest frequency. Among cases with a MYD88 L265P mutation, 34% had a simultaneous CD79B/A mutation, which was reversed among cases with a CD79B/A mutation; even more had a simultaneous MYD88 L265P mutation (43%). These data indicate that there may be crosstalk between chronically-active BCR and MYD88 signaling (8). In a previous study, co-immunoprecipitation experiments revealed that in HEK293 cells, ectopically expressed BTK associated with MYD88 (18). In addition, it was also reported that BTK interacted with MYD88 in lipopolysaccharide-stimulated macrophages to promote the activation of MYD88-dependent pathways (19). In the present study, co-immunoprecipitation experiments where performed to investigate whether endogenous MYD88 interacts with BTK in ABC DLBCL cells, with activation of the MYD88 signaling pathway driven by the L265P mutation. Robust MYD88 co-immunoprecipitation with BTK was observed in L265P-expressing ABC DLBCL cells (OCI-LY10 and TMD8) (Fig. 4A). This binding between BTK and MYD88 was also confirmed through reverse pull-down studies (Fig. 4B). It was further demonstrated that the binding of BTK with MYD88 was inhibited following treatment with ST2825 in OCI-LY10 and TMD8 cells (Fig. 4C). In addition, treatment with ST2825 decreased the ratio of phos-/total BTK expression in OCI-LY10 and TMD8 ABC DLBCL cell lines (Fig. 3A).

MYD88 and BCR signaling often converge at IκB, leading to constitutive activation of the NF-κB pathway in ABC DLBCL cells. In addition, a recent study demonstrated that MYD88, the BCR and TLR9 form a multi-protein supercomplex, which drives pro-survival-associated NF-κB signaling in ibrutinib-responsive DLBCL cells (20). Furthermore, the present study identified the direct interaction between BTK and MYD88 in ABC DLBCL cell lines. Thus, the potential dual inhibition of MYD88 and BTK signaling and its synergistic effects on ABC DLBCL cell death were investigated. The combination of ST2825 and the BTK inhibitor ibrutinib promoted OCI-LY10 and TMD8 cell death (Figs. 5A and B, S2A and B). The inhibitory effect for each combination at different doses was visualized using a heat map. Analysis of the CI indicated that ST2825 and ibrutinib were synergistic at most doses, though greater synergistic effects were observed at higher doses of ST2825 (≥1.3 µM) for both OCI-LY10 and TMD8 cells (Figs. 5C and S2C). The enhanced cytotoxicity was associated with stronger inhibition of NF-κB activity in tumor cells treated by dual inhibition, compared with that of each drug alone (Figs. 5D and S2D).

BCL-2 upregulation is prevalent in B-cell lymphoma and promotes tumor cell survival by blocking apoptosis (21,22). In addition, high BCL-2 expression levels were associated with poor clinical outcome in patients with ABC DLBCL (23). Given that BCL-2 serves a key role in the underlying oncogenic mechanism of B-cell lymphoma, BCL-2-targeted therapy has developed rapidly in recent years (24). In the present study, combination treatment with ST2825 and the BCL-2 inhibitor ABT-199 enhanced OCI-LY10 and TMD8 cell death (Figs. 6A and B, S3A and B). Analysis of the CI indicated that ST2825 and ibrutinib were synergistic at higher doses of ST2825 (≥1.3 µM) and ABT-199 (≥0.5 µM) in OCI-LY10 cells (Fig. 5C). Similar results were revealed in TMD8 cells (Fig. S3C). This combination treatment also consistently promoted apoptosis, as indicated by the increased number of apoptotic cells (including early and late apoptotic cells) detected by flow cytometry, compared with treatment with the individual drugs alone (Figs. 6D, S3D and S4).