RPMI 1640, fetal bovine serum (FBS), 0.4% trypan blue vital stain, and an antibiotic–antimycotic mixture were obtained from Life Technologies (Carlsbad, CA). Rabbit anti-phospho-BTK (#5082), rabbit anti-BTK (#8547), rabbit anti-phospho-PLCγ2 (#3871), rabbit anti-PLCγ2 (#3872) were obtained from Cell Signaling Technology (Danvers, MA). Antibodies against GAPDH, Hoechst 33342, 3-(4,5- dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), Goat anti-rabbit and goat anti-mouse horseradish peroxidase conjugates were purchased from GE Healthcare (Piscataway, NJ) and goat anti-rabbit Alexa Fluor 594 was purchased from Molecular Probes (Eugene, OR). The ADP-Glo Kinase assay kit along with BTK enzyme was procured from Promega Corporation (Madison, WI, United States). Cell viability dye Viva Fix 498/521 (#1351115) and mouse anti-human CD138 Alexa Fluor 647 (#MCA2459A647) were obtained from Bio-Rad (Hercules, California). CD19 (#130-113-646), CD27 (#130-113-629), CD38(#130-113-429), and CD138 (#130-051-301) magnetic beads were procured from Miltenyi Biotec (Auburn, California). The TaqMan primers specific for GAPDH (Hs02786624_g1), BTK (Hs00975865_m1), Nanog (Hs02387400_g1), Gli1 (Hs00171790_m1), and Myc (Hs00153408_m1) were obtained from Life Technologies (Carlsbad, CA). KS151 was synthesized and characterized at Penn State College of Medicine, Hershey, PA. A 25 mmol/L solution of KS151 was prepared in dimethyl sulfoxide, stored as small aliquots at –20 °C, and then diluted as needed in cell culture medium.

Human MM cells MM.1R, MM. 1S, U266, RPMI 8226 cells, and BM stromal cells (BMSCs) were purchased from the American Type Culture Collection (ATCC) (Manassas, VA). Bortezomib resistant cells MM.1S BTZ-R and RPMI8226 BTZ-R were a generous gift from Dr. Nathan Dolloff, The Medical University of South Carolina, Charleston, SC, USA. MM.1R, MM. 1S, and RPMI 8226 cells were cultured in RPMI 1640 growth media (Cellgro, Manassas, VA) supplemented with heat-inactivated 10% Fetal Bovine Serum (Sigma, St. Louis, MO), 100 I.U./ml penicillin, and 100 μg/ml streptomycin (Cellgro, Manassas, VA). U266 cells were cultured in RPMI 1640 with 15% FBS. BMSCs were procured from ATCC and cultured in Dulbecco’s Modified Eagle Medium (DMEM) (ATCC) supplemented with heat-inactivated 10% Fetal Bovine Serum (Sigma), 100 I.U./ml penicillin, and 100 μg/ml streptomycin (Cellgro, Manassas, VA). All cells were cultured in a 37°C and 5% CO₂ incubator.

CD138 ⁻ cells were isolated by two methods using the S3e™ Bio-Rad Cell Sorter and magnetic beads. MM.1R, and RPMI 8226 cells were washed and resuspended in cold PBS with 3% FBS. Then cells were stained with cell viability dye VivaFix498/521 and mouse anti-human CD138 Alexa Fluor 647 and incubated in the dark on ice for 25 min. After washing twice, cells were sorted using S3e™ Bio-Rad following the manufacturer’s protocol (Itoua Maïga et al., 2014). Alternatively, CD138 magnetic beads were used to isolate CD138 ⁺ cells. Human MM cells MM.1R and RPMI 8226 cells were fractioned by a MACS separator, using CD138 magnetic beads. After washing cells from the media, cells were resuspended in 0.5% MACS buffer and incubated with CD138 microbeads at 4°C for 15 min. After washing, CD138⁻ cells were isolated using an MS column according to the manufacturer’s protocol (Pandey et al., 2017). Similarly, CD19⁺CD138 ⁻ , and CD19⁺CD27⁺, CD38 ⁻ cells were isolated using S3e™ Bio-Rad Cell Sorter following the manufacturer’s protocol. Briefly, after suctioning out media, cells were washed with PBS that contains 3% FBS. Cells were incubated with respective antibodies in the dark on ice for 25 min and then washed twice with PBS and sorted using S3e™ Bio-Rad.

The three-dimensional crystal structure of the human Bruton’s tyrosine kinase hBTK (PDB: 3GEN) receptor was retrieved from the Protein Data Bank and used as a receptor for docking (Marcotte et al., 2009). In order to prepare the protein for docking, protein structure was prepared by stripping off water molecules and other bound heteroatoms from structure coordinates file using PyMOL software and hydrogen atoms were added into the receptor by using MGL Tools and converted in PDBQT format for further analysis (Schrödinger and DeLano, 2020; Trott and Olson, 2010). Chem Draw 20.0 was used to create the 3D structures of analogs in mol2 format, which was then converted into protein data bank (PDB) format using PyMOL software.

Molecular docking aids in predicting a ligand’s preferred binding pose with a protein and analyzing inhibitory interactions between them. Here, a molecular docking studies were performed using Autodock Vina (Trott and Olson, 2010). A variety of searches were conducted in order to provide the best possible conformation. The docking grid size in X, Y, Z dimensions was set at size_x = 50.00 A⁰, size_y = 50.00 A⁰ and size_z = 50.00 A⁰, respectively and centered at center_ x = -17.03, center_y = 10.28 and center_ z = 21.22 with a 0.375 A⁰ grid spacing and exhaustiveness were set to 24. KS151 was prepared by setting the active torsions and detecting the root of the molecules to adjust according to the defined pocket of the target protein. The docking studies were carried out with rigid protein and the molecular docking outcomes were analyzed based on binding energy (kcal/mol) and binding interactions profile. The schematic diagrams of protein ligand interactions were generated using UCSF Chimera (https://doi.org/10.1002/jcc.20084) and analyzed using Ligplot + v.2.1 (Laskowski and Swindells, 2011).

The MD simulations were carried out up to 20 ns to analyze the stability of the hBTK-KS151 complex using GROMACS 2018.1 running on an Intel(R) Core (TM) i5-9400 CPU machine installed with Ubuntu 18.04 Linux. In all runs, the GROMOS43a1 force field was used (Van Der Spoel et al., 2005). The PRODRG server was used to generate ligand topology files (Schüttelkopf and van Aalten, 2004). Before simulation, both the hBTK and hBTK-KS151 complexes were solvated in the TIP3P model in a triclinic box with a minimum spacing of 2.5 Å distance between any protein atoms to the closest box edge. To neutralize the system Na⁺ ions were added, and 50,000 steps of steepest descent energy minimization were performed. Systems were equilibrated using NVT and NPT ensembles for 100 ps at 300K and 1 bar of pressure using a modified Berendsen thermostat (tau_T = 0.1 ps) and a Parrinello–Rahman barostat (tau_p = 2 ps). The final simulation run lasted for 20 ns, and the dynamic trajectories of the root mean square deviation (RMSD), the root mean square fluctuation (RMSF), radius of gyration (R_g), solvent accessible surface area (SASA), hydrogen bonding, and projections of C_α atoms were visualized by the XMGRACE software package (Turner, 2005).

The chemicals and solvents were purchased from commercial vendors. Reactions were carried out using dried glassware under an atmosphere of nitrogen. Reaction progress was monitored with analytical thin-layer chromatography (TLC) on aluminum backed precoated silica gel 60 F254 plates (E. Merck). Column chromatography was carried out using silica gel 60 (230-400 mesh, E. Merck) with the solvent system indicated in the procedures. Nuclear magnetic resonance (NMR) spectra were recorded using a Bruker Avance II 500 spectrometer.

KS151 was synthesized as shown in Figure 2. Briefly, to a solution of 4-iodobenzaldehyde (1) (7.5 mmol) in dioxane (30 ml) was added 2,3-dichlorophenol (2) (7.5 mmol), cuprous iodide (0.75 mmol), N, N-dimethylglycine hydrochloride (2.25 mmol), and cesium carbonate (15 mmol). The mixture was stirred at 105 °C in a nitrogen atmosphere for 18 h. The mixture was concentrated and extracted with ethyl acetate, washed with water, and dried over MgSO₄. The crude product was purified by a silica gel column using hexane and ethyl acetate (9:1) as the eluent to yield the intermediate (3) (yield 82%). Compound (3) (3.37 mmol) was refluxed with 5-Nitro-2-oxindole (3.37 mmol) and piperidine in ethanol for 8 h and filtered to obtain the product (4) (yield 86%). Compound (4) (1.75 mmol) was dissolved in anhydrous DMF (30 ml), and K₂CO₃ (2.2 mmol) was added and stirred for 1 h. 1,4-Bis(bromomethyl)benzene (7 mmol) was added slowly with constant stirring until compound 4 had been consumed (monitored by TLC). The reaction mixture was poured into cold water and extracted with ethyl acetate. The ethyl acetate layer was washed with water and brine and dried over MgSO4. The solvent was removed by rotavaporation, and the crude product was purified by silica gel column chromatography using hexanes/ethyl acetate (80:20) as the eluent to yield the intermediate (5) as yellow crystals (yield 79%). To compound 5 (1.04 mmol), thiourea (1.0 mmol) and ethanol (25 ml) were added and heated to reflux until the starting material had been consumed (monitored by TLC). The solvent was removed under a vacuum. The final product 4-((3-(4-(2,3-dichlorophenoxy) benzylidene)-5-nitro-2-oxoindolin-1-yl) methyl) benzyl carbamimidothioate hydrobromide (6) (KS151) was recrystallized in ethanol-ethyl acetate to afford KS151 (yield 70%). The identity of KS151 was confirmed by nuclear magnetic resonance and mass spectral analysis, and purity (>99%) was quantified by high-performance liquid chromatography analysis. Yellow solid, yield: 70%; ¹H NMR (500 MHz, DMSO-d6): δ 9.17 (s, 2H), 8.98 (s, 2H), 8.8 (d, J = 2.5 Hz, 1H), 8.6 (d, J = 9.0 Hz, 1H), 8.41 (d, J = 2.0 Hz, 1H), 8.32 (m, 4H), 8.01 (s, 1H), 7.89 (d, J = 8 Hz, 1H), 7.12–7.60 (m, 6H), 5.10 (s, 2H), 4.45 (s, 2H); MS (ESI) m/z 687 [M + H].

The assay was performed as described earlier (Pandey et al., 2017). Various concentrations of KS151 (0.5–1 μM) were used in the reaction. Briefly, various amounts of KS151 were diluted in kinase assay buffer consisting of BTK enzyme, substrate, and adenosine triphosphate (ATP) (total volume, 5 ml). The assay was performed in a 96-well plate. The reaction mixture was incubated at room temperature for 1 h. After incubation, 5 µL of ADP-Glo reagent was added, and the reaction was incubated for an additional 40 min at room temperature. Adding 10 µL of kinase detecting reagent, followed by 30 min of room temperature incubation, terminated the reaction. Finally, luminescence was detected using an Infinite M200 Pro multi plate reader (Tecan, Switzerland). Kinase activity was normalized by vehicle control (dimethylsulfoxide, DMSO) and a graph was performed using Prism version 5 software (GraphPad, La Jolla, CA).

RNA was extracted from cells using an RNA extraction kit (Invitrogen) and converted to cDNA using an Invitrogen kit (#12183018A) as per the manufacturer’s instructions (Pandey et al., 2017). RNA was quantified using an ND-1000 spectrophotometer (Nanodrop Technologies, Wilmington, DE). Reverse transcription (RT) was done to obtain complementary DNA, and samples were diluted to prepare for qPCR as previously described (Pandey et al., 2017; Elbezanti et al., 2020). Quantitative polymerase chain reaction (qPCR) experiments were performed using an AB Applied Biosystems Step One Pulse Real-Time PCR System and TaqMan universal PCR Master Mix (Applied Biosystems, Foster City, CA). The housekeeping gene GAPDH was used for all delta-delta Ct calculations as described earlier (Ding et al., 2010).

Human MM cells were treated with KS151 for 24 h. After incubation, cells were washed and lysed using RIPA buffer supplemented with protease and phosphatase inhibitor cocktails (Joglekar et al., 2015). After incubation on ice for 1h, cell extracts were vortexed and the cell lysate was centrifuged at 15,000 rpm for 10 min. The supernatant was kept at –80°C until further use. Protein was quantified using a BCA Protein Assay kit (Pierce Scientific, Rockford, IL). Heat denatured protein samples were separated using a 4-12% NuPAGE gel (Thermo Scientific, Rockford, IL, United States). Proteins were transferred to PVDF membranes at 30 V for 90 min. After blocking with 5% non-fat dry milk for 1h, membranes were incubated overnight at 4°C with primary antibodies. After washing, the secondary antibody was added and incubated for 3 h at room temperature. After washing, the membrane was exposed to an enhanced chemiluminescent substrate for detection of horseradish peroxidase (HRP) (PierceTM ECL Western; Thermo Scientific. Rockford, IL, United States). Bands were visualized and quantitated using the Chemi DocTM MP imaging system (Bio-Rad, Hercules, CA).

The effect of KS151 and the other drugs on cell viability was determined using an MTT assay by a method described earlier (Pandey et al., 2017). Briefly, 5,000 cells of human MM were plated in triplicate in 96-well plates and treated with various treatments for 72 h (Pandey et al., 2017). After incubation for 72h, 20 µL of a freshly prepared MTT solution (5 mg/ml) was added and incubated for 3 h at 37°C. The 96-well plates were centrifuged, the supernatant was removed, and 100 µL of dimethyl sulfoxide (DMSO) was added to each well to dissolve the formazan crystals. The optical density of the resulting solution was measured by delta value (570-630 nm) using a 96-well multi-mode microplate reader (BioTek Technologies, Winooski, VT, United States).

To study apoptosis, a Muse^® Cell Analyzer (Luminex Corporation, Austin) was used. MM cells were treated with either KS151 or in combination with other drugs for 24 h before subjected to a Luminex Annexin V Live Dead Assay (Luminex Corporation, Austin). Annexin V live dead assay was performed essentially following the manufacturer’s protocol. Briefly, cells were plated at a density of 2 × 10⁵ per well and then treated and incubated for 24 h. Then, cells were scraped and mixed with Annexin V stain (Luminex Corporation) and incubated for 20 min. The Muse Cell Analyzer was used to analyze the data. The lower left quadrant [annexin V negative and 7-amino-actinomycin negative (7-AAD)] reflects the live cells, while the lower right quadrant (annexin V+ and 7-AAD ⁻ ) contains early apoptotic cells. The late apoptotic and dead cells are in the upper right quadrant (annexin V⁺ and 7-AAD⁺). The upper left quadrant represents the dead cells (annexin V and 7-AAD⁺).

Cells were plated in 12 well plates and treated with 5 µM of KS151 for 24 h. Then 100 µL of cell suspension was loaded into the funnel that is used for the Cytospin. Cells were centrifuged for 3 min at 800 rpm, followed by fixing with and 4% formaldehyde for 15 min. After washing three times with TBST, the cells were incubated with 0.2% Triton while rocking for 18 min. Then, after 1 h of blocking with 3% BSA, 1:200 primary antibody was added in 3% BSA overnight at 4°C. After washing three times with phosphate buffered saline (PBS), secondary antibody, goat anti-rabbit IgG (H + L) Alexa Fluor ™ Plus 488 (Red A32731, Invitrogen) was added at a 1:200 dilution and kept rocking for 1 h at room temperature. Slides were mounted with ProLong ™ Diamond Antifade Mount with DAPI (#P36971, Invitrogen, Carlsbad, CA) and covered with a coverslip, then imaged with the Nikon A1R GaAsP Laser Scanning Confocal Microscope (Nikon Eclipse Ti inverted using a 60x objective). The Integrated density of cells was calculated using the ImageJ software as previously described (Leite et al., 2020).

A Co-culture of MM cells with bone marrow stromal cells (BMSCs) was performed by a method described before (Bar-Natan et al., 2017). In brief, 2 × 10⁵ BMSCs were seeded in six well plates overnight. After incubation, 2 × 10⁵ MM cell lines were plated on top of BMSCs and treated with 5 μM of KS151 for 24 h. The cells were collected using a scraper, and RNA was harvested, and qPCR was performed as described above.

Statistical significance of the results was analyzed by either unpaired two-tailed Student’s t-test, or one-way analysis of variance (ANOVA), or two-way ANOVA using GraphPad Prism software. Each graph represents the average of at least three replicates with error bars on the graph representing standard deviation* p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001.

First, we performed in silico analysis using the three-dimensional crystal structure of the human BTK (PDB: 3GEN) retrieved from the Protein Data Bank (Marcotte et al., 2009). The phenoxyphenyl group of isatin of KS99 was overlaid with the same fragment of ibrutinib, which completely entered the hydrophobic pocket and formed a face-to-edge π-stacking interaction with surrounding amino acid residues. The central core of ibrutinib with 4-aminopyrazolo [3,4-days] pyrimidine maintains the potency of inhibiting BTK; we selected isatin as the new scaffold to replace the pharmacophore of ibrutinib by employing the theory of bioisostere. In silico analysis, illustrated in Figure 1B, revealed KS151 docked with hBTK, showed potential binding against hBTK with a docking score of -10.3 kcal/mol. KS151 forms backbone peptide bonds of residues Met477 and Ala478 at a distance of 3.20 A⁰ and 3.33 A⁰, respectively. Amino acid residues, Leu408, Val416, Met449, Leu528, and Phe540 were involved in hydrophobic interactions while Thr474, Tyr476, Asn526 and Asp539 are in close proximity to create hydrophilic-environment within the active site pocket. The 1,4-bis (bromomethyl) benzene was selected as a linker to connect the scaffold and isothiourea warheads.1,4-bis (bromomethyl) benzene was used as a linker because it provides better stability in the active site pocket.

To further probe the stability and dynamics of the hBTK-KS151 complex in active site pocket, a molecular dynamics simulation calculation was carried out for 20 ns. The trajectories of hBTK and hBTK-KS151 complex were studied to understand the flexibility and stability of the ligand complex. Root mean square displacements (RRMSD), root mean square fluctuations (RMSF), radius of gyration (R_g), and hydrogen bond contact were evaluated as a function of time. The average RMSD values of hBTK alone and the KS151 complex are 0.23 nm, 0.37 nm respectively. During the MD simulation, the overall residual motions of the docked complex were computed. The RMS fluctuations for the amino acid residues of the hBTK and KS151 complex are in the same pattern without any major deviations. Due to ligand binding, it is not highly reflective. The RMSF plot indicated stable interactions in the binding pocket throughout the simulations. The R_g of a protein is a measure of its compactness. If a protein is stably folded, its R_g value is likely to remain stable. If a protein unfolds, its R_g will change over time. Radius of gyration values for the complex were 1.84 nm compared to 1.86 nm of hBTK indicating a gain in compactness upon ligand binding. The average SASA value of the complex was estimated at 121.9 compared to 129.7 for hBTK alone. The intermolecular H-bonding between KS151 and hBTK is critical for the complex’s stability. Hydrogen bonding interactions between KS151 and various residues in binding pocket ranges from three to four at different time intervals of the KS151 simulated MD trajectories indicating stable interaction patterns. The results of these analyses are summarized in Figures 1C–H.

We synthesized sixteen analogs as shown in Supplementary Figure S1 by following the scheme shown in Figure 2. All analogs were purified by silica gel column and purity was determined by analytical high-pressure liquid chromatography followed by NMR and mass spectrometry. The efficacies of 16 analogs were tested against BTK by performing an in vitro kinase assay. Most analogs did not show any significant effect on the kinase activities except analog 16 (KS151), thus we identified KS151 as a lead molecule. Indeed, KS151 inhibited 44 and 49.8% of kinase activity at 0.5 and 1 μM respectively, whereas ibrutinib inhibited 25.9% and 32.2% of kinase activity at 0.5 and 1 μM respectively (Figure 3A). This suggests that KS151 is more potent than ibrutinib. A broader range of drug concentration was tested as shown in Supplementary Figure S2A. We next determined whether KS151 is a selective BTK inhibitor. We tested the specificity of KS151 by performing a semi-quantitative, sandwich-based array following the manufacturer’s protocol (RayBiotech Life, Peachtree Corners, GA, United States) for detecting 71 tyrosine kinase phosphorylation signatures. We were able to detect only 23 kinases in the human MM cell line U266. KS151 had either minor inhibitory or inducible effect on several kinases such as ABL1, AXL, EphA6, EphA7, EphA8, NGFR, and JAK2, whereas the maximum inhibitory effect of KS151 was observed on BTK (Figure 3B). Furthermore, KS151 had no significant effects on rest of the detectable kinases (Supplementary Figure S2B). This suggests that KS151 selectively targets BTK and has little or no effect on the cell kinases that were tested.

To choose the best cell line model to study the effect of KS151, we tested various MM cell lines for BTK expression. We extracted proteins from untreated U266, MM. 1R, MM. 1S, and RPMI 8226 MM cell lines and investigated BTK expression using western blot (Figure 4A). We found that RPMI 8226 has the highest level of endogenous BTK followed by MM.1S, MM.1R, and with U266 having the least expression. The expression of the mRNA of BTK was also assessed in each cell line. Similar to the western blot results, we found that RPMI 8226 has the highest BTK expression followed by MM.1S, and MM. 1R (Figure 4B). We further investigated whether KS151 inhibits the phosphorylation of BTK and its downstream target PLCγ2. The autophosphorylation of BTK at Y223 was inhibited by KS151 in a dose-dependent manner. Furthermore, KS151 inhibited phospho-PLCγ2 without affecting the PLCγ2 (Figure 4C). To examine whether KS151 has a cytotoxic effect on BTK, we tested the effect of KS151 for shorter period of time on lipopolysaccharide (LPS) induced BTK expression. KS151 did not change the BTK expression of unstimulated U266 cells, however, it was able to inhibit LPS induced BTK to the basal level (Supplementary Figure S3). We also used confocal microscopy to test the effect of KS151 on phospho-BTK. We stained fixed MM cells for phospho-BTK after treating cells with 5 µM KS151 for 24 h. The total signal intensity was calculated. We found that KS151-treated cells had a lower integrated density of phospho-BTK compared to the untreated control (Figures 4D,E). We calculated the median reading of integrated density. Interestingly, 5 µM of KS151 was able to inhibit phospho-BTK by 56.48% in the RPMI 8226 cell line. Interestingly, treatment of KS151 inhibits the nuclear translocation of BTK. It is important to mention that BTK translocation in different compartments has been observed (Mohamed et al., 2000; Webb et al., 2000).

Once the specificity and functionality of KS151 was established, we tested its cytotoxic potential and compared it with existing chemotherapeutic agents via the MTT assay (Pandey et al., 2017). MM cell lines MM.1S, MM.1R, U266, and RPMI 8226 were treated with various concentrations of KS151 (0.1–25 μM) for 72 h. An MTT cell viability assay confirmed a dose-dependent decrease in cell viability of MM cell lines by KS151, which inhibited the growth of MM.1S, MM.1R, U266, and RPMI 8226 with an IC₅₀ value of 12.07, 8.4, 15.48, and 9.07 μM, respectively (Figure 5A). Furthermore, we compared its effect to chemotherapeutic compounds that are FDA-approved to treat MM. We compared KS151 to different classes of chemotherapeutic agents such as BTK inhibitors (acalabrutinib, and ibrutinib), alkylating agents (cyclophosphamide, and melphalan), immunomodulatory agents (lenalidomide), histone deacetylase (HDAC) inhibitor (panobinostat), and plant alkaloid “topoisomerase II inhibitor” (etoposide). Interestingly, we found KS151 more potent than most of these medications in MM.1R and RPMI 8226 MM cell lines (Figure 5B). Furthermore, we tested KS151 cytotoxicity in normal dermal fibroblast using MTT assay. KS151 showed no toxic effect towards non-malignant cells (Supplementary Figure S4). Because the acquired resistance against bortezomib is the main reason for relapse in MM, we further tested the extent to which KS151 is effective in bortezomib-resistant cells. We compared the efficacy of KS151 with other BTK inhibitors (acalabrutinib, and ibrutinib) in bortezomib resistant cell lines. MM.1S-bortezomib-resistant (MM.1S Btz-R), and RPMI 8226- bortezomib-resistant (RPMI 8226-Btz-R) cell lines were treated with various concentrations of KS151 (0.1–25 μM) for 72 h. As reflected in Figure 5C, KS151 was found to be more effective in these resistant cells than acalabrutinib, and ibrutinib. The IC₅₀ value of KS151 was 7.38 μM, and 5.13 μM in MM.1S Btz-R, and RPMI 8226-Btz-R cell lines respectively, whereas in comparison, IC₅₀ was much higher for ibrutinib (16.7 μM, 15.28 μM) and acalabrutinib (more than 25 μM) in both cell lines (Figure 5C). To investigate whether the decrease in MM cell viability after KS151 treatment was due to apoptosis induction, we assessed the apoptosis-inducing potential of KS151 in MM cells by Annexin V + 7AAD staining. Various MM cells were treated with increasing concentrations of KS151 for 24 h, and apoptotic assay was performed using the Muse Cell Analyzer. KS151 induced the apoptosis in a dose-dependent manner in various MM cell lines (Figure 5D-E). The level of mRNA and protein expression of BTK was assessed in various bortezomib-resistant cells, surprisingly in our hand both mRNA and protein level were less in bortezomib-resistant cells as compared to naïve cells, whether bortezomib resistance plays any role in the BTK expression we did not determine and it is worth exploring (Supplementary Figure S5A,B).

Because bortezomib, lenalidomide, and panobinostat are widely used to treat MM (Buda et al., 2021), we first investigated the extent to which KS151 augments the efficacy of therapeutic drugs in MM cell lines. We assessed the apoptosis-inducing potential of KS151 in combination with bortezomib, lenalidomide, and panobinostat on MM cells by Annexin V + 7AAD staining. As shown in Figures 6A,B the addition of KS151 to bortezomib had a significant additive effect in MM.1R and RPMI 8226 cells that contain a higher expression of BTK suggesting that BTK inhibition is required to improve the therapeutic outcome of bortezomib treatment. As reflected from Figures 6A,B, although current treatments may kill the tumor bulk, they can leave behind therapy-resistant stem cell-like population, which serves as a reservoir for disease recurrence (Chim et al., 2018; Papadas and Asimakopoulos, 2018). Therefore, approaches to target stem cell-like cells to prevent cancer resistance are important for managing MM (Franqui-Machin et al., 2015; Papadas and Asimakopoulos, 2018). Our goal was to test whether KS151 compromises the survival of stem cell-like cells. We isolated CD138⁻ cells (which possess stem cell-like properties as discussed above) from human MM.1R, and the RPMI 8226 cell line and treated them with KS151. We found that KS151 induced the apoptosis in both CD138⁻ MM.1R, and CD138⁻ RPMI 8226 cells (Supplementary Figure S6A). We further tested whether a combination of KS151 with bortezomib is an effective strategy against CD138⁻ cells. We treated CD138⁻ MM.1R, and CD138⁻ RPMI 8226 cells with 5 µM of KS151. The combination of KS151 further potentiates the response of bortezomib suggesting that KS151-based treatment combination is effective against stem cell-like cells, which are the main source of MM relapse (Figures 6C,D). Our molecule is not only specific to the CD138⁻ population, also effective in killing other stem cell populations such as CD19⁺CD138⁻ (Matsui et al., 2004; Kellner et al., 2013), and, CD19⁺CD27⁺ CD38⁻ population (Rasmussen et al., 2004). As shown in Figures 6E,F, KS151 has an additive effect on these populations when treated along with bortezomib. Furthermore, KS151 also sensitizes MM.1S Btz-R and RPMI 8226-Btz-R cells to bortezomib when cells are treated in combination with KS151(Supplementary Figure S6C). We also tested the combination of KS151 with lenalidomide, and panobinostat that are other widely used to treat MM. Similarly, KS151 potentiated the effect of lenalidomide and panobinostat in human MM.1R, and the RPMI 8226 cell line, as well as CD138⁻ MM.1R, and CD138⁻ RPMI 8226 cells (Supplementary Figure S7A-D).

BTK regulates the expression of stem genes (OCT4, SOX2, Nanog, and MYC) (Yang et al., 2015; Zhao et al., 2017). Given our results in Figures 6C–E, we examined if KS151 impacts the expression of stem genes in stem cell-like cells (CD138⁻). Therefore, we investigated whether KS151 can also inhibit stemness markers in CD138⁻ MM.1R and CD138⁻ RPMI 8226 cells. CD138⁻ cells were isolated from MM.1R, and RPMI8226 using magnetic beads. Then, cells were treated with 5 μM of KS151. After 24 h, RNA was isolated from cells, and RT reaction and qPCR were performed. We found that KS151 inhibited the stemness markers effectively in both stem cell-like cells. The stemness marker Gli1, Myc, and SOX2 were inhibited by 25.6, 74.9., and 31.6% respectively in CD138⁻ MM.1R cells (Figure 7A). Similarly, the stemness marker Gli1, Myc, and SOX2 were inhibited by 44, 32, and 13.9% respectively in CD138⁻ RPMI 8226 cells (Figure 7B). Altogether, the results in Figures 7A-B suggest that KS151 targets MM cells with stem cell-like properties. It has been demonstrated that BM microenvironment (BMM) regulates the stemness of MM cells via BTK signaling pathways (Zhao et al., 2017). To examine the efficacy of KS151 in the relevant microenvironmental context, we tested the efficacy of KS151 when MM cells co-cultured with BMSCs. We co-cultured MM.1R and RPMI 8226 cells on top of BMSCs in a 1:1 ratio. Then, cells were treated with 5 μM of KS151. After 24 h, RNA was isolated from cells and RT-qPCR was performed using specific primers for Nanog, and Gli1. As shown in Figure 7 C-D, KS151 was able to suppress stemness markers even in the presence of BMSCs. Nanog and Gli1 were inhibited 67.48%, and 46.2% in MM.1R co-culture, respectively. Likewise, Nanog and Gli1 were inhibited by 24.8%, and 56.5%, respectively, in RPMI 8226 and BMSCs co-culture, suggesting that novel BTK inhibitor regulates the stem cell-like phenotypes by modulating the microenvironment (Figures 7C,D ).