Selective Inhibition of Bruton's Tyrosine Kinase by a Designed Covalent Ligand Leads to Potent Therapeutic Efficacy in Blood Cancers Relative to Clinically Used Inhibitors

Abstract

Bruton's tyrosine kinase (BTK) is a member of the TEC-family kinases and crucial for the proliferation and differentiation of B-cells. We evaluated the therapeutic potential of a covalent inhibitor (JS25) with nanomolar potency against BTK and with a more desirable selectivity and inhibitory profile compared to the FDA-approved BTK inhibitors ibrutinib and acalabrutinib. Structural prediction of the BTK/JS25 complex revealed sequestration of Tyr551 that leads to BTK's inactivation. JS25 also inhibited the proliferation of myeloid and lymphoid B-cell cancer cell lines. Its therapeutic potential was further tested against ibrutinib in preclinical models of B-cell cancers. JS25 treatment induced a more pronounced cell death in a murine xenograft model of Burkitt's lymphoma, causing a 30-40% reduction of the subcutaneous tumor and an overall reduction in the percentage of metastasis and secondary tumor formation. In a patient model of diffuse large B-cell lymphoma, the drug response of JS25 was higher than that of ibrutinib, leading to a 64% "on-target" efficacy. Finally, in zebrafish patient-derived xenografts of chronic lymphocytic leukemia, JS25 was faster and more effective in decreasing tumor burden, producing superior therapeutic effects compared to ibrutinib. We expect JS25 to become therapeutically relevant as a BTK inhibitor and to find applications in the treatment of hematological cancers and other pathologies with unmet clinical treatment.

Figure 1

Putative structure of BTK covalently inhibited and cell viability assays. (a) Chemical structure of JS25 and other BTK inhibitors used. (b) The energetically best poses for BTK as determined by docking calculations. (c) Overlay of 10 frames of BTK/JS25 complex sampled from 0.5 μs MD simulations, together with the distance between the sidechain of Leu408 and the aromatic ring (Ph-SO₂Me) of JS25, and the geometry of the sidechain (χ¹ dihedral angle) of Tyr551 throughout MD simulations. (d) Overlay of 10 frames of BTK/Ibrutinib complex sampled from 0.5 μs MD simulations, together with the geometry of the sidechain (χ1 dihedral angle) of Tyr551 through MD simulations. BTK is shown as blue ribbons, and carbon atoms of the ligand and Tyr551 are shown in green and purple, respectively. (e) Cell viability of Raji, (f) DoHH-2, (g) WA-C3CD5+, (h) Mo1034, (i) MOLM-13, (j) HL-60, (k) HEK293T, (l) JURKAT, and (m) HBEC-5i. Cells were treated with serial doses of acalabrutinib, ibrutinib, and JS25 for 72 h. Error bars correspond to the standard deviation of the mean, n = 3 technical replicates. (n) Degradation analysis of BTK after treating Raji cells with JS25. (o) Translocation profile of different compounds (15 μM) at 1 h and (p) 24 h. Experiments were performed in triplicates on at least three different days using independently grown cell cultures. Error bars correspond to the standard deviation of the mean.

Figure 2

JS25 treatment inhibits the tumor growth of Burkitt’s lymphoma and induces selective ex vivo cytotoxicity in primary DLBCL samples. (a) Schematic representation of the in vivo assay. Blue arrows indicate days of treatment. (b) Tumor size and (c) body weight were monitored periodically. (d, e) Quantification and analysis of the metastases and tumor formation observed (n = 5/group), ((b) **p = 0.0018 and 0.0090, (d) **p = 0.0086, *p = 0.0418, (e) *p = 0.0386). Statistical analysis was conducted by one-way ANOVA, followed by Dunnett’s test for significance: not significant (ns) p > 0.05; *p < 0.05 (*); **p < 0.01. (f) Example of neoplastic cells observed in the liver of the control and JS25-treated groups. (g) Schematic representation of the ex vivo experiment. (h) JS25 and (i) ibrutinib/drug response score (DRS) calculated as 1-mean of the RCF. Each concentration point for each sample was performed in four replicates at 72 h incubation time point. Blue indicates DRS > 0, and white indicates DRS < 0. DRS scores > 0 indicate “on-target” cytotoxic response, and <0 indicates general cytotoxicity or “off-target” cytotoxic response.

Figure 3

Comparison of the therapeutic effects of BTK inhibitors in zebrafish patient-derived xenografts of CLL disease. (a) Representative scheme of the zPDX assay. (b) Percentage of CD19+CD5+ cells within the CD45+ population from PBMCs of each CLL patient. (c–c′) Representative zPDX confocal image on where the therapeutic effects of the different compounds were analyzed (white rectangle). (d–o) Representative confocal images for each zPDX. Percentage of zPDXs with tumor ((p) ****p < 0.0001, (r) **p = 0.0080, ****p <0.0001, (t) *p = 0.0183, **p = 0.0054, ****p < 0.0001) and tumor burden ((q) ****p <0.0001, (s) *p = 0.0188, **p = 0.0045, ****p <0.0001, (u) ****p < 0.0001). The outcomes are expressed as AVG (b, p, r, t) and AVG ± SEM (fold induction-normalized values to controls) (q, s, u). Data are from one independent experiment, and the number of xenografts analyzed for tumor burden is indicated in the representative images. The number of total zPDXs analyzed at the end of the assay to generate the tumor incidence is indicated below the respective charts. Each dot represents one zebrafish xenograft. Statistical analysis was performed using Fisher’s exact test (tumor incidence) and an unpaired test (tumor burden). Statistical results: ns > 0.05, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, and ****p ≤ 0.0001. Scale bar represents 50 μm.