The Bruton tyrosine kinase inhibitor PCI-32765 thwarts chronic lymphocytic leukemia cell survival and tissue homing in vitro and in vivo

Abstract

B-cell receptor (BCR) signaling is a critical pathway in the pathogenesis of several B-cell malignancies, including chronic lymphocytic leukemia (CLL), and can be targeted by inhibitors of BCR-associated kinases, such as Bruton tyrosine kinase (Btk). PCI-32765, a selective, irreversible Btk inhibitor, is a novel, molecularly targeted agent for patients with B-cell malignancies, and is particularly active in patients with CLL. In this study, we analyzed the mechanism of action of PCI-32765 in CLL, using in vitro and in vivo models, and performed correlative studies on specimens from patients receiving therapy with PCI-32765. PCI-32765 significantly inhibited CLL cell survival, DNA synthesis, and migration in response to tissue homing chemokines (CXCL12, CXCL13). PCI-32765 also down-regulated secretion of BCR-dependent chemokines (CCL3, CCL4) by the CLL cells, both in vitro and in vivo. In an adoptive transfer TCL1 mouse model of CLL, PCI-32765 affected disease progression. In this model, PCI-32765 caused a transient early lymphocytosis, and profoundly inhibited CLL progression, as assessed by weight, development, and extent of hepatospenomegaly, and survival. Our data demonstrate that PCI-32765 effectively inhibits CLL cell migration and survival, possibly explaining some of the characteristic clinical activity of this new targeted agent.

Figures

Figure 1

PCI-32765 inhibits anti-IgM and NLC-mediated prosurvival signals in CLL cells. (A) Contour plots of a representative CLL sample, depicting CLL cell viabilities after 24, 48, and 72 hours of incubation with anti-IgM in the presence or absence of PCI-32765, as indicated above the plots. The gates in each of the plots highlight the viable cell populations. (B-C) Bar diagrams representing the mean relative CLL cell viabilities after 24, 48, and 72 hours of anti-IgM stimulation in the presence or absence of PCI-32765 (B, n = 11) or in NLC cocultures (C, n = 15). Viabilities were normalized to the relative viability of control samples with medium (100%) to account for differences in spontaneous apoptosis in samples from different patients. Displayed are means ± SEM. *P ≤ .05.

Figure 2

3H-thymidine incorporation by CLL cells in coculture with NLCs is decreased after treatment with PCI-32765.3H-thymidine uptake in CLL cells cocultured with NLC was measured with a scintillation counter. Cells were either left untreated (control) or incubated with 2 different concentrations of PCI-32765 (0.5 or 1.0μM) for 24 hours (left panel) or 48 hours (right panel), respectively. The uptake of the tritiated nucleoside was significantly decreased in cells treated with the inhibitor. In untreated CLL cells, a slight increase between 24 and 48 hours was observed, suggesting proliferation in a small subset of CLL cells cocultured with NLC. In CLL cells treated with PCI-32765 this proliferation was not observed. Shown are mean values ± SEM of 9 patients. *P ≤ .05.

Figure 3

PCI-32765 down-regulates CCL3 and CCL4 secretion in vitro (in NLC cocultures) and in vivo in plasma samples of CLL patients undergoing therapy with PCI-32765. (A) Incubation with PCI-32765 significantly inhibits the secretion of CCL3 and CCL4 in CLL-NLC cocultures. The bars in this graph represent mean ± SEM CCL3 (left graph) and CCL4 (right graph) concentrations in supernatants of CLL-NLC cocultures from 11 different CLL samples in pg/mL. *P ≤ .05. (B) CCL3 (left graph) and CCL4 (right graph) plasma concentrations in CLL patients before and during therapy with PCI-32765. Displayed are individual plasma concentrations, indicated by the dots, and mean ± SD plasma concentrations, as indicated by the horizontal lines. CCL3 and CCL4 plasma concentrations were measured at the time points indicated on the horizontal axes. * indicates significant changes with P ≤ .05.

Figure 4

Inhibition of CLL cell chemotaxis and actin polymerization. (A) Displayed is the mean ± SEM relative migration of CLL cells from 8 different patients toward CXCL12 (left graph) and CXCL13 (right graph) in the presence or absence (medium control) of different concentrations of PCI-32765, as indicated on the horizontal axes. White bars depict background migration toward wells without chemokine. (B) Displayed is the relative F actin content of CLL cells after stimulation with CXCL12 staining with FITC-labeled phalloidin in the presence or absence of PCI-32765 (or AMD3100 as control) at the time points indicated on the horizontal axis. * indicates P ≤ .05 compared with the control with chemokine; medium.

Figure 5

Abrogation of signaling downstream of the BCRs and of CXCR4 and CXCR5 receptors. Displayed are immunoblots from CLL cells from 1 representative patient of 9 patients, that were either unstimulated or stimulated for 10 minutes with anti-IgM (αIgM, left row), CXCL12 (middle row), or CXCL13 (right row) in the presence or absence of PCI-32765 as indicated. P indicates immunoblotting for the active, phosphorylated form, whereas T represents the nonphosphorylated, total amount of the respective proteins.

Figure 6

Delayed CLL disease progression by PCI-32765 in TCL1 model. (A) Peripheral blood (100 μL) from mice treated with control vehicle (n = 4), 2.5 (n = 3), or 25 mg/kg/d (n = 4) PCI-32765 were incubated with B220 and CD5 antibodies. Numbers of CLL cells were determined using calibrated counting beads. Results shown are the mean CLL cell counts ± SD with RMANOVA statistic analysis on control versus treated mice. *P < .01, ***P < .0001. (B) All 11 mice were killed 16 days after treatment. Top: example of each group of mice after 16 days of treatment. Spleen (middle) and liver (bottom) tissue sections were assessed histologically after H&E staining. Less lymphocyte infiltration was observed in 25 mg/kg/d PCI-32765 treated mice compared with control mice. (C) Spleens from control or PCI-32765 treated mice were compared for size and weight. The right panel shows the mean spleen weight ± SD with the unpaired t test on control versus treated mice. *P < .05. (D) Spleen cells from mice treated for 16 days with vehicle or PCI-32765 were left untreated (−) or treated with 25μM pervanadate (PVD) for 4 minutes (+). Intracellular staining of phosphorylated PLCγ2 (pPLCγ2) was then performed and analyzed by flow cytometry on gated B220+CD5+ cells. Results are the mean fluorescence intensity (MFI) of phosphorylated-PLCγ2 ± SD with the unpaired t test on control versus treated mice. *P < .05.