Cyclin-Dependent Kinase Inhibitor P1446A Induces Apoptosis in a JNK/p38 MAPK-Dependent Manner in Chronic Lymphocytic Leukemia B-Cells

Abstract

CDK (cyclin-dependent kinase) inhibitors have shown remarkable activity in CLL, where its efficacy has been linked to inhibition of the transcriptional CDKs (7 and 9) and deregulation of RNA polymerase and short-lived pro-survival proteins such as MCL1. Furthermore, ER (endoplasmic reticulum) stress has been implicated in CDK inhibition in CLL. Here we conducted a pre-clinical study of a novel orally active kinase inhibitor P1446A in CLL B-cells. P1446A inhibited CDKs at nanomolar concentrations and induced rapid apoptosis of CLL cells in vitro, irrespective of chromosomal abnormalities or IGHV mutational status. Apoptosis preceded inactivation of RNA polymerase, and was accompanied by phosphorylation of stress kinases JNK (c-Jun N-terminal kinase) and p38 MAPK (mitogen-activated protein kinase). Pharmacologic inhibitors of JNK/p38 MAPK conferred protection from P1446A-mediated apoptosis. Treatment with P1446A led to a dramatic induction of NOXA in a JNK-dependent manner, and sensitized CLL cells to ABT-737, a BH3-mimetic. We observed concurrent activation of apoptosis stress-inducing kinase 1 (ASK1) and its interaction with inositol-requiring enzyme 1 (IRE1) and tumor necrosis factor receptor-associated factor 2 (TRAF2) in CLL cells treated with P1446A, providing insights into upstream regulation of JNK in this setting. Consistent with previous reports on limited functionality of ER stress mechanism in CLL cells, treatment with P1446A failed to induce an extensive unfolded protein response. This study provides rationale for additional investigations of P1446A in CLL.

Fig 1. CDK inhibitor P1446A induces apoptosis of the CLL B-cells independent of IGHV mutational status and cytogenetics.

(A) PBMC's from patients with CLL (N = 10) were incubated with 0.05–5 μM P1446A or vehicle control for 24 hours and assayed for apoptosis. Here and subsequently, apoptosis was determined by Annexin V staining within the CD19⁺ subset of cells. Horizontal lines represent the mean. (B) CLL B-cells were incubated with 0.05–1.5 μM P1446A for 24 hours. IC₅₀ value was determined for CLL samples with mutated (N = 13) and unmutated IGHV (N = 10). (C) CLL B-cells (N = 62) were incubated with 1.5 μM P1446A for 24 hours. Cytogenetics markers were determined by fluorescent in situ hybridization [no marker, 11q, trisomy 12, 13q, 17p]. Horizontal lines represent the mean. (D) PBMCs from patients with CLL (N = 37) or healthy volunteers (N = 6) were incubated with 1.5 μM P1446A or vehicle control for 24 hours. Since we noted a significant variation in baseline apoptosis between patient samples, normalization to the time-matched untreated controls was performed to more clearly reflect the drug-induced apoptosis. *- p<0.05 compared to untreated control.

Fig 2. P1446A inhibits interphase CDKs in CLL B-cells.

(A-C) CLL cells were incubated with P1446A at the indicated concentrations for 4 or 8 hours. As a control, cells were incubated with dinaciclib for 4 hours. Whole-cell protein lysates were subjected to immunoblotting. Representative images and densitometry data from four individual patients are shown. *- p<0.05 compared to untreated control. (D) CLL cells from 6 individual patients were incubated with 0–1.5 μM P1446A for 6 hours. Total RNA was isolated from CD19⁺ CLL B-cells, reverse-transcribed and subjected to real-time PCR with the indicated probes (in duplicates). Results were normalized to 18S levels. Data are the mean ± SE.

Fig 3. P1446A induces JNK and p38 MAPK in CLL cells.

(A-D) CLL cells obtained from four individual patients were incubated with P1446A (0.5–1.5 μM) for 4 or 8 hours. Whole-cell protein lysates were subjected to immunoblotting. Representative images and densitometry data are shown. For densitometry, a fold increase in expression of pATF2 (C) and pJNK/cPARP (D) referenced to total levels of protein at each dose and time point were normalized to their expression in untreated samples (E) CLL cells obtained from six individual patients were incubated with 20 μM JNK inhibitor VIII, 20 μM SP600125, 10 μM SB203580 or vehicle control for 1 hour and then with 1 μM P1446A for 24 hours. Apoptosis was determined by Annexin V and 7-AAD staining within the CD19⁺ subset of cells. *–p<0.05; **—p<0.01 vs. control. (F) CLL cells were transfected with siRNA's against ASK1 or control siRNA using Amaxa program X-05 and subsequently incubated with 1 μM P1446A or vehicle control for 24 h. Whole-cell lysates were subjected to immunoblotting. A representative blot of four independent experiments is shown, along with densitometry data (*- p<0.05 vs. control). (G) CLL cells were incubated with 1 μM P1446A or vehicle control for 6 hours. Protein lysates were incubated with antibody against ASK1 or isotype control (not shown) and subjected to immunoblotting. A representative blot of three independent experiments is shown.

Fig 4. The UPR in response to P1446A treatment is limited.

(A, B) CLL cells obtained from four individual patients were incubated with P1446A (0.5–1.5 μM) for the indicated time intervals. Whole-cell protein lysates were subjected to immunoblotting. Representative images from 1 of 3 independent experiments are shown. (C) Cells were incubated with the indicated concentration of P1446A or 1 μM thapsigargin (thap) for 6 hours, total RNA was isolated, reverse-transcribed and subjected to PCR using XBP1-specific primers.

Fig 5. P1446A induces NOXA and cooperates with ABT-737 to reverse apoptosis resistance.

(A-B) CLL cells from 6 individual patients were incubated with the indicated doses of P1446A for 6 hours with or without 20 μM SP600125. Total RNA was isolated from CD19⁺ CLL B-cells, reverse-transcribed and subjected to real-time PCR with the indicated probes (in duplicates). Results were normalized to 18S levels. Data are the mean ± SE. (C) CLL cells (N = 3) were incubated with 20 μM SP600125 or vehicle control for 1 hour and then with 1 μM P1446A for 6 hours. Whole-cell protein lysates were subjected to immunoblotting. (D) CLL cells (N = 6) were co-cultured with CD40L-expressing stroma for 24 hours and subsequently incubated with the indicated doses of P1446A and ABT-737 or vehicle control for 24 hours. Apoptosis was determined by Annexin V and 7-AAD staining within the CD19⁺ subset of cells. **—p<0.01 vs. control.