Abstract
Introduction: While in CLL cells phosphorylation of STAT3 on serine 727 residues is constitutive, phosphorylation of STAT3 on tyrosine 705 residues is inducible. Cytokines, such as IL-6, or IgM antibodies that activate CLL cells' BCR, induce tyrosine phosphorylated (p) STAT3. However, whereas IL-6 induces tyrosine pSTAT3 phosphorylation within 15 minutes, IgM induces pSTAT3 within ≥ 2-4 hours. The reason for the delayed IgM-induced phosphorylation is unknown. Like STAT3, the transcription factor NF-κB is constitutively activated in CLL cells and stimulation of the BCR activates NF-κB. Whether BCR stimulation upsurges NF-κB's transcriptional activity has not been elucidated. Because IL-6 is an NF-κB-target gene and, like IL-6, IgM antibodies induce tyrosine pSTAT3, we wondered whether prolonged stimulation with IgM antibodies induces tyrosine pSTAT3 via NF-κB-mediated induction of IL-6 in CLL cells.
Methods: We incubated peripheral blood CLL cells in the presence or absence of IgM antibodies or IL-6, and harvested the cells at different time points. Total RNA was extracted using TRIzol (Life technology), cDNA was synthesized with Super Script First synthesis System for RT-PCR (Invitrogen), and NF-κB-target gene expression was quantified using RT-PCR (Invitrogen Life Sciences). To measure the levels of tyrosine pSTAT3 we used flow cytometry and to assess binding of NF-κB (p65) to DNA we utilized an electromobility shift assay (EMSA) using an NF-κB-binding site labelled DNA probe.
Results: The transcriptional activity of NF-κB was studied using a PCR array that profiles the expression of 83 NF-κB-target genes. To reduce the 'noise' from stochastic variability in gene expression we first identified a core of genes that are expressed in cells from all patients' samples. To that aim we ranked the Ct values in each array and considered all genes that were amplified earlier than the cycle in the 75th percentile. Using this approach we identified 35 genes (42% of genes represented in the array) that were amplified in all 6 patients' samples. Annotation analysis revealed that the key pathways common to these 35 genes included 'Positive regulation of the NF-κB cascade', 'Inflammation' and 'Negative regulation of apoptosis'. Applying stringent criteria we identified 5 genes common to all cases that were amplified prior to the cycle representing the 25th percentile. Most amplified genes detected in all samples prior to stimulation (28/35, 80%) were also detected after 4 h of IgM stimulation, confirming that NF-κB is constitutively activated in CLL cells. However, 19 addition genes (19/83, 23%of the genes in the array) were detected in all IgM-stimulated but not in unstimulated cells. Remarkably, IL-6 was detected in all cases only after IgM stimulation. Furthermore, the delta-delta Ct method identified an IgM-induced time-dependent increment in IL-6 and IL-8, suggesting that IL-6 expression is dependent on stimulation of the BCR. Indeed IL-6 neutralizing antibodies significantly reduced the levels of tyrosine pSTAT3 in CLL cells incubated for 18 h with IgM antibodies. In addition, EMSA studies using CLL cells from 4 different patients showed that stimulation of the BCR with IgM antibodies increased the binding of NF-κB to DNA in a time-dependent manner. Moreover, the JAK2 inhibitor Ruxolitinib attenuated the NF-κB-DNA binding, suggesting that long exposure to IgM antibodies induces activation of NF-κB, a process mediated in part by IL-6 that activates the JAK2/STAT3 pathway.
Conclusions: The BCR of CLL cells is stimulated in the bone marrow and lymph nodes. However, whereas the immediate effects of BCR stimulation have been excessively studied, the successive effect BCR stimulation is poorly understood. We found that stimulation of the BCR induces tyrosine phosphorylation of STAT3 via NF-κB-mediated induction of IL-6, a process that requires protracted BCR stimulation. Although NF-κB is constitutively activated in CLL cells, continuous activation of the BCR further activates NF-κB. Continuous stimulation of the BCR increases the levels of IL-6 that, upon binding to its receptor, activates STAT3 that in turn activates NF-κB. Taken together, our data suggest that agents, such as Ruxolitinib, that inhibit the successive effects of BCR activation, would become effective therapeutic agents in CLL.
Disclosures
Rozovski: Novartis: Other: Advisory board. Wierda:Glaxo-Smith-Kline Inc.: Research Funding; Celgene Corp.: Consultancy.
The PRDM14–CtBP1/2–PRC2 complex regulates transcriptional repression during the transition from primed to naïve pluripotency