Acute lymphoblastic leukemia (ALL) is the most common hematological malignancy in children. Although risk-adaptive therapy, CNS-directed chemotherapy and supportive care have improved the survival of ALL patients, disease relapse is still the leading cause of cancer-related death in children. Therefore, new drugs or novel multi-drug combinations are needed as frontline treatments for high-risk patients and as salvage agents for relapsed disease. T-cell ALL (T-ALL) is a subset of ALL that exhibits activating mutations of NOTCH1 in more than 50% of the patients. However, the use of gamma-secretase inhibitors to reduce NOTCH1 activity has not been successful in patients due to limited response and toxicity. Therefore, identification of genetic factors that cooperate with T-ALL leukemogenesis is needed for the development of alternative therapies.

KLF4 is a transcription factor that functions as a tumor suppressor or an oncogene depending on cellular context. Our data showed significant reduction of KLF4 transcripts in lymphoblasts from T-ALL patients compared to blood and bone marrow cells from healthy individuals. In consistent with reduced KLF4 levels, these patients exhibit hyper-methylation of CpG islands located between nt -811 and +1190 relative to KLF4 transcription start site. From these findings we hypothesized that KLF4 has tumor suppressor function in T-ALL leukemogenesis. To test our hypothesis, we transduced 5-FU treated bone marrow (BM) cells from control (Klf4fl/fl), Klf4 null (Klf4fl/fl; Vav-iCre) and Klf4 heterozygous (Klf4fl/+; Vav-iCre) mice with retrovirus carrying a NOTCH1 activating mutant (L1601P-ΔP) and then transplanted these BM cells into irradiated recipient mice. In contrast to controls, mice transplanted with transduced Klf4-null BM cells developed T-ALL with significantly higher penetrance (Klf4 null 76.5% v.s. control 21.3%) and shorter latency (Klf4 null 93 days v.s. control 130 days). Interestingly, Klf4 heterozygous group shows similar survival kinetics as Klf4 null group, suggesting that Klf4 haploinsufficiency is enough to accelerate onset of leukemia. To investigate the effect of Klf4 deletion in established leukemia cells, we transplanted NOTCH1 L1601P-ΔP transduced BM cells from Klf4fl/fl; CreER+ mice to induce leukemia. Post-transplantation deletion of the Klf4 gene by tamoxifen administration was able to accelerate T-ALL development compared to mice injected with vehicle. On the cellular level, loss of KLF4 led to increased proliferation of leukemia cells as assessed by in vivo BrdU incorporation, which correlated with decreased levels of p21 protein. Limited dilution transplantation of primary leukemia cells into secondary recipients showed a 9-fold increase of leukemia initiating cells (LIC) frequency in Klf4null leukemia cells compared to controls, suggesting that KLF4 controls expansion of LIC in T-ALL.

To elucidate molecular mechanism underlying KLF4 regulation in T-ALL cells, we performed microarray and ChIP-Seq in control and Klf4 null CD4+CD8+ leukemia cells. Combined analyses revealed 202 genes as KLF4 direct targets, of which 11 genes are also deregulated in human T-ALL cells by comparing with published microarray datasets. One of the top upregulated genes is Map2k7, which encodes a kinase upstream of the JNK pathway. Immunoblots in leukemia cells confirmed increased expression of MAP2K7 protein and enhanced phosphorylation of its downstream targets JNK and ATF2. To further investigate the role of JNK pathway in T-ALL, we tested JNK inhibitor SP600125 in human T-ALL cell lines (KOPTK1, DND41, CCRF-CEM, MOLT3). Interestingly, SP600125 showed dose-dependent cytotoxicity in all human T-ALL cell lines tested regardless of their NOTCH1 status.

Overall our results showed for the first time that KLF4 functions as a tumor suppressor in T-ALL by regulating proliferation of leukemia cells and frequency of LIC. Additional study elucidated that KLF4 suppresses the JNK pathway via direct transcriptional regulation of MAP2K7. Moreover, the vulnerability of human T-ALL cell lines to JNK inhibition provides a novel target for future therapy in T-ALL patients.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution