Tsz Kan Fung

Cyclin A in cell cycle control and cancer

Abstract. Cyclin A is particularly interesting among the cyclin family because it can activate two different cyclin-dependent kinases (CDKs) and functions in both S phase and mitosis. An embryonic form of cyclin A that is only essential for spermatogenesis is also present in some organisms. In S phase, phosphorylation of components of the DNA replication machinery such as CDC6 by cyclin A-CDK is believed to be important for initiation of DNA replication and to restrict the initiation to only once per cell cycle. In mitosis, the precise role of cyclin A is still obscure, but it may contribute to the control of cyclin B stability. Cyclin A starts to accumulate during S phase and is abruptly destroyed before metaphase. The synthesis of cyclin A is mainly controlled at the transcription level, involving E2F and other transcription factors. Removal of cyclin A is carried out by ubiquitin-mediated proteolysis, but whether the same anaphase-promoting complex/cyclosome targeting subunits are used as for cyclin B is debatable. Consistent with its role as a key cell cycle regulator, expression of cyclin A is found to be elevated in a variety of tumors.

Sorafenib treatment of FLT3-ITD+ acute myeloid leukemia: favorable initial outcome and mechanisms of subsequent nonresponsiveness associated with the emergence of a D835 mutation

Internal tandem duplication (ITD) of the fms-related tyrosine kinase-3 (FLT3) gene occurs in 30% of acute myeloid leukemias (AMLs) and confers a poor prognosis. Thirteen relapsed or chemo-refractory FLT3-ITD(+) AML patients were treated with sorafenib (200-400 mg twice daily). Twelve patients showed clearance or near clearance of bone marrow myeloblasts after 27 (range 21-84) days with evidence of differentiation of leukemia cells. The sorafenib response was lost in most patients after 72 (range 54-287) days but the FLT3 and downstream effectors remained suppressed. Gene expression profiling showed that leukemia cells that have become sorafenib resistant expressed several genes including ALDH1A1, JAK3, and MMP15, whose functions were unknown in AML. Nonobese diabetic/severe combined immunodeficiency mice transplanted with leukemia cells from patients before and during sorafenib resistance recapitulated the clinical results. Both ITD and tyrosine kinase domain mutations at D835 were identified in leukemia initiating cells (LICs) from samples before sorafenib treatment. LICs bearing the D835 mutant have expanded during sorafenib treatment and dominated during the subsequent clinical resistance. These results suggest that sorafenib have selected more aggressive sorafenib-resistant subclones carrying both FLT3-ITD and D835 mutations, and might provide important leads to further improvement of treatment outcome with FLT3 inhibitors.

Cyclin F Is Degraded during G2-M by Mechanisms Fundamentally Different from Other Cyclins

Cyclin F, a cyclin that can form SCF complexes and bind to cyclin B, oscillates in the cell cycle with a pattern similar to cyclin A and cyclin B. Ectopic expression of cyclin F arrests the cell cycle in G2/M. How the level of cyclin F is regulated during the cell cycle is completely obscure. Here we show that, similar to cyclin A, cyclin F is degraded when the spindle assembly checkpoint is activated and accumulates when the DNA damage checkpoint is activated. Cyclin F is a very unstable protein throughout much of the cell cycle. Unlike other cyclins, degradation of cyclin F is independent of ubiquitination and proteasome-mediated pathways. Interestingly, proteolysis of cyclin F is likely to involve metalloproteases. Rapid destruction of cyclin F does not require the N-terminal F-box motif but requires the COOH-terminal PEST sequences. The PEST region alone is sufficient to interfere with the degradation of cyclin F and confer instability when fused to cyclin A. These data show that although cyclin F is degraded at similar time as the mitotic cyclins, the underlying mechanisms are entirely distinct.

Ubiquitination of p53 at multiple sites in the DNA-binding domain.

The tumor suppressor p53 is negatively regulated by the ubiquitin ligase MDM2. The MDM2 recognition site is at the NH2-terminal region of p53, but the positions of the actual ubiquitination acceptor sites are less well defined. Lysine residues at the COOH-terminal region of p53 are implicated as sites for ubiquitination and other post-translational modifications. Unexpectedly, we found that substitution of the COOH-terminal lysine residues did not diminish MDM2-mediated ubiquitination. Ubiquitination was not abolished even after the entire COOH-terminal regulatory region was removed. Using a method involving in vitro proteolytic cleavage at specific sites after ubiquitination, we found that p53 was ubiquitinated at the NH2-terminal portion of the protein. The lysine residue within the transactivation domain is probably not essential for ubiquitination, as substitution with an arginine did not affect MDM2 binding or ubiquitination. In contrast, several conserved lysine residues in the DNA-binding domain are critical for p53 ubiquitination. Removal of the DNA-binding domain reduced ubiquitination and increased the stability of p53. These data provide evidence that in addition to the COOH-terminal residues, p53 may also be ubiquitinated at sites in the DNA-binding domain. (Mol Cancer Res 2006;4(1)15–25)

Synthetic lethal targeting of oncogenic transcription factors in acute leukemia by PARP inhibitors.

Acute myeloid leukemia (AML) is mostly driven by oncogenic transcription factors, which have been classically viewed as intractable targets using small-molecule inhibitor approaches. Here we demonstrate that AML driven by repressive transcription factors, including AML1-ETO (encoded by the fusion oncogene RUNX1-RUNX1T1) and PML-RARα fusion oncoproteins (encoded by PML-RARA) are extremely sensitive to poly (ADP-ribose) polymerase (PARP) inhibition, in part owing to their suppressed expression of key homologous recombination (HR)-associated genes and their compromised DNA-damage response (DDR). In contrast, leukemia driven by mixed-lineage leukemia (MLL, encoded by KMT2A) fusions with dominant transactivation ability is proficient in DDR and insensitive to PARP inhibition. Intriguingly, genetic or pharmacological inhibition of an MLL downstream target, HOXA9, which activates expression of various HR-associated genes, impairs DDR and sensitizes MLL leukemia to PARP inhibitors (PARPis). Conversely, HOXA9 overexpression confers PARPi resistance to AML1-ETO and PML-RARα transformed cells. Together, these studies describe a potential utility of PARPi-induced synthetic lethality for leukemia treatment and reveal a novel molecular mechanism governing PARPi sensitivity in AML.

The N-terminal regulatory domain of cyclin A contains redundant ubiquitination targeting sequences and acceptor sites

Cyclin A is destroyed during mitosis by the ubiquitin-proteasome system. Like cyclin B, a destruction box (D-box) motif is required for the destruction of cyclin A. However, Cyclin A degradation is more complicated than cyclin B because cyclin A’s D-box motif is more extensive and proteolysis involves complex signaling in some organisms. In this study, we found that in addition to the D-box, the region between residues 123-157 also contributed to the ubiquitination and degradation of human cyclin A. Indeed, removal of the bulk of the N-terminal regulatory domain was needed to completely stabilize cyclin A and eliminate ubiquitination. A putative second RxxL motif around residue 138 played only a minor role in cyclin A degradation. To distinguish between sequences recognized by the ubiquitination machinery and the ubiquitin acceptor sites per se, we utilized a novel approach involving in vitro cleavage of cyclin A after ubiquitination. We found that several lysine residues proximal to the D-box (Lys37, Lys54, and Lys68) were ubiquitin acceptor sites. Cyclin A lacking the three lysine residues was degraded slower than the wild-type protein. Although these lysines were normally used, ubiquitination could shift to other cryptic sites when the preferred sites were unavailable, suggesting the exact positions of the ubiquitin chains also contributed to degradation. Together, these data revealed that ubiquitination does not occur randomly on cyclin A and open up questions on the precise function of the D-box.

Targeting Aberrant Epigenetic Networks Mediated by PRMT1 and KDM4C in Acute Myeloid Leukemia

Summary Transcriptional deregulation plays a major role in acute myeloid leukemia, and therefore identification of epigenetic modifying enzymes essential for the maintenance of oncogenic transcription programs holds the key to better understanding of the biology and designing effective therapeutic strategies for the disease. Here we provide experimental evidence for the functional involvement and therapeutic potential of targeting PRMT1, an H4R3 methyltransferase, in various MLL and non-MLL leukemias. PRMT1 is necessary but not sufficient for leukemic transformation, which requires co-recruitment of KDM4C, an H3K9 demethylase, by chimeric transcription factors to mediate epigenetic reprogramming. Pharmacological inhibition of KDM4C/PRMT1 suppresses transcription and transformation ability of MLL fusions and MOZ-TIF2, revealing a tractable aberrant epigenetic circuitry mediated by KDM4C and PRMT1 in acute leukemia.

Biophysical characterization of hematopoietic cells from normal and leukemic sources with distinct primitiveness

This letter reported the biophysical characterization of immunophenotypically distinct hematopoietic cells from normal and leukemic sources, through manipulation with optical tweezers at single cell level. The results show that the percentage of cells that are stretchable and their deformability are significantly higher in the more primitive cell populations. This study provides the evidence that normal and leukemic hematopoietic cell populations with distinct primitiveness exhibit differential biophysical properties. These findings raise a hypothesis that the high deformability may be related to the unique functions and activities of primitive hematopoietic cells.

Selective treatment of mixed-lineage leukemia leukemic stem cells through targeting glycogen synthase kinase 3 and the canonical Wnt/β-catenin pathway.

Purpose of reviewLeukemia carrying mutation of the mixed-lineage leukemia (MLL) gene is particularly refractory to current treatment, and is associated with frequent relapse. We will review the biology of MLL leukemia, and explore the potential of targeting multiple signaling pathways deregulated in MLL leukemic stem cells (LSCs). Recent findingsGlycogen synthase kinase 3 (GSK3) plays a critical role in mediating Hox/MEIS1 transcriptional program and its inhibition shows promise in suppressing leukemia carrying MLL fusions or aberrant Hox expression. However, recent evidence indicates that GSK3 inhibition can be overcome by hyperactivation of the canonical Wnt signaling pathway in MLL LSCs, whereas suppression of &bgr;-catenin resensitizes MLL LSCs to the GSK3 inhibitor treatment. These results suggest a differential GSK3 dependence in different subsets of leukemic populations during disease development. SummaryOn the basis of the results from preclinical model studies, a combination treatment targeting both GSK3 and the canonical Wnt signaling pathway emerges as a promising avenue to eradicate MLL LSCs. Future effort in identifying the key tractable components along these signaling pathways will be critical for the development of effective inhibitors to target this aggressive disease.

Sorafenib treatment of FLT3-ITD+ acute myeloid leukemia: favorable initial outcome and mechanisms of subsequent non-responsiveness associated with a D835 mutation

Internal tandem duplication (ITD) of the fms-related tyrosine kinase-3 (FLT3) gene occurs in 30% of acute myeloid leukemias (AMLs) and confers a poor prognosis. Thirteen relapsed or chemo-refractory FLT3-ITD(+) AML patients were treated with sorafenib (200-400 mg twice daily). Twelve patients showed clearance or near clearance of bone marrow myeloblasts after 27 (range 21-84) days with evidence of differentiation of leukemia cells. The sorafenib response was lost in most patients after 72 (range 54-287) days but the FLT3 and downstream effectors remained suppressed. Gene expression profiling showed that leukemia cells that have become sorafenib resistant expressed several genes including ALDH1A1, JAK3, and MMP15, whose functions were unknown in AML. Nonobese diabetic/severe combined immunodeficiency mice transplanted with leukemia cells from patients before and during sorafenib resistance recapitulated the clinical results. Both ITD and tyrosine kinase domain mutations at D835 were identified in leukemia initiating cells (LICs) from samples before sorafenib treatment. LICs bearing the D835 mutant have expanded during sorafenib treatment and dominated during the subsequent clinical resistance. These results suggest that sorafenib have selected more aggressive sorafenib-resistant subclones carrying both FLT3-ITD and D835 mutations, and might provide important leads to further improvement of treatment outcome with FLT3 inhibitors.