The Relevance of Aurora Kinase Inhibition in Hematological Malignancies
In 2017 alone, nearly 24.5 million new cases of cancer were diagnosed worldwide, and more than 9.5 million people died from the disease. Cancer can be described as a heterogeneous set of diseases that share the main clinical characteristic of uncontrolled cell proliferation in the form of a solid tumor or as individual cells in the bloodstream (1,2).
Even though cancer is characterized as a multifactorial disease, the accumulation of genetic mutations, as well as the deregulation of the cellular cycle, are both well-established occurrences in cancerous cells. Failure in the regulation of the cell cycle is associated with malignant neoplasms since it is fundamental for genome maintenance, repairing DNA damage and, consequently, preventing the proliferation of cells with malignant potential (3,4).
As one of the main factors in cell-cycle control, the activity of protein kinases is to be highlighted due to their vital roles in several phases of cell division. In recent years, it has been demonstrated that alterations in cellular expression levels of many kinases are associated with malignant cancer phenotypes, pointing them out as possible therapeutic targets for these diseases. One such kinase family of possible clinical relevance are the aurora kinases (5,6).
Role of Aurora Kinases in Cell Division
The aurora kinases are a family of serine/threonine protein kinases that play a central role in eukaryotic cell division. Humans express three different types of aurora kinases: aurora kinase A (AURKA), aurora kinase B (AURKB) and aurora kinase C (AURKC). AURKA and AURKB are the most relevant to mitosis, being expressed in most somatic cell types, while AURKC is mostly expressed in germ cells due to its role in meiosis regulation (7,8).
From the first reports of them being essential in cell division to the recent years of thorough molecular studies, the auroras have been described as oncogenes relevant to many human malignancies (9,10). Their gene amplification and overexpression is detected particularly in solid tumors, however, their correlation with leukemia development has also been reported (10,11).
The family name as aurora was conceived when it was first discovered that aurora-mutated
Prior to bipolar spindle formation, from late S phase to pro-metaphase, AURKA is mainly concentrated at centrosomes (13,14) due to an increase in cyclin-dependent kinase 11 (CDK11) activity (15). AURKA is then a part of important processes for centrosome maturation such as pericentriolar matrix recruitment and microtubule nucleation and stabilization (16,17).
The role of AURKA in maturation occurs through phosphorylation loops and protein-protein interactions with many different kinases. One such pathway is AURKA-mediated phosphorylation of a large tumor suppressor 2 (LATS2) and ajuba LIM protein (AJUBA) complex which acts on an auto-phosphorylation loop that directs AURKA to the centrosomes and induces the recruitment of a centrosomal pool of γ-tubulin, an essential step for microtubule nucleation and later spindle assembly (17-20).
During metaphase and throughout mitosis, the bipolar spindle assembly begins and AURKA also localizes to spindle microtubules, as well as the centrosomes (21,22). AURKA localization to spindle microtubules is dependent on activity of TPX2 microtubule nucleation factor (TPX2) which binds to and activates AURKA at the spindles, inhibiting dephosphorylation by protein phosphatase 1 (PP1) (22-24).
Aurora-related proteins have been shown to phosphorylate kinesin family member 11 (EG5), an important microtubule-motor protein responsible for centrosome separation, in the
Besides its role in cellular spatial organization, AURKA also exerts a series of nuclear functions related to the control of mitotic checkpoint G2/M (
Along with the regulatory proteins inner centromere protein, survivin and borealin, AURKB forms the chromosomal passenger complex (CPC) and takes part in regulating crucial pathways for chromosome condensation, sister chromatid separation and cytokinesis (30).
During the G2 phase, at mitosis onset, AURKB is responsible for phosphorylation of histone H3 (Figure 2), an event that coincides with heterochromatin condensation and proper chromosome formation (31-33).
After chromosome condensation, during prophase in early mitosis, the activation of phosphorylated histones H3 and H2A recruits AURKB to the centromeres (35-37). It has been suggested that recruitment of AURKB happens in a two-step process where H2A phosphorylation by BUB1 mitotic checkpoint serine/threonine protein kinase (BUB1) firstly creates a pool of kinetochore-bound AURKB and then H3 phosphorylation by haspin kinase translocates this AURKB pool to the inner centromere, where it is mainly located during prophase (38).
Until the onset of anaphase, AURKB phosphorylation of the KMN network, an association of the conserved kinetochore proteins kinetochore null protein 1 (KNL1), mis-segregation 12 (MIS12) and nuclear division cycle 80 (NDC80), takes part in regulating kinetochore-microtubule attachment stabilization (39,40). While during early mitosis the attachment turnover is high, ensuring quickly destabilization of incorrect attachments, it tends to slow down alongside mitosis progression after correct and stable attachments are formed and the spindle assembly checkpoint is satisfied, allowing for increased activity of anaphase-promoting complex/cyclosome and proper chromosomal separation in anaphase (38,40-42).
AURKB also exerts major roles in cytokinesis, from initial anaphase through telophase and the end of mitosis, being relocated from inner centromeres to the spindle midzone by the action of mitotic kinesin MKLP2 and then interacting with kinesins KIF4A and KIF2A to promote microtubule bundling and consequent central spindle formation, as well as proper cleavage furrow completion (43-45). It has been shown that after chromosomaI segregation, the activity of AURKB at the spindle midzone functions as a checkpoint for the regulation of incorrectly segregated chromatin, inhibiting the formation of tetraploid daughter cells and protecting lagging chromosomes from breakage (45,46).
The activity and metabolic role of AURKC is still poorly understood when compared to the other two aurora kinases expressed by humans, being mostly described by its overlapping functions with AURKB (47,48). Although poorly expressed in most somatic cells, high AURKC expression in germ cells and its ability to interact with chromosomal passenger complex proteins has led to the understanding that AURKC exerts many functions that overlap with activities of AURKB in meiotic cells, being responsible for the regulation of kinetochore-microtubule attachment and proper sister chromatid segregation (48,49).
AURKC has also been observed to be highly expressed in pre-implantation embryo cells, being able to carry out normal cell division in AURKB-null mice with death occurring only after implantation takes place, making it clear that AURKC may be relevant in the division of somatic cells (50,51). While the non-overlapping functions of AURKB and AURKC are not fully elucidated, both kinases are known to form distinct complexes with chromosomal passenger complex proteins and act through distinct mechanisms, while still being able to interfere with each other’s activities, as shown by experiments of selective inhibition of one or both kinases (52).
Aurora Family as Targets in Oncohematological Therapy
The finding of overexpression of aurora kinases in cancer and their role as major regulators of the cell cycle quickly led to the idea that their inhibition might be a potential pathway when treating oncological patients. Over the past couple of decades, the quest to design and test molecules capable of inhibiting aurora activities fueled many pre-clinical and clinical studies, resulting in the development of a series of pan-aurora inhibitors as well as selective inhibitors of one of the three human kinases (53). While some aurora kinase inhibitors (AKIs) demonstrate promising results as single treatments, it is also important to perceive their potential as synergistic compounds, being able to restore sensitivity to chemotherapy agents in tumor cells and interact with other targeted therapies for increased efficacy (54).
In this scenario, hematological malignant disorders as leukemia presents overexpression of AURK, and the use of AKIs seems to be an effective pharmacological approach to treating these diseases (14).
AURKA and AURKB have an important role in cell-cycle progression, in fact, the transcription of
Most studies evaluating the mechanism of AKI action in hematologic cell lines focus on acute myeloid leukemia (AML). This neoplasia is characterized by clonal expansion of undifferentiated myeloid precursors, leading to impaired hematopoiesis and bone marrow failure (73). In
Internal tandem duplication of FMS-like receptor tyrosine kinase 3 (
Translocation t(8;21) is frequent in 4-8% of patients with AML and results in the chimeric protein RUNX family transcription factor 1-RUNX1 partner transcriptional co-repressor 1 (
Acute lymphoblastic leukemia (ALL) is the most common leukemia diagnosed in adults and the second most frequent acute subtype in children worldwide, this neoplasia presents chromosomal and genetic alterations related to the differentiation and proliferation of T and B precursor cells (79,80). AURKA and AURKB are overexpressed in samples from patients with ALL (55,65,81) and some research described the mechanism of action of AKI in ALL cell lines alone or in combination with other antineoplastics agents.
Genetic alterations as t(4;11) with mixed lineage leukemia (MLL) fusion–associated gene
The Philadelphia chromosome (Ph+) is characterized by the translocation between chromosomes 9 and 22, generating the chimeric gene BCR–ABL1 (85,86). Although this translocation is more described in chronic myeloid leukemia (CML), the presence of Ph affects around 3-5% of children and 25% of adults with ALL, and is related to a poor prognosis (87-89). The chimeric
Due to the efficacy of AKIs in leukemia cell lines
CML is basically, but not only, characterized by the presence of Ph+ cells, since around 95% of cases present this translocation (94,95). Thus, the malignant potential of CML cells is related to the
Lymphomas are a group of malignant diseases that can be derived from constituent cells of the lymphoid tissue, lymphocytes and histiocytes. Malignant lymphomas are divided into Hodgkin’s and non-Hodgkin’s lymphoma (NHL) and are more common in the head and neck region, but NHL can also be found in extranodal regions (25% of the cases), with or without lymph node involvement. NHL is more frequent than Hodgkin’s disease and corresponds to around 90% of head and neck malignancies (105,106). Of NHL subtypes, follicular and diffuse large B-cell lymphomas are the most frequently diagnosed, corresponding to about 20% and 30%, respectively. Other subtypes account for fewer than 10% of the cases (107).
Studies have shown that AURKA is overexpressed in several NHL subtypes when compared to normal tissues and B-cells, Chowdhury
Moreover, the pan-AKI AT9283 exhibited highly cytotoxic effects in NHL cells and, when used in combination with docetaxel, enhanced apoptosis, as well as survival rates and reduced tumor volume in the mouse mantle cell lymphoma xenograft model (68). The synergistic effect of AURK inhibition was observed with other antitumoral agents that target the cell cycle and microtubular constituents. The disruption in replication machinery leads to selective cancer cell death, enhancing the antitumoral potential of AKI in oncohematological treatment (108,109).
In all clinical trials presented, the most prevalent therapeutic option was alisertib (113,114,118,120,122,123), a selective inhibitor of AURKA that is also under investigation for relevance in the treatment of non-hematological malignancies (125). While it was most effective when treating patients with AML, this finding may also be correlated to its use as a synergistic compound alongside conventional induction chemotherapy (113,114). A study cohort by Goldberg
While most investigated drugs were AURKA-selective inhibitors or pan-AKIs, AURKB was selectively targeted by the use of barasertib in two studies (119,124). Although its efficacy was demonstrated in the treatment of patients with diffuse large B-cell lymphoma, the low progression-free survival of 60 days attested to by Collins
Other inhibitors that were reported in only one study include AMG 900, AT9283, ENMD-2076 and MK-0457. These drugs had only modest to no efficacy when utilized as single agents and in low, tolerable doses, although ENMD-2076 monotherapy still achieved a 25% overall response rate when treating patients with AML and CML, and should be taken into account when considering possible synergistic interactions (115-117,121).
Overall, the usage of AKIs did not cause unexpected or unmanageable adverse effects in patients and the main ones were related to blood or gastrointestinal disorders (115,117,119-121,123), which is in accordance with the expected effects of aurora inhibition in non-malignant rapidly dividing cells due to mitosis impairment (127,128).
The interest in understanding the involvement of aurora in biological pathways in non-malignant as well as in malignant cells has only grown over the years. While there is still much to be elucidated, it is clear that these kinases and their upstream and downstream regulators play a key role in mitosis and oncogenesis, motivating their investigation as potential targets for oncological treatments. Even though the clinical trials using AKIs as a monotherapy for hematological disorders have not shown great results, the manageable toxicity and potentially synergistic effects still render them a focus of interest for future investigations in combinatorial clinical trials.
Conflicts of Interest
The Authors declare no conflicts of interest regarding this study.
Machado CB, da Silva EL and Moreira-Nunes CA, performed the study design; Machado CB and Nogueira BMD, prepared the figures; Machado CB, da Silva EL, Nogueira BMD, da Silva JBS, Moraes-Filho MO, Moraes MEA, Montenegro RC and Moreira-Nunes CA wrote the article. All Authors read and approved the final article.
This study was supported by Brazilian funding agencies National Counsel of Technological and Scientific Development (CNPq; to ELS, RCM, MEAM, MOMF and CAMN).