NEKs, NIMA-Related Kinases
The NIMA-related kinase, or “Nek,” family constitutes approximately 2% of all human kinases. They are related in sequence, as their name suggests, to NIMA (699 residues, 80 kDa), a serine/threonine protein kinase present in the filamentous fungus, Aspergillus nidulans. Ron Morris identified the gene, nimA, through analysis of a temperature-sensitive loss-of-function mutant that was never in mitosis (nim) when cells were incubated at the restrictive temperature (Morris 1975). Loss of NIMA activity led to G2 arrest, while overexpression of NIMA drove cells into a premature mitosis from any point in the cell cycle (Oakley and Morris 1983; Osmani et al. 1988, 1991). Mechanistically, NIMA is likely to regulate multiple aspects of mitotic entry, with evidence that its activity is required for nuclear pore disassembly, relocalization of the master regulator, cdc2-cyclin B, to the nucleus and spindle pole body (SPB), and chromatin condensation. NIMA is subsequently degraded in an anaphase-promoting complex/cyclosome (APC/C)-dependent manner and this is necessary for mitotic exit (reviewed in O’Connell et al. 2003; O’Regan et al. 2007).
Aspergillus cells are syncytial and undertake a semi-closed mitosis. It may therefore have evolved control mechanisms that are specific for this form of cell division. However, an early suggestion that NIMA-related kinases might be conserved in other eukaryotes came from expressing Aspergillus NIMA in cells from diverse species, including humans, and observing cell cycle defects (Lu and Hunter 1995; O’Connell et al. 1994). With the complete sequencing of many genomes, it is now clear that kinases related to NIMA by sequence are indeed present in most eukaryotes (Parker et al. 2007). However, so far, only the NIM-1 protein from the highly related filamentous fungus, Neurospora crassa, has been proven to be a functional homologue of NIMA capable of rescuing an Aspergillus nimA mutant suggesting the possibility of functional divergence. Indeed, the budding yeast, Saccharomyces cerevisiae, and fission yeast, Schizosaccharomyces pombe, each have one NIMA-related kinase in their genome, Kin3 and Fin1, respectively, but these are not essential for mitotic entry. However, careful studies revealed that Fin1 does contribute to the timing of mitotic onset through regulating the localization of the polo-like kinase, Plo1, to the SPB, which in turn promotes activation of cdc2-cyclin B (Grallert and Hagan 2002). Fin1 also contributes to mitotic spindle formation and mitotic exit (Grallert and Hagan 2002; Grallert et al. 2004). Thus, Neks from different species do play roles in mitotic progression.
Surprisingly, some lower eukaryotes have many genes encoding Nek kinases. For example, the unicellular organisms Chlamydomonas and Tetrahymena have 10 and 39 Nek genes, respectively. The key to this expansion of Nek genes appears to lie in an alternative non-mitotic function for Nek kinases, that is, in ciliogenesis. Chlamydomonas produce two elongated cilia, or flagella, that allow the organism to swim in response to environmental stimuli. To date, only two of the Chlamydomonas Nek genes have been studied in depth, Fa2p and Cnk2p, but loss of either protein affects both flagella disassembly and cell cycle progression (Bradley and Quarmby 2005; Mahjoub et al. 2002). Meanwhile, Tetrahymena has hundreds of cilia that fall into different classes depending on their location and length, and all of the Neks tested so far in this organism localize to cilia and regulate cilia length (Wloga et al. 2006).
Mitotic Neks: Nek2, Nek6, Nek7, and Nek9
Of all the human Nek kinases, Nek2 is the most closely related by sequence to NIMA being 48% identical within the catalytic domain. For this reason, it is to date the most well-studied member of this family. It also shares a number of other important properties with NIMA, including cell cycle–dependent expression, APC/C-dependent degradation and localization to the centrosome, the higher eukaryotic microtubule organizing center and equivalent of the fungal SPB (Fry 2002; Hayward and Fry 2006). However, like the yeast Neks and unlike Aspergillus NIMA, human Nek2, as far as one can tell, is not essential for mitotic entry.
Nek2 is a ubiquitous protein with at least three splice variants, Nek2A (445 residues; 51 kDa), Nek2B (384 residues; 44 kDa), and Nek2C (437 residues; 50 kDa; also called Nek2A-T). These all share the N-terminal kinase domain followed by a leucine zipper coiled-coil motif that promotes dimerization and activation. Nek2A then contains a second coiled-coil motif at its C-terminus. Nek2B is somewhat shorter than Nek2A and lacks the second coiled-coil, whereas Nek2C is identical to Nek2A apart from missing an 8 residue sequence that lies between the leucine zipper and C-terminal coiled-coil. The relative expression of the three variants differs depending on cell type and developmental stage, although generally Nek2A appears to be the predominant isoform. The major function of Nek2A is to regulate centrosome organization through the cell cycle, as described below. Similarly, Nek2B is required for assembly and maintenance of centrosome structure and this may be particularly important during early embryonic development as, in Xenopus, it is the only variant expressed at this stage (Uto and Sagata 2000). Interestingly, a role for Nek2B has been proposed in mitotic exit in human cells where its depletion leads to cytokinesis failure (Fletcher et al. 2005). The 8 residue deletion that leads to Nek2C creates a functional nuclear localization signal not present in Nek2A or Nek2B, raising the possibility of nuclear-specific functions for this variant (Wu et al. 2007).
Overall, Nek2 expression and activity are regulated in a cell cycle–dependent manner (Fry et al. 1995; Schultz et al. 1994). The protein is almost undetectable during G1 but accumulates abruptly at the G1/S transition and remains high until late G2. This is a combined result of transcriptional repression and APC/C-mediated degradation in G1 and transcriptional upregulation in S/G2 (Hayward and Fry 2006). However, Nek2A activity is further regulated by a number of key structural and sequence-specific features of the kinase. The N-terminal kinase domain contains a number of sites which undergo autophosphorylation upon dimerization through the leucine zipper (Rellos et al. 2007). It is likely that phosphorylation at least at some of these sites, particularly those in the activation loop, is required for kinase activation. The C-terminal non-catalytic domain also contains autophosphorylation sites, although the functions of these remain unclear. The C-terminal coiled-coil motif has recently been defined as a putative SARAH domain, which enables interaction with the Hippo pathway proteins, hSav1 and the Mst2 kinase (Mardin et al. 2010). Mst2 acts as an upstream activator of Nek2, phosphorylating sites in the C-terminal region that regulate localization of Nek2 to the centrosome. The C-terminal domain also contains a site that mediates direct interaction of Nek2 with the phosphatase, PP1 (Helps et al. 2000). PP1 negatively regulates Nek2 through dephosphorylation, while Nek2 may be able to inhibit PP1 by phosphorylation. This creates a very sensitive bistable switch that allows rapid Nek2 activation, once PP1 activity starts to decrease at the onset of mitosis (Eto et al. 2002). Nek2 is also negatively regulated by the focal adhesion scaffolding protein, HEF1, although the mechanism is not known (Pugacheva and Golemis 2005). Finally, the C-terminal domain of Nek2 contains two destruction motifs that target it for APC/C-mediated degradation in early mitosis (Hayes et al. 2006).
Nek2 localizes to the centrosome throughout the cell cycle (Fry et al. 1998b). An additional fraction is present in the cytoplasm where it colocalizes with and is trafficked along microtubules (Hames et al. 2005). This localization is dependent on the region between the leucine zipper and SARAH domain that encompasses sites phosphorylated by Mst2 (Hames et al. 2005; Mardin et al. 2010). The primary role of Nek2 at the centrosome is as a regulator of centrosome cohesion. Overexpression of active Nek2 induces premature centrosome splitting during interphase (Fry et al. 1998b), while expression of inactive Nek2 or depletion by RNAi inhibits centrosome separation and promotes monopolar spindle formation (Faragher and Fry 2003; Fletcher et al. 2005; Mardin et al. 2010). Nek2 promotes loss of cohesion or “disjunction” of centrosomes through phosphorylation of C-Nap1 and, possibly rootletin and β-catenin (Bahmanyar et al. 2008; Fry et al. 1998a). These proteins, together with Cep68, form a flexible linker structure that extends between the proximal ends of the parental centrioles and which must be dismantled at the onset of mitosis to allow centrosome separation and spindle formation to occur. In addition to the role of Nek2 in centrosome separation, Nek2 has been implicated in regulating the microtubule organizing capacity of the centrosome through phosphorylation of Nlp and centrobin, chromatin condensation through phosphorylation of HMGA2, spindle checkpoint signaling through phosphorylation of Hec1, and nuclear pore complex disassembly through phosphorylation of Nup98 (Laurell et al. 2011; O’Regan et al. 2007).
Nek9 is a 979 residue protein of 113 kDa that, following its N-terminal catalytic domain, has a C-terminal regulatory region comprising a central Regulator of Chromosome Condensation 1 (RCC1)-like domain and a C-terminal coiled-coil motif. It is expressed throughout the cell cycle, but becomes phosphorylated and activated specifically in mitosis (Roig et al. 2002). Nek9 is subject to complex but, as yet, poorly understood regulation. It is thought that during interphase Nek9 adopts an autoinhibited conformation with the RCC1-like domain blocking access to the catalytic site. However, at the G2/M transition, a number of events lead to release of this conformation and activation of the kinase. These are likely to include dimerization, autophosphorylation (especially on T210 in the activation loop), and, potentially, activation by upstream kinases such as Cdk1 and Plk1 (Roig et al. 2002; Bertran et al. 2011). In terms of localization, Nek9 is mainly cytoplasmic, although it can also be found in the nucleus. However, its activated form appears to be specifically concentrated on spindle poles during mitosis (Roig et al. 2005).
Early functional studies on Nek9 by overexpression of wild-type or mutant constructs, or by inhibition through antibody microinjection, led to the hypothesis that Nek9 activity contributes to mitotic spindle formation (Roig et al. 2002). How exactly it does this remains unclear, but Nek9 can interact with the γ-tubulin ring complex (γ-TuRC) that nucleates microtubules both from the centrosome and within the spindle (Roig et al. 2005). Hence, one is tempted to speculate that Nek9 directly contributes to microtubule nucleation in mitosis. However, depletion of Nek9 from Xenopus egg extracts prevents the formation of spindles via either the centrosome- or chromatin-mediated pathways (Roig et al. 2005). This suggests that Nek9 function is unlikely to be restricted to regulating the microtubule nucleating activity of γ-tubulin. Nek9 also interacts with BICD2, a protein associated with microtubule-dependent motor proteins, raising the possibility that BICD2 might target Nek9 to microtubules, as well as potentially being a substrate (Holland et al. 2002). Furthermore, the presence of the RCC1-like domain, together with the demonstration that the Ran GTPase can bind to Nek9, points to a potential role in Ran-mediated spindle formation (Roig et al. 2002). Some reports suggest that Nek9 could have functions outside of mitosis. For example, Nek9 was reported to associate with the FACT complex, a chromatin modifying complex involved in replication and transcription (Tan and Lee 2004). This could explain the presence of a nuclear fraction of Nek9 in interphase and, although not consistent with the timing of bulk Nek9 activation, it is possible that the FACT complex activates a restricted pool of Nek9 in the nucleus, which in turn might be inhibited by the adenovirus E1A protein.
One major route through which Nek9 almost certainly regulates mitotic spindle organization is through interaction, phosphorylation, and activation of two other NIMA-related kinases, Nek6 and Nek7. Indeed, Nek9 was first identified through its association with Nek6 (Roig et al. 2002). Subsequently, Nek9 was shown to phosphorylate sites within the activation loop of Nek6 and this, together with the high degree of similarity between Nek6 and Nek7, led to the proposal that these three Neks form a mitotic cascade in which Nek9 acts upstream of Nek6 and Nek7 (Belham et al. 2003). Lately, it has been demonstrated that Nek9 may also activate Nek7, and by analogy Nek6, through an allosteric mechanism independent of phosphorylation (Richards et al. 2009).
Nek6 and Nek7 are the smallest family members being only 313 (36 kDa) and 302 (35 kDa) residues, respectively. As a result they comprise little more than a catalytic domain with only a short (30–40 residue) N-terminal extension. By sequence, they are highly related to each other, sharing 86% amino acid identity within the catalytic domain, although the N-termini of the two proteins are not conserved. Often considered as a pair because of their similarity, both Nek6 and Nek7 are, like their upstream activator Nek9, cell cycle regulated with maximal activity in mitosis. Moreover, while they give slightly different localization patterns, with Nek6 weakly associated to spindle fibers and Nek7 more concentrated on spindle poles, RNAi depletion experiments demonstrate that they are both essential not only for assembly of a robust mitotic spindle, but also potentially for completion of cytokinesis and cell abscission (Kim et al. 2007; O’Regan and Fry 2009; Yin et al. 2003; Yissachar et al. 2006). In support of a late mitotic role for these kinases, Nek6 activation, judged with a phosphospecific antibody against a key activation loop residue, peaks at the time of cytokinesis (Rapley et al. 2008), while mouse embryonic fibroblasts derived from Nek7−/− embryos show defects indicative of cytokinesis failure (Salem et al. 2010).
How these kinases regulate spindle formation and cytokinesis remains to be defined. However, despite their sequence similarity and requirement for both events, it is likely that their respective roles and substrates differ. Systems approaches have identified a large number of putative interacting partners and substrates for Nek6 (Ewing et al. 2007; Vaz Meirelles et al. 2010). These include proteins involved in chromatin condensation, microtubule binding, and nuclear pore complex organization. On this basis, Nek6 has been proposed to be a high confidence hub kinase with an expansive network of substrates involved in diverse cellular processes. Moreover, a number of these interactions depend on the N-terminal extension of Nek6 explaining why Nek6 and Nek7 may target different proteins (Vaz Meirelles et al. 2010). However, to date, the only substrates that have been studied in any detail for Nek6 and Nek7 are the kinesin motor, Eg5, which is phosphorylated by Nek6 promoting spindle pole separation (Rapley et al. 2008), and the nuclear pore complex component, Nup98, whose combined phosphorylation by Cdk1 and potentially multiple Neks, including Nek2, Nek6, and Nek7, promotes nuclear pore complex disassembly upon mitotic entry (Laurell et al. 2011).
Checkpoint Neks: Nek10 and Nek11
Nek10 is a protein of 1,125 residues in length (129 kDa) that is unique in this family, in that unlike the other Neks, Nek10 has its catalytic domain in the center of the protein with long N- and C-terminal non-catalytic regions. The catalytic domain is flanked by two coiled-coil motifs, while in addition the N-terminal region contains four armadillo repeats. Nek10 is one of the least characterized members of the family at the present time with, for example, no data yet on its localization. However, the first report into its function has placed Nek10 within the G2/M DNA damage checkpoint, suggesting that Nek10 is required for Erk1/2 activation in response to UV-induced damage (Moniz and Stambolic 2011). In unperturbed cells, Nek10 forms a trimeric complex, interacting with Mek1 via Raf-1. In response to UV treatment, Mek1 undergoes autophosphorylation and activation in a Nek10-dependent manner; this leads to phosphorylation of Erk1/2 and G2/M arrest. Importantly, though, UV treatment increases neither association of Nek10 with Raf-1 and Mek1, nor Nek10 activity, and there is no evidence that Nek10 directly phosphorylates Raf-1 or Mek1. Hence, it seems likely that other factors or events contribute to the mechanism by which Nek10 modulates Mek1 signaling.
Nek11 exists as at least four splice variants: Nek11 Long (Nek11L; 645 residues, 74 kDa), Nek11 Short (Nek11S; 470 residues, 54 kDa), and two isoforms that are present in databases but have yet to be reported, Nek11C (482 residues, 55 kDa) and Nek11D (599 residues, 69 kDa). Expression of at least the Nek11L isoform is cell cycle regulated being highest from S through to the G2/M phases of the cell cycle (Noguchi et al. 2002). Localization studies revealed that Nek11 localizes to the nucleus, and possibly nucleolus, in interphase cells, as well as spindle microtubules in prometaphase and metaphase cells (Noguchi et al. 2002, 2004). It was also suggested that Nek11 might be phosphorylated by the Nek2 kinase, converting Nek11 into an active conformation (Noguchi et al. 2004); however, this has yet to be verified. More convincingly, Nek11 is implicated in the DNA damage response (DDR). Nek11 activity is increased in response to stalled DNA replication and genotoxic stresses, such as ionizing radiation (IR), and this activation is blocked by caffeine, an inhibitor of the DDR kinases, ATM and ATR (Melixetian et al. 2009; Noguchi et al. 2002). The role of Nek11 in the G2/M DNA damage checkpoint appears to be a central one. Upon DNA damage, ATM and ATR phosphorylate and activate Chk1; this in turn activates Nek11 by phosphorylation of Ser273 (Melixetian et al. 2009). Both Chk1 and Nek11 then phosphorylate Cdc25A on residues that promote binding of the β-TrCP E3 ubiquitin ligase. Ultimately, this results in the proteasomal degradation of Cdc25A arresting the cell in G2. While this model is highly attractive, it has been argued that casein kinase 1, and not Nek11, is the major kinase that regulates the phospho-dependent recruitment of β-TrCP. It is also worth noting that, in Xenopus, Erk1/2 can target Cdc25A for degradation through the β-TrCP pathway following genotoxic stress. Clearly, further work is required to determine the relative importance of these kinases in mediating G2/M arrest in response to different forms of DNA damage.
Ciliary Neks: Nek1 and Nek8
Nek1 was the first Nek to be identified in mammals and is also the largest member of the family, being composed of 1,258 residues (145 kDa). It was initially reported to have dual serine-threonine and tyrosine kinase activity in vitro, although this is unlikely to be the case in vivo (Letwin et al. 1992). The first clue to its function came almost a decade later when, completely unexpectedly, mutations of the gene encoding Nek1 were found to be causative in two mouse models for polycystic kidney disease (PKD), named kat (for kidney, anemia, testis) and kat 2J (Upadhya et al. 2000; Vogler et al. 1999). Cystic kidney diseases are now known to be common hallmarks of ciliopathies, whereby the underlying defect is in the formation or function of the primary cilium. Indeed, this was the first indication that Nek kinases may have some role in ciliogenesis. Since then, Nek1 has been localized to the primary cilia in a number of different cell types (Mahjoub et al. 2005; Shalom et al. 2008; White and Quarmby 2008). Here, it is proposed to negatively regulate ciliogenesis, as overexpression of wild-type and certain truncated forms of Nek1 inhibit ciliogenesis, whereas overexpression of mutants predicted to be catalytically inactive do not (White and Quarmby 2008). Interestingly, Nek1 may also be involved in cell cycle control and, specifically, the DDR. Like Nek11, Nek1 activity is elevated in response to IR and Nek1-deficient cells are sensitive to DNA damage (Polci et al. 2004). Furthermore, although to date no bona fide substrates have been identified for Nek1, a number of proteins involved in DNA double-strand break repair were found as Nek1 binding partners (Surpili et al. 2003).
Nek8 is a smaller protein than Nek1, being 692 residues in length (80 kDa). Interestingly, Nek8 shares a very similar domain organization to Nek9, with Nek8 also having an RCC1-like domain. On this basis, one might expect that Nek8 would have a role in mitosis. However, subsequent to identification of the mouse Nek1 PKD model, a missense mutation in the RCC1-like domain of Nek8 was identified in the jck mouse model of autosomal recessive juvenile PKD (Liu et al. 2002). Consistent with this, Nek8 was also found to localize to the primary cilia (Quarmby and Mahjoub 2005; Sohara et al. 2008). In fact, Nek8 concentrates in the proximal region of the cilia, known as the inversin compartment, a localization that is dependent on the inversin protein (Shiba et al. 2010). Inversin is a protein encoded by a gene that is mutated in the human childhood kidney disorder, nephronophthisis (NPHP). This disease is a typical ciliopathy with at least nine candidate Nphp genes identified so far. Importantly, Nek8 turns out to be Nphp9, thus implicating the Neks in human inherited disease for the first time (Otto et al. 2008). Mutations in Nek8 found in NPHP patients lead to loss of Nek8 localization from the cilia (Trapp et al. 2008). Nek8 has also been found to interact with polycystin-2 (PC-2), a causative gene for human autosomal dominant PKD (ADPKD), with abnormal phosphorylation of PC-2 detected in cells from the jck mouse (Sohara et al. 2008). To date, though, much still remains to be learnt about the cellular basis of renal cyst formation. Thus, it remains unclear whether defects in Nek1 and Nek8 cause cystic kidney disease directly, by interfering with the structure of the primary cilium itself, or more indirectly, by abrogating cilia-dependent signaling. It is also formally possible that these Neks somehow relay signals between the cilium, at the time when the cell is quiescent, and the mitotic spindle, once it reenters the cell cycle.
Signal Transduction Neks: Nek3
In contrast to most other Neks, there is no clear indication that Nek3 (506 residues, 58 kDa) is a cell cycle–dependent kinase with conflicting reports over whether Nek3 expression in elevated in dividing or quiescent cells. Localization studies indicate that Nek3 is predominantly cytoplasmic, with no evidence to date that Nek3 associates with centrosomes (Tanaka and Nigg 1999). Unexpectedly, however, yeast two-hybrid and co-immunoprecipitation studies found a direct interaction between Nek3 and members of the Vav family of guanine nucleotide exchange factors that was enhanced in response to signaling from the prolactin receptor (Miller et al. 2005). These studies suggested that prolactin receptor stimulation induces Nek3 kinase activity causing Vav2 to interact with the kinase domain of Nek3 and become phosphorylated. Phosphorylated Vav proteins then activate downstream signaling targets involved in tumor progression. Indeed, overexpression of Nek3 potentiated prolactin-mediated cytoskeletal reorganization of cells; however, if Nek3 was depleted then cytoskeletal reorganization was attenuated, as was cell migration and invasion. This is therefore the first report of a Nek kinase being involved in growth-related signaling events. However, Nek3 has also been reported to more directly regulate cytoskeletal dynamics in neurons by altering levels of acetylated tubulin, thus raising the possibility of a role for Nek3 in neuronal disorders (Chang et al. 2009).
Nek4 and Nek5
Little research has been carried out on Nek4 (781 residues, 90 kDa). However, a recent study implies that this kinase might also play a role in microtubule regulation, as changes in Nek4 expression led to altered sensitivity of cells to microtubule poisons (Doles and Hemann 2010). There is no specific published data yet on Nek5 (708 residues, 81 kDa). Intriguingly, though, Nek5 was identified in a microarray analysis of FOXJ1 target genes. This transcription factor governs motile cilia assembly by regulating genes involved in cilia biogenesis and function, suggesting that Nek5 may contribute to ciliogenesis in multiciliated cells.
Nek Kinases and Cancer
There is increasing evidence implicating Nek kinases in cancer. Most commonly, this involves upregulated expression, although a few rare mutations have been identified in cancer genome screens. However, whether elevated expression or mutation contributes to the transformed phenotype remains an important but currently unanswered question. The current data on Nek kinases and cancer can be summarized as follows. Nek1 may be required to protect genome stability as cells deficient in Nek1 form tumors in mice (Chen et al. 2011); this would be consistent with it having a role in the DNA damage checkpoint. Nek2 expression levels are frequently upregulated in a wide range of cancer cell lines and primary tumors, and the Nek2 gene has been reported to be amplified in breast and gastric cancers (Hayward and Fry 2006; Kokuryo et al. 2007; Suzuki et al. 2010; Tsunoda et al. 2009). Nek3 is enriched in breast carcinomas (Miller et al. 2005, 2007), while the Nek3 gene is located in a chromosomal region that is frequently deleted in several types of human cancer and a polymorphism in the Nek3 gene is linked to prostate cancer. Nek4 appears to be frequently deleted in lung cancer, although interestingly this might make tumors more sensitive to particular microtubule poisons (Doles and Hemann 2010). The Nek6 gene is found at a locus for which loss of heterozygosity is associated with several cancers, while, like Nek2, the expression of Nek6 is widely elevated in cancer cell lines and tumors (Nassirpour et al. 2010). Nek7 is also overexpressed in breast, colon, and larynx cancers (Capra et al. 2006), and mutations in Nek7 have been identified in lung and ovarian cancers. Nek8, although primarily associated with ciliogenesis, is upregulated in some breast tumors. Although this could suggest that Nek8 has alternative functions in cell cycle control, a link between cilia-dependent signaling and tumorigenesis is beginning to emerge. A few point mutations have been identified in Nek9, although their relevance remains unclear, and a link between Nek9 expression and human cancers is not yet well established. On the other hand, Nek10 has been identified as a candidate breast cancer susceptibility gene and mutations of this kinase have been reported in several human cancers, including lung. Finally, Nek11 expression is increased during colorectal cancer development suggesting that this kinase too is implicated in cancer progression (Sørensen et al. 2010).
In summary, of the 11 human Nek kinases, Nek2, Nek6, Nek7, and Nek9 function in mitotic regulation, Nek10 and Nek11 in the DDR, Nek3 in signal transduction, and Nek1 and Nek8 in ciliogenesis. Currently, Nek4 and Nek5 have no assigned function. Hence, although these kinases are related to each other in sequence, it is clear that they are regulated in quite different manners and function in diverse processes. However, based on studies in both lower and higher eukaryotes, the common underlying theme is that many, albeit perhaps not all, Neks contribute to microtubule organization during cell cycle progression and/or ciliogenesis. Importantly, advances in understanding how this family contributes to cell cycle events make these kinases attractive targets for therapeutic interventions in human cancer. RNAi-mediated depletion of Nek2 and Nek6, for example, has been found to inhibit proliferation of cancer cell lines and tumor xenografts, and selective pharmacological inhibitors are beginning to be generated (Hayward et al. 2010; Jeon et al. 2010; Jin et al. 2006; Kokuryo et al. 2007; Nassirpour et al. 2010; Qiu et al. 2009; Suzuki et al. 2010; Tsunoda et al. 2009; Whelligan et al. 2010; Wu et al. 2008). Hence, although a better understanding of the basic biology of NIMA-related kinases is still required, there is growing evidence that Neks could serve as important targets for the management of cancers. Finally, the identification of mutations in Nek8 as causative for an inherited human disease emphasizes the need for further research into the processes and pathways in which this family of kinases operate.
- Ewing RM, Chu P, Elisma F, Li H, Taylor P, Climie S, McBroom-Cerajewski L, Robinson MD, O’Connor L, Li M, Taylor R, Dharsee M, Ho Y, Heilbut A, Moore L, Zhang S, Ornatsky O, Bukhman YV, Ethier M, Sheng Y, Vasilescu J, Abu-Farha M, Lambert JP, Duewel HS, Stewart II B, Kuehl K, Hogue K, Colwill K, Gladwish B, Muskat R, Kinach SL, Adams MF, Moran GB, Morin TT, Figeys D. Large-scale mapping of human protein-protein interactions by mass spectrometry. Mol Syst Biol. 2007;3:89.PubMedPubMedCentralCrossRefGoogle Scholar
- Hayward DG, Newbatt Y, Pickard L, Byrne E, Mao G, Burns S, Sahota NK, Workman P, Collins I, Aherne W, Fry AM. Identification by high-throughput screening of viridin analogs as biochemical and cell-based inhibitors of the cell cycle-regulated Nek2 kinase. J Biomol Screen. 2010;15:918–27.PubMedPubMedCentralCrossRefGoogle Scholar
- Vaz Meirelles G, Ferreira Lanza DC, da Silva JC, Santana Bernachi J, Paes Leme AF, Kobarg J. Characterization of hNek6 interactome reveals an important role for its short N-terminal domain and colocalization with proteins at the centrosome. J Proteome Res. 2010;9:6298–316.PubMedCrossRefGoogle Scholar
- Whelligan DK, Solanki S, Taylor D, Thomson DW, Cheung K-M, Boxall K, Mas-Droux C, Barillari C, Burns S, Grummitt CG, Collins I, van Montfort RLM, Aherne GW, Bayliss R, Hoelder S. Aminopyrazine inhibitors binding to an unusual inactive conformation of the mitotic kinase Nek2: SAR and structural characterization. J Med Chem. 2010;53(21):7682–98.PubMedPubMedCentralCrossRefGoogle Scholar