The protein kinase NLK was originally identified as nemo, a gene involved in Drosophila melanogaster eye development. D. melanogaster compound eyes are made up of hexagonal units called ommatidia. A mutation in nemo affects the movement of ommatidia cells, causing them to acquire a squared shape (Choi and Benzer 1994) which is reflected in the name “nemo,” Korean for square. In 1998, Brott et al. isolated the mouse homolog of nemo and named it “Nemo-like kinase (NLK).” Meneghini et al. (1999) and Rocheleau et al. (1999) found that the Caenorhabditis elegans loss-of-intestine-1 (lit-1) mutant, which lacks the endoderm, has a mutation in a homolog of NLK. This homolog phosphorylates Posterior pharynx defect protein 1 (POP-1) to relieve POP-1-mediated inhibition of gene expression required for endoderm formation. POP-1 is a homolog of the mammalian T-cell factor/lymphoid enhancer factor (Tcf/Lef), a key component of the Wnt/β-catenin signaling pathway. Ishitani et al. (1999) also discovered that mammalian NLK could phosphorylate and inhibit Tcf/Lef in human embryonic kidney 293 (HEK293) cells. This was the first discovery of a molecular function of NLK.
Roles of NLK-Mediated Phosphorylation in the Nucleus
As mentioned above, C. elegans POP-1, a member of the Tcf/Lef protein family, was identified as the first NLK substrate. NLK phosphorylates and regulates vertebrate Tcf/Lefs. Vertebrates have four Tcf/Lef members (five in zebrafish), including Tcf7 (Tcf1), Tcf7L1 (Tcf3), Tcf7L2 (Tcf4), and Lef1. NLK phosphorylates Tcf7L1, Tcf7L2, and Lef1 on conserved Ser/Thr-Pro sequences (e.g., Thr155-Pro156 and Ser166-Pro167 on human Lef1). Although Tcf7 contains such conserved Ser/Thr-Pro sequences, NLK is unable of phosphorylating it (Ishitani et al. 1999, 2003a; Ota et al. 2012). NLK-mediated Tcf/Lef phosphorylation positively or negatively regulates Tcf/Lef activity. In human embryonic kidney 293 (HEK293) and cervical cancer (HeLa) cells, overexpression of NLK reduced DNA binding and inhibited the transcriptional activity of Tcf7L2 and Lef1 (Ishitani et al. 1999, 2003a; Ota et al. 2012). In addition, Yamada et al. (2006) reported that NLK promoted NLK-associated RING finger protein (NARF)-mediated ubiquitination and the subsequent proteasomal degradation of Tcf7L2 and Lef1 in HEK293 cells. On the contrary, NLK is essential for the activation of Lef1-mediated transcription in neural progenitor cells (NPCs) and related cells. In NPC-like mammalian cell lines, rat pheochromocytoma tumor (PC12) cells, mouse neuroblastoma (neuro-2a) cells, and zebrafish midbrain NPCs, histone deacetylase HDAC1 binds to Lef1 and inhibits its transcriptional activity. NLK-mediated Lef1 phosphorylation stimulates the dissociation of Lef1 from HDAC1, thereby promoting Lef1 transcriptional activity. Such positive regulation promotes the proliferation of zebrafish midbrain NPCs and consequent midbrain size expansion. Consistent with this finding, knockdown of zebrafish type-II NLK, Nlk2, reduced Lef1 activity and cell proliferation in the developing midbrain, whereas co-knockdown of HDAC1 reversed this effect (Ota et al. 2012).
A biochemical screening for NLK substrates yielded the Notch signaling transcription factors, Notch1 and Notch3 (Ishitani et al. 2010). NLK phosphorylates the intracellular domain of Notch1 (Notch1-ICD) on seven Ser residues within conserved Ser-Pro motifs. Phosphorylation at these sites prevents the formation of a ternary complex between Notch1-ICD, the DNA-binding factor CSL (CBF-1, Suppressor of Hairless, and LAG-1) and a member of the Mastermind family of transcriptional coactivators. Consequently, Notch1-ICD-mediated gene expression is inhibited in a variety of mammalian cultures, including HEK293, HeLa, neuro-2a, PC12, and the colorectal cancer SW480 cells. NLK can also regulate Notch1 activity in vivo. In zebrafish, Notch1 signaling maintains NPCs in an undifferentiated state, preventing their commitment to neurons. Knockdown of zebrafish type-I NLK, Nlk1, enhanced Notch1 signaling and blocked NPC differentiation in the zebrafish neural plate. Accordingly, Nlk1-mediated downregulation of Notch1 signaling was likely to be essential for proper neurogenesis in the zebrafish neural plate (Ishitani et al. 2010). A recent report showed that NLK-mediated Notch1 inhibition contributed to natural killer cell development (Cichocki et al. 2011). Thus, NLK negatively regulates Notch1 signaling in vivo. In contrast, the role of NLK-mediated Notch3 phosphorylation remains unclear, although Ishitani et al. (2010) previously reported that NLK could promote Notch3 activity in a kinase-dependent manner in neuro-2a cells.
NLK can phosphorylate other families of transcriptional regulators, including c-Myb, signal transducer and activator of transcription 3 (STAT3), Forkhead box protein O1 (Foxo1), and MEF2A (Fig. 2). c-Myb is a transcription factor that regulates hematopoietic stem cell proliferation and differentiation. NLK promotes c-Myb degradation via phosphorylation of 15 Ser/Thr-Pro motifs in monkey kidney CV-1 cells (Kanei-Ishii et al. 2004). Kurahashi et al. (2005) showed that NLK could phosphorylate the c-Myb-related protein A-Myb, thus promoting the dissociation of A-Myb from the transcriptional coactivator CBP and attenuation of A-Myb-dependent transcription. STAT3 is an important transcription factor that mediates cytokine signaling. NLK phosphorylates STAT3 on Ser-727 to promote downstream interleukin-6 (IL-6) signaling in human hepatocellular carcinoma (HepG2) cells (Kojima et al. 2005). Ohkawara et al. (2004) reported that X. laevis type-I NLK, NLK1, regulated mesoderm formation via STAT3 Ser-727 phosphorylation. Kim et al. (2010) showed that NLK phosphorylates Foxo1, which controlled the expression of genes involved in apoptosis, cell cycle, stress response, longevity, DNA repair, and glucose metabolism. NLK-mediated phosphorylation negatively regulates the transcriptional activity of Foxo1 by promoting its nuclear export in monkey kidney (Cos-1) cells. Satoh et al. (2007) reported that X. laevis NLK1 phosphorylated the MEF2A transcription factor to control embryonic anterior head formation. Interestingly, Ota et al. (2011) showed that NLK1 overexpression could induce the phosphorylation of key regulators of translational control, including Pumilio1, Pumilio2, and cytoplasmic polyadenylation element-binding protein (CPEB) in X. laevis oocytes, suggesting that NLK may phosphorylate not only transcriptional but also translational regulators.
Recent D. melanogaster genetic analyses have revealed that Nemo phosphorylates a variety of transcriptional regulators, such as even-skipped (Eve), mothers against decapentaplegic (Mad), eyes absent (Eya), and period (Fig. 2). The phosphorylation of the Eve homeobox transcription repressor contributes to embryonic segment patterning (Braid et al. 2010). Mad is the D. melanogaster homolog of the bone morphogenetic protein (BMP) signaling transcription factor Smad. Mad phosphorylation negatively regulates BMP signaling during wing development (Zeng et al. 2007). Eya is a transcriptional regulator essential for eye development, which is activated by Nemo-dependent phosphorylation (Morillo et al. 2012). Period is a circadian clock component, whose stability is enhanced by Nemo-dependent phosphorylation (Chiu et al. 2011).
Occasionally, NLK can work without phosphorylating its substrate. NLK binds to ATF5, a member of the ATF/CREB protein family of transcription factors, thus stabilizing ATF5 in a kinase-independent manner downstream of IL-1 signaling (Zhang et al. 2015). Finally, NLK stabilizes and activates the transcription factor p53, albeit without phosphorylating it, in response to DNA damage (Zhang et al. 2014). Thus, NLK plays a variety of roles in the nucleus.
Roles of NLK-Mediated Phosphorylation Outside the Nucleus
Given that in many mammalian cell lines, exogenous NLK localizes mainly to the nucleus (Brott et al. 1998; Ishitani et al. 2009), nuclear functions of NLK have been well studied. Interestingly, our immunostaining analyses revealed that endogenous NLK localized to the cytoplasm and not to the nucleus in neuro-2a, PC12, HEK293, and HeLa cells. Moreover, stimulation with nerve growth factor (NGF) not only promoted the translocation of endogenous NLK into the nucleus and leading edges of the cell, but also induced the enzymatic activation of NLK (Ishitani et al. 2009; T.I. and S.I. unpublished observations). These findings suggest that NLK functions not only in the nucleus but also in the cytoplasm and near the plasma membrane. Consistent with this idea, Ishitani et al. (2009) showed that NLK phosphorylated the focal adhesion adaptor protein paxillin on Ser-126 and the microtubule-associated protein 1B (MAP1B) at the leading edge of NGF-treated PC12 cells (Fig. 2). D. melanogaster Nemo localizes to the plasma membrane and phosphorylates the β-catenin homolog armadillo during eye development (Fig. 2; Mirkovic et al. 2011), indicating that invertebrate NLK functions also near the plasma membrane. However, because mammalian NLK cannot phosphorylate β-catenin (Ishitani et al. 2003a; T.I. and S.I. unpublished observations), the relationship between NLK and β-catenin does not appear to be evolutionarily conserved.
Recently, Yuan et al. (2015) have reported that NLK plays an important role in stress response in the cytoplasm. In nutrient-rich conditions, the Rag1 GTPase binds to and activates the mechanistic target of rapamycin (mTORC1) and promotes cell growth by stimulating transcription, translation, and anabolism. To counteract osmotic and oxidative stress, NLK phosphorylates Raptor, an mTORC1 subunit, on Ser-863 (Fig. 2) and disrupts the interaction between Rag1 and mTORC1, thus inhibiting mTORC1 activation.
Regulation of NLK Activity
As described above, NLK is an atypical MAPK. The activity of typical MAPKs is tightly regulated, and overexpression is insufficient to activate them. MAPK kinase (MAPKK)-dependent phosphorylation of Thr and Tyr residues in the TXY motif is required to activate typical MAPKs. However, overexpression is sufficient for activation of NLK kinase (Brott et al. 1998; Ishitani et al. 2011), suggesting that NLK does not require the intervention of upstream kinases. Given that the negatively charged glutamic acid can mimic a phosphorylated amino acid, it is possible that Ser/Thr kinase(s)-mediated phosphorylation of Thr-286 in the TQE sequence leads to NLK activation. Ishitani et al. (2011) showed that exogenous NLK formed homodimers and then autophosphorylated Thr-286 in a trans-manner in HEK293 cells. Substitution of Thr-286 with a valine or phosphatase treatment reduced NLK kinase activity. The conserved C-terminal domain is required for NLK homodimerization. Mutation of Cys-425 in the C-terminal domain of NLK prevented NLK dimerization and hampered its kinase activity (Ishitani et al. 2011). Thus, NLK can be self-activated via homodimerization. Ishitani et al. (2011) also reported that inactive endogenous NLK was trapped in large heterologous complexes around the Golgi in unstimulated PC12 cells. Addition of NGF stimulated the release of NLK from the complex, its homodimerization leading to autophosphorylation on Thr-286 and subsequent activation (Ishitani et al. 2011) (Fig. 3). However, the detailed nature of the large Golgi complex remains unclear.
In addition to NGF, several extracellular factors have been identified as NLK activators. Endogenous NLK can be activated by epidermal growth factor (EGF) or Wnt-3a in PC12 cells (Ishitani et al. 2009; Ota et al. 2012) and by Activin A or Wnt-5a in HEK293 cells (Ohkawara et al. 2004; Ishitani et al. 2003b). Exogenous NLK can be activated by treatment with IL-6 or transforming growth factor-β (TGF-β) in HepG2 cells (Kojima et al. 2005) and by osmotic or oxidative stress in HEK293 cells (Yuan et al. 2015). Several intracellular modulators of NLK activity have also been identified. One of these is TAK1 MAP3K, which promotes the autophosphorylation and activation of NLK/LIT-1 in C. elegans (Meneghini et al. 1999; Ishitani et al. 1999). TAK1 appears to activate NLK indirectly. Kanei-Ishii et al. (2004) reported that TAK1 phosphorylated and activated homeodomain-interacting protein kinase 2 (Hipk2), which then bound to and activated NLK. Ohnishi et al. (2010) reported that p38 MAPK phosphorylated and activated NLK downstream of TAK1. Although TAK1 is involved in Wnt-5a- and IL-6-dependent activation of NLK (Ishitani et al. 2003b; Kojima et al. 2005), NGF and EGF activate NLK in a TAK1-independent manner through Ras GTPase (Ishitani et al. 2009; T.I. and S.I. unpublished observations). Kim et al. (2012) recently described dimerization partner 1 (DP1) as an inhibitor of NLK. DP1 binds directly to and inhibits NLK, consequently enhancing Wnt/β-catenin signaling in X. laevis early embryos.
NLK is regulated at the protein level by noncoding RNAs. MicroRNAs miR-181a and miR-181b reduce NLK protein levels and prevent NLK-mediated suppression of Notch signaling, thereby promoting human natural killer cell development in vitro (Cichocki et al. 2011). In hepatocellular carcinoma (HCC) cell lines, miR-181 family members (miRNA-181a/b/c/d) and miR-101 negatively regulate NLK expression (Ji et al. 2009; Shen et al. 2014). In addition, miR-288a inhibits NLK expression in rat cardiomyocyte (H9C2) cells (Huang et al. 2016). Furthermore, a cancer-related long noncoding RNA, HOTAIR, has been shown to inhibit NLK transcription in glioma cells (Zhou et al. 2015). Because the regulation of NLK enzymatic activity is relatively loose, organisms may have evolved a variety of systems to control NLK expression.
To better understand NLK function and regulation, it would be advantageous to develop specific chemical inhibitors against NLK. At present, NLK activation can be blocked only with lithium chloride and lithium carbonate (Ishitani et al. 2009; T.I. and S.I. unpublished observations). However, lithium is not a specific inhibitor, and development of specific NLK inhibitors is awaited.
Physiological Roles of NLK
The NLK gene family plays important roles in central nervous system (CNS) development in vertebrates. X. laevis NLK1 regulates anterior brain formation through MEF2A phosphorylation in early embryos (Satoh et al. 2007). D. rerio Nlk1 controls neural plate anterior-posterior patterning via positive regulation of Wnt/β-catenin signaling (Thorpe and Moon 2004) and promotes the differentiation of NPCs into neurons by blocking Notch signaling (Ishitani et al. 2010). In contrast, D. rerio Nlk2 contributes to midbrain size expansion through Lef1 phosphorylation and consequent stimulation of Wnt/β-catenin signaling (Ota et al. 2012). Kortenjann et al. (2001) reported that type-II NLK-deficient mice suffered from various neurological abnormalities such as cerebellar ataxia, indicating that type-II NLK might contribute to brain development in mammals. Thus, type-I NLKs seem to control early brain (neural plate) development, whereas type-II NLK appears to regulate the later stages.
Mouse type-II NLK is also involved in non-neural tissue development, including lung morphogenesis, adipogenesis, and hematopoiesis. Ke et al. (2016) reported that type-II NLK-deficient mice became cyanotic because of lung maturation defects, exhibiting elevated vascular endothelial growth factor (VEGF) expression, and small and compressed alveoli (Ke et al. 2016). Kortenjann et al. (2001) showed that bone marrow adipogenesis was enhanced, whereas the number of hematopoietic cells was reduced in type-II NLK-deficient mice.
NLK in Disease
A number of studies have reported a correlation between cancer development and NLK expression or activity. NLK negatively regulates the expression of androgen receptor (AR) and the nuclear steroid hormone receptor Nurr1, both of which correlate with prostate cancer progression. Consistent with this, NLK expression has been found to decrease during prostate cancer progression (Emami et al. 2009; Wang et al. 2016). Similarly, NLK expression is inversely correlated with glioma grade (Cui et al. 2011; Sa et al. 2015). NLK can negatively regulate Wnt/β-catenin signaling through Lef1 phosphorylation and suppress mesenchymal activity of glioma cells; however, in high-grade gliomas in which NLK expression is very low, Wnt/β-catenin signaling and mesenchymal activities are upregulated (Sa et al. 2015). Thus, NLK exhibits tumor suppressor activities in this type of cancer. By contrast, in HCCs, upregulation of NLK enhances the expression of cyclin D1, a core component of cell cycle regulation. NLK RNAi has been reported to reduce HCC viability (Jung et al. 2010). In addition, Mendes-Pereira et al. (2012) showed that NLK RNAi reduced the viability of PTEN-deficient breast, colorectal, bladder, and ovary cancer, as well as in melanoma cells. Consequently, NLK represents a potential therapeutic target for these tumor types, whose treatment would benefit from the existence of specific chemical NLK inhibitors.
A recent study using a mouse disease model has revealed that NLK is involved in polyglutamine diseases, including spinal and bulbar muscular atrophy (SBMA) and spinocerebellar ataxia type 1(SCA1). Polyglutamine diseases represent a group of neurodegenerative disorders caused by expanded CAG repeats encoding a long polyQ tract in the corresponding proteins. SBMA is caused by polyQ-expanded AR (polyQ-AR), resulting in motor neuron degeneration and muscle bulk loss. SCA1 is caused by polyQ-expanded Ataxin1 (polyQ-Ataxin1) and leads to cerebellum and motor neuron degeneration. Interestingly, Lim and coworkers at Yale University reported that NLK phosphorylated polyQ-AR and polyQ-Ataxin1 (Fig. 2). Moreover, loss of one copy of the NLK gene rescued neuropathological phenotypes in SBMA and SCA1 model mice (Ju et al. 2013; Todd et al. 2015). Although the mechanisms by which NLK promotes SBMA and SCA1 are not fully understood, NLK may be a potential target for SBMA and SCA1 therapy.
NLK is an evolutionarily conserved atypical MAPK, which plays multifaceted roles in invertebrate and vertebrate development. NLK controls gene expression and cytoskeletal architecture by phosphorylating and regulating a variety of cell signaling components and cytoskeleton-associated proteins. As a result, it contributes to cell growth, proliferation, differentiation, and morphological changes during early embryonic patterning, brain development, lung morphogenesis, adipogenesis, and hematopoiesis. Although activated NLK phosphorylates many substrates, it is unlikely that it does so simultaneously. Instead, NLK probably phosphorylates and regulates specific substrates depending on the context. However, the detailed mechanisms of such selective regulation are still not clear. What is partly known is the process of NLK activation: NLK is released from a large Golgi complex in response to upstream stimulation, leading to dimerization and self-activation via autophosphorylation on Thr-286. However, the nature of this large complex and the mechanism by which NLK dissociates from it remain unclear, and these will be the focus of future studies. Recent reports have shown that NLK inhibition can ameliorate tumorigenesis and neurodegenerative diseases. Although specific chemical inhibitors against NLK have not been reported yet, they may be a useful tool not only in basic biological research but also in the treatment of multiple diseases.
- Braid LR, Lee W, Uetrecht AC, Swarup S, Papaianni G, Heiler A, Verheyen EM. Nemo phosphorylates even-skipped and promotes eve-mediated repression of odd-skipped in even parasegments during Drosophila embryogenesis. Dev Biol. 2010;343:178–89. doi:10.1016/j.ydbio.2010.04.008.PubMedCrossRefGoogle Scholar
- Ishitani T, Ninomiya-Tsuji J, Nagai S, Nishita M, Meneghini M, Barker N, Waterman M, Bowerman B, Clevers H, Shibuya H, Matsumoto K. The TAK1-NLK-MAPK-related pathway antagonizes signalling between beta-catenin and transcription factor TCF. Nature. 1999;399:798–802. doi:10.1038/21674.PubMedCrossRefGoogle Scholar
- Ishitani T, Kishida S, Hyodo-Miura J, Ueno N, Yasuda J, Waterman M, Shibuya H, Moon RT, Ninomiya-Tsuji J, Matsumoto K. The TAK1-NLK mitogen-activated protein kinase cascade functions in the Wnt-5a/Ca(2+) pathway to antagonize Wnt/beta-catenin signaling. Mol Cell Biol. 2003b;23:131–9.PubMedPubMedCentralCrossRefGoogle Scholar
- Ji J, Yamashita T, Budhu A, Forgues M, Jia HL, Li C, Deng C, Wauthier E, Reid LM, Ye QH, Qin LX, Yang W, Wang HY, Tang ZY, Croce CM, Wang XW. Identification of microRNA-181 by genome-wide screening as a critical player in EpCAM-positive hepatic cancer stem cells. Hepatology. 2009;50:472–80. doi:10.1002/hep.22989.PubMedPubMedCentralCrossRefGoogle Scholar
- Jung KH, Kim JK, Noh JH, Eun JW, Bae HJ, Xie HJ, Ahn YM, Park WS, Lee JY, Nam SW. Targeted disruption of Nemo-like kinase inhibits tumor cell growth by simultaneous suppression of cyclin D1 and CDK2 in human hepatocellular carcinoma. J Cell Biochem. 2010;110:687–96. doi:10.1002/jcb.22579.PubMedCrossRefGoogle Scholar
- Kanei-Ishii C, Ninomiya-Tsuji J, Tanikawa J, Nomura T, Ishitani T, Kishida S, Kokura K, Kurahashi T, Ichikawa-Iwata E, Kim Y, Matsumoto K, Ishii S. Wnt-1 signal induces phosphorylation and degradation of c-Myb protein via TAK1, HIPK2, and NLK. Genes Dev. 2004;18:816–29. doi:10.1101/gad.1170604.PubMedPubMedCentralCrossRefGoogle Scholar
- Ke H, Masoumi KC, Ahlqvist K, Seckl MJ, Rydell-Törmänen K, Massoumi R. Nemo-like kinase regulates the expression of vascular endothelial growth factor (VEGF) in alveolar epithelial cells. Sci Rep. 2016;6. doi:10.1038/srep23987.Google Scholar
- Kojima H, Sasaki T, Ishitani T, Iemura S, Zhao H, Kaneko S, Kunimoto H, Natsume T, Matsumoto K, Nakajima K. STAT3 regulates Nemo-like kinase by mediating its interaction with IL-6-stimulated TGFbeta-activated kinase 1 for STAT3 Ser-727 phosphorylation. Proc Natl Acad Sci U S A. 2005;102:4524–9. doi:10.1073/pnas.0500679102.PubMedPubMedCentralCrossRefGoogle Scholar
- Mirkovic I, Gault WJ, Rahnama M, Jenny A, Gaengel K, Bessette D, Gottardi CJ, Verheyen EM, Mlodzik M. Nemo kinase phosphorylates β-catenin to promote ommatidial rotation and connects core PCP factors to E-cadherin-β-catenin. Nat Struct Mol Biol. 2011;18:665–72. doi:10.1038/nsmb.2049.PubMedPubMedCentralCrossRefGoogle Scholar
- Sa JK, Yoon Y, Kim M, Kim Y, Cho HJ, Lee JK, Kim GS, Han S, Kim WJ, Shin YJ, Joo KM, Paddison PJ, Ishitani T, Lee J, Nam DH. In vivo RNAi screen identifies NLK as a negative regulator of mesenchymal activity in glioblastoma. Oncotarget. 2015;6:20145–59. doi:10.18632/oncotarget.3980.PubMedPubMedCentralCrossRefGoogle Scholar
- Todd TW, Kokubu H, Miranda HC, Cortes CJ, La Spada AR, Lim J. Nemo-like kinase is a novel regulator of spinal and bulbar muscular atrophy. Elizab Theatr. 2015;4:e08493. doi:10.7554/eLife.08493.Google Scholar
- Yamada M, Ohnishi J, Ohkawara B, Iemura S, Satoh K, Hyodo-Miura J, Kawachi K, Natsume T, Shibuya H. NARF, an nemo-like kinase (NLK)-associated ring finger protein regulates the ubiquitylation and degradation of T cell factor/lymphoid enhancer factor (TCF/LEF). J Biol Chem. 2006;281:20749–60. doi:10.1074/jbc.M602089200.PubMedCrossRefGoogle Scholar
- Zhang ZY, Li SZ, Zhang HH, Wu QR, Gong J, Liang T, Gao L, Xing NN, Liu WB, Du RL, Zhang XD. Stabilization of ATF5 by TAK1-Nemo-like kinase critically regulates the interleukin-1β-stimulated C/EBP signaling pathway. Mol Cell Biol. 2015;35:778–88. doi:10.1128/MCB.01228-14.PubMedPubMedCentralCrossRefGoogle Scholar