A fundamental process in embryonic development is the establishment of body axes, including the anteroposterior (A-P) axis, the dorsoventral (D-V) axis, and the left-right (L-R) axis. Back to 1940s, when signaling pathways and molecular mechanisms behind axis formation remained a mystery, a spontaneous mutation called Kinky (Fu Ki ) at the mouse Fused locus attracted attention due to its impact on a variety of developmental processes, especially on the formation of embryonic A-P axis(Gluecksohn-Schoenheimer 1949). Around 1990s, two other spontaneous alleles of the mouse Fused gene, Fused (Fu Fu ) and Knobbly (Fu Kb ), and one transgenic insertional mutant Fu Tg1 were identified (Jacobs-Cohen et al. 1984; Perry et al. 1995; Mary et al. 1996), all showing defects in the formation of embryonic axis. These Axin null mutants normally led to lethality at embryonic day 8–10, and in addition to causing embryonic axis duplications, they also led to neuroectodermal and cardiac abnormalities (Jacobs-Cohen et al. 1984; Perry et al. 1995; Mary et al. 1996). Meanwhile, during 1990s, the requirement of Wnt signaling pathway for the development of embryonic axes became increasingly clear and several studies confirmed the essential role of β-catenin, the central player of canonical Wnt signaling, in the establishment of embryonic axes (Heasman et al. 1994; Fagotto et al. 1996; Miller and Moon 1996). In 1997, Zeng and colleagues revealed involvement of the Fused gene in the Wnt signaling pathway and showed that the Fused gene affects the formation of embryonic axis through a negative regulation of the Wnt pathway (Zeng et al. 1997). They renamed the Fused gene as Axin for ax is in inhibition to distinguish it from another unrelated Drosophila gene fuse (Zeng et al. 1997). Soon after that, a number of studies confirmed Axin as an essential component of Wnt pathway and these studies also uncovered Axin as a key scaffold protein in a complex for directing β-catenin degradation (Behrens et al. 1998; Hart et al. 1998; Ikeda et al. 1998; Itoh et al. 1998; Kishida et al. 1998; Nakamura et al. 1998; Sakanaka et al. 1998). In 1998, two independent studies identified a homolog of Axin, designated as Conduction and Axil respectively (Behrens et al. 1998; Yamamoto et al. 1998). Soon, it was confirmed that Conductin and Axil are the same protein with similar function to that of Axin by negatively regulating Wnt pathway (Kikuchi 1999; Mai et al. 1999). To differentiate these two Axins, Axin is thereafter also known as Axin1, with Conductin/Axil as Axin2.
Structure of Axin
Axin in Canonical Wnt Signaling
Dual Roles of Axin in Transmitting Canonical Wnt Signals
The Wnt signaling pathway is highly conserved in organisms ranging from nematode worms to mammals, and it controls many fundamental physiological processes such as cell growth, tissue morphogenesis, and embryonic development (MacDonald et al. 2009). The signal transduction is launched by binding of a group of Wnt proteins to their membrane receptors, and according to whether β-catenin is essential for the signal transduction, the Wnt signaling pathway has been so far classified into two classes: canonical Wnt pathway (also known as Wnt/β-catenin pathway) and noncanonical Wnt pathway (including planar cell polarity pathway and Wnt/calcium pathway). In Wnt/β-catenin pathway (hereinafter referred to as Wnt pathway), when there is no Wnt stimulation, the central player β-catenin is sequestered in a complex called “β-catenin destruction complex” (hereinafter referred to as destruction complex), in which β-catenin is subjected to GSK3β/CK1-mediated phoshorylation followed by TRCP (transducin repeat containing E3 ubiquitin protein ligase)-mediated ubiquitination and subsequent proteasome-dependent degradation. When Wnt ligands bind to the membrane receptors consisted of frizzled and LRP5/6 (LDL receptor related protein 5/6), the destruction complex is functionally inhibited, which allows β-catenin to escape degradation and subsequently enter the nucleus where it forms a transcriptional complex with LEF1/TCF4 to induce target gene expression. Axin interacts directly with multiple key components in the destruction complex such as β-catenin, APC, and GSK3β (glycogen synthase kinase 3β), by which Axin brings these proteins together to efficiently regulate the activity of β-catenin.
Posttranslational Modifications of Axin
As the concentration-limiting component of the destruction complex, the function of Axin is subject to multiple posttranslational modifications including phosphorylation, ubiquitination, methylation, and sumolyation (Song et al. 2014). Generally, phosphorylated Axin catalyzed by GSK3β shows enhanced stability and increased affinities with its binding partners, while dephoshorylation has the reverse effect on Axin stability and function (Jho et al. 1999; Willert et al. 1999; Yamamoto et al. 1999; Strovel et al. 2000). In addition to GSK3β, CKI was also observed to phosphorylate Axin both in vitro and in vivo (Gao et al. 2002). Another study suggested that phosphorylation of Axin by CKI may promote Axin-GSK3β interaction, leading to a more active destruction complex; while protein phosphatase 1 (PP1) dephosphorylates Axin and reverses the effect conferred by CKI (Luo et al. 2007). Moreover, phosphorylation of Axin by Cyclin A/CDK2 was reported to increase its association with β-catenin (Kim et al. 2004). Recently, Axin was found to be also phosphorylated by CDK5, and this phosphorylation facilitates its interaction with GSK3β, which plays an essential role for axon development (Fang et al. 2011). On the other hand, PP2A and PP2C are the other two phosphatases that may mediate dephosphorylation of Axin, thereby suppressing the function of Axin and facilitating disassembly of the destruction complex (Willert et al. 1999; Strovel et al. 2000). A recent study by Kim and colleagues provides some insights into the mechanisms behind how phosphorylation/dephosphorylation may affect the interaction of Axin with its partners (Kim et al. 2013). Their studies indicate that phosphorylation of Axin may release an auto-inhibited conformation of Axin, and the Axin protein in this “open” state may have increased affinities with its binding partners, such as β-catenin and LRP5/6 (Kim et al. 2013).
In 2009, Stegmeier and colleagues revealed that poly-ADP-ribose modification (PARsylation) of Axin by Tankyrase (TNKS) plays a critical role in regulating Axin stability and function (Huang et al. 2009). Further studies indicated that PARsylated Axin is probably recognized by the E3 ligase RNF146 for ubiquitination and subsequent proteasomal degradation (Huang et al. 2009; Callow et al. 2011; Zhang et al. 2011). On the other hand, ubiquitin-specific protease 34 (USP34) is reported to counteract this TNKS-RNF146-meidated reaction on Axin, thereby promoting nuclear accumulation of Axin to positively influence β-catenin-mediated transcription (Lui et al. 2011). Ubiquitin ligase Smurf1 and its homologs protein Smurf2 are E3 ligases which both targets Axin for ubiquitination. However these two Smurfs exert distinct effects on Axin function: Smurf2-mediated ubiquitination causes Axin degradation and thus prompts Wnt signal transduction, while Smurf1-mediated ubiquitination of Axin disrupts the interaction of Axin with LRP5/6 rather than leads to its degradation, thereby downregulating Wnt signaling (Kim and Jho 2010; Fei et al. 2013). In addition to ubiquitination, sumoylation is also reported to play a role in regulating Axin function. An early study in 2002 showed that sumoylation at the extreme C-terminal six amino acids of Axin is required for Axin-mediated JNK activation but dispensable for Wnt signaling (Rui et al. 2002). Later, Kim and colleagues suggested that this sumoylation at the C-terminus protects Axin from ubiquitination and thus enhances its stability (Kim et al. 2008). In 2011, Cha and colleagues reported methylation of Axin catalyzed by PRMT1, which leads to enhanced Axin stability and increased Axin-GSK3β interaction (Cha et al. 2011).
Axin Functions beyond Wnt Signaling
In TGF-β signaling, Axin interacts directly with Smad2/3, facilitating Smad2/3 activation and thus TGF-β signal transduction (Furuhashi et al. 2001). In addition to reinforcing the function of the positive regulators of TGF-β signaling like Smad2/3, Axin is also found to facilitate TGF-β signal transduction through suppressing the function of the negative regulator Smad7. Axin brings Smad7 and the ubiquitin E3 ligase Arkadia together, thereby promoting Arkadia-mediated degradation of Smad7 and prompting TGF-β signaling (Liu et al. 2006). Axin is also involved in the regulation of the basal activity of Smad3 through facilitating GSK3β-mediated Smad3 phosphorylation and its subsequent degradation in the absence of TGF-β signals (Guo et al. 2008).
Axin interacts with a couple of important components in JNK pathway. Axin utilizes distinct regions to interact with MEKK1 and MEKK4 and this may confer Axin the ability to regulate JNK activation according to distinct signals or in different cellular contexts (Zhang et al. 1999; Luo et al. 2003). On the other hand, the interaction between Axin and the I-mfa domain proteins HIC and I-mfa blocks Axin-MEKK interaction, thus decreasing Axin-mediated JNK activation (Kusano and Raab-Traub 2002). The interaction of HIC and I-mfa with Axin may also affect the activity of Axin in Wnt signaling, through affecting the formation of Axin-GSK3β-catenin complex (Kusano and Raab-Traub 2002). As mentioned before, Axin is heavily sumoylated at its extreme C-terminal six amino acids. Axin binds to the SUMO-1-conjugating enzyme E3 PIAS (protein inhibitor of activated STAT) including PIAS1, PIASxβ, and PIASy through its C-terminal region. Sumoylation of Axin by these PIAS family proteins is essential for Axin to promote JNK activation (Rui et al. 2002).
Axin interacts directly with p53. In addition, Axin bridges both positive and negative regulators to the p53 complex to regulate p53 activation. Knockdown of Axin by siRNA leads to impaired activation of p53, highlighting the functional significance of Axin in p53 pathway (Rui et al. 2004; Li et al. 2007). Axin directly interacts with HIPK2, a key kinase in activating p53, which brings HIPK2 close to the Axin-bound p53 protein to facilitated HIPK2-mediated p53 phosphorylation and activation. Meanwhile, Axin-HIPK2 interaction may induce a conformational change of HIPK2, increasing HIPK2 activity to phosphorylate and activate p53 (Rui et al. 2004). In addition to p53 and HIPK2, Axin is reported to bind directly to other regulators including PML (Promyelocytic leukemia protein), Tip60, Pirh2, (Li et al. 2007, 2009, 2011) and Daxx (Death domain-associated protein 6). As a key scaffold protein, Axin brings these proteins close to form a functional complex, in which these proteins cooperate with each other to keep p53 activity under tight control.
In addition to the above signaling pathways, Axin is also involved in regulating other pathways. For example, Axin directly interacts with LKB1 and AMPK holoenzyme, facilitating LKB1-mediated phosphorylation of AMPK and AMPK activation (Zhang et al. 2013). Axin also joins up GSK3β, Pin, and PP2A-B56 to form a degradation for c-Myc, facilitating proteasomal degradation of c-Myc and suppressing cell proliferation. Of note, the studies of Axin in different pathways consistently relate Axin to a function of promoting apoptosis, suggesting that Axin may serve as a general regulator to suppress cellular proliferation.
Axin as a Drug Target
Considering the significant role of Wnt signaling in many aspects of controlling cell behaviors, it is not surprising that dysregulation of Wnt signaling would lead to tumors as well as other diseases like bone disease and metabolic disease (Clevers and Nusse 2012). So far, mutation or aberrant expression of components of Wnt signaling has been closely related to the generation and progression of a number of human cancers (Polakis 2012). A striking example is in colorectal cancer, in which alterations of APC have been observed in ~90% of tumor samples (Miyaki et al. 1994; Najdi et al. 2011). Around 2000, several mutations in Axin1 and Axin2 were firstly identified in hepatocellular carcinoma (HCC) and hepatoblastoma (HB) (Taniguchi et al. 2002). Soon thereafter, impaired activation of Axin was implicated in a variety of other human cancers such as lung cancer (Hughes and Brady 2006; Kanzaki et al. 2006), breast cancer (Ozaki et al. 2005), colon cancer (Lammi et al. 2004; Parveen et al. 2011; Yang et al. 2013), and prostate cancer (Yardy et al. 2009). Despite these findings, the role of Axin as a drug target has not been realized until the discovery of Tankyrases (TNKS) inhibitors. In a large-scale screening to find drug leads against Wnt signaling, a small-molecule inhibitor IWR-1 was identified (Chen et al. 2009). Later, another inhibitor XAV939 was identified and it was then revealed that both IWR-1 and XAV939 target TNKS and inhibit the enzyme activity of TNKS (Huang et al. 2009). As mentioned above, TNKS mediates PARsylation of Axin, leading to ubiquitination and degradation of Axin (Huang et al. 2009). Thus inhibition of TNKS could improve Axin stability and promote Axin-mediated degradation of β-catenin. Since identification of the initial TNKS inhibitors of IWR-1 and XAV939, a couple of new TNKS inhibitors have been identified including WIKI4, JW55, G007-LK, and G244-LM, and they all show promising antitumor activity through stabilizing Axin (Huang et al. 2009; James et al. 2012; Waaler et al. 2012; Lau et al. 2013). In 2012, the 3D structure of TNKS in complex with Axin was determined (Morrone et al. 2012). This structure is expected to facilitate the development of more specific modulators that target the TNKS-Axin interface. In addition to TNKS inhibitors, two other small-molecule compounds, SKL2001 and HLY78, have been identified through high-throughput screening with Axin as their cellular target. However, unlike TNKS inhibitors, these two compounds promote activation of Wnt signaling instead of blocking it, although they do it through different mechanisms (Gwak et al. 2012; Wang et al. 2013). SKL2001 disrupts the interaction between Axin and β-catenin, thus increasing the stability of β-catenin; while HLY78 induces a conformational change of Axin, leading to an enhanced Axin-LRP5/6 interaction. SKL2001 and HLY78 show promising activities in the aspects like stem cell renewal and bone fracture repair, in which enhancement of Wnt signaling is preferred.
Axin was initially established as a negative regulator of Wnt signaling. In the Wnt-off state, Axin interacts with multiple key components in the destruction complex to facilitate the degradation of β-catenin; whereas in the Wnt-on state, Axin moves to interact with the membrane receptor LRP5/6, by which Axin brings GSK3β/CK1 close to LRP5/6, leading to LRP5/6 phosphorylation and thus promoting Wnt signal transduction. Axin is well known as a key component of Wnt signaling, however its master scaffolding function is hardly limited to the Wnt signaling. Current findings have related Axin to several other signaling pathways such as JNK, P53, and TGF-β. With the discoveries of new binding partners of Axin, more functions of Axin are expected to be uncovered in the future. Based on the current knowledge, Axin may serve as a connecting point among different signaling pathways for a negative regulation of cell growth. Aberrant activation of Axin has so far been implicated in various human cancers especially in hepatocellular and colorectal cancer. The discovery of TNKS inhibitors which show potent antitumor activities by increasing Axin stability further points out Axin as a promising druggable target for cancer treatment. Axin has no enzymatic activity itself and it functions through interacting with different sets of binding partners to affect various signaling pathways that are closely related to pathogenesis of human disease. Hence modulators targeting those Axin-mediated protein-protein interactions (PPIs) may have potentials as drug target for treating human disease. However, before that, more studies and clinical data are required to classify the specific roles of those Axin-mediated PPIs in human disease.
- Fiedler M, Mendoza-Topaz C, Rutherford TJ, Mieszczanek J, Bienz M. Dishevelled interacts with the DIX domain polymerization interface of Axin to interfere with its function in down-regulating beta-catenin. Proc Natl Acad Sci USA. 2011;108:1937–42. doi: 10.1073/pnas.1017063108.PubMedCentralCrossRefPubMedGoogle Scholar
- Ikeda S, Kishida S, Yamamoto H, Murai H, Koyama S, Kikuchi A. Axin, a negative regulator of the Wnt signaling pathway, forms a complex with GSK-3beta and beta-catenin and promotes GSK-3beta-dependent phosphorylation of beta-catenin. EMBO J. 1998;17:1371–84. doi: 10.1093/emboj/17.5.1371.PubMedCentralCrossRefPubMedGoogle Scholar
- Li Q, He Y, Wei L, Wu X, Wu D, Lin S, et al. AXIN is an essential co-activator for the promyelocytic leukemia protein in p53 activation. Oncogene. 2011;30:1194–204. doi: 10.1038/onc.2010.499.
- Lyon MF, Rastan S, Brown SDM, editors. Genetic variants and strains of the laboratory mouse. Oxford: Oxford University Press; 1996.Google Scholar
- Polakis P. Wnt signaling in cancer. Cold Spring Harb Perspect Biol 2012;4. doi: 10.1101/cshperspect.a008052.
- Waaler J, Machon O, Tumova L, Dinh H, Korinek V, Wilson SR, et al. A novel tankyrase inhibitor decreases canonical Wnt signaling in colon carcinoma cells and reduces tumor growth in conditional APC mutant mice. Cancer Res. 2012;72:2822–32. doi: 10.1158/0008-5472.CAN-11-3336.CrossRefPubMedGoogle Scholar