MK-STYX is part of a family of two known proteins in humans with a STYX domain. The other protein, simply named STYX, is also catalytically inactive, due to a Cys → Gly mutation within its active site (Wishart and Dixon 1998). Reversion of the glycine back to a cysteine is sufficient to restore catalytic activity, demonstrating the preservation of the phosphatase domain and binding pocket in this protein (Wishart et al. 1995). Due to this structural preservation, it has been speculated that the STYX domain could bind to phosphorylated proteins, analogous to an SH2 domain binding phosphotyrosine residues. This would be a particularly interesting possibility, as DUSPs are able dephosphorylate both tyrosine residues, as well as serine/threonine residues due to a particularly shallow pocket which accommodates the active site. Indeed, a recent study demonstrated that the mutation of two residues within the active site, the serine → cysteine along with the −1 position, is sufficient to promote phosphatase activity of MK-STYX (Hinton et al. 2010). These data suggest that while the enzyme is catalytically inactive, it retains a phosphatase binding pocket that could be important in its cellular functions.
Another interesting possibility of MK-STYX function is to regulate the localization or enzymatic activity of an active phosphatase, presumably a DUSP. Surprisingly, there are numerous phosphatases present in the human genome which are rendered catalytically inactive due to mutations at various points within their active sites (Tonks 2006). This includes the D2 domains of many receptor protein tyrosine phosphatases (R-PTPs), which are implicated in controlling enzymatic activity. Interestingly, a family of lipid phosphatases, the myotubularins, consists of 14 different members, 6 of which are catalytically inactive (Begley and Dixon 2005). Importantly, these catalytically inactive myotubularins are critical for modulating enzymatic activity and/or cellular sublocalization of their active counterparts, creating an additional regulatory layer into lipid phosphatase biology (Begley et al. 2006; Robinson et al. 2008). While a phosphatase interactor for STYX and/or MK-STYX has yet to be found, the possibility that these proteins could be similar to a regulatory module for active phosphatases is an attractive model with clear precedence within the cell.
MK-STYX and STYX Domains Throughout Evolution
MK-STYX seems to be a relatively recent molecular addition in the evolution of animals, as definitive homologues of the protein can only be found within the phylum Chordata, which includes deuterosomes such as the sea cucumber as well as zebrafish, mice, and humans. It is not conserved, however, in insects, C. elegans, or yeast. It is interesting to note, however, that “STYX-domain” containing proteins do exist in these organisms (Wishart and Dixon 1998). While they are probably not direct homologues of MK-STYX, it is notable that catalytically inactive phosphatases have been utilized in even early life forms, though little is known about the functionality of these genes.
As MK-STYX is relatively uncharacterized, it comes as no surprise that little is known about its regulation at a transcriptional or posttranscriptional level. There are multiple studies, however, that have demonstrated that MK-STYX transcript levels increase in the context of the Ewing’s Sarcoma fusion product, EWS-FLI (Guillon et al. 2009; Siligan et al. 2005). EWS-FLI is a transcription factor with aberrant-binding capacities which is known to be sufficient in causing Ewing’s Sarcoma. While MK-STYX transcript levels have been shown to increase to this aberrant fusion protein, presumably through transcriptional upregulation, little is known of the physiological significance of this observation, as MK-STYX has also been hypothesized to function as a tumor suppressor (see RNAi phenotypes, below).
MK-STYX has also been identified as being transcriptionally upregulated by the induction of the p21 protein (Chang et al. 2000), though this would have to be an indirect form of regulation, as p21 itself is not a direct transcriptional regulator. It is interesting, however, in light of the potential tumor suppressive functions of MK-STYX uncovered by MacKeigan et al., that p21 could upregulate MK-STYX to halt cell cycle progression or promote an apoptotic phenotype. The broad applicability p21-mediated control of MK-STYX expression will have to be confirmed and explored in much more detail before a clear understanding of these implications is uncovered.
RNAi-Mediated Phenotypes of MK-STYX Knockdown
Interestingly, MK-STYX has been identified as having numerous phenotypes in multiple RNAi screens designed to study very different cellular processes. The first study aimed to identify novel kinases and phosphatases involved in cellular survival or apoptotic potential (MacKeigan et al. 2005). The authors used siRNA sequences to all known and putative human kinases and phosphatases and transfected them into cancer cells. RNAi-mediated loss of MK-STYX promoted the most highly chemoresistant phenotype of all enzymes assayed. Importantly, this chemoresistance was shown in response to multiple drugs with different mechanisms of action, implicating a general cellular mechanism of chemoresistance. As chemoresistance is a highly significant clinical problem for patients with advanced and recurrent cancers, studies on the significance of this gene in treatment response and/or prediction of response rate could be an important future direction in oncology research.
A second RNAi-screening paper identified MK-STYX as potentially tumor suppressive in the context of breast cancer. This paper demonstrated that the RNAi-mediated loss of MK-STYX promoted a highly aberrant migratory phenotype in MCF-10A cells, a nontransformed mammary epithelial cell line (Simpson et al. 2008). This seemed to be coupled with a striking loss in cell polarity, which is a typical feature of cancer cells.
An additional independent study identified loss of MK-STYX within a set of genes whose downregulation is associated with breast cancer metastasis to the brain (Bos et al. 2009). In the study, two cell lines were passaged in vivo to create a daughter cell line which was highly metastatic to the brain. Importantly, both cell models had statistically significant downregulation of MK-STYX in the metastatic cell lines relative to the parental lines. This data, coupled with the RNAi-screening data, suggests that MK-STYX could be a potent tumor and/or metastasis suppressor in breast cancers.
While no molecular mechanism was worked out for either of these phenotypes, it is interesting to note that the loss of MK-STYX seems to promote both resistance to therapy, as well as a prometastatic phenotype. These data suggest that the loss of MK-STYX could be an important event in the later stages of cancer progression and could be a valuable therapeutic target if these observations validate in follow-up studies.
MK-STYX and Stress Granule Formation
Currently, the only known protein identified as an interaction partner of MK-STYX is the RNA-binding protein G3BP1. A recent study has shown that MK-STYX can bind to endogenous G3BP, a protein intimately involved in the formation and maintenance of stress granules within the cell (Hinton et al. 2010). Importantly, MK-STYX overexpression alleviated stress granule formation within cells, implicating a functional role to this interaction. Interestingly, the authors noted that the interaction between MK-STYX and G3BP is abrogated when MK-STYX is reverted to an active enzyme through the mutation of two key residues within its active site, suggesting that the MK-STYX-G3BP1 interaction axis is mediated through its STYX domain, potentially through a substrate-trapping mechanism. It will be interesting in the future to understand whether G3BP is phosphorylated and whether this phosphorylation is what mediates the interaction of these two proteins.
Summary and Future Directions
MK-STYX, while relatively uncharacterized, is suggested to have numerous independent and interesting cellular phenotypes. Many studies remain to be done to elucidate how a single gene could have such pleiotropic effects; subcellular localization, regulation, and turnover, as well as identification of interaction partners will be critical for these analyses. Although many facts remain to be uncovered about this gene, the few studies that have been done on this protein suggest that it could play a very interesting role in the etiology of diseases, such as cancer. As such, studies in the future should take note of this interesting gene, and efforts should be made to uncover its cellular function and what role it plays in the etiology of disease.