MICAL redox enzymes and actin remodeling: New links to classical tumorigenic and cancer pathways

ABSTRACT

MICAL Redox enzymes have recently emerged as direct regulators of cell shape and motility – working through specific reversible post-translational oxidation of actin to disassemble and remodel the cytoskeleton. Links are also now emerging between MICALs and cancer, including our recent results that regulation of MICAL sensitizes cancer cells to the cancer drug Gleevec. Targeting this new actin regulatory enzyme system may thus provide new therapeutic options for cancer treatment.

Cancer is a complicated disease that affects all tissues of the body and is characterized by the accumulation of a variety of adverse genetic alterations in normal cells. At their core, however, human cancers share a small number of attributes: including the ability to modulate cell proliferation, replicate endlessly, avoid programmed cell death, promote angiogenesis, and mobilize throughout the body.¹ Defining the molecular signals underlying these acquired traits is therefore imperative to the development of effective strategies to prevent, diagnose, and treat cancer.

There is a growing realization that cells utilize similar molecular machinery regulating their proliferation, differentiation, mobility, and death. For example, there is an expanding list of proteins initially identified for their functional role in axon guidance and neuronal development that are now being characterized for their roles in the morphogenesis of other tissues. These so called “axon guidance cues” and their receptors – including the classically designated axon guidance cues Ephrins and their Eph receptors, Slits and their Robo receptors, Netrins and their DCC and Unc5 receptors, and Semaphorins and their Neuropilin and Plexin receptors – have all been identified as being involved in the formation of normal tissue and in the development of cancerous tissue.² This co-involvement has prompted us to investigate how our research observations into the molecular and biochemical mechanisms controlling cell morphology, motility, and navigation (e.g.,^3–5) might offer new strategies for cancer prevention, diagnosis, and treatment.

We have focused on using the Semaphorins and Plexins⁶ as a model to understand how cells change their shape and are guided to their targets in response to guidance cues. Recently, we identified a new family of proteins the MICALs – and we find that the MICALs are Redox enzymes that mediate Semaphorin (Sema) signaling (Fig. 1A; reviewed in ref⁷). Furthermore, we find that Mical proteins bind to actin and utilize their Redox enzymatic activity to directly disassemble actin filaments – utilizing F-actin as a direct substrate, which the MICALs stereospecifically oxidize on two specific actin methionine residues to dismantle F-actin (Fig. 1B;^3–5). This work identified MICAL as an unusual new regulator of the actin cytoskeleton, the structure underlying many aspects of both normal and cancer cell behavior.

Figure 1.

New links between members of the MICAL family of Redox actin remodeling enzymes and Abl tyrosine kinases – including roles in orchestrating cancer cell invasion, colony formation, and survival. (A) The MICAL family of multidomain monooxygenase (Redox) enzymes are present in invertebrates and vertebrates with one family member in Drosophila (D) and three family members (1, 2, and 3) present in mammals including humans. In addition to their Redox domain, which binds the co-factor flavin adenine dinucleotide (FAD) and utilizes the co-enzyme nicotinamide adenine dinucleotide phosphate (NADPH) in Redox reactions modifying actin, MICALs also have a calponin homology (CH) domain, a Lin11, Isl-1 & Mec-3 (LIM) domain, a stretch of proline-rich (PxxP) residues, and a C-terminus in which they interact with the Semaphorin guidance receptor Plexin (Plexin interacting region [PIR]). MICALs can also interact with other proteins such as Rab small GTPases using regions in their C-terminus (see ⁷ for review). Regions C terminal to the LIM domain are also variable in length between the MICALs. N (N-terminus), (C) C-terminus. (B) Semaphorins (Sema) and their Plexin receptors directly associate with Mical and utilize Mical to control the F-actin cytoskeleton.³ Mical (yellow) works by binding to F-actin (green) and employing its NADPH-dependent Redox activity to stereospecifically oxidize (O) the methionine (M) 44 and M47 residues on actin filament subunits.⁴ This oxidation results in both the disassembly of F-actin and the reduced ability of oxidized actin subunits to reassemble into filaments.⁴ Our recent results reveal that the non-receptor tyrosine kinase and oncogene Abl (cyan) also plays a role in regulating Mical's actin dismantling activity.⁹ In particular, we find that the PxxP motif of Mical associates with the SH3 domain of Abl. We also find that Mical serves as a substrate for the Abl kinase, which phosphorylates (P) a specific tyrosine residue (Y500) within the Redox domain of Mical. We find that this phosphorylation increases the enzymatic activity of Mical, enhancing Mical's F-actin dismantling ability. Our results also uncover that growth factors (purple) can further amplify the F-actin dismantling effects of Mical – including combinatorially acting with Semaphorins – and that together this signaling network directs multiple cellular effects, including extending and shaping cellular processes, guiding axons, and orchestrating cancer cell invasion, colony formation, and survival.

We and others have now begun to explore the myriad of questions surrounding this unique family of enzymes. For example, we initially uncovered Mical for its ability to direct the effects of Semaphorin/Plexin guidance cues (Fig. 1B), but we have also wondered if Mical is involved in more classically studied signaling pathways. We have therefore been conducting genetic, molecular, and biochemical screens to look for proteins that play a role with Mical in its enzymatic F-actin regulatory effects (e.g.,^{5, 8, 9}). Recently, we identified a physical interaction between Mical and the Abl non-receptor tyrosine kinase (Fig. 1B;⁹). Turning to biochemical assays, our results revealed that Abl phosphorylates Mical to directly amplify Mical's Redox-mediated F-actin disassembly (Fig. 1B;⁹). Furthermore, our use of genetic assays demonstrated that Abl increases Mical-mediated F-actin disassembly, cellular remodeling, and repulsive axon guidance.⁹ Interestingly, we also found that Abl, which is well-known to be linked to growth factors and their receptors, provides a means by which growth factors can impinge on Semaphorin guidance cue signaling, where they can together enhance the effects of Mical on the cytoskeleton (Fig. 1B;⁹).

In light of Abl's enhanced expression and activation in multiple tumors including breast, lung, colorectal, gastric, prostate, and melanoma,¹⁰ we wondered if Mical and Abl interactions were also important for cancer cell behaviors (Fig. 1B). Utilizing different breast cancer cell lines and in vivo xenograft assays as models revealed that MICAL-1 knockdown resulted in changes to the cytoskeletal organization, morphology, invasion, colony formation, and survival of cancer cells.⁹ Thus, these results point to a role for Mical in regulating different cancer behaviors – and also support previous work with MICALs including genome-wide studies revealing mutations in MICALs in cancer patients as well as roles in growth and survival of melanoma cells, prostate cancer progression, and gastric, breast, and renal cancers (reviewed in ref⁷).

Abl kinases contribute pathologically to multiple different cancers including leukemia, for which Gleevec (imatinib mesylate, STI571), an Abl kinase inhibitor, is a well-known therapeutic for the treatment of Abl-related cancers. Since MICALs also have known effects on several of these Abl-related cancers, we tested the interaction between Mical and Gleevec, using different breast cancer cell lines as models. Treating highly invasive MDA-MB-231 breast cancer cells with Gleevec in combination with MICAL-1 knockdown, promoted a ∼60% reduction in invasion compared with Gleevec treatment alone.⁹ We also saw a similar dramatic increase in sensitivity to Gleevec when MICAL-1 was knocked down in non-invasive, colony-forming MCF-7 breast cancer cells.⁹ Since de novo point mutations within Abl tyrosine kinases are one of the major hurdles to overcome in effective Abl-related cancer treatment, identification of other targets is needed for effective personalized therapy for Abl-related cancer patients. Since we find that decreasing MICAL-1 levels sensitizes cancer cells to Gleevec treatment, future work should focus on finding specific inhibitors of MICALs and testing if those inhibitors can be used in cancer treatment. Furthermore, in light of numerous studies revealing that overexpression of growth factors make tumor cells more aggressive, links between Mical – and its dramatic F-actin remodeling ability – and growth factors (Fig. 1B;⁹) should be further explored as targets for curtailing the mobility, metastasis, and proliferative capabilities of cancer cells.

Acknowledgments

We apologize for those whose work we could not directly cite because of space and reference limitations. This work was supported by NIH (MH085923) and Welch Foundation (I-1749) grants to J.R.T.