Nonmuscle Myosin II
Heavy chain: NMMHC-II; MHCII; zipper (in Drosophila). Three isoforms are expressed in mammalian cells, also known as: (1) MHCII-A; myosin-9= MHA= FTNS= EPSTS= BDPLT6= DFNA17= NMMHCA= NMHC-II-A= NMMHC-IIA; (2) MHCII-B=myosin-10= NMMHCB= NMMHC-IIB; (3) MHCII-C=myosin-14= DFNA4= DFNA4A= FP17425= MHC16= MYH17= NMHC II-C= NMHC-II-C= PNMHH= myosin. Arabic numbers are now used too, e.g., myosin 2
Regulatory light chain: RLC; MLC; MRLC; MYRL; spaghetti squash (in Drosophila). Three isoforms are expressed in mammalian cells, also known as: (1) SM (smooth muscle)-RLC= MYL9= LC20= MLC-2C= MLC2= MRLC1= MYRL2; (2) MYL12A= HEL-S-24= MLC-2B= MLCB= MRCL3= MRLC3= MYL2B; (3) MYL12B= MLC-B=MRLC2
Essential light chain: MYL6= ELC= MLC= MLC-ALK= MLC1 (in Drosophila)
Functional unit (2×HC+2×RLC+2×ELC): NMII, NMMII, cytoplasmic myosin II
The discovery of nonmuscle myosin II (NMII) is linked to the story of muscle contraction (reviewed in Szent-Gyorgyi 2004). It started in 1864, when Köhler purified a viscous substance (“myosin”) from muscle, which he thought related to the contractile properties of muscle. In 1939, myosin was identified as an ATPase made of two major components, myosin and actomyosin. This led to the seminal idea that myosin and actin formed two types of interacting filaments. In the 1950s, proteolytic studies revealed the main functional parts of myosin: a heavy meromyosin that interacted with actin and contained the ATPase activity and a light meromyosin that enabled the formation of myosin filaments. The myosin light chains were also identified around this time. These experiments, together with crystallographic and functional evidence of the different bands that comprise the muscle sarcomere, allowed Huxley to propose that muscle contraction was due to the sliding of actin and myosin filaments with respect to each other. The concept that myosin and ATP existed in different binding states was incorporated into this early model to formulate the myosin swinging cross-bridge model, which is the core of the function of muscle (and nonmuscle) myosin II activity (Huxley 1969). This model was the culmination of a hundred years of investigation, and it also spurred the interest of many cell biologists and physiologists in contractile events observed in nonmuscle cells, for example migrating cells and clot formation and retraction, and also in morphogenetic movements, for example gastrulation. This led to the rapid identification of nonmuscle myosin II (NMII) in platelets (Adelstein et al. 1971) and also in amoebae (Pollard and Korn 1973a, b). NMII is a ubiquitous protein expressed in all mammalian tissues as well as in simple organisms. Like muscle myosin II, NMII is endowed with ATPase activity and the ability to form filaments with actin. The main mechanism of action of NMII is also similar to that of muscle myosin II. It involves an ATP-powered conformational change of the myosin head (Spudich 2001). Such movement is conjoined with its interaction with actin. The conformational change of the myosin head while tethered to the actin produces the movement of the filament or exerts tension on it. This property has positioned NMII at the center of the mechanotransduction field as the major force and tension generator inside nonmuscle cells (Aguilar-Cuenca et al. 2014). In this chapter, we outline the major mechanisms controlling NMII ATPase and actin-binding activity and its assembly to form filaments as well as the emerging picture of downstream pathways that are regulated by NMII-generated mechanical force.
NMII Structure and Assembly
Regarding the coiled coil and nonhelical tail domain, these regions are thought to support the lateral interaction of the NMII hexamer with the equivalent region of other NMII hexamers to assemble antiparallel filaments (mini-filaments) that contain multiple hexamers (Ricketson et al. 2010), forming proto-sarcomeric structures in many cell types, particularly epithelial and mesenchymal cells. The size and stability of these mini-filaments depend of the isoform of the heavy chain. Although small filaments with a mixed composition have been described in cultured cells (Beach et al. 2014; Billington et al. 2015), many NMII filaments present in live cells contain only one type of hexamer. However, assemblies of different NMII isoforms can assemble along the same actin bundles, forming a characteristic stippled pattern observed by confocal and super-resolution microscopy (Vicente-Manzanares et al. 2007; Beach et al. 2014).
Mechanism of Filament Sliding and Force Generation
Two major regulatory checkpoints of NMII exist. One is the direct control of the actin binding/ATPase activity/ conformational movement of the NMII hexamer, which mainly takes place at the head and neck regions, including the RLC. The other is the control of the assembly state of the NMII hexamers (NMII filament growth/ disassembly equilibrium) as well as their subcellular localization. These two levels of control constitute the core of this review and are described in the following section.
Regulation of NMII Assembly and Activity
RLC phosphorylation is the central regulatory event that controls NMII activity. The regulatory hotspot of the RLC is Ser19, which is responsible for the extension of the 10S NMII hexamer into the 6S assembly-competent form (Rosenfeld et al. 1994). It also promotes the activation of the motor ATPase activity of the NMII (Adelstein and Conti 1975). Additional phosphorylation at Thr18 further increases the actomyosin enzymatic activity and favors filament formation. In live cells, NMII phosphorylated in both Thr18 and Ser19 mainly appears in thick, stable actomyosin bundles, whereas NMII phosphorylated only in Ser19 is predominantly located in thinner, more dynamic filaments (Vicente-Manzanares and Horwitz 2010). Different kinases promote the phosphorylation of the RLC in Thr18 and Ser19. A major activating pathway emerges from the small Rho GTPase RhoA. When bound to GTP, RhoA binds to and activates ROCK (RhO-associated Coiled-coil Kinase). ROCK may phosphorylate RLC directly on Ser19 (Amano et al. 1996). However, the main function of ROCK in this context is the phosphorylation and inactivation of MYPT1, which is the catalytic subunit of a phosphatase specific for the RLC of NMII. In this manner, ROCK triggers NMII activation by promoting RLC phosphorylation directly and also by preventing its dephosphorylation. In addition, MYPT1 can be regulated in other ways, for example it can be phosphorylated and inhibited by ZIPK (Ichikawa et al. 1996). Also, the Arf GEFs Big1 and Big2 regulate its incorporation into complexes, thereby controlling its access to NMII (Le et al. 2013). CitK (citron kinase) is a RhoA-dependent cell cycle-related kinase that phosphorylates both Ser19 and Thr18 and controls the mitotic function of NMII (Yamashiro et al. 2003). Another Rho family member, Cdc42, activates MRCK (Myotonic dystrophy-Related Coiled-Coil Kinase), which also phosphorylates RLC on Thr18 and Ser19 (Leung et al. 1998). NMII is also directly regulated by calcium through the activation of the myosin light chain kinase (MLCK). This kinase is activated by Ca2+-calmodulin binding, and it phosphorylates RLC preferentially on Ser19 (Sellers et al. 1981). These kinases constitute a spatially segregated regulatory network, by which NMII is differentially activated throughout the cell. For example, MRCK and MLCK activate NMII closer to the leading edge (Totsukawa et al. 2004; Tan et al. 2008), whereas ROCK drives the activation of NMII in thick actomyosin bundles, which define the trailing edge of the cell (Totsukawa et al. 2004). Additional regulatory sites exist in the RLC, for example Ser1/Ser2. The phosphorylation of these Ser by PKC inhibits the function of NMII by decreasing the binding of NMII to actin filaments (Nishikawa et al. 1984; Komatsu and Ikebe 2007).
Regarding the control of NMII assembly into filaments, this is spatially mediated by the last loops of the α-helix coiled-coil rod domain and C-terminal nonhelical tail of the MHCII (∼last 200 amino acids of the MHCII). These regions interact laterally (mainly in an antiparallel fashion) with binding motifs in other NMII hexamers to mediate filament assembly. This interaction relies on intermolecular electrostatic interactions between negatively and positively charged regions that reside in this region of the MHCII. Importantly, this region contains most regulatory residues that underlie isoform specificity, thereby promoting homotypic interactions. This means that many mini-filaments contain only one type of paralog. Several kinases can regulate this activity, including TRPM (transient receptor potential melastatin)-6 and -7, members of the PKC (protein kinase C) family and CK2 (casein kinase 2). Regulation of these residues through phosphorylation introduces negative charges in selected regions of the different isoforms, which potentially destabilizes the electrostatic interactions between the chains and generally causes NMII disassembly and actomyosin filament instability that leads to cytoskeletal reorganization (Ricketson et al. 2010). TRPM7 also regulates NMII through the regulation of Mg2+ homeostasis (Stritt et al. 2016). These phosphorylations control the interaction of NMII with additional proteins. For example, NMII-A interacts with S100A4 (Mts1, metastatin-1) through a binding motif that comprises the last loops of the coiled-coil domain and the nonhelical tail domain. This interaction is regulated by phosphorylation of Ser1943 of NMII-A. Although this residue is not within the S100A4-NMIIA binding interface, its phosphorylation and the subsequent addition of negative charge may alter the conformation of the C-terminus of NMII-A, thereby preventing the interaction (Dulyaninova et al. 2005). Another interacting protein is Lgl (Lethal giant larvae), which is a tumor suppressor protein that binds to the C-terminal of both MHCII-A and -B under the control of aPKCζ and inhibits NMII-A assembly in live cells (Dahan et al. 2012). Myosin binding protein H (MYBPH) can directly interact with NMII to inhibit cell migration (Hosono et al. 2012). Additional regulators of NMII include various tropomyosin isoforms. Although muscle tropomyosin inhibits muscle myosin II binding to actin filaments, nonmuscle tropomyosin isoforms have variable effects on NMII function, depending on the tropomyosin and NMII isoform (Barua et al. 2014). Also, supervillin and anillin can bind NMII and control its function in cell division (Smith et al. 2013).
Pharmacological Control of NMII Function
A specific NMII inhibitor exists, blebbistatin (Straight et al. 2003). The (−) isomer of blebbistatin blocks the ATPase activity of NMII (Shu et al. 2005). In this manner, blebbistatin initially blocks contraction (the power stroke does not occur) and later detaches NMII from actin (Kovacs et al. 2004). Blebbistatin is not isoform-specific, and although its original form was phototoxic upon blue or UV light illumination (Kolega 2004), a recent derivative, para-nitroblebbistatin, is not (Kepiro et al. 2014). Pharmacological interventions are usually carried out at the regulatory level. For example, there are several specific inhibitors for ROCK, for example Y27632 and HA-1077 (fasudil) (Uehata et al. 1997). The latter is used in several countries to treat vasospasm, stroke, and cognitive decline in Alzheimer’s patients. There is an inhibitor for MLCK, ML-7. Inhibition of RhoA activity by ADP-ribosylation with C3 exoenzyme from Clostridium botulinum phenocopies many of the effects of direct NMII inhibition (Ridley and Hall 1992), but it has additional effects due to the role of RhoA as a controller of additional pathways involved in gene expression (Hill et al. 1995). Pharmacological activation of NMII can be elicited using a toxin produced by the Japanese sponge Discodermia calyx, calyculin A. This is a phosphatase inhibitor with moderate to low specificity for MYPT1 (Ishihara et al. 1989). On the other hand, no specific inhibitors of NMII assembly have been described yet. A recent study has identified a small molecule, 4-hydroxyacetophenone (4-HAP), which stabilizes NMII-B assemblies (Surcel et al. 2015), but the mechanism of such an effect remains unclear at present.
NMII-Dependent, Force-Sensitive Pathways
Mechanical regulation of gene expression is another point of great interest. Several transcription factors are activated in response to NMII-dependent mechanical signals, for example Yap/Taz or Nkx2.5 (Piccolo et al. 2014; Dingal et al. 2015). The precise mechanism of activation is still unclear, but it requires the translocation of the transcription factor from the cytoplasm to the nucleus, which is promoted by NMII-dependent forces. It also requires the binding of the transcription factor to its DNA targets in the nucleus, which depends on the degree of chromatin extension. Chromatin extension could be controlled by NMII-dependent forces through the connection of the cytoplasmic cytoskeleton to the nucleoskeleton through nuclear nesprin/SUN/KASH complexes (Fig. 4b, c). The connection of nucleoskeleton proteins to chromatin suggests that these forces may expose loops of condensed chromatin DNA, which would become accessible to mechano-sensitive transcription factors, enabling the expression of specific genes in response to mechanical cues. This is how stiff environments are thought to promote stem cell differentiation into osteoblasts: a stiff substrate would activate NMII, which would transmit mechanical forces to chromatin through nuclear complexes, exposing DNA binding sites related to genes involved in osteogenesis. Although the mechanisms that control transcription factor shuttling in and out of the nucleus in response to mechanical forces are still unclear, some mediators have been identified in specific cell types, for example E-cadherin/β-catenin in epithelia (Benham-Pyle et al. 2015).
Mechanically driven gene expression is clearly an area of rapid development that aims to define the mechano-responsive elements in the nucleus that drive chromatin exposure under different mechanical stimuli. It is interesting to note that some of the mechano-responsive transcription factors identified so far, for example Yap/Taz and Nkx2.5, are also implicated in morphogenetic pathways that control the growth of tissues to their proper size. This suggests that the relationship between the shape of individual cells and tissues as a whole is conjoined by the convergence of mechanical signals and genetic responses.
A mutation in the gene Myh10 has been recently identified, with a phenotype consisting of intrauterine growth restriction, microcephaly, developmental delay, failure to thrive, congenital bilateral hip dysplasia, cerebral and cerebellar atrophy, hydrocephalus, and congenital diaphragmatic hernia (CDH) (Tuzovic et al. 2013). This mutation phenocopies some aspects of mice strains carrying loss-of-function mutations to this gene (Ma et al. 2004). Finally, autosomal dominant hearing impairment (DFNA4) is associated to a mutation in Myh14. Myh14 encodes the heavy chain isoform II-C. This isoform is expressed in the cochlea and several mutations are associated to DFNA4, including 20c>a = S7X (stop mutation in the motor domain); 1126 g>t = G376C and 2176c>a = R726S (both in the motor domain); and 2926c>t = L976F (in the coiled-coil domain) (Donaudy et al. 2004).
Nonmuscle myosin II has emerged as a central protein at the interface between chemistry and physics in terms of the control of the cellular behavior. This is mainly due to its ability to generate mechanical force and actin filament work, as well as its actin cross-linking properties. Two major mechanisms of control exist, related to the actin-binding and ATPase activity of the molecule as well as the stability of the NMII subunits in filaments. In addition to the human syndromes associated to single-site mutations in the MHCII, NMII plays central roles in cancer and inflammation as it controls cancer cell division, the physical integrity of tumor masses, their dissemination, and their implantation at distant sites from the original tumor. It also determines the mechanical properties of the extracellular microenvironment, which may promote, or inhibit, the development of these diseases. Finally, NMII also participates in leukocyte proliferation and migration, that is, the systemic response of multicellular organisms against external aggression or anomalous growth. Although too central to be a useful pharmacological target to treat disease so far, its complex regulation together with its ubiquity in terms of cellular function and tissue expression makes NMII a crucial mediator of most cellular responses and a lynchpin of the biology of complex organisms.
We regret the large number of original articles that were left out due to space constraints. Dr. Vicente-Manzanares would like to thank Prof. Robert S. Adelstein, Dr. Mary Anne Conti and Dr. Xuefei Ma for critical reading of the manuscript and their advice, help and warm friendship over the years. This work was funded by grant SAF2014-54705-R from MINECO. Miguel Vicente-Manzanares is a Ramon y Cajal Assistant Professor (RYC2010-06094).
- Cremo CR, Hartshorne DJ. Myosins: a superfamily of molecular motors. In: Coluccio LM, editor. Watertown: Springer; 2007. p. 171–222.Google Scholar