Concerted actions of DnaA complexes with DNA-unwinding sequences within and flanking replication origin oriC promote DnaB helicase loading

Unwinding of the replication origin and loading of DNA helicases underlie the initiation of chromosomal replication. In Escherichia coli, the minimal origin oriC contains a duplex unwinding element (DUE) region and three (Left, Middle, and Right) regions that bind the initiator protein DnaA. The Left/Right regions bear a set of DnaA-binding sequences, constituting the Left/Right-DnaA subcomplexes, while the Middle region has a single DnaA-binding site, which stimulates formation of the Left/Right-DnaA subcomplexes. In addition, a DUE-flanking AT-cluster element (TATTAAAAAGAA) is located just outside of the minimal oriC region. The Left-DnaA subcomplex promotes unwinding of the flanking DUE exposing TT[A/G]T(T) sequences that then bind to the Left-DnaA subcomplex, stabilizing the unwound state required for DnaB helicase loading. However, the role of the Right-DnaA subcomplex is largely unclear. Here, we show that DUE unwinding by both the Left/Right-DnaA subcomplexes, but not the Left-DnaA subcomplex only, was stimulated by a DUE-terminal subregion flanking the AT-cluster. Consistently, we found the Right-DnaA subcomplex–bound single-stranded DUE and AT-cluster regions. In addition, the Left/Right-DnaA subcomplexes bound DnaB helicase independently. For only the Left-DnaA subcomplex, we show the AT-cluster was crucial for DnaB loading. The role of unwound DNA binding of the Right-DnaA subcomplex was further supported by in vivo data. Taken together, we propose a model in which the Right-DnaA subcomplex dynamically interacts with the unwound DUE, assisting in DUE unwinding and efficient loading of DnaB helicases, while in the absence of the Right-DnaA subcomplex, the AT-cluster assists in those processes, supporting robustness of replication initiation.

The initiation of bacterial DNA replication requires local duplex unwinding of the chromosomal replication origin oriC, which is regulated by highly ordered initiation complexes. In Escherichia coli, the initiation complex contains oriC, the ATP-bound form of the DnaA initiator protein (ATP–DnaA), and the DNA-bending protein IHF (Fig. 1, A and B), which promotes local unwinding of oriC (1, 2, 3, 4). Upon this oriC unwinding, two hexamers of DnaB helicases are bidirectionally loaded onto the resultant single-stranded (ss) region with the help of the DnaC helicase loader (Fig. 1B), leading to bidirectional chromosomal replication (5, 6, 7, 8). However, the fundamental mechanism underlying oriC-dependent bidirectional DnaB loading remains elusive.Open in a separate windowFigure 1Schematic structures of oriC, DnaA, and the initiation complexes. A, the overall structure of oriC. The minimal oriC region and the AT-cluster region are indicated. The sequence of the AT-cluster−DUE (duplex-unwinding element) region is also shown below. The DUE region (DUE; pale orange bars) contains three 13-mer repeats: L-DUE, M-DUE, and R-DUE. DnaA-binding motifs in M/R-DUE, TT(A/G)T(T), are indicated by red characters. The AT-cluster region (AT cluster; brown bars) is flanked by DUE outside of the minimal oriC. The DnaA-oligomerization region (DOR) consists of three subregions called Left-, Middle-, and Right-DOR. B, model for replication initiation. DnaA is shown as light brown (for domain I–III) and darkbrown (for domain IV) polygons (right panel). ATP–DnaA forms head-to-tail oligomers on the Left- and Right-DORs (left panel). The Middle-DOR (R2 box)-bound DnaA interacts with DnaA bound to the Left/Right-DORs using domain I, but not domain III, stimulating DnaA assembly. IHF, shown as purple hexagons, bends DNA >160° and supports DUE unwinding by the DnaA complexes. M/R-DUE regions are efficiently unwound. Unwound DUE is recruited to the Left-DnaA subcomplex and mainly binds to R1/R5M-bound DnaA molecules. The sites of ssDUE-binding B/H-motifs V211 and R245 of R1/R5M-bound DnaA molecules are indicated (pink). Two DnaB homohexamer helicases (light green) are recruited and loaded onto the ssDUE regions with the help of the DnaC helicase loader (cyan). ss, single stranded.The minimal oriC region consists of the duplex unwinding element (DUE) and the DnaA oligomerization region (DOR), which contains specific arrays of 9-mer DnaA-binding sites (DnaA boxes) with the consensus sequence TTA[T/A]NCACA (Fig. 1A) (3, 4). The DUE underlies the local unwinding and contains 13-mer AT-rich sequence repeats named L-, M-, and R-DUE (9). The M/R-DUE region includes TT[A/G]T(A) sequences with specific affinity for DnaA (10). In addition, a DUE-flanking AT-cluster (TATTAAAAAGAA) region resides just outside of the minimal oriC (Fig. 1A) (11). The DOR is divided into three subregions, the Left-, Middle-, and Right-DORs, where DnaA forms structurally distinct subcomplexes (Fig. 1A) (8, 12, 13, 14, 15, 16, 17). The Left-DOR contains high-affinity DnaA box R1, low-affinity boxes R5M, τ1−2, and I1-2, and an IHF-binding region (17, 18, 19, 20). The τ1 and IHF-binding regions partly overlap (17).In the presence of IHF, ATP–DnaA molecules cooperatively bind to R1, R5M, τ2, and I1-2 boxes in the Left-DOR, generating the Left-DnaA subcomplex (Fig. 1B) (8, 17). Along with IHF causing sharp DNA bending, the Left-DnaA subcomplex plays a leading role in DUE unwinding and subsequent DnaB loading. The Middle-DOR contains moderate-affinity DnaA box R2. Binding of DnaA to this box stimulates DnaA assembly in the Left- and Right-DORs using interaction by DnaA N-terminal domain (Fig. 1B; also see below) (8, 12, 14, 16, 21). The Right-DOR contains five boxes (C3-R4 boxes) and cooperative binding of ATP–DnaA molecules to these generates the Right-DnaA subcomplex (Fig. 1B) (12, 18). This subcomplex is not essential for DUE unwinding and plays a supportive role in DnaB loading (8, 15, 17). The Left-DnaA subcomplex interacts with DnaB helicase, and the Right-DnaA subcomplex has been suggested to play a similar role (Fig. 1B) (8, 13, 16).In the presence of ATP–DnaA, M- and R-DUE adjacent to the Left-DOR are predominant sites for in vitro DUE unwinding: unwinding of L-DUE is less efficient than unwinding of the other two (Fig. 1B) (9, 22, 23). Deletion of L-DUE or the whole DUE inhibits replication of oriC in vitro moderately or completely, respectively (23). A chromosomal oriC Δ(AT-cluster−L-DUE) mutant with an intact DOR, as well as deletion of Right-DOR, exhibits limited inhibition of replication initiation, whereas the synthetic mutant combining the two deletions exhibits severe inhibition of cell growth (24). These studies suggest that AT-cluster−L-DUE regions stimulate replication initiation in a manner concerted with Right-DOR, although the underlying mechanisms remain elusive.DnaA consists of four functional domains (Fig. 1B) (4, 25). Domain I supports weak domain I–domain I interaction and serves as a hub for interaction with various proteins such as DnaB helicase and DiaA, which stimulates ATP–DnaA assembly at oriC (26, 27, 28, 29, 30). Two or three domain I molecules of the oriC–DnaA subcomplex bind a single DnaB hexamer, forming a stable higher-order complex (7). Domain II is a flexible linker (28, 31). Domain III contains AAA+ (ATPase associated with various cellular activities) motifs essential for ATP/ADP binding, ATP hydrolysis, and DnaA–DnaA interactions in addition to specific sites for ssDUE binding and a second, weak interaction with DnaB helicase (1, 4, 8, 10, 19, 25, 32, 33, 34, 35). Domain IV bears a helix-turn-helix motif with specific affinity for the DnaA box (36).As in typical AAA+ proteins, a head-to-tail interaction underlies formation of ATP–DnaA pentamers on the DOR, where the AAA+ arginine-finger motif Arg285 recognizes ATP bound to the adjacent DnaA protomer, promoting cooperative ATP–DnaA binding (Fig. 1B) (19, 32). DnaA ssDUE-binding H/B-motifs (Val211 and Arg245) in domain III sustain stable unwinding by directly binding to the T-rich (upper) strand sequences TT[A/G]T(A) within the unwound M/R-DUE (Fig. 1B) (8, 10). Val211 residue is included in the initiator-specific motif of the AAA+ protein family (10). For DUE unwinding, ssDUE is recruited to the Left-DnaA subcomplex via DNA bending by IHF and directly interacts with H/B-motifs of DnaA assembled on Left-DOR, resulting in stable DUE unwinding competent for DnaB helicase loading; in particular, DnaA protomers bound to R1 and R5M boxes play a crucial role in the interaction with M/R-ssDUE (Fig. 1B) (8, 10, 17). Collectively, these mechanisms are termed ssDUE recruitment (4, 17, 37).Two DnaB helicases are thought to be loaded onto the upper and lower strands of the region including the AT-cluster and DUE, with the aid of interactions with DnaC and DnaA (Fig. 1B) (25, 38, 39). DnaC binding modulates the closed ring structure of DnaB hexamer into an open spiral form for entry of ssDNA (40, 41, 42, 43). Upon ssDUE loading of DnaB, DnaC is released from DnaB in a manner stimulated by interactions with ssDNA and DnaG primase (44, 45). Also, the Left- and Right-DnaA subcomplexes, which are oriented opposite to each other, could regulate bidirectional loading of DnaB helicases onto the ssDUE (Fig. 1B) (7, 8, 35). Similarly, recent works suggest that the origin complex structure is bidirectionally organized in both archaea and eukaryotes (1, 46). In Saccharomyces cerevisiae, two origin recognition complexes containing AAA+ proteins bind to the replication origin region in opposite orientations; this, in turn, results in efficient loading of two replicative helicases, leading to head-to-head interactions in vitro (46). Consistent with this, origin recognition complex dimerization occurs in the origin region during the late M-G1 phase (47). The fundamental mechanism of bidirectional origin complexes might be widely conserved among species.In this study, we analyzed various mutants of oriC and DnaA in reconstituted systems to reveal the regulatory mechanisms underlying DUE unwinding and DnaB loading. The Right-DnaA subcomplex assisted in the unwinding of oriC, dependent upon an interaction with L-DUE, which is important for efficient loading of DnaB helicases. The AT-cluster region adjacent to the DUE promoted loading of DnaB helicase in the absence of the Right-DnaA subcomplex. Consistently, the ssDNA-binding activity of the Right-DnaA subcomplex sustained timely initiation of growing cells. These results indicate that DUE unwinding and efficient loading of DnaB helicases are sustained by concerted actions of the Left- and Right-DnaA subcomplexes. In addition, loading of DnaB helicases are sustained by multiple mechanisms that ensure robust replication initiation, although the complete mechanisms are required for precise timing of initiation during the cell cycle.     

Cytoplasmic diffusion: molecular motors mix it up

Random motion within the cytoplasm gives rise to molecular diffusion; this motion is essential to many biological processes. However, in addition to thermal Brownian motion, the cytoplasm also undergoes constant agitation caused by the activity of molecular motors and other nonequilibrium cellular processes. Here, we discuss recent work that suggests this activity can give rise to cytoplasmic motion that has the appearance of diffusion but is significantly enhanced in its magnitude and which can play an important biological role, particularly in cytoskeletal assembly.The cytoplasm of eukaryotic cells is a highly dynamic and out-of-equilibrium material that undergoes continual restructuring. This is largely driven by active processes such as polymerization of cytoskeletal filaments and forces generated by molecular motors. Such activity usually results in directed movement within cells; examples include the slow retrograde flow at the leading edge during cell crawling (Fisher et al., 1988; Waterman-Storer and Salmon, 1997; Cai et al., 2006) and the transport of motor-bound vesicles along cytoskeletal filaments (Vale, 2003). Given the intrinsically small, micrometer scales involved, the cytoplasm is also subject to the thermal agitation of Brownian motion (Brown, 1828; Einstein, 1905). Random though these thermal fluctuations may be, they are implicated in force generation by polymerization (Peskin et al., 1993; Mogilner and Oster, 1996), as well as the elastic response of cytoskeletal networks (MacKintosh et al., 1995; Gardel et al., 2004; Storm et al., 2005). Moreover, thermal fluctuations give rise to the diffusive transport of small molecules throughout the cell, without which molecular signaling would be impossible. However, it is becoming increasingly clear that nonthermal forces, such as those resulting from motor protein activity, can also lead to strongly fluctuating intracellular motion; this motion differs significantly from the directed motion commonly associated with motor activity (Caspi et al., 2000; Lau et al., 2003; Bursac et al., 2005). These active fluctuations remain poorly understood, and the relative contributions of thermal fluctuations compared with active fluctuations in living cells are only now being fully explored.

Particle-based probes of intracellular motion

To elucidate the nature of fluctuating motion in cells, many studies have analyzed the “passive” fluctuating motion of micrometer-sized spherical probe particles. If such particles were embedded in a viscous liquid driven by thermal Brownian fluctuations, they would exhibit random, diffusive motion, as shown schematically in Fig. 1 a (blue particle). Quantitatively, this means that the distance the particle has moved, Δx, after some time interval, τ, is described by , where the angled brackets indicate an average over many particles, and the diffusion coefficient, D, is given by the Stokes-Einstein equation:which depends on the viscosity η and particle radius a (Einstein, 1905). This reflects the fundamentally thermal origin of diffusion in an equilibrium liquid, depending as well on the temperature (T) and Boltzmann''s constant (k_B). This diffusive time dependence is shown schematically by the blue line in Fig. 1 a. Such motion is in stark contrast with steady particle motion in one direction with constant velocity v, illustrated by the black curve in Fig. 1 a; because , this motion is described by . In principle, one can track the motion of inert tracer particles in cells to determine whether their motion is diffusive with the appropriate diffusion coefficient. However, inside cells this picture is complicated by the fact that the cytoplasm is generally not a simple viscous liquid but rather a structured viscoelastic material (Luby-Phelps et al., 1987; Fabry et al., 2001). In a viscoelastic material, thermal fluctuations do not lead to ordinary diffusion, but rather “subdiffusive” motion, characterized by a different time-dependence: , where α < 1, as shown schematically by the red curve in Fig. 1 a. Thus, studies showing diffusive, or even “superdiffusive” motion (α > 1), within the viscoelastic cytoplasm suggested this random motion is not thermally induced (Caspi et al., 2000). However, other studies assume that random intracellular motion is thermally induced, and use this assumption to extract the mechanical properties of the cell from tracer particle motion (Tseng et al., 2002; Panorchan et al., 2006). To quantitatively clarify this, several recent studies have analyzed both measurements of the “passive” random motion of tracer particles, as well as direct, active measurements of the viscoelastic properties of the cytoplasm, for example, using a magnetic tweezer to pull on cells (Lau et al., 2003; Bursac et al., 2005). These studies have concluded that the random motion not only has an unexpected time dependence but is also significantly larger in magnitude than would be expected for purely thermal fluctuations, suggesting that biological activity can indeed give rise to random fluctuating motion that dominates over thermal fluctuations.Open in a separate windowFigure 1.Quantifying fluctuating motion. (a) Schematic showing different types of particle motion, characterized by the mean square displacement . The particle can exhibit directed (“ballistic”) motion, as shown by the black particle; here α = 2, as shown by the black curve. Particles can also exhibit random diffusive-like behavior, as shown by the blue particle, with α = 1 (blue line). Particles constrained by a viscoelastic network (red particle) often exhibit subdiffusive behavior, with α < 1 (red curve). (b) Data showing the bending motion of a fluctuating intracellular microtubule. The amplitude of a bend, ∝a_q, randomly fluctuates in time, as shown in the bottom inset. The mean squared amplitude difference as a function of lag time displays diffusive-like behavior (α = 1). (c) A CHO cell fixed and stained to reveal the nucleus (blue) and microtubules (green). Bottom inset shows schematically the variables used in the Fourier analysis of microtubule bending. Top inset shows fluctuations in a GFP-tubulin–transfected Cos7 cell over a time difference of 1.6 s. Microtubules that have fluctuated to a new position are in red and the earlier position is in green.Considerable new insight into the underlying physical origin of these motor-driven fluctuations was obtained from studies of a reconstituted actin network incorporating myosin II motors (Mizuno et al., 2007), oligomerized into processive motor assemblies similar to those found in the cellular cytoskeleton (Cai et al., 2006). The mechanical resistance of the actin network was precisely characterized using optical tweezers to actively pull on particles within the network. The fluctuating motion of the particles was also measured, both with and without motors present. By comparing these two measurements, the contribution of the thermal motion could be distinguished from that of the nonthermal motion, showing clearly that myosin motors can give rise to strong random fluctuating motion within the network. Interestingly, these random myosin-generated forces can also lead to a pronounced stiffening of the actin network, increasing the rigidity by as much as 100-fold (Mizuno et al., 2007; unpublished data).

Microtubule bending dynamics reflects active fluctuations

The random fluctuating motion arising from the activity of cytoskeletal motor proteins can have important biophysical consequences. In an attempt to directly measure the underlying fluctuating forces, a different but complementary approach has recently been developed. It uses endogenous cytoskeletal microtubules as probes. Microtubules are stiff biopolymer filaments that are present in almost every animal cell and are physically linked to other components of the cytoskeleton (Rodriguez et al., 2003; Rosales-Nieves et al., 2006). Thus, as with spherical probe particles, their motion reflects forces and fluctuations of the network (Waterman-Storer and Salmon, 1997; Odde et al., 1999). However, in contrast to spherical probes, microtubules also exhibit a local bending motion, whose amplitude can, like a simple elastic spring, be used to determine the applied force (Fig. 1 b). Microtubules in cells indeed appear highly bent, as can be seen, for example, in the fluorescence image of the microtubule network within an adherent CHO cell, shown in Fig. 1 c. These bends fluctuate dynamically in time, which can be seen by subtracting sequential images, as shown in the top inset of Fig. 1 c.This motion was studied in cells by tracking individual fluorescent microtubules, using both Cos7 and CHO cells (Brangwynne et al., 2007b). The curves defined by the microtubules are analyzed using a Fourier analysis technique developed to characterize the bending fluctuations of isolated biopolymers in thermal equilibrium (Gittes et al., 1993; Brangwynne et al., 2007a). This analysis is based on the fact that each curve can be represented as a sum of simpler sinusoidal curves, each of a different amplitude (a_q) and wavelength (λ). The wavelength is typically written in terms of its wave vector, q = 2π/λ, as sketched in the bottom inset of Fig. 1 c. Using this approach, intracellular motion was analyzed by calculating the mean-squared difference in amplitude using , as a function of lag time τ.This quantity is analogous to that calculated for fluctuating particles, , but here effectively measures the fluctuating motion of the microtubule at different wavelengths. In thermal equilibrium, the amplitude difference will grow with τ up to a maximum value given bywhere k_BT is the thermal energy scale and κ is the microtubule bending rigidity. The ratio of these defines the persistence length:which represents the length scale at which thermal fluctuations completely change the direction of the filament; for microtubules, l_p is on the order of 1 mm (Gittes et al., 1993). This establishes the maximum amplitude of bending fluctuations that can be induced by thermal agitation and allows thermal and motor-induced fluctuations within the cell to be distinguished.For microtubules in cells, the amplitude of the fluctuations was found to grow roughly linearly in time, the behavior expected for simple Brownian diffusion (Fig. 1 b). However, strikingly, the maximum bending amplitude in cells is much larger than that expected for thermally induced bends for l_p ≈ 1 mm. This is most apparent for small wavelength bends, q > 1 μm⁻¹, as shown by the blue triangles in Fig. 2 a, which are significantly above the maximum thermal amplitude shown by the solid line. Thus, although random intracellular motion can exhibit features similar to random Brownian motion, it appears inconsistent with a purely thermal origin.Open in a separate windowFigure 2.Microtubule bending in vivo and in vitro. (a) Thermal microtubules in an in vitro network of F-actin exhibit a roughly q⁻² spectrum of fluctuations (solid line), although wave vectors smaller than q ∼0.4 μm⁻¹ have not reached their maximum fluctuations on this time scale (τ = 2 s; red squares). In the presence of myosin II motors (blue squares), the bending fluctuations are significantly larger than thermal on short wavelengths (high q). Curves are means of 10 filaments. Intracellular microtubule fluctuations show similar behavior (blue triangles), with amplitudes larger than thermal on short wavelengths, as shown by the mean of 23 filaments from a CHO cell (τ = 2 s). (b) Microtubules embedded in an in vitro actin network in thermal equilibrium exhibit small fluctuations (inset, red squares, q ∼0.3 μm⁻¹), which evolve subdiffusively, i.e., , because of the elasticity of the surrounding actin network (red squares, mean of ∼10 filaments). In myosin-driven networks, the fluctuations are significantly larger and steplike (inset, blue squares, q ∼0.3 μm⁻¹). These large nonthermal fluctuations are diffusive in character, i.e., (blue squares, mean of ∼10 filaments). (c) The bending fluctuations of microtubules in vitro are highly localized and relax rapidly as shown in the top right inset (78 ms between each frame, top to bottom). These localized bends can be well fit to the expected shape resulting from transverse point forces (Landau and Lifshitz, 1986; Brangwynne et al., 2008). A localized bend with the fit to the theoretical form (red line) is shown in the top inset. From these fits, a distribution of localized force pulses with a mean magnitude of ∼10 pN is found (main plot).In these experiments in living cells, there are several unknown variables that could play a role. For example, the persistence length of microtubules may vary within the cell, caused by a possible length dependence (Pampaloni et al., 2006) or arising from the effects of microtubule-associated proteins (Felgner et al., 1997). A simplified in vitro cytoskeleton was therefore developed, incorporating purified microtubules in a model actin network (Brangwynne et al., 2008). In the absence of motor proteins or other sources of nonequilibrium activity, microtubules embedded in the actin network are subject to only thermal forces that result in small bending fluctuations; as with the motion of spherical probe particles, the viscoelasticity of the surrounding network leads to subdiffusive behavior of the bending amplitudes, , as shown in Fig. 2 b. These fluctuations display a small maximum amplitude corresponding to l_p ≈ 1 mm, as shown by the red squares in Fig. 2 a. When myosin II motor assemblies are added to the actin network, the embedded microtubules show dramatically different behavior: they bend significantly more, and the bends are highly localized. However, although this behavior is driven by processive motor activity, the microtubule bends fluctuate randomly in time, with localized bends growing and shrinking rapidly, as illustrated by the typical time series shown in the inset of Fig. 2 b; this behavior is similar to that found using spherical probe particles (Mizuno et al., 2007). These microtubule bends appear to result from localized transverse forces (Landau and Lifshitz, 1986), as sketched in the bottom inset of Fig. 2 c. Analogous to a simple spring (force proportional to displacement, F = kΔx), the maximum force was directly determined from the amplitude of these bends, revealing force pulses on the order of 10 pN, which is consistent with that of a few myosin motors acting together (Finer et al., 1994). Interestingly, short wavelength bends can also arise from compressive forces acting within cells (Brangwynne et al., 2006).The dynamic, localized microtubule bends observed in vitro lead to fluctuations in the bending amplitudes that are large and distinctly nonthermal at short wavelengths (large q), as shown by the comparison of thermal (Fig. 2 a, red squares) and motor-driven (blue squares) fluctuations for q ≥ 0.2 μm⁻¹. At longer wavelengths, however, the fluctuations are indistinguishable from those of thermally excited filaments, as expected for microtubules whose lateral motion is restricted by the surrounding elastic environment. Surprisingly, with added myosin motors, the actively driven bends fluctuate in a diffusive-like manner, , as shown in Fig. 2 b (blue squares). This nonthermal diffusive behavior can be understood in terms of steplike or on–off dynamics in the forces applied by individual motor assemblies as they bind and unbind to cytoskeletal filaments (Mizuno et al., 2007; MacKintosh and Levine, 2008). Thus, even processive motor activity has stochastic features that can give rise to diffusive-like motion of cytoplasmic components. Although it is likely that myosin II is not the only motor contributing to this behavior in cells, these experiments help elucidate the underlying biophysical processes.

Implications of active fluctuations for transport and cytoskeletal assembly

These random nonthermal force fluctuations appear to be a ubiquitous feature of living cells. They can, therefore, play an important role in a variety of cellular processes. For example, as a result of large microtubule bending fluctuations, the tips of growing microtubules also undergo large fluctuations in directional orientation, leading to highly curved microtubule shapes that appear to be “frozen-in” by the surrounding elastic cytoskeleton, as shown in the example in Fig. 3 a. These random fluctuations in tip orientation are analogous to random thermal fluctuations and give rise to microtubule bends with a thermal-like dependence on q, as shown in Fig. 3 b. However, the corresponding nonequilibrium persistence length is reduced to ∼30 μm, ∼100-fold less than the thermal persistence length. Thus, the microtubule network is more bent by ∼100-fold in these cells as compared with isolated microtubules in solution. Although the effects of microtubule binding proteins or defects in the tubulin lattice could also contribute, tip fluctuations alone are sufficient to explain these large bends. Driven fluctuations can thus play a significant role in determining cytoskeletal architecture.Open in a separate windowFigure 3.Nonequilibrium fluctuations affect the growth dynamics and final structure of the microtubule network. (a) A microtubule within a GFP-tubulin–transfected Cos7 cell, highlighted in red, can be seen growing toward the bottom right, in the direction indicated by the yellow line. In the second frame, the filament experiences a naturally occurring bending fluctuation caused by internal forces, indicated by the arrow. As a result, the orientation of the microtubule tip changes and the microtubule grows upward, giving rise to a long wavelength bend. Frame times are 0, 36, 46, 53, and 92 s, top to bottom. (b) Inset schematic shows how lateral bending fluctuations of microtubules will cause fluctuations in the microtubule tip during growth, giving rise to curved polymerization trajectories. The Fourier spectrum of a single representative microtubule in thermal equilibrium is shown by the green circles, calculated from the maximum variance of the fluctuations. This microtubule has a persistence length, l_p = 4 mm. The Fourier amplitude of an ensemble of microtubules in CHO cells, , is shown by the blue squares.The enhanced diffusive dynamics that result from motor activity may also affect the rates of biochemical reactions that take place on the cytoskeletal scaffold (Forgacs et al., 2004), as well as those usually thought to be limited by thermal diffusion. Thus, “active” cytoplasmic diffusion could represent a kind of microscopic mixing that enables rapid diffusion of vesicles and small molecules.Thermal fluctuations have been known to play a fundamental role in the behavior of all nonliving matter since Einstein''s seminal work in 1905 (Einstein, 1905), in a paper explaining how thermal forces give rise to the diffusive motion first observed by Brown in 1828 (Brown, 1828). We propose that active intracellular force fluctuations represent the biological analogue, controlling cell behavior while subject to biochemical regulation. Interestingly, this may actually be closer to Brown''s initial concept of a vital microscopic activity, in contrast to the nonliving, thermal motion that bears his name (Brown, 1828). This active motion is clearly a ubiquitous and important phenomenon in living cells; indeed, it may be that active processes contribute to virtually all randomly fluctuating, diffusive-like motion in cells.     

Ratiometric Imaging of the T-Cell Actin Cytoskeleton Reveals the Nature of Receptor-Induced Cytoskeletal Enrichment

The T-cell actin cytoskeleton mediates adaptive immune system responses to peptide antigens by physically directing the motion and clustering of T-cell receptors (TCRs) on the cell surface. When TCR movement is impeded by externally applied physical barriers, the actin network exhibits transient enrichment near the trapped receptors. The coordinated nature of the actin density fluctuations suggests that they are composed of filamentous actin, but it has not been possible to eliminate de novo polymerization at TCR-associated actin polymerizing factors as an alternative cause. Here, we use a dual-probe cytoskeleton labeling strategy to distinguish between stable and polymerizing pools of actin. Our results suggest that TCR-associated actin consists of a relatively high proportion of the stable cytoskeletal fraction and extends away from the cell membrane into the cell. This implies that actin enrichment at mechanically trapped TCRs results from three-dimensional bunching of the existing filamentous actin network.The T-cell actin cytoskeleton is critical for proper antigen recognition by the mammalian adaptive immune system. During T-cell receptor (TCR) triggering by antigen peptides presented on major histocompatibility proteins (pMHCs) on the surfaces of antigen-presenting cells (APCs), the T-cell actin cytoskeleton adopts a pattern of centrosymmetric retrograde flow (1–3). This simultaneously promotes further TCR triggering (4) and rearranges various T-cell membrane proteins and their APC counterparts into an organized cell-cell interface termed the immunological synapse (IS) (5–7). During this process, TCRs form microclusters that move to the center of the IS in an actin-dependent manner (8,9). When engineered physical barriers interrupt the centripetal motion of TCR clusters, actin flow slows near the pinned microclusters, and the cytoskeletal network transiently accumulates and dissipates at the sites (10,11). The amplitude and duration of the induced cytoskeletal fluctuations are much greater than would be expected for a random distribution of independent objects, indicating that the actin in the local environment is coordinated. Whether this coordination arises from a rearrangement in the existing F-actin network or represents de novo polymerization of the cytoskeleton, as predicted by the association of TCRs with actin polymerizing factors (12), remains unclear. Here, we use a dual-probe cytoskeleton labeling approach that has previously been applied to distinguish between stable and dynamic populations of actin by exploiting the different relative affinities of monomeric actin and actin-binding proteins toward each population (13). This strategy reveals that TCR-associated actin is composed primarily of the stable cytoskeletal fraction and that local enrichment results from three-dimensional bunching of the existing filamentous actin network.Primary T cells from mice transgenic for the AND TCR were triggered using synthetic APCs consisting of supported lipid bilayers functionalized with pMHC and the integrin ligand intercellular adhesion molecule 1. Nanopatterned metal grids on the bilayer substrate acted as diffusion barriers that prevented lateral transport of TCR-pMHC complexes (14,15). Transient enrichment of actin at TCR clusters trapped at these barriers was visualized using fluorescent fusions of actin itself (mKate2-β-actin) and the F-actin binding domain of utrophin (EGFP-UtrCH). Such a dual-probe strategy theoretically allows for discrimination between different pools of actin: dynamic populations characterized by high polymerization and/or short filament fragments tend to be relatively better labeled by direct actin fusions whereas stable populations composed of longer filaments can support higher labeling by fluorescent fusions of F-actin binding proteins. This visualization method has been validated in Xenopus oocytes, where it distinguishes actin populations during wound healing (13). It has not been explicitly applied to T cells; however, simultaneous labeling of the Jurkat cell cytoskeleton using EGFP-actin and Alexa 568-phalloidin reveals distinct populations of actin consistent with the results expected from Xenopus (13,16).Our results show that the T-cell periphery is relatively enriched in mKate2-β-actin (Fig. 1 C, box 1), while EGFP-UtrCH dominates toward the center of the IS (Fig. 1 C, box 2). We infer from this probe distribution that the cytoskeleton at the T-cell periphery is composed of short fragments and is a site of active polymerization, whereas at the center of the IS, actin filaments are longer and predominantly stable. This is consistent with previous models of the T-cell actin network (3,16). An effective way to highlight each of these cytoskeletal regions is to consider the relative ratios of the two probes at each location. In this case, a high UtrCH/actin ratio corresponds to stable actin, and a high actin/UtrCH ratio corresponds to dynamic actin (Fig. 1 D). When T cells are treated with cytochalasin D, an inhibitor of actin polymerization, the overall UtrCH/actin ratio of the cell decreases as would be expected from a general decrease in polymerized actin (see Movie S7 and Movie S8 in the Supporting Material). However, it should be noted that photobleaching can also shift the UtrCH/actin ratio over time. We limit quantitative analysis of the ratio to its spatial gradients at a single time point, but such analysis is possible in systems that permit rigorous calibration for probe expression and photobleaching.Open in a separate windowFigure 1Ratiometric imaging of the cytoskeleton in live T cells distinguishes between dynamic and stable actin populations. (A) mKate2-β-actin, (B) EGFP-UtrCH, and (C) merged images of a triggered T cell show different actin pools. The cutouts in panel C correspond to (1) a region high in dynamic actin featuring short, polymerizing filaments and/or actin monomers and (2) a region with a stable actin population featuring longer filaments to which UtrCH can bind. (D) The UtrCH/actin ratio image highlights pools of relatively high UtrCH (red) or actin (blue). (Scale bars: 5 μm.)Actin enrichment at trapped TCR clusters incorporates both mKate2-β-actin (Fig. 2, A and C) and EGFP-UtrCH (Fig. 2, B and C). The relative UtrCH/actin ratio at these sites (Fig. 2 D, box 2) is quite high relative to nearby background areas (Fig. 2 D, box 1), indicating that the actin is derived primarily from the stable actin population.Open in a separate windowFigure 2Receptor-induced cytoskeletal enrichment at sites of pinned TCRs corresponds to a primarily stable actin fraction. (A) mKate2-β-actin, (B) EGFP-UtrCH, and (C) merged images of a triggered T cell interacting with a nanopatterned supported lipid bilayer show actin enrichment corresponding to putative sites of pinned TCRs. (D) The UtrCH/actin ratio is high at sites displaying actin enrichment, indicating a primarily stable actin fraction in (1) these regions compared to (2) nearby background areas. (Scale bars: 5 μm.)The three-dimensional distribution of TCR-associated actin was analyzed in dual-labeled live T cells using a spinning disk confocal microscope. The recordings show actin extending away from the cell membrane in the vicinity of trapped TCRs, while the rest of the actin cytoskeleton remains relatively flat (Fig. 3 and see Fig. S1 in the Supporting Material). These protrusions of actin away from the membrane surface are predominantly composed of stable, filamentous actin, as indicated by their relatively high UtrCH/actin ratio (Fig. 3 B).Open in a separate windowFigure 3Three-dimensional ratiometric imaging shows that actin enrichment extends away from the cell membrane. Single planes from (A) merged mKate2-β-actin and EGFP-UtrCH and (B) UtrCH/actin ratio three-dimensional stacks show actin enrichment at the cell membrane. Cutouts represent Z projections passing through sites of (1) enrichment and (2) nearby background regions. The color distribution in panel B is analogous to that in Figs. 1D and ​and22D, and is omitted for clarity. (Scale bar: 5 μm in the x axis only. Scale box: 1 μm.)Our interpretation of these results is that the filamentous actin network is relatively dense at sites of pinned TCRs. This is the simplest explanation out of several possibilities, one of which is formin-mediated mKate2-β-actin-deficient actin nucleation (17). Filament bunching at pinned TCRs can arise from consistent biophysical properties without assuming heterogeneity between the biochemistry of these receptors and other actin-associated proteins such as those at the cell edge, where locally high probe ratios are absent.Although TCRs are intentionally trapped as part of this experimental strategy, it is likely APCs can naturally impede TCR ligand mobilities under certain circumstances, and this has been shown to impact T-cell signaling (18,19). Actin architecture near cell surface proteins has been extensively studied in focal adhesions of fibroblasts (20), but the lack of stress fibers in T cells makes it unlikely that the two structures are similar. Thus, receptor-induced cytoskeletal enrichment at TCR clusters adds to the catalog of actin behaviors in situ, which is conveniently probed by techniques such as ratiometric dual-probe imaging in live cells. These techniques can be coupled to various spatial analysis algorithms to further extend their utility.     

Focus Issue on Plant Cell Walls: Sequencing around 5-Hydroxyconiferyl Alcohol-Derived Units in Caffeic Acid O-Methyltransferase-Deficient Poplar Lignins

Caffeic acid O-methyltransferase (COMT) is a bifunctional enzyme that methylates the 5- and 3-hydroxyl positions on the aromatic ring of monolignol precursors, with a preference for 5-hydroxyconiferaldehyde, on the way to producing sinapyl alcohol. Lignins in COMT-deficient plants contain benzodioxane substructures due to the incorporation of 5-hydroxyconiferyl alcohol (5-OH-CA), as a monomer, into the lignin polymer. The derivatization followed by reductive cleavage method can be used to detect and determine benzodioxane structures because of their total survival under this degradation method. Moreover, partial sequencing information for 5-OH-CA incorporation into lignin can be derived from detection or isolation and structural analysis of the resulting benzodioxane products. Results from a modified derivatization followed by reductive cleavage analysis of COMT-deficient lignins provide evidence that 5-OH-CA cross couples (at its β-position) with syringyl and guaiacyl units (at their O-4-positions) in the growing lignin polymer and then either coniferyl or sinapyl alcohol, or another 5-hydroxyconiferyl monomer, adds to the resulting 5-hydroxyguaiacyl terminus, producing the benzodioxane. This new terminus may also become etherified by coupling with further monolignols, incorporating the 5-OH-CA integrally into the lignin structure.Lignins are polymeric aromatic constituents of plant cell walls, constituting about 15% to 35% of the dry mass (Freudenberg and Neish, 1968; Adler, 1977). Unlike other natural polymers such as cellulose or proteins, which have labile linkages (glycosides and peptides) between their building units, lignins’ building units are combinatorially linked with strong ether and carbon-carbon bonds (Sarkanen and Ludwig, 1971; Harkin, 1973). It is difficult to completely degrade lignins. Lignins are traditionally considered to be dehydrogenative polymers derived from three monolignols, p-coumaryl alcohol 1h (which is typically minor), coniferyl alcohol 1g, and sinapyl alcohol 1s (Fig. 1; Sarkanen, 1971). They can vary greatly in their composition in terms of their plant and tissue origins (Campbell and Sederoff, 1996). This variability is probably determined and regulated by different activities and substrate specificities of the monolignol biosynthetic enzymes from different sources, and by the carefully controlled supply of monomers to the lignifying zone (Sederoff and Chang, 1991).Open in a separate windowFigure 1.The monolignols 1, and marker compounds 2 to 4 resulting from incorporation of novel monomer 15h into lignins: thioacidolysis monomeric marker 2, dimers 3, and DFRC dimeric markers 4.Recently there has been considerable interest in genetic modification of lignins with the goal of improving the utilization of lignocellulosics in various agricultural and industrial processes (Baucher et al., 2003; Boerjan et al., 2003a, 2003b). Studies on mutant and transgenic plants with altered monolignol biosynthesis have suggested that plants have a high level of metabolic plasticity in the formation of their lignins (Sederoff et al., 1999; Ralph et al., 2004). Lignins in angiosperm plants with depressed caffeic acid O-methyltransferase (COMT) were found to derive from significant amounts of 5-hydroxyconiferyl alcohol (5-OH-CA) monomers 15h (Fig. 1) substituting for the traditional monomer, sinapyl alcohol 1s (Marita et al., 2001; Ralph et al., 2001a, 2001b; Jouanin et al., 2004; Morreel et al., 2004b). NMR analysis of a ligqnin from COMT-deficient poplar (Populus spp.) has revealed that novel benzodioxane structures are formed through β-O-4 coupling of a monolignol with 5-hydroxyguaiacyl units (resulting from coupling of 5-OH-CA), followed by internal trapping of the resultant quinone methide by the phenolic 5-hydroxyl (Ralph et al., 2001a). When the lignin was subjected to thioacidolysis, a novel 5-hydroxyguaiacyl monomer 2 (Fig. 1) was found in addition to the normal guaiacyl and syringyl thioacidolysis monomers (Jouanin et al., 2000). Also, a new compound 3g (Fig. 1) was found in the dimeric products from thioacidolysis followed by Raney nickel desulfurization (Lapierre et al., 2001; Goujon et al., 2003).Further study with the lignin using the derivatization followed by reductive cleavage (DFRC) method also confirmed the existence of benzodioxane structures, with compounds 4 (Fig. 1) being identified following synthesis of the authentic parent compounds 9 (Fig. 2). However, no 5-hydroxyguaiacyl monomer could be detected in the DFRC products. These facts imply that the DFRC method leaves the benzodioxane structures fully intact, suggesting that the method might therefore be useful as an analytical tool for determining benzodioxane structures that are linked by β-O-4 ethers. Using a modified DFRC procedure, we report here on results that provide further evidence for the existence of benzodioxane structures in lignins from COMT-deficient plants, that 5-OH-CA is behaving as a rather ideal monolignol that can be integrated into plant lignins, and demonstrate the usefulness of the DFRC method for determining these benzodioxane structures.Open in a separate windowFigure 2.Synthesis of benzodioxane DFRC products 12 (see later in Fig. 6 for their structures). i, NaH, THF. ii, Pyrrolidine. iii, 1g or 1s, benzene/acetone (4/1, v/v). iv, DIBAL-H, toluene. v, Iodomethane-K₂CO₃, acetone. vi, Ac₂O pyridine.     

Rapid reverse phase-HPLC assay of HMG-CoA reductase activity

Radioisotope-based and mass spectrometry coupled to chromatographic techniques are the conventional methods for monitoring HMG-CoA reductase (HMGR) activity. Irrespective of offering adequate sensitivity, these methods are often cumbersome and time-consuming, requiring the handling of radiolabeled chemicals or elaborate ad-hoc derivatizing procedures. We propose a rapid and versatile reverse phase-HPLC method for assaying HMGR activity capable of monitoring the levels of both substrates (HMG-CoA and NADPH) and products (CoA, mevalonate, and NADP⁺) in a single 20 min run with no pretreatment required. The linear dynamic range was 10–26 pmol for HMG-CoA, 7–27 nmol for NADPH, 0.5–40 pmol for CoA and mevalonate, and 2–27 nmol for NADP⁺, and limit of detection values were 2.67 pmol, 2.77 nmol, 0.27 pmol, and 1.3 nmol, respectively.HMG-CoA reductase (HMGR) is the enzyme that catalyze the four-electron reductive deacylation of HMG-CoA to CoA and mevalonate (Fig. 1) (1). This reaction is the controlling step in the biosynthesis of sterols and isoprenoids (2, 3); hence, a large number of studies on the modulation of HMGR activity are continuously performed in the effort of developing new drugs in the treatment of hypercholesterolemic disorders (1).Open in a separate windowFig. 1.Schematic representation of HMGR enzymatic reaction.HMGR activity is conventionally assayed using elaborate radiochemical assay (4–9), chromatographic techniques coupled with mass spectrometry (10–15), or spectrophotometrically by monitoring the decrease in the absorbance of cofactor NADPH at 340 nm (16).Herein, as an alternative for laboratories with no access to the expensive LC/MS equipment, we propose a rapid and adequately sensitive HPLC-based method capable of monitoring both the levels of all the species involved in the equilibrium in a single analysis and the kinetics of HMGR-catalyzed reactions.     

Structural basis of the super- and hyper-relaxed states of myosin II

Super-relaxation is a state of muscle thick filaments in which ATP turnover by myosin is much slower than that of myosin II in solution. This inhibited state, in equilibrium with a faster (relaxed) state, is ubiquitous and thought to be fundamental to muscle function, acting as a mechanism for switching off energy-consuming myosin motors when they are not being used. The structural basis of super-relaxation is usually taken to be a motif formed by myosin in which the two heads interact with each other and with the proximal tail forming an interacting-heads motif, which switches the heads off. However, recent studies show that even isolated myosin heads can exhibit this slow rate. Here, we review the role of head interactions in creating the super-relaxed state and show how increased numbers of interactions in thick filaments underlie the high levels of super-relaxation found in intact muscle. We suggest how a third, even more inhibited, state of myosin (a hyper-relaxed state) seen in certain species results from additional interactions involving the heads. We speculate on the relationship between animal lifestyle and level of super-relaxation in different species and on the mechanism of formation of the super-relaxed state. We also review how super-relaxed thick filaments are activated and how the super-relaxed state is modulated in healthy and diseased muscles.

The super-relaxed state of myosinAnimal life is characterized by the constant need for food, which provides the raw materials for making ATP used in the body’s energy-requiring processes. A substantial amount of energy is expended by skeletal muscle, which in humans amounts to 30–40% of body mass (Janssen et al., 2000). During contraction, ATP is rapidly consumed by the molecular motor, myosin II, as it pulls on actin filaments to produce force and movement. But ATP is also used at a significant basal rate even in the resting state. Producing ATP is costly for the cell, so there is a substantial evolutionary advantage to minimizing its waste. Animals have adapted to this requirement by evolving a mechanism that reduces the consumption of ATP to minimal levels in relaxed (RX) muscle. This mechanism is known as super-relaxation or the super-relaxed state (SRX).SRX is a biochemical state in which ATP turnover by a portion of the myosin heads in the RX state of muscle thick filaments is much slower (∼10 times) than that of myosin II molecules in solution (Stewart et al., 2010; Cooke, 2011; Nag and Trivedi, 2021). This inhibited state, in equilibrium with a faster (RX) state, was suggested by early studies showing that the metabolic rate of live, resting muscle was much less than that expected from the ATP turnover rate of purified myosin (Ferenczi et al., 1978) and similarly that ATPase activity of RX myofibrils was also lower than expected from isolated myosin (Myburgh et al., 1995). A slow rate of ATP turnover was later directly demonstrated in studies of skinned muscle fibers (Stewart et al., 2010). The single ATP turnover rate is typically measured as the rate of binding of fluorescent ATP (mantATP) to, or release from, myosin heads in solution or in skinned fiber bundles (Hooijman et al., 2011; McNamara et al., 2015); conceptually, because P_i release is rate limiting, the apparent rates of ATP binding and ADP release effectively reveal the overall ATP turnover rate. Experimental curves are generally interpreted in terms of two exponentials, one with a slow rate (SRX) and the other with a faster but still slow (RX) rate of ATP use (Cooke, 2011), although additional exponentials can sometimes be resolved (Hooijman et al., 2011; Naber et al., 2011).The SRX state is ubiquitous and thought to be fundamental to muscle function, acting as a mechanism for parking and switching off—like a car—energy-consuming myosin motors when they are not being used. It has been detected in all muscles where it has been studied: vertebrate (including human) and invertebrate, fast and slow skeletal, and cardiac muscle (Stewart et al., 2010; Cooke, 2011; Hooijman et al., 2011; Naber et al., 2011; McNamara et al., 2015; Phung et al., 2020). A highly inhibited rate of ATP turnover also characterizes the switched-off state of vertebrate smooth muscle and nonmuscle myosin II molecules (Cross et al., 1988). SRX plays a critical role in muscle energetics, conserving ATP in resting as well as contracting muscles and providing a reserve of myosin heads for enhanced contractility in cardiac muscles (Cooke, 2011; Hooijman et al., 2011; Brunello et al., 2020; Ma et al., 2021b).What are the key features of the SRX state?The SRX state has two main parameters: (1) the rate of ATP use (i.e., ATP turnover rate, often expressed as its inverse, the ATP turnover time) and (2) the fraction of molecules exhibiting this rate. Both parameters are essential to understanding the significance of SRX in the energy balance of muscle: myosin heads with a very slow rate would be of little consequence if not present in substantial numbers. The rate of ATP use has been attributed to the specific conformation of the myosin head, which can impede, or not, the release of the products of ATP hydrolysis (ADP and P_i; Anderson et al., 2018), and the fraction of heads with this rate in muscle appears to be related to the intra- and intermolecular interactions that the heads undergo in thick filaments, as discussed below.ATP turnover by myosin heads can vary by six orders of magnitude, depending on the muscle and its state of activity (Cooke, 2011; Naber et al., 2011). These fall into four main groups, which, on a logarithmic scale, differ internally by small amounts, but by an order of magnitude or more between groups (Fig. 1 A). The fastest use of ATP is in contracting muscle (rate ≈ 10 s⁻¹), where actin activates the release of hydrolysis products in a process coupled to the power stroke of the myosin head. The advantages of force and movement are paid for by a rapid consumption of ATP. RX heads use ATP at least 100 times more slowly: some at the RX rate, similar to that of purified myosin in solution (<0.1 s⁻¹), and others at an SRX rate, 10 times slower (Fig. 1, A and B). As might be predicted from the two parameters of the SRX, nature has evolved different ways of saving energy in this state: decreasing the ATP turnover rate, increasing the fraction with the low rate, or a combination of the two. These strategies appear adapted to the different systems in which they are found. The SRX rate typically occurs in 50% or more of the heads in a thick filament, the balance being RX heads (Cooke, 2011). In certain muscles, there is an even slower rate (10 times slower than SRX), which we refer to as “hyper-relaxed” (HRX), present together with SRX and RX heads (Naber et al., 2011).Open in a separate windowFigure 1.ATP turnover rates of myosin and proposed relation to head organization. (A) Logarithmic plot of rates, varying from very slow (hyper-relaxed), to slow (super-relaxed), to fast (disordered-relaxed), to very fast (actin-activated). (B) Expanded plot of relaxed rates and proposed relation to head configuration and interactions. The cartoons show the IHM as found in single molecules (upper) and thick filaments (lower); in filaments, only molecules in a single crown are shown. The colored ellipses are the interacting motor domains. The smaller, gray domains are the ELCs and RLCs, and the gray lines are the proximal region of the myosin tail (S2). Molecules with a sufficient length of S2 (25 heptads [25 hep]; ∼250 Å) form an IHM, while single heads and two-headed molecules with only two heptads of tail show no interactions. Molecules in filaments have SRX and DRX turnover rates similar to those of isolated molecules (25-heptad IHM; cf. vertebrate IHM) or more inhibited (HRX) rates. Heads turn over ATP at fast (F; noninteracting), slow (S; interacting), or very slow (VS; more interactions) rates. Slow (attached) FHs of the IHM can detach and sway out (red, double-headed arrows) and, while detached, can equilibrate between the slow and fast conformations, similar to S1 (blue, reversible arrows). The “tail” in tarantula and the uncolored IHM in scallop provide additional, intermolecular interactions leading to the very slow rate; this rate is also seen in 10S myosin through intramolecular interactions with segment 2 of its own tail. The different lengths of the bars in A correspond to the lengths in B, which are drawn to include the range of rates for HRX, SRX, and DRX; the key point is the close clustering of rates within these groups and the large gaps between them.What is the structural basis of the SRX state?The early finding that myosin filaments in muscle had a slower ATP turnover rate than myosin II molecules in solution led to the idea that interactions of heads possible in the polymer (e.g., with the thick filament backbone), but not in the monomer, inhibited their activity (Myburgh et al., 1995; Stewart et al., 2010; Cooke, 2011). The first study to unequivocally show head organization in a thick filament indeed revealed such interactions—of heads with the backbone, with each other within a molecule, and with each other between molecules (Woodhead et al., 2005). Of particular note was a folding back of the two heads onto the proximal part of the tail (subfragment 2 [S2]), with which they interacted, and intramolecular interaction between the heads through their motor domains. This structure became known as the “interacting-heads motif” (IHM; Fig. 1 B, dotted box; Fig. 2 B; Woodhead et al., 2005; Alamo et al., 2008). Such interaction between heads had first been seen in isolated myosin molecules (Wendt et al., 2001; Burgess et al., 2007) and was thought to inhibit their activity. The two heads were called “blocked” and “free” (BH and FH, respectively), referring to their actin-binding capability (Wendt et al., 2001). In this model, actin binding by the BH would be blocked by interaction of its actin-binding site with the FH. While the FH actin-binding site was exposed, movement of its converter domain, needed for ATP product release, would be inhibited by binding to the BH (Wendt et al., 2001). The overall result would be inhibition of activity of both heads, but by different mechanisms. Interaction of the folded-back heads with S2 would further inhibit the motions required for ATPase activity (Woodhead et al., 2005).Open in a separate windowFigure 2.Structural basis of SRX. (A) Proposed conformations of myosin heads (S1), in equilibrium with each other, underlying SRX (left, bent) and DRX (right, straight) ATP turnover rates (Anderson et al., 2018). MD, motor domain. (B) IHM of cardiac HMM showing BH and FH (based on Protein Data Bank accession no. 5TBY). Ellipses show regions of interaction between BH and FH motor domains (black ellipse), FH and S2 (black circle), and BH and S2 (white ellipse; interaction occurs on rear side of BH). (C) Thick filament (tarantula; EM Data Bank accession no. 1950) showing IHMs lying along four coaxial helices (three on front marked with arrows) creating intermolecular interactions (ellipses) between FH of one IHM and regulatory domain of IHM above. M-line would be at the top of the image. The reconstruction shows the average positions of the heads in the filament; however, the FHs are thought to be dynamic, leaving and returning to the IHM (Fig. 1 B; see text). Models in this figure were created with UCSF Chimera (Pettersen et al, 2004).The IHMs in RX thick filaments are organized in helical arrays, with intermolecular interactions of IHMs with each other along the helices as well as with the filament backbone (Woodhead et al., 2005; Zoghbi et al., 2008; Zhao et al., 2009; Pinto et al., 2012; Woodhead et al., 2013; Sulbarán et al., 2015). Helical ordering requires the closed conformation of the myosin heads (Xu et al., 2003; Zoghbi et al., 2004; Xu et al., 2009) and is thought to be a signature of the IHM. It can be disrupted in multiple ways (increased salt level, phosphorylation of the myosin regulatory light chains [RLCs], substitution of GTP or ADP for ATP, lowering of temperature), and this is accompanied by reduction of the SRX state in each case (Stewart et al., 2010; Cooke, 2011). Conversely, conditions that enhance the IHM, such as treatment with the myosin inhibitor blebbistatin (Zhao et al., 2008; Xu et al., 2009; Fusi et al., 2015), enhance the SRX (Wilson et al., 2014; Fusi et al., 2015). These correlations led to the view that IHMs in the ADP.Pi prepowerstroke state (Xu et al., 1999), organized in helices and bound to the core of the thick filament, may be the structural basis of the SRX state (Stewart et al., 2010; Cooke, 2011; Wilson et al., 2014; Fusi et al., 2015; Alamo et al., 2016).Despite this suggestive evidence, however, recent single ATP turnover measurements of myosin constructs have shown that slow turnover of ATP occurs not only in thick filaments but also, to a small extent (∼10–20% of molecules), in isolated myosin heads (S1; Anderson et al., 2018; Rohde et al., 2018; Gollapudi et al., 2021b; the earlier, steady-state solution ATPase observations were not capable of revealing these low rates). Thus, neither the filamentous form of myosin nor the IHM is required for the SRX rate of ATP turnover, as originally thought. It has been suggested, instead, that myosin heads in solution exist in an equilibrium between a strongly bent (“closed”) conformation (in which the lever arm is tilted more toward the prestroke direction similar to that in the IHM structure; 10–20% of molecules) and a more extended (“open”) structure (Anderson et al., 2018) and that inhibition of phosphate release and thus ATP turnover by the closed structure could account for the observation of a small level of SRX in S1 (Figs. 1 B and 2 A). If this is the case, what is the role of the IHM in SRX?Different degrees of SRX correlate with different levels of head organizationMyosin heads can exist in several structural forms: single heads, two-headed myosin molecules or constructs, and the polymeric myosin filaments found in muscle. Studies show that SRX increases (greater inhibition or greater number of inhibited molecules) as myosin heads are incorporated into more complex structures. As mentioned, ∼10–20% of isolated myosin heads in solution have a slow rate (SRX) and are thought to be in equilibrium with the balance of heads having a turnover rate similar to the conventionally measured ATPase (RX heads; Anderson et al., 2018; Rohde et al., 2018; Gollapudi et al., 2021b). When myosin heads are present in two-headed constructs containing 25 heptads of tail (∼250 Å; Figs. 1 B and 2 B) or full-length S2, similar SRX and RX rates are observed, but the fraction of SRX heads increases to ∼25–30% (at physiological ionic strength; Anderson et al., 2018; Rohde et al., 2018; Gollapudi et al., 2021b), suggesting that the presence of two heads and/or the tail stabilizes the SRX head structure. Importantly, if the tail is short (only two heptads; Fig. 1 B), the rates and fractions are similar to those of S1 (Anderson et al., 2018), demonstrating the importance of the tail in creating the SRX. Modeling and experiments show that 25 heptads (or more) of tail is long enough for molecules to form an IHM with folded-back, interacting FHs and BHs (Figs. 1 B and 2 B; Anderson et al., 2018). Head–head and head–tail interactions would stabilize the heads in their inhibited conformations (Fig. 2 B), consistent with the increase in SRX fraction. The two-heptad-long tail is too short for these interactions (Fig. 1 B), and the molecule does not show evidence of head interactions (Nag et al., 2017): the heads behave like isolated S1 in solution, with only ∼10% having the SRX rate (Fig. 1 B; Anderson et al., 2018). We conclude that the stabilizing interactions in the IHM involving the tail and the two heads increase the number of heads in the SRX conformation.Studies of whole muscle fibers (vertebrate fast skeletal) show that when myosin molecules are in native thick filaments in their helically ordered IHM conformations (Zoghbi et al., 2008; AL-Khayat et al., 2013), the RX and SRX rates are again similar to those of S1 (Fig. 1 B), but the fraction of SRX now reaches 60–75% (Stewart et al., 2010; Cooke, 2011). A similar, greatly increased fraction of SRX heads is found in slow skeletal and cardiac muscle fibers and myofibrils (Stewart et al., 2010; Hooijman et al., 2011; McNamara et al., 2017; Nelson et al., 2020; Gollapudi et al., 2021a). In filaments, IHMs undergo intermolecular interactions that cannot occur with single molecules (Zoghbi et al., 2008; AL-Khayat et al., 2013). These include interactions between heads of different IHMs along the myosin helices (Fig. 2 C), interaction of heads with tails in the filament backbone, and interaction with myosin-binding protein C (MyBP-C) and titin, lying on the filament surface (Nag et al., 2017). We suggest that these interactions further stabilize the FHs and BHs of the IHMs, without significantly affecting their conformation, leading to the high percentage of SRX in filaments (Anderson et al., 2018).These observations are consistent with the concept suggested earlier that the SRX rate is a consequence of a specific conformation of myosin heads, which inhibits ATP turnover, and with stabilization of this conformation by intra- and intermolecular interactions of the heads (Anderson et al., 2018). When heads are attached to each other in myosin constructs with a sufficient length of tail, forming the IHM, the inhibited conformation is stabilized by intramolecular interactions of the two heads with each other and with the myosin tail, approximately doubling (25–30%) the SRX fraction found in single heads (10–20%). When IHMs are incorporated into thick filaments, the inhibited heads are further stabilized by intermolecular interactions, with a further doubling of the SRX fraction to 60–75%. Thus, while the IHM is not a requirement for the SRX state, in real muscle, it strongly enhances it.Are the RX and SRX rates associated with specific heads in the IHM? It is known experimentally that the BH is more stably associated with the IHM than the FH. EM images of smooth muscle heavy meromyosin (HMM; a soluble, proteolytic fragment of myosin containing the two heads and the proximal third of the tail) show molecules where the BH is attached to S2 but the FH has dissociated (Burgess et al., 2007). This and studies of tarantula thick filaments (Brito et al., 2011; Sulbarán et al., 2013) suggest that the FH can dissociate from the IHM and become mobile (“swaying”; Fig. 1 B, red double-headed arrows), with a duty ratio that defines the fraction of time spent in the dissociated state (Alamo et al., 2017). When unconstrained in this way, the FH presumably acts essentially like S1 in solution, equilibrating between its minor SRX (bent) and its predominant RX (straight) rate of ATP turnover (Fig. 1 B, reversible blue arrows). Thus, the more stable BH would account for most of the SRX rate and the dissociated FH for the faster, RX rate. When the FH is docked in the IHM, stabilized in its inhibited form by interactions with the BH and S2 in the IHM, it may slow to approximately the BH SRX rate (Fig. 1 B; Alamo et al., 2016). The BH and docked FH may have similar enough rates that they are not resolved by the two-exponential analysis; there is in fact some evidence that additional exponentials give an improved fit (Hooijman et al., 2011), but this has not been explored in detail. The idea that some heads can move freely for a portion of the time, leading to their disordering in thick filaments, and the correlation of disorder with the RX rate, has led to the concept of the “disordered relaxed state” (DRX; Wilson et al., 2014). Thus, heads in thick filaments are typically referred to as SRX or DRX.What is the structural basis of the HRX state?Studies of invertebrate (tarantula) striated muscle show that the SRX state in some species can be enhanced by a further (10 times) slowing of the ATP turnover rate compared with the SRX (Fig. 1) and that this occurs in up to 50% of the heads (Naber et al., 2011). We refer to this as the HRX state. The HRX state has not been observed in vertebrate striated muscle thick filaments or myosin molecules. Is there a structural explanation for the greater inhibition of ATP turnover in tarantula muscle? The cryo-EM reconstruction of tarantula filaments shows that the IHMs in this species are arranged in four coaxial helices, with crowns of four IHMs spaced 145 Å apart axially (Woodhead et al., 2005). Each IHM shows the conventional structure of two heads interacting with each other through their motor domains and with the proximal portion of the myosin tail. Importantly, the reconstruction also suggests an additional key interaction that is apparently absent in vertebrate thick filaments (see below). S2, after leaving its IHM, travels along the filament toward the bare zone, passing near the next IHM along the filament axis, through a groove formed by the BH SRC homology 3 (SH3) and converter domains (Fig. 3 A; and Fig. 4, A and B; Woodhead et al., 2005). This location suggests that S2 could sterically interfere with movement of the BH converter region of this IHM that is required for product release. It could thus act as an additional constraint on the already inhibited conformation of the BH that further inhibits ATP turnover (each BH would thus interact with two S2s, its own and that from the axially adjacent IHM). In this case, we suggest that intermolecular interaction does not simply stabilize the inhibited head conformation but creates a new and additional physical barrier to motions of the BH required for ATPase activity, leading to the highly inhibited HRX turnover rate. The reconstruction shows that only the BHs are affected in this way, which could directly explain the finding that 50% of heads are in the HRX state (Naber et al., 2011).Open in a separate windowFigure 3.Comparison of intermolecular interactions in 3-D reconstructions of different thick filaments. Reconstructions are fitted with IHM atomic models (Protein Data Bank accession nos. 3JBH or 5TBY) in each case. All thick filaments (except insect flight muscle; Hu et al., 2016) show interactions between IHMs along the helical tracks (arrows), involving the FH motor domain at one level and the BH lever arm at the next. These interactions are the same at every level in tarantula and scallop (true helical structures) but different at different levels in vertebrates, which are quasi-helical. Additional interactions vary between species. (A) Tarantula shows interaction of S2 from one crown of heads with SH3 and converter domains of the next level (circles; see Fig. 4, A and B). IHM (Protein Data Bank accession no. 3JBH) was fitted to four levels of heads in cryo-EM reconstruction (EMD accession no. 1950; gray). (B) Similar fitting of IHM (Protein Data Bank accession no. 5TBY) to vertebrate (human) cardiac negative stain reconstruction of C-zone (EMD accession no. 2240) shows well-ordered IHMs at two of every three levels of heads (strong map density for pink and cyan IHMs), while the third level (yellow) is poorly ordered (weak map density, suggesting substantial IHM mobility; Zoghbi et al., 2008; AL-Khayat et al., 2013). S2 cannot be fitted unambiguously due to low resolution of map and lack of internal detail with negative stain; within these limitations, there is no obvious S2–head interaction between crowns (circles; see Video 1; cf. Video 2 for 3-D view). (C) In scallop cryo-EM reconstruction, S2 is not resolved, but the tight azimuthal crowding of IHMs around the circumference at each crown suggests potential intermolecular interaction between the SH3 domain of a BH and the motor domain of the neighboring FH (circles). Filaments are oriented with M-line at top; their different symmetries (fourfold, threefold, and sevenfold rotational symmetry, respectively) cause the varying views of the IHMs in the different filaments. All reconstructions are at the same scale. The human has a smaller diameter due to radial shrinkage occurring during negative staining and to the smaller number of molecules (n = 3) at each level. The scallop and tarantula cryo-reconstructions also have different diameters: scallop has seven molecules at each level forming a shell above the filament backbone (Woodhead et al., 2013), while tarantula has four molecules at each level, closer to the backbone (Woodhead et al., 2005), leading to a smaller diameter. Models in this figure were created with UCSF Chimera (Pettersen et al, 2004).Open in a separate windowFigure 4.Comparison of tail interactions with the SH3 and converter domains in atomic models of tarantula thick filaments and 10S smooth muscle myosin. (A and B) Tarantula thick filament (Protein Data Bank accession no. 3JBH). (C and D) 10S myosin (Protein Data Bank accession no. 6XE9). (A and C) Front views at same scale. (B and D) End views (same scale) obtained by rotating A and C 90° around the x axis so that pink IHM in A is closest to viewer (i.e., looking toward M-line). S2 in tarantula filament and segment 2 (Seg2) in 10S myosin both pass through a groove formed by the SH3 and converter (Cnv) domains of the BH motor domain (MD). The looser fit to the groove in tarantula may be due to the lower resolution of the reconstruction used to obtain the atomic model (20 Å resolution; cf. 4.3 Å for 10S myosin). Models in this figure were created with UCSF Chimera (Pettersen et al, 2004).Several observations support this structural interpretation of the HRX state. Vertebrate thick filaments have a symmetry and organization of myosin heads that is different from those in tarantula. Although no cryo-EM reconstruction of vertebrate filaments is available, two negative stain reconstructions of the C-zone (the middle section of each filament half, containing MyBP-C) clearly show the well-known quasi-helical arrangement of myosin heads characteristic of vertebrates (Zoghbi et al., 2008; AL-Khayat et al., 2013). Fitting of the IHM into the density map reveals well-ordered IHMs at two of every three levels of heads, while the third level is relatively disordered (Zoghbi et al., 2008; AL-Khayat et al., 2013). S2 cannot be fitted with certainty due to the low resolution of the map and ambiguities of negative stain; however, there is no obvious interaction of S2 from one crown with heads in the next (Fig. 3 B; Video 1; cf. Video 2). Correspondingly, vertebrate filaments do not exhibit an HRX state. There is also no HRX state in isolated S1- or S2-headed myosin constructs forming the IHM (Anderson et al., 2018; Rohde et al., 2018). These observations are consistent with the conclusion that in tarantula, intermolecular interaction with S2 from the neighboring crown is responsible for the HRX state. Note that the “ultra-relaxed” state induced in myosin by the inhibitor blebbistatin (Gollapudi et al., 2021a) has a very different origin from the HRX—the former due to internal stabilization of switch 2 in the myosin head in the closed state, inhibiting phosphate release (Zhao et al., 2008), the latter to the external structural constraints on head movements that we have described here.Video 1.Human cardiac thick filament reconstruction (EMD accession no. 2240) fitted with human cardiac atomic model (Protein Data Bank accession no. 5TBY) to show apparent absence of interaction between S2 from one level of heads and the SH3 and converter domains (red) of the next level. M-line at top. See also Fig. 3 B. Compare with Video 2.Video 2.Tarantula thick filament reconstruction (EMD accession no. 1950) fitted with tarantula IHM atomic model (Protein Data Bank accession no. 3JBH) to show 3-D view of interaction between S2 from one level of heads and the SH3 and converter domains (red) of the next level. M-line at top. See also Fig. 3 A and Fig. 4, A and B.Additional evidence comes from smooth muscle and nonmuscle myosin molecules, which can exist in a switched-off state in which the ATP turnover rate is similar to that in tarantula—also an HRX state (Cross et al., 1988). Strikingly, these myosins have a folded conformation, in which the two heads form an IHM, and the tail folds into three segments, the middle segment wrapping around the BH (Fig. 1 B; Suzuki et al., 1982; Trybus et al., 1982; Craig et al., 1983; Burgess et al., 2007; Yang et al., 2019). Cryo-EM analysis of this conformation shows intimate contact of the tail with the BH motor as it runs through the groove formed by the BH SH3 and converter domains (Fig. 4, C and D; Scarff et al., 2020; Yang et al., 2020). The regions of tail contact with the BH in these single molecules are very similar to those in the tarantula filament and would sterically impede BH converter movement, contributing to the HRX turnover rate of this inhibited form of the myosin molecule (Fig. 4). Although the inhibitory regions of the tail (S2 in tarantula, the middle segment of the tail in smooth muscle and nonmuscle myosin) and the nature of the interaction (intermolecular in one, intramolecular in the other) are both different, the likely inhibitory effects on BH converter movement and ATPase activity appear to be similar, supporting the importance of this interaction in generating the HRX state. HMM from these myosins lacks the distal two-thirds of the tail and therefore the interaction of the middle segment with the BH. Correspondingly, HMM exhibits an SRX but not an HRX rate (Cross et al., 1988). Together these observations provide strong support for the notion that interaction of the myosin tail with the SH3/converter region of the BH is the structural basis of hyper-relaxation.Another invertebrate muscle in which a putative HRX state has been observed is the scallop striated adductor. When thick filaments lose their ATP, helical ordering of the heads is lost (we would suggest by straightening of the heads in the apo state, so that the IHM can no longer form). EM shows that this process takes up to 30 min in scallop (Vibert and Craig, 1985), consistent with an HRX turnover rate. Scallop thick filaments differ from tarantula in having sevenfold rather than fourfold rotational symmetry (Vibert and Craig, 1983). Cryo-EM shows that this tightly packs the IHMs around the circumference of each crown, creating intermolecular interactions within crowns involving the SH3 domain of the BH and the motor domain of its neighbor FH (Figs. 1 B and 3 C; Woodhead et al., 2013; the heads in tarantula crowns are more widely spaced and do not show these interactions). While the reconstruction does not provide detail on possible tail interactions between crowns (as in tarantula), these additional intermolecular interactions may constrain the heads within a crown, impeding structural changes of the BH and FH and accounting for the hyper-slow release of products from scallop thick filaments through a mechanism quite different from that in tarantula.How is the IHM formed?Formation of the IHM depends on several key properties of myosin II: flexing of the heads and the head–tail junction and interaction of the motor domains with each other and with the proximal region of S2. BH–S2 interaction (where S2 runs over the bent BH) appears to be the primary binding interaction of the IHM (Alamo et al., 2016), as it is observed in smooth muscle HMM even when the FH is not bound to the BH (Burgess et al., 2007); FH–S2 interaction without interaction of the BH and BH–FH interaction without S2 are not observed. Interaction of S2 with the BH can only occur when the BH folds back onto the tail, and then only when the BH is in the strongly bent (Rosenfeld et al., 1994; Alamo et al., 2008, 2016; Scarff et al., 2020; Yang et al., 2020), nucleotide-trapping state, putatively with the SRX rate (Anderson et al., 2018). Thus, we picture the myosin molecule as having flexible heads (in a bent–straight equilibrium biased 90% toward the straight [DRX] conformation), which are flexibly attached to the tail, with the following sequence for formation of the IHM. If a transiently bent head comes in contact with S2, it binds and is stabilized in its bent (SRX) state, thereby becoming a BH (i.e., a precursor IHM; Alamo et al., 2016). The other head, flexing around its head–tail junction and in a similar bent–straight equilibrium, can now be caught (becoming an FH) when its converter region contacts the captured BH motor domain, stabilizing the FH bent conformation (Liu et al., 2003; Alamo et al., 2016; Scarff et al., 2020; Yang et al., 2020). This interaction is strengthened by contact of loops on the FH with S2 (Alamo et al., 2008; Scarff et al., 2020; Yang et al., 2020).Structural observations show that both heads of the IHM are strongly bent (Wendt et al., 2001; Woodhead et al., 2005; Scarff et al., 2020; Yang et al., 2020), while the bent structure in isolated heads may be uncommon. The sequence proposed above suggests how the IHM can be formed despite these constraints. If the bent structure occurs only rarely (e.g., 10% of the time), the likelihood of two bent heads coming together will be low: this likelihood is greatly increased by initial stabilization of one bent head (the BH) binding to S2 (forming the precursor IHM) and then binding of the second head (the FH) to the already bent BH of the precursor. This kinetic argument would explain why head–head interaction without involvement of the tail has not been observed, consistent with mutational studies (Adhikari et al., 2019). The final IHM is a tripartite structure with multiple weak interactions (BH–FH, BH–S2, and FH–S2) that inhibit ATPase activity as well as actin-binding capability (Scarff et al., 2020; Yang et al., 2020). These interactions are easily broken and in equilibrium with the noninteracting form. The result in isolated sarcomeric myosin soluble fragments at physiological ionic strength is an overall 25% SRX and 75% DRX rate (Anderson et al., 2018; Rohde et al., 2018; Gollapudi et al., 2021b). In filaments, individual IHMs (with the inhibited, bent structure of the individual heads) are stabilized by intermolecular interactions occurring in the polymer (as described earlier), increasing the fraction of SRX heads above 50%.Is there a relationship of the SRX state with animal lifestyle?We have suggested that nature uses two ways to enhance the SRX state: increase in fraction of inhibited heads and increase in degree of inhibition. The particular mechanism may be adapted to the lifestyle of the animal (Naber et al., 2011). Tarantulas spend long periods of time immobile and can survive many months without food. Minimizing ATP use during this time by hyper-relaxation of their BHs and SRX of most of their FHs would be an evolutionary advantage (Naber et al., 2011). They would nevertheless be ready for a rapid switch to the active state through the small numbers of swaying FHs that sense thin filament activation when muscle is stimulated (Brito et al., 2011). Scallops can swim quickly for short bursts by jet propulsion, using their striated adductor muscles to rapidly close their shell (e.g., to escape predators), but they remain stationary during extended periods of filter feeding (Speiser and Wilkens, 2016). In this low-activity state, the shell is held partially closed through contraction of the tonic smooth adductor muscle, which enters a catch state, maintaining force with little energy expenditure (Chantler, 2016). The striated adductor is 10 times more massive than its smooth counterpart (Naidu, 1987) and a potential metabolic drain during the long periods when the muscle is not in use: minimizing ATP use during these nonswimming periods through hyper-relaxation of their heads would be advantageous. The folded form of vertebrate smooth muscle and nonmuscle myosin is thought to serve as a storage molecule, which can be activated to form functional filaments as required through phosphorylation of their RLCs (Cross et al., 1988). The complete switching off of ATPase activity through hyper-relaxation when the molecule is not in use would again provide an evolutionary advantage.Vertebrate striated muscle thick filaments exhibit SRX but not hyper-relaxation. Interaction of IHMs with each other or with the filament backbone, MyBP-C, or titin enhances the fraction of heads in the SRX state but does not significantly increase the level of inhibition of the heads. SRX appears to be developed most strongly in the C-zone (probably in the two crowns of heads showing IHMs in each three-crown repeat), implicating MyBP-C in this inhibition (McNamara et al., 2016, 2019; Nelson et al., 2020). Myosin heads in the D-zone and in the non-IHM crown in the C-zone are less ordered (Zoghbi et al., 2008; AL-Khayat et al., 2013; Brunello et al., 2020) and may be the main source of DRX heads detected by single turnover experiments (Nelson et al., 2020). Vertebrates may make a trade-off of greater (though still low) ATP use in the RX state (SRX with some DRX heads) for the ability to switch on instantly as needed and for fine-tuning of the contractile response through interactions with other proteins such as MyBP-C.How are thick filaments activated from the SRX state?We have painted a picture in which most myosin heads in the thick filaments of resting muscle are tacked down on the filament surface in helices of IHMs in an SRX (or HRX) state. However, muscles must be ready for immediate activation from this energy-saving state. The other part of the picture therefore includes a small fraction of heads dissociated from the IHM at any particular moment. These relatively few, mobile, constitutively on (“sentinel”) heads are presumed to constantly explore the interfilament space, able to instantly sense thin filament activation and bind to actin when myosin-binding sites are exposed (by Ca²⁺-induced tropomyosin movement; Gordon et al., 2000), leading to initial tension development (Linari et al., 2015; Irving, 2017). We suggest that these are the transiently dissociated (swaying) FHs of the IHMs. In vertebrates, the sentinel heads could also include the less well-ordered heads at every third crown of the C-zone (Zoghbi et al., 2008; AL-Khayat et al., 2013) or the disordered heads in the D-zone (Brunello et al., 2020; Nelson et al., 2020). This essential role for a small number of mobile heads as the trigger for thick filament activity, when thin filaments are switched on, is paid for by the increase in ATP consumption over that used by SRX heads. What happens next depends on the type of muscle.In tarantula, the small amount of thick filament activity in an initial twitch appears to be enhanced in additional twitches by Ca²⁺-induced activation of myosin light chain kinase, which phosphorylates the FH RLCs, leading to increased release of the FHs from the IHMs, reduction in the HRX fraction and increase in the DRX fraction (Naber et al., 2011), and increased force production (Padrón et al., 2020). Prolonged high Ca²⁺ (e.g., in a tetanus) subsequently phosphorylates the BHs, resulting in their release and further enhancement of force (Padrón et al., 2020).Vertebrates have a finely tuned, graded response to activation. X-ray studies show that low-force isotonic contractions use only a small fraction of available heads (Linari et al., 2015). The majority remain helically ordered in their IHM configurations, continuing to save energy—even during contraction—representing a highly efficient use of ATP. As force increases, stress on the thick filaments rises, resulting in release of more heads, which produce greater force, in a positive feedback loop (Linari et al., 2015; Irving, 2017). In this mechanosensing model for thick filament activation, the filament is stretched by ∼1%, which may be sufficient to weaken intermolecular contacts between heads along the helical tracks that help to maintain the IHM conformations (Irving, 2017). Destabilization of the IHMs by stress could thus release the additional heads needed for strong force production. When contraction is switched off by removal of Ca²⁺ from the cytosol, IHMs rapidly go back to their helically ordered arrangement (Linari et al., 2015; cf. Ma et al., 2019), quickly returning the muscle to its energy-saving SRX state.In scallop, the thick filaments are directly activated by Ca²⁺ binding to the essential light chains (ELCs) on the myosin heads rather than through Ca²⁺ activation of the thin filaments (Szent-Györgyi et al., 1999; Chantler, 2016). Ca²⁺ binding causes breakage of the IHM intra- and intermolecular interactions, cooperatively releasing heads from their SRX/HRX state (Szent-Györgyi et al., 1999; Zhao and Craig, 2003; Jung et al., 2008; Chantler, 2016). Released Ca²⁺-activated heads can move freely, similarly to phosphorylated heads of tarantula, interacting with actin to generate force. As this process involves binding of Ca²⁺ to the myosin heads and not the slower, enzymatic phosphorylation of the light chains, full activation can occur rapidly (Zhao and Craig, 2003), making all heads immediately available for the powerful twitches that produce the strong swimming motions of this species. In the absence of a thin filament switch in scallops, sentinel heads may not be required as the heads directly sense Ca²⁺ activation.Interactions between IHMs along their helical tracks are common to all thick filaments in the SRX/HRX state. Reconstructions of tarantula, scallop, and vertebrate thick filaments, representing three distinct mechanisms of activation, all show such interactions, typically between the FH motor domain of one IHM and the BH regulatory domain of its neighbor in the next crown closer to the filament center (Figs. 2 C and ​and3;3; Woodhead et al., 2005; Zoghbi et al., 2008; AL-Khayat et al., 2013; Woodhead et al., 2013). As described earlier, these intermolecular interactions appear to stabilize the IHM conformation, enhancing the fraction of SRX heads in filaments. Concomitantly, they may also underlie rapid thick filament activation, as they provide a direct physical path for cooperative disruption of the IHM and thus exit from the SRX state (Stewart et al., 2010; McNamara et al., 2015). Cooperative exit from the SRX state has been experimentally demonstrated in skeletal and cardiac muscle incubated with mantATP chased by ADP (Stewart et al., 2010; Cooke, 2011; Gollapudi et al., 2021b). Cooperative activation is especially well developed in scallop thick filaments (Szent-Györgyi et al., 1999; Chantler, 2016), where it may underlie the rapid Ca²⁺ activation leading to the strong swimming motions of this species. The intermolecular interactions along helices and around each crown in scallops (Fig. 3 C; Woodhead et al., 2013) connect all heads in an extensive network that could be rapidly disrupted by Ca²⁺ binding.How is SRX modulated?The stability of the IHM and the level of SRX can be modulated in several ways, including myosin RLC phosphorylation and, in vertebrates, phosphorylation of MyBP-C (Stewart et al., 2010; Nag and Trivedi, 2021). RLC phosphorylation in tarantula greatly reduces the fraction of SRX and HRX heads and their ATP turnover times (Naber et al., 2011) while increasing DRX heads. This correlates with a decrease in helical ordering and extension of heads from the filament backbone (suggesting disruption of the IHM), as demonstrated by x-ray diffraction (Padrón et al., 2020) and EM (Craig et al., 1987). Phosphorylation not only activates heads but also maintains a memory of activation following the extensive phosphorylation that occurs in a tetanus; this can greatly potentiate subsequent contraction (Padrón et al., 2020). Disruption of the IHM and extension of heads toward neighboring actin filaments in the RX period following a tetanus presumably enables the stronger and more rapid interaction with actin of a post-tetanic contraction (Padrón et al., 2020). While the phosphorylated RX state that follows a tetanus would temporarily consume more ATP, it could provide a survival benefit by enabling stronger contractions when escaping predators or capturing prey. Following such periods of activity, the RLCs again become dephosphorylated, and the energy-saving SRX/HRX state, with ordered, interconnected IHMs lying along the filament surface, returns (Padrón et al., 2020), characterizing the long periods of inactivity in the life of the tarantula, when ATP savings are critical.Vertebrate skeletal muscle RLCs can also be phosphorylated, and phosphorylation correlates with post-tetanic potentiation (Sweeney et al., 1993) together with disordering of myosin helices and extension of heads from the filament surface (Levine et al., 1996; Yamaguchi et al., 2016), again suggestive of IHM disruption. As with tarantula, a corresponding reduction in the SRX state with phosphorylation (Stewart et al., 2010; Cooke, 2011; Gollapudi et al., 2021b), with a temporarily greater resting ATP consumption, pays for the greater contractility available following phosphorylation.Vertebrates have a second means of modulating the SRX state, which may provide finer control of thick filament activation/relaxation than with invertebrates. MyBP-C binds to myosin in the middle one-third of each half of the thick filament (the C-zone; Craig and Offer, 1976; Flashman et al., 2004; Luther et al., 2008), and several lines of evidence demonstrate that the SRX state is more pronounced in these regions (McNamara et al., 2016, 2019; Nelson et al., 2020; Nag and Trivedi, 2021), reaching as high as 90% (Nelson et al., 2020); this correlates with the clearest delineation of IHMs in reconstructions of the thick filament C-zone (Zoghbi et al., 2008; AL-Khayat et al., 2013). Toward the tips of the filament (the distal or D-zone), heads are less ordered (Brunello et al., 2020; R. Craig, unpublished EM data), and the SRX state is diminished (Nelson et al., 2020). In MyBP-C knockout mice, the IHM configuration is weakened or abolished (Zoghbi et al., 2008), and the SRX state is disrupted (McNamara et al., 2016). Thus MyBP-C appears to enhance the SRX state, apparently by stabilizing the IHM. This stabilization is further modulated by phosphorylation of MyBP-C, occurring in the heart in response to β-adrenergic stimulation. Enhancement of cardiac contractility by cardiac MyBP-C (cMyBP-C) phosphorylation may result in part from depression of cMyBP-C’s stabilizing effect on the SRX, which coincides with weakening of the IHM (Kensler et al., 2017; Caremani et al., 2019b; Irving and Craig, 2019; McNamara et al., 2019). These data overall imply that MyBP-C enhances energy saving by stabilizing the IHM structure and that this is modulated in the heart by cMyBP-C phosphorylation (McNamara et al., 2019).Importantly, in the healthy heart, RLC and MyBP-C phosphorylation are not zero but ∼50% (Chang et al., 2015) and ∼60% of maximum (Previs et al., 2012), respectively. This would suggest a partial weakening of the SRX state (compared with zero phosphorylation) during normal cardiac activity, which may poise myosin heads optimally between sequestration in the IHM (to save energy) and availability for interaction with actin to generate force, with fine-tuning of these levels available upon further phosphorylation. Thus, we assume that the level of SRX of a muscle (degree of IHM formation) is tuned to the physiological needs of the moment, with the goal over time of minimizing energy consumption within these limits. Strikingly, phosphorylation levels of MyBP-C and RLC can both decrease in heart failure (El-Armouche et al., 2007; Toepfer et al., 2013), which would predict a higher fraction of SRX heads. This may reduce the need for energy under these adverse circumstances but may also contribute to the compromised contractility of the failing heart.Animals can save energy when food supplies are scarce or weather conditions adverse by a reduction in body temperature. This occurs naturally in ectotherms (cold-blooded animals) when exterior temperatures drop and by hibernation in some endotherms (warm-blooded animals), where body temperature and metabolism reduce to low values. Does enhanced SRX play a role in energy conservation under these circumstances? X-ray diffraction of both mammalian and tarantula muscle at low temperatures (10°C) suggests that the number of myosin motors in the helically ordered, IHM conformation decreases substantially compared with temperatures nearer physiological levels (Malinchik et al., 1997; Xu et al., 1997; Caremani et al., 2019a, 2021; Ma et al., 2021a); modeling of tarantula suggests that it is specifically the FHs that are disordered, while the BHs remain ordered (Ma et al., 2021a). This disordering suggests that the SRX state may actually be reduced, rather than enhanced, in cold temperatures. Other factors may contribute to energy conservation by muscle under cold conditions (Caremani et al., 2021; Ma et al., 2021a): myosin heads become refractory to actin binding at low temperature (Caremani et al., 2019a), and the disordered FHs, containing ADP.Pi at the active site, may transition toward the ATP conformation, inhibiting ATP turnover (Xu et al., 1999; Ma et al., 2021a). The impact of torpor (a form of hibernation) on the SRX state has recently been explicitly studied in the 13-lined ground squirrel (Ictidomys tridecemlineatus; Toepfer et al., 2020). During torpor, core body temperature drops to 5°C, metabolic rate to 3% of basal levels, and heart rate from 311 to 6 beats per min; hummingbirds undergo a similar dramatic reduction in metabolic rate to save energy each night (Shankar et al., 2020). The fraction of SRX heads found in cardiac muscle removed from the ground squirrel during torpor was reported to increase from 65 to 75%, contrary to expectation from the low-temperature x-ray studies of mammalian muscle described above that would imply a decrease in the number of IHMs. This discrepancy could be due to performance of the SRX measurements at 21°C (Toepfer et al., 2020) rather than the low temperatures used in the x-ray experiments that revealed helical disordering. The latter would presumably reflect the thick filament structure most accurately at actual torpor temperatures. We speculate that the phenomenon of iguanas (ectotherms) falling from trees and becoming immobile when environmental temperatures are abnormally low (Stroud et al., 2020) may be caused in part by the refractory impact of temperature on the ability of their myosin heads to bind to actin; a similar effect may have contributed to extinction of the dinosaurs during the global cooling that followed the asteroid impact of 66 million years ago (Vellekoop et al., 2014).Modulation of the SRX and its putative structural correlate, the IHM, may play a critical role in a number of cardiac diseases and their treatment, summarized elsewhere (Alamo et al., 2017; Nag et al., 2017; Yotti et al., 2019; Trivedi et al., 2020; Daniels et al., 2021; Schmid and Toepfer, 2021). Hypertrophic cardiomyopathy (HCM) is an inherited disease caused by mutations in sarcomeric proteins, including myosin, and characterized by hypercontractility and the inability to fully relax during diastole (Ashrafian et al., 2011). Recent studies show that mutations in the myosin heavy chain (MYH7; accounting for ∼40% of HCM cases) strongly cluster in regions of the myosin molecule involved in the interfaces of the IHM (Alamo et al., 2017; Nag et al., 2017), and lead to a substantial decrease in SRX and increased energy use (Anderson et al., 2018; Adhikari et al., 2019; Sarkar et al., 2020). Disruption of the IHM by such mutations, at the head–head or BH–S2 interface, could release more heads for interaction with actin, accounting for the hypercontractility and impaired relaxation observed (Alamo et al., 2017; Anderson et al., 2018; Spudich, 2019; Sarkar et al., 2020). Experimental evidence for this proposal has been obtained (Adhikari et al., 2019; Sarkar et al., 2020). Mutations in MyBP-C leading to HCM also appear to partially disrupt the SRX state of the myosin heads (Toepfer et al., 2019; Nag and Trivedi, 2021), leading to an increase in the number of DRX heads, which may contribute to the observed hypercontractility (McNamara et al., 2017). Depending on the mutation, cMyBP-C may have a reduced affinity for myosin, or the amount of MyBP-C incorporated into the thick filament may be reduced. It has been proposed that either mechanism would reduce the strength of the cMyBP-C–myosin interaction, releasing heads from the filament backbone (DRX heads), which could be partially responsible for the hypercontractile phenotype observed in patients with these mutations (Colson et al., 2007; Toepfer et al., 2019; Nag and Trivedi, 2021).Impaired relaxation due to HCM mutations in myosin (and other myofibrillar proteins) can be compensated by a recently developed drug, mavacamten, which has been shown to increase the fraction of myosin heads in the SRX state (Anderson et al., 2018; Rohde et al., 2018; Spudich, 2019; Nag and Trivedi, 2021), specifically in the D-zone of thick filaments (Nelson et al., 2020), concomitant with an increased number of molecules folded into the IHM conformation (Anderson et al., 2018; Rohde et al., 2018; Gollapudi et al., 2021b). In thick filaments of intact cardiac muscle, mavacamten increases the degree of quasi-helical ordering of myosin heads, consistent with an increase in the stability or fraction of molecules in the IHM conformation (Anderson et al., 2018), supporting our overall contention that the IHM is the main basis of the SRX state in muscle. Mavacamten inhibits the ATPase activity of isolated S1, specifically by inhibiting phosphate release (Green et al., 2016), suggesting that it enhances the bent (SRX) state of the myosin head, which, based on our earlier reasoning, would lead to the observed increase in the fraction of molecules in the IHM (Anderson et al., 2018).Future directionsWe currently know the structure of the IHM at ∼15 Å resolution in tarantula (Yang et al., 2016) and insect (Hu et al., 2016) filaments and ∼4 Å resolution in single smooth muscle myosin molecules (Scarff et al., 2020; Yang et al., 2020). Improvements in resolution should enhance visualization of the side-chain interactions that stabilize the SRX state in both filament and monomer. For filaments, cryo-EM of tarantula offers the greatest potential, owing to the stability of its helices. To better understand how mutations in the IHM interfaces lead to the hypercontractility of HCM, cryo-EM of vertebrate cardiac thick filaments is required to improve the resolution beyond the current ∼40 Å (obtained with negative staining; Zoghbi et al., 2008; AL-Khayat et al., 2013), a difficult task owing to the lability and quasi-helical symmetry of the vertebrate thick filament head array. Differences in the levels of SRX in the C- and D-zones of the thick filaments (Nelson et al., 2020) suggest differences in IHM stability or interactions, which will need to be analyzed by cryo-EM, again a significant task for the reasons stated above and because of the small size of these zones. The most detailed insights into IHM intramolecular interactions should come from a high-resolution structure of the IHM in isolated cardiac myosin molecules, which is urgently needed. This could potentially be obtained by x-ray crystallography or cryo-EM of IHM constructs (e.g., 25 heptad; Fig. 1; Anderson et al., 2018), although this is likely to be hampered by the relative instability of the structure. Incubation with mavacamten may improve its stability (Anderson et al., 2018). How interaction of MyBP-C with myosin enhances the SRX state is another fertile but challenging area of investigation, which may require cryo-electron tomography of thick filaments or intact myofibrils (Burbaum et al., 2021) or single particle cryo-EM studies of MyBP-C–IHM complexes. Experiments suggest that thick filaments in muscle are in equilibrium with a pool of myosin monomer, which may play a role in thick filament assembly/disassembly during development, hypertrophy, and myosin turnover (Saad et al., 1986; Katoh et al., 1998; Ojima, 2019). Myosin monomers at physiological ionic strength form IHMs with a folded tail structure and slow ATP turnover rate, which may act as a transport form from ribosome to filament (Ankrett et al., 1991; Katoh et al., 1998; Jung et al., 2008). Whether the SRX/IHM structure in thick filaments affects the monomer–polymer equilibrium is an area for future investigation.ConclusionThe SRX state of thick filaments plays a crucial role in the energy balance of muscle and is ubiquitous across the animal kingdom. It results from a conformation of the myosin head in which ATP turnover is strongly inhibited, minimizing ATP use. This conformation is stabilized by intramolecular interactions when it is incorporated into the IHM and by additional intermolecular interactions when IHMs are assembled into thick filaments, increasing the fraction of energy-saving SRX heads. Thus, we propose that IHMs, helically organized along the thick filament surface in the relaxed state, are the major basis of SRX in living muscle. An HRX state, found in thick filaments of some invertebrates and in the folded (storage) form of smooth muscle and nonmuscle myosin molecules, is also based on the IHM but in these cases results from additional intra/intermolecular interactions. A filament in which every head was in SRX, while maximally saving ATP, would not be useful in contraction: at any time, a small fraction of heads is dissociated (DRX or sentinel heads) and available to sense thin filament activation and generate initial force. The SRX state is down-regulated in situ by RLC and MyBP-C phosphorylation in response to physiological requirements (activation), which disrupt the IHM, releasing heads for interaction with actin. Mutations in myosin or MyBP-C causing HCM disrupt the SRX state, causing hypercontractility, which can be reversed by drugs that stabilize the IHM and thus the SRX.Online supplemental materialVideo 1 shows human cardiac thick filament reconstruction fitted with a human cardiac atomic model to demonstrate an apparent absence of interaction between S2 from one level of heads and the SH3 and converter domains of the next level. Video 2 shows tarantula thick filament reconstruction fitted with a tarantula IHM atomic model to show that in this species there is an interaction between S2 from one level of heads and the SH3 and converter domains of the next level.     

Chlorophyll f-driven photosynthesis in a cavernous cyanobacterium

Chlorophyll (Chl) f is the most recently discovered chlorophyll and has only been found in cyanobacteria from wet environments. Although its structure and biophysical properties are resolved, the importance of Chl f as an accessory pigment in photosynthesis remains unresolved. We found Chl f in a cyanobacterium enriched from a cavernous environment and report the first example of Chl f-supported oxygenic photosynthesis in cyanobacteria from such habitats. Pigment extraction, hyperspectral microscopy and transmission electron microscopy demonstrated the presence of Chl a and f in unicellular cyanobacteria found in enrichment cultures. Amplicon sequencing indicated that all oxygenic phototrophs were related to KC1, a Chl f-containing cyanobacterium previously isolated from an aquatic environment. Microsensor measurements on aggregates demonstrated oxygenic photosynthesis at 742 nm and less efficient photosynthesis under 768- and 777-nm light probably because of diminished overlap with the absorption spectrum of Chl f and other far-red absorbing pigments. Our findings suggest the importance of Chl f-containing cyanobacteria in terrestrial habitats.The textbook concept that oxygenic phototrophs primarily use radiation in the visible range (400–700 nm) has been challenged by several findings of unique cyanobacteria and chlorophylls (Chl) over the past two decades (Miyashita et al., 1996; Chen et al., 2010; Croce and van Amerongen, 2014) Unicellular cyanobacteria in the genus Acaryochloris primarily employ Chl d for oxygenic photosynthesis at 700–720 nm (Miyashita et al., 1996) and thrive in shaded habitats with low levels of visible light but replete of near-infrared radiation (NIR, >700 nm, Kühl et al., 2005; Behrendt et al., 2011, 2012). Furthermore, Chl f was recently discovered in filamentous (Chen et al., 2010; Airs et al., 2014; Gan et al., 2014) and unicellular cyanobacteria (Miyashita et al., 2014), enabling light harvesting even further into the NIR region up to ∼740 nm, often aided by employing additional far-red light-absorbing pigments such as Chl d and phycobiliproteins (Gan et al., 2014). Whereas the biochemical structure (Willows et al., 2013) and biophysical properties (Li et al., 2013; Tomo et al., 2014) of Chl f have been studied in detail, the actual importance of this new chlorophyll for photosynthesis is hardly explored (Li et al., 2014).Chlorophyll f has been found in cyanobacteria originating from aquatic/wet environments: the filamentous Halomicronema hongdechloris from stromatolites in Australia (Chen et al., 2012), a unicellar morphotype (Strain KC1) from Lake Biwa in Japan (Akutsu et al., 2011; Miyashita et al., 2014) and a filamentous Leptolyngbya sp. strain (JSC-1, Gan et al., 2014) from a hot-spring and in a unicellular Chlorogloeopsis fritschii strain from rice paddies (Airs et al., 2014). In this study, we report on a unicellular Chl f-containing cyanobacterium originating from a wet cavernous habitat and demonstrate its capability of NIR-driven oxygenic photosynthesis. Enrichments of the new cyanobacterium were obtained from a dense dark green-blackish biofilm dominated by globular morphotypes of Nostocaceae growing on moist limestone outside Jenolan Caves, NSW, Australia. The sampling site was heavily shaded even during mid-day with low irradiance levels of 400- to 700-nm light varying from 0.5 to 5 μmol photons m⁻² s⁻¹. Biofilms were carefully scraped off the substratum and kept in shaded zip-lock bags in a moist atmosphere until further processing. Samples were then incubated at 28 °C in a f/2 medium under NIR at 720 nm (∼10 μmol photons m⁻² s⁻¹) yielding conspicuous green cell aggregates after several months of incubation. Repeated transfer of the aggregates into fresh medium resulted in a culture predominated by green cell clusters (Figure 1a), exhibiting orange-red fluorescence upon excitation with blue light (Figure 1b). Transmission electron microscopy revealed that the green clusters consisted of slightly elongated unicellular cyanobacteria (∼1- to 2-μm wide and ∼2- to 3-μm long), with stacked thylakoids and embedded in a joint polymer matrix (Figure 1c). Hyperspectral microscopy (Kühl and Polerecky, 2008) of the clusters revealed distinct troughs in the transmission spectra at absorption maxima indicative of Chl a (675–680 nm) and Chl f (∼720 nm; Figure 1d, red line). In situ spectral irradiance measurements at the sampling site showed strong depletion of visible wavelengths in the 480- to 710-nm range (Figure 1d, gray line), whereas highest light levels were found in the near-infrared region of the solar spectrum at 710–900 nm. The presence of Chl a and f was further confirmed in enrichment cultures using high-performance liquid chromatography-based pigment analysis (Figure 1e, Supplementary Figure S1), while no Chl d was detected. In addition, weak spectral signatures of carotenoids and phycobilins, with absorption occurring at ∼495 and 665 nm, were evident in the hyperspectral data. Cyanobacteria, including those producing Chl d/f, are known to actively remodel their pigment content in response to the available light spectrum (Stomp et al., 2007; Chen and Scheer, 2013; Gan et al., 2014) and Chl d/f has almost exclusively been found in cyanobacteria grown under far-red light and not under visible light (Kühl et al., 2005; Chen et al., 2010; Airs et al., 2014; Gan et al., 2014; Li et al., 2014; Miyashita et al., 2014). Recent work describes this acclimation response as ‘Far-Red Light photoacclimation'' (FaRLiP), which, in strain JSC-1, comprises a global change in gene expression and structural remodeling of the PSII/PSI core proteins and phycobilisome constituents (Gan et al., 2014). The extent to which this arrangement results in optimized photosynthetic performance is only known for the NIR (=710 nm)-acclimated strain JSC-1, where exposure to wavelengths >695 nm resulted in 40% higher O₂ evolution rates as compared with cells that were previously adapted to red light (645 nm; Gan et al., 2014). Yet the discrimination of actinic wavelengths and their relative effect on gross photosynthesis in Chl f-containing cells needs further investigation. Using an O₂ microsensor and the light–dark shift method (Revsbech et al., 1983) on embedded Chl f-containing aggregates, we found maximal gross photosynthesis rates (∼1.06 μmol O₂ cm⁻³ s⁻¹) to occur at irradiances of ∼250 μmol photons m⁻² s⁻¹ of 742 nm (half-bandwidth, HBW, 25 nm, Figures 2a and b) with light saturation to occur very early at ∼35 μmol photons m⁻² s⁻¹. Further red-shifted actinic light, that is, 768 nm (HBW 28 nm) and 777 nm (HBW 30 nm), yielded lower O₂ evolution rates, which, in all likelihood, are an effect of the diminished overlap with far-red light-absorbing pigments, including Chl f (Figures 2a and b). As O₂ evolution rates were measured on non-axenic cell aggregates, 16S rDNA amplicon sequencing was employed to determine the microbial diversity found within the enrichment culture. This revealed the presence of a variety of bacterial types, including anoxygenic phototrophs, yet all sequences for known oxygenic phototrophs in the data set (∼9.3% of all reads on the order level, Supplementary Figure S2) formed a single operational taxonomic unit (OTU) closely affiliated with the Chl f-containing strain KC1 (Miyashita et al., 2014, Figure 2c).Open in a separate windowFigure 1Imaging and pigment analysis of Chl f-containing cyanobacteria isolated from a cavernous low-light environment. (a) Representative bright field microscope image of cultured cells grown under 720 nm NIR. (b) Fluorescence image of the same cells as in a, excited at 450–490 nm, with emission being detected at >510 nm. (c) Transmission electron microscopy of a Chl f-containing cyanobacterium with densely stacked thylakoid membranes. (d) Transmittance spectrum of cell aggregate determined by hyperspectral imaging (red line). Ambient light conditions at the site of isolation (gray line), as measured by a spectroradiometer. Note the Chl f-specific in vivo absorption at ∼720 nm in the transmittance spectrum (dotted line). Small insert picture denotes the cells and area of interest (black arrow) from which the spectrum was taken. (e) In vitro absorption spectrum of Chl f extracted from enrichment cultures and analyzed via high-performance liquid chromatography. The two Chl f-specific absorption peaks (404 and 704 nm in acetone:MeOH solvent) are indicated.Open in a separate windowFigure 2Taxonomic affiliation and O₂ evolution of Chl f-containing cells as determined by O₂ microelectrode measurements and 16 S rDNA amplicon sequencing. (a) Emission spectra of narrow-band light-emitting diodes (LEDs) used in this study, with peak emissions at 742, 768 and 777 nm indicated by a–c, respectively. (b) Gross photosynthesis measured via an O₂ microsensor placed in a clump of agarose-embedded Chl f-containing cells. Different NIR irradiance was administered by the LEDs in a and by altering the distance of the LEDs to the embedded cells. (c) Phylogenetic affiliation of known Chl f and/or Chl d-containing cyanobacteria (highlighted in gray) and their respective habitat/place of isolation. Taxonomy was determined by clustering all known oxygenic phototrophs found in enrichment cultures from this study (at order level) into a single OTU (=292 bp length, see Supplementary Materials for details). Phylogeny was inferred using Maximum-likelihood in conjunction with the GTR +I +G nucleotide substitution model, tree stability was tested using bootstrapping with 100 replicates. The analysis involved 39 nucleotide sequences each truncated to a length of 292 bp. Here, the green-sulphur bacterium Chlorobium tepidum TLS was chosen as the outgroup.This advocates that cells from our enrichment culture are related to KC1 cells and supports, in conjunction with further morphological-, physiological- and ultrastructural evidence, that Chl f is extending the usable light spectrum for oxygenic photosynthesis in a cavernous low-light environment. Given the lifestyle and known habitats of recognized Chl d/f-producing cyanobacteria (Figure 2c), we propose that many, if not all, surface-associated cyanobacteria are intrinsically capable of producing far-red light-absorbing pigments and to actively employ them in oxygenic photosynthesis as a result of FaRLiP or similar, yet unknown, mechanisms.     

Mitofusins: ubiquitylation promotes fusion

Mitochondrial genes including Mfn2 are at the center of many diseases, underscoring their potential as a therapeutical target. The Chen group now identified 15-oxospiramilactone as a chemical inhibitor of the mammalian deubiquitylase USP30, acting on Mfn1 and Mfn2.Mitofusins, Fzo1 in yeast and Mfn1 and Mfn2 in mammals, are ubiquitylated and this post-translational modification has both positive and negative consequences on mitochondrial fusion¹. The process of ubiquitylation requires enzymes belonging to three classes of proteins called E1, E2 and E3, which catalyze a cascade of successive steps leading to the covalent attachment of the modifier to its target protein². Deubiquitylating enzymes render this modification reversible, thus offering further possibilities for regulation². Ubiquitylation of mitofusins leads to their proteolyic breakdown, inhibiting fusion of mitochondria that consequently undergo fragmentation (Figure 1, left panel)^1,3. For example in response to mitochondrial depolarization or apoptotic stimuli, E3 ligases like Parkin and Huwe1 ubiquitylate and target Mfn1 and Mfn2 to the proteasome (Figure 1, left panel)^3,4. However, ubiquitylation of mitofusins is a dual process and a non-proteolytic role of mitofusin ubiquitylation that promotes mitochondrial fusion is now emerging¹. This opposing mechanism was first described in yeast, where the isopeptidases Ubp12 and Ubp2 that deubiquitylate Fzo1 have been identified⁵. Inhibition and activation of mitochondrial fusion by ubiquitylation enable different morphologies of mitochondria ranging from a multitude of small organelles to a hyperconnected network (Figure 1)⁵. In a recent paper published in Cell Research, Yue et al.⁶ reveal that a similar process is present in mammalian cells. The authors report that the isopeptidase USP30 acts on ubiquitylated forms of Mfn1 and Mfn2 that stimulate mitochondrial fusion (Figure 1, right panel). This discovery identifies for the first time in mammals a positive role of ubiquitylation in the regulation of Mfn1 and Mfn2 fusion activity⁶.Open in a separate windowFigure 1Dual roles of ubiquitylation and deubiquitylation of mitofusins Mfn1 and Mfn2, the key effectors for mitochondrial fusion, in regulating mitochondrial fusion. On one hand, ubiquitylation of Mfn1 and Mfn2 by E3 ligases like Parkin or Huwe1 targets their proteasomal degradation and inhibits mitochondrial fusion, which results in mitochondrial fragmentation due to unopposed fission events. On the other hand, ubiquitylation of Mfn1 and Mfn2 by an unknown E3 ligase enhances their activity and promotes mitochondrial fusion. This positive regulation is counteracted by the deubiquitylase USP30, targeted by the small molecule inhibitor 15-oxospiramilactone.Moreover, Yue et al.⁶ identified the first small molecule inhibitor of mitochondrial fusion, 15-oxospiramilactone, which targets USP30 in both human and mouse cell lines. 15-oxospiramilactone is a semi-synthetic diterpene alkaloid of 330 Da that can be chemically synthetized through an oxidation reaction from spiramines extracted from the roots of a Chinese herbal medicine Spiraea japonica (Rosaceae). Inhibition of USP30 increased ubiquitylation of Mfn1 and Mfn2 and led to an elongation of the mitochondrial network (Figure 1, right panel)^6,7. USP30 is a cysteine ubiquitin isopeptidase N-terminally anchored to the outer membrane of mitochondria, which was previously shown to regulate mitochondrial morphology dependent on Mfn1 and Mfn2⁷. USP30 knockdown leads to mitochondrial elongation, a phenotype rescued by ectopic expression of wild-type USP30, while the catalytically inactive mutant C77S USP30 failed to revert⁷. Yue et al.⁶ show that 15-oxospiramilactone directly interacts with USP30, which also depends on its catalytically active cysteine, and inhibits the DUB activity of USP30 on tetraubiquitin chains. Moreover, they demonstrate that inhibition of USP30 and subsequent mitochondrial elongation are due to stimulated mitochondrial fusion activity, apparently with no influence on mitochondrial fission⁶. Concomitantly, cells showed increased ubiquitylation of Mfn1 and Mfn2 without significant changes in protein turnover of these two proteins⁶. Therefore, in analogy to findings in yeast, ubiquitylation of Mfn1 and Mfn2 can either signal them to activate mitochondrial fusion or in contrast promote their proteasomal degradation, resulting in mitochondrial fission (Figure 1).Importantly, 15-oxospiramilactone reverts the mitochondrial fragmentation phenotype of single Mfn-knockout (Mfn1^−/− or Mfn2^−/−) cells, suggesting that mitochondrial fusion depends on the ubiquitylation of both mitofusin proteins⁶. In yeast, the importance of ubiquitylation was proven by directly attaching a deubiquitylase to Fzo1, which resulted in a non-ubiquitylated and non-functional Fzo1 protein⁵. In addition, the identification and the subsequent mutagenesis study of the ubiquitylation sites in Fzo1 confirmed an interplay between ubiquitylation and oligomerization in mitochondrial fusion in S. cerevisiae⁵. Impairing the yeast E3 ligase SCF^Mdm30 inhibited mitochondrial fusion and, conversely, ablation of UBP12 led to more fusion events^5,8. Given this new identification of USP30 as the functional orthologue of the yeast Ubp12, future studies will certainly aim at the identification of the E3 ligase counterpart of SCF^Mdm30 and ubiquitylation sites in Mfn1 and Mfn2. In addition to USP30 inhibition, other conditions leading to mitochondrial hyperfusion have been previously observed, such as mild stress conditions that increase reactive oxygen species (ROS)⁹. Importantly, oxidative stress and mitochondrial fusion are directly linked as ROS induces disulphide switching of Mfn2 to oligomeric forms that promote mitochondrial fusion⁹. It would be interesting to investigate whether 15-oxospiramilactone also affects the generation of disulphide-mediated mitofusin oligomers, thus activating mitochondrial fusion.Mutations in Mfn2 are causative for the Charcot-Marie-Tooth type 2A neuropathy, an autosomal dominant disorder of the peripheral nervous system that mainly affects axons and lower extremities¹. Deficiencies in Parkin and Mfn2 ubiquitylation were also linked to Parkinson''s disease³. In addition to neuropathies, Mfn2 is associated to other diseases like cardiomyophathies and diabetes¹. Yue et al.⁶ found that 15-oxospiramilactone reverted phenotypes arising from the lack of Mfn1 or Mfn2. It restored the normal distribution of mtDNA, allowed recovery of the ΔΨm and increased the ATP levels and OXPHOS capacity of the rebuilt mitochondrial network. Therefore, this study potentiates 15-oxospiramilactone for therapeutical benefit. The anti-cancer properties of 15-oxospiramilactone, also named S3 or NC043, have been previously reported^10,11. It inhibits Wnt/β-catenin signaling and colon cancer cell tumorigenesis in a xenograft model¹⁰. Moreover, 15-oxospiramilactone increases Bim expression and apoptosis to inhibit tumor growth from Bax^−/−/Bak^−/− cells implanted in mice¹¹. However, Yue et al.⁶ show that the effect of 15-oxospiramilactone in mitochondrial fusion is independent of apoptosis and suggest that the difference is due to drug concentration. Indeed, previous anti-cancer studies used 15-oxospiramilactone at a concentration range of 3.75-15 μM^10,11, whereas 2 μM suffice to inhibit USP30⁶. Further studies are needed to address the clinical relevance of 15-oxospiramilactone and USP30 in Mfn2-associated diseases.     

Big steps toward understanding dynein

Dynein is a microtubule-based molecular motor that is involved in various biological functions, such as axonal transport, mitosis, and cilia/flagella movement. Although dynein was discovered 50 years ago, the progress of dynein research has been slow due to its large size and flexible structure. Recent progress in understanding the force-generating mechanism of dynein using x-ray crystallography, cryo-electron microscopy, and single molecule studies has provided key insight into the structure and mechanism of action of this complex motor protein.It has been 50 years since dynein was discovered and named by Ian Gibbons as a motor protein that drives cilia/flagella bending (Gibbons, 1963; Gibbons and Rowe, 1965). In the mid-1980s, dynein was also found to power retrograde transport in neurons (Paschal and Vallee, 1987). Subsequently, the primary amino acid sequence of the cytoplasmic dynein heavy chain, which contains the motor domain, was determined from the cDNA sequence (Mikami et al., 1993; Zhang et al., 1993). Like other biological motors, such as kinesins and myosins, the amino acid sequence of the dynein motor domain is well conserved. There are 16 putative genes that encode dynein heavy chains in the human genome (Yagi, 2009). Among these is one gene encoding cytoplasmic dynein heavy chain and one encoding retrograde intraflagellar transport dynein heavy chain, while the rest encode for heavy chains of axonemal dyneins. Most of the genes encoding the human dynein heavy chain have a counterpart in Chlamydomonas reinhardtii, which suggests that their functions are conserved from algae to humans.Dynein is unique compared with kinesin and myosin because dynein molecules form large molecular complexes. For example, one axonemal outer arm dynein molecule of C. reinhardtii is composed of three dynein heavy chains, two intermediate chains, and more than ten light chains (King, 2012). Mammalian cytoplasmic dynein consists of two heavy chains and several smaller subunits (Fig. 1 A; Vallee et al., 1988; Allan, 2011). The cargoes of cytoplasmic dynein are various membranous organelles, including lysosomes, endosomes, phagosomes, and the Golgi complex (Hirokawa, 1998). It is likely that one cytoplasmic dynein heavy chain can adapt to diverse cargos and functions by changing its composition.Open in a separate windowFigure 1.Atomic structures of cytoplasmic dynein. (A) Schematic structure of cytoplasmic dynein complex, adapted from Allan (2011). (B) The primary structure of cytoplasmic dynein. (C and D) The atomic model of D. discoideum cytoplasmic dynein motor domain (PDB accession no. 3VKG) overlaid on a microtubule (EMDB-5193; Sui and Downing, 2010) according to the orientation determined by Mizuno et al. (2007) (C) Side view. (D) View from the plus end of microtubule. (E) Schematic domain structure of dynein.Dynein must have a distinct motor mechanism from kinesin and myosin, because it belongs to the AAA+ family of proteins and does not have the conserved amino acid motifs, called the switch regions, present in kinesins, myosins, and guanine nucleotide-binding proteins (Vale, 1996). Therefore, studying dynein is of great interest because it will reveal new design principles of motor proteins. This review will focus on the mechanism of force generation by cytoplasmic and axonemal dynein heavy chains revealed by recent structural and biophysical studies.

Anatomy of dynein

To understand the chemomechanical cycle of dynein based on its molecular structure, it is important to obtain well-diffracting crystals and build accurate atomic models. Recently, Kon and colleagues determined the crystal structures of Dictyostelium discoideum cytoplasmic dynein motor domain, first at 4.5-Å resolution (Kon et al., 2011), and subsequently at 2.8 Å (without the microtubule binding domain) and 3.8-Å (wild type) resolution (Kon et al., 2012). Carter and colleagues also determined the crystal structures of the Saccharomyces cerevisiae (yeast) cytoplasmic dynein motor domain, first at 6-Å resolution (Carter et al., 2011), and later at 3.3–3.7-Å resolution (Schmidt et al., 2012). According to these crystal structures as well as previous EM studies, the overall structure of the dynein heavy chain is divided into four domains: tail, linker, head, and stalk (Fig. 1, B–E). Simply put, each domain carries out one essential function of a motor protein: the tail is the cargo binding domain, the head is the site of ATP hydrolysis, the linker is the mechanical amplifier, and the stalk is the track-binding domain.The tail, which is not part of the motor domain and is absent from crystal structures, is located at the N-terminal ∼1,400 amino acid residues and involved in cargo binding (gray in Fig. 1, B and E). The next ∼550 residues comprise the “linker” (pink in Fig. 1, B–E), which changes its conformation depending on the nucleotide state (Burgess et al., 2003; Kon et al., 2005). This linker domain was first observed by negative staining EM in combination with single particle analysis of dynein c, an isoform of inner arm dynein from C. reinhardtii flagella (Burgess et al., 2003). According to the crystal structures, the linker is made of bundles of α-helices and lies across the AAA+ head domain, forming a 10-nm-long rod-like structure (Fig. 1, C and D). Recent class averaged images of D. discoideum cytoplasmic dynein show that the linker domain is stiff along its entire length when undocked from the head (Roberts et al., 2012). The head (motor) domain of dynein is composed of six AAA+ (ATPase associated with diverse cellular activities) modules (Neuwald et al., 1999; color-coded in Fig. 1, B–E). Although many AAA+ family proteins are a symmetric homohexamer (Ammelburg et al., 2006), the AAA+ domains of dynein are encoded by a single heavy chain gene and form an asymmetric heterohexamer. Among the six AAA+ domains, hydrolysis at the first AAA domain mainly provides the energy for dynein motility (Imamula et al., 2007; Kon et al., 2012). The hexameric ring has two distinct faces: the linker face and the C-terminal face. The linker face is slightly convex and the linker domain lies across this side (Fig. 1 D, left side). The other side of the ring has the C-terminal domain (Fig. 1 D, right side).The stalk domain of dynein was identified as the microtubule-binding domain (MTBD; Gee et al., 1997). It emanates from the C-terminal face of AAA4 and is composed of antiparallel α-helical coiled-coil domain (yellow in Fig. 1, B–E). The tip of the stalk is the actual MTBD. Interestingly, the crystal structures revealed another antiparallel α-helical coiled coil that emerges from AAA5 (orange in Fig. 1, B–E), and this region is called the buttress (Carter et al., 2011) or strut (Kon et al., 2011), which was also observed as the bifurcation of the stalk by negative-staining EM (Burgess et al., 2003; Roberts et al., 2009). The tip of the buttress/strut is in contact with the middle of the stalk and probably works as a mechanical reinforcement of the stalk.

The chemomechanical cycle of dynein

Based on structural and biochemical data, a putative chemomechanical cycle of dynein is outlined in Fig. 2 (A–E). In the no-nucleotide state, dynein is bound to a microtubule through its stalk domain, and its tail region is bound to cargoes (Fig. 2 A). The crystal structures of yeast dynein are considered to be in this no-nucleotide state. When ATP is bound to the AAA+ head, the MTBD quickly detaches from the microtubule (Fig. 2 B; Porter and Johnson, 1983). The ATP binding also induces “hinging” of the linker from the head (Fig. 2 C). According to the biochemical analysis of recombinant D. discoideum dynein (Imamula et al., 2007), the detachment from the microtubule (Fig. 2, A and B) is faster than the later hinging (Fig. 2, B and C). As a result of these two reactions, the head rotates or shifts toward the minus end of the microtubule (for more discussion about “rotate” versus “shift” see the “Dyneins in the axoneme” section) and the MTBD steps forward. The directionality of stepping seems to be mainly determined by the MTBD, because the direction of dynein movement does not change even if the head domain is rotated relative to the microtubule by insertion or deletion of the stalk (Carter et al., 2008). In the presence of ADP and vanadate, dynein is considered to be in this state (Fig. 2 C).Open in a separate windowFigure 2.Presumed chemomechanical cycle and stepping of dynein. (A–E) Chemomechanical cycle of dynein. The pre- and post-power stroke states are also called the primed and unprimed states, respectively. The registries of the stalk coiled coil are denoted as α and β according to Gibbons et al. (2005). (F and G) Processive movement of kinesin (F) and dynein (G). (F) Hand-over-hand movement of kinesin. A step by one head (red) is always followed by the step of another head (green). The stepping of kinesin is on one protofilament of microtubule. (G) Presumed stepping of dynein. The step size varies and the interhead separation can be large. A step by one head (red) is not always flowed by the step of another head (green). (H) A model of strain-based dynein ATPase activation. (G, top) Without strain, the gap between the AAA1 and AAA2 is open and the motor domain cannot hydrolyze ATP. (G, bottom) Under a strain imposed between MTBD and tail (thin black arrows), the gap becomes smaller (thick black arrows) and turns on ATP hydrolysis by dynein.After the MTBD rebinds to the microtubule at the forward site (Fig. 2 D), release of hydrolysis products from the AAA+ head is activated (Holzbaur and Johnson, 1989) and the hinged linker goes back to the straight conformation (Fig. 2 E; Kon et al., 2005). The crystal structure of D. discoideum dynein is considered to be in the state after phosphate release and before ADP release. This straightening of the linker is considered to be the power-generating step and brings the cargo forward relative to the microtubule.

The MTBD of dynein

As outlined in Fig. 2, the nucleotide state of the head domain may control the affinity of the MTBD to the microtubule. Conversely, the binding of the MTBD to the microtubule should activate the ATPase activity of the head domain. This two-way communication is transmitted through the simple ∼17-nm-long α-helical coiled-coil stalk and the buttress/strut, and its structural basis has been a puzzling question.Currently there are three independent MTBD atomic structures in the Protein Data Bank (PDB): One of the crystal structures of the D. discoideum dynein motor domain contains the MTBD (Fig. 3 A), and Carter et al. (2008) crystallized the MTBD of mouse cytoplasmic dynein fused with a seryl tRNA-synthetase domain (Fig. 3 C). The MTBD structure of C. reinhardtii axonemal dynein was solved using nuclear magnetic resonance (PDB accession no. 2RR7; Fig. 3 B). The MTBD is mostly composed of α-helices and the three structures are quite similar to each other within the globular MTBD (Fig. 3). Note that dynein c has an additional insert at the MTBD–microtubule interface (Fig. 3 B, inset), whose function is not yet clear. The three structures start to deviate from the junction between the MTBD and the coiled-coil region of the stalk (Fig. 3, A–C, blue arrowheads). Particularly, one of the stalk α-helix (CC2) in D. discoideum dynein motor domain appears to melt at the junction with the MTBD (Fig. 3 A, red arrowhead). This structural deviation suggests that the stalk coiled coil at the junction is flexible, which is consistent with the observation by EM (Roberts et al., 2009).Open in a separate windowFigure 3.Atomic models of the MTBD of dynein. (A) D. discoideum cytoplasmic dynein (PDB accession no. 3VKH). (B) C. reinhardtii dynein c (PDB accession no. 2RR7). The inset shows the side view, highlighting the dynein c–specific insert. (C) Mouse cytoplasmic dynein (PDB accession no. 3ERR). (D) Mouse cytoplasmic dynein fit to the MTBD–microtubule complex derived from cryo-EM (PDB accession no. 3J1T). All the MTBD structures were aligned using least square fits and color-coded with a gradient from the N to C terminus. CC1, coiled coil helix 1; CC2, coiled coil helix 2. The blue arrowheads points to the junction between MTBD and the stalk, where a well-conserved proline residue (colored pink) is located. In C and D, two residues (isoleucine 3269 and leucine 3417) are shown as spheres. The two residues form hydrophobic contacts in the β-registry (C), whereas they are separated in the α-registry (D) because of the sliding between the two α-helices (blue and red arrows). Conformational changes observed in the mouse dynein MTBD in complex with a microtubule by cryo-EM are shown by black arrows. Note that the cryo-EM density map does not have enough resolution to observe sliding between CC1 and CC2. The sliding was modeled based on targeted molecular dynamics (Redwine et al., 2012).Various mechanisms have been proposed to explain how the affinity between the MTBD and a microtubule is controlled. Gibbons et al. (2005) proposed “the helix-sliding hypothesis” (for review see Cho and Vale, 2012). In brief, this hypothesis proposes that the sliding between two α-helices CC1 and CC2 (Fig. 3, C and D; blue and red arrows) may control the affinity of this domain to a microtubule. When Gibbons’s classification (Gibbons et al., 2005) of the sliding state is applied to the three MTBD structures, the stalk in the D. discoideum dynein motor domain is in the “α-registry” state (not visible in Fig. 3 A because of the melting of CC2), which corresponds to the strong binding state. However, the mouse cytoplasmic and C. reinhardtii axonemal MTBDs have the “β-registry” stalk (Fig. 3 C), which corresponds to the weak binding state.To observe conformational changes induced by the α-registry and/or microtubule binding, Redwine et al. (2012) solved the structure of mouse dynein MTBD in complex with a microtubule at 9.7-Å resolution using cryo-EM and single particle analysis. The MTBD was coupled with seryl tRNA-synthetase to fix the stalk helix in the α-registry. At this resolution, α-helices are visible, and they used molecular dynamics to fit the crystal structure of mouse MTBD (β-registry) to the cryo-EM density map. According to this result, the first helix H1 moves ∼10 Å to a position that avoids a clash with the microtubule (Fig. 3 D, black arrows). This also induces opening of the stalk helix (CC1). Together with mutagenesis and single-molecule motility assays, Redwine et al. (2012) proposed that this new structure represents the strong binding state. Currently, it is not clear why the MTBD structure of D. discoideum dynein motor domain (α-registry, Fig. 3 A) is not similar to the new α-registry mouse dynein MTBD, and this problem needs to be addressed by further studies.

Structures around the first ATP binding site

Another central question about motor proteins is how Ångstrom-scale changes around the nucleotide are amplified to generate steps >8 nm. For dynein, the interface between the first nucleotide-binding pocket and the linker seem to be the key force-generating element (Fig. 4). The crystal structures of dynein give us clues about how nucleotide-induced conformational changes may be transmitted to and amplified by the linker domain.Open in a separate windowFigure 4.Structures around the first ATP binding site. (A) Schematic domain structure of the head domain. Regions contacting the linker domain are colored purple. (B) AAA submodules surrounding the first nucleotide-binding pocket (PDB accession no. 3VKG, chain A). The linker is connected to AAA1 domain by the “N-loop.” To highlight that the two finger-like structures are protruding, the shadow of the atomic structure has been cast on the plane parallel to the head domain. (C) Interaction between the linker and the two finger-like structures. The pink arrowhead points to the hinge-like structure of the linker. The pink numbers indicates the subdomain of the linker.The main ATP catalytic site is located between AAA1 and AAA2 (Fig. 4, A and B). There are four ADP molecules in the D. discoideum dynein crystal structures, but the first ATP binding site alone drives the microtubule-activated ATPase activity, based on biochemical experiments on dyneins whose ATP binding sites were mutated (Kon et al., 2012).One AAA+ module is composed of a large submodule and a small α submodule (Fig. 4 B). The large α/β submodule is located inside of the ring and the small α submodule is located outside. The large submodule bulges toward the linker face, and the overall ring forms a dome-like shape (Fig. 1 D).The main ATP catalytic site is surrounded by three submodules: AAA1 large α/β, AAA1 small α, and AAA2 large α/β (Fig. 4, A and B). Based on the structural changes of other AAA+ proteins (Gai et al., 2004; Suno et al., 2006; Wendler et al., 2012), the gap between AAA1 and AAA2 modules is expected to open and close during the ATPase cycle.In fact, the size of the gap varies among the dynein crystal structures. The crystal structures of yeast dyneins show a larger gap between AAA1 and AAA2, which might be the reason why no nucleotide was found in the binding pocket. Although Schmidt et al. (2012) soaked the crystals in a high concentration of various nucleotides (up to 25 mM of ATP), no electron densities corresponding to the nucleotide were observed at the first ATP binding site. Among dynein crystal structures, one of D. discoideum dynein (PDB accession no. 3VKH, chain A) has the smallest gap, but it is still considered to be in an “open state” because the arginine finger in the AAA2 module (Fig. 4 B, red) is far from the phosphates of ADP. Because the arginine finger is essential for ATP hydrolysis in other AAA+ proteins (Ogura et al., 2004), the gap is expected to close and the arginine finger would stabilize the negative charge during the transition state of ATP hydrolysis.The presumed open/close conformational change between AAA1 and AAA2 would result in the movement of two “finger-like” structures protruding from the AAA2 large α/β submodule (Fig. 4 B). The two finger-like structures are composed of the H2 insert β-hairpin and preSensor I (PS-I) insert. In D. discoideum dynein crystal structure, the two finger-like structures are in contact with the “hinge-like cleft” of the linker (Fig. 4 C, pink arrowhead). The hinge-like cleft is one of the thinnest parts of the linker, where only one α-helix is connecting between the linker subdomains 2 and 3.In the yeast crystal structures, which have wider gaps between AAA1 and AAA2, the two finger-like structures are not in direct contact with the linker and separated by 18 Å. Instead, the N-terminal region of the linker is in contact with the AAA5 domain (Fig. 4 A). To test the functional role of the linker–AAA5 interaction, Schmidt et al. (2012) mutated a residue involved in the interaction (Phe3446) and found that the mutation resulted in severe motility defects, showing strong microtubule binding and impaired ATPase activities. In D. discoideum dynein crystals, there is no direct interaction between AAA5 and the linker, which suggests that the gap between AAA1 and AAA2 may influence the interaction between the head and linker domain. The contact between the linker and AAA5 may also influence the gap around AAA5, because the gap between AAA5 and AAA6 is large in yeast dynein crystal, whereas the one between AAA4 and AAA5 is large in D. discoideum dynein.The movement of two finger-like structures would induce remodeling of the linker. According to the recent cryo-EM 3D reconstructions of cytoplasmic dynein and axonemal dynein c (Roberts et al., 2012), the linker is visible across the head and there is a large gap between AAA1 and AAA2 in the no-nucleotide state. This linker structure is considered to be the “straight” state (Fig. 2, A and E). In the presence of ADP vanadate, the gap between AAA1 and AAA2 is closed and the N-terminal region of linker is near AAA3, which corresponds to the pre-power stroke “hinged” state (Fig. 2, C and D). The transition from the hinged state to the straight state of the linker is considered to be the force-generating step of dynein.

Processivity of dynein

As the structure of dynein is different from other motor proteins, dynein’s stepping mechanism is also distinct. Both dynein and kinesin are microtubule-based motors and move processively. Based on the single molecule tracking experiment with nanometer accuracy (Yildiz et al., 2004), it is widely accepted that kinesin moves processively by using its two motor domain alternately, called the “hand-over-hand” mechanism. To test whether dynein uses a similar mechanism to kinesin or not, recently Qiu et al. (2012) and DeWitt et al. (2012) applied similar single-molecule approaches to dynein.To observe the stepping, the two head domains of yeast recombinant cytoplasmic dynein were labeled with different colors and the movement of two head domains was tracked simultaneously. If dynein walks by the hand-over-hand mechanism, the step size would be 16 nm and the stepping of one head domain would always be followed by the stepping of another head domain (alternating pattern), and the trailing head would always take a step (Fig. 2 F). Contrary to this prediction, both groups found that the stepping of the head domains is not coordinated when the two head domains are close together. These observations indicated that the chances of a leading or trailing head domain stepping are not significantly different (Fig. 2 G; DeWitt et al., 2012; Qiu et al., 2012).This stepping pattern predicts that the distance between the head domains can be long. In fact, the distance between the two head domains is on average ∼18 ± 11 nm (Qiu et al., 2012) or 28.4 ± 10.7 nm (DeWitt et al., 2012), and as large as ∼50 nm (DeWitt et al., 2012). When the two head domains are separated, there is a tendency where stepping of the trailing head is preferred over that of the forward head.In addition, even though the recombinant cytoplasmic dynein is a homodimer, the two heavy chains do not function equally. While walking along the microtubule, the leading head tends to walk on the right side, whereas the trailing head walks on the left side (DeWitt et al., 2012; Qiu et al., 2012). This arrangement suggests that the stepping mechanism is different between the two heads. In fact, when one of the two dynein heavy chains is mutated to abolish the ATPase activity at AAA1, the heterodimeric dynein still moves processively (DeWitt et al., 2012), with the AAA1-mutated dynein heavy chain remaining mostly in the trailing position. This result clearly demonstrates that allosteric communication between the two AAA1 domains is not required for processivity of dynein. It is likely that the mutated head acts as a tether to the microtubule, as it is known that wild-type dynein can step processively along microtubules under external load even in the absence of ATP (Gennerich et al., 2007).These results collectively show that dynein moves by a different mechanism from kinesin. It is likely that the long stalk and tail allow dynein to move in a more flexible manner.

Dyneins in the axoneme

As mentioned in the introduction, >10 dyneins work in motile flagella and cilia. The core of flagella and cilia is the axoneme, which is typically made of nine outer doublet microtubules and two central pair microtubules (“9 + 2,” Fig. 5 A). The axonemes are found in various eukaryotic cells ranging from the single-cell algae C. reinhardtii to human. Recent extensive cryo-electron tomography (cryo-ET) in combination with genetics revealed the highly organized and complex structures of axonemes that are potentially important for regulating dynein activities (Fig. 5, C and D; Nicastro et al., 2006; Bui et al., 2008, 2009, 2012; Heuser et al., 2009, 2012; Movassagh et al., 2010; Lin et al., 2012; Carbajal-González et al., 2013; Yamamoto et al., 2013).Open in a separate windowFigure 5.Arrangement of axonemal dyneins. (A) The schematic structure of the motile 9 + 2 axoneme, viewed from the base of flagella. (B) Quasi-planar asymmetric movement of the 9 + 2 axoneme typically observed in trachea cilia or in C. reinhardtii flagella. (C and D) 3D structure of a 96-nm repeat of doublet microtubules in the distal/central region of C. reinhardtii flagella (EMDB-2132; Bui et al., 2012). N-DRC, the nexin-dynein regulatory complex; ICLC, intermediate chain/light chain complex. Inner arm dynein subspecies are labeled according to Bui et al. (2012) and Lin et al. (2012). To avoid the confusion with the linker domain of dynein, the structures connecting between outer and inner arm dyneins are labeled as “connecters,” which are normally called “linkers.” Putative ATP binding sites of outer arm dynein determined by biotin-ADP (Oda et al., 2013) are indicated by orange circles. The atomic structure of cytoplasmic dynein is placed into the β-heavy chain of outer arm dynein and its enlarged view is shown in the inset. (D) Two doublet microtubules, viewed from the base of flagella.The basic mechanochemical cycles of axonemal dyneins are believed to be shared with cytoplasmic dynein. Dynein c is an inner arm dynein of C. reinhardtii and used extensively to investigate the conformational changes of dynein, as shown in Fig. 2 (A–E), by combining EM and single-particle analysis (Burgess et al., 2003; Roberts et al., 2012). Structural changes of axonemal dyneins complexed with microtubules are also observed by quick-freeze and deep-etch EM (Goodenough and Heuser, 1982; Burgess, 1995), cryo-EM (Oda et al., 2007), negative-staining EM (Ueno et al., 2008), and cryo-ET (Movassagh et al., 2010). According to these studies, the AAA+ head domains are constrained near the A-tubule in the no-nucleotide state. In the presence of nucleotide, the head domains move closer to the B-tubule and/or the minus end of microtubule, and their appearance becomes heterogeneous, which is consistent with the observation of isolated dynein c that shows greater flexibility between tail and stalk in the ADP/vanadate state (Burgess et al., 2003).One of the controversies about the structural changes of axonemal dyneins is whether their stepping involves “rotation” or “shift” of the head (Fig. 2, B to D). The stalk angle relative to the microtubule seems to be a constant ∼60° irrespective of the nucleotide state (Ueno et al., 2008; Movassagh et al., 2010). This angle is similar to the angle obtained from cryo-EM study of the MTBD–microtubule complex (Redwine et al., 2012). Based on these observations, Ueno et al. (2008) and Movassagh et al. (2010) hypothesize that the “shift” of the head pulls the B-microtubule toward the distal end. However, Roberts et al. (2012) propose that the “rotation” of head and stalk is involved in the stepping based on the docking of dynein c head into an averaged flagella tomogram obtained by Movassagh et al. (2010). This issue needs to be resolved by more reliable and high-resolution data, but these two models may not be mutually exclusive. For example, averaged tomograms may be biased toward the microtubule-bound stalk because tomograms are aligned using microtubules.To interpret these structural changes of axonemal dyneins, docking atomic models of dynein is necessary. According to Roberts et al. (2012), the linker face of inner arm dynein c is oriented outside of axoneme (Fig. 5 D). For outer arm dyneins, we used cryo-EM in combination with biotin-ADP-streptavidin labeling and showed that the ATP binding site, most likely AAA1, is on the left side of the AAA+ head (Fig. 5 C; Oda et al. (2013)). Assuming that the stalks extend out of the plane toward the viewer, the linker face of outer arm dynein is oriented outside of axoneme (Fig. 5 C, inset; and Fig. 5 D). If it were the opposite, the AAA1 would be located on the right side of the AAA+ head. In summary, both inner and outer arm dynein seem to have the same arrangement, with their linker face oriented outside of the axoneme (Fig. 5 D).A unique characteristic of axonemal dyneins is that these dyneins are under precise temporal and spatial control. To generate a planer beating motion (Fig. 5 B), dyneins should be asymmetrically controlled, because the dyneins located on doublets 2–4 drive the effective stroke, whereas the ones on doublets 6–8 drive the recovery stroke (Fig. 5 A). Based on the cryo-ET observation of axonemes, Nicastro et al. (2006) proposed that “linkers” between dyneins provide hard-wiring to coordinate motor activities. Because the linkers in axonemes are distinct structures from the linker domain of dynein, for clarity, here we call them “connecters.” According to the recent cryo-ET of proximal region of C. reinhardtii flagella (Bui et al., 2012), there are in fact asymmetries among nine doublets that are localized to the connecters between outer and inner arm dynein, called the outer-inner dynein (OID) connecters (Fig. 5, A and C). Recently we identified that the intermediate chain 2 (IC2) of outer arm dynein is a part of the OID connecters, and a mutation of the N-terminal region of IC2 affects functions of both outer and inner arm dyneins (Oda et al., 2013), which supports the idea that the connecters between dyneins are involved in axonemal dynein regulation.

Closing remarks

Thanks to the crystal structures, we can now design and interpret experiments such as single molecule assays and EM based on the atomic models of dynein. Our understanding of the molecular mechanism and cellular functions of dyneins will be significantly advanced by these experiments in the near future.One important direction of dynein research is to understand the motor mechanisms closer to the in vivo state. For example, the step sizes of cytoplasmic dynein purified from porcine brain is ∼8 nm independent of load (Toba et al., 2006). This result suggests that intermediate and light chain bound to the dynein heavy chain may modulate the motor activity of dynein. To address such questions, Trokter et al. (2012) reconstituted human cytoplasmic dynein complex from recombinant proteins, although the reconstituted dynein did not show robust processive movement. Further studies are required to understand the movement of cytoplasmic dynein. Similarly, axonemal dyneins should also be studied using mutations in a specific gene that does not affect the overall flagella structure, rather than depending on null mutants that cause the loss of large protein complexes.Detailed full chemomechanical cycle of dynein and its regulation are of great importance. Currently, open/closed states of the gap between AAA1 and AAA2 are not clearly correlated with the chemomechanical cycle of dynein. Soaking dynein crystal with nucleotides showed that the presence of ATP alone is not sufficient to close the gap, at least in the crystal (Schmidt et al., 2012). This result suggests that other factors such as a conformational change of the linker are required. For other motors, ATP hydrolysis is an irreversible chemical step, which is often “gated” by strain. In the case of kinesin, ATP is hydrolyzed by a motor domain only when a forward strain is applied by the other motor domain through the neck linker (Cross, 2004; Kikkawa, 2008). A similar strain-based gating mechanism may play important roles in controlling the dynein ATPase. Upon MTBD binding to the forward binding site, a strain between MTBD and tail would be applied to the dynein molecule. The Y-shaped stalk and strut/buttress under the strain would force the head domain to close the gap between AAA4 and AAA5 (Fig. 2 H). Similarly, the linker under the strain would be hooked onto the two finger-like structures and close the gap between AAA1 and AAA2 (Fig. 2 H). The gap closure then triggers ATP hydrolysis by dynein. This strain-based gating of dynein is consistent with the observation that the rate of nonadvancing backward steps, which would depend on ATP hydrolysis, is increased by load applied to dynein (Gennerich et al., 2007). To explain cilia and flagella movement, the geometric clutch hypothesis has been proposed (Lindemann, 2007), which contends that the forces transverse (t-force) to the axonemal axis act on the dynein to regulate dynein activities. In the axoneme, dynein itself can be the sensor of the t-force by the strain-based gating mechanism. Further experiments are required to test this idea, but the strain-based gating could be a shared property of biological motors.     

Pharmacological targeting of guanosine monophosphate synthase suppresses melanoma cell invasion and tumorigenicity

Malignant melanoma possesses one of the highest metastatic potentials among human cancers. Acquisition of invasive phenotypes is a prerequisite for melanoma metastases. Elucidation of the molecular mechanisms underlying melanoma invasion will greatly enhance the design of novel agents for melanoma therapeutic intervention. Here, we report that guanosine monophosphate synthase (GMPS), an enzyme required for the de novo biosynthesis of GMP, has a major role in invasion and tumorigenicity of cells derived from either BRAF^V600E or NRAS^Q61R human metastatic melanomas. Moreover, GMPS levels are increased in metastatic human melanoma specimens compared with primary melanomas arguing that GMPS is an attractive candidate for anti-melanoma therapy. Accordingly, for the first time we demonstrate that angustmycin A, a nucleoside-analog inhibitor of GMPS produced by Streptomyces hygroscopius efficiently suppresses melanoma cell invasion in vitro and tumorigenicity in immunocompromised mice. Our data identify GMPS as a powerful driver of melanoma cell invasion and warrant further investigation of angustmycin A as a novel anti-melanoma agent.Malignant melanoma is one of the most aggressive types of human cancers. Its ability to metastasize in combination with resistance to conventional anticancer chemotherapy makes melanoma extremely difficult to cure, and the median survival of patients with metastatic melanoma is 8.5 months.^{1, 2, 3} A better understanding of the biology behind melanoma aggressiveness is imperative to facilitate the development of novel anti-melanoma strategies.Melanoma and other cancers cells have been shown to strongly rely on de novo nucleotide biosynthesis^{4, 5} and often overexpress several biosynthetic enzymes involved in these pathways.^{6, 7, 8, 9} Recently, we have identified a fundamental connection between melanoma invasion and biosynthesis of guanylates,⁸ suggesting that distortion of the guanylate metabolism facilitates melanoma progression.Guanosine monophosphate reductase (GMPR) reduces GMP to one of its precursors, inosine monophosphate (IMP), and depletes intracellular GTP pools (Figure 1a). We have recently demonstrated that GMPR suppresses melanoma cell invasion and growth of human melanoma cell xenografts. These findings tightly linked guanylate production to the invasive potential of melanoma cells.⁸Open in a separate windowFigure 1GMPS contributes to the invasive capability of melanoma cells. (a) Simplified schematic of the metabolic pathway for guanylates production. (b) SK-Mel-103 and SK-Mel-28 cells were transduced with a control vector or two independent shRNAs to GMPS and tested for invasion through Matrigel (left panel). Where indicated, cells were incubated with 100 μM guanosine for 24 h before the assay and guanosine supplementation was maintained throughout the experimental procedure. The data represent the average ± S.E.M. of at least two independent experiments. GMPS suppression was verified by immunoblotting (right panel). (c) Cells transduced as in (a) were plated on coverslips coated with FITC-conjugated gelatin. After 16 h cells were fixed with 4% PFA and stained for actin (rhodamine-conjugated phallodin) and nuclei (Hoechst). Where indicated, cells were incubated with 100 μM guanosine for 24 h before the assay and guanosine supplementation was maintained throughout the experimental procedure. At least 25 cells/sample were imaged to assess the number of cells with gelatin degradation. The data represent the average ± S.E.M. of two independent experiments. *P<0.05, **P<0.001 compared with control; ^#P<0.05, ^##P<0.001 compared with untreated cells. Statistics performed with Student''s t-Test. See also Supplementary Figure S1Of the several enzymes involved in guanylate biosynthesis, inositol monophosphate dehydrogenases 1 and 2 (IMPDH-1, -2), functional antagonists of GMPR (Figure 1a), have been targeted clinically with several drugs including the most specific one, mycophenolic acid (MPA) and its salt, mycophenolate mofetil (MMF).^{10, 11, 12, 13} Nonetheless, prior studies demonstrated that MPA possesses poor anti-tumor activity,^{14, 15} and it is primarily used as an immunosuppressing agent in organ transplantation.^{10, 11, 12}GMP synthase (GMPS) is the other functional antagonist of GMPR. GMPS catalyzes the amination of xanitol monophosphate (XMP) to GMP to promote GTP synthesis (Figure 1a).^{16, 17} Most of the studies on GMPS have been performed in bacteria, yeast, and insects, where GMPS has been shown to have a key role in sporulation, pathogenicity, and axon guidance, respectively.^{18, 19, 20} Mammalian GMPS has been the subject of several studies addressing its unconventional (GMP-unrelated) roles in the regulation of activity of ubiquitin-specific protease 7 (USP7).^{21, 22, 23, 24} However, because of the newly revealed importance of guanylate metabolism in control of melanoma cell invasion and tumorigenicity,⁸ GMPS emerges as an attractive target for anti-cancer therapy.In the late 1950s, a specific inhibitor of bacterial GMPS, angustmycin A (also known as decoyinine), has been isolated from Streptomyces hygroscopius as a potential antibiotic with sporulation-inducing activity in Bacillus subtilis.^{25, 26, 27, 28, 29} Its anti-tumor activity has never been experimentally explored. In the current study, we investigated the role of GMPS in regulation of melanoma invasion and tumorigenicity, and explored the possibility of targeting GMPS by angustmycin A as a novel anti-melanoma strategy.     

Conformations of voltage-sensing domain III differentially define NaV channel closed- and open-state inactivation

Voltage-gated Na⁺ (Na_V) channels underlie the initiation and propagation of action potentials (APs). Rapid inactivation after Na_V channel opening, known as open-state inactivation, plays a critical role in limiting the AP duration. However, Na_V channel inactivation can also occur before opening, namely closed-state inactivation, to tune the cellular excitability. The voltage-sensing domain (VSD) within repeat IV (VSD-IV) of the pseudotetrameric Na_V channel α-subunit is known to be a critical regulator of Na_V channel inactivation. Yet, the two processes of open- and closed-state inactivation predominate at different voltage ranges and feature distinct kinetics. How inactivation occurs over these different ranges to give rise to the complexity of Na_V channel dynamics is unclear. Past functional studies and recent cryo-electron microscopy structures, however, reveal significant inactivation regulation from other Na_V channel components. In this Hypothesis paper, we propose that the VSD of Na_V repeat III (VSD-III), together with VSD-IV, orchestrates the inactivation-state occupancy of Na_V channels by modulating the affinity of the intracellular binding site of the IFMT motif on the III-IV linker. We review and outline substantial evidence that VSD-III activates in two distinct steps, with the intermediate and fully activated conformation regulating closed- and open-state inactivation state occupancy by altering the formation and affinity of the IFMT crevice. A role of VSD-III in determining inactivation-state occupancy and recovery from inactivation suggests a regulatory mechanism for the state-dependent block by small-molecule anti-arrhythmic and anesthetic therapies.

IntroductionVoltage-gated Na⁺ (Na_V) channels initiate excitation in neurons and myocytes, enabling rapid conduction over large distances that is decoupled from intracellular Ca²⁺ signaling (Hille, 2001). During excitation, Na_V channel opening releases a large inward Na⁺ current (I_Na) that is followed by rapid inactivation, which occurs within milliseconds and makes most channels nonconductive. This inactivation from the open state is required to allow outward repolarizing currents to bring the cell membrane back to the resting potential. However, inactivation recovery is not instantaneous at hyperpolarized membrane potentials, nor is it limited to depolarized potentials. For example, in cardiac myocytes, Na_V channels remain inactivated for 10s to 100s of milliseconds following action potential repolarization, rendering the myocyte refractory to excitation for a brief period and preventing reentrant arrhythmia (Zipes et al., 2017). In addition, neuronal memory of previous excitation can be conferred by inactivation of Na_V channels that reduces the subsequent firing rate (Marom, 1998; Toib et al., 1998). At modest depolarized membrane potentials (less than −30 mV), Na_V channels can also become inactivated before the activation gate opening but require longer depolarizations (Aldrich et al., 1983; Bean, 1981). This closed-state inactivation results in fewer available channels to initiate excitation. Thus, the regulation of Na_V channel inactivation by time and membrane potential is essential for the functioning of excitable cells, allowing neurons and myocytes to appropriately respond to changes in membrane potential that span multiple time domains (Silva, 2014).Mammalian Na_V channels are formed by a single protein with four homologous repeats (I–IV; Fig. 1 A), each constituting six membrane-spanning segments (S1–S6). Within each repeat contains a voltage-sensing domain (VSD) that consists of segments S1–S4. The Na⁺-selective pore is formed jointly by the S5 and S6 segments (Yu and Catterall, 2003), with the region between S5 and S6 constituting the P-loop responsible for Na⁺ ion selectivity. Mammalian Na_V channels additionally feature intracellular linkers and a unique C-terminal domain (CTD), which is known to bind various accessory subunits, including calmodulin and fibroblast growth factor homologous factors (Abriel, 2010). Given the complex time and voltage dependence of Na_V channel inactivation, it must be connected to multiple different conformational states, whose occupancy is determined by the positions of the VSDs, the state of the channel pore, and the position of the CTD (Ulbricht, 2005).Open in a separate windowFigure 1.The Na_V channel inactivation mechanism.(A) A schematic representation of mammalian Na_V channel shows four homologous repeats (I–IV) and the IFMT motif on the III–IV linker (yellow). (B) The structure of rNa_V1.5 (PDB accession no. 6UZ3) shows the IFMT motif enclosed by repeat III S5 and repeat IV S4–S5 linker and S5 and S6 segments, causing an allosteric block during inactivation.Indeed, several reports have connected the activation of the VSD-IV to the onset of fast inactivation after channel opening (Capes et al., 2013; Goldschen-Ohm et al., 2013). Mutations in VSD-IV also suggest a role in closed-state inactivation (Kambouris et al., 2000; Chahine et al., 1994; Groome et al., 2011). Charge neutralization mutations within VSD-IV cause a large hyperpolarizing shift in voltage-dependent channel availability and a large fraction of inactivated channels at voltages where the channels are closed (Capes et al., 2013; Brake et al., 2021 Preprint). Early experiments demonstrated that the addition of intracellular pronase removed Na_V channel inactivation and revealed the participation of intracellular components (Armstrong et al., 1973; Armstrong, 1981; Salgado et al., 1985). The model of Na_V channel inactivation was then suggested to resemble the “ball-and-chain” model of K⁺ channel N-type inactivation, where a ball that is attached to the inner part of the channel causes inactivation by occluding the pore (Armstrong and Bezanilla, 1977; Hoshi et al., 1990). Subsequent experiments that disrupted the intracellular linker between repeats III and IV produced functional channels but lacked inactivation (Stühmer et al., 1989; Vassilev et al., 1988; Vassilev et al., 1989), further confining the sites of the inactivation gate to the III-IV linker. Finally, site-specific mutagenesis identified a hydrophobic cluster of amino acids, Ile-Phe-Met-Thr (IFMT), as an essential component for the inactivation mechanism (Hartmann et al., 1994), and a “hinged-lid” model was proposed (Eaholtz et al., 1994; Kellenberger et al., 1997; Rohl et al., 1999; West et al., 1992). In this model, the loop between two hinged points serves as a rigid lid that folds over the channel pore, with the IFMT motif acting as a hydrophobic latch to stabilize the inactivated conformation. Additional mutagenesis and the discovery of inherited proarrhythmic mutations within cardiac Na_V channels further suggested that the binding site or the “receptor” for the IFMT motif involves repeats III and IV S4–S5 linkers and IV S6 segment (Smith and Goldin, 1997; McPhee et al., 1995; McPhee et al., 1998). Recent structures of eukaryotic Na_V channels, however, reveal that the IFMT motif resembles more of a wedge that squeezes into the crevice formed by repeat III S5 and repeat IV S4–S5 linker and S5 and S6 segments (Yan et al., 2017; Pan et al., 2018, 2019; Shen et al., 2019; Jiang et al., 2020; Li et al., 2021) and allosterically blocks the Na_V channel conduction pore (Fig. 1 B).Apart from the inactivation gate, the VSD-III, the CTD, and the state of the channel pore have also been implicated in Na_V channel inactivation (Cha et al., 1999; Deschênes et al., 2001; Hsu et al., 2017; Mantegazza et al., 2001; Motoike et al., 2004; Pitt and Lee, 2016, Mangold et al., 2017). Based on new structural data in combination with numerous previous functional studies, we hypothesize that the activation of the VSD-III, in addition to the VSD-IV, facilitates different inactivated states by modulating the binding affinity of IFMT crevice. In this Hypothesis paper, we will first outline the models of each inactivated state. Then, we will present the structural and electrophysiological evidence that supports our hypothesis and leads to the proposed model. Finally, we will discuss the model implications on the modulation of Na_V channel function by accessory subunits and therapeutic drugs.The models of open- and closed-state Na_V channel inactivationWe propose a VSD-III and VSD-IV-centric inactivation model (created at https://biorender.com; Fig. 2), which incorporates structural motifs throughout the channel. In this model, the VSD-III can activate in two distinct conformations over different voltage ranges as detected by fluorescent tracking of voltage sensor conformational change upon membrane depolarization (Chanda and Bezanilla, 2002; Zhu et al., 2017) and various supportive electrophysiological evidence (Varga et al., 2015; Hsu et al., 2017). The intermediate and fully activated conformations of VSD-III then unmask distinct crevices with low and high affinity for the IFMT motif, as suggested by eukaryotic Na_V channel structures (Yan et al., 2017; Jiang et al., 2020), that leads to unique kinetics of closed- versus open-state inactivation. Different inactivated states are therefore connected to distinct activated conformations of VSD-III.Open in a separate windowFigure 2.The models of open- and closed-state inactivation based on two VSD-III depolarized conformations.(A) During open-state inactivation, strong membrane depolarization leads to the activation of repeat I and II VSDs and the intermediate activation of VSD-III, resulting in the opening of the activation gate and the conduction of Na⁺ ions (I). Activation of the VSD-IV exposes the low-affinity binding site for the IFMT motif (II). Further translation of the VSD-III into its fully activated conformation establishes the stable inactivated configuration with high-affinity IFMT motif binding (III). (B) For closed-state inactivation at hyperpolarized membrane potential, VSD-III activates while repeats I and II VSDs are at rest, resulting in closed activation gate and no I_Na (I). Subsequent VSD-IV activation forms the low-affinity binding site for the IFMT motif (II). Since VSD-III occupies primarily in its intermediate conformation, the stable inactivated configuration is not established.During the process of open-state inactivation (Fig. 2 A), the VSDs of repeats I and II rapidly move outward into their activated position upon membrane depolarization. At the same time, VSD-III adopts its intermediate conformation, and the activation gate opens (step I), allowing the channel to pass current. Shortly thereafter, VSD-IV activates and a partial crevice for the IFMT motif is formed. The III–IV linker may also be released from CTD interaction upon VSD-IV activation, as suggested by the structure of the α-scorpion toxin AaH2-bound hNa_V1.7 VSD-IV–Na_VPaS hybrid, where the intracellular end of resting IV S4 (K7 and R8) interacts with the CTD, causing the sequestration of the III–IV linker (Clairfeuille et al., 2019). The IFMT motif can move into the hydrophobic cleft formed by repeat IV S4–S5 linker and S6 segment and establish low-affinity inactivation binding (step II). Concurrently, further translation of VSD-III into its fully activated conformation repositions the III S4–S5 linker to increasingly interact with the helix along III–IV linker, providing additional support for the stable inactivated conformation (step III). The binding of the IFMT motif leads to a rearrangement around S6 helices and an allosteric block of the conduction pathway. During fast inactivation, the VSDs of repeats I and II are free to deactivate upon membrane hyperpolarization. VSD-III and VSD-IV, however, are immobilized by the IFMT motif. Importantly, the deactivation of VSD-III and VSD-IV determines inactivation recovery time course, with the slowness of VSD-III deactivation playing a rate-limiting role (Hsu et al., 2017).For closed-state inactivation, at membrane potentials less than −30 mV (Fig. 2 B), VSD-III activates to its intermediate conformation while VSD-I and VSD-II are at rest (step I). When VSD-IV moves up to its activated conformation, the crevice for IFMT motif is partially exposed and low-affinity IFMT binding is established (step II). Because of the hyperpolarized potential, VSD-III primarily occupies only its intermediate conformation, incapable of providing supportive interactions around the inactivation gate, and thus the channel is unable to adopt a stable inactivated configuration. The inactivated state with low-affinity IFMT motif binding results in the slow kinetics of closed-state inactivation. As the VSD-IV activates at more positive potentials than the VSD-III (Varga et al., 2015), the rate limiting of closed-state inactivation is reflected by the overlap in voltage range at which VSD-IV activates and the steady-state inactivation occurs (Capes et al., 2013).In the following section, we will review supporting evidence that underlies the key aspects and implications of our proposed hypothesis.Differential Na_V channel–inactivated states are determined by the IFMT binding affinity that is modulated by the conformations of repeats III and IV VSDsThe closed and open Na_V channel inactivation states differ not only in the voltage range over which they are occupied but also prominently in their kinetics, where inactivation proceeds at a much faster rate from the open-channel conformation (Aldrich et al., 1983; Goldman, 1995). As shown by structural and functional studies, inactivation is due to the binding of the IFMT motif to a crevice (Fig. 1 B). If the final step in both closed- and open-state inactivation is identical, then how can these two inactivation processes feature distinct kinetics? One possible explanation is that the conformation of IFMT binding site is dynamic and its binding affinity can be tuned. The crevice is formed by the S4–S5 linker of repeat IV, S5 of repeat III, and the S5 and S6 segments of repeat IV with additional interactions from repeat III S4–S5 linker aiding in the stabilization of the IFMT motif binding, as indicated by many recent structures (Yan et al., 2017; Pan et al., 2018; Shen et al., 2019; Pan et al., 2019; Jiang et al., 2020) and past mutagenesis studies (Kellenberger et al., 1997; McPhee et al., 1995, 1998; Smith and Goldin, 1997).In the resting state model of rat Na_V1.5 (rNa_V1.5), fitted after the resting bacterial Na_VAb structure (Wisedchaisri et al., 2019), the IFMT crevice is blocked by the intracellular end of repeat III S4 and the S4–S5 linkers of repeats III and IV (Jiang et al., 2020). The outward translation of repeat III and IV VSDs is therefore a prerequisite for rendering the inactivation crevice available. The movement of III– or IV–S4 segments will pull on their corresponding S4–S5 linkers and position them accordingly. In other words, the extent of VSD activation (i.e., the number of gating charges that moves from an internal to an external side) directly affects the placement of the S4–S5 linker, which in repeats III and IV form the integral component of the IFMT crevice. Thus, different combinations of the VSD-III and VSD-IV depolarized conformations may give rise to different binding affinities of the IFMT crevice, resulting in the distinct inactivation kinetics observed over different membrane potentials. We expect that there are also differences in the contribution from VSD-III and VSD-IV in regulating the IFMT crevice between different Na_V channel homologues (Chanda and Bezanilla, 2002; Varga et al., 2015; Brake et al., 2021 Preprint).Facilitation of inactivation by the depolarized VSD-IV conformation has been convincingly shown via multiple charge neutralization and toxin studies (Capes et al., 2013; Chen et al., 1996; Kontis et al., 1997; Kühn and Greeff, 1999; Clairfeuille et al., 2019). Further studies have revealed that VSD-III may also play a key role in modulating IFMT binding. During fast inactivation, a fraction of the gating charge was found to be immobilized by the inactivation gate (Armstrong and Bezanilla, 1975; Bezanilla and Armstrong, 1974; Bezanilla et al., 1982). These gating charges were identified as components of the repeat III and IV VSDs, suggesting their interaction with the III–IV linker (Armstrong and Bezanilla, 1977; Cha et al., 1999). Specific disruption of the inactivation particle through an IFM/ICM mutation did not eliminate the immobilization of VSD-IV, yet entirely abolished VSD-III immobilization (Sheets and Hanck, 2005), emphasizing a functional linkage between the IFMT motif and the VSD-III depolarized conformation.The role of VSD-III in regulating Na_V channel closed- and open-state inactivation was additionally highlighted through various protocols. Particularly, biasing VSD-III to the depolarized conformation by tethering extracellular MTSEA-biotin on repeat III S4 through an R3C mutation dramatically reduced Na_V channel availability via enhanced closed-state inactivation and its slope factor but showed minimal effects on the peak I–V relationship (Sheets and Hanck, 2007). The R1128C (R4C on repeat III S4 segment) mutation in rat skeletal muscle sodium channel (rNa_V1.4) promoted closed-state inactivation, leading to reduced steady-state availability (Groome et al., 2011, 2014a). Similar mutations (R4H on repeat III S4 segment) in human skeletal muscle Na_V channel (hNa_V1.4 R1135H) and human cardiac sodium channel (hNa_V1.5 R1309H) showed enhanced entry into inactivated states with prolonged recovery from inactivation (Groome et al., 2014b; Wang et al., 2016). When the three outermost gating charges of VSD-III were neutralized, the fraction of inactivated channels at subthreshold potentials (−60 and −50 mV), increased and the delay to the onset of open-state inactivation after a conditioning pulse decreased, but to a lesser extent than the neutralization of VSD-IV (Capes et al., 2013). The charge neutralization in VSD-III of Na_V1.5 caused the similar significant shift in steady-state inactivation curve toward hyperpolarizing potentials, which is an indicative of increased closed-state inactivation, as observed when the VSD-IV charges were neutralized (Brake et al., 2021 Preprint). All these results illustrate that the modulation of the repeat III S4 either through a chemical manipulation or a reduction in the number of VSD-III gating charges that facilitate VSD-III activation enhance closed-state inactivation and reduce channel availability, emphasizing the significant role of VSD-III activated conformations in regulating the states of Na_V channel inactivation.Interestingly, open-state and closed-state inactivation feature distinct kinetics: open-state inactivation kinetics show little change in rate over the voltages at which it occurs (Aldrich et al., 1983), whereas the time constant of closed-state inactivation is highly voltage dependent (Goldman, 1995; Sheets and Hanck, 1995). An inactivation gate peptide (KIFMK) that mimics the IFMT motif and restores the fast inactivation in defective mutant Na_V channels illustrates weak voltage dependence in terms of inactivation time constants at depolarized potential of more than −30 mV (Eaholtz et al., 1994; Peter et al., 1999). This observation, together with the absence of gating current component with a time course that tracks inactivation, suggests that inactivation after channel opening has no intrinsic voltage dependence (Armstrong, 2006). We hypothesize that the voltage dependence of closed-state inactivation arises from movements of the VSDs of repeats III and IV exposing the crevice for IFMT binding, while during open-state inactivation, rapid and complete VSD-III and VSD-IV activation fully reveals the crevice and inactivation kinetics are determined by voltage-independent III-IV linker diffusion. Previously, closed-state inactivation was postulated to occur when the repeats III and IV VSDs move to depolarized conformations, while the VSDs of repeats I and II remain at rest (Armstrong, 2006). We further propose that the position of the VSD-III is crucial in determining the conformation of the IFMT binding site and that inactivation state occupancy is derived from the multiple steps traversed by the VSD-III as it activates. In the next section, we will examine studies that imply different VSD-III activated conformations.Unique VSD-III activated conformation determines the differential binding of the IFMT motifRecent structures of eukaryotic Na_V channels featured components of Na_V channel inactivation machinery that were missing from prior structures of bacterial Na_V channels. The first eukaryotic Na_V channel structure from the American cockroach, Na_VPaS, captured closed pore domain with all four VSDs in an “activated” conformation (Shen et al., 2017), defined by two positively charges within the S4 segment having crossed the hydrophobic constriction site (HCS), a cluster of hydrophobic residues that prevents ion leakage in voltage sensor (Yang et al., 1996; Starace and Bezanilla, 2004; Tao et al., 2010; Catterall, 2014). The III–IV linker, although similar in length to other eukaryotic Na_V channels, lacks the key hydrophobic motif required for inactivation and is shown to interact with the CTD. Subsequently, the electric eel Na_V1.4 (eeNa_V1.4), human Na_V1.2 (hNa_V1.2), hNa_V1.4, hNa_V1.7, and rat Na_V1.5 (rNa_V1.5) structures were resolved. These structures are all highly similar, with high resolution of key elements that are critical for inactivation mechanism. In these structures, the III–IV linker contains the inactivation motif (LFM for eeNa_V1.4) binding to the crevice formed by repeat III S5, repeat IV S4–S5 linker, and S5 and S6 segments (Fig. 1 B), likely representing the inactivated state structure (Yan et al., 2017; Pan et al., 2018; Shen et al., 2019; Pan et al., 2019; Jiang et al., 2020). All four VSDs are also captured in the activated positions, albeit some with different conformations in the Na_VPaS structure. The CTD is not well resolved and is thus omitted.Comparison between Na_VPaS and other eukaryotic Na_V channels reveals VSD-III can activate in two steps. First, the positions of the VSD-III differ in the extent of their depolarized conformation or activation. Even though Na_VPaS channel is nonfunctional and may not represent a relevant physiological state, Na_VPaS contains highly conserved transmembrane segments that are similar to and superimpose well with other eukaryotic Na_V channels. The number of arginines on repeat III S4 segment is also comparable among two groups. Differences observed may provide an opportunity for dissecting Na_V channel mechanism. In Na_VPaS, two gating charges on repeat III S4 are transferred across the HCS, while in other structures, a total of four charges on III S4 segment are displaced from the intracellular to the extracellular side (Fig. 3 A). These structural variations reveal that the VSD-III might adopt two distinct activated conformations correlating to different numbers of charges across the HCS. The trajectory of the VSD-III activation resembles a sigmoidal curve, with an intracellular tip moving laterally away from the pore and the helix rotating upward and bending downward at the extracellular end (Yan et al., 2017).Open in a separate windowFigure 3.A comparison between Na_VPaS and rNa_V1.5 structures. (A) Structures of eukaryotic Na_V channel reveal two distinct conformations of the depolarized VSD-III, varying in the number of gating charges (K and R) across the HCS. In Na_VPaS (left, light blue; PDB accession no. 5X0M), the activated VSD-III transfers two positively charged residues from an internal to an external side. The other VSD-III activated conformation from rNa_V1.5 (right, cyan; PDB accession no. 6UZ3) captures the total of four gating charges transfer. (B) An overlay of two Na_V channel structures shows that the VSD-III fully activated conformation (cyan) is needed to facilitate the binding of the IFMT motif.Two-step VSD-III activation is further supported by functional experiments that tracked Na_V channel VSD movement optically via the voltage-clamp fluorometry (VCF) technique (Zhu et al., 2016; Rudokas et al., 2014; Chanda and Bezanilla, 2002; Mannuzzu et al., 1996). The VCF protocol was developed to observe the conformational dynamics of individual S4 segments and can reveal structurally correlated kinetic information (Cowgill and Chanda, 2019). By attaching a fluorophore to the extracellular S3–S4 linker site through a cysteine mutation, the movement of the proximal S4 segment can be detected upon the changes in fluorescence emission caused by differences in the surrounding environment (Cha et al., 1999b; Glauner et al., 1999; Bezanilla, 2000; Chanda and Bezanilla, 2002). Simultaneous recordings of fluorescence emission and ionic current identified the concurrent activation of repeat I–III VSDs during Na_V channel activation, with a slight delay of VSD-IV activation lagging the rise of the I_Na (Fig. 4 A). The VSD-III is most sensitive to hyperpolarized potentials, having the half-maximal voltage of the fluorescence–voltage curve at the most negative potential among all VSDs (Fig. 4 B; Varga et al., 2015). The VSD-III is already activated at highly negative potentials where closed-state inactivation occurs.Open in a separate windowFigure 4.Experimental data supporting a two-step VSD-III activation model. (A) VCF recordings of rNa_V1.4 show simultaneous initiation of repeats I–III VSD activation (blue, red, and green) but a delay in VSD-IV activation (black), which is reduced with increasing voltages. Adapted from Chanda and Bezanilla (2002). (B) Fluorescence–voltage (FV) curves of hNa_V1.5 VCF constructs show that VSD-III activates at hyperpolarized potential, earlier than other repeat VSDs. Adapted from Varga et al. (2015). LFS, large fluorescence signal. (C) VCF recording of S1113C in rNa_V1.4 elicits two stages of VSD-III movement, with an initial increase followed by a decrease in fluorescence emission. Adapted from Chanda and Bezanilla (2002). (D) Coexpression of hNa_V1.5 and β3-subunit at high ratio (1:4 and 1:6) yields two distinct components of VSD-III activation kinetics, as detected by fluorescence emission (inset) and fluorescence–voltage curves. Adapted from Zhu et al. (2017). (E) A correlation between VSD-III deactivation time constants and the depolarization duration in WT hNa_V1.5 suggests the multiple steps of VSD-III activation, which is dependent on the IFMT motif as illustrated by the loss of correlation in IQM mutation. Adapted from Hsu et al. (2017). Error bars represent the standard error of the mean from the sample size of 3–6 measurements.Two stages of VSD-III activation were demonstrated in the VCF recordings of rNa_V1.4 VSD-III with a fluorophore attached to S1113C, showing an initial increase in the fluorescence emission followed by a reduction (Fig. 4 C; Chanda and Bezanilla, 2002). Similarly, in another independent recording of hNa_V1.5 in the presence of high β3-subunit expressions, the fluorescence emission tracking VSD-III activation at M1296C showed two components with distinct kinetics (Fig. 4 D; Zhu et al., 2017). The curves also displayed two elements of VSD-III activation, implying that VSD-III activates with different kinetics over different voltage ranges. Recently, the VCF recordings of neonatal form Na_V1.5 (Na_V1.5e) reported the biphasic relationship of VSD-III fluorescence–voltage curve reaching its maximum at −50 mV and declining with more depolarized potentials (Brake et al., 2021 Preprint). Finally, the time constant of VSD-III deactivation, unlike other VSDs, varied with depolarization duration, hinting at the multiple-step VSD-III activation mechanism (Fig. 4 E; Hsu et al., 2017). If the VSD activation progressed to only a single conformation, then prolonged depolarization would not affect the time it takes to return to its resting position. Instead, when the S4 segment travels across two distinct states, longer depolarizing pulses bias the transition to the second step and result in longer deactivation times. Altogether, extensive structural and functional evidence strongly supports two-step VSD-III activation.The two conformations of VSD-III activation lead to an interesting consequence that each activation step may dictate a different conformation of the IFMT crevice. Because of the highly similar structures in the transmembrane core and the same length of III-IV linker across eukaryotic Na_V channels, the relative positions of repeat III S4 and its ensuing S4–S5 linker can provide relevant, useful insights. In the Na_VPaS structure, the position of the III S4 segment orients the S4–S5 linker so that it interferes with IFMT binding and likely obstructs the formation of the IFMT crevice (Fig. 3 B). The short section on the III–IV linker before the IFMT motif is also restricted from moving near the membrane, further constraining the range over which the IFMT motif can travel. In contrast, further translation of the III S4 in other eukaryotic structures pulls the S4–S5 linker up and away from the pore such that a crevice is exposed and direct interactions with the IFMT motif can be established (Fig. 3 B; Jiang et al., 2020). Consecutive helices on the III–IV linker provide additional interactions with the S4–S5 linkers of repeats III and IV that further stabilize inactivated state. Specifically, charged residues near the IFMT motif allosterically interact with repeats III and IV VSDs (Groome et al., 2003; Groome et al., 2007).Electrostatic interactions from the III–IV linker cause the immobilization of repeats III and IV VSDs during prolonged depolarization (Armstrong and Bezanilla, 1977; Kuo and Bean, 1994; Cha et al., 1999). The recovery from inactivation is thus affected by the kinetics of VSD-III and VSD-IV deactivation. When the IFMT motif was mutated to IQMT, fast inactivation was greatly impaired, and the deactivation of repeats III and IV VSDs was significantly accelerated (Hsu et al., 2017). The time constant of VSD-III deactivation that normally increases with depolarizing pulse duration no longer shows such correlation in the IQMT mutant channel (Fig. 4 E), suggesting that IFMT binding specifically stabilizes the second VSD-III activated state (Hsu et al., 2017). The same study also showed that inactivation recovery after >200 ms depolarization is defined exclusively by the deactivation of VSD-III (Hsu et al., 2017). This result coincides with the second step VSD-III activation, transferring the highest number of gating charges across the HCS and hence likely returning last to the resting position. Other arrhythmogenic mutations along the IFMT binding sites on S4–S5 linkers of III and IV (N1325S, A1330, and N1659A) also showed enhanced recovery from inactivation, accompanying the decreased time constant of VSD-III deactivation (Hsu et al., 2017). Together, the differential states of VSD-III activation are likely to provide the basis for distinct Na_V channel inactivation kinetics at hyperpolarized and depolarized potentials.Physiological and pharmacological modulation of repeat III VSD activation affects Na_V channel inactivationMultiple factors involved in the regulation of Na_V channel inactivation, such as the β-subunits, drugs and toxins, were shown to modulate VSD-III activation. Resolved structures of hNa_V1.2, hNa_V1.4, and hNa_V1.7 with coexpressed β1-subunit identified its docking site near the VSD-III with interactions between the Ig domain of β1-subunit and repeat IV L6 (extracellular linker between P-loop and S6) and repeat I L5 (extracellular linker between S5 and P-loop) loops (Pan et al., 2018; Shen et al., 2019; Pan et al., 2019). Such interaction sites imply β-subunit modulation of VSDs III and IV kinetics. Careful study of the mechanism of noncovalent β-subunits (β1 and β3) demonstrated a unique impact on the initiation of repeat III and IV VSDs activation, which results in distinct modulatory effects on the hNa_V1.5 (Zhu et al., 2017). Consistent with the resolved structures, coexpression of β3-subunit shifts VSD-III and VSD-IV activation, leading to altered steady-state inactivation, activation, and fast inactivation kinetics. Switching the charge of a β3-subunit transmembrane glutamic acid (E176K) altered VSD-III activation of hNa_V1.5 and affected the channel recovery from inactivation (Salvage et al., 2019). To the contrary, the β1-subunit regulates Na_V channel inactivation solely through the modulation of the VSD-IV, resulting in a shift in steady-state inactivation, while having no impact on I_Na activation or fast inactivation kinetics (Zhu et al., 2017). This unique interaction between cardiac Na_V channel and β1-subunit is further supported by rNa_V1.5 and hNa_V1.5 structures that failed to resolve the coexpressed β1-subunit (Jiang et al., 2020; Li et al., 2021). The different regulating mechanisms between β1 and β3 subunits on hNa_V1.5, therefore, emphasize the significance of the VSD-III conformation in determining the state and kinetics of Na_V channel inactivation.The action of the local anesthetic (LA) lidocaine has also been intimately linked to VSD-III function. LAs are normally used for local and regional anesthesia and the treatment of excitatory pathologies, including epilepsy and cardiac arrhythmia. Their significant therapeutic effect is achieved through use-dependent block, an increase in channel blocking over repetitive activation. This effect is due to the higher-affinity binding when the channel is in the open and inactivated states (Bean, 1981), resulting in a cumulative increase of inactivated channels over successive pulses of depolarization. The modulated receptor hypothesis postulates that the accessibility to the high-affinity binding site depends on the state of the channel, which in turn involves voltage-dependent VSD activation (Hille, 2001). Lidocaine stabilizes VSD-III in its activated state in rNa_V1.4 channel coexpressed with β1-subunit (Muroi and Chanda, 2009). The prepositioning of both repeats III and IV VSDs in the outward position (through MTSEA-biotin modified R3C III and R2C IV) enhances lidocaine block significantly in hNa_V1.5 α-subunit alone, and when the inactivation gate is disrupted via IFM/ICM mutation, the stabilization of individual III or IV S4 increases lidocaine affinity similar to when both VSDs are stabilized, suggesting an equally significant contribution of VSD-IV as VSD-III to lidocaine block (Sheets and Hanck, 2003; Sheets and Hanck, 2007). The discrepancy that is observed between two studies by Muroi et al. and Sheets et al. could be attributed to the intrinsic properties of different Na_V channel isoforms as well as the presence of β1-subunit. We found that lidocaine induces hyperpolarizing shifts in VSD-III and VSD-IV activations, but β1-subunit coexpression subdues the VSD-IV modulation (Zhu et al., 2021). The inactivation gate, though not necessary for drug binding, drastically reduces drug binding affinity when it is removed (Bennett et al., 1995; Sheets and Hanck, 2007). How the inactivation gate contributes to the kinetics of lidocaine block is still not fully understood (Fozzard et al., 2011).It is, nevertheless, interesting to note the parallel molecular determinants between lidocaine and the IFMT motif binding affinity. Both are dependent upon the positions of VSD-III and VSD-IV. Even though lidocaine is found to bind in the inner pore of Na_V channel through interaction with phenylalanine and tyrosine on the repeat IV S6 segment (Ragsdale et al., 1994; Yarov-Yarovoy et al., 2002; F1759 and Y1766 in Na_V1.5) and directly blocks Na⁺ conductance, binding of the IFMT motif on the rim of the activation gate might also shape the local interaction around the lidocaine-binding site. For one, the IFMT binding stabilizes the outward positions of VSD-III and VSD-IV, which are integral to the drug receptor formation. Thus, the VSD translocation of repeats III and IV could have twofold implications, directly through the VSD-pore coupling and indirectly through the binding of the IFMT motif. According to our hypothesis, the preactivation of VSD-III by lidocaine would favor the high-affinity binding of the IFMT motif and facilitate the inactivation, which in turn enables the high drug affinity–binding mode. A chemical accessibility study of the inactivation gate position during lidocaine block in rNa_V1.4 F1304C (IFMT motif) by MTSET modification showed that lidocaine favors the inactivation gate closure, hypothetically by promoting the transition along the activation pathway (Vedantham and Cannon, 1999).Finally, a study of a lidocaine derivative, mexiletine, also shows that the level of VSD-III activation strongly affects drug efficacy (Zhu et al., 2019; Moreno et al., 2019). Mexiletine preferentially blocks late I_Na, which is enhanced in many cardiac pathologies, including long-QT type 3 syndrome. Investigation into varied mexiletine sensitivity of different long-QT type 3 mutations identified a correlation to the fraction of activated VSD-III. These results resonate with the findings from the lidocaine studies on the significance of VSD-III position on a determination of LA like drug affinity. Taken together, these studies of Na_V channel accessory subunits and its blockers emphasize the essential role of VSD-III activation in the regulation of Na_V channel inactivation kinetics.Future perspectiveThere is often a singular focus on individual channel domains in the regulation of the Na_V channel inactivation, consistent with the m³h gating model of Hodgkin and Huxley, where a single “h” gate mediates inactivation. However, a myriad of studies has proved the process to be more elaborate and complicated than the simplified representation of this model, and intricate inactivation is not unique to Na_V channels. Homotetrameric potassium channels and prokaryotic Na_V channels display complexity in the coupling of inactivation to channel activation (Hoshi et al., 1990; Yang et al., 2018). Toxin and pharmacological studies suggest the possibility of a drug-induced alteration in Na_V channel conducting conformation (Baumgarten et al., 1991) or a unique drug-bound inactivated state (Finol-Urdaneta et al., 2019a; Finol-Urdaneta et al., 2019b; Ong et al., 2000). Distinct channel isoforms also tend to respond to drug and toxin differently, such as the differential sensitivity to tetrodotoxin or a unique external lidocaine-binding site in the cardiac Na_V channel (Baumgarten et al., 1991). More importantly, variabilities in the voltage-dependent activation and inactivation curves among different Na_V isoforms suggest their distinctive propensity to closed-state inactivation that likely stems from the unique sequence of VSDs activation (Brake et al., 2021 Preprint) determined by their rates of activation and the midpoint of activation curve (Capes et al., 2013). Because of the heterogeneity and complexity of the channel kinetics, combined results from multiple methods should be interpreted with careful consideration.Through multiple Na_V channel structures and electrophysiological studies combined with the fluorescent labeling technique, we argue here that VSD-III, together with VSD-IV, governs the binding affinity for the inactivation gate and thus affects the different inactivated states. Still missing, however, is the discrete structure of state-dependent conformations such as those of the resting-state and closed-state inactivation, where some VSDs are not activated. One possible means to overcome this obstacle is the use of toxin, small-molecule modulators, or biochemical methods to force certain states of the channel (Clairfeuille et al., 2019; Wisedchaisri et al., 2019; Moreno et al., 2019). Linking channel structure to channel functional states is also challenging but is achievable via dynamic tools such as FRET. FRET is a distance-dependent energy transfer between two fluorophores, namely the excited donor and the emitting acceptor (Sekar and Periasamy, 2003), that can be attached to any protein or peptide. This method allows for the measurement of molecular proximity between two different parts within angstrom distances and thus is often used to test molecular interactions. When combined with patch clamp, FRET measurement could be used to determine the interactions between the channel’s different domains at various functional states. Integrating such knowledge to the available structures would provide the means to derive a state-dependent structure and a dynamic change in its conformation. Recent work by Kubota et al. (2017) used a lanthanide-based resonance energy transfer, which allows distance measurement, to map the voltage sensor positions of rNa_V1.4 during the resting and inactivated states. To gain better insight into Na_V channel inactivation, the ability to label the intracellular linkers or region is critical. Conventional fluorescent proteins and genetically encoded tags produce large stable fluorescent signals but are disruptive to the protein structure due to their bulkiness (Toseland, 2013). To overcome such limitations, genetic code expansion through site-specific mutation of fluorescent unnatural amino acids such as ANAP (Shandell et al., 2019; Kalstrup and Blunck, 2017) could serve as potential alternatives. Using these techniques to track activation gating and intracellular inactivation particle binding over different membrane potentials would allow differentiation of the states of inactivation, namely closed and open, and ultimately dissection of the regulatory components required for the physiological function of Na_V channels.     

Mechanisms of spindle positioning

Accurate positioning of spindles is essential for asymmetric mitotic and meiotic cell divisions that are crucial for animal development and oocyte maturation, respectively. The predominant model for spindle positioning, termed “cortical pulling,” involves attachment of the microtubule-based motor cytoplasmic dynein to the cortex, where it exerts a pulling force on microtubules that extend from the spindle poles to the cell cortex, thereby displacing the spindle. Recent studies have addressed important details of the cortical pulling mechanism and have revealed alternative mechanisms that may be used when microtubules do not extend from the spindle to the cortex.Mitotic and meiotic spindles are precisely positioned within eukaryotic cells for several reasons. In animal cells, spindle position determines the location of contractile ring assembly (Green et al., 2012). Thus, placing a spindle in the center of the cell will result in daughter cells of equal size, whereas positioning the spindle asymmetrically results in daughter cells of different sizes. In oocytes, the extreme asymmetrical positioning of the meiotic spindle allows expulsion of three fourths of the chromosomes into two tiny polar bodies while preserving most of the cytoplasm in the egg for the developing zygote. In polarized cells, where proteins and RNAs are asymmetrically distributed before division, the orientation of the spindle relative to the polarity axis determines whether the daughter cells will have the same or different developmental fates. An excellent review of the developmental context of spindle positioning is provided by Morin and Bellaïche (2011). In budding yeast (Slaughter et al., 2009) and plants (Rasmussen et al., 2011), the site of cytokinesis is determined before the spindle forms. In these organisms, the spindle must be oriented relative to the predetermined division plane to ensure that both daughter cells receive a complete chromosome complement.The majority of research on the mechanisms of spindle positioning has focused on cell types that have “astral” microtubules. Astral microtubules have minus ends embedded in the spindle poles and plus ends extending outward, away from the spindle toward the cell cortex (Fig. 1, A and B). Astral microtubules have been proposed to mediate spindle positioning by generating pulling forces at the cortex or pulling forces against the cytoplasm. The minus ends of astral microtubules are embedded in the pericentriolar material that surrounds the centrioles of animal cells or in the spindle pole bodies of fungi. This attachment is essential for pulling forces on the astral microtubules to move the spindle. However, late stage oocytes of several animal phyla and all cells of flowering plants lack centrioles and lack obvious astral microtubules. Thus, these cell types have evolved alternative spindle positioning mechanisms. Here we review recent advances in both astral and nonastral spindle positioning mechanisms.Open in a separate windowFigure 1.Mitotic spindle movements in the C. elegans zygote. (A) Schematic diagram of a single-celled C. elegans embryo showing cortical pulling by cytoplasmic dynein during pronuclear centration and rotation. The nuclei move toward the anterior (left) so that the spindle assembles in the center of the embryo. (B) Schematic diagram of a single-celled C. elegans embryo showing cortical pulling by dynein during anaphase. The spindle moves to the posterior (right) so that cytokinesis generates two cells of different sizes. The squares highlight a dynein molecule that pulls toward the posterior before spindle displacement, then pulls toward the anterior after spindle displacement. (C) Schematic drawing of cytoplasmic pulling that contributes to centering the pronuclei. (D) Illustration of a spindle that was centered at metaphase but in which both poles moved all the way to the posterior end of the embryo. This occurs in zyg-8 mutants (Gönczy et al., 2001), cls-1,2 (RNAi) embryos (Espiritu et al., 2012), and zyg-9(ts) mutants shifted to a nonpermissive temperature at metaphase (Bellanger et al., 2007), possibly because astral microtubules are too short to reach force generators that would pull toward the anterior.

Cortical versus cytoplasmic pulling

The most prominent model of spindle positioning involves a cortical pulling mechanism. In this model, the minus end–directed microtubule motor protein, cytoplasmic dynein, is attached to the cell cortex and exerts pulling forces on the plus ends of astral microtubules that reach the cortex. In the single-celled Caenorhabditis elegans embryo at early prophase, complexes of GPR-1,2 (G protein regulator) and LIN-5 (abnormal cell lineage), positive regulators of cytoplasmic dynein, are more concentrated at the anterior cortex of the embryo (Fig. 1 A; Park and Rose, 2008), resulting in greater net pulling force toward the anterior. This results in net movement of the pronuclei to the center of the embryo and rotation of the centrosome–pronuclear complex so that the metaphase spindle forms in the center of the embryo with its poles oriented along the anterior-posterior axis of the embryo (Fig. 1 B). During metaphase and early anaphase, GPR-1,2 and LIN-5 become more concentrated at the posterior end of the embryo, resulting in movement of the spindle toward the posterior so that cytokinesis generates daughter cells of different sizes (Fig. 1 B; Grill et al., 2001, 2003). Depending on the relative distribution of active force generators at the cortex, this mechanism can also lead to centering the spindle within the cell to allow symmetric cytokinesis as occurs in HeLa cells (Kiyomitsu and Cheeseman, 2012) and LLC-Pk1 cells (Collins et al., 2012). Cortical pulling might involve pulling on the sides of microtubules that bend as they approach the cortex (Fig. 1 A, 1) or end-on interactions that require coupling of microtubule depolymerization with pulling (Fig. 1 A, 2), as occurs at kinetochores during anaphase A (McIntosh et al., 2010).Cortical pulling differs from a cytoplasmic pulling mechanism most clearly proposed by Kimura and Kimura (2011) and diagrammed in Fig. 1 C. In this cytoplasmic pulling mechanism, the viscous drag on membranous organelles transported toward the minus ends of astral microtubules by cytoplasmic dynein generates a force in the opposite direction, toward the cortex. Depletion of RAB-5, RAB-7, or RILP-1 (RAB-7 interacting lysosomal protein homologue) blocks dynein-dependent organelle transport and slows the velocity of pronuclear centration in C. elegans without affecting other dynein-dependent movements (Kimura and Kimura, 2011). Elimination of cortical pulling by depleting GPR-1,2 also slows pronuclear centration (Park and Rose, 2008), which suggests that cortical pulling and cytoplasmic pulling each contribute 50% of the velocity of pronuclear centration. Unlike end-on cortical pulling, cytoplasmic pulling force is proportional to the length of the astral microtubules because more organelles will be transported on a long microtubule than a short microtubule (Fig. 1 C). This generates a self-centering mechanism as the length of the astral microtubules equalize when the pronuclei reach the center of the zygote (Fig. 1 A; Hamaguchi and Hiramoto, 1986). Cytoplasmic pulling may predominate in very large zygotes where astral microtubules clearly do not reach the cortex but where both pronuclei and the mitotic spindle are centered (Mitchison et al., 2012).

Evidence for cortical pulling.

The cortical pulling mechanism requires that microtubules extend from the spindle to the cortex and form a contiguous structure that is mechanically robust enough that the force generators do not pull the minus ends of the microtubules out of the spindle pole or cause the plasma membrane to buckle inward. Experimental evidence for this contiguous mechanical linkage comes from experiments in oocytes of the marine annelid, Chaetopterus. Insertion of a glass needle into the meiotic spindle allowed pulling the spindle away from the cortex, which caused inward buckling of the cortex. Further pulling resulted in sudden release of the spindle from the cortex, restoration of cortical shape, and concomitant disappearance of a birefringent aster extending between the spindle and cortex (Lutz et al., 1988). The second requirement for a cortical pulling mechanism is that force is generated at the cortex. Using a laser to cut the central spindle of an early anaphase C. elegans embryo, Grill et al. (2001) showed that spindle poles are pulled from outside the spindle rather than pushed from inside the spindle during posterior spindle displacement. Fragmentation of centrosomes with a laser (Grill et al., 2003) revealed that astral microtubules freed from the spindle move outward toward the cortex. Either cortical pulling or cytoplasmic pulling could explain this result; however, the asymmetric distribution of fragment velocities is controlled by proteins that are localized at the cortex, GPR-1,2 (Grill et al., 2003) and LET-99 (lethal-99; Krueger et al., 2010). In a key experiment, Redemann et al. (2010) showed that microtubule plus ends pull tubular invaginations of the plasma membrane inward when cortical stiffness is partially reduced. This experiment showed that astral microtubules are pulling on the cortex during spindle displacement, which would not occur if forces were generated by movement of cytoplasmic organelles along the sides of microtubules.

End-on versus side-on microtubule–cortex interactions.

How do microtubules interact with force generators at the cortex? Astral microtubules might polymerize to the cortex then bend along it so that cortical motors interact with the side of the microtubule to generate force (Fig. 1 A, 1). This type of interaction would be consistent with the in vitro gliding motility of microtubules generated by cytoplasmic dynein immobilized on glass coverslips (Paschal et al., 1987; Vallee et al., 1988) and is the only type of dynein-dependent cortical microtubule interaction observed in budding yeast (Adames and Cooper, 2000). Studies of C. elegans embryos, however, support end-on microtubule contacts (Fig. 1 A, 2) as the functional contact with cortical force generators for posterior displacement of the early anaphase spindle. Live imaging of YFP-tubulin in optical sections at the surface of the embryo during anaphase revealed dots rather than lines, indicating that the microtubules contacting the cortex are <200 nm in length (Labbé et al., 2003; Kozlowski et al., 2007). When short microtubule fragments are generated by ectopic katanin activity, dynein-dependent gliding of microtubule “lines” on the cortex is frequently observed (Gusnowski and Srayko, 2011), indicating that cortical dynein is capable of moving microtubules along the cortex via side-on interactions in wild-type embryos but that it does not during posterior displacement of the anaphase spindle. A likely explanation comes from the finding that astral microtubule plus ends undergo catastrophe (switch to depolymerization) on average 1.4 s after polymerizing to the cortex (Kozlowski et al., 2007). Thus astral microtubule plus ends do not have time to polymerize along the cortex to establish extensive side-on contacts. Support for this idea comes from depletion of the conserved plasma membrane protein EFA-6 (exchange factor for Arf) from the C. elegans embryo. In the absence of EFA-6, the residence time of microtubule plus ends at the cortex increases fivefold, astral microtubules form extensive lateral contacts with the cortex, and centrosomes exhibit movements consistent with excessive dynein-dependent cortical pulling (O’Rourke et al., 2010). Side-on contacts of astral microtubules with the cortex occur later during telophase in the wild-type C. elegans embryo (Kozlowski et al., 2007), but the nature of this switch has not been addressed.The two distinct activities of cytoplasmic dynein, end-on pulling and walking along the side of a microtubule, have been genetically separated in C. elegans. Cortical pulling forces during early anaphase require the redundant cortical dynein activators GPR-1 and -2 (Grill et al., 2003), which bind to LIN-5 (Gotta et al., 2003). GPR-1,2/LIN-5 is anchored in the plasma membrane via the myristoyl and palmitoyl lipid modifications of the redundant Gα proteins GOA-1 and GPA-16 (Gotta et al., 2003; Park and Rose, 2008; Kotak et al., 2012). The complex of GPR-1 and LIN-5 interacts with the dynein light chain DYRB-1 (Couwenbergs et al., 2007) and the dynein regulator LIS-1 (human lissencephaly gene related; Nguyen-Ngoc et al., 2007). GPR-1 and -2, however, are not required for dynein-dependent gliding of severed microtubule fragments along the cortex (Gusnowski and Srayko, 2011), dynein-dependent transport of membranous organelles along the sides of microtubules (Kimura and Kimura, 2011), dynein-dependent positioning of the acentriolar C. elegans meiotic spindle (van der Voet et al., 2009), or dynein-dependent centration of the male pronucleus (Kimura and Kimura, 2011). These GPR-independent activities of cytoplasmic dynein likely do not require end-on pulling.Recently, end-on pulling by cytoplasmic dynein has been reconstituted in vitro with a purified preparation of artificially dimerized budding yeast cytoplasmic dynein. Laan et al. (2012) immobilized purified cytoplasmic dynein on microfabricated barriers and observed the interaction of centrosome-nucleated microtubules as they approached these dynein-coated barriers. Microtubule plus ends hitting a dynein-coated barrier switched to catastrophe with high frequency but the microtubule depolymerization rate after the catastrophe was reduced. The result was an extended period of interaction between the depolymerizing plus end and the dynein-coated barrier. Plus-end depolymerization pulled the centrosome toward the barrier, and in similar reactions the pulling force was measured as high as 5 pN. Side-on interactions with the barrier were not observed. Whereas ATP was required for this end-on pulling, it is not clear if the energy source is ATP hydrolysis–driven stepping by dynein or if the energy source is GTP hydrolysis–driven depolymerization of the microtubule. In the latter case, ATP might only be required to prevent rigor binding of dynein to the microtubule. Indeed, an artificial rigor binding of a streptavidin-coated bead to a depolymerizing biotinylated microtubule plus end resulted in a pulling force that is restricted to an extremely short distance (Grishchuk et al., 2005). In vitro reconstitution of pulling force coupled to depolymerizing microtubule plus ends was first demonstrated with beads coated with kinesin-1 or a nonmotile kinesin chimera (NK350; Lombillo et al., 1995). Like barrier-bound cytoplasmic dynein, kinesin-coated beads slowed the depolymerization rate of microtubule plus ends, whereas NK350-coated beads enhanced the depolymerization rate of bound plus ends. ATP enhanced depolymerization-coupled pulling for both kinesin-1 and NK350, just as it did for barrier-bound cytoplasmic dynein. The in vitro pulling reaction reconstituted by Laan et al. (2012) seems unlikely to be the same reaction that pulls on plus ends in the anaphase budding yeast cell, as these are through side-on interactions (Adames and Cooper, 2000). In vitro reconstitution of a GPR/LIN-5–dependent, end-on pulling reaction with purified metazoan cytoplasmic dynein may reveal mechanisms acting on spindles in vivo.Another interesting contrast between end-on versus side-on cortical pulling reactions is suggested by differences in the dependence on cortical F-actin. F-actin is required for cortical rigidity to prevent end-on microtubule contacts from pulling membrane tubules inward instead of moving the spindle pole outward (Redemann et al., 2010). Side-on pulling by cortical dynein in budding yeast, however, does not require F-actin (Theesfeld et al., 1999; Heil-Chapdelaine et al., 2000a). The curvature of microtubules gliding on the bud cortex indicates that the microtubule is engaged with multiple dynein molecules distributed over several microns of cortex, and this distribution of force might allow effective pulling against a less rigid cortex. Alternatively, rigidity of the yeast plasma membrane might be mediated by oligomers of BAR domain proteins like Num1 (nuclear migration; Tang et al., 2012) or eisosomes (Walther et al., 2006; Olivera-Couto et al., 2011), by osmotic pressure, or by attachment of the plasma membrane to the cell wall.

Why the spindle is not pulled all the way to the cortex with more active force generators.

What prevents the C. elegans centrosome–pronuclear complex from moving all the way to the anterior cortex where the concentration of cortical force generators is highest during prophase (Fig. 1 A), and what prevents the spindle from moving all the way to the posterior cortex, which has the highest concentration of active force generators during anaphase (Fig. 1, B and D)? Increasing the concentration of GPR-1,2/LIN-5 at the anterior cortex causes pronuclei to move further toward the anterior, but they still do not crash into the anterior cortex (Panbianco et al., 2008). During metaphase/anaphase (Fig. 1 B), weak cortical pulling on the anterior aster might oppose strong pulling on the posterior aster. Supporting this idea, spindle severing results in the posterior aster moving further posterior than when the spindle is intact (Grill et al., 2001), but the posterior pole still does not move all the way to the posterior cortex. Monopolar spindles move toward the posterior in a GPR1,2-dependent manner but then reverse direction and oscillate along the anterior-posterior axis (Krueger et al., 2010). Laan et al. (2012) found that centrosome-nucleated microtubule asters could accurately self-center within microchambers whose walls are coated with dynein. Their mathematical modeling suggested that self-centering was achieved because of a balance between cortical pulling by dynein and pushing by polymerizing microtubules (Dogterom and Yurke, 1997) that have not yet engaged a dynein molecule. Thus, cortical pushing by astral microtubules might buffer cortical pulling in vivo. Grill and Hyman (2005) suggested another simple solution. When a spindle pole moves to the posterior, it passes a subset of cortical force generators that were initially pulling toward the posterior (Fig. 1 B, box). After passage of the spindle pole, these force generators pull toward the anterior. Failure in any of these buffering mechanisms might explain why both spindle poles move to the posterior cortex (Fig. 1 D) in mutants that have either short astral microtubules or fewer astral microtubules reaching the cortex (Gönczy et al., 2001; Bellanger et al., 2007; Espiritu et al., 2012).Recent experiments in HeLa and LLC-Pk1 cells have revealed a more sophisticated feedback mechanism that effectively centers the mitotic spindle. Kiyomitsu and Cheeseman (2012) found that HeLa cell mitotic spindles oscillate back and forth along their pole-to-pole axis. They found that dynein/dynactin formed a lateral crescent on the cortex when the spindle was far from that lateral cortex (Fig. 2 A). As the spindle moved toward the dynein crescent, the dynein crescent disappeared as the spindle approached and a new crescent appeared on the opposite lateral cortex (Fig. 2 C). They found that polo kinase 1 (Plk1), which is concentrated on spindle poles, causes dynein/dynactin to dissociate from the LGN–NuMA–Gαi complex (Leu-Gly-Asn repeat enriched protein–nuclear mitotic apparatus protein–Gα; homologues of GPR-1,2–LIN-5–Gα), which explains why the dynein crescent disappears once the spindle pole gets close to the cortex. They also found that the GTP-Ran gradient produced by chromosome-bound RCC1 (regulator of chromosome condensation) was responsible for inhibiting LGN–NuMA association with the cortex, and hence dynein, above the central spindle (Fig. 2). The RCC1 pathway explains why the spindles move only along their pole–pole axis. The two pathways combine to center the spindle in two different axes. Similar spindle oscillations with dynein/dynactin crescents forming only when the spindle pole is far from the cortex were reported in LLC-Pk1 epithelial cells (Collins et al., 2012). Because spindles are small relative to the two-dimensional flattened area of these cells, the role of this pathway in centering spindles to allow symmetric cytokinesis is much more obvious than in HeLa cells.Open in a separate windowFigure 2.How to center a spindle. Schematic diagram of a metaphase HeLa cell where the spindle oscillates along its pole–pole axis to maintain a centered position to allow symmetric cytokinesis. (A) When the left spindle pole is close to the cortex, Plk1 on the pole (red) causes dynein (green) to dissociate from LGN–NuMA complexes (purple; human homologues of GPR-1,2/LIN-5). (B and C) The spindle moves to the right because of the higher concentration of LGN–NuMA–dynein complexes on the right cortex. When chromosomes are close to the cortex as in A, the GTP-Ran gradient from the chromosomes causes dissociation of LGN/NuMA from the cortex. This second system centers the spindle in the axis perpendicular to the pole–pole axis.There are normal situations where movement of one spindle pole all the way to the cortex occurs. This exaggerated movement that results in one set of astral microtubules being splayed onto the cortex is observed in fourth cleavage sea urchin embryos (Holy and Schatten, 1991) and for the female meiotic spindles of Chaetopterus (Lutz et al., 1988) and Spisula solidissima (Pielak et al., 2004).

The budding yeast model.

The S. cerevisiae mitotic spindle is positioned relative to a preformed bud neck by two sequential and partially redundant cortical pulling pathways to ensure that chromosomes are deposited into both mother and daughter cells. Spindle positioning is also monitored by a budding yeast–specific checkpoint (Caydasi et al., 2010). Deletion of genes in any one positioning pathway results in a viable yeast strain, whereas double mutants bearing deletions of genes in both positioning pathways or of genes in one positioning pathway plus the checkpoint results in lethality (Miller et al., 1998). In the early pathway, the microtubule plus-end tip tracking protein, Bim1 (binding to microtubules; EB1 homologue), binds to the yeast-specific adaptor protein Kar9 (karyogamy; Korinek et al., 2000; Lee et al., 2000; Miller et al., 2000), which binds to the yeast myosin V (Myo2; Yin et al., 2000). Myosin V transports the growing microtubule plus end toward the bud tip on polarized actin cables (Hwang et al., 2003). This results in a unique “sweeping” or “pivoting” of the growing astral microtubule toward the bud neck (Fig. 3 A; Adames and Cooper, 2000). Because Bim1 only binds to growing microtubule plus ends (Zimniak et al., 2009), this mechanism alone cannot bring the spindle to the bud neck because the polymerizing astral microtubule would push the spindle back into the mother cell. To prevent the astral microtubules from becoming too long, the kinesin-8 family member, Kip3, passively tracks the plus end until the plus end reaches the bud cortex (Fig. 3 A). Kip3 then switches to a plus-end depolymerase and shortens the astral microtubule to pull the spindle to the bud neck (Fig. 3, B and C; Gupta et al., 2006). Purified Kip3 has unique biochemical properties that contribute to its in vivo function. In vitro, Kip3 accumulates at plus-end tips via its plus end–directed motor activity. Longer microtubules accumulated more Kip3 at their plus ends than short microtubules simply because there are more lateral binding sites on a long microtubule and Kip3 switches to a plus-end depolymerase only when the microtubule has reached a threshold length (Varga et al., 2009). Plus-end pulling by the fission yeast homologue of Kip3 has also been reconstituted in vitro (Grissom et al., 2009). More recent work suggests that Kip3-dependent cortical pulling requires the concerted action of the cortical protein, Bud6 (BUD site selection), and the plus end tracker, Bim1, as well as cytoplasmic dynein (ten Hoopen et al., 2012).Open in a separate windowFigure 3.Spindle positioning in budding yeast. Schematic diagram of the two sequential spindle positioning pathways of budding yeast. In the early pathway (A–C), myosin V transports the plus end of an astral microtubule toward the bud tip on a polarized actin cable. Once the plus end has reached the bud cortex, the plus-end depolymerase, KIP3, is activated to allow pulling of the spindle pole toward the bud neck. In the late pathway (D–F), the plus end–directed microtubule motor Kip2 transports dynein to the plus ends of microtubules via the adaptor protein Bik1 (D). Dynein can be targeted to plus ends by two additional Bik1-dependent mechanisms (see text). When dynein reaches the bud cortex on a polymerizing microtubule plus end (E), contact with the cortical protein Num1 allows dynein to pull the spindle toward the bud (F). During early anaphase (D and E), dynein is not loaded onto microtubules in the mother cell. During late anaphase (F), dynein is loaded on microtubules in the mother cell to prevent movement of the spindle all the way into the bud.The late pathway is initiated when the yeast-specific dynein inhibitor She1 (sensitivity to high expression) is removed from astral microtubules at the metaphase–anaphase transition (Woodruff et al., 2009). She1 appears to act specifically by preventing recruitment of dynactin to microtubules (Bergman et al., 2012) and by inhibiting dynein motility (Markus et al., 2012). In the late pathway, cytoplasmic dynein is targeted to growing microtubule plus ends by the plus-end tracking protein Bik1 (bilateral karyogamy defect; Sheeman et al., 2003), which itself is targeted to plus ends by three partially redundant mechanisms. In the first mechanism, Bik1 is transported as cargo by the plus end–directed kinesin Kip2 (Carvalho et al., 2004). Cytoplasmic dynein is thus transported as cargo to the bud cortex by Bik1–Kip2 complexes (Fig. 3 D). In the absence of Kip2, Bik1 (and therefore dynein) can still track growing microtubule plus ends either through a second mechanism that requires the C-terminal tyrosine residue of α-tubulin or a third mechanism that requires the plus end–tracking protein Bim1 (Caudron et al., 2008). This is partially consistent with results of reconstitution experiments with purified proteins showing that the Bik1 homologue, CLIP170, tracks growing plus ends through a mechanism that involves binding to both the Bim1 homologue, EB1, and the C-terminal tyrosine-containing motif of α-tubulin (Bieling et al., 2008). When a microtubule plus end carrying dynein contacts the yeast-specific cortical protein Num1, Num1 apparently stimulates off-loading of the dynein tail onto the cortex so that the dynein motor domains engage the microtubule in a cortical pulling reaction (Fig. 3, E and F). Deletion of Num1 results in accumulation of inactive dynein at plus ends (Lee et al., 2003; Sheeman et al., 2003; Markus and Lee, 2011). In contrast with the GPR-dependent end-on cortical interactions observed in C. elegans, dynein-dependent cortical pulling is mediated by lateral sliding of astral microtubules along the yeast bud cortex (Fig. 3, E and F; Adames and Cooper, 2000). Also, unlike cortical GPR/LIN-5 in animal cells, cortical Num1 is distributed in patches throughout both mother and daughter cells (Heil-Chapdelaine et al., 2000b) and even participates in mitochondrial positioning and fission throughout the cell (Cerveny et al., 2007; Hammermeister et al., 2010). If there is no asymmetrically distributed cortical activator of dynein, how is dynein-mediated pulling directed specifically toward the bud cortex?

Why both spindle poles are not pulled to the same cortex in budding yeast.

The budding yeast spindle, rather than the cortex, is asymmetrical. At metaphase, Kar9 is asymmetrically localized on the spindle pole oriented toward the bud and on the plus ends of astral microtubules emanating from that bud-proximal spindle pole (Liakopoulos et al., 2003). This alone would explain why only one spindle pole moves toward the bud but then leaves the question of how Kar9 asymmetry is established. Cepeda-García et al. (2010) found that Kar9 at spindle poles became symmetrical and reduced in concentration after depolymerization of actin cables, depolymerization of microtubules, or disruption of the myosin V–Kar9 interaction. When microtubules were repolymerized, Kar9 was initially symmetrical on both spindle poles and quickly repolarized onto the first microtubule to make a functional cortical contact. These results suggested a positive feedback loop in which functional cortical pulling by Bim1–Kar9–Myo2 complexes causes loading of additional Kar9. Cytoplasmic dynein also accumulates preferentially on the plus-end tips that reach the bud neck first (Fig. 3, D and E) and on the bud-proximal spindle pole during metaphase. The asymmetrical accumulation of dynein on the bud-proximal microtubules and bud-proximal spindle pole requires kinases that are found at the bud neck. Thus, the asymmetry of dynein localization may be generated by a positive feedback loop, as suggested for Kar9 asymmetry. After the spindle is pulled into the bud neck, during anaphase, dynein becomes symmetrical on the plus ends emanating from both poles (Fig. 3 F; Grava et al., 2006). This regulation prevents both poles from moving toward the bud during metaphase and then prevents the spindle from being pulled all the way into the bud during anaphase. Asymmetric localization of dynein on spindle poles or microtubule plus ends has not yet been reported in animal cells.

Other spindle positioning mechanisms

In the examples of the C. elegans and budding yeast mitotic spindles, long astral microtubules are in contact with the cell cortex. In cells where spindles have no astral microtubules, other mechanisms must be at work. Female meiotic spindles are universally positioned with one spindle pole contacting the oocyte cortex so that one set of chromosomes can be eliminated in a polar body through an extremely asymmetrical division (Fabritius et al., 2011). Female meiotic spindles of at least three animal phyla (Chordata, Nematoda, and Arthropoda), however, have no centrioles in their spindle poles and no apparent astral microtubules. Work in C. elegans and mice suggests that different species have evolved different mechanisms for acentriolar meiotic spindle positioning.

Parallel metaphase meiotic spindles.

The metaphase I and metaphase II spindles of C. elegans (Fig. 4 A) and the metaphase II spindle of mouse (Fig. 4 B) are positioned at the cortex in a parallel orientation, with both poles equidistant from the cortex. In C. elegans, this parallel cortical positioning requires microtubules, kinesin-1, and a worm-specific kinesin-1–binding partner, KCA-1 (Yang et al., 2003, 2005), but is independent of F-actin (Yang et al., 2003). Kinesin-1 and microtubules are also required to move the nucleus to the cortex before germinal vesicle breakdown and to drive transport of yolk granules inward from the cortex, which results in a circular streaming pattern. It has been suggested that kinesin-1 may only move the germinal vesicle to the cortex and that an additional, unidentified pathway moves the spindle over the remaining distance to the cortex and establishes the parallel orientation (McNally et al., 2010). The mouse metaphase II spindle is maintained in a similar parallel orientation at the cortex, but this positioning requires the actin nucleator ARP2/3. ARP2/3 also drives streaming of actin filaments and cytoplasm in a pattern that has been proposed to push the spindle into the cortex to maintain parallel cortical position (Fig. 4 B; Yi et al., 2011).Open in a separate windowFigure 4.A plethora of nonastral spindle positioning mechanisms. (A) Metaphase C. elegans meiotic spindles are positioned in a parallel orientation at the cortex by microtubules and kinesin-1. (B) The mouse metaphase II spindle may be positioned by actin-dependent cytoplasmic streaming. Pole-first migration of the mouse meiosis I spindle to the cortex may be mediated by cargo transport on parallel actin filaments by spindle pole–bound myosin II (C), myosin II–based contraction of anti-parallel actin filaments (D), or pushing forces generated by polymerizing actin filaments nucleated by formin molecules on the spindle (E) or nucleated by formin molecules on the cortex (F). Red arrows indicate the pointed ends of actin filaments. (G) One spindle pole of the early anaphase C. elegans meiotic spindle may be transported to the cortex as cargo by dynein on polarized cytoplasmic microtubules.

Pole first migration of the mouse meiosis I spindle.

Unlike the C. elegans germinal vesicle, the mouse germinal vesicle is centered in the egg at germinal vesicle breakdown. Thus, the meiosis I spindle assembles near the center of the egg then migrates in a pole-first orientation to the nearest cortex so that it never adopts a parallel orientation (Fig. 4 C). This migration requires F-actin (Verlhac et al., 2000) and the actin nucleators Formin 2 (Dumont et al., 2007), Spire 1 and Spire 2 (Pfender et al., 2011), and ARP2/3 (Sun et al., 2011), but the mechanism of movement remains unclear. Schuh and Ellenberg (2008) demonstrated the existence of actin bundles extending between the spindle and an invagination of the cortex during spindle migration. The invagination indicated a pulling mechanism, and Schuh and Ellenberg (2008) suggested that myosin II on the spindle poles might walk on a discontinuous actin network with barbed ends oriented toward the cortex (Fig. 4 C). In support of this model, the Rho kinase inhibitor ML7 eliminated spindle pole staining by an antibody specific for phosphorylated myosin regulatory light chain and blocked spindle migration (Schuh and Ellenberg, 2008). However, Li et al. (2008) found that the myosin ATPase inhibitor, blebbistatin, had no effect on spindle migration even though it completely blocked polar body extrusion (cytokinesis). Because myosin II typically acts by forming bipolar thick filaments that exert contractile force on antiparallel actin filaments, a myosin II–based model would make more sense if myosin II was concentrated on antiparallel actin bundles extending between the spindle and cortex (Fig. 4 D). Myosin V is more typically associated with the transport of cargo on uniformly oriented actin filaments, and a recent study has shown that myosin V drives transport of secretory vesicles outward toward the cortex in germinal vesicle stage oocytes (Schuh, 2011). This study at least suggests that the cytoplasmic actin meshwork has a net polarity with barbed ends toward the cortex, a prerequisite for a myosin cargo transport model (Fig. 4 C).Li et al. (2008) proposed a completely different mechanism in which F-actin nucleated near the chromosomes generates a cloud of F-actin that pushes the spindle toward the cortex (Fig. 4 E) in a manner analogous to Listeria monocytogenes motility (Lambrechts et al., 2008). Support for this pushing model comes from imaging of an actin cloud behind the migrating spindle and localization of Formin 2 around the spindle (Li et al., 2008). In addition, Formin 2 overexpression causes invaginations in the nuclear envelope, which is consistent with inward pushing from the cortex, rather than protrusions that would be consistent with pulling forces from the cortex (Azoury et al., 2011). Formin 2 is symmetrical around the cortex during prophase but clears from the cortex in front of the migrating spindle, which is consistent with nucleation-based pushing from behind (Fig. 4 F; Azoury et al., 2011). As previously mentioned, cortical stiffness is a prerequisite for cortical pulling mechanisms because pulling on an unsupported plasma membrane should cause the membrane to invaginate inward instead of the spindle moving outward. Strikingly, cortical stiffness of the mouse oocyte decreases sixfold during spindle migration (Larson et al., 2010). Clearly, more work is required to resolve the mechanism of pole-first spindle migration in the mouse oocyte.

Rotation of the parallel metaphase spindle to a perpendicular anaphase spindle.

Activation of the anaphase-promoting complex results in rotation of the parallel metaphase meiotic spindle to a perpendicular orientation in both meiotic divisions of C. elegans, whereas fertilization induces rotation during meiosis II in mouse. In kinesin-1–depleted C. elegans embryos, the metaphase I spindle is far from the cortex and initiates pole-first migration to the cortex at the same time that wild-type rotation initiates (Yang et al., 2005). Both spindle rotation and late spindle migration in a kinesin mutant require cytoplasmic dynein (Ellefson and McNally, 2009, 2011). These results indicate that spindle rotation is simply migration of one spindle pole toward the cortex (analogous to spindle migration during mouse meiosis I). However, unlike dynein-dependent migration of one spindle pole toward the cortex during C. elegans mitosis or HeLa cell mitosis, C. elegans meiotic spindle rotation does not require GPR-1,2 (van der Voet et al., 2009). One possible model for C. elegans meiotic spindle rotation is that cytoplasmic dynein, which accumulates on both spindle poles just before and during rotation, transports one spindle pole as cargo on cytoplasmic microtubules with minus ends anchored at the cortex (Fig. 4 G). The orientation of these microtubules is inferred from the direction of kinesin-dependent yolk granule movement (McNally et al., 2010) and hook decoration in Xenopus laevis oocytes (Pfeiffer and Gard, 1999). This model is essentially the same as the myosin cargo transport model proposed for mouse meiosis I (Fig. 4 C). Rotation of the mouse meiosis II spindle is actin dependent (Maro et al., 1984) and myosin II dependent (Matson et al., 2006; Wang et al., 2011), but the mechanism is unknown.

Unifying themes and future directions

Work in HeLa cells and budding yeast suggests that negative feedback loops might generally lead to spindle centering and that positive feedback loops might generally lead to asymmetrical spindle positioning. Whereas the recent x-ray crystal structures of cytoplasmic dynein (Carter et al., 2011; Kon et al., 2012) have revealed great insights into how the motor walks, they have revealed little about how the motor is locally activated and anchored at the cortex or how GPR-1/LIN-5 switches the motor from a side-on motor to an end-on motor. More attention needs to be focused on the distinction between cortical stiffness and cortical anchoring required in any cortical pulling mechanism. The nonastral spindle positioning mechanisms acting in oocytes of mouse and C. elegans will require a much more detailed understanding of the polarity of cytoplasmic actin filaments and cytoplasmic microtubules.     

Calmodulin regulation (calmodulation) of voltage-gated calcium channels

Calmodulin regulation (calmodulation) of the family of voltage-gated Ca_V1-2 channels comprises a prominent prototype for ion channel regulation, remarkable for its powerful Ca²⁺ sensing capabilities, deep in elegant mechanistic lessons, and rich in biological and therapeutic implications. This field thereby resides squarely at the epicenter of Ca²⁺ signaling biology, ion channel biophysics, and therapeutic advance. This review summarizes the historical development of ideas in this field, the scope and richly patterned organization of Ca²⁺ feedback behaviors encompassed by this system, and the long-standing challenges and recent developments in discerning a molecular basis for calmodulation. We conclude by highlighting the considerable synergy between mechanism, biological insight, and promising therapeutics.The ancient Ca²⁺ sensor protein calmodulin (CaM) has emerged as a pervasive modulator of ion channels (Saimi and Kung, 2002), manifesting remarkable Ca²⁺ sensing capabilities in this context (Dunlap, 2007; Tadross et al., 2008) and furnishing essential Ca²⁺ feedback in numerous biological settings (Alseikhan et al., 2002; Xu and Wu, 2005; Mahajan et al., 2008; Adams et al., 2010; Morotti et al., 2012). Such modulation was discovered via mutations in CaM that altered the motile behavior of Paramecium, stemming from blunted activation of Ca²⁺-dependent Na⁺ current or K⁺ current (Kink et al., 1990). Since then, numerous ion channels have been found to be modulated by CaM, as extensively reviewed (Budde et al., 2002; Saimi and Kung, 2002; Wei et al., 2003; Halling et al., 2006; Tan et al., 2011; Nyegaard et al., 2012). This paper focuses on the CaM regulation (calmodulation) of voltage-gated Ca²⁺ channels, which are extensively regulated by intracellular Ca²⁺ binding to CaM (Ca²⁺/CaM; Lee et al., 1999; Peterson et al., 1999; Zühlke et al., 1999; Haeseleer et al., 2000; DeMaria et al., 2001; Budde et al., 2002; Halling et al., 2006). Such feedback regulation by Ca²⁺/CaM demonstrates versatile functional capabilities, represents a prominent prototype for ion-channel modulation, and holds far-reaching biological consequences. These attributes contribute much to the standing of voltage-gated Ca²⁺ channels as the queen of ion channels, molecules that not only sculpt membrane electrical waveforms but also serve as gatekeepers of the ubiquitous second messenger Ca²⁺ (Yue, 2004).

Classic history of discovery

The earliest signs of Ca²⁺-dependent regulation of Ca²⁺ channels came from data showing that increased Ca²⁺ could accelerate the inactivation of Ca²⁺ currents in Paramecium (Brehm and Eckert, 1978), invertebrate neurons (Tillotson, 1979), and insect muscle (Ashcroft and Stanfield, 1981). These results gave birth to the concept of Ca²⁺-dependent inactivation (CDI), the then counterintuitive notion that entities other than transmembrane voltage could modulate the rapid gating of ion channels (Eckert and Tillotson, 1981).Methodological advances permitting routine isolation and recording of Ca²⁺ currents within vertebrate preparations clearly revealed the existence of CDI of cardiac L-type Ca²⁺ currents (carried by Ca_V1.2 channels) in both multicellular preparations (Kass and Sanguinetti, 1984; Mentrard et al., 1984) and isolated myocytes (Lee et al., 1985). Classic data illustrating CDI of cardiac L-type currents is shown in Fig. 1 (A–C). Fig. 1 A (Ca) shows Ca²⁺ current from frog cardiocytes, elicited by a test voltage pulse to +80 mV (Mentrard et al., 1984). CDI becomes apparent upon Ca²⁺ entry during a voltage prepulse to +80 mV (Fig. 1 B), which causes sharply attenuated Ca²⁺ current in a subsequent test pulse (diminished area of red shading). To exclude voltage depolarization itself as the cause of inactivation, a prepulse to +120 mV (Fig. 1 C) can be demonstrated to produce far less inactivation within a subsequent test-pulse current. Because this prepulse imposed strong depolarization, but diminished Ca²⁺ entry via decreased driving force, the reemergence of test Ca²⁺ current suggests that the strong inactivation evident in Fig. 1 B could be attributed to CDI. Data of this type specifies an inactivation mechanism that depends on voltage as a U-shaped function, now considered a hallmark of CDI (Eckert and Tillotson, 1981).Open in a separate windowFigure 1.Classic U-shaped signature of CDI. (A–C) Early voltage-clamp recordings from a multicellular preparation of frog atrial trabeculae cells demonstrates CDI of Ca²⁺ currents. All listed voltages are relative to resting potential E_R (∼−80 mV) Adapted from Mentrard et al. (1984). (A) Ca²⁺ currents (shaded pink) evoked in response to an +80-mV depolarizing test pulse show robust inactivation. Fast capacitive transient was estimated by blocking these Ca²⁺ currents with 3 mM Co²⁺ solution (Co). Vertical bar corresponds to 0.5 µA of current; horizontal bar, 100 ms. (B) The Ca²⁺ current in response to the test pulse is sharply attenuated when preceded by an +80 mV prepulse. This reduction in current amplitude reflects the inactivation of Ca²⁺ channels. Format is as in A. (C) Further increase in the prepulse potential to +120 mV restores Ca²⁺ current amplitude (compare with B). Here, the diminished Ca²⁺ entry during the prepulse was insufficient to trigger CDI. The format is as in A. (D–F) Single channel recordings of L-type Ca²⁺ channels from adult rat ventricular myocytes also exhibit U-shaped dependence of Ca²⁺ channel inactivation. Adapted from Imredy and Yue (1994) with permission from Elsevier. (D) Depolarizing voltage pulse to +20 mV evokes representative elementary Ca²⁺ currents. Ensemble average currents are shown on the bottom. (E) When depolarizing pulse was preceded by +40 mV prepulse, the elementary Ca²⁺ currents are sparser and the first opening is delayed. This reduction in open probability parallels the sharp attenuation of macroscopic Ca²⁺ currents seen in B. The ensemble average is shown on the bottom. (F) Further increase in prepulse voltage to +120 mV led to a reversal of this gating pattern, once again highlighting the U-shaped dependence of CDI. Ensemble average is shown on the bottom.Still, the actual nature of effects of Ca²⁺ upon gating remained unclear. Fig. 1 (D–F) reproduces the profile of CDI at the single-molecule level (Yue et al., 1990; Imredy and Yue, 1992, 1994). Resolving data such as these is technically challenging, owing to the diminutive size of unitary Ca²⁺ currents (∼0.3 pA) and the sub-millisecond gating timescale involved. These data demonstrate the existence of U-shaped inactivation in the gating of a single cardiac L-type Ca²⁺ channel fluxing Ca²⁺, as present in an adult rat ventricular myocyte (Imredy and Yue, 1994). These results established that Ca²⁺ influx of a single channel suffices to trigger CDI, and demonstrated the clear correspondence of single-channel and multicellular behavior (Fig. 1). Thus, CDI emerged as a legitimate molecular-level modulatory process.

Advent of Ca²⁺ channel calmodulation

The Ca²⁺ sensor mediating Ca²⁺ regulation of Ca²⁺ channels has long been a matter of debate, with proposals of direct binding of Ca²⁺ to the channel complex (Standen and Stanfield, 1982; Plant et al., 1983; Eckert and Chad, 1984), and of Ca²⁺-dependent phosphorylation and/or dephosphorylation of channels (Chad and Eckert, 1986; Armstrong, D., C. Erxleben, D. Kalman, Y. Lai, A. Nairn, and P. Greengard. 1988. Abstracts of papers presented at the forty-second annual meeting of the Society of General Physiologists. Abstr. 20). The road toward identification of the actual Ca²⁺ sensor began with the cloning and expression of recombinant Ca²⁺ channels, thereby enabling structure–function studies (Snutch and Reiner, 1992). This phase of discovery was initiated by a bioinformatic observation (Babitch, 1990), pointing out that the carboxy terminus of voltage-gated Ca²⁺ channels contains a region with weak homology to an EF hand Ca²⁺-binding motif (Fig. 2 A, EF1). After the recognition of EF1, the existence of a second EF hand–like motif became apparent (Fig. 2 A, EF2). This led to chimeric channel analysis (de Leon et al., 1995), wherein segments of the carboxy termini of L-type (Ca_V1.2) and R-type (Ca_V2.3) Ca²⁺ channels were swapped. These two types of channels exhibit strong and weak CDI, respectively, under conditions of strong intracellular Ca²⁺ buffering (Liang et al., 2003), making them advantageous for chimeric analysis. Results from these experiments revealed that the proximal third of the carboxy terminus (Fig. 2 A, CI region) was important for CDI, and that EF1 of Ca_V1.2 channels was essential for the strong CDI in these channels (de Leon et al., 1995). One possible explanation was that EF1 might be the Ca²⁺ binding site for CDI postulated previously (Standen and Stanfield, 1982; Plant et al., 1983; Eckert and Chad, 1984). However, mutations within the Ca²⁺ binding loop of EF1 failed to eliminate CDI (Zhou et al., 1997; Peterson et al., 2000), focusing the search for the Ca²⁺ sensor elsewhere.Open in a separate windowFigure 2.Calmodulation: the ingredients and the flavors. (A) Channel diagram illustrates overall arrangement of structural landmarks critical for CDI. The Ca²⁺-inactivation (CI) region, spanning ∼160 residues of the channel carboxy terminus, is highly conserved across Ca_V1/2 channel families and is elemental for CDI. The proximal segment (PCI) of the CI region includes the dual vestigial EF hand (EF) segments (shaded purple and blue-green). The IQ domain, a canonical CaM binding motif critical for CDI, is located just downstream of the PCI segment. The NSCaTE element in the amino terminus of Ca_V1.2/1.3 channels is an N-lobe Ca²⁺/CaM effector site. The CBD element (gray) in the carboxy terminus of Ca_V2 channels is a CaM binding segment thought to be critical for CDI. (B) The table outlines functional bipartition of CaM in Ca_V channels and the corresponding spatial Ca²⁺ selectivities. In Ca_V1.2 and Ca_V1.3 channels, both the C-lobe and N-lobe of CaM enable fast CDI with local Ca²⁺ selectivity. Latent CDI of Ca_V1.4 channels is revealed upon deletion of an autoinhibitory domain in the distal carboxy terminus. Throughout the Ca_V2 channel family, the N-lobe of CaM evokes slow CDI with global Ca²⁺ spatial selectivity. In Ca_V2.1 channels, the C-lobe of CaM supports ultra-fast facilitation (CDF) with local Ca²⁺ selectivity. No C-lobe triggered modulation has been reported for Ca_V2.2 and Ca_V2.3 channels.Targets of CaM binding themselves often resemble CaM, a bilobed molecule with each lobe composed of two EF hands (Jarrett and Madhavan, 1991). The dual vestigial EF hands in the carboxy terminus of channels (Fig. 2 A) may be seen as resembling a lobe of CaM, which suggests that CaM itself might be the Ca²⁺ sensor for CDI. Initial studies exploring this notion, however, showed that pharmacological inhibition of CaM did not eliminate CDI of L-type Ca²⁺ channels (Imredy and Yue, 1994; Zühlke and Reuter, 1998). Nonetheless, deletions within the Ca_V1.2 channel CI region identified an IQ motif (Fig. 2 A) as a critical element in CDI (Zühlke and Reuter, 1998), and mutagenesis of this IQ domain weakens CDI (Qin et al., 1999). As IQ motifs often bind Ca²⁺-free CaM (apoCaM; Jurado et al., 1999), the possibility that CaM acts as a CDI sensor reemerged. Definitive evidence came with studies involving a Ca²⁺-insensitive mutant form of CaM (CaM₁₂₃₄), in which point mutations within all four EF hands eliminate Ca²⁺ binding (Xia et al., 1998). If “preassociation” of apoCaM with target molecules were a prerequisite for subsequent regulation by Ca²⁺ binding to this apoCaM, then CaM₁₂₃₄ could act as a dominant negative to silence regulation, as seen for small-conductance Ca²⁺-activated K channels (Xia et al., 1998). Indeed, when Ca_V1.2 channels were coexpressed with CaM₁₂₃₄ or its analogues, CDI was ablated or strongly suppressed (Peterson et al., 1999; Zühlke et al., 1999). Additionally, apoCaM was found to bind to the Ca_V1.2 CI region, with critical dependence upon the IQ segment (Erickson et al., 2001; Pitt et al., 2001; Erickson et al., 2003). In retrospect, apoCaM preassociation with Ca_V1.2 channels would sterically protect CaM from pharmacological effects (Dasgupta et al., 1989), rationalizing the prior insensitivity of CDI to such small-molecule perturbation. In all, these data firmly substantiated CaM as the sensor for Ca_V1.2 channel Ca²⁺ regulation.Initially, it was believed that only Ca_V1.2 L-type channels (Fig. 2 B) were subject to CaM-mediated Ca²⁺ regulation (Zamponi, 2003). However, this system of calmodulation was gradually recognized to pertain to most of the Ca_V1 and Ca_V2 (but not Ca_V3) branches of the Ca_V channel superfamily (Liang et al., 2003; Fig. 2 B). P-type (Ca_V2.1) channels were found to be Ca²⁺ regulated (Lee et al., 1999), and a Ca²⁺/CaM binding site downstream of the IQ element (CBD; Fig. 2 A) was argued to be important. In this initial study, no role for CaM preassociation was recognized, and no role for the IQ domain was found. Moreover, Ca²⁺ regulation of Ca_V2.1 channels manifests as a facilitation of current (Ca²⁺-dependent facilitation; CDF), followed by a slowly developing CDI. Thus, it appeared possible that the Ca²⁺ regulation of Ca_V2.1 channels might diverge mechanistically from that of Ca_V1.2 channels. However, apoCaM was later found to preassociate with Ca_V2.1 channels (Erickson et al., 2001), Ca²⁺-insensitive mutant CaM molecules were found to eliminate Ca²⁺ regulation in Ca_V2.1 channels (DeMaria et al., 2001; Lee et al., 2003), and the IQ domain was determined to be structurally essential for this regulation (DeMaria et al., 2001; Lee et al., 2003; Kim et al., 2008; Mori et al., 2008). That said, the role of the CBD segment remains contentious, with some studies still arguing for this segment’s importance in CDI (Lee et al., 2003). In contrast, deletion of the entire carboxy terminus after the IQ domain (including the CBD element) completely spared CDF and CDI of Ca_V2.1 channels (DeMaria et al., 2001; Chaudhuri et al., 2005). Overall, Ca_V2.1 and Ca_V1.2 channels appear to be Ca²⁺ regulated by a largely conserved scheme.Soon thereafter, calmodulation was established for the remaining members of the Ca_V2 class of channels (Fig. 2 B, Ca_V2.2 and Ca_V2.3), which indicates that this modulatory system pertains throughout the Ca_V1-2 channel superfamily (Liang et al., 2003). Strong CaM-mediated CDI was found in Ca_V1.3 channels (Xu and Lipscombe, 2001; Shen et al., 2006; Yang et al., 2006); a latent capacity for CaM-mediated CDI was observed in Ca_V1.4 channels (Singh et al., 2006; Wahl-Schott et al., 2006); and potential indications of CaM-mediated regulation of Ca_V1.1 channels have been described (Stroffekova, 2008, 2011).A few more nuanced points nonetheless merit attention. First, some types of calmodulation are insensitive to strong intracellular Ca²⁺ buffering (e.g., Ca_V1.2 CDI), whereas others are not (Ca_V2.3 CDI; Fig. 2 B). This contrasting sensitivity enables the use of chimeric channels under increased Ca²⁺ buffering to identify key structural determinants (de Leon et al., 1995), despite the existence of calmodulation across the channel superfamily. Second, the mechanisms underlying CDF in Ca_V1.2 channels remain mysterious, as the CDF of native channels of the heart is weak to start, and attenuated by blockade of Ca²⁺ release from neighboring ryanodine receptor channels (RYRs; Wu et al., 2001). Additionally, CDF of Ca_V1.2 channels is strongly manifest only in recombinant channels containing point mutations within the IQ domain, and only when they are expressed in frog oocytes (Zühlke et al., 2000). Curiously, CDF is not observed in recombinant Ca_V1.2 channels expressed in mammalian cell lines (Peterson et al., 1999). In contrast, CDI of Ca_V1.2 channels is strong and universally observed across experimental platforms. Third, Ca²⁺/CaM-dependent dephosphorylation by calcineurin was initially suggested as a mechanism for CDI of Ca²⁺ channels (Chad and Eckert, 1986); however, this proposition has remained controversial as previously reviewed (Budde et al., 2002). Recently, calcineurin was found to preassemble with the Ca_V1.2 channel complex through the scaffolding protein AKAP79/150 bound to the distal carboxy terminus of the channel (Oliveria et al., 2007). Furthermore, two additional maneuvers were reported to impede CDI of these channels (Oliveria et al., 2012): the overexpression of a mutant AKAP79/150 incapable of binding calcineurin, and inhibition of calcineurin activity. However, others have found that direct inhibition of calcineurin has no effect on the CDI of L-type channels within multiple neuronal preparations (Branchaw et al., 1997; Victor et al., 1997; Zeilhofer et al., 1999). Moreover, deletion of the entire distal carboxy terminus (including the AKAP79/150 binding segment) of Ca_V1.2 channels preserves strong CDI (de Leon et al., 1995; Erickson et al., 2001; Crump et al., 2013). Similarly, short variants of Ca_V1.3 channels lacking the AKAP harboring distal carboxy termini (Marshall et al., 2011) also fully support CDI (Xu and Lipscombe, 2001; Shen et al., 2006; Yang et al., 2006; Bock et al., 2011). Altogether, it may be that calcineurin and AKAP79/150 are indirect and context-dependent modulators of CDI, rather than direct effector molecules (Budde et al., 2002).

Functional bipartition of CaM and selectivity for local/global Ca²⁺ sources

The calmodulatory mechanism for Ca²⁺ channels supports remarkable forms of Ca²⁺ decoding. This feature echoes earlier discoveries of “functional bipartition” of CaM in Paramecium (Kink et al., 1990; Saimi and Kung, 2002). Here, “under-excitable” behavioral strains had mutations only in the N-lobe of CaM, whereas “over-excitable” strains had mutations only in the C-lobe. Thus, one lobe of CaM can mediate signaling to one set of functions, whereas the other lobe signals to an alternative set of operations. It was unclear, however, whether this bipartition extended to mammals. The requirement that Ca²⁺ regulation of mammalian Ca²⁺ channels requires apoCaM preassociation permitted direct exploration of this question. Coexpressing channels with “half mutant” CaM molecules (CaM₁₂ and CaM₃₄, where only C- or N-terminal lobes, respectively, bind Ca²⁺) revealed that the individual lobes of CaM can evoke distinct components of channel regulation (Fig. 2 B). In most Ca_V1 channels, the C-lobe induces a kinetically rapid phase of CDI, whereas the N-lobe yields a slower component (Peterson et al., 1999; Yang et al., 2006; Dick et al., 2008). For Ca_V1.2 channels, early studies that utilized high intracellular Ca²⁺ buffering found only minimal N-lobe–mediated CDI (Peterson et al., 1999). However, when interrogated under low intracellular Ca²⁺ buffering, slow but recognizable N-lobe–mediated CDI of Ca_V1.2 channels emerges (Dick et al., 2008; Simms et al., 2013). Most strikingly, in Ca_V2.1 channels, the lobes of CaM produce opposing polarities of regulation: the C-lobe of CaM triggers a kinetically rapid CDF, whereas the N-lobe evokes a slower CDI (DeMaria et al., 2001; Lee et al., 2003). Ca_V2.2 and Ca_V2.3 channels manifest CDI triggered mainly by the N-lobe of CaM (Liang et al., 2003). Whether this bipartition orchestrates divergent classes of behavior in higher-order animals remains an intriguing and open question (Wei et al., 2003).Beyond simply splitting the Ca²⁺ signal, however, the calmodulation of mammalian Ca²⁺ channels revealed that the lobes of CaM can selectively decode Ca²⁺ in different ways. Hints as to this capability came from the differential sensitivity of calmodulation in various channels to Ca²⁺ buffering. Processes evoked by the C-lobe of CaM are invariably insensitive to introduction of strong Ca²⁺ buffering, whereas N-lobe–mediated processes in Ca_V2 channels can be eliminated by the same maneuver. Strong buffering would only spare Ca²⁺ signals near the cytoplasmic mouth of channels where strong point-source Ca²⁺ influx would overwhelm buffer capacity (Neher, 1986; Stern, 1992). This argues that local Ca²⁺ influx through individual channels triggers C-lobe signaling; in other words, the C-lobe of CaM exhibits a “local Ca²⁺ selectivity.” In contrast, a spatially global elevation of Ca²⁺ (present in the absence of strong Ca²⁺ buffering) is required for N-lobe signaling of Ca_V2 channels (DeMaria et al., 2001; Lee et al., 2003; Liang et al., 2003); thus, the N-lobe in this context exhibits a “global selectivity.” The existence of global selectivity is notable, given that local Ca²⁺ influx yields far larger Ca²⁺ increases near a channel than does globally sourced Ca²⁺ (Tay et al., 2012). One exception to this pattern is the local Ca²⁺ selectivity of the N-lobe component of CDI in Ca_V1.2/1.3 channels (see two paragraphs below). Corroboration of spatial Ca²⁺ selectivity is provided by the ability of single Ca_V1.2 channels to undergo CDI (Fig. 1, D–F). By definition, only a local Ca²⁺ source is present in the single-channel configuration; thus, the presence of CDI in this context indicates local Ca²⁺ selectivity. Additionally, single-channel records of Ca_V2.1 channels exhibit CDF (driven by C-lobe), but not CDI (triggered by N-lobe; Chaudhuri et al., 2007), which is consistent with the proposed differential selectivities for this channel. Fig. 2 B summarizes the arrangement of spatial Ca²⁺ selectivities according to the lobes of CaM.The mechanisms underlying these contrasting spatial Ca²⁺ selectivities can be interpreted as emergent behaviors of a system in which a lobe of apoCaM must transiently detach from a preassociation site before binding Ca²⁺, and where a Ca²⁺-bound lobe of CaM must associate with a channel effector site to mediate regulation (Tadross et al., 2008). When the slow Ca²⁺-unbinding kinetics of the C-lobe of CaM are imposed on this scenario, local Ca²⁺ sensitivity invariably results. In contrast, when the rapid Ca²⁺-unbinding kinetics of the N-lobe are interfaced with this architecture, global Ca²⁺ selectivity arises if the channel preferentially binds apoCaM versus Ca²⁺/CaM, as in the case of Ca_V2 channels (Fig. 2 B). When the channel preferentially interacts with Ca²⁺/CaM, local Ca²⁺ selectivity emerges, as in the context of Ca_V1 channels (Fig. 2 B). The relative roles of the affinities and kinetics of Ca²⁺ binding to the two lobes of CaM in mediating local and global Ca²⁺ selectivities have been considered at length elsewhere (Tadross et al., 2008).As an explicit example of how spatial Ca²⁺ selectivity of N-lobe regulation is specified, we consider the effects of an N-terminal spatial Ca²⁺-transforming element (NSCaTE) present only on the amino terminus of Ca_V1.2/1.3 channels (Fig. 2 A). NSCaTE is a Ca²⁺/CaM binding site whose presence enhances the aggregate channel affinity for Ca²⁺/CaM over apoCaM (Ivanina et al., 2000; Dick et al., 2008), endowing the N-lobe component of CDI of these channels with a largely local Ca²⁺ selectivity of N-lobe CDI in these channels. Elimination of the NSCaTE site in these channels, which tilts channels toward a preference for apoCaM, then switches their N-lobe CDI to a global profile. Conversely, donation of NSCaTE to Ca_V2 channels, which causes channels to favor Ca²⁺/CaM binding, then endows their N-lobe CDI with local Ca²⁺ selectivity (Dick et al., 2008). In this manner, the presence of NSCaTE can tune the spatial Ca²⁺ selectivity of N-lobe–mediated channel regulation. This overall arrangement of CaM interactions with a target molecule could endow numerous other regulatory systems with spatial Ca²⁺ selectivity.

Molecular basis of calmodulation

Elucidating the arrangement of apoCaM and Ca²⁺/CaM on Ca²⁺ channels is fundamental for the field, given the biological influence of calmodulation, and the importance of this system as an ion-channel regulatory prototype (Dunlap, 2007). This critical task, however, remains an ongoing challenge, given the >2,000 amino acids comprising the main pore-forming α₁ subunit alone, the ability of multiple peptide segments of the channel to bind CaM in vitro, and the formidable nature of obtaining atomic structures for these channels.The stoichiometry of CaM interaction with channels has been debated. Crystal structures of Ca²⁺/CaM complexed with portions of the carboxy tail CI region of Ca_V1.2 channels (Fallon et al., 2009; Kim et al., 2010) suggest that multiple CaM molecules may interact to produce the full spectrum of Ca²⁺ regulatory functions. Moreover, CaM can interact with multiple peptide segments of the channel (Ivanina et al., 2000; Tang et al., 2003; Zhou et al., 2005; Dick et al., 2008; Ben Johny et al., 2013). In contrast, covalent fusion of single CaM molecules to Ca_V1.2 channels suggests that only one CaM per channel suffices to elicit CDI (Mori et al., 2004). Moreover, live-cell FRET studies of the interactions between CaM and holochannels (Ca_V1.2) also point to a 1:1 CaM/channel ratio (Ben Johny et al., 2012). In the holochannel, the various CaM binding segments may be arranged in close proximity so as to allow the two lobes of CaM to preferentially interact with the distinct channel segments during Ca²⁺ regulation (Dick et al., 2008; Ben Johny et al., 2013). Recent biochemical studies of the NSCaTE element show that this segment preferentially interacts with a single lobe of CaM (N-lobe) at a time (Liu and Vogel, 2012). Furthermore, the NSCaTE element can interact also with Ca²⁺/CaM prebound to an IQ domain peptide (Taiakina et al., 2013). Thus, we favor the interpretation that only one CaM interacts within holochannels, although peptide fragments of Ca²⁺ channels may bind to multiple CaM molecules.The IQ domain (Fig. 3 A, blue circle) is critical for calmodulation. Mutations within this domain markedly modulate the Ca²⁺ regulation of Ca_V1.2 channels (Zühlke and Reuter, 1998; Qin et al., 1999; Zühlke et al., 1999, 2000; Erickson et al., 2003), Ca_V1.3 channels (Yang et al., 2006; Bazzazi et al., 2013; Ben Johny et al., 2013), Ca_V2.1 channels (DeMaria et al., 2001; Lee et al., 2003; Kim et al., 2008; Mori et al., 2008), and Ca_V2.2 and Ca_V2.3 channels (Liang et al., 2003). Furthermore, apoCaM preassociation with Ca_V1.2/1.3 channels requires the IQ domain (Erickson et al., 2001; Pitt et al., 2001; Erickson et al., 2003; Bazzazi et al., 2013; Ben Johny et al., 2013). Finally, Ca²⁺/CaM binds well to IQ-domain peptides of many Ca_V channels (Peterson et al., 1999; Zühlke et al., 1999; DeMaria et al., 2001; Pitt et al., 2001; Liang et al., 2003; Bazzazi et al., 2013; Ben Johny et al., 2013). All these results gave rise to the IQ-centric hypothesis shown in Fig. 3 A. Here, the IQ domain is important for both apoCaM preassociation (left) and Ca²⁺/CaM binding (effector configuration shown on the right). This IQ-centric paradigm has motivated efforts to resolve crystal structures of Ca²⁺/CaM complexed with IQ-domain peptides of various Ca_V1-2 channels (Fallon et al., 2005; Van Petegem et al., 2005; Kim et al., 2008; Mori et al., 2008), as shown in Fig. 3 (B and C). Throughout, the IQ peptide segment appears as an α-helical entity with N and C termini as labeled.Open in a separate windowFigure 3.Toward an atomic-level understanding of CDI. (A) IQ-centric view of calmodulation of Ca_V channels. In this mechanistic scheme, ApoCaM is preassociated with the IQ domain (blue). Upon Ca²⁺ binding, CaM rebinds the same IQ domain with a higher affinity, and subtle conformational rearrangements are presumed to trigger CaM-mediated channel regulation. (B, left) Crystal structure of Ca_V1.2 IQ domain peptide in complex with Ca²⁺/CaM (Protein Data Bank accession no. 2BE6). The IQ domain is colored blue. CaM is show in cyan (N-lobe, pale cyan; C-lobe, cyan). Ca²⁺ ions are depicted as yellow spheres. The key isoleucine residue (red) serves as a hydrophobic anchor for CaM. Ca²⁺/CaM adopts a parallel arrangement with the IQ domain in which the N-lobe binds closer to the amino terminus of the IQ domain and the C-lobe binds further downstream. (B, right) Crystal structure of Ca²⁺/CaM bound to mutant Ca_V1.2 IQ domain with alanines substituted for the key isoleucine-glutamine residues (accession no. 2F3Z). This double mutation abolishes CDI. Structurally, however, Ca²⁺/CaM hugs the mutant IQ domain in a similar conformation as to its interaction with the wild-type (left). (C) Crystal structure of Ca²⁺/CaM bound to Ca_V2.1 IQ domain (left, accession no. 3BXK; right, accession no. 3DVM). The IQ domain is shaded green, CaM in cyan. Ca²⁺/CaM was reported to bind to the Ca_V2.1 IQ domain in both parallel (left) and antiparallel (right) arrangements. The antiparallel arrangement in which the C-lobe of CaM binds upstream of IQ domain has led to speculation that CDF may result from this inverted polarity of Ca²⁺/CaM association. (D) Simplified configurations of the CaM/channel complex relevant for calmodulation (E, A, I_C, I_N, and I_CN). In configuration E, channels lack preassociated CaM and therefore cannot undergo CDI. Configuration A corresponds to channels bound to apoCaM, and thus capable of undergoing robust CDI. Ca²⁺ binding to the C-lobe and N-lobe of CaM leads to the inactivated configurations I_C and I_N, respectively. Ca²⁺ binding to both lobes of CaM then yields configuration I_CN, with both C and N lobes of CaM engaged toward CDI. This final transition may exhibit positive cooperativity as specified by the constant λ >> 1. Under endogenous levels of CaM, channels may reside in any of the five configurations. Strong coexpression of wild-type or mutant CaM restricts accessibility to various states. Reproduced with permission from Ben Johny et al., 2013.However, this viewpoint remains problematic in three regards. (1) The atomic structures of Ca²⁺/CaM bound to wild-type and mutant IQ peptides of Ca_V1.2 (Fig. 3 B) show that a central isoleucine (side chains explicitly shown) in the IQ element is deeply buried within the C-lobe of Ca²⁺/CaM (Van Petegem et al., 2005), and that alanine substitution at this site hardly changes structure (Fallon et al., 2005). Additionally, Ca²⁺/CaM dissociation constants for corresponding wild-type and mutant IQ peptides are nearly the same (Zühlke et al., 2000; Bazzazi et al., 2013). Why then would alanine substitution at this well-ensconced site influence the rest of the channel to blunt regulation (Fallon et al., 2005)? (2) For Ca_V1.2/1.3 channels, the N-lobe of Ca²⁺/CaM effector site appears to be an NSCaTE element (Fig. 3 A, oval) of the channel amino terminus (Dick et al., 2008; Tadross et al., 2008; Liu and Vogel, 2012), separate from the IQ element. (3) Crystal structures of Ca²⁺/CaM in complex with the IQ peptide of Ca_V2.1 channels show that CaM can adopt both a parallel (Mori et al., 2008) and an antiparallel configuration (Kim et al., 2008; Fig. 3 C, left and right, respectively). The apparent inversion in configuration of CaM binding to IQ domain between Ca_V1.2 and Ca_V2.1 has been proposed as a mechanism for the opposing polarity of Ca²⁺ regulation observed in the two channels (Kim et al., 2008; Minor and Findeisen, 2010). However, detailed structural analysis along with functional systematic alanine scanning mutagenesis and chimeric channel analysis argues that the C-lobe effector site resides at a site beyond the IQ module (Mori et al., 2008).A major concern with older IQ domain analyses is that the regulatory system was not considered conceptually as a whole, as shown in Fig. 3 D (drawn with specific reference to Ca_V1.3 channels; Ben Johny et al., 2013). Configuration E portrays channels lacking apoCaM. Such channels can open normally, but do not manifest CDI because Ca²⁺/CaM from bulk solution cannot efficiently access a channel in configuration E to induce CDI (Mori et al., 2004; Yang, P.S., M.X. Mori, E.A. Antony, M.R. Tadross, and D.T. Yue. 2007. Biophysical Journal abstracts issue. 1669-Plat; Liu et al., 2010; Findeisen et al., 2011). ApoCaM binding with configuration E gives rise to configuration A, where opening can also occur normally, but CDI can now take place. For CDI, Ca²⁺ binding to both lobes of CaM yields configuration I_CN, which underlies a fully inactivated channel with strongly reduced opening. For intermediate configurations, Ca²⁺ binding only to the C-lobe brings about configuration I_C, equivalent to a C-lobe inactivated channel; Ca²⁺ binding only to the N-lobe yields the N-lobe–inactivated arrangement (I_N). Of particular importance, ensuing entry into I_CN likely involves positively cooperative interactions specified by cooperativity factor λ >> 1.This scheme emphasizes several challenges for older analyses of the IQ domain, where CDI was mostly assessed with only endogenous CaM present. Results thus obtained would be ambiguous, because IQ-domain mutations could alter calmodulation via changes at multiple steps depicted in Fig. 3 D, whereas other interpretations largely attributed the effects to altered Ca²⁺/CaM binding with an IQ effector site. In contrast, IQ mutations could well weaken apoCaM preassociation and reduce CDI by favoring configuration E. Moreover, functional deficits caused by mutations that do attenuate interaction with one lobe of Ca²⁺/CaM may be masked by positively cooperative steps (λ in Fig. 3 D).To minimize these issues, CDI of Ca_V1.3 channels was recently characterized during strong coexpression with various mutant CaM molecules (Bazzazi et al., 2013; Ben Johny et al., 2013) as shown in Fig. 4 (A–C, top). For orientation, CDI of channels expressed with only endogenous CaM present is shown in the upper portion of Fig. 4 A. Strong CDI produces a rapid decay of whole-cell Ca²⁺ current (red trace), compared with the negligible decline of Ba²⁺ current (black trace). Because Ba²⁺ binds CaM poorly (Chao et al., 1984), the decline of Ca²⁺ versus Ba²⁺ current after 300 ms of depolarization quantifies CDI (CDI parameter plotted below in bar graphs). One can isolate the diamond-shaped subsystem lacking configuration E (Fig. 4 B, top) by leveraging mass action with strong coexpression of wild-type CaM (CaM_WT). Further simplification of CDI was obtained by strongly coexpressing channels with CaM₁₂ (Fig. 4 C, top), which empties configuration E by mass action, and forbids configurations I_N and I_CN. Thus, the isolated C-lobe component of CDI can be studied, with the signature rapid time course of current decay shown near the top of Fig. 4 C. Critically, this layout avoids interplay with cooperative λ steps in Fig. 3 D. Likewise, strongly coexpressing CaM₃₄ focuses on the slower N-lobe form of CDI, with attendant simplifications.Open in a separate windowFigure 4.Residue-level roadmap for calmodulation of Ca_V1.3 channels. Systematic alanine scanning mutagenesis of the entire CI domain of Ca_V1.3 channels. (A) CDI of wild-type and mutant Ca_V1.3 channels recombinantly expressed in HEK293 cells was measured with only endogenous CaM present. CDI observed here reflects properties of the entire system (stick figure) diagrammed in Fig. 3 D. Exemplar whole-cell current shows robust CDI for wild-type (WT) Ca_V1.3, reflecting their high affinity for apoCaM. Here and throughout, the vertical bar pertains to 0.2 nA of Ca²⁺ current (black); the Ba²⁺ current (gray) has been scaled ∼3-fold downward to aid comparison of decay kinetics. Horizontal bar, 100 ms. CDI is measured as the fractional reduction of Ca²⁺ current in comparison to Ba²⁺ at 300 ms (Ben Johny et al., 2013). Bottom, bar graph summary of CDI for alanine substitutions for indicated residues. CDI for WT channels, blue-green vertical line. Alanine substitutions in both EF hand regions and IQ domain resulted in diminished CDI. (B) Overexpression of CaM_WT isolates the behavior of the diamond-shaped subsystem. Such overexpression of CaM_WT should rescue CDI for mutations that weakened apoCaM preassociation. For WT Ca_V1.3, exemplar current shows that CDI is unaltered, as expected for a channel with high affinity for apoCaM. Bottom, bar graph summary of CDI for various mutant channels with CaM_WT overexpressed. Format is as in A. CDI is rescued for several mutations in the EF hand region (EF2) and the IQ domain, reflecting the preassociation of N-lobe and C-lobe of apoCaM. (C) Overexpression of CaM₁₂ isolates C-lobe CDI. Exemplar currents for WT channels show the fast C-lobe form of CDI. Bottom, bar graph summary shows the footprint of C-lobe CDI. Mutants in both the first EF hand and the IQ domain disrupted the C-lobe CDI (see the LGF loci in the EF hand regions and TFL and IQD loci in the IQ domain). The format is as in A. (D) Overexpression of CaM₃₄ isolates N-lobe CDI. N-lobe CDI was largely preserved by mutations in the CI region. Mutations that weakened N-lobe apoCaM preassociation exhibited enhanced N-lobe CDI (see EF2 segment). Reassuringly, alanine scanning mutagenesis of NSCaTE element resulted in specific disruption of N-lobe CDI of Ca_V1.3 channels (Tadross et al., 2008). Adapted with permission from Ben Johny et al. (2013) and Bazzazi et al. (2013).Accordingly, a new framework of CaM/channel configurations that underlie calmodulation could be deduced, as diagrammed in Fig. 5 (Ben Johny et al., 2013). For convenience, Fig. 4 compiles results from a systematic alanine scan covering the entire CI domain of the carboxyl tail of Ca_V1.3 channels (Bazzazi et al., 2013; Ben Johny et al., 2013), where substitution positions are denoted to the left of the bar graph in Fig. 4 A. The N-terminal end of the CI segment corresponds to the top of the graph, purple and blue-green regions indicate EF1 and EF2, and the lavender zone denotes the IQ domain. Electrophysiological characterization for each of the substitutions was performed for all the various subsystems (Fig. 4, A–D, top), and bar graphs plot the strength of CDI for the corresponding subsystems in each of the panels. The blue-green vertical lines reference the profile for wild-type channels. The results thus obtained were combined with CaM binding assays (not depicted) to identify meaningful structure–function correlations.Open in a separate windowFigure 5.Next-generation blueprint for CaM/CaBP regulation of Ca_V channels. (A) De novo molecular model of Ca_V1.3 CI region docked to apoCaM. The N-lobe of apoCaM is thought to preassociate with the PCI region (green), whereas the C-lobe binds to the IQ domain (blue). The NSCaTE segment (tan) is unoccupied. This model corresponds to configuration A in Fig. 3 D, where the channels are charged with an apoCaM and ready to undergo CDI. (B) Proposed model for the Ca²⁺ inactivated state of the channel. The N-lobe of Ca²⁺/CaM is shown binding to the NSCaTE segment based on a recent NMR structure (Protein Data Bank accession no. 2LQC). This configuration is believed to result in N-lobe CDI. The de novo molecular model shows C-lobe of Ca²⁺/CaM, the EF hand segment, and the IQ domain forming a tripartite complex resulting in C-lobe CDI. Overall, this model corresponds to the inactive configuration I_CN in Fig. 3 D. (C) Proposed allosteric mechanism of CaBP action. CaBP and apoCaM may dually bind the Ca_V1 channel. However, CaBP may allosterically inhibit CDI of Ca_V1 channels by interacting with the NT, III-IV loop, or PCI segment, thereby preventing Ca²⁺/CaM from reaching its effector configuration. (A–C) Adapted with permission from Yang et al. (2014).Under the regimen in Fig. 4 A, where all configurations are potentially accessible, diminished CDI by alanine substitutions occurred not only in the IQ domain, but upstream in the EF hand regions (between LGF and TLF residues). The latter effects strongly suggested that something outside the IQ domain was essential. Some of these deficits in CDI could be attributed to weakening of apoCaM preassociation with channels, because overexpressing wild-type CaM to depopulate configuration E (in Fig. 4 B) substantially recovered many of these CDI deficits. Binding assays of apoCaM with various CI segments confirmed this interpretation (Bazzazi et al., 2013; Ben Johny et al., 2013).Turning to potential deficits relating to diminished Ca²⁺/CaM action through channel effector sites, Fig. 4 (C and D) separately interrogates the C-lobe and N-lobe components of CDI. Importantly, isolating the C- and N-lobe components minimizes masking of mutational effects by positive cooperativity, allowing CDI deficits to be more sensitively observed in this regimen. That said, Fig. 4 C displays C-lobe CDI deficits in the EF hand region (strongest for LGF substitution), as well as in the IQ domain (with the strongest effect upon substitution at the central isoleucine). These outcomes support a hypothesis where the Ca²⁺/CaM effector configuration for the C-lobe component of CDI involves both of these functional hotspots, as developed two paragraphs below. For the N-lobe component of CDI (Fig. 4 D), no appreciable deficits were observed, as would be expected if interaction with the channel NSCaTE element (in the channel amino terminus) serves as the effector element.Fig. 5 A displays a proposed configuration A for apoCaM interaction with the channel. This includes a homology model of the apoCaM C-lobe complexed with the IQ domain (blue), based on analogous Na_V structures (Chagot and Chazin, 2011; Feldkamp et al., 2011). The portrayal of the apoCaM N-lobe incorporates ab initio structural prediction of the CI domain, with two vestigial EF hands (EF), and a protruding helix (preIQ subelement). The EF hand module (EF) resembles the structure of a homologous Na_V segment (Chagot et al., 2009; Wang et al., 2012), and the preIQ helix resembles a helical segment observed in crystal structures of analogous Ca_V1.2 peptides (Fallon et al., 2009; Kim et al., 2010). The atomic structure of the apoCaM N-lobe (1CFD) was interfaced with shape-complementarity docking algorithms. Regarding the proposed interaction interface of the N-lobe with the channel EF domain, we note that alanine substitution therein produced a curious enhancement of N-lobe CDI seen with certain alanine substitutions, especially in the EF region (Fig. 4 D). This feature is interesting because weakening of channel interaction with the N-lobe of apoCaM would be predicted to strengthen CDI produced by the N-lobe of Ca²⁺/CaM (Tadross et al., 2008).Fig. 5 B displays a proposed arrangement for the Ca²⁺/CaM-bound configuration I_CN. The N-lobe complex with NSCaTE is an NMR structure (Liu and Vogel, 2012). An alternative ab initio model of the PCI is computationally docked with the C-lobe of Ca²⁺/CaM (Protein Data Bank accession no. 3BXL) and IQ module, which together form a ternary complex. This ternary arrangement is consistent with the importance of both IQ and EF domains for C-lobe CDI (Fig. 4 C). Intriguingly the C-lobe configuration resembles a rather canonical CaM–peptide complex, where the channel EF module contributes a surrogate lobe of CaM. Importantly, assays of Ca²⁺/CaM binding to the IQ segment alone would correlate poorly with functional effects relating to the ternary complex, as experimental studies observe (Bazzazi et al., 2013; Ben Johny et al., 2013). Overall, the proposed framework in Fig. 5 (A and B) may aid future structural biology and structure–function work. In addition, this scheme of CaM exchange within its target molecule may generalize beyond the Ca²⁺ channel family.

Biological consequences and prospects for new disease therapies

The biological consequences of Ca²⁺ regulation by CaM promise to be wide ranging and immense. In the heart, elimination of Ca_V1.2 CDI by means of dominant-negative CaM elicits marked prolongation of ventricular action potential duration (APD), implicating CDI as a dominant control factor in specifying APD (Alseikhan et al., 2002; Mahajan et al., 2008). As APD is one of the main determinants of electrical stability and arrhythmias in the heart, pharmacological manipulation of such regulation looms as a future antiarrhythmic strategy (Mahajan et al., 2008; Anderson and Mohler, 2009). Most recently, genome-wide linkage analysis in humans has uncovered heritable and de novo CaM mutations as the probable cause of several cases of catecholaminergic polymorphic ventricular tachycardia (CPVT), with altered CaM-ryanodine receptor function implicated as a major contributing factor (Nyegaard et al., 2012). Whole-exome sequencing has revealed an association between three de novo missense CaM mutations and severe long-QT (LQT) syndrome with recurrent cardiac arrest (Crotti et al., 2013), quite plausibly acting via disruption of Ca_V1.2/1.3 channel CDI (Limpitikul et al., 2014).In brain, state-of-the-art analysis of genome-wide single-nucleotide polymorphisms (SNPs) has identified Ca_V channels as a major risk factor for several psychiatric disorders (Cross-Disorder Group of the Psychiatric Genomics Consortium, 2013). More specifically to calmodulation, Ca_V1.3 channels constitute a prominent Ca²⁺ entry portal into pacemaking and oscillatory neurons (Bean, 2007), owing to the more negative voltages required to open these ion channels. These channels are subject to extensive alternative splicing (Hui et al., 1991; Xu and Lipscombe, 2001; Shen et al., 2006; Bock et al., 2011; Tan et al., 2011) and RNA editing (Huang et al., 2012) in the carboxy tail, in ways that strongly modulate the strength of CDI (Shen et al., 2006; Liu et al., 2010; Huang et al., 2012). This fine tuning of CDI appears to be important for circadian rhythms (Huang et al., 2012). From the specific perspective of disease, Ca_V1.3 channels contribute a substantial portion of Ca²⁺ entry into substantia nigral neurons (Bean, 2007; Chan et al., 2007; Puopolo et al., 2007; Guzman et al., 2009), which exhibit high-frequency pacemaking (Chan et al., 2007) that drives dopamine release important for movement control. Notably, loss of these neurons is intimately related to Parkinson’s disease, and Ca²⁺ disturbances and overload are critical to this neurodegeneration (Bezprozvanny, 2009; Surmeier and Sulzer, 2013). Accordingly, an attractive therapeutic possibility for Parkinson’s disease involves discovery of small molecules that selectively down-regulate the opening of Ca_V1.3 versus other closely related Ca²⁺ channels (Kang et al., 2012). Thus, understanding the mechanisms underlying the modulation of Ca_V1.3 CDI is crucial, particularly to furnish specific molecular interfaces as targets of rational screens for small-molecule modulators.The fundamental mechanism of action of a natural modulator of Ca_V1.3 CDI may be especially relevant. In particular, recent discoveries identify Ca²⁺-binding proteins (CaBPs), a family of CaM-like brain molecules (Haeseleer et al., 2000) that may also bind Ca_V channels and other targets (Yang et al., 2002; Kasri et al., 2004). Like CaM, CaBPs are bilobed, with each lobe containing two EF hand Ca²⁺ binding motifs (Haeseleer et al., 2000), and a recent crystal structure shows overall similarity to CaM (Findeisen and Minor, 2010). However, whereas all four EF hands bind Ca²⁺ in CaM, one lobe is nonfunctional in CaBPs. Coexpressing CaBPs with Ca_V1 channels eliminates their CDI, thereby potentially influencing diverse biological processes (Zhou et al., 2004; Yang et al., 2006). Some have argued that CaBPs may simply compete for apoCaM on channels (Lee et al., 2002; Findeisen et al., 2013; Oz et al., 2013). More recent data argue for a different mechanism (Fig. 5 C), in which CaBP and apoCaM can both preassociate with the channel (Yang et al., 2014). In particular, by binding to channel regions that overlap Ca²⁺/CaM effector loci, CaBPs may preemptively retard the ability of Ca²⁺/CaM to reach its effector configuration (Fig. 5 B), allowing low concentrations of CaBP molecules in the CNS to still exert functional effects in the presence of much higher CaM levels (Yang et al., 2014). If this scheme is correct, the basic mechanisms of CaBP action may have implications for drug discovery, in that small molecules that target channel interaction surfaces for CaBP and/or Ca²⁺/CaM may exert potent modulatory actions.In conclusion, this overview of the calmodulation of voltage-gated Ca²⁺ channels highlights an impressive synergy among elegant molecular regulatory mechanisms, vital biological functions, and pathogenesis and potential therapy. As such, this field promises considerable mystery, enrichment, and enlightenment in the years ahead.     

Interplay between Velocity and Travel Distance of Kinesin-based Transport in the Presence of Tau

Although the disease-relevant microtubule-associated protein tau is known to severely inhibit kinesin-based transport in vitro, the potential mechanisms for reversing this detrimental effect to maintain healthy transport in cells remain unknown. Here we report the unambiguous upregulation of multiple-kinesin travel distance despite the presence of tau, via decreased single-kinesin velocity. Interestingly, the presence of tau also modestly reduced cargo velocity in multiple-kinesin transport, and our stochastic simulations indicate that the tau-mediated reduction in single-kinesin travel underlies this observation. Taken together, our observations highlight a nontrivial interplay between velocity and travel distance for kinesin transport, and suggest that single-kinesin velocity is a promising experimental handle for tuning the effect of tau on multiple-kinesin travel distance.Conventional kinesin is a major microtubule-based molecular motor that enables long-range transport in living cells. Although traditionally investigated in the context of single-motor experiments, two or more kinesin motors are often linked together to transport the same cargo in vivo (1–4). Understanding the control and regulation of the group function of multiple kinesins has important implications for reversing failure modes of transport in a variety of human diseases, particularly neurodegenerative diseases. Tau is a disease-relevant protein enriched in neurons (5,6). The decoration of microtubules with tau is known to strongly inhibit kinesin transport in vitro (7–9), but how kinesin-based transport is maintained in the presence of high levels of tau, particularly in healthy neurons, remains an important open question. To date, no mechanism has been directly demonstrated to reverse the inhibitory effect of tau on kinesin-based transport. Here we present a simple in vitro study that demonstrates the significant upregulation of multiple-kinesin travel distance with decreasing ATP concentration, despite the presence of tau.This investigation was motivated by our recent finding that single-kinesin velocity is a key controller for multiple-kinesin travel distance along bare microtubules (10). The active stepping of each kinesin motor is stimulated by ATP (11), and each kinesin motor remains strongly bound to the microtubule between successive steps (10,11). As demonstrated for bare microtubules (10), with decreasing ATP concentrations, each microtubule-bound kinesin experiences a decreased stepping rate per unit time and spends an increased fraction of time in the strongly bound state; additional unbound kinesins on the same cargo have more time to bind to the microtubule before cargo travel terminates. Thus, reductions in single-kinesin velocity increase the probability that at least one kinesin motor will remain bound to the microtubule per unit time, thereby increasing the travel distance of each cargo (10). Because this effect only pertains to the stepping rate of each individual kinesin and does not address the potential presence of roadblocks such as tau on the microtubules, we hypothesized in this study that single-kinesin velocity may be exploited to relieve the impact of tau on multiple-kinesin travel distance.We focused our in vitro investigation on human tau 23 (htau23, or 3RS tau), an isoform of tau that exhibits the strongest inhibitory effect on kinesin-based transport (7–9). Importantly, htau23 does not alter the stepping rate of individual kinesins (7,9), supporting our hypothesis and enabling us to decouple single-kinesin velocity from the potential effects of tau. We carried out multiple-kinesin motility experiments using polystyrene beads as in vitro cargos (8,10), ATP concentration as an in vitro handle to controllably tune single-kinesin velocity (10,11), and three input kinesin concentrations to test the generality of potential findings for multiple-kinesin transport. Combined with previous two-kinesin studies (10,12), our measurements of travel distance (Fig. 1 A) indicate that the lowest kinesin concentration employed (0.8 nM) corresponds to an average of ∼2–3 kinesins per cargo. Note that in the absence of tau, the observed decrease in bead velocity at the higher kinesin concentrations (Fig. 1 A) is consistent with a recent in vitro finding (13). At 1 mM ATP, htau23 reduced kinesin-based travel distance by a factor of two or more (Fig. 1, A and B). This observation is in good agreement with previous reports (7,8).Open in a separate windowFigure 1Distributions of multiple-kinesin travel distances measured at three experimental conditions, to verify the effect of tau (A and B) and to investigate the impact of single-kinesin velocity on the tau effect (B and C). Shaded bars at 8.7 μm indicate counts of travel exceeding the field of view. The mean travel distance (d; ± standard error of mean, SEM), sample size (n), and corresponding mean velocity (v; ± SEM) are indicated. MT denotes microtubule. Mean travel distance increased substantially at 20 μM ATP (C), despite the presence of htau23. This effect persisted across all three kinesin concentrations tested (left to right).Consistent with our hypothesis, reducing the available ATP concentration to 20 μM increased the multiple-kinesin travel distance by >1.4-fold for all three input kinesin concentrations (Fig. 1, B and C), despite the presence of htau23. The corresponding reduction in single-kinesin velocity with decreasing ATP concentration (10,11) is reflected in the ∼3.4-fold reduction in the measured bead velocities (Fig. 1, B and C). Therefore, the strong negative relationship between single-kinesin velocity and multiple-kinesin travel distance occurs not only for bare microtubules (10), but also for tau-decorated microtubules.What causes the observed increase in travel distance at the lower ATP concentration (Fig. 1, B and C)? In addition to the mechanism discussed above for the case of bare microtubules (10), an intriguing mechanism was suggested by recent studies of tau-microtubule interactions in which htau23 was observed to dynamically diffuse along microtubule lattices (14,15): reducing the stepping rate of a microtubule-bound kinesin may effectively increase the probability that a tau roadblock can diffuse away before the kinesin takes its next step.Perhaps surprisingly, although htau23 does not impact single-kinesin velocity (7,9), we observed a modest reduction in the average velocity of multiple-kinesin transport in experiments using tau-decorated microtubules (Fig. 1, A and B). This decreased velocity reflects a substantially larger variance in the instantaneous velocity for bead trajectories in the presence of htau23 (see Fig. S1 in the Supporting Material), as quantified by parsing each bead trajectory into a series of constant-velocity segments using a previously developed automatic software incorporating Bayesian statistics (16).To test the possibility that single-kinesin travel distance impacts multiple-kinesin velocity, we performed stochastic simulations (see the Supporting Material) that assumed N identical kinesin motors available for transport and included kinesin’s detachment kinetics (17). Previously, this model successfully captured multiple-dynein travel distances in vivo using single-dynein characteristics measured in vitro (18). In this study, we introduced one (and only one) free parameter to reflect the probability of each bound kinesin encountering tau at each step. When encountering tau, each kinesin has a 54% probability of detaching from the microtubule (interpolated from Fig. 2A of Dixit et al. (7)); the undetached kinesin is assumed to remain engaged in transport and completes its step along the microtubule despite the presence of tau.Remarkably, our simple simulation suggested that the tau-mediated reduction in single-kinesin travel is sufficient to reduce multiple-kinesin velocity (Fig. 2 A). The majority of the velocity decrease is predicted to occur at the transition from single-kinesin to two-kinesin transport (Fig. 2). Further decreases in cargo velocity with increasing motor number are predicted to be modest and largely independent of tau (Fig. 2 B). The results of our simulation remain qualitatively the same when evaluated at two bounds (40 and 65%) encompassing the interpolated 54% probability of kinesin detaching at tau (see Fig. S2).Open in a separate windowFigure 2Stochastic simulations predict a tau-dependent reduction in multiple-kinesin velocity, assuming that the only effect of tau protein is to prematurely detach kinesin from the microtubule (or, to reduce single-kinesin travel distance). (A) Average velocity of cargos carried by the indicated number of kinesins was evaluated at 1 mM ATP, and for four probabilities that a kinesin may encounter tau at each step. Mean velocity was evaluated using 600 simulated trajectories for all simulation conditions. Error bars indicate SEM. (B) Change in cargo velocity with each additional kinesin (ΔVel/kinesin) as a function of tau-encounter probability. These values were calculated from cargo velocities shown in panel A. Error bars indicate SEM.We note that our simple simulations do not consider the possibility that kinesin may pause in front of a tau roadblock, as previously reported in Dixit et al. (7). We omitted this consideration because the interaction strength between kinesin and the microtubule in such a paused state is unknown. In a multiple-motor geometry, could a paused kinesin be dragged along by the other motors bound to the same cargo? Could a tau roadblock be forcefully swept off the microtubule surface by the collective motion of the cargo-motor complex? Significant experimental innovations are necessary to specifically address these questions in future multiple-motor assays and to guide modeling efforts. Nonetheless, our simple simulation demonstrates that reducing single-kinesin travel distance is sufficient to decrease multiple-kinesin travel distance.Taken together, our observations highlight a nontrivial interplay between velocity and travel distance for kinesin-based transport in the presence of tau. We uncover a previously unexplored dual inhibition of tau on kinesin-transport: in addition to limiting cargo travel distance, the tau-mediated reduction in single-kinesin travel distance also leads to a modest reduction in multiple-kinesin velocity. We provide what we believe to be the first demonstration of the unambiguous upregulation of multiple-kinesin travel distance despite the presence of tau, via reducing single-kinesin velocity, suggesting a mechanism that could be harnessed for future therapeutic interventions in diseases that result from aberrant kinesin-based transport.