DNA,Flexibly Flexible

Investigators have constructed dsDNA molecules with several different base modifications and have characterized their bending and twisting flexibilities using atomic force microscopy, DNA ring closure, and single-molecule force spectroscopy with optical tweezers. The three methods provide persistence length measurements that agree semiquantitatively, and they show that the persistence length is surprisingly similar for all of the modified DNAs. The circular dichroism spectra of modified DNAs differ substantially. Simple explanations based on base stacking strength, polymer charge, or groove occupancy by functional groups cannot explain the results, which will guide further high-resolution theory and experiments.Real double-stranded DNA molecules differ from the idealized zero-Kelvin A, B, and Z forms. They can adopt deformed average conformations, as in bent A-tract DNA or protein-DNA complexes. The path of the DNA helix axis also varies due to thermal energy, so at very long lengths DNA behaves as a random coil. The term “long lengths” is relative to the persistence length P of the wormlike chain model. P is the average offset of the end of a chain along its initial direction, or alternatively the length over which the unit vectors μ1→ and μ2→ tangent to the helix axis lose colinearity according to〈μ1→⋅μ2→〉=〈cosθ〉=e−d12/P,where d₁₂ is the contour length from point 1 to point 2, as in Fig. 1. P can be measured by hydrodynamics (1), atomic force microscopy (AFM) (2), DNA ring closure (3) or protein-DNA looping (4), tethered particle microscopy (5), or single-molecule optical tweezers experiments (6). The long-range loss of memory of DNA direction grows out of local variations in the helix axis direction specified by roll, tilt, and twist angles that parameterize changes in the helix axis direction. For harmonic bending potentials, the bending persistence length is related to roll and tilt according toσroll2+σtilt2=2ℓ/P,where ℓ = 3.4 Å, so for P ∼ 50 nm (147 bp) the average standard deviations in the roll and tilt angles σ_roll and σ_tilt are ∼4.7°, although in real DNA, roll varies more than tilt. Similar relationships hold for twist flexibility (7).Open in a separate windowFigure 1The base modifications studied by Peters et al. (13,14) affect both Watson-Crick hydrogen bonding and groove occupancy. They used AFM, DNA ring closure, single-molecule force spectroscopy, and circular dichroism spectroscopy (not shown) to characterize the resulting changes in bending and twisting flexibility. DNA molecules are not shown to scale. To see this figure in color, go online.DNA flexibility can be studied at contour length scales from Ångstroms to microns. Flexibility at the atomic scale accessed by nuclear magnetic resonance, x-ray crystallography, cryo-electron microscopy, and molecular dynamics simulations (8) refers to many aspects of conformational variability. One active thread of research at this scale concerns interconversion among helical forms, base flipping, DNA kinking, changes in backbone torsion angles, and the sequence dependence of all of these local properties. Local fluctuations in the basepair roll, tilt, and twist angles do seem to predict the correct long-range behavior (9). A second thread asks whether the wormlike chain model holds at DNA lengths shorter than P (2,10); the active controversy concerning enhanced bendability at short lengths has recently been reviewed by Vologodskii and Frank-Kamenetskii (11). A third thread asks whether we can understand the underlying biophysical causes of long-range DNA flexibility. These presumably include base stacking, electrostatic repulsion along the backbone, changes in the counterion atmosphere (12), occupancy of the major and minor grooves by functional groups, conformational entropy, the strength of Watson-Crick hydrogen bonding, and water structure. Helical polymorphisms and the junctions between polymorphs presumably affect the sequence dependence of the persistence length.Peters et al. (13,14) have attempted to understand bending and twisting flexibility by characterizing a variety of modified nucleic acids using DNA ring closure, AFM, and optical tweezer methods, sketched in Fig. 1. In previous work (13), they used ring closure to show that major groove substituents that alter the charge on the polymer do not have substantial effects on the bending persistence length, and that the effects were not correlated in an obvious way to the stacking propensity of the modified bases. The work described in this issue of the Biophysical Journal (14) uses all three methods to demonstrate that DNA with 2-amino-adenosine (a.k.a., 2,6-diaminopurine) substituted for adenosine has an increased persistence length, whereas inosine substitution for guanosine reduces the persistence length, as would be expected if groove occupancy (or the number of Watson-Crick hydrogen bonds) affects flexibility. However, the authors did one experiment too many—when they measured the effects of the earlier major groove substituents (13) using AFM, the correlation with groove occupancy disappeared. This could be because changes in helical geometry, as evidenced by the circular dichroism spectroscopy also reported in the article, alter the grooves sufficiently to prevent a straightforward connection to flexibility.The magnitude of the effect of base modifications on P is the largest for the optical tweezers and the smallest for DNA ring closure, showing that no more than one of the experiments is perfect. The Supporting Material for both articles (13,14) offers valuable resources for the careful evaluation of experimental results and possible sources of error within and between experiments. For example, the DNA lengths and the ionic conditions required by the different methods differ. Ring closure results depend critically on the purity of the DNA and appropriate ligation conditions. Analysis of AFM results averaged several different statistical measures of decaying angular correlations and end-to-end distance, which did not individually always agree. In force spectroscopy there are variations in the bead attachment for each molecule, errors in the stretch modulus can affect the measured persistence length, force can induce DNA melting, and very few molecules can be observed. Rare kinking events proposed to explain enhanced bendability should affect the cyclization experiment most markedly; no evidence for enhanced flexibility was seen. Finally, Peters et al. (14) have observed that DNA twist and twisting flexibility seem to be more sensitive than the persistence length to base modifications.Taken as a whole, this extremely thorough series of experiments shows that we still do not understand the fundamental origins of the remarkable stiffness of double-stranded DNA. There may be compensating effects that make the dissection difficult. For example, changing the charge on the polymer may induce a corresponding adjustment in the counterion condensation atmosphere, leading to a relatively constant residual charge. Groove substituents that enhance basepair stability could enhance bendability for steric reasons. Stacking thermodynamics may not change very much for the very small bend angles at any individual basepair. Locally stiff regions may introduce nearby junctions that are flexible.The stiffness of DNA relative to other biopolymers inspired the development of DNA nanotechnology (although that field has adopted bridged synthetic constructs that are even more rigid). Further research on the biophysics, and specifically the long-range mechanical properties of DNA, will be essential as we build better models of DNA in the cell, which has evolved many proteins that act to increase apparent flexibility. The various aspects of DNA flexibility influence the protein-DNA complexes that mediate DNA’s informational role, the induction of and responses to supercoiling used for long-range communication among sites (15), and chromosome structure and genome organization.     

The Physical Foundation of Vasoocclusion in Sickle Cell Disease

The pathology of sickle cell disease arises from the occlusion of small blood vessels because of polymerization of the sickle hemoglobin within the red cells. We present measurements using a microfluidic method we have developed to determine the pressure required to eject individual red cells from a capillary-sized channel after the cell has sickled. We find that the maximum pressure is only ∼100 Pa, much smaller than typically found in the microcirculation. This explains why experiments using animal models have not observed occlusion beginning in capillaries. The magnitude of the pressure and its dependence on intracellular concentration are both well described as consequences of sickle hemoglobin polymerization acting as a Brownian ratchet. Given the recently determined stiffness of sickle hemoglobin gels, the observed obstruction seen in sickle cell disease as mediated by adherent cells can now be rationalized, and surprisingly suggests a window of maximum vulnerability during circulation of sickle cells.Human capillaries are narrower than the erythrocytes they convey. In sickle cell disease, red cells can become rigid in those capillaries, because the hemoglobin inside the red cell will aggregate into stiff polymers. This happens once the molecules deliver their oxygen, and led to the long-held view that capillary occlusion was central to the pathophysiology of the disease (1,2). This was challenged when microscopic study of animal model tissues perfused with sickle blood revealed blockages that began further downstream, in the somewhat larger venules (3–5), at the site of adherent red or white cells which diminished the vessel lumen without fully obstructing the flow. Yet no rationale has been presented for the failure of the prior assumption of capillary blockage. Microfluidic methods (6) are ideally suited to discover why cells don’t get stuck in the capillaries, yet occlude subsequent vessels, and we have constructed a system to address this question. Our measurements show that the pressure differences across capillaries in vivo can easily dislodge a cell sickled within a capillary, giving an experimental answer to the question of why sickled cells don’t stick in capillaries. It turns out that the pressure a cell can withstand is quantitatively explained by the Brownian ratchet behavior of sickle hemoglobin polymerization.We constructed single-cell channels in transparent polydimethylsiloxane, with a cross section (1.5 μm × 4 μm) that is smaller than the resting diameter of red cells (Fig. 1). These channels are much narrower than those that have been employed in other recent studies of the sickling process (7,8), and they resemble human capillaries in permitting only one cell at a time to pass through them. We used a laser photolysis method to create ligand free (deoxygenated) cells, and this requires that the hemoglobin bind CO, which can then be readily removed by strong illumination, in contrast to bound O₂ which is released with far lower efficiency than CO. The microfluidic chips were enclosed in a gas-tight chamber flushed with CO to avoid introduction of oxygen and keep the cells fully ligated before photolysis. The profiles of the channels were confirmed by microscopic observation. To confirm that liquid did not pass around the cells when they were trapped in the channels, fluorescent beads were introduced into some cell solutions. The beads did not pass the cells, nor did they approach the cell when it was occluded, verifying that no significant flow occurred around the cell when it was stuck.Open in a separate windowFigure 1An erythrocyte enters a channel (moving left to right) and is positioned in the center, where it will be photolyzed. The channel cross section is 1.5 μm × 4 μm, smaller than a resting red cell diameter.Optical measurements were carried out on a microspectrophotometer constructed on an optical table. The system employed ×32 LWD objectives (Leitz, Wetzlar, Germany), which were autofocused during collection of absorption spectra to minimize aberrations. Spectra were obtained using a series 300 camera (Photometrics, Tucson, AZ); video imaging was done with a high-speed camera (Photron, San Diego, CA). Photolysis was provided by a 2020 Argon Ion laser (Spectra Physics, Houston, TX). Sickle cells were obtained from patients at the Marian Anderson Sickle Cell Center at St. Christopher''s Hospital for Children, Philadelphia, PA by phlebotomy into EDTA-containing tubes. The blood was centrifuged at 5°C at 1200g for 4 min, and then the pellet was washed 4× with 15 volumes of buffer (120 mM NaCl, 2 mM KCl, 10 mM dibasic Na Phosphate, 7 mM monobasic Na Phosphate, 3.4 mM Na Bicarbonate, and 6 mM Dextrose) by repeated suspension and centrifugation at 30g for 4 min. This minimizes fibrinogen and platelets in the final suspension, to insure that these studies are controlled by the mechanical properties of the cells themselves.Our experiment began by parking a cell in the center of a channel (Fig. 1). The cell, its hemoglobin, and the microchannel environment all were saturated with CO. Because the thickness of the channel is known, we were able to determine the hemoglobin concentration inside the cell from its absorption spectrum (Fig. 2 A). Steady-state laser illumination then removed the CO, allowing the hemoglobin to polymerize, in which condition it remained while the laser was kept on. Removal of CO was confirmed by observing the spectral difference between COHb and deoxyHb. Photolysis of COHb generates negligible heating (9–11). During illumination, hydrostatic pressure was applied until the cell broke free.Open in a separate windowFigure 2(A) Absorption of the cell (points), fit to a standard spectrum (9). (B) Pressure to dislodge a cell sickled in the microchannel, as a function of intracellular concentration. Note that typical intracellular concentrations are ∼32 g/dL. (Line) Brownian-ratchet theory described in the text. The coefficient of friction (0.036) is within the observed range, and is the only parameter varied.The magnitude of the dislodging pressure, measured by simple height difference between input and output cell reservoirs, is shown in Fig. 2 B. The pressure needed to dislodge the cell increased with increasing intracellular Hb concentration, implying that an increased mass of polymerized hemoglobin is more difficult to dislodge. A clear concentration threshold for capture is apparent. While there is a well-known solubility below which polymers cannot form (18.5 g/dL for the 22°C of this experiment (12)), the threshold here is significantly higher.Central to explaining these observations is a Brownian ratchet mechanism (13) which derives from the metastable nature of this polymerization process. Unless disrupted, as by centrifugation, polymerization in sickle hemoglobin terminates before the thermodynamic limit of monomer solubility is reached (14,15). This arises from the fact that polymers only grow at their ends, which are easily occluded in the dense mass of polymers that form. This end obstruction leaves the system in a metastable state and fluctuations accordingly provide polymers with space into which they can incrementally grow. This Brownian ratchet has been shown to lead to dramatic fiber buckling when individual fibers are isolated in sickle cells (16). The force can be simply expressed as f = (kT/δ) ln S(c), where k is Boltzmann’s constant, T the absolute temperature, δ the net spatial elongation from addition of a single monomer, and S is the supersaturation of the solution when the metastable limit is reached, at monomer concentration c. In this calculation, c is taken as the terminal concentration, computed from our empirical finding (15) that in this metastable system the amount of polymerized hemoglobin Δ is Δ(∞) = 2/3 (c_o-c_s), rather than the expected thermodynamic limit c_o-c_s, where c_o is the initial concentration and c_s is the solubility.For determining the net force, the total number of fibers must be known, and can be calculated based on the double nucleation mechanism (17) which has been quantitatively successful in describing polymerization. The concentration of polymers [p(t)] initially grows exponentially, described by[p(t)]=(AB2J)exp(Bt),where A and B are parameters related to nucleation, and J is the polymer elongation rate, as described in Ferrone et al. (17). Because A and B are both extremely concentration-dependent (9), they will drop dramatically once monomers begin to add to polymers in any significant numbers, and thereby diminish the remaining monomer pool. Thanks to the extreme concentration dependence of the reaction, this rapidly shuts off further polymerization. This happens at approximately the 10th time (the time when the reaction has reached 1/10 of its maximum). Thus, the [p(t_1/10)] ≈ [p(∞)]. Moreover, at one-tenth of the reaction,Δ(t1/10)=12Aexp(Bt1/10)=Δ(∞)10,and thus[p(∞)]=(BJ)(Δ(∞)10)=(BJ)((co−cs)15).For computing the number of fibers, the volume of the cell was taken as 90 μm³. This calculation shows, as expected, that the number of polymers in the cell is highly concentration-dependent, and very few fibers are produced at concentrations just above solubility, but the number grows sharply as concentration rises. This is the main contribution to the threshold in holding force shown by the data.With the force per fiber, and the total number of fibers, the net force against the wall is known. With a coefficient of friction, this reveals the force that a trapped cell can withstand. If the force is divided by the cross-sectional area across which the force is applied, we get a prediction of the dislodging pressure, which can be compared to the data. For a quantitative comparison with the results, two further corrections, of order unity, were applied. Because only normal force will contribute to friction, the calculated force was determined by integrating cos θ. This integration is not over all angles (π) because of the possibility that large incidence angles of the fibers against the wall will lead to fiber runaway (18). Therefore, the integration described is taken to the runaway threshold, here ∼1 rad. Finally, it is necessary to assign a coefficient of friction. Known values span the range of 0.03–0.06 (19). We therefore selected a value within the range, 0.036, as the best match for the data. The predicted pressures match the measurements well, as the line in Fig. 2 B shows.Because the flow resistance is comparable for red cells traversing glass channels and endothelial-lined capillaries (20), we conclude that in vivo the pressures a sickled cell inside a capillary can withstand are no more than hundreds of Pa. This is significantly smaller than typical arteriovenous pressure differentials that have been measured, which range from 0.7 kPa (in hamster skin (21)) to 7.9 kPa (in rat mesentery (22)).Our measurements coupled with recent determination of the stiffness of sickle hemoglobin gels (23) provide the missing physical basis for the processes of vasoocclusion seen in ex vivo tissue and animal models of sickle cell disease, arguing that these observations indeed represent fundamental behavior of sickle cell disease. We now understand this behavior in terms of three possible outcomes, all intimately connected with kinetics:

• 1.Certain escape: A cell that does not polymerize until after passing the obstruction can reach the lungs where it reoxygenates and resets its polymerization clock.
• 2.Possible escape: A cell that polymerizes within the capillary will assume an elongated sausage shape. The forces that it can exert against the wall cannot hold it there, and it will emerge into the postcapillary venule. There it has some chance of passing a subsequent obstruction, though it might also obstruct flow were it to rotate before reaching the adherent cell, so as to present its long dimension to the reduced space it must traverse.
• 3.Certain occlusion: A cell that does not polymerize in the capillary reassumes a larger diameter as soon as it escapes. If the cell then polymerizes before it encounters a cell attached to the venule wall, this rigidified cell will not be able to squeeze past the adherent cell, because that kind of deformation takes MPa (23). This would precipitate the type of blockage that is observed. This suggests that there is a window of greatest vulnerability, toward which therapies might be addressed.

     

The tubulin code: Molecular components,readout mechanisms,and functions

Microtubules are cytoskeletal filaments that are dynamically assembled from α/β-tubulin heterodimers. The primary sequence and structure of the tubulin proteins and, consequently, the properties and architecture of microtubules are highly conserved in eukaryotes. Despite this conservation, tubulin is subject to heterogeneity that is generated in two ways: by the expression of different tubulin isotypes and by posttranslational modifications (PTMs). Identifying the mechanisms that generate and control tubulin heterogeneity and how this heterogeneity affects microtubule function are long-standing goals in the field. Recent work on tubulin PTMs has shed light on how these modifications could contribute to a “tubulin code” that coordinates the complex functions of microtubules in cells.

Introduction

Microtubules are key elements of the eukaryotic cytoskeleton that dynamically assemble from heterodimers of α- and β-tubulin. The structure of microtubules, as well as the protein sequences of α- and β-tubulin, is highly conserved in evolution, and consequently, microtubules look alike in almost all species. Despite the high level of conservation, microtubules adapt to a large variety of cellular functions. This adaptation can be mediated by a large panel of microtubule-associated proteins (MAPs), including molecular motors, as well as by mechanisms that directly modify the microtubules, thus either changing their biophysical properties or attracting subsets of MAPs that convey specific functions to the modified microtubules. Two different mechanism can generate microtubule diversity: the expression of different α- and β-tubulin genes, referred to as tubulin isotypes, and the generation of posttranslational modifications (PTMs) on α- and β-tubulin (Figs. 1 and ​and2).2). Although known for several decades, deciphering how tubulin heterogeneity controls microtubule functions is still largely unchartered. This review summarizes the current advances in the field and discusses new concepts arising.Open in a separate windowFigure 1.Tubulin heterogeneity generated by PTMs. (A) Schematic representation of the distribution of different PTMs of tubulin on the α/β-tubulin dimer with respect to their position in the microtubule lattice. Acetylation (Ac), phosphorylation (P), and polyamination (Am) are found within the tubulin bodies that assemble into the microtubule lattice, whereas polyglutamylation, polyglycylation, detyrosination, and C-terminal deglutamylation take place within the C-terminal tubulin tails that project away from the lattice surface. The tubulin dimer represents TubA1A and TubB2B (Fig. 2), and modification sites for polyglutamylation and polyglycylation have been randomly chosen. (B) Chemical structure of the branched peptide formed by polyglutamylation and polyglycylation, using the γ-carboxyl groups of the modified glutamate residues as acceptor sites for the isopeptide bonds. Note that in the case of polyglutamylation, the elongation of the side chains generates classical peptide bonds (Redeker et al., 1991).Open in a separate windowFigure 2.Heterogeneity of C-terminal tails of tubulin isotypes and their PTMs. The amino acid sequences of all tubulin genes found in the human genome are indicated, starting at the last amino acid of the folded tubulin bodies. Amino acids are represented in single-letter codes and color coded according to their biochemical properties. Known sites for polyglutamylation are indicated (Eddé et al., 1990; Alexander et al., 1991; Rüdiger et al., 1992). Potential modification sites (all glutamate residues) are indicated. Known C-terminal truncation reactions of α/β-tubulin (tub) are indicated. The C-terminal tails of the yeast Saccharomyces cerevisiae are shown to illustrate the phylogenetic diversity of these domains.

Tubulin isotypes

The cloning of the first tubulin genes in the late 1970’s (Cleveland et al., 1978) revealed the existence of multiple genes coding for α- or β-tubulin (Ludueña and Banerjee, 2008) that generate subtle differences in their amino acid sequences, particularly in the C-terminal tails (Fig. 2). It was assumed that tubulin isotypes, as they were named, assemble into discrete microtubule species that carry out unique functions. This conclusion was reinforced by the observation that some isotypes are specifically expressed in specialized cells and tissues and that isotype expression changes during development (Lewis et al., 1985; Denoulet et al., 1986). These high expectations were mitigated by a subsequent study showing that all tubulin isotypes freely copolymerize into heterogeneous microtubules (Lewis et al., 1987). To date, only highly specialized microtubules, such as ciliary axonemes (Renthal et al., 1993; Raff et al., 2008), neuronal microtubules (Denoulet et al., 1986; Joshi and Cleveland, 1989), and microtubules of the marginal band of platelets (Wang et al., 1986; Schwer et al., 2001) are known to depend on some specific (β) tubulin isotypes, whereas the function of most other microtubules appears to be independent of their isotype composition.More recently, a large number of mutations in single tubulin isotypes have been linked to deleterious neurodevelopmental disorders (Keays et al., 2007; Fallet-Bianco et al., 2008; Tischfield et al., 2010; Cederquist et al., 2012; Niwa et al., 2013). Mutations of a single tubulin isotype could lead to an imbalance in the levels of tubulins as a result of a lack of incorporation of mutant isoforms into the microtubule lattice or to incorporation that perturbs the architecture or dynamics of the microtubules. The analysis of tubulin disease mutations is starting to reveal how subtle alterations of the microtubule cytoskeleton can lead to functional aberrations in cells and organisms and might provide novel insights into the roles of tubulin isotypes that have so far been considered redundant.

Tubulin PTMs

Tubulin is subject to a large range of PTMs (Fig. 1), from well-known ones, such as acetylation or phosphorylation, to others that have so far mostly been found on tubulin. Detyrosination/tyrosination, polyglutamylation, and polyglycylation, for instance, might have evolved to specifically regulate tubulin and microtubule functions, in particular in cilia and flagella, as their evolution is closely linked to these organelles. The strong link between those modifications and tubulin evolution has led to the perception that they are tubulin PTMs; however, apart from detyrosination/tyrosination, most of them have other substrates (Regnard et al., 2000; Xie et al., 2007; van Dijk et al., 2008; Rogowski et al., 2009).

Tubulin acetylation.

Tubulin acetylation was discovered on lysine 40 (K40; Fig. 1 A) of flagellar α-tubulin in Chlamydomonas reinhardtii (L’Hernault and Rosenbaum, 1985) and is generally enriched on stable microtubules in cells. Considering that K40 acetylation per se has no effect on the ultrastructure of microtubules (Howes et al., 2014), it is rather unlikely that it directly stabilizes microtubules. As a result of its localization at the inner face of microtubules (Soppina et al., 2012), K40 acetylation might rather affect the binding of microtubule inner proteins, a poorly characterized family of proteins (Nicastro et al., 2011; Linck et al., 2014). Functional experiments in cells have further suggested that K40 acetylation regulates intracellular transport by regulating the traffic of kinesin motors (Reed et al., 2006; Dompierre et al., 2007). These observations could so far not be confirmed by biophysical measurements in vitro (Walter et al., 2012; Kaul et al., 2014), suggesting that in cells, K40 acetylation might affect intracellular traffic by indirect mechanisms.Enzymes involved in K40 acetylation are HDAC6 (histone deacetylase family member 6; Hubbert et al., 2002) and Sirt2 (sirtuin type 2; North et al., 2003). Initial functional studies used overexpression, depletion, or chemical inhibition of these enzymes. These studies should be discussed with care, as both HDAC6 and Sirt2 deacetylate other substrates and have deacetylase-independent functions and chemical inhibition of HDAC6 is not entirely selective for this enzyme (Valenzuela-Fernández et al., 2008). In contrast, acetyl transferase α-Tat1 (or Mec-17; Akella et al., 2010; Shida et al., 2010) specifically acetylates α-tubulin K40 (Fig. 3), thus providing a more specific tool to investigate the functions of K40 acetylation. Knockout mice of α-Tat1 are completely void of K40-acetylated tubulin; however, they show only slight phenotypic aberrations, for instance, in their sperm flagellum (Kalebic et al., 2013). A more detailed analysis of α-Tat1 knockout mice demonstrated that absence of K40 acetylation leads to reduced contact inhibition in proliferating cells (Aguilar et al., 2014). In migrating cells, α-Tat1 is targeted to microtubules at the leading edge by clathrin-coated pits, resulting in locally restricted acetylation of those microtubules (Montagnac et al., 2013). A recent structural study of α-Tat1 demonstrated that the low catalytic rate of this enzyme, together with its localization inside the microtubules, caused acetylation to accumulate selectively in stable, long-lived microtubules (Szyk et al., 2014), thus explaining the link between this PTM and stable microtubules in cells. However, the direct cellular function of K40 acetylation on microtubules is still unclear.Open in a separate windowFigure 3.Enzymes involved in PTM of tubulin. Schematic representation of known enzymes (mammalian enzymes are shown) involved in the generation and removal of PTMs shown in Fig. 1. Note that some enzymes still remain unknown, and some modifications are irreversible. (*CCP5 preferentially removes branching points [Rogowski et al., 2010]; however, the enzyme can also hydrolyze linear glutamate chains [Berezniuk et al., 2013]).Recent discoveries have brought up the possibility that tubulin could be subject to multiple acetylation events. A whole-acetylome study identified >10 novel sites on α- and β-tubulin (Choudhary et al., 2009); however, none of these sites have been confirmed. Another acetylation event has been described at lysine 252 (K252) of β-tubulin. This modification is catalyzed by the acetyltransferase San (Fig. 3) and might regulate the assembly efficiency of microtubules as a result of its localization at the polymerization interface (Chu et al., 2011).

Tubulin detyrosination.

Most α-tubulin genes in different species encode a C-terminal tyrosine residue (Fig. 2; Valenzuela et al., 1981). This tyrosine can be enzymatically removed (Hallak et al., 1977) and religated (Fig. 3; Arce et al., 1975). Mapping of tyrosinated and detyrosinated microtubules in cells using specific antibodies (Gundersen et al., 1984; Geuens et al., 1986; Cambray-Deakin and Burgoyne, 1987a) revealed that subsets of interphase and mitotic spindle microtubules are detyrosinated (Gundersen and Bulinski, 1986). As detyrosination was mostly found on stable and long-lived microtubules, especially in neurons (Cambray-Deakin and Burgoyne, 1987b; Robson and Burgoyne, 1989; Brown et al., 1993), it was assumed that this modification promotes microtubule stability (Gundersen et al., 1987; Sherwin et al., 1987). Although a direct stabilization of the microtubule lattice was considered unlikely (Khawaja et al., 1988), it was found more recently that detyrosination protects cellular microtubules from the depolymerizing activity of kinesin-13–type motor proteins, such as KIF2 or MCAK, thus increasing their longevity (Peris et al., 2009; Sirajuddin et al., 2014).Besides kinesin-13 motors, plus end–tracking proteins with cytoskeleton-associated protein glycine-rich (CAP-Gly) domains, such as CLIP170 or p150/glued, specifically interact with tyrosinated microtubules (Peris et al., 2006; Bieling et al., 2008) via this domain (Honnappa et al., 2006). In contrast, kinesin-1 moves preferentially on detyrosinated microtubules tracks in cells (Liao and Gundersen, 1998; Kreitzer et al., 1999; Konishi and Setou, 2009). The effect of detyrosination on kinesin-1 motor behavior was recently measured in vitro, and a small but significant increase in the landing rate and processivity of the motor has been found (Kaul et al., 2014). Such subtle changes in the motor behavior could, in conjunction with other factors, such as regulatory MAPs associated with cargo transport complexes (Barlan et al., 2013), lead to a preferential use of detyrosinated microtubules by kinesin-1 in cells.Despite the early biochemical characterization of a detyrosinating activity, the carboxypeptidase catalyzing detyrosination of α-tubulin has yet to be identified (Hallak et al., 1977; Argaraña et al., 1978, 1980). In contrast, the reverse enzyme, tubulin tyrosine ligase (TTL; Fig. 3; Raybin and Flavin, 1975; Deanin and Gordon, 1976; Argaraña et al., 1980), has been purified (Schröder et al., 1985) and cloned (Ersfeld et al., 1993). TTL modifies nonpolymerized tubulin dimers exclusively. This selectivity is determined by the binding interface between the TTL and tubulin dimers (Szyk et al., 2011, 2013; Prota et al., 2013). In contrast, the so far unidentified detyrosinase acts preferentially on polymerized microtubules (Kumar and Flavin, 1981; Arce and Barra, 1983), thus modifying a select population of microtubules within cells (Gundersen et al., 1987).In most organisms, only one unique gene for TTL exists. Consequently, TTL knockout mice show a huge accumulation of detyrosinated and particularly Δ2-tubulin (see next section). TTL knockout mice die before birth (Erck et al., 2005) with major developmental defects in the nervous system that might be related to aberrant neuronal differentiation (Marcos et al., 2009). TTL is strictly tubulin specific (Prota et al., 2013), indicating that all observed defects in TTL knockout mice are directly related to the deregulation of the microtubule cytoskeleton.

Δ2-tubulin and further C-terminal modification.

A biochemical study of brain tubulin revealed that ∼35% of α-tubulin cannot be retyrosinated (Paturle et al., 1989) because of the lack of the penultimate C-terminal glutamate residue of the primary protein sequence (Fig. 2; Paturle-Lafanechère et al., 1991). This so-called Δ2-tubulin (for two C-terminal amino acids missing) cannot undergo retyrosination as a result of structural constraints within TTL (Prota et al., 2013) and thus is considered an irreversible PTM.Δ2-tubulin accumulates in long-lived microtubules of differentiated neurons, axonemes of cilia and flagella, and also in cellular microtubules that have been artificially stabilized, for instance, with taxol (Paturle-Lafanechère et al., 1994). The generation of Δ2-tubulin requires previous detyrosination of α-tubulin; thus, the levels of this PTM are indirectly regulated by the detyrosination/retyrosination cycle. This mechanistic link is particularly apparent in the TTL knockout mice, which show massive accumulation of Δ2-tubulin in all tested tissues (Erck et al., 2005). Loss of TTL and the subsequent increase of Δ2-tubulin levels were also linked to tumor growth and might contribute to the aggressiveness of the tumors by an as-yet-unknown mechanism (Lafanechère et al., 1998; Mialhe et al., 2001). To date, no specific biochemical role of Δ2-tubulin has been determined; thus, one possibility is that the modification simply locks tubulin in the detyrosinated state.The enzymes responsible for Δ2-tubulin generation are members of a family of cytosolic carboxypeptidases (CCPs; Fig. 3; Kalinina et al., 2007; Rodriguez de la Vega et al., 2007), and most of them also remove polyglutamylation from tubulin (see next section; Rogowski et al., 2010). These enzymes are also able to generate Δ3-tubulin (Fig. 1 A; Berezniuk et al., 2012), indicating that further degradation of the tubulin C-terminal tails are possible; however, the functional significance of this event is unknown.

Polyglutamylation.

Polyglutamylation is a PTM that occurs when secondary glutamate side chains are formed on γ-carboxyl groups of glutamate residues in a protein (Fig. 1, A and B). The modification was first discovered on α- and β-tubulin from the brain (Eddé et al., 1990; Alexander et al., 1991; Rüdiger et al., 1992; Mary et al., 1994) as well as on axonemal tubulin from different species (Mary et al., 1996, 1997); however, it is not restricted to tubulin (Regnard et al., 2000; van Dijk et al., 2008). Using a glutamylation-specific antibody, GT335 (Wolff et al., 1992), it was observed that tubulin glutamylation increases during neuronal differentiation (Audebert et al., 1993, 1994) and that axonemes of cilia and flagella (Fouquet et al., 1994), as well as centrioles of mammalian centrosomes (Bobinnec et al., 1998), are extensively glutamylated.Enzymes catalyzing polyglutamylation belong to the TTL-like (TTLL) family (Regnard et al., 2003; Janke et al., 2005). In mammals, nine glutamylases exist, each of them showing intrinsic preferences for modifying either α- or β-tubulin as well as for initiating or elongating glutamate chains (Fig. 3; van Dijk et al., 2007). Two of the six well-characterized TTLL glutamylases also modify nontubulin substrates (van Dijk et al., 2008).Knockout or depletion of glutamylating enzymes in different model organisms revealed an evolutionarily conserved role of glutamylation in cilia and flagella. In motile cilia, glutamylation regulates beating behavior (Janke et al., 2005; Pathak et al., 2007; Ikegami et al., 2010) via the regulation of flagellar dynein motors (Kubo et al., 2010; Suryavanshi et al., 2010). Despite the expression of multiple glutamylases in ciliated cells and tissues, depletion or knockout of single enzymes often lead to ciliary defects, particularly in motile cilia (Ikegami et al., 2010; Vogel et al., 2010; Bosch Grau et al., 2013; Lee et al., 2013), suggesting essential and nonredundant regulatory functions of these enzymes in cilia.Despite the enrichment of polyglutamylation in neuronal microtubules (Audebert et al., 1993, 1994), knockout of TTLL1, the major polyglutamylase in brain (Janke et al., 2005), did not show obvious neuronal defects in mice (Ikegami et al., 2010; Vogel et al., 2010). This suggests a tolerance of neuronal microtubules to variations in polyglutamylation.Deglutamylases, the enzymes that reverse polyglutamylation, were identified within a novel family of CCPs (Kimura et al., 2010; Rogowski et al., 2010). So far, three out of six mammalian CCPs have been shown to cleave C-terminal glutamate residues, thus catalyzing both the reversal of polyglutamylation and the removal of gene-encoded glutamates from the C termini of proteins (Fig. 3). The hydrolysis of gene-encoded glutamate residues is not restricted to tubulin, in which it generates Δ2- and Δ3-tubulin, but has also been reported for other proteins such as myosin light chain kinase (Rusconi et al., 1997; Rogowski et al., 2010). One enzyme of the CCP family, CCP5, preferentially removes branching points generated by glutamylation, thus allowing the complete reversal of the polyglutamylation modification (Kimura et al., 2010; Rogowski et al., 2010). However, CCP5 can also hydrolyze C-terminal glutamate residues from linear peptide chains similar to other members of the CCP family (Berezniuk et al., 2013).CCP1 is mutated in a well-established mouse model for neurodegeneration, the pcd (Purkinje cell degeneration) mouse (Mullen et al., 1976; Greer and Shepherd, 1982; Fernandez-Gonzalez et al., 2002). The absence of a key deglutamylase leads to strong hyperglutamylation in brain regions that undergo degeneration, such as the cerebellum and the olfactory bulb (Rogowski et al., 2010). When glutamylation levels were rebalanced by depletion or knockout of the major brain polyglutamylase TTLL1 (Rogowski et al., 2010; Berezniuk et al., 2012), Purkinje cells survived. Although the molecular mechanisms of hyperglutamylation-induced degeneration remain to be elucidated, perturbation of neuronal transport, as well as changes in the dynamics and stability of microtubules, is expected to be induced by hyperglutamylation. Increased polyglutamylation levels have been shown to affect kinesin-1–mediated transport in cultured neurons (Maas et al., 2009), and the turnover of microtubules can also be regulated by polyglutamylation via the activation of microtubule-severing enzymes such as spastin (Lacroix et al., 2010).Subtle differences in polyglutamylation can be seen on diverse microtubules in different cell types. The functions of these modifications remain to be studied; however, its wide distribution strengthens the idea that it could be involved in fine-tuning a range of microtubule functions.

Polyglycylation.

Tubulin polyglycylation or glycylation, like polyglutamylation, generates side chains of glycine residues within the C-terminal tails of α- and β-tubulin (Fig. 1, A and B). The modification sites of glycylation are considered to be principally the same as for glutamylation, and indeed, both PTMs have been shown to be interdependent in cells (Rogowski et al., 2009; Wloga et al., 2009). Initially discovered on Paramecium tetraurelia tubulin (Redeker et al., 1994), glycylation has been extensively studied using two antibodies, TAP952 and AXO49 (Bressac et al., 1995; Levilliers et al., 1995; Bré et al., 1996). In contrast to polyglutamylation, glycylation is restricted to cilia and flagella in most organisms analyzed so far.Glycylating enzymes are also members of the TTLL family, and homologues of these enzymes have so far been found in all organisms with proven glycylation of ciliary axonemes (Rogowski et al., 2009; Wloga et al., 2009). In mammals, initiating (TTLL3 and TTLL8) and elongating (TTLL10) glycylases work together to generate polyglycylation (Fig. 3). In contrast, the two TTLL3 orthologues from Drosophila melanogaster can both initiate and elongate glycine side chains (Rogowski et al., 2009).In mice, motile ependymal cilia in brain ventricles acquire monoglycylation upon maturation, whereas polyglycylation is observed only after several weeks (Bosch Grau et al., 2013). Sperm flagella, in contrast, acquire long glycine chains much faster, suggesting that the extent of polyglycylation could correlate with the length of the axonemes (Rogowski et al., 2009). Depletion of glycylases in mice (ependymal cilia; Bosch Grau et al., 2013), zebrafish (Wloga et al., 2009; Pathak et al., 2011), Tetrahymena thermophila (Wloga et al., 2009), and D. melanogaster (Rogowski et al., 2009) consistently led to ciliary disassembly or severe ciliary defects. How glycylation regulates microtubule functions remains unknown; however, the observation that glycylation-depleted axonemes disassemble after initial assembly (Rogowski et al., 2009; Bosch Grau et al., 2013) suggests a role of this PTM in stabilizing axonemal microtubules. Strikingly, human TTLL10 is enzymatically inactive; thus, humans have lost the ability to elongate glycine side chains (Rogowski et al., 2009). This suggests that the elongation of the glycine side chains is not an essential aspect of the function of this otherwise critical tubulin PTM.

Other tubulin PTMs.

Several other PTMs have been found on tubulin. Early studies identified tubulin phosphorylation (Eipper, 1974; Gard and Kirschner, 1985; Díaz-Nido et al., 1990); however, no specific functions were found. The perhaps best-studied phosphorylation event on tubulin takes place at serine S172 of β-tubulin (Fig. 1 A), is catalyzed by the Cdk1 (Fig. 3), and might regulate microtubule dynamics during cell division (Fourest-Lieuvin et al., 2006; Caudron et al., 2010). Tubulin can be also modified by the spleen tyrosine kinase Syk (Fig. 3; Peters et al., 1996), which might play a role in immune cells (Faruki et al., 2000; Sulimenko et al., 2006) and cell division (Zyss et al., 2005; Sulimenko et al., 2006).Polyamination has recently been discovered on brain tubulin (Song et al., 2013), after having been overlooked for many years as a result of the low solubility of polyaminated tubulin. Among several glutamine residues of α- and β-tubulin that can be polyaminated, Q15 of β-tubulin is considered the primary modification site (Fig. 1 A). Polyamination is catalyzed by transglutaminases (Fig. 3), which modify free tubulin as well as microtubules in an irreversible manner, and most likely contribute to the stabilization of microtubules (Song et al., 2013).Tubulin was also reported to be palmitoylated (Caron, 1997; Ozols and Caron, 1997; Caron et al., 2001), ubiquitinated (Ren et al., 2003; Huang et al., 2009; Xu et al., 2010), glycosylated (Walgren et al., 2003; Ji et al., 2011), arginylated (Wong et al., 2007), methylated (Xiao et al., 2010), and sumoylated (Rosas-Acosta et al., 2005). These PTMs have mostly been reported without follow-up studies, and some of them are only found in specific cell types or organisms and/or under specific metabolic conditions. Further studies will be necessary to gain insights into their potential roles for the regulation of the microtubule cytoskeleton.

Current advances and future perspectives

The molecular heterogeneity of microtubules, generated by the expression of different tubulin isotypes and by the PTM of tubulin has fascinated the scientific community for ∼40 years. Although many important advances have been made in the past decade, the dissection of the molecular mechanisms and a comprehensive understanding of the biological functions of tubulin isotypes and PTMs will be a challenging field of research in the near future.

Direct measurements of the impact of tubulin heterogeneity.

The most direct and reliable type of experiments to determine the impact of tubulin heterogeneity on microtubule behavior are in vitro measurements with purified proteins. However, most biophysical work on microtubules has been performed with tubulin purified from bovine, ovine, or porcine brains, which can be obtained in large quantities and with a high degree of purity and activity (Vallee, 1986; Castoldi and Popov, 2003). Brain tubulin is a mixture of different tubulin isotypes and is heavily posttranslationally modified and thus inept for investigating the functions of tubulin heterogeneity (Denoulet et al., 1986; Cambray-Deakin and Burgoyne, 1987b; Paturle et al., 1989; Eddé et al., 1990). Thus, pure, recombinant tubulin will be essential to dissect the roles of different tubulin isoforms and PTMs.Attempts to produce recombinant, functional α- and β-tubulin in bacteria have failed so far (Yaffe et al., 1988), most likely because of the absence of the extensive tubulin-specific folding machinery (Yaffe et al., 1992; Gao et al., 1993; Tian et al., 1996; Vainberg et al., 1998) in prokaryotes. An alternative source of tubulin with less isotype heterogeneity and with almost no PTMs is endogenous tubulin from cell lines such as HeLa, which in the past has been purified using a range of biochemical procedures (Bulinski and Borisy, 1979; Weatherbee et al., 1980; Farrell, 1982; Newton et al., 2002; Fourest-Lieuvin, 2006). Such tubulin can be further modified with tubulin-modifying enzymes, such as polyglutamylases, either by expressing those enzymes in the cells before tubulin purification (Lacroix and Janke, 2011) or in vitro with purified enzymes (Vemu et al., 2014). Despite some technical limitations of these methods, HeLa tubulin modified in cells has been successfully used in an in vitro study on the role of polyglutamylation in microtubule severing (Lacroix et al., 2010).Naturally occurring variants of tubulin isotypes and PTMs can be purified from different organisms, organs, or cell types, but obviously, only some combinations of tubulin isotypes and PTMs can be obtained by this approach. The recent development of an affinity purification method using the microtubule-binding TOG (tumor overexpressed gene) domain of yeast Stu2p has brought a new twist to this approach, as it allows purifying small amounts of tubulin from any cell type or tissue (Widlund et al., 2012).The absence of tubulin heterogeneity in yeast has made budding and fission yeast potential expression systems for recombinant, PTM-free tubulin (Katsuki et al., 2009; Drummond et al., 2011; Johnson et al., 2011). However, the expression of mammalian tubulin in this system has remained impossible. This problem was then partially circumvented by expressing tubulin chimeras that consist of a yeast tubulin body fused to mammalian C-terminal tubulin tails, thus mimicking different tubulin isotypes (Sirajuddin et al., 2014). Moreover, detyrosination can be generated by deleting the key C-terminal residue from endogenous or chimeric α-tubulin (Badin-Larçon et al., 2004), and polyglutamylation is generated by chemically coupling glutamate side chains to specifically engineered tubulin chimeras (Sirajuddin et al., 2014). These approaches allowed the first direct measurements of the impact of tubulin isotypes and PTMs on the behavior of molecular motors in vitro (Sirajuddin et al., 2014) and the analysis of the effects of tubulin heterogeneity on microtubule behavior and interactions inside the yeast cell (Badin-Larçon et al., 2004; Aiken et al., 2014).Currently, the most promising development has been the successful purification of fully functional recombinant tubulin from the baculovirus expression system (Minoura et al., 2013). Using this system, defined α/β-tubulin dimers can be obtained using two different epitope tags on α- and β-tubulin, respectively. Although these epitope tags are essential for separating recombinant from the endogenous tubulin, they could also affect tubulin assembly or microtubule–MAP interactions. Thus, future developments should focus on eliminating these tags.Current efforts have brought the possibility of producing recombinant tubulin into reach. Further improvement and standardization of these methods will certainly provide a breakthrough in understanding the mechanisms by which tubulin heterogeneity contributes to microtubule functions.

Complexity of tubulin—understanding the regulatory principles.

The diversity of tubulin genes (isotypes) and the complexity of tubulin PTMs have led to the proposal of the term “tubulin code” (Verhey and Gaertig, 2007; Wehenkel and Janke, 2014), in analogy to the previously coined histone code (Jenuwein and Allis, 2001). Tubulin molecules consist of a highly structured and thus evolutionarily conserved tubulin body and the unstructured and less conserved C-terminal tails (Nogales et al., 1998). As PTMs and sequence variations within the tubulin body are expected to affect the conserved tubulin fold and therefore the properties of the microtubule lattice, they are not likely to be involved in generating the tubulin code. In contrast, modulations of the C-terminal tails could encode signals on the microtubule surface without perturbing basic microtubule functions and properties (Figs. 1 A and ​and4).4). Indeed, the highest degree of gene-encoded diversity (Fig. 2) and the highest density and complexity of PTMs (Fig. 1) are found within these tail domains.Open in a separate windowFigure 4.Molecular components of the tubulin code. Schematic representation of potential coding elements that could generate specific signals for the tubulin code. (A) The length of the C-terminal tails of different tubulin isotypes differ significantly (Fig. 2) and could have an impact on the interactions between microtubules and MAPs. (B) Tubulin C-terminal tails are rich in charged amino acid residues. The distribution of these residues and local densities of charges could influence the electrostatic interactions with the tails and the readers. (C) Although each glutamate residue within the C-terminal tails could be considered a potential modification site, only some sites have been found highly occupied in tubulin purifications from native sources. This indicates selectivity of the modification reactions, which can participate in the generation of specific modification patterns (see D). Modification sites might be distinguished by their neighboring amino acid residues, which could create specific modification epitopes. (D) As a result of the large number of modification sites and the variability of side chains, a large variety of modification patterns could be generated within a single C-terminal tail of tubulin. (E) Modification patterns as shown in D can be distinct between α- and β-tubulin. These modification patterns could be differentially distributed at the surface of the microtubule lattice, thus generating a higher-order patterning. Tub, tubulin. For color coding, see Fig. 2.Considering the number of tubulin isotypes plus all potential combinations of PTMs (e.g., each glutamate residue within the C-terminal tubulin tail could be modified by either polyglutamylation or polyglycylation, each of them generating side chains of different lengths; Fig. 4), the number of distinct signals generated by the potential tubulin code would be huge. However, as many of these potential signals represent chemical structures that are similar and might not be reliably distinguished by readout mechanisms, it is possible that the tubulin code generates probabilistic signals. In this scenario, biochemically similar modifications would have similar functional readouts, and marginal differences between those signals would only bias biological processes but not determine them. This stands in contrast to the concept of the histone code, in which precise patterns of different PTMs on the histone proteins encode distinct biological signals.The concept of probabilistic signaling is already inscribed in the machinery that generates the tubulin code. Polyglutamylases and polyglycylases from the TTLL family have preferential activities for either α- or β-tubulin and for generating different lengths of the branched glutamate or glycine chains. Although under conditions of low enzyme concentrations, as found in most cells and tissues, the enzymes seem to selectively generate their preferential type of PTM, higher enzyme concentrations induce a more promiscuous behavior, leading, for instance, to a loss of selectivity for α- or β-tubulin (van Dijk et al., 2007). Similarly, the modifying enzymes might prefer certain modification sites within the C-terminal tails of tubulin but might be equally able to modify other sites, which could be locally regulated in cells. For example, β-tubulin isotypes isolated from mammalian brain were initially found to be glutamylated on single residues (Alexander et al., 1991; Rüdiger et al., 1992), which in the light of the comparably low sensitivity of mass spectrometry at the time might rather indicate a preferential than a unique modification of these sites. Nevertheless, the neuron-specific polyglutamylase for β-tubulin TTLL7 (Ikegami et al., 2006) can incorporate glutamate onto many more modification sites of β-tubulin in vitro (Mukai et al., 2009), which clearly indicates that not all of the possible modification events take place under physiological conditions.Several examples supporting a probabilistic signaling mode of the tubulin code are found in the recent literature. In T. thermophila, a ciliate without tubulin isotype diversity (Gaertig et al., 1993) but with a huge repertoire of tubulin PTMs and tubulin-modifying enzymes (Janke et al., 2005), tubulin can be easily mutagenized to experimentally eliminate sites for PTMs. Mutagenesis of the most commonly occupied glutamylation/glycylation sites within the β-tubulin tails did not generate a clear decrease of glycylation levels nor did it cause obvious phenotypic alterations. This indicates that the modifying enzymes can deviate toward alternative modification sites and that similar PTMs on different sites can compensate the functions of the mutated site. However, when all of the key modification sites were mutated, glycylation became prominently decreased, which led to severe phenotypes, including lethality (Xia et al., 2000). Most strikingly, these phenotypes could be recovered by replacing the C-terminal tail of α-tubulin with the nonmutated β-tubulin tail. This α–β-tubulin chimera became overglycylated and functionally compensated for the absence of modification sites on β-tubulin. The conclusion of this study is that PTM- and isotype-generated signals can fulfill a biological function within a certain range of tolerance.But how efficient is such compensation? The answer can be found in a variety of already described deletion mutants for tubulin-modifying enzymes in different model organisms. Most single-gene knockouts for TTLL genes (glutamylases or glycylases) did not result in prominent phenotypic alterations in mice, even for enzymes that are ubiquitously expressed. Only some highly specialized microtubule structures show functional aberrations upon the deletion of a single enzyme. These “tips of the iceberg” are usually the motile cilia and sperm flagella, which carry very high levels of polyglutamylation and polyglycylation (Bré et al., 1996; Kann et al., 1998; Rogowski et al., 2009). It thus appears that some microtubules are essentially dependent on the generation of specific PTM patterns, whereas others can tolerate changes and appear to function normally. How “normal” these functions are remains to be investigated in future studies. It is possible that defects are subtle and thus overlooked but could become functionally important under specific conditions.A tubulin code also requires readout mechanisms. The most likely “readers” of the tubulin code are MAPs and molecular motors. Considering the probabilistic signaling hypothesis, the expected effects of the signals would be in most cases rather gradual changes, for instance, to fine-tune molecular motor traffic and/or to bias motors toward defined microtubule tracks but not to obliterate motor activity or MAP binding to microtubules. An in vitro study using recombinant tubulin chimeras purified from yeast confirmed this notion (Sirajuddin et al., 2014). By analyzing which elements of the tubulin code can regulate the velocity and processivity of the molecular motors kinesin and dynein, these researchers found that the C-terminal tails of α- and β-tubulin differentially influence the kinetic parameters of the tested motors; however, the modulation was rather modest. One of their striking observations was that a single lysine residue, present in the C-terminal tails of two β-tubulin isotypes (Figs. 2 and ​and4),4), significantly affected motor traffic and that this effect can be counterbalanced by polyglutamylation. These observations are the first in vitro evidence for the interdependence of different elements of the tubulin code and provide another indication for its probabilistic mode of signaling.

Future directions.

One of the greatest technological challenges to understanding the function of the tubulin code is to detect and interpret subtle and complex regulatory events generated by this code. It will thus be instrumental to further develop tools to better distinguish graded changes in PTM levels on microtubules in cells and tissues (Magiera and Janke, 2013) and to reliably measure subtle modulations of microtubule behavior in reconstituted systems.The current advances in the field and especially the availability of whole-organism models, as well as first insights into the pathological role of tubulin mutations (Tischfield et al., 2011), are about to transform our way of thinking about the regulation of microtubule cytoskeleton. Tubulin heterogeneity generates complex probabilistic signals that cannot be clearly attributed to single biological functions in most cases and that are not essential for most cellular processes. Nevertheless, it has been conserved throughout evolution of eukaryotes and can hardly be dismissed as not important. To understand the functional implications of these processes, we might be forced to reconsider how we define biologically important events and how we measure events that might encode probabilistic signals. The answers to these questions could provide novel insights into how complex systems, such as cells and organisms, are sustained throughout difficult and challenging life cycles, resist to environmental stress and diseases, and have the flexibility needed to succeed in evolution.     

Who’s in control? Principles of Rab GTPase activation in endolysosomal membrane trafficking and beyond

The eukaryotic endomembrane system consists of multiple interconnected organelles. Rab GTPases are organelle-specific markers that give identity to these membranes by recruiting transport and trafficking proteins. During transport processes or along organelle maturation, one Rab is replaced by another, a process termed Rab cascade, which requires at its center a Rab-specific guanine nucleotide exchange factor (GEF). The endolysosomal system serves here as a prime example for a Rab cascade. Along with endosomal maturation, the endosomal Rab5 recruits and activates the Rab7-specific GEF Mon1-Ccz1, resulting in Rab7 activation on endosomes and subsequent fusion of endosomes with lysosomes. In this review, we focus on the current idea of Mon1-Ccz1 recruitment and activation in the endolysosomal and autophagic pathway. We compare identified principles to other GTPase cascades on endomembranes, highlight the importance of regulation, and evaluate in this context the strength and relevance of recent developments in in vitro analyses to understand the underlying foundation of organelle biogenesis and maturation.

Membrane identity in the endomembrane systemOne key feature of eukaryotic cells is the presence of membrane-enclosed organelles, which constantly exchange proteins, lipids, or metabolites via vesicular transport or membrane contact sites (MCSs). Along the endomembrane system, vesicular trafficking requires vesicle budding from the donor membrane and directed transport toward and fusion with the acceptor compartment. The resulting trafficking routes form a regulated network that connects not only the internal organelles, but also the interior and exterior of the cell.The specific identity of organelles within the endomembrane system is defined by the lipid and protein composition of their membranes. This includes signaling lipids such as phosphoinositides (PIPs) and small GTPases of the Ras superfamily of small G proteins, namely of the Rab, Arf, and Arl families, which act as binding platforms for accessory proteins involved in multiple membrane trafficking processes (Balla, 2013).Rab GTPases, like other small GTPases, are key regulatory proteins that switch between an inactive GDP-bound (Rab-GDP) and an active GTP-bound (Rab-GTP) state (Barr, 2013; Goody et al., 2017; Hutagalung and Novick, 2011). Rabs are posttranslationally modified by the addition of geranylgeranyl moieties to C-terminal cysteine residues, which allow their reversible membrane association. Within the cytosol, Rab-GDP is kept soluble by binding to the chaperone-like GDP dissociation inhibitor (GDI). At the target membrane, an organelle-specific guanine nucleotide exchange factor (GEF) activates the Rab after its previous release from GDI, a process possibly supported by other factors (Dirac-Svejstrup et al., 1997). GTP binding stabilizes two loops in the Rab GTPase domain, which allows recruitment and binding of various so-called effector proteins to the Rab-GTP on the membrane. Rab GTPases are inefficient enzymes with a low intrinsic GTP hydrolysis rate and thus depend on a GTPase-activating protein (GAP) to hydrolyze bound GTP. GDI then extracts the Rab-GDP and keeps it soluble in the cytosol until the next activation cycle (Barr, 2013; Goody et al., 2017; Hutagalung and Novick, 2011). In addition to their conserved GTPase domain, Rabs contain a hypervariable C-terminal domain (HVD), which supports GEF recognition and therefore correct localization of the Rab (Thomas et al., 2018)Among various other functions, Rab GTPases are critical for the fusion of vesicles with the acceptor membrane by recruiting tethering proteins, which bring the two membranes into close proximity. Tethers, together with Sec1/Munc18 proteins, promote the folding of membrane-bound SNAREs at the vesicle and the target membrane into tetrameric coiled-coil complexes. This process further reduces the distance between the membranes, bypasses the hydration layer on membranes, and results in mixing of lipid bilayers and consequently membrane fusion (Wickner and Rizo, 2017; Ungermann and Kümmel, 2019).Organization and function of the endolysosomal pathwayEndocytosis allows the rapid adaptation of plasma membrane composition in response to changing environmental conditions by the uptake of membrane proteins from the plasma membrane, which are either transported to and finally degraded in the lysosome or sorted back to the plasma membrane, e.g., receptors after releasing their cargo within the endosomal lumen (Sardana and Emr, 2021). A third fate of endocytosed cargo is trafficking to the Golgi (Laidlaw and MacDonald, 2018). In addition, various kinds of endocytosis allow the uptake of very large particles such as bacteria during phagocytosis or fluids during pinocytosis (Huotari and Helenius, 2011; Babst, 2014). The endocytic pathway is also involved in the quality control system of plasma membrane proteins and allows degradation of damaged cell surface proteins as well as the down-regulation of nutrient transporters and receptors (Sardana and Emr, 2021). During endocytosis, membrane proteins marked by ubiquitination are incorporated into endocytic vesicles, which pinch off the plasma membrane and fuse with the tubular-shaped early endosome (EE) in the cell periphery (Fig. 1 A). The EE serves as a sorting station, at which membrane proteins are either sorted into tubular structures and brought to the recycling endosome (RE) or get incorporated into intraluminal vesicles (ILVs) with the help of four endosomal sorting complexes required for transport (ESCRTs; Sardana and Emr, 2021). A prerequisite for the degradation of cargo in the lysosome is the maturation of EEs into late endosomes (LEs) by changing the organelle surface composition, including specific Rab GTPases and PIPs, and organelle shape. The LE is eventually spherically shaped, containing multiple ILVs and a more acidified lumen. Therefore, it is also called Multivesicular Body (MVB). Upon fusion with the lysosome, ILVs and their content are degraded into precursor molecules, which are reused by the cell (Fig. 1 A; Sardana and Emr, 2021; Huotari and Helenius, 2011).Open in a separate windowFigure 1.Rab GTPases in the endolysosomal pathway.(A) Localization of key Rab GTPases along the endolysosomal pathway. Endocytic vesicles containing cargo (blue dot) or receptor proteins (red) are substrates of endocytosis. Endocytic vesicles (EV) fuse with the EE. Rabs are shown by numbers: Rab5 (green) on early EE is replaced by Rab7 (black) on multivesicular bodies (MVBs). GEFs are shown in blue. Positioning of lysosomes (Lys) depends on binding to motor proteins by either Arl8b (orange, 8b) or Rab7. Recycling occurs via REs involving Rab4, Rab11, and Rab14. MTOC, microtubule organizing center; Nuc, nucleus. (B) Spatiotemporal Rab5-to-Rab7 transition during endosomal maturation. Rab5 (green graph) is rapidly recruited to EE and replaced by Rab7. (C) Model of Rab7 GEF recruitment and activation on endosomes. Mon1-Ccz1 (or the trimeric complex additionally containing Rmc1/C18orf8/Bulli, as indicated by the unlabeled hexagon) requires Rab5-GTP for activation to promote Rab7 recruitment. For details, see text.Central functions of Rab5 and Rab7Along the endolysosomal system, several Rabs coordinate sorting and recycling processes at the EE and LE. Early endosomal Rab5 and late endosomal Rab7 are here the key Rabs conserved among species. Their spatiotemporal activation and therefore functions are tightly coordinated on the level of the MVB/LE (Fig. 1 B).In yeast, the Rab5-like GTPases Vps21, Ypt52, Ypt52, and Ypt10 and the Rab7-like Ypt7 structure the endocytic pathway (Singer-Krüger et al., 1994; Wichmann et al., 1992). In mammalian cells, Rab5 (with Rab5a, b, and c isoforms having nonredundant functions in the endocytic network; Chen et al., 2014, 2009) and Rab7 (with Rab7a and b isoforms, of which Rab7a is the main actor in transport processes along the endocytic pathway [Guerra and Bucci, 2016], whereas Rab7b has a role in the transport from endosome to the Golgi [Kjos et al., 2017; Progida et al., 2010]) are present (Wandinger-Ness and Zerial, 2014). While the overall organization of the endocytic pathway into EE and LE is conserved, yeast seems to have a more ancestral minimal endomembrane system, where the trans-Golgi network acts as EE and RE (Day et al., 2018). In mammalian cells, the more complex endolysosomal system depends on additional Rabs. Rab4 is involved in protein sorting at the EE, activation of Rab5, and recycling of cargo back to the plasma membrane (Kälin et al., 2015; Wandinger-Ness and Zerial, 2014; de Renzis et al., 2002), whereas Rab11 and Rab14 function at REs (Fig. 1 A; Linford et al., 2012; Takahashi et al., 2012). Furthermore, Rab9 is required for retrograde transport between LEs and the trans-Golgi network (Lombardi et al., 1993), and Rab32 and Rab38 function in the biogenesis of lysosome-related organelles (Bowman et al., 2019; Gerondopoulos et al., 2012; Wasmeier et al., 2006).During endosomal maturation, Rab5 is exchanged for Rab7 (Rink et al., 2005; Poteryaev et al., 2010). This Rab switch is highly conserved and a prime example of coordinated Rab turnover during organelle maturation. The rapid transition from Rab5 to Rab7 was explained by a so-called cutout switch, where activation of Rab5 fosters at a threshold value activation of Rab7, which in turn suppresses further Rab5 activation (Fig. 1 B; Del Conte-Zerial et al., 2008). Such a principle may apply to most Rab cascades (Barr, 2013).Rab5 has multiple functions on EEs (Wandinger-Ness and Zerial, 2014). It interacts with a number of effectors such as the lipid kinase Vps34, Rabaptin-5, which is found in complex with the Rab5-GEF Rabex5, Rabenosyn-5, and tethers such as the class C core vacuole/endosome tethering (CORVET) complex or EEA1. Therefore, Rab5 is critical for the homotypic fusion of EEs (Gorvel et al., 1991; Ohya et al., 2009; Christoforidis et al., 1999a, b; Perini et al., 2014; Marat and Haucke, 2016). Vps34 was initially identified in yeast (Schu et al., 1993) and exists in two heterotetrametric complexes, which differ by just one subunit (Kihara et al., 2001). Complex I resides on autophagosomes, whereas complex II functions on endosomes (Fig. 2 D). Both complexes generate a local pool of phosphatidylinositol-3-phosphate (PI3P), to which several effectors bind, including the early endosomal tether EEA1 and ESCRTs (Wallroth and Haucke, 2018). Recent structural insights revealed that Rab5 recruits and activates endosomal complex II, whereas Rab1 acts similarly on autophagosomal complex I (Tremel et al., 2021). This explains how Rab5-GTP promotes the formation of a local endosomal PI3P pool (Franke et al., 2019). Interestingly, Caenorhabditis elegans VPS-34 can recruit the Rab5 GAP TBC-2 to endosomal membranes, suggesting a possible link between PI3P generation and Rab5 inactivation (Law et al., 2017).Open in a separate windowFigure 2.Rab7 activation on autophagosomes.(A and B) Atg8-dependent Mon1-Ccz1 recruitment and activation. Atg8 (violet) recruits Mon1-Ccz1 (and likely also the trimeric GEF complex in higher eukaryotes, as indicated by the unlabeled hexagon) and allows fusion with lysosome. (C) Model of spatiotemporal Rab7 activation on autophagosomes. Maturation is prerequisite for successful fusion. (D) Comparison of proteins involved in maturation of LEs and autophagosomes.Rab7 is a key component in the late endocytic pathway (Langemeyer et al., 2018a). It is found on LEs, lysosomes, and autophagosomes and is required for the biogenesis and positioning of LEs and lysosomes, for MCSs of lysosomes with other organelles, and for the fusion of endosomes and autophagosomes with lysosomes (Fig. 1 A; Guerra and Bucci, 2016; McEwan et al., 2015; Ballabio and Bonifacino, 2020; Cabukusta and Neefjes, 2018). Even though both the metazoan Rab7 and yeast Ypt7 are activated by the homologous Mon1-Ccz1 GEF complex and are required for endosomal maturation, their function on LEs and lysosomes is not entirely conserved. In yeast, active Ypt7 directly binds the hexameric homotypic fusion and vacuole protein sorting (HOPS) tethering complex and mediates SNARE-dependent fusion of LEs or autophagosomes with vacuoles as well as homotypic vacuole fusion (Wickner and Rizo, 2017; Gao et al., 2018a, b). In higher eukaryotes, HOPS also promotes fusion between LEs and lysosomes, yet apparently does not directly interact with Rab7, but rather with the GTPases Rab2 and Arl8b (Gillingham et al., 2014; Fujita et al., 2017; Lőrincz et al., 2017; Khatter et al., 2015). How Rab7 contributes to fusion at the lysosome is still unclear. Rab7 interacts with several proteins on lysosomes, including the cholesterol sensor ORPL1 and the dynein-interacting lysosomal RILP (Jordens et al., 2001; Cantalupo et al., 2001; Rocha et al., 2009). Both proteins also bind HOPS (van der Kant et al., 2015, 2013), as does another multivalent adaptor protein, PLEKHM1 (McEwan et al., 2015), which binds both Arl8b and Rab7 (Marwaha et al., 2017). Interestingly, Arl8b in complex with its effector SKIP also binds TBC1D15, a Rab7 GAP, which may displace Rab7 from LEs before their fusion with lysosomes (Jongsma et al., 2020). It is thus possible that fusion of LEs and autophagosomes with lysosomes requires a complex coordination of the three GTPases, Rab7, Arl8b, and Rab2, with the HOPS complex and other effectors. Some of this complexity may be explained by a second function of Rab7 and Arl8b in binding adapters of the kinesin or dynein motor protein family, which connect LEs and lysosomes to the microtubule network. Thereby Rab7 and Arl8b control the positioning of these organelles to the periphery or perinuclear area via the microtubule network, which has functional implications (Fig. 1 A; Cabukusta and Neefjes, 2018; Bonifacino and Neefjes, 2017). Perinuclear lysosomes are the main places for degradation of cargo delivered by endosomes and autophagosomes, whereas peripheral lysosomes are involved in the regulation of mammalian target of rapamycin complex1 (mTORC1), the master regulator switching between cell growth and autophagy (Johnson et al., 2016; Korolchuk et al., 2011). This also may be connected to the role of lysosomes in lipid homeostasis, as Rab7 seems to control cholesterol export via the lysosomal NPC1 (van den Boomen et al., 2020; Shin and Zoncu, 2020; Castellano et al., 2017). How far the acidification state of perinuclear and peripheral lysosomes also affects their Rab7 and Arl8b mediated localization is still under debate (Ponsford et al., 2021). Thus, it is likely that Rab7 coordinates LE and lysosomal transport and fusion activity in coordination with endosomal biogenesis and cellular metabolism.GEF function and regulation in endosomal maturationThe heterodimeric complex Mon1-Ccz1 was identified as the GEF for Ypt7 in yeast and for Rab7 in higher eukaryotes (Nordmann et al., 2010; Gerondopoulos et al., 2012). The Mon1-Ccz1 complex is an effector of Rab5 (Kinchen and Ravichandran, 2010; Langemeyer et al., 2020; Cui et al., 2014; Li et al., 2015; Poteryaev et al., 2010; Singh et al., 2014), suggesting a direct link to endosomal maturation and Rab turnover (Fig. 1 B). Structural analyses uncovered how the two central longin domains in Mon1 and Ccz1 displace the bound nucleotide from Ypt7 (Kiontke et al., 2017). Unlike yeast, the metazoan Mon1-Ccz1 complex contains a third subunit termed RMC1 or C18orf8 in mammals and Bulli in Drosophila (Vaites et al., 2017; Dehnen et al., 2020; van den Boomen et al., 2020). Even though loss of this subunit impairs endosomal and autophagosomal biogenesis, this subunit does not affect GEF activity toward Rab7 in vitro (Dehnen et al., 2020; Langemeyer et al., 2020), indicating that the general GEF mechanism is conserved across species. As Rab7 is required on LEs, autophagosomes, and lysosomes, spatial recruitment and activity of the Rab7 GEF must be tightly regulated.Rab5 activates the Mon1-Ccz1 GEF complexDuring endosomal maturation, the Mon1-Ccz1 complex is recruited to Rab5- and PI3P-positive endosomes and activates Rab7 for subsequent fusion of endosomes with lysosomes (Nordmann et al., 2010; Poteryaev et al., 2010; Cabrera and Ungermann, 2013; Cabrera et al., 2014; Singh et al., 2014; Fig. 1 C). However, it was postulated that (but remained unclear how) Rab5 affects Rab7 GEF activity. The activity of GEFs is in the simplest way determined in solution, where the respective Rab, which has been loaded with a fluorescent- or radioactive-labeled nucleotide, is incubated with the GEF (Schoebel et al., 2009; Bergbrede et al., 2009). GDP or GTP addition then triggers displacement of the bound nucleotide, which results in a decrease of fluorescence or increase of radioactive signal in solution. Such in-solution assays can uncover the Rab specificity of GEFs yet cannot recapitulate the membrane context and potential regulating factors. Recent approaches therefore used liposomes and prenylated Rab:GDI complexes to address the role of membrane lipids and proteins in GEF activation (Thomas and Fromme, 2016; Thomas et al., 2018; Langemeyer et al., 2020, 2018b; Cezanne et al., 2020; Bezeljak et al., 2020). Details of these reconstituted systems are discussed below. In yeast, prenylated, membrane-bound, and GTP-loaded Rab5-like Vps21 was surprisingly inefficient as a single factor to recruit Mon1-Ccz1 to membranes, whereas addition of PIPs together with Vps21 enhanced recruitment (Langemeyer et al., 2020). However, activity of both the yeast and metazoan Rab7 GEF complexes showed a striking dependence on membrane-bound Rab5-GTP in the GEF assay, whereas PIPs alone were not sufficient to drive GEF activation. These observations demonstrate that the Mon1-Ccz1 complex depends on membrane-bound Rab5 for its Rab7 GEF activity, which nicely explains some of the previous in vivo observations on endosomal Rab5-to-Rab7 exchange (Poteryaev et al., 2010; Rink et al., 2005).This Rab exchange, which occurs similarly on phagosomes (Jeschke and Haas, 2016), is in vivo likely regulated in space and time. Time-lapse microscopy studies revealed that levels of fluorescently labeled Rab5 decreased, while fluorescently labeled Rab7 increased on the surface of a tracked endosome (Poteryaev et al., 2010; Yasuda et al., 2016). Analysis of the spatiotemporal Rab5-to-Rab7 transition in mammalian cells revealed that Rab5-positive endosomes can separate from Rab7-positive membranes, suggesting that a stepwise maturation process also occurs in some cells (Skjeldal et al., 2021). However, in all cases, only some insights on Mon1-Ccz1 regulation are presently available. Phosphorylation is one potential regulatory mechanism in GEF regulation (Kulasekaran et al., 2015). Indeed, yeast Mon1-Ccz1 is a substrate of the vacuolar casein kinase 1 Yck3 (Lawrence et al., 2014). When added to the Rab5-dependent GEF assay, Yck3-mediated phosphorylation inhibited Mon1-Ccz1 GEF activity, presumably by blocking the Rab5 interaction (Langemeyer et al., 2020). How the kinase is in turn regulated and whether this is the only mechanism of Mon1-Ccz1 GEF control is currently unknown.Rab7 activation and function in autophagyThe lysosome is also the destination of the autophagic catabolic pathway. During autophagy, portions of the cytosol, specific organelles, aggregates, or pathogens are engulfed into a double-layered membrane, which upon closure fuses with the lysosome for degradation and reuse of its content (Fig. 2 A; Zhao and Zhang, 2019; Nakatogawa, 2020). Autophagy is a versatile pathway required for adaptation of a cell’s organelle repertoire and quality control.Rab7 is found not just on LEs, but also on autophagosomes (Hegedűs et al., 2016; Gao et al., 2018a), although its precise function seems to differ between organisms (Kuchitsu and Fukuda, 2018). In yeast, the Rab7-homologue Ypt7 mediates HOPS-dependent fusion of autophagosomes with vacuoles (Gao et al., 2018a). In metazoan cells, Rab7 and its effectors PLEKHM1 and WDR91 are required for autolysosome/amphisome-lysosome fusion, yet Rab7 does not seem to directly bind HOPS during fusion of autophagosomes with lysosomes (Xing et al., 2021; McEwan et al., 2015; Gutierrez et al., 2004; Kuchitsu and Fukuda, 2018).Given the striking Rab5 dependence on endosomes in Mon1-Ccz1 activation, the question arises, how does Mon1-Ccz1-mediated Rab7 activation happen on autophagosomes? Some data suggest that yeast and metazoan Rab5 is directly involved in the autophagy process such as autophagosome closure (Ravikumar et al., 2008; Bridges et al., 2012; Zhou et al., 2019, 2017), whereas others do not find direct evidence, for instance in Drosophila (Hegedűs et al., 2016). Studies in yeast revealed that the LC3–like Atg8 protein directly binds and recruits Mon1-Ccz1 to the autophagosomal membrane during starvation, which results in Ypt7 activation as a prerequisite of HOPS-dependent fusion with the vacuole (Gao et al., 2018a; Fig. 2 B). Tight regulation of Mon1-Ccz1 GEF-activity is apparently mandatory to avoid fusion of premature autophagosomes with the vacuole (Fig. 2 C). How Mon1-Ccz1 localization to either endosomes or autophagosomes is coordinated (also with regard to similarities in organelle features; Fig. 2 D) and whether Atg8/LC3 also regulates the activity of the GEF complex are not yet known.Of note, an endosomal-like Rab5-to-Rab7 cascade also occurs on the mitochondrial outer membrane during mitophagy in metazoan cells, a selective pathway to degrade damaged mitochondria (Yamano et al., 2018). Here, Rab5 is activated by a mitochondrially localized Rab5 GEF, followed by Mon1-Ccz1 recruitment and Rab7A activation, which then orchestrates the subsequent mitophagy process. How this process is coupled to autophagosome maturation, and whether Rab7 is then again needed on the formed autophagosome, has not been addressed so far.These data nevertheless demonstrate the adjustable recruitment of Mon1-Ccz1 during endosomal maturation and autophagosome formation and even to the mitochondrial surface. Targeting of the Mon1-Ccz1 complex is likely coordinated between all these processes.A role for ER-endosome MCSs in endosome maturationEndosomes form MCSs with the ER. Such contact sites have multiple roles ranging from lipid transport to ion exchange (Scorrano et al., 2019; Reinisch and Prinz, 2021). The endosome-ER contact depends on Rab7 and contributes to transport and positioning of endosomes, supports endosomal fission, and facilitates endocytic cargo transport and cholesterol transfer between LEs and the ER (Rocha et al., 2009; Friedman et al., 2013; Rowland et al., 2014; Raiborg et al., 2015; Jongsma et al., 2016). Rab7 activation via the Mon1-Ccz1 complex is required for cholesterol export from the lysosome, likely in the context of MCSs. Rab7 binds to the NPC1 cholesterol transporter and may thus promote cholesterol export only at MCSs with the ER or other organelles (van den Boomen et al., 2020). The ER is also involved in endosome maturation, which requires an MCS between Reticulon-3L on the ER and endosomal Rab9. In fact, Rab9 is recruited shortly before the Rab5-to-Rab7 transition (Wu and Voeltz, 2021; Kucera et al., 2016). How Rab9 activation and MCS formation are coordinated with endosomal maturation has not yet been revealed. It is likely that the spatial positioning of endosomes (Fig. 1 A), their acidification, and TORC1 activity also contribute to this process (Bonifacino and Neefjes, 2017; Johnson et al., 2016).Retromer opposes Rab7 activationRetromer is a conserved heteropentameric complex that mediates the formation of vesicular carriers at the endosome and thus allows the transport of receptors back to the Golgi or plasma membrane. The complex consists of a trimeric core (Vps35, Vps26, and Vps29), which binds either a SNX1-SNX4 heterodimer or a SNX3 monomer (Simonetti and Cullen, 2018; Leneva et al., 2021; Kovtun et al., 2018). Retromer is an effector of Rab7, but also recruits the Rab7 GAP TBC1D5 in metazoan cells (Rojas et al., 2008; Kvainickas et al., 2019; Jimenez-Orgaz et al., 2018; Distefano et al., 2018; Seaman et al., 2009). This dual function of retromer may facilitate the formation of endosomal tubules after the Rab5-to-Rab7 transition, and these tubules eventually lose Rab7 once scission has occurred (Jongsma et al., 2020).It is not yet clear how conserved the Rab7-retromer-GAP connection is. Yeast retromer is also an effector of the Rab7-like Ypt7 and coordinates protein recycling at the endosome (Liu et al., 2012; Balderhaar et al., 2010), yet a role of a Rab7 GAP has not been described. However, yeast retromer also binds to the Rab5 GEFs Vps9 and Muk1 (Bean et al., 2015), which suggests that both Rab5 and Rab7 function contribute to efficient tubule formation at the endosome. Whether and how the Rab7 GEF Mon1-Ccz1 is functionally coordinated with retromer will be a topic of future studies.GEF regulation along the endomembrane systemIn the previous section, we focused mainly on the role of the Rab7 GEF in the context of endosome and autophagosome maturation. However, the timing of GEF activation and the subsequent recruitment of their target Rabs is critical for all membrane trafficking processes along the endomembrane system to guarantee maintenance of intracellular organelle organization. Rabs in turn interact with effectors, and effectors such as the lysosomal HOPS complex not only bind SNAREs but also catalyze their assembly and thus drive membrane fusion (Fig. 3 A). The spatiotemporal regulation of GEF activation is therefore at the heart of organelle biogenesis and maturation, and thus membrane trafficking. Within this section, we will now broaden our view by comparing different regulatory principles of GEFs.Open in a separate windowFigure 3.Regulatory mechanisms influence the activity of GEFs.(A) Hierarchical cascade of factors controlling membrane fusion. GEFs integrate various signals and initiate a cascade of protein activities, finally leading to membrane fusion. Signaling lipids, the presence of cargo proteins, upstream GTPases, and kinases influence the activity of GEFs and therefore determine Rab GTPase activation. Consequently, effector proteins such as tethering factors are recruited. This ultimately leads to SNARE-mediated lipid bilayer mixing and membrane fusion. (B) A Rab cascade in yeast exocytosis. Active Ypt32 and PI4P (yellow) on late Golgi compartments and secretory vesicles recruit the GEF Sec2, which in turn promotes activation and stable membrane insertion of the Rab Sec4. (C) Mon1-Ccz1 regulation by phosphorylation. Mon1-Ccz1 is recruited to and activated on LEs by coincidence detection of membrane-associated Rab5 and PI3P (red, Fig. 1 C) and promotes stable membrane insertion of Rab7. This process is terminated by Mon1-Ccz1 phosphorylation by the type I casein kinase Yck3 in yeast (orange). (D) A positive feedback loop of GEF activation on endocytic vesicles and EEs. The Rab5 GEF Rabex-5 binds ubiquitinated cargo on endocytic vesicles and is autoinhibited. Rab5 recruits Rabaptin-5, which binds Rabex-5 and releases the GEF from autoinhibition, generating a positive feedback loop. (E) Membrane factors determine GEF activity of TRAPPII at the trans-Golgi. TRAPPII activity for the Rab Ypt32 requires membrane-associated Arf1 and PI4P. (F) The length of the hypervariable domain of Golgi Rabs defines the substrate specificity for TRAPP complexes. The yeast Rab GTPases Ypt1 and Ypt32 differ in the length of their C-terminal HVD (box). TRAPPII and TRAPPIII complexes have the same active site, which is positioned away from the membrane, and thus discriminate Rab accessibility. (G) Phosphorylation as a mechanism to promote GEF activity. DENND1 GEF activity is autoinhibited, which is released by Akt-mediated phosphorylation. For details, see text.A Rab cascade in exocytosisAnother well-characterized Rab cascade is involved in the exocytic transport of secretory vesicles from the trans-Golgi network to the plasma membrane. At the trans-Golgi, the GEF transport protein particle II (TRAPPII) activates the Rab GTPase Ypt32, which then recruits the GEF Sec2 to secretory vesicles. Sec2 in turn activates the Rab Sec4, which binds the Sec15 subunit of the Exocyst tethering complex and allows vesicles to dock and fuse with the plasma membrane (Fig. 3 B; Walch-Solimena et al., 1997; Ortiz et al., 2002; Dong et al., 2007; Itzen et al., 2007). This cascade is conserved in humans. During ciliogenesis at the plasma membrane, the Ypt32 homologue Rab11 recruits the GEF Rabin 8, which in turn activates the human Sec4 homologue Rab8, a process regulated by phosphorylation (Hattula et al., 2002; Wang et al., 2015; Knödler et al., 2010). Interestingly, yeast Sec2 not only is a GEF, but also interacts with the Sec4 effector Sec15 (Medkova et al., 2006), a principle also observed in the endocytic Rab5 activation cycle, where the GEF Rabex5 interacts with the Rab5 effector Rabaptin-5. This dual role may also apply to Mon1-Ccz1, as the Mon1 homologue in C. elegans, SAND1, and yeast Mon1-Ccz1 can bind the HOPS tethering complex (Poteryaev et al., 2010; Nordmann et al., 2010).At the Golgi, phosphatidylinositol-4-phosphate (PI4P) contributes to directionality and spatiotemporal regulation of the exocytic Rab cascade. Sec2 binds both Ypt32 and PI4P on secretory vesicles via two binding sites, a process called coincidence detection. However, PI4P binding inhibits the interaction of Sec2 with Sec15. As vesicles reach the cell periphery, PI4P levels drop by the activity of Osh4, a lipid transporter, which allows Sec2 to bind the Exocyst subunit rather than Ypt32 (Ling et al., 2014; Mizuno-Yamasaki et al., 2010). In addition, Sec2 is phosphorylated by the plasma membrane–localized casein kinases Yck1 and Yck2 (Stalder et al., 2013; Stalder and Novick, 2016), resulting in effector recruitment rather than further Rab activation.Such a regulation may also apply to yeast Mon1-Ccz1. Anionic phospholipids and PI3P support Mon1-Ccz1 recruitment to liposomes and vacuoles (Langemeyer et al., 2020; Cabrera et al., 2014; Lawrence et al., 2014), whereas phosphorylation of the complex by the casein kinase Yck3 inhibits the binding of Mon1-Ccz1 to the Rab5-like Ypt10 and consequently reduces its GEF activity toward Rab7 (Fig. 3 C; Langemeyer et al., 2020). These observations suggest that the phosphorylation of GEFs by kinases may be a general regulatory principle in Rab cascades.Autoinhibition controls the Rab5 GEFAnother widely used regulatory mechanism is the autoinhibition of GEFs to control their activity. This has been analyzed in detail for the early endosomal Rab5-specific GEF Rabex-5, which interacts with the Rab5-effector Rabaptin-5 (Horiuchi et al., 1997). One factor for Rabex-5 recruitment to endocytic vesicles are ubiquitinated cargo proteins at the plasma membrane (Fig. 3 D; Mattera et al., 2006; Lee et al., 2006). Yet, isolated Rabex-5 has only low GEF activity in vitro (Delprato and Lambright, 2007). Structural analysis revealed that binding of Rabaptin-5 to Rabex-5 causes a rearrangement in the Rabex-5 C-terminus, which releases the GEF from autoinhibition and therefore facilitates nucleotide exchange of Rab5 (Delprato and Lambright, 2007; Zhang et al., 2014). On endosomes, increasing amounts of Rab5-GTP further promotes recruitment of the Rabex-5–Rabaptin-5 complex, resulting in a positive feedback loop of Rab5 activation and GEF recruitment (Lippé et al., 2001). Overall, Rabex-5 GEF activity is regulated by autoinhibition, a feedback loop with the Rab5 effector protein Rabaptin-5, and ubiquitinated cargo, which guarantees precise timing in establishing a Rab5-positive endosome. Of note, the Mon1 subunit of the Rab7 GEF can displace Rabex-5 from endosomal membranes (Poteryaev et al., 2010), which suggests a negative feedback loop of the Rab5 activation cascade once the next GEF is present.Regulation of Arf1 GEFs at different Golgi subcompartmentsThese key principles of GEF regulation in GTPase cascades are also found for Arf GTPases. Arf GTPases are soluble in their GDP-bound state by shielding their N-terminal myristate anchor in a hydrophobic pocket. Like Rabs, Arf GTPases are activated by specific GEFs, and their inactivation requires a specific GAP (Sztul et al., 2019). However, this review only highlights some key findings in the regulation of Rab GEFs and does not address regulation of the corresponding GAPs. Once activated, Arfs insert their lipid anchor and an adjacent amphipathic helix into membranes and are then able to bind effector proteins (Sztul et al., 2019). One of the best-studied Arf-GEFs is Sec7, which activates Arf1, an Arf GTPase involved in intra-Golgi trafficking (Achstetter et al., 1988). Studies on yeast Sec7 revealed that the protein is autoinhibited in solution and depends on three small GTPases—Arf1, the Rab Ypt1, and the Arf-like Arl1—for recruitment to the Golgi, a process supported by anionic lipids found in the late Golgi compartment. Importantly, the late Golgi Rabs Ypt31/32 strongly stimulate GEF activity (McDonold and Fromme, 2014; Richardson et al., 2012, 2016), indicating allosteric activation, as observed for Rab5-dependent Mon1-Ccz1 activation (Langemeyer et al., 2020). In this process, Sec7 dimerizes and promotes Arf1 recruitment and thus establishes a positive feedback loop. Interestingly, membrane binding of two additional Arf1 GEFs of the early Golgi, Gea1/2, depends on Rab1/Ypt1 and neutral membranes. Under these conditions, Gea1/2 is released from autoinhibition, although no positive feedback loop was observed (Gustafson and Fromme, 2017). Thus, Arf GEF regulation and Arf activation are tightly linked to multiple small GTPases and the membrane environment to establish Golgi compartments.Regulation and specificity of TRAPP complexes at the GolgiArf1 activation is also linked to the activation of Golgi-specific Rabs. Arf1-GTP binds to the highly conserved TRAPP GEF complexes at the Golgi (Fig. 3 E). Yeast and mammalian cells contain two TRAPP complexes. In yeast, both complexes share seven core components. TRAPPIII in addition contains Trs85, while accessory TRAPPII subunits are instead Trs130, Trs120, Trs65, and Tca17. Metazoan TRAPP complexes contain additional subunits (Lipatova and Segev, 2019).Interestingly, both complexes share the same catalytic site for Rab1/Ypt1 and Rab11/Ypt32. However, TRAPPIII provides GEF activity toward Rab1/Ypt1. Initially, it was proposed that TRAPPII can activate both Rab1/Ypt1 and Rab11/Ypt32 (Thomas et al., 2019, 2018; Thomas and Fromme, 2016; Riedel et al., 2018); however, it was recently shown that the TRAPPII complex is specific for Rab11/Ypt32 (Riedel et al., 2018; Thomas et al., 2019). Reconstitution of GEF activity on liposomes helped here to unravel TRAPP complex substrate specificity, since in solution assays are not adequate to address some of the features important for specific interactions: Rab11/Ypt32 has a longer HVD between the prenyl anchor and the GTPase domain compared with Rab1/Ypt1 (Fig. 3 F, box). The HVD not only binds TRAPPII but also stretches a longer distance from the membrane (Fig. 3 F). Thereby it allows Rab11/Ypt32, but not Rab1/Ypt1, to reach the active site of membrane-bound TRAPPII. Thus, substrate specificity is controlled by the distance of the GTPase domain from the membrane surface, since the active site seems to be located on the opposing site of the complex from the site of membrane interaction (Fig. 3 F; Thomas et al., 2019). The smaller TRAPPIII has its active site closer to the membrane, binds Ypt1 via its shorter HVD, and facilitates its activation, while Ypt32 with its longer HVD may be positioned too far away from the active site. In addition, both complexes require their respective membrane environment for optimal activity, indicating how Arf and Rab GEFs cooperate in Golgi biogenesis.The GEF DENND1 requires Arf5 for Rab35 activationRecently, another example of Arf-mediated Rab activation was reported (Kulasekaran et al., 2021). Rab35, an endocytic Rab found at the plasma membrane and REs (Sato et al., 2008; Kouranti et al., 2006), is involved in cell adhesion and cell migration by controlling the trafficking of β1-integrin and the EGF receptor (Klinkert and Echard, 2016; Allaire et al., 2013). Arf5 binds the Rab35 GEF DENND1 and stimulates its GEF activity, with dysregulation of this cascade linked to glioblastoma growth (Kulasekaran et al., 2021). DENND1 GEF activity is initially autoinhibited and relieved by phosphorylation via the central Akt kinase (Fig. 3 G; Kulasekaran et al., 2015). Similarly, another DENN-domain containing GEF, DENND3, is phosphorylated by the autophagy-specific ULK kinase and then activates Rab12, a small GTPase involved in autophagosome trafficking (Xu et al., 2015). Thus, it seems that Rab GEF activation is more generally linked to other trafficking proteins, such as Arfs, and controlled by kinases and likely also phosphatases.Lessons from reconstitutionOrganelle biogenesis and maintenance in the endomembrane system are tightly linked to the correct spatial and temporal activation of Rab GTPases. A small yeast cell gets by with 11 Rabs, while human cells encode >60 (Hutagalung and Novick, 2011). Rab activation, and therefore membrane identity, of each organelle depends on the cognate GEF. This puts GEFs into the driver’s seat of any Rab-directed function at cellular membranes. It seems that GEFs integrate, by several regulatory loops, incoming signals from various sources such as membrane composition, cargo proteins, upstream GTPases, or kinases/phosphatases (Fig. 3 A). Yet our insights on the specific membrane targeting and regulation of GEFs remain incomplete for want of available experimental approaches. We briefly discuss here how recent advances on the reconstitution of GEF-mediated Rab activation at model membranes have advanced our understanding of organelle maturation and biogenesis.Reconstitution of any reaction to uncover the essential constituents is limited by the available tools. GEFs, Rabs, Sec18/Munc1 proteins, tethering factors, and SNAREs are for instance required for membrane fusion (Fig. 3 A). Initial assays focused on SNAREs and revealed their important but rather inefficient fusogenicity (Weber et al., 1998). Further analyses uncovered critical activation steps for SNAREs (Malsam et al., 2012; Pobbati et al., 2006; Südhof and Rothman, 2009; Jahn and Scheller, 2006), yet fusion at physiological SNARE concentrations in various in vitro systems does not occur, unless assisted by chaperoning Sec1/Munc18 proteins and tethering factors (Bharat et al., 2014; Lai et al., 2017; Mima and Wickner, 2009; Ohya et al., 2009; Wickner and Rizo, 2017). Most tethers again depend on Rabs for their localization, and Rab localization to membranes requires a GEF (Cabrera and Ungermann, 2013), whose activity can be a limiting factor for fusion (Langemeyer et al., 2020, 2018b). The long avenue of understanding the mechanism and regulation of membrane fusion exemplifies the challenges in dissecting the complexity of a cellular reaction, but also demonstrates the powerful insights obtained from reconstitution of these processes.GEFs determine the localization of the corresponding Rab, and consequently, Rabs follow their GEF if they are mistargeted (Gerondopoulos et al., 2012; Blümer et al., 2013; Cabrera and Ungermann, 2013). However, these anchor-away approaches completely bypass the tight cellular regulation of GEF activation by the mistargeting and additional overexpression of the GEF protein and may allow only statements about GEF/substrate specificity. The spatiotemporal activation of each GEF at the right organelle is vital for the timing of all downstream reactions. GEFs are recruited to membranes by coincidence detection, which includes membrane lipids such as PIPs, membrane packaging defects, and peripheral membrane proteins such as upstream Rabs or other small GTPases. This recruitment is often accompanied by the release from autoinhibition, which may be triggered or inhibited by other regulatory processes such as phosphorylation. It comes as no surprise that pathogens such as Legionella and Salmonella take advantage of the central function of GEFs to establish and nourish their intracellular organellar niche by manipulating small GTPase activity (Spanò and Galán, 2018).To understand the specificity of Rab GEFs (and GAPs), mostly very simplified systems were used. Most GEF assays analyze soluble Rabs loaded with fluorescent 2′-O-(N-methylanthraniloyl) (MANT)-nucleotide or radioactively labeled GTP/GDP and soluble GEF in a test tube, where nucleotide exchange activity is observed upon addition of unlabeled nucleotide (Fig. 4 A). This strategy allows the identification of substrate (Rab) specificity of GEFs, but could also lead to misleading results, as pointed out earlier on the example of the TRAPP complexes and Rab1/Ypt1 or Rab11/Ypt32. In addition, GEF-Rab pairs negatively regulated by one of the above principles could easily be missed.Open in a separate windowFigure 4.Approaches to determine GEF activity in vitro. Methods to determine GEF activity for Mon1-Ccz1. In all approaches, Rab7 is preloaded with fluorescent MANT-GDP. Fluorescence decreases upon GEF-mediated nucleotide exchange. (A) GEF assays. (Ai) In-solution Rab GEF assay. Mon1-Ccz1 (blue, Bulli/Rmc1/C18orf8 subunit, indicated by unlabeled hexagon) and Rab7 (gray) are freely diffusible in the test tube, which results in random collision and Rab activation. (Aii) GEF-mediated activation of artificially recruited Rab7 on liposomes. Rab7 with a C-terminal 6xHis-tag is permanently immobilized on membranes containing the cationic lipid DOGS-NTA. Mon1-Ccz1 unspecifically binds to this membrane surface and activates Rab7. Diffusion is limited to the membrane surface, thus increasing chances of interactions. (Aiii) Reconstitution of Rab5-mediated Rab7 activation by Mon1-Ccz1 on liposomes. Chemically activated, prenylated Rab5 (green), delivered to the membrane by the Rab Escort Protein (REP), allows Mon1-Ccz1 recruitment and Rab7 activation from the GDI complex (see text for further details). (B) Summary of Ai–Aiii. pren., prenylation.As Rabs and GEFs function on membranes, we and others adopted strategies for measuring Rab activation by GEFs on membranes (Fig. 4 B). In a first approach, Rab and other small GTPases (Sot et al., 2013; Schmitt et al., 1994) were immobilized with C-terminal hexahistidine tags on liposomes containing the polycationic lipid 1,2-dioleoyl-sn-glycero-3-[(N-(5-amino-1-carboxypentyl)iminodiacetic acid)succinyl] (DOGS-NTA) and observed higher activity of the added GEF (Cabrera et al., 2014; Thomas and Fromme, 2016). A drawback of this technique is the artificial membrane composition. To avoid potential artifacts of unnaturally charged membranes and permanently membrane-bound Rab, recent studies relied on prenylated Rabs in complex with GDI. Reflecting the natural source of the cytoplasmic Rab pool, this complex was used as a GEF substrate in the presence of liposomes mimicking the natural membrane composition (Cezanne et al., 2020; Bezeljak et al., 2020; Langemeyer et al., 2020, 2018b; Thomas et al., 2018, 2019; Thomas and Fromme, 2016).Even though these observations are recent, the outcome and the understanding of GEF regulation is encouraging. For the Rab5 GEF complex consisting of Rabex5 and Rabaptin5, GEF-dependent Rab5 recruitment to membranes revealed a self-organizing system, nonlinear Rab5 patterning, and collective switching of the Rab5 population (Bezeljak et al., 2020; Cezanne et al., 2020). This is in agreement with mathematical modeling and predictions on bistability and ultrasensitivity of Rab networks (Del Conte-Zerial et al., 2008; Barr, 2013). For the Golgi-resident TRAPPII and TRAPPIII complexes, the membrane composition, the length of the Rab HVD, and the presence of membrane-bound Arf1 determined the GEF specificity for their Rabs (Fig. 3 F; Thomas et al., 2019, 2018; Thomas and Fromme, 2016; Riedel et al., 2018), which is nicely supported by recent structural analyses of yeast and metazoan TRAPPIII (Galindo et al., 2021; Joiner et al., 2021)Our own data uncovered that the yeast and metazoan Mon1-Ccz1(-RMC1) complex required membrane-bound Rab5-GTP to activate Rab7 out of the GDI complex (Langemeyer et al., 2020). Surprisingly, Rab5-GTP not only determined membrane binding of Mon1-Ccz1, but also activated the GEF on membranes by a yet-unknown mechanism (Fig. 1 C). Phosphorylation of yeast Mon1-Ccz1 by the casein kinase Yck3 inhibited this activation, demonstrating possible regulation of GEF activity (Fig. 3 C). Importantly, this finding agrees with the observed Rab5-to-Rab7 switch in vivo (Poteryaev et al., 2010; Rink et al., 2005).Taken together, the available tools open exciting avenues for our understanding of organelle maturation. Reconstitution will allow the investigation of an entire Rab cascade and its regulation by kinases or membrane lipids. It will be possible to determine the cross-talk with lipid kinases and observe possible regulatory loops between Rabs and PI kinases (Tremel et al., 2021). We are confident that such analyses, complemented by in vivo analyses of Rabs or other small GTPases and their GEFs, will clarify the underlying mechanism of organelle maturation and biogenesis along the endomembrane system of eukaryotic cells.     

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.     

Perspectives on: Conformational coupling in ion channels: Conformational coupling in BK potassium channels

Large conductance calcium- and voltage-dependent BK potassium channels (aka BK_Ca, MaxiK, Slo1, KCa1.1, and KCNMA1) are expressed in a wide variety of tissues throughout the body and are activated by both intracellular Ca²⁺ and membrane depolarization. Owing to these properties, BK channels participate in diverse physiological processes from electrical excitability in neurons and secretory cells, and regulation of smooth muscle tone to tuning of auditory hair cells (Vergara et al., 1998; Ghatta et al., 2006). The response to voltage and Ca²⁺ allows BK channels to integrate electrical and calcium signaling, which is central to their physiological role. Understanding how BK and other multimodal channels are regulated by and integrate diverse stimuli is not only physiologically important but also relevant to the topic of conformational coupling. As a voltage- and ligand-dependent channel, BK channels contain both voltage-sensor and ligand-binding domains as well as a gate to regulate the flow of K⁺ through the pore. Coupling of conformational changes in one domain to another provides the basis for transducing voltage and ligand binding into channel opening and, therefore, defines, together with the functional properties of the gate and sensors, the signal transduction properties of the channel. The goal of this perspective is to provide an overview on the role and molecular basis of conformational coupling between functional domains in BK channels and outline some of the questions that remain to be answered.

BK channel structure

The BK channel is a member of the superfamily of voltage-gated channels that assembles as a homotetramer of pore-forming Slo1 (α) subunits. Each subunit contains seven transmembrane segments (S0–S6), including an S5–S6 pore gate domain (PGD) and voltage-sensor domain (VSD) that is likely to include S0–S4 segments (Liu et al., 2008) with charged voltage-sensing residues in S2, S3, and S4 (Fig. 1 A; Ma et al., 2006). In addition, a large C-terminal cytoplasmic domain (CTD) consisting of two homologous regulator of K⁺ conductance domains (RCK1 and RCK2) contains binding sites for Ca²⁺ and other ligands (Hou et al., 2009). The CTDs form a tetrameric gating-ring structure whose conformation changes upon Ca²⁺ binding. Crystal structures of the isolated BK channel gating ring and related prokaryotic Ca²⁺-activated K⁺ channel MthK have been solved in the presence and absence of Ca²⁺ (Jiang et al., 2002; Ye et al., 2006; Wu et al., 2010; Yuan et al., 2010, 2012). The atomic structure of the transmembrane domain has yet to be determined but is assumed to be homologous with that of voltage-gated potassium (K_V) channels, consistent with a low resolution cryo-EM structure of the entire channel(Wang and Sigworth, 2009). Although BK channels formed from Slo1 alone are fully functional, each channel may also coassemble with up to four regulatory subunits of which several subtypes exist (β1–4 and γ1–4; Brenner et al., 2000; Yan and Aldrich, 2012). Regulatory subunits tune BK channel function in different tissues, contain one or two transmembrane segments, occupy sites adjacent to the VSD (Liu et al., 2010), and act in part to regulate voltage-dependent gating (Bao and Cox, 2005; Yan and Aldrich, 2010).Open in a separate windowFigure 1.BK channel structure. (A) Topology of the pore-forming Slo1 subunit, including VSD, PGD, and CTD. Charged residues in the VSD that are important for voltage sensing are highlighted in yellow. The CTD contains binding sites for Ca²⁺, Mg²⁺, and heme. (B) Schematic organization of functional domains in the tetrameric channel. S6 segments in the PDG are connected to the CTD gating ring through S6-RCK1 linkers.

BK channel function

The response of BK channels to voltage and Ca²⁺ is illustrated in Fig. 2 by plotting steady-state open probability (P_O; Fig. 2 A) and Log(P_O) (Fig. 2 B) versus voltage at different [Ca²⁺]_i (0–100 µM) for heterologously expressed Slo1 channels (Horrigan and Aldrich, 2002). The P_o-V relation in 0 Ca²⁺ (∼0.5 nM) shows channels can be fully activated in the absence of Ca²⁺ binding, but only at voltages approaching +300 mV. Calcium, to a first approximation, shifts P_o-V to more negative voltages (Fig. 2 A), allowing the channel to activate in a physiological voltage range. However, plotting the data on a log scale reveals that Ca²⁺ does not simply shift the curve but rather increases Log(P_O) in a nearly voltage-independent manner until P_O saturates (Fig. 2 B). This response indicates that BK channels are activated independently by voltage- and Ca²⁺-sensors. Furthermore, Log(P_O) becomes almost voltage independent at extreme negative voltages indicating that channels can open in the absence of voltage-sensor activation, a conclusion supported by gating current measurements (Horrigan and Aldrich, 1999).Open in a separate windowFigure 2.Voltage- and Ca²⁺-gating of BK channels. (A) P_O-V relations for mSlo1 estimated as G_K/G_Kmax from macroscopic tail currents after 30 ms voltage pulses in 0–100 µM Ca²⁺. (B) Log(P_o)-V relations extend P_O to <10⁻² using steady-state unitary current recordings from macropatches. A and B represent mean ± SEM and are fit (solid curves) by the HA model (Horrigan and Aldrich, 2002). The increase in Log(P_o) from 0 Ca²⁺ to saturating 100 µM Ca²⁺ at −120 mV (where voltage-sensors are in the resting state) reflects Ca²⁺-sensor/gate coupling energy (ΔΔGCOCa). The increase in Log(P_o) in 0 Ca²⁺ from −120 mV to ∼+300 mV (where voltage sensors are fully activated) reflects voltage-sensor/gate coupling (ΔΔGCOV). The values of ΔΔGCOCa and ΔΔGCOV are determined from the change in the free energy of the gate (e.g., ΔΔGCOCa=ΔGCO[100 Ca]−ΔGCO[0 Ca], where ΔG_CO = kTln[P_O/(1 ? P_O)]), and, in the case of ΔΔGCOV, the measurement at −120 mV must be extrapolated to positive voltages (dashed line) to correct for the weak voltage dependence of the C-O transition (z_L = 0.3 e). (C) Allosteric model indicates the possible conformations of the gate (C and O), voltage sensors (R and A), and Ca²⁺ sensors (X and X-Ca²⁺) in each of four subunits and allosteric factors (C, D, and E) that describe the energetic coupling between these three parts of the channel. J and L are voltage-dependent equilibrium constants with zero-voltage values J₀ and L₀ and partial charges (z_J, z_L), K = [Ca²⁺]/K_D, where K_D is the elementary Ca²⁺ dissociation constant for the closed channel.

An allosteric mechanism of BK channel gating

That BK channels exhibit basal activity in the absence of voltage-sensor activation and Ca²⁺ binding implies that sensor/gate coupling is not an obligatory process. That is, sensor activation promotes but is not required for channel opening. The ability of voltage or Ca²⁺ sensors in different subunits to influence a concerted conformational change (opening) in a nonobligatory fashion is well described in terms of allosteric mechanisms (Monod et al., 1965). A dual allosteric model (Fig. 2 C, HA model) was used to fit the steady-state data in Fig. 2 (A and B, curves), accounts for many other features of Slo1 gating (Horrigan and Aldrich, 1999, 2002; Horrigan et al., 1999), and provides a useful framework for analyzing BK channel gating in terms of domain/domain interactions. The HA (Horrigan-Aldrich) model asserts that the channel gate can undergo a closed to open (C-O) conformational change that is regulated by four independent and identical voltage- and Ca²⁺-sensors. Voltage sensors can be in a resting (R) or activated (A) conformation, whereas Ca²⁺ sensors can be Ca²⁺ free (X) or Ca²⁺ bound (X-Ca²⁺). The function of each domain is defined by equilibrium constants for gate opening (L), voltage-sensor activation (J), and Ca²⁺ binding (K). The coupling, or energy transfer, between domains is represented by allosteric factors (C, D, and E) which define the ability of a transition in one domain to affect the equilibrium constant in another.

The energetics of conformational coupling

The structures of BK and homologous channels provide important clues concerning the molecular basis of conformational coupling, as discussed below. However, many features of coupling can only be resolved through structure-function analysis using site-directed mutagenesis. A prerequisite to such analysis is to quantify coupling interactions represented by allosteric factors C, D, and E in the HA model. Mutations that alter channel activity may perturb sensors, the gate, or their coupling. Therefore, it is crucial to distinguish changes in coupling from changes in the function of sensor or gate. One way to do this is by fitting steady-state data over a wide range of voltage and [Ca²⁺] as in Fig. 2 (A and B) to determine all parameters in the HA model. However, a more direct and model-independent approach to measure coupling is to determine the energetic effect on one domain of forcing coupled domains into defined conformations (e.g., all activated or deactivated) under extreme stimulus conditions (see also Chowdhury and Chanda in this issue). For example, the total coupling between all Ca²⁺ sensors and the gate (ΔΔGCOCa=5.0 kcal mol−1) can be determined by comparing Log(P_o) in 0 Ca²⁺ and saturating 100 µM Ca²⁺ at extreme negative voltages where voltage sensors are in the resting state, and voltage-sensor/gate coupling (ΔΔGCOV=7.6 kcal mol−1) can be determined by comparing Log(P_o) at extreme negative and positive voltages in 0 Ca²⁺ (Fig. 2 B; Horrigan and Aldrich, 2002). In general, the HA model predicts that channels can occupy many different open and closed states, defined by the number of voltage- and Ca²⁺-sensors activated in each channel. But under extreme stimulus conditions, where all voltage- or Ca²⁺-sensors are either activated or deactivated, gating reduces to a single closed and open state. For a two-state process, the free energy difference between closed and open (ΔG_CO) can be defined in terms of the equilibrium constant (P_o/1-P_o) for the C-O transition (i.e., ΔG_CO = kTln[P_O/(1 ? P_O)]; Chowdhury and Chanda, 2010, 2012). Thus, changes in Log(P_o) in Fig. 2 B reflect the energetic effects of voltage- and Ca²⁺-sensor activation on the gate. Similarly, a weak coupling between voltage- and Ca²⁺-sensors (ΔΔGRACa=0.5 kcal mol−1) has been measured by comparing the effects of 0 Ca²⁺ and saturating Ca²⁺ on voltage-sensor activation measured with gating currents while the gate is closed (Horrigan and Aldrich, 2002).Although measurement of Ca²⁺-sensor/gate coupling as in Fig. 2 B is relatively straightforward, measurement of voltage-sensor/gate coupling is subject to several challenges (Fig. 3). The relationship between gating and voltage-sensor/gate coupling is illustrated in Fig. 3 A by expanding the HA model to show the four combinations of states that the gate (C and O) and voltage-sensor (R and A) can assume in a single subunit (RC, RO, AC, and AO) together with the equilibrium constants between them. The equilibrium constant for the C-O transition increases from L when the voltage sensor is in the resting state (Fig. 3 A, #1) to LD when a voltage sensor is activated (Fig. 3 A, #2). Thus the voltage-sensor/gate coupling factor D can be determined by comparing P_o at extreme voltages (Fig. 3 B, #1 and 2), with the expectation that P_o/(1-P_o) will increase by a factor of D⁴ when all four voltage sensors are activated. One challenge is the large dynamic range of these measurements, which span a seven-order-of-magnitude change in P_o and require both macroscopic and unitary current recordings in the same patch and a high level of expression that cannot always be achieved with mutant channels. A related challenge is that P_o in 0 Ca²⁺ approaches saturation near unity at +300 mV as voltage sensors become fully activated (Fig. 2 A). As P_o approaches unity, determination of the equilibrium constant P_o/(1-P_o) becomes problematic and coupling energy can therefore be underestimated, especially in the presence of Ca²⁺ or with mutants that are easier to open than the WT because P_o may saturate before voltage sensors are fully activated. Finally, the C-O conformational change, as in many ion channels, has a weak intrinsic voltage dependence (L(V)) that must be taken into account when comparing P_o at extreme voltages (Figs. 2 B and 3 B, dashed lines).Open in a separate windowFigure 3.The energetics of voltage sensor/gate coupling. (A) Voltage sensor/gate states in a single subunit. Numbers refer to data in B, C, and D used to determine the indicated equilibrium constants. (B) P_O-V determines L (1) and LD⁴ (2). Dashed lines indicate the predicted voltage dependence of P_O with all voltage sensors either in the resting or activated state and z_L = 0.3 e. (C) Normalized Q_C-V for closed channels determined from ON gating currents (inset) is fit by a Boltzmann function with z_J = 0.58 e and defines J (3). (D) Q_O-V relation for open channels defines JD (4). Q_O is estimated by fitting the foot of the q_a-V relation (q_a = kTdln(P_O)/dV) with a Boltzmann function. The shift between Q_O and Q_C relations defines the coupling factor D.A complementary approach to measure voltage-sensor/gate coupling is to determine the effect of channel opening on voltage-sensor activation. The equilibrium constant for the R-A transition increases from J when channels are closed to JD when channels are open (Fig. 3 A, #3 and 4). The equilibrium constant J is determined from the charge distribution for closed channels (Q_C-V; Fig. 3 C). Q_c can be measured by integrating ON gating currents (Fig. 3 B, inset) because voltage-sensor activation in BK channels is rapid and occurs while most channels remain closed. The equilibrium constant JD is determined from the charge distribution for open channels (Q_O-V; Fig. 3 D). Q_O cannot be determined from gating currents because BK channels close rapidly. However, Q_O can be estimated in a model-independent fashion from the log-slope of the P_o-V relation (q_a = kTdln(P_O)/dV), as shown in Fig. 3 D (Horrigan and Aldrich, 2002). The Q_O-V and Q_C-V relations have the same shape but are shifted relative to each other by a voltage directly proportional to coupling energy for a single voltage sensor (ΔV=kT ln(D)/zJ=ΔΔGCOV/4zJ; Ma et al., 2006). This method yields similar results as in Fig. 3 B and avoids challenges relating to the voltage dependence of L(V), or P_O saturation. However, gating currents are difficult to measure in BK channels and require an even higher level of expression than in Fig. 3 B.

Molecular mechanisms of conformational coupling

Although the energetics and biophysical mechanisms of sensor/gate and sensor/sensor coupling in BK channels are well defined, many fundamental questions remain concerning the molecular basis of these interactions. Where in the channel do they occur? What are the identity and nature of amino acid interactions involved? When do the interactions occur during channel activation? These questions are broad because many potential sites of domain/domain interaction exist, the conformational changes that these domains and their interfaces undergo during gating are not all understood in detail, and insight provided by the structure of BK or homologous channels is in many cases limited. For example, although voltage-sensor/gate coupling is likely to be mediated in large part by interfaces between the VSD and PGD as in Kv channels, there is also a unique interface between the VSD and CTD in BK channels that may contribute to this process. Intracellular Mg²⁺ is a BK channel activator that is coordinated by residues in both VSD and CTD and interacts electrostatically with the voltage sensor, indicating (together with cross-linking experiments) that these domains come in contact (Yang et al., 2007, 2008). The primary functional effect of Mg²⁺ is to enhance voltage-sensor/gate coupling, suggesting that VSD/CTD interaction could provide a basis for this process (Horrigan and Ma, 2008). VSD/CTD interaction must also mediate coupling between voltage and Ca²⁺ sensors. However information about the VSD/CTD interface is incomplete. Structures of BK and MthK channel CTDs and the Kv channel VSD help identifying residues that may lie at this interface. But the complete BK channel structure including the VSD/CTD interface has yet to be determined, and neither MthK nor Kv channels contain a homologous interface.Because most domain/domain interactions involved in conformational coupling remain to be identified, it is useful to consider what properties such interactions must have. In general, coupling may occur through linkers that directly connect two domains or through noncovalent interactions at domain/domain interfaces. In BK channels, Ca²⁺-sensor/gate coupling is thought to be mediated by the S6-RCK1 linker connecting CTD to PGD (Jiang et al., 2002; Niu et al., 2004). However multiple domain/domain interfaces are also involved in coupling, as noted above. Structural information has helped to identify where these interfaces are and to locate potential interaction partners. But elucidating the basis of conformational coupling is more complex than simply identifying domain/domain interactions. Coupling is necessarily a state-dependent process that depends on the conformations of two coupled domains. To understand the possible interaction mechanisms, it is useful to consider the energetics of coupling in the context of a gating cycle such as that describing voltage-sensor/gate coupling in Fig. 3 A. The coupling energy between a single voltage sensor and gate (ΔΔGCORA=kTln(D)) can be expressed in terms of the free energies of the four states involved (AC, AO, RC, and RO):ΔΔGCORA=[ΔGCOA−ΔGCOR]=[(GCA−GOA)−(GCR−GOR)]=GCA−GOA−GCR+GOR(1)Therefore, the change in coupling produced by changes in the free energy of each state is:ΔΔΔGCORA=[ΔGCA−ΔGOA−ΔGCR+ΔGOR](2)Eq. 2 indicates that coupling is enhanced (ΔΔΔGCORA>0) by decreasing GCR or GOA (stabilizing the RC or AO state) or by increasing GCA or GOR (destabilizing the AC or RO state). Coupling could also be enhanced by changing the free energy of multiple states, provided the net contribution (Eq. 2) is positive. However, interactions that are state independent or depend on the conformation of only one domain (e.g., ΔGCA=ΔGCR or ΔG0R=ΔGCR) will have no effect on coupling (i.e.,ΔΔGCORA=0).This analysis has several important implications for elucidating mechanisms of conformational coupling. First, to identify coupling interactions by structural methods requires that structures be solved in four conformations defined by two domains and have sufficient resolution to observe state-dependent changes in interaction. In practice, because changes in interaction energy may be subtle or difficult to estimate based on structure, coupling interactions would have to be verified and quantified by structure-function analysis. When only one or two structures of a channel are available, as is currently the case for BK and homologous channels, then structure-function analysis is required to determine if observed interactions participate in coupling, and residues that do not interact in an observed structure may still mediate coupling by interacting in other conformations. Second, structure-function analysis involving site-directed mutagenesis and measurements of coupling energy, as in Figs. 2 and ​and3,3, should be sufficient to identify residues involved in coupling interactions. However, such experiments must be interpreted cautiously. Allosteric proteins are often sensitive to modulation at domain interfaces by heterotropic ligands that bind and introduce interactions that do not normally exist, and mutations can have similar effects. For example, introduction of a positively charged residue near the BK channel Mg²⁺-binding site (Q397R) mimics the effect of Mg²⁺ by introducing an electrostatic interaction (Yang et al., 2007). Thus, the ability of Q397R to enhance coupling does not indicate that Q397 is normally involved in this process, just that it is located at a sensitive interface where conformational changes associated with voltage-sensor activation and channel opening both occur. Such false positives cannot be avoided simply by restricting analysis to mutations that reduce coupling because, based on Eq. 2, coupling can be reduced by introducing interactions that destabilize the RC or AO states or stabilize the AC or RO states. Intracellular heme, which inhibits voltage-sensor/gate coupling in BK channels, may act in this way (Horrigan et al., 2005). Consequently, whether a mutation increases or decreases coupling is less critical than the use of mutations such as Ala that are least likely to introduce interactions. Finally, to define the mechanism of voltage-sensor/gate coupling we must not only identify mutations that alter coupling but also characterize those changes in terms of the effect on individual states (i.e., RC, RO, AC, and AO) to determine when during activation the interactions occur. Although the change in free energy of any particular state cannot be determined directly, the equilibria between states can be evaluated as in Fig. 3 to define the change in free energy of individual states relative to each other. Such analysis has indicated that Mg²⁺ enhances voltage-sensor/gate coupling by preferentially stabilizing the AO state (Horrigan and Ma, 2008), whereas a stabilization of the RO state is consistent with the inhibitory effect of heme (Horrigan et al., 2005). It is worth noting that, depending on their state dependence, coupling interactions may influence open probability when sensors are in a resting conformation. Thus, the basal activity of BK channels in the absence of sensor activation defined by the equilibrium constant L in the HA model cannot be interpreted simply as the intrinsic stability of the gate because it also reflects any state-dependent interactions between the gate and resting sensors.

Mechanisms of voltage-sensor/gate coupling

Kv channel structures reveal two interfaces between VSD and PGD: intrasubunit contacts between the S4–S5 linker and S6 and intersubunit contacts between S4 and S5 (Long et al., 2005a, 2007). In BK channels, additional intersubunit interactions occur between the VSD and CTD (Yang et al., 2008). All of these interfaces potentially play a role in voltage-sensor/gate coupling in BK channels, but their relative contributions and mechanisms remain unknown. The S4/S5 interface is implicated because a mutation near the top of S4 (R210E) reduces coupling energy by half (Ma et al., 2006). However, we cannot rule out that this is a false positive produced by introducing rather than disrupting an interaction because other mutations at this site (R210C and R210N) appear to constitutively activate the channel preventing measurement of voltage-sensor/gate coupling (Ma et al., 2006). Similarly, the VSD/CTD interface is implicated indirectly based on the ability of Mg²⁺, Q397R, and heme to modulate coupling by introducing interactions at or near this interface.Interactions between the S4–S5 linker and S6 are widely considered to underlie voltage-sensor/gate coupling, also known as electromechanical coupling, in Kv channels (Lu et al., 2002; Tristani-Firouzi et al., 2002; Long et al., 2005b; Chowdhury and Chanda, 2012) and are also likely to be important in BK channels (Sun et al., 2012). However, even in Kv channels many of the questions raised above remain to be answered, such as what are the individual residues and nature of interactions that contribute to coupling and when do they occur during gating? The importance of these questions can be illustrated by comparing three hypothetical mechanisms of voltage-sensor/gate coupling that include S4–S5/S6 interactions and are consistent with the allosteric nature of voltage gating in Slo1 (Fig. 4). Fig. 4 (A–C) depicts the four combinations of states the gate and voltage-sensor can assume in a single subunit. The possibility that S4–S5/S6 forms a rigid connection that forces sensor and gate to move as a unit can be ruled out because voltage sensors can activate while channels remain closed. However, it is conceivable that the S4–S5 linker remains bound to S6 at all times, whereas flexibility in adjoining regions allows sensor and gate to move as if coupled by a spring (Fig. 4 A). Another possibility is the S4–S5 linker binds to the open gate only when the voltage sensor is activated, stabilizing the AO state (Fig. 4 B). Alternatively, the S4–S5 linker of the resting voltage sensor might clash with the open gate to destabilize the RO state (Fig. 4 C). These and other mechanisms could account for the ability of voltage-sensor activation to promote opening but make different predictions concerning the source of coupling and the role of S4–S5/S6 interaction. Fig. 4 A suggests that many parts of the S4–S5 linker could contribute to coupling by influencing the mechanical properties of the linkage. In this case, S4–S5/S6 interaction acts merely as a passive connection that affects coupling only to the extent that it is intact or disrupted. In contrast, Fig. 4 (B and C) predicts that the S4–S5/S6 interaction energy is the main determinant of coupling energy. However, the panels in Fig. 4 differ in the nature of the proposed interaction (Fig. 4 B, binding; Fig. 4 C, steric hindrance) and the state dependence of the interaction. The analysis of coupling energy and equilibria for such gating cycles, as outlined above, should allow such mechanisms and predictions to be distinguished in BK channels.Open in a separate windowFigure 4.Three models of voltage-sensor/gate coupling. Interaction of S4–S5 linker with the S6 gate in a single subunit. The voltage sensor can be in a resting (R) or activated (A) state, and the gate is closed (C) or open (O). (A) Flexible linkage with S4–S5/S6 interacting in all states. (B) S4–S5/S6 binding stabilizes AO state. (C) Steric hindrance destabilizes the RO state.Despite substantial homology between BK and K_V channels, important differences also exist that could be relevant to voltage-sensor/gate coupling. First, BK channels are weakly voltage dependent compared with Kv channels and exhibit a different pattern and contribution of voltage-sensing residues in the VSD, suggesting that conformational changes associated with voltage-sensor activation may differ (Ma et al., 2006). Second, the ability of intracellular blockers and Cys-modifying reagents to access the inner pore of closed BK channels (Wilkens and Aldrich, 2006; Zhou et al., 2011) suggests that the BK channel gate may be formed by the selectivity filter (Cox and Hoshi, 2011; Thompson and Begenisich, 2012), as in cyclic-nucleotide–gated channels (Contreras et al., 2008), rather than by crossing of S6 segments at the inner mouth of the pore, as in Kv channels (del Camino and Yellen, 2001). These differences do not require that voltage-sensor/gate coupling occurs through fundamentally different mechanisms. For example, if the BK channel gate is in the selectivity filter it is still likely to be strongly coupled to S6 movement, as in CNG channels (Flynn and Zagotta, 2001), and therefore subject to control by S4–S5/S6 interaction. Indeed, there is considerable evidence for S6 movement associated with BK channel opening (Li and Aldrich, 2006; Wu et al., 2009; Chen and Aldrich, 2011). However, the coupling mechanisms for BK and Kv channels may differ in detail, and a selectivity gate in the BK channel could potentially support a larger role for S4/S5 interaction in voltage-sensor/gate coupling. It should also be noted that regulatory β and γ subunits have been reported to modulate voltage-sensor/gate coupling (Bao and Cox, 2005; Yan and Aldrich, 2010); thus, additional state-dependent interactions may form in BK channels between the VSD and regulatory subunits.

Mechanisms of coupling Ca²⁺ sensors to the gate and voltage sensors

Of the three coupling interactions in the HA model, Ca²⁺-sensor/gate coupling is easiest to measure and best understood at a molecular level. Nonetheless, many questions and controversies remain to be resolved concerning the conformational changes that occur upon Ca²⁺ binding and the mechanisms coupling these changes to the gate and voltage sensors.Ca²⁺-dependent activation is generally consistent with the idea, originally proposed for MthK, that Ca²⁺ binding causes a conformational change in the gating ring that opens the channel by pulling on the RCK1-S6 linker (Jiang et al., 2002). Crystal structures in the presence and absence of Ca²⁺ suggest that MthK and BK channel gating rings expand in diameter by 8 and 12 Å, respectively, upon Ca²⁺ binding (Ye et al., 2006; Wu et al., 2010; Yuan et al., 2012). Experimental evidence in BK channels shows a monotonic relationship between channel activation and linker length, suggesting that the linker is under constant tension in the presence or absence of Ca²⁺, like a spring (Niu et al., 2004). Likewise, effects on Ca²⁺-sensor/gate coupling and Ca² sensitivity of mutations in the N-terminal AC region of the BK channel RCK1 domain suggest that the flexibility of this region is important for transmitting Ca²⁺-dependent conformational changes to the gate (Yang et al., 2010). Although the latter results are consistent with the linker hypothesis, they also have raised the possibility that conformational changes in the AC region could be coupled to the gate through direct contact with the PGD. Consistent with this possibility, the N-terminal half of RCK1 undergoes a substantial reorientation relative to the rest of the BK channel gating ring upon Ca²⁺ binding (Yuan et al., 2012). However, determining whether a CTD/PDG interface exists and contributes to coupling may have to wait until a complete BK channel structure is available.The BK channel CTD contains two high affinity Ca²⁺-binding sites (one in each RCK domain) that have been identified by site-directed mutagenesis (Bao et al., 2004; Zhang et al., 2010). By using mutations to eliminate each site individually and carefully measuring the effects of Ca²⁺ on P_o at −80 mV (similar to Fig. 2 B), Sweet and Cox (2008) determined that the contribution to Ca²⁺-sensor/gate coupling of the RCK1 site (3.74 kCal mol⁻¹) was slightly greater than that of the RCK2 Ca²⁺ bowl site (3.04 kCal mol⁻¹), but the RCK2 site has a higher affinity for Ca²⁺. In addition, the summed contribution of the individual sites exceeded their combined effect in the WT channel (ΔΔGCOCa=5.0 kcal mol−1; Fig. 2 B), consistent with negative cooperativity between sites in the same subunit (Sweet and Cox, 2008). In contrast, another group performing similar experiments at +50 mV concluded that positive cooperativity exists between the two binding sites (Qian et al., 2006). In the latter case, positive cooperativity could potentially be accounted for if both Ca²⁺ sites were coupled to the voltage sensor because the voltage sensor is not held in a resting state at +50 mV. However Sweet and Cox (2008) concluded that the RCK1 site is solely responsible for Ca²⁺-sensor/voltage-sensor coupling. A recent study combining Ca²⁺ site mutations and voltage clamp fluorometry to monitor the effects of Ca²⁺ on steady-state voltage-sensor activation and P_o reached conclusions similar to that of Sweet and Cox (2008) regarding intrasubunit cooperativity and Ca²⁺-sensor/voltage-sensor coupling (Savalli et al., 2012). However, this study also suggested that Ca²⁺ site mutations may have effects other than elimination of Ca²⁺ binding. This caveat, together with the fact that none of the studies mentioned measured Ca²⁺-sensor/voltage-sensor coupling directly, suggests that the extent to which the individual Ca²⁺-binding sites are coupled to the voltage sensor or to each other, as well as the molecular mechanisms mediating such interactions, is still open to question.Although the CTD contains two high affinity Ca²⁺-binding sites, only the Ca²⁺ bowl is occupied in the Ca²⁺-bound gating-ring structure (Yuan et al., 2012), raising questions of to what extent this structure resembles the Ca²⁺-saturated conformation in the intact channel. In principal, the two Ca²⁺ sites could have independent effects on activation by stabilizing a single Ca²⁺-bound open conformation. However, mutations in the AC region of RCK1 selectively alter the Ca²⁺ sensitivity of the RCK1 site (Yang et al., 2010), suggesting that some conformational changes may be coupled to RCK1 occupancy alone. This, together with the fact that the gating ring of the intact channel is expected to contact the VSD, suggests that significant differences in structure could exist between the crystal structure and intact Ca²⁺-saturated gating ring. A related question is to what extent gating ring expansion is determined by channel opening (i.e., RCK1-S6 linker tension) versus Ca²⁺ binding. Because gating ring expansion is observed upon Ca²⁺ binding in isolated gating rings, it seems reasonable to suppose that expansion and channel opening might not occur simultaneously. If the gating ring could expand into a high-Ca²⁺-affinity conformation while the channel is closed, it could have important implications for gating models and for the possible state dependence of CTD/VSD coupling. However, there is as yet no evidence supporting this possibility. Indeed Sweet and Cox (2008) concluded that their results were best fit by assuming gating-ring expansion is tightly coupled to channel opening, consistent with the HA model.A final question relates to conformational events that occur in Ca²⁺-binding sites during channel opening. As in any allosteric model of ligand-dependent gating, the HA model predicts that Ca²⁺-binding sites must have a higher affinity for Ca²⁺ in the open than the closed conformation. Therefore, understanding how Ca²⁺ coordination changes upon channel opening is fundamental to the mechanism of Ca²⁺-sensor/gate coupling. In principal, the contribution of individual Ca²⁺-coordinating residues to state-dependent binding can be evaluated by determining the effect on coupling energy of mutating such sites (Purohit et al., 2012). Previous studies have generally identified likely Ca²⁺-coordination sites based on a reduced sensitivity of mutated channels to Ca²⁺ in the physiological range (≤100 µM). However, experiments at higher [Ca²⁺] may be required to saturate mutated sites and determine if Ca²⁺-sensor/gate coupling is altered.

Summary

The BK channel is an important example of a voltage- and ligand-gated channel whose function depends on conformational coupling between multiple domains. Many questions remain about the molecular basis of domain–domain coupling, but this channel represents a favorable system for studying such processes owing to its unique functional properties and methods that have allowed the energetics of coupling to be studied in detail. Combining these methods with emerging structural information about domain/domain interfaces and conformational changes in the channel should provide further insight into coupling mechanisms that are important in BK channels and in other voltage-gated or ligand-gated channels. BK channels also constitute a powerful system for understanding the interplay between ligand- and voltage-dependent gating. Defining the interactions that mediate coupling between voltage and Ca²⁺ sensors in this channel should provide unique insight into processes that may be relevant to other multimodal channels such as HCN or TRP.This Perspectives series includes articles by Andersen, Colquhoun and Lape, and Chowdury and Chanda.     

ATG8ylation of proteins: A way to cope with cell stress?

The ATG8 family of proteins regulates autophagy in a variety of ways. Recently, ATG8s were demonstrated to conjugate directly to cellular proteins in a process termed “ATG8ylation,” which is amplified by mitochondrial damage and antagonized by ATG4 proteases. ATG8s may have an emerging role as small protein modifiers.

ATG8 proteins directly conjugate to cellular proteinsAutophagy describes the capture of intracellular material by autophagosomes and their delivery to lysosomes for destruction (Kaur and Debnath, 2015). This process homeostatically remodels the intracellular environment and is necessary for an organism to overcome starvation (Kaur and Debnath, 2015). The autophagy pathway is coordinated by autophagy-related (ATG) proteins that are controlled by diverse post-translational modifications (e.g., phosphorylation, acetylation, ubiquitination, and lipidation; Ichimura et al., 2000; McEwan and Dikic, 2011). Recently, a previously uncharacterized post-translational modification termed “ATG8ylation” was uncovered (Agrotis et al., 2019; Nguyen et al., 2021). ATG8ylation is the direct covalent attachment of the small ubiquitin-like family of ATG8 proteins to cellular proteins (Agrotis et al., 2019; Nguyen et al., 2021). Until now, the only known instances of ATG8 conjugation to proteins were of a transient nature, as E1- and E2-like intermediates with ATG7 and ATG3, respectively, as a way of ligating ATG8 to the lipid phosphatidylethanolamine during autophagy (Ichimura et al., 2000). Therefore, ATG8ylation may represent an underappreciated regulatory mechanism for many cellular proteins that coordinate pathways such as mitophagy.ATG8s play many roles in the autophagy pathwayDuring canonical autophagy, the ATG8 family (comprising LC3A, -B, and -C and GABARAP, -L1, and -L2) undergoes molecular processing that concludes with their attachment to phosphatidylethanolamine, enabling proper construction of autophagosomes and subsequent autophagosome–lysosome fusion (Nguyen et al., 2016). The ATG4 family of cysteine proteases (ATG4A, -B, -C, and -D) cleaves ATG8 proteins immediately after a conserved glycine residue in their C terminus in a process dubbed “priming,” which leads to the formation of ATG8-I (Skytte Rasmussen et al., 2017; Tanida et al., 2004). ATG7 then attaches to the exposed glycine residue of ATG8-I via a thioester linkage to form an E1 ubiquitin-like complex that transfers ATG8-I to ATG3 in a similar way to generate an E2-like complex (Ichimura et al., 2000). The ATG5–ATG12–ATG16L1 complex then catalyzes the E3-like transfer of ATG8-I from ATG3 to phosphatidylethanolamine to form ATG8-II, which is the lipidated species that is incorporated into double membrane–bound compartments such as autophagosomes (Hanada et al., 2007). The lipidation of ATG8s and their recruitment to the phagophore are not essential for the formation of autophagosomes but are important for phagophore expansion, the selective capture of autophagic substrates, and autophagosome–lysosome fusion (Kirkin and Rogov, 2019; Nguyen et al., 2016). Intriguingly, ATG8 lipidation is multifaceted, as ATG8s can be alternatively lipidated with phosphatidylserine (instead of phosphatidylethanolamine) to enable their recruitment to single membrane–bound compartments during LC3-associated phagocytosis, influenza infection, and lysosomal dysfunction (Durgan et al., 2021).The discovery of ATG8ylationKey insights into ATG8ylation came from the observation that various ATG8s form high-molecular-weight species in cells following the expression of their primed forms that have their C-terminal glycine exposed (for example, LC3B-G), bypassing the need for cleavage by ATG4 (Agrotis et al., 2019; Nguyen et al., 2021). Indeed, on an immunoblot, ATG8⁺ “smears” resemble that of ubiquitinated proteins (Agrotis et al., 2019; Nguyen et al., 2021). Traditionally, in the autophagy field, ATG8⁺ smears were thought to arise from poor antibody specificity. However, in light of recent findings, this widely accepted interpretation has been challenged, given that ATG8⁺ smears are enriched following ATG8 overexpression and disappear in the absence of ATG8s (Agrotis et al., 2019; Nguyen et al., 2021). Smearing has also been detected after immunoprecipitation of epitope-tagged ATG8s from cell extracts under denaturing conditions, ruling out noncovalent interactions accounting for this upshift (Agrotis et al., 2019; Nguyen et al., 2021). Further, smearing is not abolished by deubiquitinase treatment, arguing strongly against ATG8 ubiquitination as the cause (Nguyen et al., 2021). Everything considered, the most plausible explanation is that ATG8 itself undergoes covalent linkage to cellular proteins, akin to ubiquitin and NEDD8 modifiers, which are structurally similar to ATG8s. Remarkably, the protease ATG4 antagonizes the ATG8ylation state of many proteins (Agrotis et al., 2019; Nguyen et al., 2021).ATG4 displays isoform-specific proteolytic cleavage of ATG8ATG4 is required for the formation of autophagosomes, but its protease activity is not (Nguyen et al., 2021). The protease activity of ATG4 is, however, required for ATG8 processing, such as priming ahead of lipidation and de-lipidation, which removes excess ATG8 from autophagosomes and other membranes (Nguyen et al., 2021; Tanida et al., 2004; Fig. 1 A). Apart from these functions, ATG4 regulates the deubiquitinase-like removal of ATG8 from cellular proteins (de-ATG8ylation; Agrotis et al., 2019; Nguyen et al., 2021; Fig. 1 A). Consistent with this role, deletion of all four ATG4 isoforms (A, B, C, and D) increases the abundance of ATG8ylated proteins (Nguyen et al., 2021). In contrast, overexpression of ATG4B has the opposite effect, but only if its protease activity is intact (Agrotis et al., 2019). As such, ATG4 inhibits the ATG8ylation state of many proteins, which is likely to modulate their downstream functions.Open in a separate windowFigure 1.The many roles of ATG4 in ATG8 processing. (A) Molecular processing of ATG8 proteins by ATG4 illustrating its roles in priming, de-lipidation, and de-ATG8ylation. The structure of LC3B (Protein Data Bank accession no. 1V49) was used to denote ATG8 (G, glycine; PE, phosphatidylethanolamine). (B) Heatmap summarizing relationships between ATG4 isoforms and ATG8 family members. Data were summarized for qualitative interpretation (Agrotis et al., 2019; Li et al., 2011; Nguyen et al., 2021). Int., intermediate; N.d., not determined. (C) Graphical summary of questions moving forward with ATG8ylation (P, phosphorylation).ATG4 is an important “gatekeeper” for ATG8 conjugation events. ATG4 primes ATG8s to expose their C-terminal glycine, which is required for conjugation to proteins or lipids; however, ATG4 also catalyzes de-ATG8ylation and de-lipidation events, respectively (Agrotis et al., 2019; Nguyen et al., 2021; Tanida et al., 2004). Because the C-terminal glycine of a single ATG8 is occupied when conjugated to a protein or lipid, it is unlikely that ATG8ylated proteins directly engage with phagophore membranes in the same way as ATG8-II. Indeed, protease protection assays with recombinant ATG4B reveal that de-ATG8ylation of cell lysates remains unchanged with or without organellar membrane disruption, suggesting that ATG8ylated proteins are largely cytoplasmic facing rather than intraluminal (Agrotis et al., 2019). Paradoxically, however, ATG8ylation is enhanced by lysosomal V-type ATPase inhibition, which blocks the degradation of lysosomal contents, indicating that ATG8ylated substrates may undergo lysosome-dependent turnover (Agrotis et al., 2019; Nguyen et al., 2021). One explanation for these differences may be that the process of ATG8ylation is itself sensitive to lysosomal dysfunction.Functional relationships between ATG4s and ATG8sIsoforms of ATG4 show clear preferences for proteolytically processing ATG8 subfamilies (i.e., LC3s and GABARAPs) for de-ATG8ylation and priming upstream of phosphatidylethanolamine ligation (Agrotis et al., 2019; Li et al., 2011; Nguyen et al., 2021; Fig. 1 B). ATG4A strongly reduces the abundance of proteins that have been ATG8ylated with the GABARAP family while promoting ligation of GABARAPs to phosphatidylethanolamine (Agrotis et al., 2019; Nguyen et al., 2021; Fig. 1 B). In contrast, ATG4B strongly reduces the abundance of proteins that have been ATG8ylated with LC3 proteins while promoting ligation of LC3s to phosphatidylethanolamine (Agrotis et al., 2019; Nguyen et al., 2021; Fig. 1 B). In comparison, ATG4C and -D lack obvious de-ATG8ylation activity, although the latter weakly promotes phosphatidylethanolamine ligation to GABARAPL1 only (Nguyen et al., 2021). These functional similarities between ATG4 isoforms are consistent with both their sequence and structural homology (i.e., ATG4A and -B are most similar; Maruyama and Noda, 2018; Satoo et al., 2009). Structurally, ATG4B adopts an auto-inhibited conformation with its regulatory loop and N-terminal tail blocking substrate entry to its proteolytic core (Maruyama and Noda, 2018). LC3B induces conformational rearrangements in ATG4B that involve displacement of its regulatory loop and its N-terminal tail, with the latter achieved by an interaction between the ATG8-interacting region in its N-terminal tail with a second copy of LC3B that functions allosterically (Maruyama and Noda, 2018; Satoo et al., 2009). These rearrangements permit entry of LC3B into the proteolytic core of ATG4B, where cleavage of LC3B following its C-terminal glycine occurs (Li et al., 2011; Maruyama and Noda, 2018). ATG4B^L232 is directly involved in LC3B binding and its selectivity for LC3s (Satoo et al., 2009). This residue corresponds to ATG4A^I233 and, when substituted for leucine, gives ATG4A^I233L the ability to efficiently process LC3 proteins, whereas without this mutation it preferentially processes GABARAPs (Satoo et al., 2009). Moreover, the ATG8–ATG4 interaction is necessary for the de-ATG8ylation of cellular proteins, as an LC3B-G^Q116P mutant that cannot bind to ATG4 leads to widespread ATG8ylation (Agrotis et al., 2019). Altogether, these observations hint toward a common mechanism of ATG8 cleavage that regulates priming, de-lipidation, and de-ATG8ylation.Mitochondrial damage promotes ATG8ylationATG8ylation of cellular proteins appears to be enhanced by mitochondrial depolarization and inhibition of the lysosomal V-type ATPase (Agrotis et al., 2019; Nguyen et al., 2021). This may be the consequence of acute ATG4A and -B inhibition, given that cells lacking all ATG4 isoforms display an increased abundance of ATG8ylated proteins and are insensitive to further increase by mitochondrial depolarization or lysosomal V-type ATPase inhibition (Agrotis et al., 2019; Nguyen et al., 2021). Indeed, mitochondrial depolarization leads to activation of ULK1, which phosphorylates ATG4B^S316 to inhibit its protease activity (Pengo et al., 2017). Similarly, mitochondrial depolarization stimulates TBK1 activation, which prevents de-lipidation of ATG8s by blocking the ATG8–ATG4 interaction through phosphorylation of LC3C^S93/S96 and GABARAP-L2^S87/S88 (Herhaus et al., 2020; Richter et al., 2016). As such, ATG8 phosphorylation may render ATG8ylated substrates more resistant to de-ATG8ylation by ATG4s. This may be analogous to how chains of phosphorylated ubiquitin^S65 are more resistant to hydrolysis by deubiquitinating enzymes than unphosphorylated ones (Wauer et al., 2015). Moreover, ATG8ylation is insensitive to nutrient deprivation and pharmacological inhibition of mTOR, which rules out a functional contribution of this process to starvation-induced autophagy (Agrotis et al., 2019). Therefore, ATG8ylation may be a unique aspect of mitophagy (and perhaps also other forms of selective autophagy) given that depolarization potently activates Parkin-dependent mitophagy (Agrotis et al., 2019; Nguyen et al., 2021).Substrates of ATG8ylationBased on ATG8⁺ smearing, ATG4 regulates the de-ATG8ylation of numerous proteins (Agrotis et al., 2019; Nguyen et al., 2021). For the majority, their identity, induced structural and functional changes, and the cellular contexts during which these modifications occur await exploration. Considering that the ATG8 interactome is well characterized, it is likely that at least some ATG8ylated proteins have been mistaken for ATG8-binding partners (Behrends et al., 2010). Given their E2- and E3-like roles in ATG8 lipidation, it is remarkable that ATG3 and ATG16L1 are themselves modified by ATG8ylation (Agrotis et al., 2019; Hanada et al., 2007; Ichimura et al., 2000; Nguyen et al., 2021). Lysine mutagenesis indicates that ATG3^K243 is the “acceptor” site for ATG8ylation (Agrotis et al., 2019). ATG3^K243 is essential for its conjugation to either LC3B or ATG12 and is required for autophagosomes to form around damaged mitochondria (Agrotis et al., 2019; Radoshevich et al., 2010). This also raises the possibility that key functions originally attributed to ATG3–ATG12 conjugation may be, at least in part, due to ATG3–ATG8 conjugation. Because multiple high-molecular-weight species of ATG3 are enriched following immunoprecipitation of primed LC3B-G from cells lacking ATG4B, it is likely that ATG3 is either mono-ATG8ylated at several sites or poly-ATG8ylated (Agrotis et al., 2019). ATG8ylation of ATG3 may also reflect the stabilization of its E2-like intermediate (Ichimura et al., 2000). ATG8ylation of ATG16L1 may regulate whether canonical or noncanonical autophagy pathways are activated (Durgan et al., 2021; Nguyen et al., 2021). In line with this possibility, the WD40 domain mutant of ATG16L1^K490A prevents lipidation of ATG8s with phosphatidylserine (i.e., during noncanonical autophagy pathways) but not phosphatidylethanolamine (i.e., during canonical autophagy; Durgan et al., 2021). Moreover, given that ATG8ylation of protein targets correlates with the activation of mitophagy, it is tempting to speculate that it may stimulate the E2-/E3-like activity of the ATG8 conjugation machinery to amplify mitochondrial capture and destruction.Concluding remarksThe finding that numerous cellular proteins are modified by ATG8ylation poses several questions about how signaling networks are coordinated during selective autophagy (i.e., mitophagy). Whether ATG8ylation is augmented by mitochondrial injury per se or is the consequence of mitophagy activation is yet to be determined, as is whether this phenomenon occurs during other types of selective autophagy (e.g., ER-phagy, ribophagy, and lysophagy; Kirkin and Rogov, 2019; Fig. 1 C). While the in vivo relevance of ATG8ylation is not yet understood, it is plausible that this process could be altered in diseases with defective mitophagy (e.g., Parkinson’s disease and atherosclerosis). Exploring the mechanistic aspects of ATG8ylation (e.g., ATG8 ligases and regulatory proteins, linkage types, acceptor sites, etc.) and de-ATG8ylation by ATG4 will improve our understanding about how this modifier alters the structure and biological function of cellular proteins (Fig. 1 C). By identifying ATG8ylated substrates, or the ATG8ylome, insights into whether ATG8ylation is a ubiquitous epiphenomenon or a post-translational modification that is selective to proteins of distinct biological function(s) will become clearer (Fig. 1 C). Considering the similarity of ATG8s with bona fide modifier proteins (e.g., ubiquitin and ubiquitin-like proteins) and the diversity of their substrates (e.g., lipid species and proteins), only now are we beginning to understand the functional complexities of the ATG8 protein family.     

Maturation Stress Generation in Poplar Tension Wood Studied by Synchrotron Radiation Microdiffraction

Tension wood is widespread in the organs of woody plants. During its formation, it generates a large tensile mechanical stress, called maturation stress. Maturation stress performs essential biomechanical functions such as optimizing the mechanical resistance of the stem, performing adaptive movements, and ensuring long-term stability of growing plants. Although various hypotheses have recently been proposed, the mechanism generating maturation stress is not yet fully understood. In order to discriminate between these hypotheses, we investigated structural changes in cellulose microfibrils along sequences of xylem cell differentiation in tension and normal wood of poplar (Populus deltoides × Populus trichocarpa ‘I45-51’). Synchrotron radiation microdiffraction was used to measure the evolution of the angle and lattice spacing of crystalline cellulose associated with the deposition of successive cell wall layers. Profiles of normal and tension wood were very similar in early development stages corresponding to the formation of the S1 and the outer part of the S2 layer. The microfibril angle in the S2 layer was found to be lower in its inner part than in its outer part, especially in tension wood. In tension wood only, this decrease occurred together with an increase in cellulose lattice spacing, and this happened before the G-layer was visible. The relative increase in lattice spacing was found close to the usual value of maturation strains, strongly suggesting that microfibrils of this layer are put into tension and contribute to the generation of maturation stress.Wood cells are produced in the cambium at the periphery of the stem. The formation of the secondary wall occurs at the end of cell elongation by the deposition of successive layers made of cellulose microfibrils bounded by an amorphous polymeric matrix. Each layer has a specific chemical composition and is characterized by a particular orientation of the microfibrils relative to the cell axis (Mellerowicz and Sundberg, 2008). Microfibrils are made of crystalline cellulose and are by far the stiffest constituent of the cell wall. The microfibril angle (MFA) in each layer is determinant for cell wall architecture and wood mechanical properties.During the formation of wood cells, a mechanical stress of a large magnitude, known as “maturation stress” or “growth stress” (Archer, 1986; Fournier et al., 1991), occurs in the cell walls. This stress fulfills essential biomechanical functions for the tree. It compensates for the comparatively low compressive strength of wood and thus improves the stem resistance against bending loads. It also provides the tree with a motor system (Moulia et al., 2006), necessary to maintain the stem at a constant angle during growth (Alméras and Fournier, 2009) or to achieve adaptive reorientations. In angiosperms, a large tensile maturation stress is generated by a specialized tissue called “tension wood.” In poplar (Populus deltoides × Populus trichocarpa), as in most temperate tree species, tension wood fibers are characterized by the presence of a specific layer, called the G-layer (Jourez et al., 2001; Fang et al., 2008), where the matrix is almost devoid of lignin (Pilate et al., 2004) and the microfibrils are oriented parallel to the fiber axis (Fujita et al., 1974). This type of reaction cell is common in plant organs whose function involves the bending or contraction of axes, such as tendrils, twining vines (Bowling and Vaughn, 2009), or roots (Fisher, 2008).The mechanism at the origin of tensile maturation stress has been the subject of a lot of controversy and is still not fully understood. However, several recent publications have greatly improved our knowledge about the ultrastructure, chemical composition, molecular activity, mechanical state, and behavior of tension wood. Different models have been proposed and discussed to explain the origin of maturation stress (Boyd, 1972; Bamber, 1987, 2001; Okuyama et al., 1994, 1995; Yamamoto, 1998, 2004; Alméras et al., 2005, 2006; Bowling and Vaughn, 2008; Goswami et al., 2008; Mellerowicz et al., 2008). The specific organization of the G-layer suggests a tensile force induced in the microfibrils during the maturation process. Different hypotheses have been proposed to explain this mechanism, such as the contraction of amorphous zones within the cellulose microfibrils (Yamamoto, 2004), the action of xyloglucans during the formation of microfibril aggregates (Nishikubo et al., 2007; Mellerowicz et al., 2008), and the effect of changes in moisture content stimulated by pectin-like substances (Bowling and Vaughn, 2008). A recent work (Goswami et al., 2008) argued an alternative model, initially proposed by Münch (1938), which proposed that the maturation stress originates in the swelling of the G-layer during cell maturation and is transmitted to the adjacent secondary layers, where the larger MFAs allow an efficient conversion of lateral stress into axial tensile stress. Although the proposed mechanism is not consistent with the known hygroscopic behavior of tension wood, which shrinks when it dries and not when it takes up water (Clair and Thibaut, 2001; Fang et al., 2007; Clair et al., 2008), this hypothesis focused attention on the possible role of cell wall layers other than the G-layer. As a matter of fact, many types of wood fibers lacking a G-layer are known to produce axial tensile stress, such as normal wood of angiosperms and conifers (Archer, 1986) and the tension wood of many tropical species (Onaka, 1949; Clair et al., 2006b; Ruelle et al., 2007), so that mechanisms strictly based on an action of the G-layer cannot provide a general explanation for the origin of tensile maturation stress in wood.In order to further understanding, direct observations of the mechanical state of the different cell wall layers and their evolution during the formation of the tension wood fibers are needed. X-ray diffraction can be used to investigate the orientation of microfibrils (Cave, 1966, 1997a, 1997b; Peura et al., 2007, 2008a, 2008b) and the lattice spacing of crystalline cellulose. The axial lattice spacing d₀₀₄ is the distance between successive monomers along a cellulose microfibril and reflects its state of mechanical stress (Clair et al., 2006a; Peura et al., 2007). If cellulose microfibrils indeed support a tensile stress, they should be found in an extended state of deformation. Under this assumption, the progressive development of maturation stress during the cell wall formation should be accompanied by an increase in cellulose lattice spacing. Synchrotron radiation allows a reduction in the size of the x-ray beam to some micrometers while retaining a strong signal, whereby diffraction analysis can be performed at a very local scale (Riekel, 2000). This technique has been used to study sequences of wood cell development (Hori et al., 2000; Müller et al., 2002). In this study, we report an experiment where a microbeam was used to analyze the structural changes of cellulose in the cell wall layers of tension wood and normal wood fibers along the sequence of xylem cell differentiation extending from the cambium to mature wood (Fig. 1). The experiment was designed to make this measurement in planta, in order to minimize sources of mechanical disturbance and be as close as possible to the native mechanical state (Clair et al., 2006a). The 200 and 004 diffraction patterns of cellulose were analyzed to investigate the process of maturation stress generation in tension wood.Open in a separate windowFigure 1.Schematic of the experimental setup, showing the x-ray beam passing perpendicular to the longitudinal-radial plane of wood and the contribution of the 004 and 200 crystal planes to the diffraction pattern recorded by the camera. [See online article for color version of this figure.]     

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.