E XP ER I ME NTAL C E LL RE S E ARCH
] (]]]]) ]]]–]]]
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/yexcr
Review Article
ER sheet–tubule balance is regulated by an array of actin filaments and microtubules Merja Joensuua, Eija Jokitaloa,b,n a
Institute of Biotechnology, Cell and Molecular Biology Program, University of Helsinki, 00014 Helsinki, Finland Electron Microscopy Unit, University of Helsinki, 00014 Helsinki, Finland
b
article information Article Chronology: Received 2 April 2015 Accepted 13 April 2015 Keywords: Endoplasmic reticulum Actin cytoskeleton 3D-EM Morhology Dynamics Sheet–tubule balance Microtubules
Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ER consist of dynamic and static subdomains. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ER, actin cytoskeleton and microtubules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ER sheet transformations are coupled to a subset of dynamic actin filaments . . . . . . . . Proteins involved in ER–actin cytoskeleton interaction. . . . . . . . . . . . . . . . . . . . . . . . . . . Myo1c regulation of actin arrays is conveyed to sheet morphology. . . . . . . . . . . . . . . . . Dynamic microtubules and actin filament arrays counterbalance sheet–tubule balance . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
n
Corresponding author at: Institute of Biotechnology, Cell and Molecular Biology Program, University of Helsinki, 00014 Helsinki, Finland. E-mail address: eija.jokitalo@helsinki.fi (E. Jokitalo).
http://dx.doi.org/10.1016/j.yexcr.2015.04.009 0014-4827/& 2015 Elsevier Inc. All rights reserved.
Please cite this article as: M. Joensuu, E. Jokitalo, ER sheet–tubule balance is regulated by an array of actin filaments and microtubules, Exp Cell Res (2015), http://dx.doi.org/10.1016/j.yexcr.2015.04.009
1 2 2 3 5 5 6 7 7
2
E XP E RI ME N TAL CE L L R ES E ARC H
Introduction Endoplasmic reticulum (ER) is a single continuous membrane network containing multiple subdomains with different structures and functions. Gradually over decades of research, various aspects of ER morphology, organization and dynamics in different organisms and cell cultures at interphase and upon cell division have been revealed, together highlighting the significant plasticity of the organelle. It is important to understand that depending on cell type and organism, the ratio between individual subdomains (i.e. sheets and tubules) and network distribution varies significantly, and therefore drawing generalized conclusions out of observations made in one model system should be done carefully. In addition, the structure of ER subdomains varies between different cell types, e.g. the average size of the sheets, and the presence/absence of sheet fenestrations (i.e. small perforations passing through the sheets). Fenestrations are especially observed in cell types with abundant (Huh-7 and HeLa [1]) or large (plasma membrane associated ER in Saccharomyces cerevisiae [2]) sheets, and are absent or sparse in cell types where the network is predominantly tubular and sheets smaller (CHO-K1 and NRK-52E [1,3]). While some discrepancy related to the ER sheet fenestrations [1,4,5] and the mechanism for ER inheritance [1,3,6] exist, the differing views can be explained by the use of different model systems and cells. Fenestration can be observed irrespective of fixation method (i.e. chemical fixation vs. high pressure freezing, HPF) [1,5] or expression of an ER-marker [1,5], but as their diameter falls beyond the resolution limit of conventional light microscopy (LM), they can, therefore, be visualized only with high-resolution microscopy. ER network polygons (i.e. the space between sheets and tubules) and fenestrations present different structures, as fenestrations are considerably smaller openings on the sheets. In contrast to nuclear pores found from nuclear envelope (NE), which are uniform in size and formed by a special set of proteins, fenestrations do not appear to have a specific pore structure and they vary somewhat in shape and size, averaging from 7573 nm in diameter [1], but are similarly found on planar sheets. ER architecture is modified to meet specific cellular needs, implying that several means must exist to accomplish such adaptations. While ER structure is in constant flux, it does not move en masse, and the network movement is achieved through dynamics of individual subdomains and through network remodeling [7–12]. While integral membrane proteins are indispensable for the ER structure [13], the dynamics and the network remodeling are accomplished through interactions with the cytoskeleton. The interplay between tubules and microtubules (MT) has been well described [7–12,14–18], but the maintenance of sheets and the role of actin cytoskeleton on ER structure in mammalian cells have remained obscure. Here, we focus on the ER association with actin cytoskeleton determining ER sheet structure and dynamics in cultured mammalian cells. In our studies, we chose to use Huh-7 as our main model system because the highly abundant ER in these active secretory cells is composed mainly of large sheets spreading throughout the cytoplasm while tubules are found around the Golgi complex and in peripheral area of the cell [1,5].
ER consist of dynamic and static subdomains In most cell types, ER tubules and sheets create a polygonal network, which can be altered by changing branching patterns or
] (]]]]) ]]]–]]]
by remodeling of tubules into sheets and vice versa [5,7–9,12,19]. Studies done on ER dynamics have mostly concentrated on describing the tubular dynamics and transformations. The ground work on ER dynamics in animal cells was done already in 1988 by Lee and Chen [7], who described the remodeling the ER network by tubular fusion and junction sliding, which create new branch points and polygons, or result in polygon elimination from the network. In addition to these dynamic events, several additional events are common in both plant and animal cells: tubules retract, kink and deform, new sheets are created by filling the polygon with membrane, and sheets are created or altered by polygon opening within sheets [5,8,9,11,20,21]. Sheets transform into tubules by extensive fenestration [1] or by consumption of the sheet membrane [5]. Based on Huh-7 live-cell imaging, we recently showed that while sheets can transform into tubules, split into smaller sheets and fuse with one another to form larger sheets, a majority of sheets are relatively static structures [5]. In contrast, the majority of the tubules appear to be in a constant dynamic state [8–10,16]. By adapting a method that calculates the sheet's center-of-mass, we were able to show that sheet movements do not exhibit any clear directionality [5], differing from the previously described tubular movement [8–10,15], as the sheets mainly fluctuate in small area. In addition, the average sheet velocity (0.15 mm/s70.23 SD [5]) differs from that of tubules [8,10,15]. In HeLa and NRK-52E, the sheets are relatively static and the average sheet velocities were roughly the same compared to Huh-7 (Our unpublished results). These results demonstrate that sheets remain rather static in otherwise dynamic network, implying that the current view of the ER dynamics cannot be generalized to the whole network. Moreover, it is likely that the sheet repositioning in the cells is likely to occur through formation of a new sheet rather than transport of an existing one. Based on differing tubule and sheet dynamics, it is clear that multiple means must exist to accomplish the dynamics described above. While the purpose of the remodeling is unclear, it is thought that network structure, and perhaps, ultimately, the distribution of ER's functions into specific subdomains, are a consequence of the balance between dynamics, transitions and persistence of sheets and tubules [20].
ER, actin cytoskeleton and microtubules In vitro formation of reticular ER network has been shown to occur independent of MTs and actin [22] but requires the presence of curvature-inducing proteins [13,22,23]. It is, however, unclear if sheets were present in the formed ER network. Furthermore, a cell-free assembly of ER tubules and ribosomecovered sheets has been reported in in vitro experiments using rat hepatocyte microsomes [24]. The authors demonstrated that GTP is required for reconstitution of ribosome-covered sheets and that both GTP and ATP are required for tubule formation. However, to our view, the formed ER network appeared uncharacteristic and it is plausible that the formation of characteristic polygonal ER network requires interplay of several factors – including the different cytoskeletal filaments. Although MTs and the associated protein machinery are the main mechanism for tubular ER remodeling in interphase animal cells [8–10,14–16,25,26], MTs and ER do not have identical distribution in cells [10,21,27] and there is a body of work
Please cite this article as: M. Joensuu, E. Jokitalo, ER sheet–tubule balance is regulated by an array of actin filaments and microtubules, Exp Cell Res (2015), http://dx.doi.org/10.1016/j.yexcr.2015.04.009
E X PE R IM EN TA L C ELL R E S EA RC H
indicating that ER can also interact with actin cytoskeleton in animal cells. In addition of being the major track for ER remodeling in yeast [28–30], plants [11,31–34] and cell cycle dependently in C. elegans [35], there is a growing number of reports indicating that actin cytoskeleton plays a significant role in ER structure in other cell types as well: ER has been shown to align along actin fibers in kidney epithelial cells [26], and interact with actin filaments in insect photoreceptor cells [27,36], squid axons [37] and mouse embryonic fibroblasts [38]. In addition, retraction of ER to cell center upon MT depolymerization has been shown to be actin-based in frog kidney cell line [21]. Furthermore, actin cytoskeleton has also been linked to the regulation of Ca2þ homeostasis, showing that the Ca2þ release from ER is modulated by actin polymerization in rat hepatocytes [39]. In agreement, in 3D-models derived from Huh-7 cells subjected to HPF and highresolution electron tomography (ET), ER resides in close proximity with both actin filaments and MTs [5] (Fig. 1A). Based on 3D-EM imaging utilizing serial block face scanning electron microscopy (SB-EM), we often observe actin filaments traversing polygons (Fig. 1B), and MTs passing through fenestrations (Fig. 1C). This is consistent with the proposed role of fenestrations as a passage sites for cytoplasmic molecules, reducing the diffusion barrier that large intact sheets would otherwise pose. We recently demonstrated that latrunculin A treatment, a drug that inhibits actin polymerization leading to actin depolymerization, shifts the ER network towards tubular morphology and results in decrease in sheet prevalence and formation of sheet remnants Huh-7 and HeLa cells [5]. In contrast, the depolymerization of non-acetylated MTs with nocodazole [10] shifts the ER structural balance to opposite direction and leads to formation of interconnected sheet-mass at the cell center [5,9,40]. It is noteworthy that the total amount of ER did not change upon treatments, indicating that the observed changes in the ER morphology resulted from structural conversion of sheets and tubules and not from consumption of the ER [5]. In addition to depolymerization experiments, also stabilization/ bundling of MTs with paclitaxel [41] (Fig. 1D, Our unpublished results), or actin filaments with jasplakinolide [42] (Fig. 1E, Our unpublished results), lead to a striking ER phenotypes with extremely long and unbranched sheets at the central parts of the cells or ER network distribution defect with areas void of ER in the cell periphery and, respectively. These observations led us to suggest that regulation of sheet–tubule balance requires dynamic cytoskeletal filaments and that any perturbation on the filament systems induces changes in the ER subdomains they are connected to [5]. Previously, it has been shown that ER network is reconstructed through an iterative process of extension, branching, and intersection of new ER tubules along MTs but not actin after nocodazole washout in Vero cells [40]. However, the formation of sheets was not disclosed in this study and the role of actin on the maintenance of ER cannot be excluded. In agreement with suggestion that distinct ER domains might interact with different cytoskeletal elements [43], our results imply that actin cytoskeleton interacts specifically with a subset of ER, namely the sheets, instead of the whole network. In addition to interactions with actin, there is also evidence that sheets might interact with MTs. Sheet-localizing Climp-63 and p180 have been shown to bind directly to MTs and the overexpression of these proteins leads to MT bundling [62,63]. However, the role of Climp-63 as a luminal ER spacer does not
] (]]]]) ]]]–]]]
3
rely on its MT-binding and as the Climp-63 depletion has not been shown to abolish sheets, the physiological significance of the MT interaction with sheets remains open [25,44,45]. Interestingly, motor protein kinesin1 drives the ER tubular movement towards the MT plus end via kinectin, which has been shown to localize to sheets and, when over-expressed, causes sheet proliferation [46], indicating that while the major role assigned for kinesin1 is to accomplish tubular movement, kinesin1-kinectin interaction might have an additional role on sheets. It is, therefore, possible that different subsets of sheets and tubules exist, as demonstrated for specific set of tubules contacting maturing endosomes via acetylated MTs [10] and MT stabilization and bundling affecting only on centrally located sheets (Fig. 1D). Together, these results indicate a variety of possibilities for controlling the ER dynamics and morphology in different organisms and that this plasticity becomes in even more profound when taking into account the cell cycle and differentiation processes.
ER sheet transformations are coupled to a subset of dynamic actin filaments Using a combination of live-cell confocal imaging and highresolution EM, we recently revealed that ER interacts with a specific subset of actin filaments [5]. These dynamic short actin arrays and foci were found throughout the cytoplasm in the vicinity of ER, differing from relatively longer actin filaments, i.e. stress fibers or cortical actin, which had no clear interactions, nor simultaneous movement, with the ER. Our analysis of actin array dynamics revealed that they are connected to sheet movements and transformations: The relocation, disappearance, or formation of the actin arrays frequently preceded ER polygon closure or opening, respectively, leading to subsequent sheet transformation, i.e. sheets fusing to form larger sheets or splitting into smaller sheets. We also observed that the appearance of a new polygon within sheets followed closely the formation of an actin array. Sheet transformations could be induced by mimicking the natural loss of actin arrays from the vicinity of sheets latrunculin A treatment, showing a three-fold decrease in the sheet persistence. These effects could not be observed after MT depolymerization [5]. Based on center-of-mass analysis, actin depolymerization-induced sheet transformations were accompanied by increased lateral movement of sheets, i.e. sheets moved in a larger area, while the sheet velocity remained roughly the same (0.11 μm/ s70.04 SD) as in control. Acetylated MTs have been shown to act as tracks for ER tubule sliding [10], but the hyperacetylation of MTs with trichostatin A did not induce changes in the lateral movement of sheets and, moreover, actin depolymerization did not induce changes in the tubular dynamics. While in HeLa and NRK-52E the sheet persistence is lower than in Huh-7, the sheets correspond to actin depolymerization similarly to Huh-7 sheets (Our unpublished data). Together our results provide a mechanistic view on interplay between ER and actin, suggesting that the loss of actin filament arrays from the polygons allow sheets to move in a larger area and that the lateral movement is subsequently accompanied by increased sheet transformations. Our results indicate that the dynamics of actin arrays and the sheet transformations are interdependent and are in agreement with previous reports demonstrating that actin cytoskeleton interacts
Please cite this article as: M. Joensuu, E. Jokitalo, ER sheet–tubule balance is regulated by an array of actin filaments and microtubules, Exp Cell Res (2015), http://dx.doi.org/10.1016/j.yexcr.2015.04.009
4
E XP E RI ME N TAL CE L L R ES E ARC H
with sheets [26,27,35–37], while MTs have a role on the dynamics of ER tubules [7–9,14–18]. Furthermore, our results are consistent with previous studies showing that polygon filling within an sheet leads to
] (]]]]) ]]]–]]]
larger sheets [20] and that, at steady state, the ER network is balanced by polygon closure [7], while the mechanism behind these transformation has remained unclear.
E X PE R IM EN TA L C ELL R E S EA RC H
Proteins involved in ER–actin cytoskeleton interaction While a growing body of evidence indicate that ER interacts with actin cytoskeleton in wide variety of cell types, the reports on ER– actin interactions in mammalian cells have been mainly descriptive, and the mechanisms by which actin contributes to ER dynamics and/or morphology has remained unclear. The first protein identified in the ER–actin interphase was the inverted formin 2 (INF2) which is peripherally bound to the cytoplasmic face of the ER and possesses a unique ability to nucleate and depolymerize actin filaments [47]. Dominant-negative expression of mutated INF2 with abolished depolymerization activity caused ER to collapse around the nucleus, with accumulation of actin filaments around the collapsed ER. While the authors claim that INF2 knockdown does not cause apparent changes in the ER morphology, the study was done at in LM level and the ER ultrastructure was not studied. Interestingly, based on nerve biopsy, a recent study describes an ER expansion in axons and Schwann cells of intermediate type Charcot-Marie-Tooth neuropathy patients with INF2-mutation [48]. We recently described a novel role for unconventional motor protein myosin 1c (myo1c) in the interphase of ER and actin arrays. Myo1c was identified based on its depletion effects on the actin arrays and ER and from a screen of 4200 known human actin-binding proteins. Myo1c depletion or dominant-negative expression of mutated myo1c with abolished actin-binding (EGFP-myo1cΔABL) resulted in collapse of the short dynamic actin arrays to punctate foci while the overexpression of wt myo1c caused an induction of curved, bundled and unorganized actin arrays, indicating that the functional actin-binding domain of myo1c is required for creation and/or maintenance of ERassociated dynamic actin arrays. The localization of wt myo1c to the ER-associated dynamic actin arrays and under plasma membrane was shown to be dependent on the functional actin binding domain. Furthermore, the myo1c was observed to move
] (]]]]) ]]]–]]]
5
in-conjunction with actin arrays [5]. In agreement, myo1c and its rat ortholog, myr2, have been suggested to link actin cytoskeleton to cellular membranes and localize to actin-rich sites at the cell periphery and to discrete punctate within the cytoplasm [49–54] and myo1c has previously been suggested to maintain and organize actin filaments [55–61]. Providing further support to our results, myosin I has been shown to bind only dynamic actin filament population and not tropomyosin-stabilized microfilaments [62]. Another unconventional motor protein, myosin Va, has been shown to transport ER from the dendritic shaft to dendritic spine in Purkinje neurons, independent of MTs. The absence of ER in the dendritic spines of myosin Va knockout neurons can rescued by re-introducing wild-type protein, but not by its mutant [63–65]. In addition, myosins have been shown to be capable of shortrange ER mobility in vitro [66] and, in S. cerevisiae, the generation of ER tubules have been shown to be dependent on myosin 4p [67]. However, actin depolymerization does not collapse ER network in yeast [29], and even in mammalian cells ER retracts only with some delay upon MT depolymerization [9]. Thus, additional mechanisms, other than those based on MT and/or actin cytoskeleton, are likely to exist.
Myo1c regulation of actin arrays is conveyed to sheet morphology Based on our studies, the regulatory effects of myo1c on actin arrays were conveyed to changes in the sheet morphology, providing further evidence on the coupling of ER and actin cytoskeleton. We showed that the depletion of myo1c, and the subsequent loss of actin arrays, led to uncharacteristic ER network appearance and loss of sheets, similarly to latrunculin A-treated cells. Based on confocal live-cell imaging the loss of sheets could be rescued by re-introduction of wt myo1c, but not by EGFP– myo1cΔABL. The depletion-results were further supported by 3Dmodels obtained from SB-EM datasets where the uneven dis-
Fig. 1 – (A) 3D-model of wt Huh-7 shows that ER (yellow) resides in close proximity with actin filaments (red) and microtubules (light blue). Two successive 250-nm sections were prepared using HPF/FS, and subjected to ET and modeling. A few sheet fenestrations are marked with arrowheads for identification. Bar 0.5 lm. (B) SB-EM model of Huh-7/ssHRP-KDEL shows actin filaments (red) passing through polygons defined by the surrounding ER (yellow). Sheet fenestrations, considerably smaller in size than polygons, are indicated with arrowheads. Image is presented in orthogonal view and the 0.5 lm bar applies to center of the image. (C) MT passing through ER sheet fenestration. (a) A slice image of the tomogram shows a fenestration (arrowhead) in a transverse section of a sheet of Huh-7/ ssHRP-KDEL. (c–d) Double magnifications of the boxed area show slice images depicting a MT (open arrowheads) extending through the same fenestration (arrowhead). (e) A 3D-model of the boxed area of ER (yellow) shows the same fenestration (arrowhead) through which the MT (light blue) passes. Bar is 1 lm in (a) and 0.5 lm in (e). (D) MT stabilization and bundling induced a striking ER phenotype. (a) ER in the central part of the control Huh-7/ssHRP-KDEL cell is composed of large fenestrated (arrowheads) sheets with a complex 3Dstructure. (b) Wide-field image of the effect of paclitaxel treatment in live Huh-7/Hsp47-GFP. Long ER profiles (open arrowheads) are mainly observed at central parts of the cell. (c) Electron micrograph of paclitaxel-treated wt Huh-7 shows induction of long, straight and non-gapped ER profiles (arrows) that are closely associated with MT bundles (arrowheads) and directly continuous with fenestrated ER profiles (*). (d) SB-EM 3D-model of paclitaxel-treated Huh-7/ssHRP-KDEL shows planar and long ER (yellow) sheets lacking fenestrations from regions in close proximity of MTs (open arrow heads), but which are directly connected to the typical fenestrated (arrowheads) sheets. Bars are 10 lm (b), 0.5 lm in (c) and 1.0 lm in (a and d). (E) Actin stabilization and bundling affects the ER network distribution and sheet morphology. (a) A top view, (c) side-view and (b) a representative sheet of SB-EM 3D-model of ER network (yellow) and NE (blue) after jasplakinolide treatment in Huh-7/ssHRP-KDEL are shown. (d–e) Two block-face images taken at indicated heights from the beginning of the dataset are shown, and the NE and the boxed area subjected to 3D-modeling are indicated. (f) NE and areas void of ER (asterisks) are indicated in a wide-field image of wt Huh-7, labeled against endogenous calreticulin, after jasplakinolide treatment. Bar is 10 lm in (f) and 1 lm in (a–e). All cells expressing ssHRP-KDEL were cytochemically stained. (B, C, Da, Dc, Dd, E) were chemically fixed. Please cite this article as: M. Joensuu, E. Jokitalo, ER sheet–tubule balance is regulated by an array of actin filaments and microtubules, Exp Cell Res (2015), http://dx.doi.org/10.1016/j.yexcr.2015.04.009
6
E XP E RI ME N TAL CE L L R ES E ARC H
] (]]]]) ]]]–]]]
Fig. 2 – Dynamic microtubules and actin filament arrays counterbalance endoplasmic reticulum sheet–tubule balance. Perturbation of dynamic MTs or actin cytoskeleton affects the ER sheet–tubule balance. Actin depolymerization (latrunculin A) or stabilization (jasplakinolide) shift the ER balance towards tubules and lead to sheet remnants, whereas MT depolymerization (nocodazole) shifts the balance towards sheets and results in an induction of sheets and the MT stabilization and bundling (paclitaxel) results in long and less fenestrated straight sheets. Myo1c manipulations (myo1c depletion or dominant-negative expression of EGFP-myo1cΔABL with mutated actin-binding domain) affect the structure of actin arrays and the effect is conveyed to ER sheet morphology similarly to actin depolymerization effects. Tubular ER phenotype of myo1c-depleted Huh-7 cells can be rescued with re-introduction of myo1c, but not by its mutated form (EGFP–myo1cΔABL). LM images present chemically fixed wt Huh-7 cells, immunolabeled for endogenous calreticulin. SB-EM models of representative after indicated treatments are shown in yellow in the top row and are derived from chemically fixed and cytochemically stained Huh-7/ssHRP-KDEL.
tribution and sheet remnants were even more evident. The overexpression of myo1c shifted the sheet–tubule ratio toward sheets, with few tubules remaining and, instead of individual sheets, an interlinked sheet mass was observed. In contrast, dominant-negative expression of EGFP–myo1cΔABL led to almost complete structural shift of ER toward a reticular network, as only few peripheral sheets were observed [5]. Immunoreactivity studies of myr2 from isolated sucrose gradient membrane fractions have showed that Myr2 associates with fractions enriched with ER and PM, but not with other organelles [50]. Myo1c has been shown to bind specifically and with high affinity to phosphatidylinositol 4,5-bisphosphate (PIP2) via its pleckstrin homology (PH) domain, and to lesser extent to other lipids [52,68], and based on live cell imaging experiments, myo1c localization in the distinct punctate in cells is dependent on its binding to PIP2 [52,69]. In accordance, Myr2 has been shown to localize to lipid rafts enriched with PIP2 [70]. It has been suggested that the myo1c–PIP2 interaction may serve to concentrate myo1c at regions of new actin polymerization, providing mechanism for recruitment of the motor to areas driving membrane–actin interactions [55,57]. As PIP2 has been shown to be enriched at ER [71,72], it could be envisioned that PIP2 recruits myo1c to ER. This interaction, maybe together with other factors, might facilitate the recruitment of dynamic actin to ER to drive sheet transformations and to facilitate the sheet persistence. In agreement with this proposal, myo1 and actin cytoskeleton have been shown to provide mechanical stability to microvillar ER [73].
The actual mechanism for ER–actin interplay, however, remains open; while myo1c might serve to localize actin filaments to the vicinity of ER, these actin arrays might either form passive physical restriction barriers at polygons or yet unidentified factors could be needed to mediate the interaction between ER and actin filaments.
Dynamic microtubules and actin filament arrays counterbalance sheet–tubule balance It has been suggested that the ER structure at cell periphery is determined by a balance of MT and actin and their motor proteins [5,21]. Based on our results, actin cytoskeleton and MT manipulations have opposing effects on ER morphology, indicating that there is a tug-of-war between dynamic actin filaments and MTs on sheet–tubule ratio (Fig. 2). MT manipulations shifts the morphological balance towards sheets and result in stacked sheets (upon nocodazole treatment) or straight, extremely long and less fenestrated sheets (upon MT stabilization by paclitaxel [41]) compared to control sheets. Actin depolymerization or by latrunculin A or jasplakinolide treatment, respectively, shifted the balance towards tubular morphology and resulted in sheet remnants. These drug treatments had opposing effects on MTs and actin -either disassembling or stabilizing the filaments. Why, then, the effects of the treatments caused similar effects on the sheet–tubule ratio? One explanation could be that the regulation
Please cite this article as: M. Joensuu, E. Jokitalo, ER sheet–tubule balance is regulated by an array of actin filaments and microtubules, Exp Cell Res (2015), http://dx.doi.org/10.1016/j.yexcr.2015.04.009
E X PE R IM EN TA L C ELL R E S EA RC H
of sheet–tubule balance requires dynamic cytoskeletal filaments. In accordance, our results showed that myo1c regulates the dynamic actin filament arrays, which in-turn, regulate the sheet persistence. Myo1c depletion or dominant-negative expression of EGFP–myo1cΔABL, shifted the sheet–tubule balance towards tubular morphology and resulted in sheet remnants. Myo1c depletion phenotype could be rescued by over-expressing the wt myo1c, which shifted the sheet–tubule balance towards sheets-resembling the over-expression of wt myo1c. As the cytoplasm is full of MT- and actin-binding proteins that provide identity to cytoskeletal filaments in response to various intra- and extracellular cues, the interplay between ER subdomains and different cytoskeletal elements provide a variety of regulatory possibilities. In addition to integral membrane proteins that support specific structural features, interaction with the cytoskeleton provide a more global and dynamic mechanism to control the ER architecture in cells. MT-driven tubular ER is suited to host dynamic functions and provide a mechanism to rearrange the overall network distribution within the cell. In contrast, more stationary and persistent sheets might be more suitable for protein synthesis and quality control or other ER functions. The functional importance of ER network transformations is an important research question for the future.
Acknowledgments The authors gratefully acknowledge funding from The Academy of Finland (Project no. 131650) and the Biocenter Finland. Drs. Helena Vihinen, Maija Puhka and Ilya Belevich (Institute of Biotechnology, Electron Microscopy Unit, University of Helsinki) are acknowledged for their contributions especially on 3D-EM and HPF.
references [1] M. Puhka, M. Joensuu, H. Vihinen, I. Belevich, E. Jokitalo, Progressive sheet-to-tubule transformation is a general mechanism for endoplasmic reticulum partitioning in dividing mammalian cells, Mol. Biol. Cell 23 (2012) 2424–2432, http://dx. doi.org/10.1091/mbc.E10-12-0950. [2] M. West, N. Zurek, A. Hoenger, G.K. Voeltz, A 3D analysis of yeast ER structure reveals how ER domains are organized by membrane curvature, J. Cell Biol. 193 (2011) 333–346, http://dx.doi. org/10.1083/jcb.201011039. [3] M. Puhka, H. Vihinen, M. Joensuu, E. Jokitalo, Endoplasmic reticulum remains continuous and undergoes sheet-to-tubule transformation during cell division in mammalian cells, J. Cell Biol. 179 (2007) 895–909, http://dx.doi.org/10.1083/ jcb.200705112. [4] T. Shemesh, et al., A model for the generation and interconversion of ER morphologies, Proc. Natl. Acad. Sci. USA 111 (2014) E5243–E5251, http://dx.doi.org/10.1073/pnas.1419997111. [5] M. Joensuu, et al., ER sheet persistence is coupled to myosin 1cregulated dynamic actin filament arrays, Mol. Biol. Cell 25 (2014) 1111–1126, http://dx.doi.org/10.1091/mbc.E13-12-0712. [6] L. Lu, M.S. Ladinsky, T. Kirchhausen, Cisternal organization of the endoplasmic reticulum during mitosis, Mol. Biol. Cell 20 (2009) 3471–3480, http://dx.doi.org/10.1091/mbc.E09-04-0327. [7] C. Lee, L.B. Chen, Dynamic behavior of endoplasmic reticulum in living cells, Cell 54 (1988) 37–46. [8] C.M. Waterman-Storer, E.D. Salmon, Endoplasmic reticulum membrane tubules are distributed by microtubules in living cells using three distinct mechanisms, Curr. Biol. 8 (1998) 798–806.
] (]]]]) ]]]–]]]
7
[9] M. Terasaki, L.B. Chen, K. Fujiwara, Microtubules and the endoplasmic reticulum are highly interdependent structures, J. Cell Biol. 103 (1986) 1557–1568. [10] J.R. Friedman, B.M. Webster, D.N. Mastronarde, K.J. Verhey, G.K. Voeltz, ER sliding dynamics and ER-mitochondrial contacts occur on acetylated microtubules, J. Cell Biol. 190 (2010) 363–375, http: //dx.doi.org/10.1083/jcb.200911024. [11] I. Sparkes, J. Runions, C. Hawes, L. Griffing, Movement and remodeling of the endoplasmic reticulum in nondividing cells of tobacco leaves, Plant Cell 21 (2009) 3937–3949, http://dx.doi.org/ 10.1105/tpc.109.072249. [12] W. Knebel, H. Quader, E. Schnepf, Mobile and immobile endoplasmic reticulum in onion bulb epidermis cells: short- and longterm observations with a confocal laser scanning microscope, Eur. J. Cell Biol. 52 (1990) 328–340. [13] G.K. Voeltz, W.A. Prinz, Y. Shibata, J.M. Rist, T.A. Rapoport, A class of membrane proteins shaping the tubular endoplasmic reticulum, Cell 124 (2006) 573–586, http://dx.doi.org/10.1016/j. cell.2005.11.047. [14] H. Bannai, T. Inoue, T. Nakayama, M. Hattori, K. Mikoshiba, Kinesin dependent, rapid, bi-directional transport of ER subcompartment in dendrites of hippocampal neurons, J. Cell Sci. 117 (2004) 163–175, http://dx.doi.org/10.1242/jcs.00854. [15] I. Grigoriev, et al., STIM1 is a MT-plus-end-tracking protein involved in remodeling of the ER, Curr. Biol. 18 (2008) 177–182, http://dx.doi.org/10.1016/j.cub.2007.12.050. [16] M.J. Wozniak, et al., Role of kinesin-1 and cytoplasmic dynein in endoplasmic reticulum movement in VERO cells, J. Cell Sci. 122 (2009) 1979–1989, http://dx.doi.org/10.1242/jcs.041962. [17] J. Kumar, H. Yu, M.P. Sheetz, Kinectin, an essential anchor for kinesin-driven vesicle motility, Science 267 (1995) 1834–1837. [18] A.V. Nikonov, H.P. Hauri, B. Lauring, G. Kreibich, Climp-63mediated binding of microtubules to the ER affects the lateral mobility of translocon complexes, J. Cell Sci. 120 (2007) 2248– 2258, http://dx.doi.org/10.1242/jcs.008979. [19] H. Quader, A. Hofmann, E. Schnepf, Reorganization of the endoplasmic reticulum in epidermal cells of onion bulb scales after cold stress: involvement of cytoskeletal elements, Planta 177 (1989) 273–280, http://dx.doi.org/10.1007/BF00392816. [20] L.R. Griffing, Networking in the endoplasmic reticulum, Biochem. Soc. Trans. 38 (2010) 747–753, http://dx.doi.org/10.1042/ BST0380747. [21] M. Terasaki, T.S. Reese, Interactions among endoplasmic reticulum, microtubules, and retrograde movements of the cell surface, Cell Motil. Cytoskelet. 29 (1994) 291–300, http://dx.doi.org/ 10.1002/cm.970290402. [22] L. Dreier, T.A. Rapoport, In vitro formation of the endoplasmic reticulum occurs independently of microtubules by a controlled fusion reaction, J. Cell Biol. 148 (2000) 883–898. [23] J. Hu, et al., A class of dynamin-like GTPases involved in the generation of the tubular ER network, Cell 138 (2009) 549–561, http://dx.doi.org/10.1016/j.cell.2009.05.025. [24] C. Lavoie, J. Lanoix, F.W. Kan, J. Paiement, Cell-free assembly of rough and smooth endoplasmic reticulum, J. Cell Sci. 109 (Pt 6) (1996) 1415–1425. [25] D.R. Klopfenstein, F. Kappeler, H.P. Hauri, A novel direct interaction of endoplasmic reticulum with microtubules, EMBO J. 17 (1998) 6168–6177, http://dx.doi.org/10.1093/emboj/17.21.6168. [26] J.M. Sanger, J.S. Dome, B. Mittal, A.V. Somlyo, J.W. Sanger, Dynamics of the endoplasmic reticulum in living non-muscle and muscle cells, Cell Motil. Cytoskelet. 13 (1989) 301–319, http: //dx.doi.org/10.1002/cm.970130408. [27] K. Sturmer, O. Baumann, B. Walz, Actin-dependent light-induced translocation of mitochondria and ER cisternae in the photoreceptor cells of the locust Schistocerca gregaria, J. Cell Sci. 108 (Pt 6) (1995) 2273–2283. [28] D. Pruyne, A. Legesse-Miller, L. Gao, Y. Dong, A. Bretscher, Mechanisms of polarized growth and organelle segregation in
Please cite this article as: M. Joensuu, E. Jokitalo, ER sheet–tubule balance is regulated by an array of actin filaments and microtubules, Exp Cell Res (2015), http://dx.doi.org/10.1016/j.yexcr.2015.04.009
8
E XP E RI ME N TAL CE L L R ES E ARC H
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40] [41]
[42]
[43]
[44]
[45]
[46]
yeast, Annu. Rev. Cell Dev. Biol. 20 (2004) 559–591, http://dx.doi. org/10.1146/annurev.cellbio.20.010403.103108. W.A. Prinz, et al., Mutants affecting the structure of the cortical endoplasmic reticulum in Saccharomyces cerevisiae, J. Cell Biol. 150 (2000) 461–474. K.L. Fehrenbacher, D. Davis, M. Wu, I. Boldogh, L.A. Pon, Endoplasmic reticulum dynamics, inheritance, and cytoskeletal interactions in budding yeast, Mol. Biol. Cell 13 (2002) 854–865, http://dx.doi.org/10.1091/mbc.01-04-0184. S.L. Gupton, D.A. Collings, N.S. Allen, Endoplasmic reticulum targeted GFP reveals ER organization in tobacco NT-1 cells during cell division, Plant Physiol. Biochem.: Ppb/Soc. Francaise Physiol. Veg. 44 (2006) 95–105, http://dx.doi.org/10.1016/j. plaphy.2006.03.003. H. Ueda, et al., Myosin-dependent endoplasmic reticulum motility and F-actin organization in plant cells, Proc. Natl. Acad. Sci. USA 107 (2010) 6894–6899, http://dx.doi.org/10.1073/ pnas.0911482107. P. Boevink, et al., Stacks on tracks: the plant Golgi apparatus traffics on an actin/ER network, Plant J.: Cell Mol. Biol. 15 (1998) 441–447. F. Lv, et al., GhCFE1A, a dynamic linker between the ER network and actin cytoskeleton, plays an important role in cotton fibre cell initiation and elongation, J. Exp. Bot. 66 (2015) 1877–1889, http://dx.doi.org/10.1093/jxb/eru530. D. Poteryaev, J.M. Squirrell, J.M. Campbell, J.G. White, A. Spang, Involvement of the actin cytoskeleton and homotypic membrane fusion in ER dynamics in Caenorhabditis elegans, Mol. Biol. Cell 16 (2005) 2139–2153, http://dx.doi.org/10.1091/mbc.E04-080726. O. Baumann, B. Lautenschlager, The role of actin filaments in the organization of the endoplasmic reticulum in honeybee photoreceptor cells, Cell Tissue Res. 278 (1994) 419–432. S.A. Kuznetsov, D.T. Rivera, F.F. Severin, D.G. Weiss, G.M. Langford, Movement of axoplasmic organelles on actin filaments from skeletal muscle, Cell Motil. Cytoskelet. 28 (1994) 231–242, http: //dx.doi.org/10.1002/cm.970280306. C.D. Lynch, et al., Filamin depletion blocks endoplasmic spreading and destabilizes force-bearing adhesions, Mol. Biol. Cell 22 (2011) 1263–1273, http://dx.doi.org/10.1091/mbc.E10-08-0661. Y.J. Wang, R.B. Gregory, G.J. Barritt, Maintenance of the filamentous actin cytoskeleton is necessary for the activation of storeoperated Ca2þ channels, but not other types of plasmamembrane Ca2þ channels, in rat hepatocytes, Biochem. J. 363 (2002) 117–126. C. Lee, M. Ferguson, L.B. Chen, Construction of the endoplasmic reticulum, J. Cell Biol. 109 (1989) 2045–2055. A.M. Yvon, P. Wadsworth, M.A. Jordan, Taxol suppresses dynamics of individual microtubules in living human tumor cells, Mol. Biol. Cell 10 (1999) 947–959. A. Holzinger, Jasplakinolide: an actin-specific reagent that promotes actin polymerization, Methods Mol. Biol. 586 (2009) 71– 87, http://dx.doi.org/10.1007/978-1-60761-376-3_4. O. Baumann, Structural Interactions of Actin-Filaments and Endoplasmic-Reticulum in Honeybee Photoreceptor Cells, Cell Tissue Res. 268 (1992) 71–79, http://dx.doi.org/10.1007/ Bf00338055. D.R. Klopfenstein, et al., Subdomain-specific localization of CLIMP-63 (p63) in the endoplasmic reticulum is mediated by its luminal alpha-helical segment, J. Cell Biol. 153 (2001) 1287– 1300. C. Vedrenne, D.R. Klopfenstein, H.P. Hauri, Phosphorylation controls CLIMP-63-mediated anchoring of the endoplasmic reticulum to microtubules, Mol. Biol. Cell 16 (2005) 1928–1937, http: //dx.doi.org/10.1091/mbc.E04-07-0554. N. Santama, C.P. Er, L.L. Ong, H. Yu, Distribution and functions of kinectin isoforms, J. Cell Sci. 117 (2004) 4537–4549, http://dx.doi. org/10.1242/jcs.01326.
] (]]]]) ]]]–]]]
[47] E.S. Chhabra, V. Ramabhadran, S.A. Gerber, H.N. Higgs, INF2 is an endoplasmic reticulum-associated formin protein, J. Cell Sci. 122 (2009) 1430–1440, http://dx.doi.org/10.1242/jcs.040691. [48] A. Roos, et al., Inverted formin 2-related Charcot-Marie-Tooth disease: extension of the mutational spectrum and pathological findings in Schwann cells and axons, J. Peripher. Nerv. Syst. . (2015)http://dx.doi.org/10.1111/jns.12106. [49] M.C. Wagner, B. Barylko, J.P. Albanesi, Tissue distribution and subcellular localization of mammalian myosin I, J. Cell Biol. 119 (1992) 163–170. [50] C. Ruppert, et al., Localization of the rat myosin I molecules myr 1 and myr 2 and in vivo targeting of their tail domains, J. Cell Sci. 108 (Pt 12) (1995) 3775–3786. [51] K.D. Novak, M.D. Peterson, M.C. Reedy, M.A. Titus, Dictyostelium myosin I double mutants exhibit conditional defects in pinocytosis, J. Cell Biol. 131 (1995) 1205–1221. [52] D.E. Hokanson, J.M. Laakso, T. Lin, D. Sept, E.M. Ostap, Myo1c binds phosphoinositides through a putative pleckstrin homology domain, Mol. Biol. Cell 17 (2006) 4856–4865, http://dx.doi.org/ 10.1091/mbc.E06-05-0449. [53] M.C. Wagner, B.A. Molitoris, ATP depletion alters myosin I beta cellular location in LLC-PK1 cells, Am. J. Physiol. 272 (1997) C1680–1690. [54] O. Reizes, B. Barylko, C. Li, T.C. Sudhof, J.P. Albanesi, Domain structure of a mammalian myosin I beta, Proc. Natl. Acad. Sci. USA 91 (1994) 6349–6353. [55] A. Bose, et al., Unconventional myosin Myo1c promotes membrane fusion in a regulated exocytic pathway, Mol. Cell. Biol. 24 (2004) 5447–5458, http://dx.doi.org/10.1128/MCB.24.12.54475458.2004. [56] G.N. Hagan, Y. Lin, M.A. Magnuson, J. Avruch, M.P. Czech, A Rictor-Myo1c complex participates in dynamic cortical actin events in 3T3-L1 adipocytes, Mol. Cell. Biol. 28 (2008) 4215– 4226, http://dx.doi.org/10.1128/MCB.00867-07. [57] T.J. Diefenbach, et al., Myosin 1c and myosin IIB serve opposing roles in lamellipodial dynamics of the neuronal growth cone, J. Cell Biol. 158 (2002) 1207–1217. [58] Y. Fan, S.M. Eswarappa, M. Hitomi, P.L. Fox, Myo1c facilitates G-actin transport to the leading edge of migrating endothelial cells, J. Cell Biol. 198 (2012) 47–55, http://dx.doi.org/10.1083/ jcb.201111088. [59] E. Arif, et al., Motor protein Myo1c is a podocyte protein that facilitates the transport of slit diaphragm protein Neph1 to the podocyte membrane, Mol. Cell. Biol. 31 (2011) 2134–2150, http: //dx.doi.org/10.1128/MCB.05051-11. [60] P. Garg, R. Verma, D. Nihalani, D.B. Johnstone, L.B. Holzman, Neph1 cooperates with nephrin to transduce a signal that induces actin polymerization, Mol. Cell. Biol. 27 (2007) 8698– 8712, http://dx.doi.org/10.1128/MCB.00948-07. [61] J.L. Maravillas-Montero, P.G. Gillespie, G. Patino-Lopez, S. Shaw, L. Santos-Argumedo, Myosin 1c participates in B cell cytoskeleton rearrangements, is recruited to the immunologic synapse, and contributes to antigen presentation, J. Immunol. 187 (2011) 3053–3063, http://dx.doi.org/10.4049/jimmunol.1004018. [62] N. Tang, E.M. Ostap, Motor domain-dependent localization of myo1b (myr-1), Curr. Biol. 11 (2001) 1131–1135. [63] Y. Takagishi, et al., The dilute-lethal (dl) gene attacks a Ca2þ store in the dendritic spine of Purkinje cells in mice, Neurosci. Lett. 215 (1996) 169–172. [64] P.C. Bridgman, Myosin Va movements in normal and dilutelethal axons provide support for a dual filament motor complex, J. Cell Biol. 146 (1999) 1045–1060. [65] W. Wagner, S.D. Brenowitz, J.A. Hammer 3rd, Myosin-Va transports the endoplasmic reticulum into the dendritic spines of Purkinje neurons, Nat. Cell Biol. 13 (2011) 40–48, http://dx.doi. org/10.1038/ncb2132. [66] T. Wollert, D.G. Weiss, H.H. Gerdes, S.A. Kuznetsov, Activation of myosin V-based motility and F-actin-dependent network
Please cite this article as: M. Joensuu, E. Jokitalo, ER sheet–tubule balance is regulated by an array of actin filaments and microtubules, Exp Cell Res (2015), http://dx.doi.org/10.1016/j.yexcr.2015.04.009
E X PE R IM EN TA L C ELL R E S EA RC H
formation of endoplasmic reticulum during mitosis, J. Cell Biol. 159 (2002) 571–577, http://dx.doi.org/10.1083/jcb.200204065. [67] P. Estrada, et al., Myo4p and She3p are required for cortical ER inheritance in Saccharomyces cerevisiae, J. Cell Biol. 163 (2003) 1255–1266, http://dx.doi.org/10.1083/jcb.200304030. [68] M. Hirono, C.S. Denis, G.P. Richardson, P.G. Gillespie, Hair cells require phosphatidylinositol 4,5-bisphosphate for mechanical transduction and adaptation, Neuron 44 (2004) 309–320, http: //dx.doi.org/10.1016/j.neuron.2004.09.020. [69] D.E. Hokanson, E.M. Ostap, Myo1c binds tightly and specifically to phosphatidylinositol 4,5-bisphosphate and inositol 1,4,5-trisphosphate, Proc. Natl. Acad. Sci. USA 103 (2006) 3118–3123, http://dx.doi.org/10.1073/pnas.0505685103.
] (]]]]) ]]]–]]]
9
[70] B. Barylko, G. Jung, J.P. Albanesi, Structure, function, and regulation of myosin 1C, Acta Biochim. Pol. 52 (2005) 373–380. [71] G. Griffiths, et al., The dynamic nature of the Golgi complex, J. Cell Biol. 108 (1989) 277–297. [72] F.Zambrano, S. Fleischer, B. Fleischer, Lipid composition of the Golgi apparatus of rat kidney and liver in comparison with other subcellular organelles, Biochim. Biophys. Acta 380 (1975) 357–369. [73] T.D. Pollard, S.K. Doberstein, H.G. Zot, Myosin-I, Annu. Rev. Physiol. 53 (1991) 653–681, http://dx.doi.org/10.1146/annurev. ph.53.030191.003253.
Please cite this article as: M. Joensuu, E. Jokitalo, ER sheet–tubule balance is regulated by an array of actin filaments and microtubules, Exp Cell Res (2015), http://dx.doi.org/10.1016/j.yexcr.2015.04.009
] (]]]]) ]]]–]]]
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/yexcr
Review Article
ER sheet–tubule balance is regulated by an array of actin filaments and microtubules Merja Joensuua, Eija Jokitaloa,b,n a
Institute of Biotechnology, Cell and Molecular Biology Program, University of Helsinki, 00014 Helsinki, Finland Electron Microscopy Unit, University of Helsinki, 00014 Helsinki, Finland
b
article information Article Chronology: Received 2 April 2015 Accepted 13 April 2015 Keywords: Endoplasmic reticulum Actin cytoskeleton 3D-EM Morhology Dynamics Sheet–tubule balance Microtubules
Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ER consist of dynamic and static subdomains. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ER, actin cytoskeleton and microtubules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ER sheet transformations are coupled to a subset of dynamic actin filaments . . . . . . . . Proteins involved in ER–actin cytoskeleton interaction. . . . . . . . . . . . . . . . . . . . . . . . . . . Myo1c regulation of actin arrays is conveyed to sheet morphology. . . . . . . . . . . . . . . . . Dynamic microtubules and actin filament arrays counterbalance sheet–tubule balance . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
n
Corresponding author at: Institute of Biotechnology, Cell and Molecular Biology Program, University of Helsinki, 00014 Helsinki, Finland. E-mail address: eija.jokitalo@helsinki.fi (E. Jokitalo).
http://dx.doi.org/10.1016/j.yexcr.2015.04.009 0014-4827/& 2015 Elsevier Inc. All rights reserved.
Please cite this article as: M. Joensuu, E. Jokitalo, ER sheet–tubule balance is regulated by an array of actin filaments and microtubules, Exp Cell Res (2015), http://dx.doi.org/10.1016/j.yexcr.2015.04.009
1 2 2 3 5 5 6 7 7
2
E XP E RI ME N TAL CE L L R ES E ARC H
Introduction Endoplasmic reticulum (ER) is a single continuous membrane network containing multiple subdomains with different structures and functions. Gradually over decades of research, various aspects of ER morphology, organization and dynamics in different organisms and cell cultures at interphase and upon cell division have been revealed, together highlighting the significant plasticity of the organelle. It is important to understand that depending on cell type and organism, the ratio between individual subdomains (i.e. sheets and tubules) and network distribution varies significantly, and therefore drawing generalized conclusions out of observations made in one model system should be done carefully. In addition, the structure of ER subdomains varies between different cell types, e.g. the average size of the sheets, and the presence/absence of sheet fenestrations (i.e. small perforations passing through the sheets). Fenestrations are especially observed in cell types with abundant (Huh-7 and HeLa [1]) or large (plasma membrane associated ER in Saccharomyces cerevisiae [2]) sheets, and are absent or sparse in cell types where the network is predominantly tubular and sheets smaller (CHO-K1 and NRK-52E [1,3]). While some discrepancy related to the ER sheet fenestrations [1,4,5] and the mechanism for ER inheritance [1,3,6] exist, the differing views can be explained by the use of different model systems and cells. Fenestration can be observed irrespective of fixation method (i.e. chemical fixation vs. high pressure freezing, HPF) [1,5] or expression of an ER-marker [1,5], but as their diameter falls beyond the resolution limit of conventional light microscopy (LM), they can, therefore, be visualized only with high-resolution microscopy. ER network polygons (i.e. the space between sheets and tubules) and fenestrations present different structures, as fenestrations are considerably smaller openings on the sheets. In contrast to nuclear pores found from nuclear envelope (NE), which are uniform in size and formed by a special set of proteins, fenestrations do not appear to have a specific pore structure and they vary somewhat in shape and size, averaging from 7573 nm in diameter [1], but are similarly found on planar sheets. ER architecture is modified to meet specific cellular needs, implying that several means must exist to accomplish such adaptations. While ER structure is in constant flux, it does not move en masse, and the network movement is achieved through dynamics of individual subdomains and through network remodeling [7–12]. While integral membrane proteins are indispensable for the ER structure [13], the dynamics and the network remodeling are accomplished through interactions with the cytoskeleton. The interplay between tubules and microtubules (MT) has been well described [7–12,14–18], but the maintenance of sheets and the role of actin cytoskeleton on ER structure in mammalian cells have remained obscure. Here, we focus on the ER association with actin cytoskeleton determining ER sheet structure and dynamics in cultured mammalian cells. In our studies, we chose to use Huh-7 as our main model system because the highly abundant ER in these active secretory cells is composed mainly of large sheets spreading throughout the cytoplasm while tubules are found around the Golgi complex and in peripheral area of the cell [1,5].
ER consist of dynamic and static subdomains In most cell types, ER tubules and sheets create a polygonal network, which can be altered by changing branching patterns or
] (]]]]) ]]]–]]]
by remodeling of tubules into sheets and vice versa [5,7–9,12,19]. Studies done on ER dynamics have mostly concentrated on describing the tubular dynamics and transformations. The ground work on ER dynamics in animal cells was done already in 1988 by Lee and Chen [7], who described the remodeling the ER network by tubular fusion and junction sliding, which create new branch points and polygons, or result in polygon elimination from the network. In addition to these dynamic events, several additional events are common in both plant and animal cells: tubules retract, kink and deform, new sheets are created by filling the polygon with membrane, and sheets are created or altered by polygon opening within sheets [5,8,9,11,20,21]. Sheets transform into tubules by extensive fenestration [1] or by consumption of the sheet membrane [5]. Based on Huh-7 live-cell imaging, we recently showed that while sheets can transform into tubules, split into smaller sheets and fuse with one another to form larger sheets, a majority of sheets are relatively static structures [5]. In contrast, the majority of the tubules appear to be in a constant dynamic state [8–10,16]. By adapting a method that calculates the sheet's center-of-mass, we were able to show that sheet movements do not exhibit any clear directionality [5], differing from the previously described tubular movement [8–10,15], as the sheets mainly fluctuate in small area. In addition, the average sheet velocity (0.15 mm/s70.23 SD [5]) differs from that of tubules [8,10,15]. In HeLa and NRK-52E, the sheets are relatively static and the average sheet velocities were roughly the same compared to Huh-7 (Our unpublished results). These results demonstrate that sheets remain rather static in otherwise dynamic network, implying that the current view of the ER dynamics cannot be generalized to the whole network. Moreover, it is likely that the sheet repositioning in the cells is likely to occur through formation of a new sheet rather than transport of an existing one. Based on differing tubule and sheet dynamics, it is clear that multiple means must exist to accomplish the dynamics described above. While the purpose of the remodeling is unclear, it is thought that network structure, and perhaps, ultimately, the distribution of ER's functions into specific subdomains, are a consequence of the balance between dynamics, transitions and persistence of sheets and tubules [20].
ER, actin cytoskeleton and microtubules In vitro formation of reticular ER network has been shown to occur independent of MTs and actin [22] but requires the presence of curvature-inducing proteins [13,22,23]. It is, however, unclear if sheets were present in the formed ER network. Furthermore, a cell-free assembly of ER tubules and ribosomecovered sheets has been reported in in vitro experiments using rat hepatocyte microsomes [24]. The authors demonstrated that GTP is required for reconstitution of ribosome-covered sheets and that both GTP and ATP are required for tubule formation. However, to our view, the formed ER network appeared uncharacteristic and it is plausible that the formation of characteristic polygonal ER network requires interplay of several factors – including the different cytoskeletal filaments. Although MTs and the associated protein machinery are the main mechanism for tubular ER remodeling in interphase animal cells [8–10,14–16,25,26], MTs and ER do not have identical distribution in cells [10,21,27] and there is a body of work
Please cite this article as: M. Joensuu, E. Jokitalo, ER sheet–tubule balance is regulated by an array of actin filaments and microtubules, Exp Cell Res (2015), http://dx.doi.org/10.1016/j.yexcr.2015.04.009
E X PE R IM EN TA L C ELL R E S EA RC H
indicating that ER can also interact with actin cytoskeleton in animal cells. In addition of being the major track for ER remodeling in yeast [28–30], plants [11,31–34] and cell cycle dependently in C. elegans [35], there is a growing number of reports indicating that actin cytoskeleton plays a significant role in ER structure in other cell types as well: ER has been shown to align along actin fibers in kidney epithelial cells [26], and interact with actin filaments in insect photoreceptor cells [27,36], squid axons [37] and mouse embryonic fibroblasts [38]. In addition, retraction of ER to cell center upon MT depolymerization has been shown to be actin-based in frog kidney cell line [21]. Furthermore, actin cytoskeleton has also been linked to the regulation of Ca2þ homeostasis, showing that the Ca2þ release from ER is modulated by actin polymerization in rat hepatocytes [39]. In agreement, in 3D-models derived from Huh-7 cells subjected to HPF and highresolution electron tomography (ET), ER resides in close proximity with both actin filaments and MTs [5] (Fig. 1A). Based on 3D-EM imaging utilizing serial block face scanning electron microscopy (SB-EM), we often observe actin filaments traversing polygons (Fig. 1B), and MTs passing through fenestrations (Fig. 1C). This is consistent with the proposed role of fenestrations as a passage sites for cytoplasmic molecules, reducing the diffusion barrier that large intact sheets would otherwise pose. We recently demonstrated that latrunculin A treatment, a drug that inhibits actin polymerization leading to actin depolymerization, shifts the ER network towards tubular morphology and results in decrease in sheet prevalence and formation of sheet remnants Huh-7 and HeLa cells [5]. In contrast, the depolymerization of non-acetylated MTs with nocodazole [10] shifts the ER structural balance to opposite direction and leads to formation of interconnected sheet-mass at the cell center [5,9,40]. It is noteworthy that the total amount of ER did not change upon treatments, indicating that the observed changes in the ER morphology resulted from structural conversion of sheets and tubules and not from consumption of the ER [5]. In addition to depolymerization experiments, also stabilization/ bundling of MTs with paclitaxel [41] (Fig. 1D, Our unpublished results), or actin filaments with jasplakinolide [42] (Fig. 1E, Our unpublished results), lead to a striking ER phenotypes with extremely long and unbranched sheets at the central parts of the cells or ER network distribution defect with areas void of ER in the cell periphery and, respectively. These observations led us to suggest that regulation of sheet–tubule balance requires dynamic cytoskeletal filaments and that any perturbation on the filament systems induces changes in the ER subdomains they are connected to [5]. Previously, it has been shown that ER network is reconstructed through an iterative process of extension, branching, and intersection of new ER tubules along MTs but not actin after nocodazole washout in Vero cells [40]. However, the formation of sheets was not disclosed in this study and the role of actin on the maintenance of ER cannot be excluded. In agreement with suggestion that distinct ER domains might interact with different cytoskeletal elements [43], our results imply that actin cytoskeleton interacts specifically with a subset of ER, namely the sheets, instead of the whole network. In addition to interactions with actin, there is also evidence that sheets might interact with MTs. Sheet-localizing Climp-63 and p180 have been shown to bind directly to MTs and the overexpression of these proteins leads to MT bundling [62,63]. However, the role of Climp-63 as a luminal ER spacer does not
] (]]]]) ]]]–]]]
3
rely on its MT-binding and as the Climp-63 depletion has not been shown to abolish sheets, the physiological significance of the MT interaction with sheets remains open [25,44,45]. Interestingly, motor protein kinesin1 drives the ER tubular movement towards the MT plus end via kinectin, which has been shown to localize to sheets and, when over-expressed, causes sheet proliferation [46], indicating that while the major role assigned for kinesin1 is to accomplish tubular movement, kinesin1-kinectin interaction might have an additional role on sheets. It is, therefore, possible that different subsets of sheets and tubules exist, as demonstrated for specific set of tubules contacting maturing endosomes via acetylated MTs [10] and MT stabilization and bundling affecting only on centrally located sheets (Fig. 1D). Together, these results indicate a variety of possibilities for controlling the ER dynamics and morphology in different organisms and that this plasticity becomes in even more profound when taking into account the cell cycle and differentiation processes.
ER sheet transformations are coupled to a subset of dynamic actin filaments Using a combination of live-cell confocal imaging and highresolution EM, we recently revealed that ER interacts with a specific subset of actin filaments [5]. These dynamic short actin arrays and foci were found throughout the cytoplasm in the vicinity of ER, differing from relatively longer actin filaments, i.e. stress fibers or cortical actin, which had no clear interactions, nor simultaneous movement, with the ER. Our analysis of actin array dynamics revealed that they are connected to sheet movements and transformations: The relocation, disappearance, or formation of the actin arrays frequently preceded ER polygon closure or opening, respectively, leading to subsequent sheet transformation, i.e. sheets fusing to form larger sheets or splitting into smaller sheets. We also observed that the appearance of a new polygon within sheets followed closely the formation of an actin array. Sheet transformations could be induced by mimicking the natural loss of actin arrays from the vicinity of sheets latrunculin A treatment, showing a three-fold decrease in the sheet persistence. These effects could not be observed after MT depolymerization [5]. Based on center-of-mass analysis, actin depolymerization-induced sheet transformations were accompanied by increased lateral movement of sheets, i.e. sheets moved in a larger area, while the sheet velocity remained roughly the same (0.11 μm/ s70.04 SD) as in control. Acetylated MTs have been shown to act as tracks for ER tubule sliding [10], but the hyperacetylation of MTs with trichostatin A did not induce changes in the lateral movement of sheets and, moreover, actin depolymerization did not induce changes in the tubular dynamics. While in HeLa and NRK-52E the sheet persistence is lower than in Huh-7, the sheets correspond to actin depolymerization similarly to Huh-7 sheets (Our unpublished data). Together our results provide a mechanistic view on interplay between ER and actin, suggesting that the loss of actin filament arrays from the polygons allow sheets to move in a larger area and that the lateral movement is subsequently accompanied by increased sheet transformations. Our results indicate that the dynamics of actin arrays and the sheet transformations are interdependent and are in agreement with previous reports demonstrating that actin cytoskeleton interacts
Please cite this article as: M. Joensuu, E. Jokitalo, ER sheet–tubule balance is regulated by an array of actin filaments and microtubules, Exp Cell Res (2015), http://dx.doi.org/10.1016/j.yexcr.2015.04.009
4
E XP E RI ME N TAL CE L L R ES E ARC H
with sheets [26,27,35–37], while MTs have a role on the dynamics of ER tubules [7–9,14–18]. Furthermore, our results are consistent with previous studies showing that polygon filling within an sheet leads to
] (]]]]) ]]]–]]]
larger sheets [20] and that, at steady state, the ER network is balanced by polygon closure [7], while the mechanism behind these transformation has remained unclear.
E X PE R IM EN TA L C ELL R E S EA RC H
Proteins involved in ER–actin cytoskeleton interaction While a growing body of evidence indicate that ER interacts with actin cytoskeleton in wide variety of cell types, the reports on ER– actin interactions in mammalian cells have been mainly descriptive, and the mechanisms by which actin contributes to ER dynamics and/or morphology has remained unclear. The first protein identified in the ER–actin interphase was the inverted formin 2 (INF2) which is peripherally bound to the cytoplasmic face of the ER and possesses a unique ability to nucleate and depolymerize actin filaments [47]. Dominant-negative expression of mutated INF2 with abolished depolymerization activity caused ER to collapse around the nucleus, with accumulation of actin filaments around the collapsed ER. While the authors claim that INF2 knockdown does not cause apparent changes in the ER morphology, the study was done at in LM level and the ER ultrastructure was not studied. Interestingly, based on nerve biopsy, a recent study describes an ER expansion in axons and Schwann cells of intermediate type Charcot-Marie-Tooth neuropathy patients with INF2-mutation [48]. We recently described a novel role for unconventional motor protein myosin 1c (myo1c) in the interphase of ER and actin arrays. Myo1c was identified based on its depletion effects on the actin arrays and ER and from a screen of 4200 known human actin-binding proteins. Myo1c depletion or dominant-negative expression of mutated myo1c with abolished actin-binding (EGFP-myo1cΔABL) resulted in collapse of the short dynamic actin arrays to punctate foci while the overexpression of wt myo1c caused an induction of curved, bundled and unorganized actin arrays, indicating that the functional actin-binding domain of myo1c is required for creation and/or maintenance of ERassociated dynamic actin arrays. The localization of wt myo1c to the ER-associated dynamic actin arrays and under plasma membrane was shown to be dependent on the functional actin binding domain. Furthermore, the myo1c was observed to move
] (]]]]) ]]]–]]]
5
in-conjunction with actin arrays [5]. In agreement, myo1c and its rat ortholog, myr2, have been suggested to link actin cytoskeleton to cellular membranes and localize to actin-rich sites at the cell periphery and to discrete punctate within the cytoplasm [49–54] and myo1c has previously been suggested to maintain and organize actin filaments [55–61]. Providing further support to our results, myosin I has been shown to bind only dynamic actin filament population and not tropomyosin-stabilized microfilaments [62]. Another unconventional motor protein, myosin Va, has been shown to transport ER from the dendritic shaft to dendritic spine in Purkinje neurons, independent of MTs. The absence of ER in the dendritic spines of myosin Va knockout neurons can rescued by re-introducing wild-type protein, but not by its mutant [63–65]. In addition, myosins have been shown to be capable of shortrange ER mobility in vitro [66] and, in S. cerevisiae, the generation of ER tubules have been shown to be dependent on myosin 4p [67]. However, actin depolymerization does not collapse ER network in yeast [29], and even in mammalian cells ER retracts only with some delay upon MT depolymerization [9]. Thus, additional mechanisms, other than those based on MT and/or actin cytoskeleton, are likely to exist.
Myo1c regulation of actin arrays is conveyed to sheet morphology Based on our studies, the regulatory effects of myo1c on actin arrays were conveyed to changes in the sheet morphology, providing further evidence on the coupling of ER and actin cytoskeleton. We showed that the depletion of myo1c, and the subsequent loss of actin arrays, led to uncharacteristic ER network appearance and loss of sheets, similarly to latrunculin A-treated cells. Based on confocal live-cell imaging the loss of sheets could be rescued by re-introduction of wt myo1c, but not by EGFP– myo1cΔABL. The depletion-results were further supported by 3Dmodels obtained from SB-EM datasets where the uneven dis-
Fig. 1 – (A) 3D-model of wt Huh-7 shows that ER (yellow) resides in close proximity with actin filaments (red) and microtubules (light blue). Two successive 250-nm sections were prepared using HPF/FS, and subjected to ET and modeling. A few sheet fenestrations are marked with arrowheads for identification. Bar 0.5 lm. (B) SB-EM model of Huh-7/ssHRP-KDEL shows actin filaments (red) passing through polygons defined by the surrounding ER (yellow). Sheet fenestrations, considerably smaller in size than polygons, are indicated with arrowheads. Image is presented in orthogonal view and the 0.5 lm bar applies to center of the image. (C) MT passing through ER sheet fenestration. (a) A slice image of the tomogram shows a fenestration (arrowhead) in a transverse section of a sheet of Huh-7/ ssHRP-KDEL. (c–d) Double magnifications of the boxed area show slice images depicting a MT (open arrowheads) extending through the same fenestration (arrowhead). (e) A 3D-model of the boxed area of ER (yellow) shows the same fenestration (arrowhead) through which the MT (light blue) passes. Bar is 1 lm in (a) and 0.5 lm in (e). (D) MT stabilization and bundling induced a striking ER phenotype. (a) ER in the central part of the control Huh-7/ssHRP-KDEL cell is composed of large fenestrated (arrowheads) sheets with a complex 3Dstructure. (b) Wide-field image of the effect of paclitaxel treatment in live Huh-7/Hsp47-GFP. Long ER profiles (open arrowheads) are mainly observed at central parts of the cell. (c) Electron micrograph of paclitaxel-treated wt Huh-7 shows induction of long, straight and non-gapped ER profiles (arrows) that are closely associated with MT bundles (arrowheads) and directly continuous with fenestrated ER profiles (*). (d) SB-EM 3D-model of paclitaxel-treated Huh-7/ssHRP-KDEL shows planar and long ER (yellow) sheets lacking fenestrations from regions in close proximity of MTs (open arrow heads), but which are directly connected to the typical fenestrated (arrowheads) sheets. Bars are 10 lm (b), 0.5 lm in (c) and 1.0 lm in (a and d). (E) Actin stabilization and bundling affects the ER network distribution and sheet morphology. (a) A top view, (c) side-view and (b) a representative sheet of SB-EM 3D-model of ER network (yellow) and NE (blue) after jasplakinolide treatment in Huh-7/ssHRP-KDEL are shown. (d–e) Two block-face images taken at indicated heights from the beginning of the dataset are shown, and the NE and the boxed area subjected to 3D-modeling are indicated. (f) NE and areas void of ER (asterisks) are indicated in a wide-field image of wt Huh-7, labeled against endogenous calreticulin, after jasplakinolide treatment. Bar is 10 lm in (f) and 1 lm in (a–e). All cells expressing ssHRP-KDEL were cytochemically stained. (B, C, Da, Dc, Dd, E) were chemically fixed. Please cite this article as: M. Joensuu, E. Jokitalo, ER sheet–tubule balance is regulated by an array of actin filaments and microtubules, Exp Cell Res (2015), http://dx.doi.org/10.1016/j.yexcr.2015.04.009
6
E XP E RI ME N TAL CE L L R ES E ARC H
] (]]]]) ]]]–]]]
Fig. 2 – Dynamic microtubules and actin filament arrays counterbalance endoplasmic reticulum sheet–tubule balance. Perturbation of dynamic MTs or actin cytoskeleton affects the ER sheet–tubule balance. Actin depolymerization (latrunculin A) or stabilization (jasplakinolide) shift the ER balance towards tubules and lead to sheet remnants, whereas MT depolymerization (nocodazole) shifts the balance towards sheets and results in an induction of sheets and the MT stabilization and bundling (paclitaxel) results in long and less fenestrated straight sheets. Myo1c manipulations (myo1c depletion or dominant-negative expression of EGFP-myo1cΔABL with mutated actin-binding domain) affect the structure of actin arrays and the effect is conveyed to ER sheet morphology similarly to actin depolymerization effects. Tubular ER phenotype of myo1c-depleted Huh-7 cells can be rescued with re-introduction of myo1c, but not by its mutated form (EGFP–myo1cΔABL). LM images present chemically fixed wt Huh-7 cells, immunolabeled for endogenous calreticulin. SB-EM models of representative after indicated treatments are shown in yellow in the top row and are derived from chemically fixed and cytochemically stained Huh-7/ssHRP-KDEL.
tribution and sheet remnants were even more evident. The overexpression of myo1c shifted the sheet–tubule ratio toward sheets, with few tubules remaining and, instead of individual sheets, an interlinked sheet mass was observed. In contrast, dominant-negative expression of EGFP–myo1cΔABL led to almost complete structural shift of ER toward a reticular network, as only few peripheral sheets were observed [5]. Immunoreactivity studies of myr2 from isolated sucrose gradient membrane fractions have showed that Myr2 associates with fractions enriched with ER and PM, but not with other organelles [50]. Myo1c has been shown to bind specifically and with high affinity to phosphatidylinositol 4,5-bisphosphate (PIP2) via its pleckstrin homology (PH) domain, and to lesser extent to other lipids [52,68], and based on live cell imaging experiments, myo1c localization in the distinct punctate in cells is dependent on its binding to PIP2 [52,69]. In accordance, Myr2 has been shown to localize to lipid rafts enriched with PIP2 [70]. It has been suggested that the myo1c–PIP2 interaction may serve to concentrate myo1c at regions of new actin polymerization, providing mechanism for recruitment of the motor to areas driving membrane–actin interactions [55,57]. As PIP2 has been shown to be enriched at ER [71,72], it could be envisioned that PIP2 recruits myo1c to ER. This interaction, maybe together with other factors, might facilitate the recruitment of dynamic actin to ER to drive sheet transformations and to facilitate the sheet persistence. In agreement with this proposal, myo1 and actin cytoskeleton have been shown to provide mechanical stability to microvillar ER [73].
The actual mechanism for ER–actin interplay, however, remains open; while myo1c might serve to localize actin filaments to the vicinity of ER, these actin arrays might either form passive physical restriction barriers at polygons or yet unidentified factors could be needed to mediate the interaction between ER and actin filaments.
Dynamic microtubules and actin filament arrays counterbalance sheet–tubule balance It has been suggested that the ER structure at cell periphery is determined by a balance of MT and actin and their motor proteins [5,21]. Based on our results, actin cytoskeleton and MT manipulations have opposing effects on ER morphology, indicating that there is a tug-of-war between dynamic actin filaments and MTs on sheet–tubule ratio (Fig. 2). MT manipulations shifts the morphological balance towards sheets and result in stacked sheets (upon nocodazole treatment) or straight, extremely long and less fenestrated sheets (upon MT stabilization by paclitaxel [41]) compared to control sheets. Actin depolymerization or by latrunculin A or jasplakinolide treatment, respectively, shifted the balance towards tubular morphology and resulted in sheet remnants. These drug treatments had opposing effects on MTs and actin -either disassembling or stabilizing the filaments. Why, then, the effects of the treatments caused similar effects on the sheet–tubule ratio? One explanation could be that the regulation
Please cite this article as: M. Joensuu, E. Jokitalo, ER sheet–tubule balance is regulated by an array of actin filaments and microtubules, Exp Cell Res (2015), http://dx.doi.org/10.1016/j.yexcr.2015.04.009
E X PE R IM EN TA L C ELL R E S EA RC H
of sheet–tubule balance requires dynamic cytoskeletal filaments. In accordance, our results showed that myo1c regulates the dynamic actin filament arrays, which in-turn, regulate the sheet persistence. Myo1c depletion or dominant-negative expression of EGFP–myo1cΔABL, shifted the sheet–tubule balance towards tubular morphology and resulted in sheet remnants. Myo1c depletion phenotype could be rescued by over-expressing the wt myo1c, which shifted the sheet–tubule balance towards sheets-resembling the over-expression of wt myo1c. As the cytoplasm is full of MT- and actin-binding proteins that provide identity to cytoskeletal filaments in response to various intra- and extracellular cues, the interplay between ER subdomains and different cytoskeletal elements provide a variety of regulatory possibilities. In addition to integral membrane proteins that support specific structural features, interaction with the cytoskeleton provide a more global and dynamic mechanism to control the ER architecture in cells. MT-driven tubular ER is suited to host dynamic functions and provide a mechanism to rearrange the overall network distribution within the cell. In contrast, more stationary and persistent sheets might be more suitable for protein synthesis and quality control or other ER functions. The functional importance of ER network transformations is an important research question for the future.
Acknowledgments The authors gratefully acknowledge funding from The Academy of Finland (Project no. 131650) and the Biocenter Finland. Drs. Helena Vihinen, Maija Puhka and Ilya Belevich (Institute of Biotechnology, Electron Microscopy Unit, University of Helsinki) are acknowledged for their contributions especially on 3D-EM and HPF.
references [1] M. Puhka, M. Joensuu, H. Vihinen, I. Belevich, E. Jokitalo, Progressive sheet-to-tubule transformation is a general mechanism for endoplasmic reticulum partitioning in dividing mammalian cells, Mol. Biol. Cell 23 (2012) 2424–2432, http://dx. doi.org/10.1091/mbc.E10-12-0950. [2] M. West, N. Zurek, A. Hoenger, G.K. Voeltz, A 3D analysis of yeast ER structure reveals how ER domains are organized by membrane curvature, J. Cell Biol. 193 (2011) 333–346, http://dx.doi. org/10.1083/jcb.201011039. [3] M. Puhka, H. Vihinen, M. Joensuu, E. Jokitalo, Endoplasmic reticulum remains continuous and undergoes sheet-to-tubule transformation during cell division in mammalian cells, J. Cell Biol. 179 (2007) 895–909, http://dx.doi.org/10.1083/ jcb.200705112. [4] T. Shemesh, et al., A model for the generation and interconversion of ER morphologies, Proc. Natl. Acad. Sci. USA 111 (2014) E5243–E5251, http://dx.doi.org/10.1073/pnas.1419997111. [5] M. Joensuu, et al., ER sheet persistence is coupled to myosin 1cregulated dynamic actin filament arrays, Mol. Biol. Cell 25 (2014) 1111–1126, http://dx.doi.org/10.1091/mbc.E13-12-0712. [6] L. Lu, M.S. Ladinsky, T. Kirchhausen, Cisternal organization of the endoplasmic reticulum during mitosis, Mol. Biol. Cell 20 (2009) 3471–3480, http://dx.doi.org/10.1091/mbc.E09-04-0327. [7] C. Lee, L.B. Chen, Dynamic behavior of endoplasmic reticulum in living cells, Cell 54 (1988) 37–46. [8] C.M. Waterman-Storer, E.D. Salmon, Endoplasmic reticulum membrane tubules are distributed by microtubules in living cells using three distinct mechanisms, Curr. Biol. 8 (1998) 798–806.
] (]]]]) ]]]–]]]
7
[9] M. Terasaki, L.B. Chen, K. Fujiwara, Microtubules and the endoplasmic reticulum are highly interdependent structures, J. Cell Biol. 103 (1986) 1557–1568. [10] J.R. Friedman, B.M. Webster, D.N. Mastronarde, K.J. Verhey, G.K. Voeltz, ER sliding dynamics and ER-mitochondrial contacts occur on acetylated microtubules, J. Cell Biol. 190 (2010) 363–375, http: //dx.doi.org/10.1083/jcb.200911024. [11] I. Sparkes, J. Runions, C. Hawes, L. Griffing, Movement and remodeling of the endoplasmic reticulum in nondividing cells of tobacco leaves, Plant Cell 21 (2009) 3937–3949, http://dx.doi.org/ 10.1105/tpc.109.072249. [12] W. Knebel, H. Quader, E. Schnepf, Mobile and immobile endoplasmic reticulum in onion bulb epidermis cells: short- and longterm observations with a confocal laser scanning microscope, Eur. J. Cell Biol. 52 (1990) 328–340. [13] G.K. Voeltz, W.A. Prinz, Y. Shibata, J.M. Rist, T.A. Rapoport, A class of membrane proteins shaping the tubular endoplasmic reticulum, Cell 124 (2006) 573–586, http://dx.doi.org/10.1016/j. cell.2005.11.047. [14] H. Bannai, T. Inoue, T. Nakayama, M. Hattori, K. Mikoshiba, Kinesin dependent, rapid, bi-directional transport of ER subcompartment in dendrites of hippocampal neurons, J. Cell Sci. 117 (2004) 163–175, http://dx.doi.org/10.1242/jcs.00854. [15] I. Grigoriev, et al., STIM1 is a MT-plus-end-tracking protein involved in remodeling of the ER, Curr. Biol. 18 (2008) 177–182, http://dx.doi.org/10.1016/j.cub.2007.12.050. [16] M.J. Wozniak, et al., Role of kinesin-1 and cytoplasmic dynein in endoplasmic reticulum movement in VERO cells, J. Cell Sci. 122 (2009) 1979–1989, http://dx.doi.org/10.1242/jcs.041962. [17] J. Kumar, H. Yu, M.P. Sheetz, Kinectin, an essential anchor for kinesin-driven vesicle motility, Science 267 (1995) 1834–1837. [18] A.V. Nikonov, H.P. Hauri, B. Lauring, G. Kreibich, Climp-63mediated binding of microtubules to the ER affects the lateral mobility of translocon complexes, J. Cell Sci. 120 (2007) 2248– 2258, http://dx.doi.org/10.1242/jcs.008979. [19] H. Quader, A. Hofmann, E. Schnepf, Reorganization of the endoplasmic reticulum in epidermal cells of onion bulb scales after cold stress: involvement of cytoskeletal elements, Planta 177 (1989) 273–280, http://dx.doi.org/10.1007/BF00392816. [20] L.R. Griffing, Networking in the endoplasmic reticulum, Biochem. Soc. Trans. 38 (2010) 747–753, http://dx.doi.org/10.1042/ BST0380747. [21] M. Terasaki, T.S. Reese, Interactions among endoplasmic reticulum, microtubules, and retrograde movements of the cell surface, Cell Motil. Cytoskelet. 29 (1994) 291–300, http://dx.doi.org/ 10.1002/cm.970290402. [22] L. Dreier, T.A. Rapoport, In vitro formation of the endoplasmic reticulum occurs independently of microtubules by a controlled fusion reaction, J. Cell Biol. 148 (2000) 883–898. [23] J. Hu, et al., A class of dynamin-like GTPases involved in the generation of the tubular ER network, Cell 138 (2009) 549–561, http://dx.doi.org/10.1016/j.cell.2009.05.025. [24] C. Lavoie, J. Lanoix, F.W. Kan, J. Paiement, Cell-free assembly of rough and smooth endoplasmic reticulum, J. Cell Sci. 109 (Pt 6) (1996) 1415–1425. [25] D.R. Klopfenstein, F. Kappeler, H.P. Hauri, A novel direct interaction of endoplasmic reticulum with microtubules, EMBO J. 17 (1998) 6168–6177, http://dx.doi.org/10.1093/emboj/17.21.6168. [26] J.M. Sanger, J.S. Dome, B. Mittal, A.V. Somlyo, J.W. Sanger, Dynamics of the endoplasmic reticulum in living non-muscle and muscle cells, Cell Motil. Cytoskelet. 13 (1989) 301–319, http: //dx.doi.org/10.1002/cm.970130408. [27] K. Sturmer, O. Baumann, B. Walz, Actin-dependent light-induced translocation of mitochondria and ER cisternae in the photoreceptor cells of the locust Schistocerca gregaria, J. Cell Sci. 108 (Pt 6) (1995) 2273–2283. [28] D. Pruyne, A. Legesse-Miller, L. Gao, Y. Dong, A. Bretscher, Mechanisms of polarized growth and organelle segregation in
Please cite this article as: M. Joensuu, E. Jokitalo, ER sheet–tubule balance is regulated by an array of actin filaments and microtubules, Exp Cell Res (2015), http://dx.doi.org/10.1016/j.yexcr.2015.04.009
8
E XP E RI ME N TAL CE L L R ES E ARC H
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40] [41]
[42]
[43]
[44]
[45]
[46]
yeast, Annu. Rev. Cell Dev. Biol. 20 (2004) 559–591, http://dx.doi. org/10.1146/annurev.cellbio.20.010403.103108. W.A. Prinz, et al., Mutants affecting the structure of the cortical endoplasmic reticulum in Saccharomyces cerevisiae, J. Cell Biol. 150 (2000) 461–474. K.L. Fehrenbacher, D. Davis, M. Wu, I. Boldogh, L.A. Pon, Endoplasmic reticulum dynamics, inheritance, and cytoskeletal interactions in budding yeast, Mol. Biol. Cell 13 (2002) 854–865, http://dx.doi.org/10.1091/mbc.01-04-0184. S.L. Gupton, D.A. Collings, N.S. Allen, Endoplasmic reticulum targeted GFP reveals ER organization in tobacco NT-1 cells during cell division, Plant Physiol. Biochem.: Ppb/Soc. Francaise Physiol. Veg. 44 (2006) 95–105, http://dx.doi.org/10.1016/j. plaphy.2006.03.003. H. Ueda, et al., Myosin-dependent endoplasmic reticulum motility and F-actin organization in plant cells, Proc. Natl. Acad. Sci. USA 107 (2010) 6894–6899, http://dx.doi.org/10.1073/ pnas.0911482107. P. Boevink, et al., Stacks on tracks: the plant Golgi apparatus traffics on an actin/ER network, Plant J.: Cell Mol. Biol. 15 (1998) 441–447. F. Lv, et al., GhCFE1A, a dynamic linker between the ER network and actin cytoskeleton, plays an important role in cotton fibre cell initiation and elongation, J. Exp. Bot. 66 (2015) 1877–1889, http://dx.doi.org/10.1093/jxb/eru530. D. Poteryaev, J.M. Squirrell, J.M. Campbell, J.G. White, A. Spang, Involvement of the actin cytoskeleton and homotypic membrane fusion in ER dynamics in Caenorhabditis elegans, Mol. Biol. Cell 16 (2005) 2139–2153, http://dx.doi.org/10.1091/mbc.E04-080726. O. Baumann, B. Lautenschlager, The role of actin filaments in the organization of the endoplasmic reticulum in honeybee photoreceptor cells, Cell Tissue Res. 278 (1994) 419–432. S.A. Kuznetsov, D.T. Rivera, F.F. Severin, D.G. Weiss, G.M. Langford, Movement of axoplasmic organelles on actin filaments from skeletal muscle, Cell Motil. Cytoskelet. 28 (1994) 231–242, http: //dx.doi.org/10.1002/cm.970280306. C.D. Lynch, et al., Filamin depletion blocks endoplasmic spreading and destabilizes force-bearing adhesions, Mol. Biol. Cell 22 (2011) 1263–1273, http://dx.doi.org/10.1091/mbc.E10-08-0661. Y.J. Wang, R.B. Gregory, G.J. Barritt, Maintenance of the filamentous actin cytoskeleton is necessary for the activation of storeoperated Ca2þ channels, but not other types of plasmamembrane Ca2þ channels, in rat hepatocytes, Biochem. J. 363 (2002) 117–126. C. Lee, M. Ferguson, L.B. Chen, Construction of the endoplasmic reticulum, J. Cell Biol. 109 (1989) 2045–2055. A.M. Yvon, P. Wadsworth, M.A. Jordan, Taxol suppresses dynamics of individual microtubules in living human tumor cells, Mol. Biol. Cell 10 (1999) 947–959. A. Holzinger, Jasplakinolide: an actin-specific reagent that promotes actin polymerization, Methods Mol. Biol. 586 (2009) 71– 87, http://dx.doi.org/10.1007/978-1-60761-376-3_4. O. Baumann, Structural Interactions of Actin-Filaments and Endoplasmic-Reticulum in Honeybee Photoreceptor Cells, Cell Tissue Res. 268 (1992) 71–79, http://dx.doi.org/10.1007/ Bf00338055. D.R. Klopfenstein, et al., Subdomain-specific localization of CLIMP-63 (p63) in the endoplasmic reticulum is mediated by its luminal alpha-helical segment, J. Cell Biol. 153 (2001) 1287– 1300. C. Vedrenne, D.R. Klopfenstein, H.P. Hauri, Phosphorylation controls CLIMP-63-mediated anchoring of the endoplasmic reticulum to microtubules, Mol. Biol. Cell 16 (2005) 1928–1937, http: //dx.doi.org/10.1091/mbc.E04-07-0554. N. Santama, C.P. Er, L.L. Ong, H. Yu, Distribution and functions of kinectin isoforms, J. Cell Sci. 117 (2004) 4537–4549, http://dx.doi. org/10.1242/jcs.01326.
] (]]]]) ]]]–]]]
[47] E.S. Chhabra, V. Ramabhadran, S.A. Gerber, H.N. Higgs, INF2 is an endoplasmic reticulum-associated formin protein, J. Cell Sci. 122 (2009) 1430–1440, http://dx.doi.org/10.1242/jcs.040691. [48] A. Roos, et al., Inverted formin 2-related Charcot-Marie-Tooth disease: extension of the mutational spectrum and pathological findings in Schwann cells and axons, J. Peripher. Nerv. Syst. . (2015)http://dx.doi.org/10.1111/jns.12106. [49] M.C. Wagner, B. Barylko, J.P. Albanesi, Tissue distribution and subcellular localization of mammalian myosin I, J. Cell Biol. 119 (1992) 163–170. [50] C. Ruppert, et al., Localization of the rat myosin I molecules myr 1 and myr 2 and in vivo targeting of their tail domains, J. Cell Sci. 108 (Pt 12) (1995) 3775–3786. [51] K.D. Novak, M.D. Peterson, M.C. Reedy, M.A. Titus, Dictyostelium myosin I double mutants exhibit conditional defects in pinocytosis, J. Cell Biol. 131 (1995) 1205–1221. [52] D.E. Hokanson, J.M. Laakso, T. Lin, D. Sept, E.M. Ostap, Myo1c binds phosphoinositides through a putative pleckstrin homology domain, Mol. Biol. Cell 17 (2006) 4856–4865, http://dx.doi.org/ 10.1091/mbc.E06-05-0449. [53] M.C. Wagner, B.A. Molitoris, ATP depletion alters myosin I beta cellular location in LLC-PK1 cells, Am. J. Physiol. 272 (1997) C1680–1690. [54] O. Reizes, B. Barylko, C. Li, T.C. Sudhof, J.P. Albanesi, Domain structure of a mammalian myosin I beta, Proc. Natl. Acad. Sci. USA 91 (1994) 6349–6353. [55] A. Bose, et al., Unconventional myosin Myo1c promotes membrane fusion in a regulated exocytic pathway, Mol. Cell. Biol. 24 (2004) 5447–5458, http://dx.doi.org/10.1128/MCB.24.12.54475458.2004. [56] G.N. Hagan, Y. Lin, M.A. Magnuson, J. Avruch, M.P. Czech, A Rictor-Myo1c complex participates in dynamic cortical actin events in 3T3-L1 adipocytes, Mol. Cell. Biol. 28 (2008) 4215– 4226, http://dx.doi.org/10.1128/MCB.00867-07. [57] T.J. Diefenbach, et al., Myosin 1c and myosin IIB serve opposing roles in lamellipodial dynamics of the neuronal growth cone, J. Cell Biol. 158 (2002) 1207–1217. [58] Y. Fan, S.M. Eswarappa, M. Hitomi, P.L. Fox, Myo1c facilitates G-actin transport to the leading edge of migrating endothelial cells, J. Cell Biol. 198 (2012) 47–55, http://dx.doi.org/10.1083/ jcb.201111088. [59] E. Arif, et al., Motor protein Myo1c is a podocyte protein that facilitates the transport of slit diaphragm protein Neph1 to the podocyte membrane, Mol. Cell. Biol. 31 (2011) 2134–2150, http: //dx.doi.org/10.1128/MCB.05051-11. [60] P. Garg, R. Verma, D. Nihalani, D.B. Johnstone, L.B. Holzman, Neph1 cooperates with nephrin to transduce a signal that induces actin polymerization, Mol. Cell. Biol. 27 (2007) 8698– 8712, http://dx.doi.org/10.1128/MCB.00948-07. [61] J.L. Maravillas-Montero, P.G. Gillespie, G. Patino-Lopez, S. Shaw, L. Santos-Argumedo, Myosin 1c participates in B cell cytoskeleton rearrangements, is recruited to the immunologic synapse, and contributes to antigen presentation, J. Immunol. 187 (2011) 3053–3063, http://dx.doi.org/10.4049/jimmunol.1004018. [62] N. Tang, E.M. Ostap, Motor domain-dependent localization of myo1b (myr-1), Curr. Biol. 11 (2001) 1131–1135. [63] Y. Takagishi, et al., The dilute-lethal (dl) gene attacks a Ca2þ store in the dendritic spine of Purkinje cells in mice, Neurosci. Lett. 215 (1996) 169–172. [64] P.C. Bridgman, Myosin Va movements in normal and dilutelethal axons provide support for a dual filament motor complex, J. Cell Biol. 146 (1999) 1045–1060. [65] W. Wagner, S.D. Brenowitz, J.A. Hammer 3rd, Myosin-Va transports the endoplasmic reticulum into the dendritic spines of Purkinje neurons, Nat. Cell Biol. 13 (2011) 40–48, http://dx.doi. org/10.1038/ncb2132. [66] T. Wollert, D.G. Weiss, H.H. Gerdes, S.A. Kuznetsov, Activation of myosin V-based motility and F-actin-dependent network
Please cite this article as: M. Joensuu, E. Jokitalo, ER sheet–tubule balance is regulated by an array of actin filaments and microtubules, Exp Cell Res (2015), http://dx.doi.org/10.1016/j.yexcr.2015.04.009
E X PE R IM EN TA L C ELL R E S EA RC H
formation of endoplasmic reticulum during mitosis, J. Cell Biol. 159 (2002) 571–577, http://dx.doi.org/10.1083/jcb.200204065. [67] P. Estrada, et al., Myo4p and She3p are required for cortical ER inheritance in Saccharomyces cerevisiae, J. Cell Biol. 163 (2003) 1255–1266, http://dx.doi.org/10.1083/jcb.200304030. [68] M. Hirono, C.S. Denis, G.P. Richardson, P.G. Gillespie, Hair cells require phosphatidylinositol 4,5-bisphosphate for mechanical transduction and adaptation, Neuron 44 (2004) 309–320, http: //dx.doi.org/10.1016/j.neuron.2004.09.020. [69] D.E. Hokanson, E.M. Ostap, Myo1c binds tightly and specifically to phosphatidylinositol 4,5-bisphosphate and inositol 1,4,5-trisphosphate, Proc. Natl. Acad. Sci. USA 103 (2006) 3118–3123, http://dx.doi.org/10.1073/pnas.0505685103.
] (]]]]) ]]]–]]]
9
[70] B. Barylko, G. Jung, J.P. Albanesi, Structure, function, and regulation of myosin 1C, Acta Biochim. Pol. 52 (2005) 373–380. [71] G. Griffiths, et al., The dynamic nature of the Golgi complex, J. Cell Biol. 108 (1989) 277–297. [72] F.Zambrano, S. Fleischer, B. Fleischer, Lipid composition of the Golgi apparatus of rat kidney and liver in comparison with other subcellular organelles, Biochim. Biophys. Acta 380 (1975) 357–369. [73] T.D. Pollard, S.K. Doberstein, H.G. Zot, Myosin-I, Annu. Rev. Physiol. 53 (1991) 653–681, http://dx.doi.org/10.1146/annurev. ph.53.030191.003253.
Please cite this article as: M. Joensuu, E. Jokitalo, ER sheet–tubule balance is regulated by an array of actin filaments and microtubules, Exp Cell Res (2015), http://dx.doi.org/10.1016/j.yexcr.2015.04.009
