Perico, C., & Sparkes, I. (2018). Plant organelle dynamics: cytoskeletalcontrol and membrane contact sites. New Phytologist, 220(2), 381-394.https://doi.org/10.1111/nph.15365

Peer reviewed version

Link to published version (if available):10.1111/nph.15365

Link to publication record in Explore Bristol ResearchPDF-document

This is the author accepted manuscript (AAM). The final published version (version of record) is available onlinevia Wiley at https://nph.onlinelibrary.wiley.com/doi/full/10.1111/nph.15365 . Please refer to any applicable termsof use of the publisher.

University of Bristol - Explore Bristol ResearchGeneral rights

This document is made available in accordance with publisher policies. Please cite only the publishedversion using the reference above. Full terms of use are available: http://www.bristol.ac.uk/pure/user-guides/explore-bristol-research/ebr-terms/

For Peer Review

Plant organelle dynamics: cytoskeletal control and

membrane contact sites

Journal: New Phytologist

Manuscript ID NPH-TR-2018-26701.R1

Manuscript Type: TR - Commissioned Material - Tansley Review

Date Submitted by the Author: n/a

Complete List of Authors: Perico, Chiara; University of Bristol, School of Biological Sciences Sparkes, Imogen; University of Bristol, School of Biological Sciences

Key Words: actin, dynamics, membrane contact site, myosin, organelle, tether, cytoskeleton

Manuscript submitted to New Phytologist for review

For Peer Review

1

Plant organelle dynamics: cytoskeletal control and membrane contact sites 1

2

3

Chiara Perico1, Imogen Sparkes

1* 4

1 School of Biological Sciences, University of Bristol, 24 Tyndall Avenue, Bristol, BS8 1TQ, UK 5

6

*Corresponding author 7

i.sparkes@bristol.ac.uk 8

+44 (0)117 39 41211 9

10

Word count: 7599 (excluding summary, reference list and figure legends) 11

4 figures: Fig1, 2 and 4 in colour, Fig 3 in black and white 12

13

Summary 14

15

Organelle movement and positioning are correlated with plant growth and development. Movement 16

characteristics are seemingly erratic yet respond to external stimuli including pathogens and light. 17

Given these clear correlations, we still do not understand the specific roles that movement plays in 18

these processes. There are few exceptions including organelle inheritance during cell division, and 19

photorelocation of chloroplasts to prevent photodamage. The molecular and biophysical 20

components that drive movement can be broken down into cytoskeletal components, motor 21

proteins and tethers which allow organelles to physically interact with one another. Our 22

understanding of these components and concepts has exploded over the past decade, with recent 23

technological advances allowing an even more in depth profiling. Here, we provide an overview of 24

the cytoskeletal and tethering components and discuss the mechanisms behind organelle movement 25

in higher plants. 26

27

Keywords: Actin, dynamics, cytoskeleton, membrane contact sites, myosin, organelle, tether 28

29

30

31

Page 1 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

2

Introduction 32

33

A defining characteristic of eukaryotic life is subcellular compartmentalisation in the form of 34

membrane bounded organelles. Organelles allow spatial separation of components required to carry 35

out key biochemical reactions in distinct microenvironments. However, biochemical pathways can 36

span multiple organelles requiring exchange of key metabolites. How can exchange occur if 37

organelles are spatially separated within the cytoplasmic milieu? How can an organelle deliver 38

components to targeted sites? Ultimately, how do organelles ‘communicate’ with one another, and 39

what controls the choreography of such events? 40

Since the early observations of cytoplasmic streaming events, it is clear that organelles whilst being 41

discrete entities are not static, they are highly motile. A basic analogy can be drawn between a cell 42

and an island, where organelles are cars with cytoskeletal components acting as roads. For the island 43

(cell) to function there must be clear communication allowing delivery and transfer of goods in a 44

coordinated manner. The emerging picture is that in eukaryotic systems organelle movement is 45

dynamic with physical interactions occurring between organelles allowing ‘communication’ and 46

exchange to occur. These exchanges occur at membrane contact sites (MCS). Exchange between 47

organelles also occurs via vesicle mediated pathways (Bonifacino & Glick, 2004). Ultimately, 48

organelle movement is therefore driven by an interplay between physical interaction between 49

organelles and motor driven activity to allow separation and independent motion. 50

The combination of technological advances in imaging, genetic manipulation of model systems and 51

introduction of a multitude of fluorophores has enabled a relatively fast description of basic 52

parameters relating to organelle dynamics and interactions. Here, we introduce the elements that 53

control organelle dynamics in higher plant cells, and why it is important that we unpick the 54

molecular mechanisms controlling movement and positioning of organelles. 55

56

Basic movement characteristics 57

58

Cytoplasmic streaming is an actin dependent event resulting in the bulk flow of cytoplasm in a 59

directed manner (Shimmen, 2007; Tominaga & Ito, 2015). In cell types undergoing polarised tip 60

growth (root hairs and pollen tubes) streaming events are ordered with flows along the outer wall 61

towards the tip changing direction and flowing down the central axis (Hepler et al., 2001). This 62

process is called reverse fountain streaming. Other cell types however display less ordered ‘patterns’ 63

of movement with streams rotating and bifurcating in a seemingly random manner. Modelling the 64

process of cytoplasmic streaming events indicates that actin dynamics themselves (sliding of actin 65

Page 2 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

3

filaments, polymerisation / depolymerisation) cannot generate the required shear forces in the 66

cytoplasm to generate bulk flow. However, the movement of a net like structure, which effectively 67

has wider ‘filaments’ than actin, being dragged through the cytoplasm could produce bulk flow of 68

the cytoplasm (Nothnagel & Webb., 1982). These studies infer that it is the movement of the ER 69

which generates the observed flows. Based on correlations between ER and actin rearrangements, 70

Ueda et al. postulated that movement of the ER may act to rearrange the actin filaments (Ueda et 71

al., 2010). More recently, data indicated that changes in ER network movement were correlated 72

with changes in streaming rates based on movement of discrete organelles such as Golgi and 73

mitochondria (Stefano et al., 2014). In summary, cytoplasmic streaming is complex and the 74

molecular mechanism is likely driven through an interplay between actin and organelle movement. 75

Note, while studies in Chara have provided advances in our understanding of cytoplasmic streaming 76

(Nothnagel & Webb, 1982; Kachar & Reese, 1988; Woodhouse & Goldstein, 2013; Goldstein et al. 77

2008), these may not directly map on to angiosperms. For example, differences in myosin 78

mechanochemical properties, organelle and actin filament positioning may all impact streaming, and 79

the modelling of such events. 80

81

Initially it was proposed that organelle movement was a passive process with organelles moving 82

along in cytoplasmic streams. With the development of tracking algorithms and implementation of 83

fluorescent probes targeted to organelles, it quickly became clear that individual organelles can 84

display a multitude of movement characteristics within a short time frame, and that movement 85

outside of the streams is directed and not passive. For example, smaller seemingly ‘spheroid’ 86

organelles observed with a light microscope, such as Golgi, mitochondria and peroxisomes undergo 87

‘stop-go’ movement; fast motion with acceleration, deceleration, pausing and changing direction 88

with speeds reported up to 8-10µm/s (Boevink et al., 1998; Nebenführ et al., 1999; Logan & Leaver, 89

2000; Jedd & Chua, 2002; Mano et al., 2002; Mathur et al., 2002; Van Gestel et al., 2002). It has been 90

suggested that sites of organelle pausing occurs at cross-over points between actin and microtubule 91

filaments, which can occur around ER-PM anchor points (see later). The ER is a network of 92

interconnected tubules and cisternae which readily interconvert. Movement here is through tubule 93

extension and retraction, tubule fusion and lateral sliding to form new polygons, with possible 94

polygonal filling to generate new cisternae (Sparkes, I et al., 2009). In addition to movement 95

(remodelling) of the ER network structure, movement of membrane proteins (surface flow) also 96

occurs (Sparkes, I et al., 2009; Runions , J et al., 2006). An example of ER (green) and Golgi 97

(magenta) movement over a 30 second period in Arabidopsis leaf epidermal cells is depicted in 98

Figure 1. Comparing 3 individual frames (0, 15 and 30 seconds) indicates the dramatic level of 99

Page 3 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

4

movement of both the ER network and individual Golgi bodies; merge image (h and l) highlights 100

Golgi and regions of the ER which have remained relatively stationary (white) compared with other 101

regions within the network / discrete Golgi. It is important to note that the vast majority of discrete 102

organelles appear to be in close association with the ER (Fig1a-c). Whether this reflects a true 103

physical connection between organelles, or is an indirect effect due to the large central vacuole 104

effectively forcing cytoplasmic contents together or organelles occupying the same cytoskeletal 105

track is unclear (see later). 106

107

Actin and associated motors, myosins, play a primary role in plant organelle movement and 108

positioning. 109

Organelle re-positioning within the plant cell cytoplasm largely depends on the actin cytoskeleton 110

and on myosins, the associated motor proteins (Boevink et al., 1998; Nebenführ et al., 1999; Mathur 111

et al., 2002; Van Gestel et al., 2002; Sparkes et al., 2008; Avisar, Dror et al., 2009). Myosin motors 112

generally consist of an N-terminal motor domain which binds actin in an ATP dependent manner, a 113

neck domain with calmodulin-binding IQ-repeats and a region of dimerization, and a C-terminal 114

globular tail domain (GTD) involved in cargo binding. 115

Actin monomers (G-actin) polymerise under physiological conditions to form filaments (F-actin) 116

which constitute a dynamic, rail-like structure. It is generally thought that myosin dimers are 117

recruited to organelle membranes and proceed on these tracks in a step-wise ATP-dependent 118

manner towards the faster growing plus end, dragging the organelle along the actin filament. 119

Disruption of actin with depolymerising drugs such as Latrunculin B (LatB) or Cytochalasin D (CytD) 120

halts organelle movement and, more generally, cytoplasmic streaming (Nebenführ et al., 1999; 121

Mathur et al., 2002) . While these studies have indeed highlighted the central role of the actin 122

network in organelle dynamics, it does not indicate regulatory mechanisms to allow specific control 123

of the movement and positioning of one organelle over another. This is provided through myosin 124

interaction, and organelle interactions at membrane contact sites. 125

Based on the structure of the conserved motor domain, two sub-classes of myosin have been 126

identified in the model plant Arabidopsis thaliana: class VIII and class XI, which include 4 and 13 127

members respectively (Foth et al., 2006; Peremyslov et al., 2011). Myosins from class VIII have been 128

observed to localise at the cell plate (Damme et al., 2004), plasmodesmata (Golomb et al., 2008), 129

nucleolus (Avisar, Dror et al., 2009) and in association with phragmoplast microtubules (Wu & 130

Bezanilla, 2014). In higher plants they do not appear to control organelle movement, and, as 131

Page 4 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

5

suggested by Ryan and Nebenführ, may act as tensors interacting with actin to control tension 132

within the network (Ryan & Nebenführ, 2018). Class XI myosins contain a conserved Dilute domain 133

and are similar to class V myosins from mammals and fungi (Foth et al., 2006). Class V and XI 134

myosins are involved in organelle movement. In higher plants, the role of class XI myosins has been 135

determined through knock out, RNAi and expression of dominant negative myosin fusions. The latter 136

is proposed to act in a dominant negative manner as fusions lack the motor domain which could 137

prevent movement, yet retain the cargo binding domain. Transient over-expression of dominant 138

negative tail and IQ-tail of all the 17 A.thaliana myosins in N.tabacum and N.benthamiana leaf 139

epidermal cells highlighted the role of multiple myosins in controlling Golgi, mitochondria and 140

peroxisome dynamics (Sparkes et al., 2008; Avisar, Dror et al., 2009). Tail and IQ-tail of myosins XI-1 141

(MYA1), XI-2 ( MYA2), XI-C, XI-E, XI-I, and XI-K significantly decrease Golgi displacement rate in both 142

Nicotiana species, without affecting actin organisation or colocalising to the Golgi membrane (Avisar, 143

Dror et al., 2009). Interestingly, IQ-tail of the same myosins also inhibit mitochondria movement 144

(Avisar, Dror et al., 2009), whilst myosin XI-E and XI-K tail fusions were observed to stop peroxisome 145

movement (Sparkes et al., 2008), thus suggesting that a single myosin can control the dynamics of 146

multiple organelles. It was reported that Arabidopsis knock-outs (KO) in single myosin XI genes didn’t 147

display a noticeable phenotype, apart from xi-k and xi-2 plants, which display a reduced root hair 148

growth (Peremyslov, Valera V. et al., 2008). Much like what happens upon the expression of XI-K and 149

XI-2 dominant negative mutants, Golgi, peroxisomes and mitochondria movement drastically 150

decrease in leaves of xi-k and xi-2 Arabidopsis lines (Peremyslov, Valera V. et al., 2008). Considering 151

that multiple class XI myosins appear to control global organelle movement, with XI-K and XI-2 152

having the most marked effect, implies a certain degree of functional redundancy within the class XI 153

myosin gene family. 154

Considering that several myosins control global organelle movement, one would hypothesis that 155

they would be located to the surface of the organelle in question. Myosin localisation data through 156

expression of fluorescent fusions to full length and truncated myosins, and immunohistochemistry 157

data have not provided a clear answer to this issue. The vast majority of data for class XI myosins 158

indicate diffuse cytoplasmic localisation, or to unknown puncta / vesicles; IQ-Tail and tail fusions are 159

diffuse or on unknown puncta (Avisar, D. et al., 2009), ‘DIL’ and ‘PAL’ domains are frequently on 160

unknown puncta (Sattarzadeh et al., 2011; Sattarzadeh et al., 2013), XI-K full length fusion is on 161

unknown highly motile vesicles (Peremyslov et al., 2012). There are few exceptions including 162

fluorescent fusions of XI-I residing on the nuclear envelope (Avisar, D. et al., 2009; Tamura et al., 163

2013), expression of portions of the GTD of XI-1 located to either peroxisomes and / or Golgi (Li & 164

Nebenführ, 2007), and immunohistochemistry indicating XI-2 on peroxisomes and additional puncta 165

Page 5 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

6

(Hashimoto et al., 2005). How can we reconcile this disconnect between myosin localisation and 166

activity on organelle movement? There are several hypothesis; (1) the number of functional myosin 167

molecules required to generate movement is small and so expression of truncations may bind and 168

saturate receptors on the organelle surface, which are ‘masked’ and indistinguishable from the 169

higher levels present in the cytoplasm, (2) cytoplasmic levels titrate out components required to 170

recruit endogenous myosins to the surface of the target organelle, in effect inhibiting its action. 171

Considering that results from expression of truncated myosin fusions are corroborated by knockout 172

(T-DNA) and knock down (RNAi) studies highlights all three experimental approaches are bona fide 173

and results are not artefactual. Given that seemingly opposing methods of study, which either 174

downregulate (T-DNA, RNAi) or ‘upregulate’ myosin levels (albeit a non-functional motorless 175

variant), lead to the same experimental outcome may be due to truncations heterodimerising with 176

endogenous myosins and effectively preventing formation of functional myosin dimers. In addition, 177

expression of XI-K tail fusions containing point mutations indicates that two arginine residues control 178

the dominant negative phenotype (Avisar et al., 2012). Here, it was suggested that perhaps the tail 179

fusions bind to the head domains of endogenous myosins inhibiting action. This is based on the 180

arginine residues in XI-K corresponding with Lysine residues in myoVa which are important for 181

affecting tail to head interactions which render myoVa inactive (Li et al., 2008). Similar head-tail 182

interactions in plants myosins have not been confirmed. The exact mode of action of complete 183

myosin molecules and the dominant negative results displayed through the expression of myosin 184

truncations, will remain a mystery until a greater understanding of intramolecular (such as possible 185

head to tail) and intermolecular (myosin dimerization) interactions, and components which regulate 186

myosin recruitment and activity are determined. 187

Mechanisms of myosin recruitment: a tightly regulated system? 188

Andreas Nebenführ’s group have begun to decipher intra- and intermolecular interactions and 189

effects on organelle localisation (Li & Nebenführ, 2007; Li & Nebenführ, 2008). Here, the suggestion 190

is that regions within the GTD specify the target organelle, and that dimerization stabilises 191

association with the organelle surface; portions of the XI-1 GTD locate to ether Golgi or peroxisomes, 192

expression of full length tail (including GTD) dimerise and stabilise binding to organelles. While these 193

results provide interesting insight into the mechanism behind myosin recruitment, we are still some 194

way off from understanding how this occurs within a complete, rather than truncated, myosin 195

molecule, and whether interacting proteins mediate conformational changes within the myosin 196

hetero/homodimers to enable targeting to specific organelles. 197

Page 6 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

7

The large number of myosins in plant cells, and their seemingly overlapping functions, suggests that 198

identification of additional components at the myosin-organelle interface is required to provide 199

specificity in recruitment and regulation of myosin activity. Studies from mammals and yeast suggest 200

a myosin recruitment process mediated by receptors and adaptor proteins, either through a single 201

membrane protein (Fagarasanu et al., 2006) or larger complexes including Rab GTPases, receptors 202

and adaptors (Tang et al., 2003; Yan et al., 2005; Hume et al., 2007; Pylypenko et al., 2013). For 203

instance, myosin Myo2p of Saccharomyces cerevisiae is required for both peroxisome and vacuole 204

inheritance. Coordination of these organelles towards the bud is mediated by different receptors 205

and adaptors, which are independently regulated, at the interface between Myo2p and the 206

peroxisomal (Fagarasanu et al., 2006) or vacuolar (Tang et al., 2003) membranes. Interestingly, 207

Arabidopsis XI-2 was shown to interact in vitro with the small GTPases AtRabD1 and AtRabC2a via 208

distinct regions in its tail domain (Hashimoto et al., 2008). 209

There are few examples showing interactor binding to a myosin affects the movement of the 210

organelle to which it is bound. A.thaliana myosin XI-I specifically interacts with the proteins WIT1 211

and WIT2 located to the outer nuclear membrane (Fig.2a). Lack of either myosin XI-I or one of the 212

two WIT proteins diminishes dark-induced nuclear positioning (Tamura et al.). Notably, expression of 213

XI-I tail fusion also influences the dynamics of other organelles (Avisar, Dror et al., 2009), whereas 214

lack of WIT1/WIT2 only influences nuclear movement. Therefore, interaction between XI-I and 215

WIT1/2 controls nuclear migration. Other examples of myosin interactors include DECAPPING 216

PROTEIN 1 on P-bodies and RISAP on TGN (Steffens et al., 2014; Stephan, O. et al., 2014). 217

Recent advances in myosin recruitment studies in A.thaliana led to the identification of two new 218

classes of myosin-binding proteins: the MyoB family of potential myosin receptors (Peremyslov et 219

al., 2013; Peremyslov et al., 2015) and the MadA/MadB adaptor proteins (Kurth et al., 2017). A 220

yeast-2-hybrid (Y2H) screen on a cDNA library using the globular tail domain of myosin XI-K from 221

A.thaliana as the bait led to the identification of the MyoB family of potential myosin receptors 222

(Peremyslov et al., 2013). Whilst the lengths and architectures of the 16 identified potential 223

receptors are variable, they all share a plant specific myosin-binding DUF593 domain (Domain of 224

Unknown Function 593), originally associated with the product of the Floury1 gene in Z.mays and 225

hence named “Zein-binding domain”(Holding et al., 2007). DUF593 proteins are also present in rice 226

(Oryza sativa) (Holding et al., 2007) and in Nicotiana species (Stephan, Octavian et al., 2014; 227

Peremyslov et al., 2015). The interpretation of the role of MyoB potential receptors in myosin 228

recruitment has so far been quite challenging. Unlike other interactors including WIT1/WIT2, 229

fluorescent fusions of MyoB do not localise to known organelles. Instead, they are on highly motile 230

Page 7 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

8

unidentified membrane bounded punctae identified as unknown vesicles (Peremyslov et al., 2013; 231

Peremyslov et al., 2015; Kurth et al., 2017). The MyoB vesicles move at higher average speeds than 232

organelles such as Golgi, mitochondria and peroxisomes (Sparkes et al., 2008; Avisar, Dror et al., 233

2009; Peremyslov et al., 2015). Transformation of MyoB1-YFP into Columbia, xi-k, and xi-k/xi-1/xi-2 234

backgrounds showed that their movement and localisation is myosin-dependent (Peremyslov et al., 235

2013); MyoB1-YFP punctae proceed at high speeds in linear tracks in Columbia wild type 236

background, whereas in knockout backgrounds, and especially in the triple knockout, MyoB1-YFP 237

shows a more diffuse localisation with a few immotile punctae. Moreover, the DUF593 domain 238

seems to directly mediate the binding of MyoB potential receptors with myosins; DUF593 of 239

AtMyoB2 binds to myosin XI-K GTD in vitro (Peremyslov et al., 2013), although it is not known where 240

this interaction takes place. Similarly, the TGN-localised potential myosin receptor NtRISAP, interacts 241

in vitro with the GTD of a tobacco pollen tube myosin XI through DUF593 (Stephan, Octavian et al., 242

2014). Transient over-expression of the DUF593 domain from NbMyoB1 and NbMyoB2 in 243

N.benthamiana leads to a strong decrease in Golgi, peroxisomes and mitochondria mean velocities 244

(Peremyslov et al., 2015), perhaps suggesting a similar effect to that obtained through over-245

expression of dominant negative myosin tails (Avisar, Dror et al., 2009). Single gene knock-out (KO) 246

of myob1-4 genes does not lead to any noticeable phenotype, whereas triple 247

(myob1/myob2/myob3) and quadruple (myob1/myob2/myob3/myob4) KOs present delayed 248

flowering and reduced height (Peremyslov et al., 2013; Peremyslov et al., 2015). Interestingly, xi-249

k/xi-2/myob1/myob2/myob3 (5KO) plants show exacerbated developmental defects when 250

compared to either xi-k/xi-2 (2KO) or myob1/myob2/myob3 (3KO) plants (Peremyslov et al., 2015), 251

suggesting that MyoB proteins and myosins might be effectively involved in the same processes. 252

Peremyslov et al (Peremyslov et al., 2015) suggest that the MyoB compartment could play a major 253

role in driving cytoplasmic streaming, actively moving on the actin filament at high velocities in a 254

myosin-dependent way and thus creating the motive force for other organelles to move into the 255

cytoplasm. Whether the MyoB vesicles actually define a novel membrane compartment or a 256

specialised subdomain of a known organelle remains unclear. 257

Two novel classes of myosin-binding proteins, recently uncovered by Kurth et al. (Kurth et al., 2017), 258

are the MadA and MadB proteins. These two independent families, formed of four members each, 259

are structurally different from one another and from the MyoB family (Peremyslov et al., 2013; 260

Kurth et al., 2017) but, just as the MyoB and myosin families they appear to be conserved in all land 261

plants (Kurth et al., 2017). Aside from MadA1, which localises to the nucleoplasm, Mad proteins 262

locate to motile punctae and their localisation is myosin-dependent, similar to certain MyoBs 263

(Peremyslov et al., 2013; Kurth et al., 2017). Interestingly, single KO plants for the MadB1 gene show 264

Page 8 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

9

reduced root hair length (Kurth et al., 2017), much like what had been described for xi-2 and xi-k 265

mutants (Peremyslov, Valera V. et al., 2008). To determine whether Mad and MyoB proteins interact 266

with myosins a yeast two hybrid screen using XI-K, XI-1, XI-I, XI-F and XI-C as bait was performed. It 267

was found that numerous permutations of Mad and MyoB interactions with myosins appeared to 268

occur. Assuming that these interactions are representative of those in planta could infer that 269

control of organelle movement occurs through multiple rounds of myosin interaction with different 270

interactors to fine tune control within a short time frame. Alternatively, these combinations could 271

reflect tissue specific or developmental regulated differences or activation under certain stimuli. 272

How can we reconcile observations of organelle movement with the effects of myosins and myosin 273

interactors to generate a mechanistic model accounting for movement? The more traditional view of 274

myosins being recruited to the surface of the organelle whose movement it controls appears to hold 275

for XI-I’s control of nuclei movement (Fig 2a, b). A model for passive motion of organelles through 276

movement in cytoplasmic streams has been suggested based on MyoB / MadA-B work (Fig 2c). 277

AtMyoB2 interacts with the GTD of XI-K, both collocate to highly motile structures which move faster 278

than other spheroid organelles. XI-K is also implicated in cytoplasmic streaming. These observations 279

have led to the proposal that MyoB/MadA-B vesicles bound by myosin drive streaming, which in turn 280

affects organelle movement in a passive manner (Fig 2c). Alternatively, perhaps MyoB/MadA-B 281

vesicles are in fact decorating subdomains of organelles, such as the ER, which in turn drive 282

movement in a directed manner (Fig 2b,d). 283

284

Microtubules, associated motors, and interplay with actin 285

Two types of motor proteins are associated with microtubules: kinesins, plus-end directed, and 286

dyneins, minus-end directed. So far, sixty-one potential kinesin encoding genes have been identified 287

in the A.thaliana genome, 21 of which are suggested to be minus-end directed kinesins based on the 288

presence of a particular peptide within the neck region (Reddy & Day, 2001), which could suffice for 289

the apparent absence of dynein genes. Notably, not all plants lack genes potentially encoding for 290

dyneins: for example, four predicted dynein heavy chains have been identified in the O.sativa 291

genome (King, 2002). Similarly to myosins, kinesins general architecture is composed of a motor 292

domain, a stalk and a tail and they are classified into multiple subfamilies, named kinesin-1 to 293

kinesin-14, based on similarities in their motor domain (Lawrence et al., 2004). Unlike in mammals 294

and some fungi, microtubules and possibly kinesins appear to play a minor, but significant, role in 295

organelle movement in plants through short range rather than long distance trafficking associated 296

Page 9 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

10

with actomyosin system (reviewed in Brandizzi & Wasteneys, 2013; Cai et al., 2015). Treatments 297

with microtubule depolymerizing drugs doesn’t seem to affect long range movement of organelles 298

(Nebenführ et al., 1999; Lovy-Wheeler et al., 2007), but might affect their distribution (Van Gestel et 299

al., 2002; Idilli et al., 2013). It has been suggested that here, microtubule dependent motion allows 300

organelles to pause. Golgi bodies are known to move in the cell in an acto-myosin dependent way 301

(Boevink et al., 1998; Nebenführ et al., 1999; Sparkes et al., 2008), and to pause at specific sites in 302

correspondence of cortical microtubules (Crowell et al., 2009). Similar observations have been made 303

for mitochondria and peroxisomes pausing at actin microtubule junctions. However, pausing may 304

not be microtubule dependent events as it occurred upon microtubule depolymerisation but was 305

noted to occur at ER junctions (Hamada et al., 2012). Given that stable ER junctions are possible sites 306

of tethering to the PM (see later), perhaps organelles pause either due to a physical blockage from 307

the ER, actin, microtubule complex, or perhaps pausing serves a functional role through 308

communication at ER-PM sites (Bayer et al., 2017). In addition to actin, microtubules have been 309

shown to contribute to ER tubule elongation and anchoring in A.thaliana (Hamada et al., 2014), and 310

exo- and endocytosis and endosomal movement in tobacco pollen tube tip (Idilli et al., 2013). 311

Moreover, kinesins have been detected on organelle membranes; Kinesin-13A on Golgi stacks (Lu, 312

Ling et al., 2005) and Golgi-associated vesicles (Wei et al., 2009), KAC1 and KAC2 on chloroplast 313

(Suetsugu et al., 2010), which are interestingly able to bind actin. In At kinesin-13 mutant plants, 314

Golgi stacks aggregate in trichomes and in other epidermal cells, with trichomes often exhibiting 4 315

branches instead of 3 (Lu, L. et al., 2005). Although some Golgi stacks were found to be associated 316

with microtubules, it is currently unknown whether Kinesin-13 has an active role in moving Golgi 317

along microtubules. 318

Interestingly, there may be functional cooperation and coordination between actin and 319

microtubules which could affect organelle dynamics. Studies in A.thaliana seedlings treated with 320

cytoskeleton stabilizing and de-stabilizing drugs suggest a coordinated re-organisation of actin and 321

microtubules (Sampathkumar et al., 2011). It has been observed that some plant formins, proteins 322

traditionally associated with regulating actin dynamics, can also associate and bind to microtubules 323

(Deeks et al., 2010; Wang et al., 2013; Sun et al., 2017). The rice formin OsFH15, for example, was 324

shown to bind to and induce bundling of both actin and microtubules, being also able to cross-link 325

the two cytoskeletal networks (Sun et al., 2017). Therefore, actin and microtubules are intimately 326

linked and coordinated networks. Myosins, classically thought of as ‘binding’ organelles and 327

controlling their movement directly, may in fact control organelle movement indirectly through 328

controlling actin dynamics (Peremyslov et al., 2010; Cai et al., 2014; Madison et al., 2015). Therefore, 329

could myosins affect both actin and microtubule networks directly or indirectly respectively? 330

Page 10 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

11

331

In conclusion, both actin and microtubules are important regulators of organelle dynamics. Future 332

research into coordinated activities, either through myosin / kinesin or dynamics of their 333

corresponding cytoskeletal tracks, will help decipher the molecular cues resulting in the observed 334

seemingly chaotic movement displayed in cells undergoing diffuse growth. 335

336

Role of organelle interactions: tales of tethers 337

338

Organelle interactions occur at membrane contact sites (MCS), where membranes from apposing 339

organelles are within 10-30nm of one another. In mammals and yeast numerous interactions at 340

MCS’ between various organelle pairs, and the molecular components required for physical 341

association, have been elucidated. These interactions play important roles in exchange of 342

components such as lipids and calcium (Prinz, 2014). Therefore, physical interaction between 343

organelles plays a functional role in communication between organelles, and therefore affects 344

organelle positioning. 345

In plants, the role of MCS is currently limited to the identification of a physical association of a 346

handful of organelle pairings (such as ER-Golgi, peroxisome-chloroplast), with scant identification of 347

the molecular components involved at the tether site (eg. CASP for ER-Golgi, retromer components 348

for oil body-peroxisome). Here, we will focus on each organelle pairing in turn, provide a brief 349

rational for the functional role for association, and the evidence that interaction occurs. Note, 350

correlation of movement between two closely opposed organelles has been assumed to reflect 351

physical tethering between compartments. Isolation and characterisation of bona fide molecular 352

tethering components will provide more conclusive evidence on interaction between organelles. 353

The potential role for membrane extensions seen emanating from, and in between, certain 354

organelles will also be discussed. It is important to note that whilst our understanding of 355

interactions in plants appears rudimentary at this stage compared with other model systems, the 356

development and application of novel biophysical tools (such as optical tweezers and pressure wave 357

technology) in plants will allow for a greater understanding of the regulation of interactions as the 358

molecular components are isolated. 359

360

ER-Golgi 361

Page 11 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

12

The secretory pathway begins with protein production in the ER and subsequent processing in the 362

Golgi. The functional relationship and observed close associations between these two organelles 363

lead to the hypothesis that physical association, be it through membrane continuity or large 364

proteineous tethering complex, must occur. Whilst trafficking of cargo between the two organelles is 365

cytoskeleton independent, movement of the two organelles are correlated and are cytoskeleton 366

dependent processes (see earlier, Brandizzi et al., 2002; Runions et al., 2006; Stefano et al., 2014). 367

To probe physical association between the two, optical tweezers attached to either a confocal 368

microscope or TIRF imaging system have been utilised. The principle behind optical tweezers is that 369

it uses a focused infrared beam which allows the user to physically trap and move an object which 370

has a different refractive index to the surrounding media (Sparkes et al., 2018; Sparkes, 2016). 371

Unlike the ER, Golgi bodies are amenable to being trapped in intact Arabidopsis leaf epidermal cells. 372

Physical association between the ER and Golgi would result in a trapped Golgi body dragging the ER. 373

If Golgi bodies were not physically associated with the ER, then the subsequent movement of a 374

trapped Golgi would separate it from the underlying ER. Qualitative assessment indicated that the 375

ER did, in fact, move with the Golgi and, on occasion, Golgi could be separated from the ER (Sparkes, 376

IA et al., 2009). More recently, semi-quantitative work has shown that CASP, a Golgi associated 377

protein, affects association between the two organelles (Osterrieder et al., 2017). These results, and 378

the application of optical tweezers, clearly highlight how we can determine rates of association 379

between organelles, and the relative strength of such through affecting the molecular components 380

involved in the tethering process. Future studies are required to elucidate components, in addition 381

to CASP, that are involved in the tethering of Golgi to the ER. Furthermore, application of a fully, 382

rather than partially, quantifiable optical trapping platform (Gao et al., 2016) will address the 383

relative importance of CASP and interactors in the tethering process (Sparkes et al., 2018). 384

385

Nucleus-chloroplast 386

Movement of nuclei and chloroplasts are controlled through short actin filament assembly at the 387

organelle surface. Tethering between nuclei and chloroplasts has been inferred through photo 388

induced migration of chloroplasts affecting the movement of associated nuclei (Higa et al., 2014). 389

Physical interaction between these two organelles is likely required for retrograde signalling. Elegant 390

signalling studies have shown that hydrogen peroxide levels in chloroplasts juxtaposed to the 391

nucleus alter in response to light, which ultimately changes nuclear expression of genes required for 392

photosynthesis (Exposito-Rodriguez et al., 2017). Identification and characterisation of molecular 393

tethers required for interaction between the nucleus and chloroplast will provide conclusive 394

evidence of physical association between the two compartments. 395

Page 12 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

13

396

ER-peroxisomes / oil bodies-peroxisomes 397

Peroxisome biogenesis is a contentious topic with two overarching proposed mechanisms; de novo 398

synthesis from the ER or fission from pre-existing peroxisomes. In plants there is very little evidence 399

for the former (Kao et al., 2018). Live cell imaging indicates that peroxisomes, and the resulting 400

membrane emanation peroxules, are closely aligned with the ER (Sinclair et al., 2009; Barton et al., 401

2013). It is unclear if this represents a direct physical association, or simply occupying the same actin 402

filament, or due to spatial constrictions imposed by the large central vacuole of cytoplasmic content, 403

in effect, pushing content together. 404

405

The ER is the site of production for oil bodies. Energy production through β-oxidation of the fatty 406

acids derived from oil bodies occurs in plant peroxisomes. Recent evidence has emerged indicating 407

close association between oil bodies and peroxisomes, which may be mediated by the retromer 408

complex (Thazar-Poulot et al., 2015). This retromer mediated interaction appeared to occur at 409

peroxisome membrane extensions (peroxules, see later). Changes in metabolic requirements 410

through provision of sucrose alters the frequency of oil body-peroxisome juxtaposition (Cui et al., 411

2016). In combination these results suggest that peroxisomes are associated with oil bodies 412

during β-oxidation, but it remains to be seen whether this interaction is direct or indirectly mediated 413

by an ER bridge. 414

415

Peroxisome-chloroplast 416

Peroxisomes and chloroplasts are functionally linked through the photorespiratory pathway, which 417

also spans mitochondria. Early micrographs again highlight close association between these 418

organelles. By altering the molecular components which drive chloroplast motion (chup1), 419

neighbouring peroxisomes appeared to move with chloroplasts as though physically attached 420

(Oikawa et al., 2003). Clear evidence for physical association came through optical tweezer (Fig 3) 421

and pressure wave studies (Oikawa et al., 2015; Gao et al., 2016). The latter generates a wave which 422

can affect, and effectively ‘push’ all subcellular components within the pressure wave radius, 423

whereas optical tweezers targets individual organelles with submicron accuracy. Both studies 424

indicated that peroxisomes are attached to chloroplasts, with pressure wave studies inferring that 425

interaction maybe affected by photosynthetic rate. Given that photosynthesis and photorespiration 426

are linked processes, it is unlikely that photosynthesis per se drives the organelle association and is 427

perhaps more likely a correlation. Isolating the molecular components which drive these interactions 428

will allow a deeper understanding of regulation and subsequent role of peroxisome-chloroplast 429

Page 13 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

14

interaction. Considering the functional repertoire of plant peroxisomes has expanded beyond earlier 430

concepts of lipid catabolism (β-oxidation), reactive oxygen species (ROS) production and scavenging , 431

it is not impossible to propose that novel functional roles may still exist requiring shuttling between 432

chloroplasts and peroxisomes (Kao et al., 2018). For example, β-oxidation is not only required for 433

energy mobilisation in germinating oilseeds, but also plays roles in production of hormones including 434

Jasmonic acid (JA). JA synthesis spans both peroxisomes and chloroplasts. Considering that 435

metabolism is a complex process, by controlling physical location of organelles we may be able to 436

alter metabolic balance. 437

438

ER-plasma membrane 439

Early work by Ridge et al. indicated static regions within the ER network (Ridge et al., 1999). More 440

recently, through the development of a quantitative platform to measure static elements within the 441

highly dynamic ER network (termed persistency mapping), static tubules, nodes and cisternal regions 442

were isolated which comprised ~5% of the ER network in tobacco leaf epidermal cell (Sparkes, I et 443

al., 2009). By observing the movement around these nodes, it appears as the though the ER network 444

‘wrapped’ itself and anchored in place to generate new geometric configurations of the network. 445

The premise of anchoring was again shown using optical tweezers; moving trapped Golgi bodies 446

around ER nodes to wrap and anchor the ER in place (Sparkes, IA et al., 2009). These anchor points 447

were in fact regions of the ER pinned to the plasma membrane. Plasmolysis studies result in the 448

plasma membrane being ‘pulled’ away from the cell wall to generate Hechtian strands (Wang et al., 449

2016). These strands highlight that the PM is fixed to the cell wall at junctions. The Hechtian strands 450

are filled with ER which reach up to the tip of the strand where the PM meets the cell wall (Wang et 451

al., 2016; Cheng et al. 2017). This arrangement is indicative of the ER being anchored to the PM, 452

which is fixed to the cell wall at discrete regions (Wang & Hussey, 2015). This arrangement could 453

enable fast transmission of signals between the extracellular and intracellular environments, and 454

also act to constrain and maintain an even distribution of the ER within the cell cortex (Wang & 455

Hussey, 2015; Bayer et al., 2017). Interestingly, attempts to model ER network formation based on 456

parameters including static and mobile ER node distribution shows a high correlation with formation 457

in vivo (Lin et al., 2014; Lin et al., 2017). As previously mentioned anchoring is through actin and 458

microtubules, and requires a complex of proteins including Net3, VAP27, synaptotagmin 71 and 459

Syp73 (Bayer et al., 2017; Wang et al., 2014, 2016, 2017; Wang & Hussey, 2017; Perez-Sancho et al., 460

2015; Cao et al., 2016). This complex could possibly affect the movement of other organelles, or act 461

as a functional nexus in the cell allowing organelles to ‘communicate’. Future studies to determine 462

how node formation is regulated, turned over and how the ER appears to randomly ‘hook’ and 463

Page 14 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

15

‘unhook’ itself from the PM will promote our understanding of the functional role of ER-PM 464

interactions. 465

466

467

Stromules, peroxules and matrixules: Tubular extensions as connections to other organelles? 468

Stromules, peroxules and matrixules are membrane protrusions containing luminal content 469

emanating from chloroplasts, peroxisomes and mitochondria respectively. Several hypothesis have 470

been proposed for the functional role of such extensions; to increase surface area, to allow 471

organelles to communicate and exchange components at the tip of the extension, or / and to anchor 472

organelles to one another. These hypothesis are not mutually exclusive. For example communication 473

could occur at MCS’ between two organelles, if the organelles start to move at differing rates then a 474

physical manifestation of maintaining contact would be formation of extensions (Figure 4). 475

Interestingly, peroxule formation occurs upon trapping and pulling peroxisomes away from 476

chloroplasts (Gao et al., 2016) (Fig 3). Quantification of this event indicated that peroxules at the 477

chloroplast-peroxisome contact site are anchored with a degree of tension being induced. Modelling 478

of the forces required to ‘pull’ peroxisomes back to the chloroplast gives an indication of the force 479

required by myosins to pull peroxisomes away from chloroplasts in vivo. 480

481

Evidence for transfer occurring at the tips of extensions is contentious, and is based on studies of 482

exchange between stromules from neighbouring plastids (Hanson & Sattarzadeh, 2013; Mathur et 483

al., 2013). Membrane extensions also form in response to other stimuli including ROS and is 484

reviewed in (Mathur et al., 2012). More recently stromule formation has been linked with pathogen 485

infection (Caplan et al., 2015). 486

487

Stromules and peroxules appear to co-align with the ER network, leading to the proposal that they 488

are physically attached to the ER network (Schattat et al., 2011; Barton et al., 2013). Peroxules are 489

single extensions unlike stromules which can form branched networks. The ER is pervasive 490

throughout the cortex, organelles in highly vacuolated cells are constrained and can appear to 491

neighbour one another at random, and movement of both the ER and these organelles are largely 492

driven by acto-myosin dependent processes. These observations make it difficult to determine 493

whether there is a true physical association between the ER and these other organelles, or whether 494

they are ‘connected’ through association to the same actin filament. Future studies are required to 495

reconcile which of these scenarios is correct. 496

497

Page 15 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

16

Interestingly, a biochemical complementation approach highlighted that the ER and chloroplasts can 498

exchange nonpolar metabolites (Mehrshahi et al., 2013). This exchange was suggested to likely occur 499

at closely opposed membrane sites. Photostimulation studies highlighted a potential interaction site 500

between the two organelles which may affect movement of the ER network and protein within 501

(Griffing L, 2011). Optical tweezer studies in laser ablated cells indicated a physical association 502

between the ER and chloroplasts (Andersson et al., 2007). Future studies to pinpoint molecular 503

tethering components will determine if there is a true physical association between these two 504

organelles. 505

506

Summary model to describe organelle movement in higher plants 507

To take into account the level of evidence for physical interaction between organelles and the role of 508

actomyosin system on movement, Figure 4 proposes how movement of tethered organelles is 509

coordinated and regulated. Here, movement of discrete untethered organelles are depicted as being 510

controlled independently of one another (Fig 4a). This could be through specific myosin-organelle 511

interaction mediated by myosin recruitment factors (receptors, adaptors previously mentioned) and 512

relates to Fig 2b. For organelles tethered to one another (Fig 4b-d), movement could be regulated by 513

one of the two organelles in turn pulling the other (Fig 4b, organelle A pulls tethered organelle B), or 514

movement of the two tethered organelles could be co-regulated by coordinated action of myosins 515

on each organelle (Fig 4c). Here, co-regulation could move the two organelles in the same direction 516

through interaction of organelle specific myosins binding and traversing different F-actin strands. In 517

addition, it is interesting to point out that if one organelle moves faster than the other, or changes 518

direction of motion, here it may result in changes in morphology of the tethered organelles resulting 519

in membrane extensions such as peroxules, matrixules and stromules. Evidence supporting this 520

could be drawn from increased peroxisome speeds under ROS stress which also induces peroxule 521

formation; are these organelles simply trying to maintain contact with other organelles? Given that 522

the ER may be involved in multiple organelle interactions then here movement of organelles could 523

be mediated through a three way organelle interaction complex; organelle A moves which causes ER 524

movement which in turn affects organelle B’s movement (Fig 4d). In addition, organelle movement 525

could possibly be driven through direct interaction with actin, with actin polymerisation itself driving 526

organelle movement. For example, proteins such as Net3 and Syp73 connect the ER to actin. These 527

models do not include cytoplasmic streaming, an additional complication which needs to be taken 528

into account to aid our understanding of the local and global regulation of organelle movement and 529

interaction. Could streaming increase the chance of organelle collisions which then ‘stick’ to one 530

another if they contain the correct tethering components? Could the scenario in Fig 4d explain 531

Page 16 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

17

cytoplasmic streaming; movement of organelle A and B controls ER movement which in turn 532

generates viscous drag to entrain the cytoplasmic flow? To begin to unpick which, if any, of these 533

suggested models occurs will require further understanding of the tethers involved in allowing 534

interaction of organelles and how this impacts on organelle movement and positioning. 535

536

537

Why is organelle movement important? 538

539

Recent advances suggest that organelle movement plays a fundamental role in plant growth and 540

development and in response to biotic and abiotic stresses/stimuli. Readers are directed towards 541

recent reviews covering this topic in more depth (Duan & Tominaga, 2018; Ryan & Nebenführ, 542

2018). In brief, here, myosin mutants with reduced organelle movement can display a reduction in 543

aerial tissue growth and therefore biomass, altered flowering timing and reduced seed set. These 544

gross morphological phenotypes are due to changes in cell size rather than cell number, and as 545

shown in root hairs, reduced polar tip growth (Peremyslov, V. V. et al., 2008; Prokhnevsky et al., 546

2008; Peremyslov et al., 2010). This is also seen in pollen tubes and so may reflect reduced 547

fertilisation affecting seed set (Madison et al., 2015). In addition, there are also alterations in 548

trichome growth (Ojangu et al., 2007; Ojangu et al., 2012). 549

Key players in organelle movement and actin dynamics include myosins XI-K, XI-1 and XI-2. 550

Considering that these myosins have also been implicated in cytoplasmic streaming, could these 551

phenotypes be a direct result of alterations in streaming rates rather than directed organelle 552

movement per se? Here, the hypothesis would be that streaming mixes the cytoplasm and that 553

organelle movement is passive in the process. Work presented by Tominaga et al. tested this 554

hypothesis by generating transgenic Arabidopsis expressing XI-2 chimeras which either had a higher 555

or slower ‘speed’ (Tominaga et al., 2013). Here, they found that higher ‘speed’ motor resulted in 556

faster growth compared with plants expressing the ‘lower’ speed variant, and that growth rates 557

related to cell size. They concluded that cytoplasmic streaming is therefore a determinant of cell 558

size. While these results show a clear correlation between streaming and cell size, it does not negate 559

that size may be due to effecting the movement of the organelles rather than mixing of the 560

cytoplasm. It is difficult to reconcile the relative contributions of these two processes (cytoplasmic 561

mixing versus organelle movement) on cell growth given our current molecular tools. Specifically 562

targeting and affecting the movement of individual classes of organelles will begin to unpick these 563

processes and relative impacts on cell growth. Clear evidence, however, for the role of chloroplast 564

and nuclear movement has been elucidated. 565

Page 17 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

18

566

Both chloroplasts and nuclei positioning changes in response to light, a process called 567

photorelocation (Tsuboi et al., 2007; Kodama et al., 2008; Wada, 2013; Wada, 2016). Chloroplasts 568

locate to the periclinal walls of mesophyll cells of A.thaliana under low blue light intensities 569

(accumulation response) in order to maximise light capture and, therefore, photosynthesis. They 570

then migrate to anticlinal walls (avoidance response) in conditions of high blue light intensities in 571

order to avoid photodamage (Kasahara et al., 2002). Phototropin1 (phot1) and Phototropin2 (phot2) 572

are the blue light receptors involved in perceiving the intensity of blue light: both phot1 and phot2 573

are involved in the accumulation response (Sakai et al., 2001), whereas only phot2 is involved in the 574

avoidance response (Kagawa et al., 2001). Interestingly, phot2 is also involved in nuclei blue light 575

avoidance response (Iwabuchi et al., 2010). 576

The position that organelles occupy in the cell also appears to be finely regulated. During root hair 577

growth, for instance, the nucleus moves forward maintaining the same distance from the tip 578

(Chytilova et al., 2000; Ketelaar et al., 2002), until the growth process is complete. Drug treatments 579

with actin depolymerising drugs and immobilisation of the nucleus by optical trapping (Ketelaar et 580

al., 2002), suggest that correct positioning of the nucleus from the growing tip is essential. Under 581

certain growth conditions, wild type plants can produce and extend two root hairs from the same 582

single trichoblast cell. These root hairs grow and elongate in a coordinated manner albeit with a 583

slight lag in growth of the nuclear ‘free’ root hair (Jones & Smirnoff, 2006). These events are rare, 584

and question whether nuclear positioning is completely essential for root hair elongation. 585

586

587

Conclusions and future perspectives 588

589

We have provided an overview of the molecular mechanisms underpinning organelle movement and 590

positioning in plants; largely actomyosin dependent and interaction between organelles. Whilst 591

there appears to be a degree of functional redundancy between the myosin class XI family, it should 592

be noted that there appears to be a certain degree of specific control over ER dynamics (Griffing et 593

al., 2014). Could a similar level of control occur within subpopulations of discrete organelles such as 594

Golgi, peroxisomes and mitochondria? When observing movement of these organelles it is usual to 595

monitor a fluorophore targeted to the organelle through a targeting sequence which doesn’t 596

necessarily represent a fully functional native protein. The population of a certain class of organelle 597

Page 18 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

19

therefore appears homogenous. In reality, could organelle groups be more heterogenous, and could 598

this heterogeneity reflect the movement characteristics of the organelle itself? For example, a cargo 599

laden organelle may need to move faster to exchange and release cargo unlike another which has a 600

lower cargo load, or a different cargo load. Future studies into organelle heterogeneity and how this 601

regulates movement will unveil this exciting possibility that organelles regulate their own movement 602

patterns through signalling cascades. This has already begun to occur through studies of the effects 603

of ROS on organelle dynamics. To be able to pull out patterns of movement, and relate them to a 604

functional response will therefore require development and application of novel sensors (such as 605

ratiometric ATP probe (De Col et al., 2017), novel applications of imaging techniques (such as optical 606

tweezers for testing and quantifying organelle interactions), and mathematical and biophysical 607

models. The latter will provide predictive power to determine which parts of the system provide the 608

ultimate control over organelle dynamics. For example, models of ER network formation have 609

provided a first principle approximation of the biophysical properties required to form a dynamic 610

model of network formation which fits in vivo ER network dynamics (Lemarchand et al., 2014; Lin et 611

al., 2015; Griffing et al., 2017; Lin et al., 2017). 612

613

614

Acknowledgements 615

Owing to space limitations we apologise to those whose work we could not include herein. IS and CP 616

are supported by the Gatsby foundation and the Leverhulme Trust. Some of the work reported 617

herein was supported by the BBSRC and STFC. 618

619

Authors contributions 620

621

Both authors contributed equally to the work 622

623

References 624

Andersson MX, Goksor M, Sandelius AS. 2007. Optical manipulation reveals strong attracting forces 625

at membrane contact sites between endoplasmic reticulum and chloroplasts. The Journal of 626

Biological Chemistry 282: 1170-1174. 627

Avisar D, Abu-Abied M, Belausov E, Sadot E. 2012. Myosin XIK is a major player in cytoplasm 628

dynamics and is regulated by two amino acids in its tail. Journal of Experimental Botany 629

63(1): 241-249. 630

Avisar D, Abu-Abied M, Belausov E, Sadot E, Hawes C, Sparkes IA. 2009. A comparative study of the 631

involvement of 17 Arabidopsis myosin family members on the motility of Golgi and other 632

organelles. Plant Physiology 150(2): 700-709. 633

Page 19 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

20

Barton K, Mathur N, Mathur J. 2013. Simultaneous live-imaging of peroxisomes and the ER in plant 634

cells suggests contiguity but no luminal continuity between the two organelles. Frontiers in 635

Physiology 4: 196. 636

Bayer EM, Sparkes I, Vanneste S, Rosado A. 2017. From shaping organelles to signalling platforms: 637

the emerging functions of plant ER-PM contact sites. Current Opinion in Plant Biology 40: 89-638

96. 639

Boevink P, Oparka K, Cruz SS, Martin B, Betteridge A, Hawes C. 1998. Stacks on tracks: the plant 640

Golgi apparatus traffics on an actin/ER network. The Plant Journal 15(3): 441-447. 641

Bonifacino JS, Glick BS. 2004 The mechanisms of vesicle budding and fusion. Cell 116 (2):152-66 642

Brandizzi F, Snapp EL, Roberts AG, Lippincott-Schwartz J, Hawes C. 2002. Membrane protein 643

transport between the endoplasmic reticulum and the Golgi in tobacco leaves is energy 644

dependent but cytoskeleton independent: evidence from selective photobleaching. Plant 645

Cell 14(6): 1293-1309. 646

Brandizzi F, Wasteneys GO. 2013. Cytoskeleton-dependent endomembrane organization in plant 647

cells: an emerging role for microtubules. The Plant Journal 75(2): 339-349. 648

Cai C, Henty-Ridilla JL, Szymanski DB, Staiger CJ. 2014. Arabidopsis myosin XI: a motor rules the 649

tracks. Plant Physiology 166(3): 1359-1370. 650

Cai G, Parrotta L, Cresti M. 2015. Organelle trafficking, the cytoskeleton, and pollen tube growth. 651

Journal of Integrative Plant Biology 57(1): 63-78. 652

Cao P, Renna L, Stefano G, Brandizzi F. 2016. SYP73 Anchors the ER to the Actin Cytoskeleton for 653

Maintenance of ER Integrity and Streaming in Arabidopsis. Curr Biol 26(23): 3245-3254. 654

Caplan JL, Kumar AS, Park E, Padmanabhan MS, Hoban K, Modla S, Czymmek K, Dinesh-Kumar SP. 655

2015. Chloroplast stromules function during innate immunity. Developmental Cell 34(1): 45-656

57. 657

Cheng X, Lang I, Adeniji OS, Griffing L. 2017. Plasmolysis-deplasmolysis causes changes in 658

endoplasmic reticulum form, movement, flow, and cytoskeletal association. Journal of 659

Experimental Botany, 68(15):4075–4087 660

Chytilova E, Macas J, Sliwinska E, Rafelski SM, Lambert GM, Galbraith DW. 2000. Nuclear dynamics 661

in Arabidopsis thaliana. Molecular Biology of the Cell 11(8): 2733-2741. 662

Crowell EF, Bischoff V, Desprez T, Rolland A, Stierhof Y-D, Schumacher K, Gonneau M, Höfte H, 663

Vernhettes S. 2009. Pausing of Golgi bodies on microtubules regulates secretion of cellulose 664

synthase complexes in Arabidopsis. Plant Cell 21(4): 1141-1154. 665

Cui S, Hayashi Y, Otomo M, Mano S, Oikawa K, Hayashi M, Nishimura M. 2016. Sucrose production 666

mediated by lipid metabolism suppresses the physical interaction of peroxisomes and oil 667

bodies during germination of Arabidopsis thaliana. Journal of Biological Chemistry 291(38): 668

19734-19745. 669

Damme DV, Bouget F-Y, Poucke KV, Inzé D, Geelen D. 2004. Molecular dissection of plant 670

cytokinesis and phragmoplast structure: a survey of GFP-tagged proteins. The Plant Journal 671

40(3): 386-398. 672

De Col V, Fuchs P, Nietzel T, Elsasser M, Voon CP, Candeo A, Seeliger I, Fricker MD, Grefen C, 673

Moller IM, et al. 2017. ATP sensing in living plant cells reveals tissue gradients and stress 674

dynamics of energy physiology. eLife 6. 675

Deeks MJ, Fendrych M, Smertenko A, Bell KS, Oparka K, Cvrčková F, Žárský V, Hussey PJ. 2010. The 676

plant formin AtFH4 interacts with both actin and microtubules, and contains a newly 677

identified microtubule-binding domain. Journal of Cell Science 123(8): 1209-1215. 678

Duan Z, Tominaga M. 2018. Actin-myosin XI: an intracellular control network in plants. Biochemical 679

and Biophysical Research Communications, in press. 680

Exposito-Rodriguez M, Laissue PP, Yvon-Durocher G, Smirnoff N, Mullineaux PM. 2017. 681

Photosynthesis-dependent H2O2 transfer from chloroplasts to nuclei provides a high-light 682

signalling mechanism. Nature Communications 8(1): 49. 683

Page 20 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

21

Fagarasanu A, Fagarasanu M, Eitzen GA, Aitchison JD, Rachubinski RA. 2006. The peroxisomal 684

membrane protein Inp2p is the peroxisome-specific receptor for the myosin V motor Myo2p 685

of Saccharomyces cerevisiae. Developmental Cell 10(5): 587-600. 686

Foth BJ, Goedecke MC, Soldati D. 2006. New insights into myosin evolution and classification. 687

Proceedings of the National Academy of Sciences of the United States of America 103(10): 688

3681-3686. 689

Gao H, Metz J, Teanby NA, Ward AD, Botchway SW, Coles B, Pollard M, Sparkes I. 2016. In vivo 690

quantification of peroxisome tethering to chloroplasts in tobacco epidermal cells using 691

optical tweezers. Plant Physiology 170: 263-272. 692

Golomb L, Abu-Abied M, Belausov E, Sadot E. 2008. Different subcellular localizations and functions 693

of Arabidopsis myosin VIII. BMC Plant Biology 8(1): 3. 694

Goldstein RE, Tuval I, van de Meent JW. 2008. Microfluidics of cytoplasmic streaming and its 695

implications for intracellular transport. Proceedings of the National Academy of Sciences USA 696

105(1):3663-3667 697

Griffing LR, Gao HT, Sparkes I. 2014. ER network dynamics are differentially controlled by myosins 698

XI-K, XI-C, XI-E, XI-I, XI-1, and XI-2. Frontiers in Plant Science 5: 218. 699

Griffing LR, Lin C, Perico C, White RR, Sparkes I. 2017. Plant ER geometry and dynamics: biophysical 700

and cytoskeletal control during growth and biotic response. Protoplasma 254(1): 43-56. 701

Hamada T, Tominaga M, Fukaya T, Nakamura M, Nakano A, Watanabe Y, Hashimoto T, Baskin TI. 702

2012. RNA processing bodies, peroxisomes, Golgi bodies, mitochondria, and endoplasmic 703

reticulum tubule junctions frequently pause at cortical microtubules. Plant and Cell 704

Physiology 53(4): 699-708. 705

Hamada T, Ueda H, Kawase T, Hara-Nishimura I. 2014. Microtubules contribute to tubule 706

elongation and anchoring of endoplasmic reticulum, resulting in high network complexity in 707

Arabidopsis. Plant Physiology 166(4): 1869-1876. 708

Hanson MR, Sattarzadeh A. 2013. Trafficking of proteins through plastid stromules. Plant Cell 25(8): 709

2774-2782. 710

Hashimoto K, Igarashi H, Mano S, Nishimura M, Shimmen T, Yokota E. 2005. Peroxisomal 711

localization of a myosin XI isoform in Arabidopsis thaliana. Plant and Cell Physiology 46(5): 712

782-789. 713

Hashimoto K, Igarashi H, Mano S, Takenaka C, Shiina T, Yamaguchi M, Demura T, Nishimura M, 714

Shimmen T, Yokota E. 2008. An isoform of Arabidopsis myosin XI interacts with small 715

GTPases in its C-terminal tail region. Journal of Experimental Botany 59(13): 3523-3531. 716

Hepler PK, Vidali L, Cheung AY. 2001. Polarized cell growth in higher plants. Annual Review of Cell 717

and Developmental Biology 17: 159-187. 718

Higa T, Suetsugu N, Kong SG, Wada M. 2014. Actin-dependent plastid movement is required for 719

motive force generation in directional nuclear movement in plants. Proceedings of the 720

National Academy of Sciences of the United States of America 111(11): 4327-4331. 721

Holding DR, Otegui MS, Li B, Meeley RB, Dam T, Hunter BG, Jung R, Larkins BA. 2007. The maize 722

Floury1 gene encodes a novel endoplasmic reticulum protein involved in zein protein body 723

formation. Plant Cell 19(8): 2569-2582. 724

Hume AN, Ushakov DS, Tarafder AK, Ferenczi MA, Seabra MC. 2007. Rab27a and MyoVa are the 725

primary Mlph interactors regulating melanosome transport in melanocytes. Journal of Cell 726

Science 120(17): 3111-3122. 727

Idilli AI, Morandini P, Onelli E, Rodighiero S, Caccianiga M, Moscatelli A. 2013. Microtubule 728

depolymerization affects endocytosis and exocytosis in the tip and influences endosome 729

movement in tobacco pollen tubes. Molecular Plant 6(4): 1109-1130. 730

Iwabuchi K, Minamino R, Takagi S. 2010. Actin reorganization underlies phototropin-dependent 731

positioning of nuclei in Arabidopsis leaf cells. Plant Physiology 152(3): 1309. 732

Page 21 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

22

Jedd G, Chua NH. 2002. Visualization of peroxisomes in living plant cells reveals acto-myosin-733

dependent cytoplasmic streaming and peroxisome budding. Plant and Cell Physiology 43(4): 734

384-392. 735

Jones M, Smirnoff N. 2006. Nuclear dynamics during the simultaneous and sustained tip growth of 736

multiple root hairs arising from a single root epidermal cell. Journal of Experimental Botany 737

57(15):4269-4275 738

Kachar B, Reese TS. 1988. The mechanism of cytoplasmic streaming in characean algal cells: sliding 739

of endoplasmic reticulum along actin filaments. Journal of Cell Biology 106(5):1545-52. 740

741

Kagawa T, Sakai T, Suetsugu N, Oikawa K, Ishiguro S, Kato T, Tabata S, Okada K, Wada M. 2001. 742

Arabidopsis NPL1: A phototropin homolog controlling the chloroplast high-light avoidance 743

response. Science 291(5511): 2138. 744

Kao YT, Gonzalez KL, Bartel B. 2018. Peroxisome function, biogenesis, and dynamics in plants. Plant 745

Physiology 176(1): 162-177. 746

Kasahara M, Kagawa T, Oikawa K, Suetsugu N, Miyao M, Wada M. 2002. Chloroplast avoidance 747

movement reduces photodamage in plants. Nature 420: 829. 748

Ketelaar T, Faivre-Moskalenko C, Esseling JJ, de Ruijter NCA, Grierson CS, Dogterom M, Emons 749

AMC. 2002. Positioning of nuclei in Arabidopsis root hairs: an actin-regulated process of tip 750

growth. The Plant Cell 14(11): 2941-2955. 751

King SM. 2002. Dyneins motor on in plants. Traffic 3(12): 930-931. 752

Kodama Y, Tsuboi H, Kagawa T, Wada M. 2008. Low temperature-induced chloroplast relocation 753

mediated by a blue light receptor, phototropin 2, in fern gametophytes. Journal of Plant 754

Research 121(4): 441-448. 755

Kurth EG, Peremyslov VV, Turner HL, Makarova KS, Iranzo J, Mekhedov SL, Koonin EV, Dolja VV. 756

2017. Myosin-driven transport network in plants. Proceedings of the National Academy of 757

Sciences of the United States of America 114(8): E1385-E1394. 758

Lawrence CJ, Dawe RK, Christie KR, Cleveland DW, Dawson SC, Endow SA, Goldstein LSB, Goodson 759

HV, Hirokawa N, Howard J, et al. 2004. A standardized kinesin nomenclature. Journal of Cell 760

Biology 167(1): 19-22. 761

Lemarchand L, Euler R, Lin C, Sparkes I 2014. Modeling the geometry of the endoplasmic reticulum 762

network. In: Dediu AH, Martin-Vide C, Truthe B eds. Algorithms for Computational Biology. 763

AlCoB 2014. Lecture Notes in Computer Science, vol 8542. Springer, Cham. 764

Li J, Nebenführ A. 2007. Organelle targeting of myosin XI is mediated by two globular tail domains 765

with separate cargo binding sites. Journal of Biological Chemistry 282: 20593-20602. 766

Li JF, Nebenführ A. 2008. Inter-dependence of dimerisation and organelle binding in myosin XI. Plant 767

Journal 55: 478-490. 768

Li XD, Jung HS, Wang Q, Ikebe R, Craig R, Ikebe M. 2008. The globular tail domain puts on the brake 769

to stop the ATPase cycle of myosin Va. Proceedings of the National Academy of Sciences of 770

the United States of America 105(4): 1140-1145. 771

Lin C, Lemarchand L, Euler R, Sparkes I. 2015. Modeling the geometry and dynamics of the 772

Endoplasmic Reticulum network. IEEE/ACM Transactions on Computational Biology and 773

Bioinformatics 15 (2): 377-386. 774

Lin C, White RR, Sparkes I, Ashwin P. 2017. Modeling endoplasmic reticulum network maintenance 775

in a plant cell. Biophysical Journal 113(1): 214-222. 776

Lin C, Zhang Y, Sparkes I, Ashwin P. 2014. Structure and dynamics of ER: minimal networks and 777

biophysical constraints. Biophysical Journal 107(3): 763-772. 778

Logan DC, Leaver CJ. 2000. Mitochondria-targeted GFP highlights the heterogeneity of mitochondria 779

shape, size and movement within living plant cells. Journal of Experimental Botany 51: 865-780

871. 781

Lovy-Wheeler A, Cárdenas L, Kunkel JG, Hepler PK. 2007. Differential organelle movement on the 782

actin cytoskeleton in lily pollen tubes. Cell Motility and the Cytoskeleton 64(3): 217-232. 783

Page 22 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

23

Lu L, Lee Y-RJ, Pan R, Maloof JN, Liu B. 2005. An internal motor kinesin is associated with the Golgi 784

apparatus and plays a role in trichome morphogenesis in Arabidopsis. Molecular Biology of 785

the Cell 16(2): 811-823. 786

Madison SL, Buchanan ML, Glass JD, McClain TF, Park E, Nebenführ A. 2015. Class XI myosins move 787

specific organelles in pollen tubes and are required for normal fertility and pollen tube 788

growth in Arabidopsis. Plant Physiology 169(3): 1946-1960. 789

Mano S, Nakamori C, Hayashi M, Kato A, Kondo M, Nishimura M. 2002. Distribution and 790

characterization of peroxisomes in arabidopsis by visualization with GFP: Dynamic 791

morphology and actin- dependent movement. Plant and Cell Physiology 43(3): 331-341. 792

Mathur J, Barton KA, Schattat MH. 2013. Fluorescent protein flow within stromules. Plant Cell 25(8): 793

2771-2772. 794

Mathur J, Mammone A, Barton KA. 2012. Organelle extensions in plant cells. Journal of Integrative 795

Plant Biology 54(11): 851-867. 796

Mathur J, Mathur N, Hülskamp M. 2002. Simultaneous visualization of peroxisomes and cytoskeletal 797

elements reveals actin and not microtubule-based peroxisome motility in plants. Plant 798

Physiology 128(3): 1031-1045. 799

Mehrshahi P, Stefano G, Andaloro JM, Brandizzi F, Froehlich JE, DellaPenna D. 2013. 800

Transorganellar complementation redefines the biochemical continuity of endoplasmic 801

reticulum and chloroplasts. Proceedings of the National Academy of Sciences of the United 802

States of America 110(29): 12126-12131. 803

Nebenführ A, Gallagher LA, Dunahay TG, Frohlick JA, Mazurkiewicz AM, Meehl JB, Staehelin LA. 804

1999. Stop-and-Go movements of plant Golgi stacks are mediated by the acto-myosin 805

system. Plant Physiology 121(4): 1127-1141. 806

Nothnagel EA, Webb. WW. 1982. Hydrodynamic models of viscous coupling between motile myosin 807

and endoplasm in Characean algae. Journal of Cell Biology 94: 444-454. 808

Oikawa K, Kasahara M, Kiyosue T, Kagawa T, Suetsugu N, Takahashi F, Kanegae T, Niwa Y, Kadota 809

A, Wada M. 2003. Chloroplast unusual positioning 1 is essential for proper chloroplast 810

positioning. Plant Cell 15(12): 2805-2815. 811

Oikawa K, Matsunaga S, Mano S, Kondo M, Yamada K, Hayashi H, Kagawa T, Kadota A, Sakamoto 812

H, Higashi S, et al. 2015. Physical interaction between peroxisomes and chloroplasts 813

elucidated by in situ laser analysis. 1(4): 15035. 814

Ojangu EL, Jarve K, Paves H, Truve E. 2007. Arabidopsis thaliana myosin XIK is involved in root hair 815

as well as trichome morphogenesis on stems and leaves. Protoplasma 230(3-4): 193-202. 816

Ojangu EL, Tanner K, Pata P, Jarve K, Holweg C, Truve E, Paves H. 2012. Myosins XI-K, XI-1, and XI-2 817

are required for development of pavement cells, trichomes, and stigmatic papillae in 818

Arabidopsis. BMC Plant Biology 12(81): doi: 10.1186/1471-2229-1112-1181. 819

Osterrieder A, Sparkes IA, Botchway SW, Ward A, Ketelaar T, De Ruijter NCA, Hawes C. 2017. 820

Stacks off tracks: A role for the golgin AtCASP in plant endoplasmic reticulum - Golgi 821

apparatus tethering Journal of Experimental Biology 68(13): 3339-3350. 822

Peremyslov VV, Cole RA, Fowler JE, Dolja VV. 2015. Myosin-powered membrane compartment 823

drives cytoplasmic streaming, cell expansion and plant development. PLOS ONE 10(10): 824

e0139331. 825

Peremyslov VV, Klocko AL, Fowler JE, Dolja VV. 2012. Arabidopsis myosin XI-K localizes to the 826

motile endomembrane vesicles associated with F-actin. Frontiers in Plant Science 3: 184. 827

Peremyslov VV, Mockler TC, Filichkin SA, Fox SE, Jaiswal P, Makarova KS, Koonin EV, Dolja VV. 828

2011. Expression, splicing, and evolution of the myosin gene family in plants. Plant 829

Physiology 155(3): 1191-1204. 830

Peremyslov VV, Morgun EA, Kurth EG, Makarova KS, Koonin EV, Dolja VV. 2013. Identification of 831

myosin XI receptors in Arabidopsis defines a distinct class of transport vesicles. Plant Cell 832

25(8): 3022-3038. 833

Page 23 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

24

Peremyslov VV, Prokhnevsky AI, Avisar D, Dolja VV. 2008. Two class XI myosins function in 834

organelle trafficking and root hair development in Arabidopsis thaliana. Plant Physiology 835

146: 1109-1116. 836

Peremyslov VV, Prokhnevsky AI, Dolja VV. 2010. Class XI myosins are required for development, cell 837

expansion, and F-actin organization in Arabidopsis. Plant Cell 22(6): 1883-1897. 838

Perez-Sancho J, Vanneste S, Lee E, McFarlane HE, Esteban Del Valle A, Valpuesta V, Friml J, Botella 839

MA, Rosado A. 2015. The Arabidopsis synaptotagmin1 is enriched in endoplasmic reticulum-840

plasma membrane contact sites and confers cellular resistance to mechanical stresses. Plant 841

Physiology 168(1): 132-143. 842

Prinz WA. 2014. Bridging the gap: membrane contact sites in signaling, metabolism, and organelle 843

dynamics. Journal of Cell Biology 205(6): 759-769. 844

Prokhnevsky AI, Peremyslov VV, Dolja VV. 2008. Overlapping functions of the four class XI myosins 845

in Arabidopsis growth, root hair elongation and organelle motility. Proceedings of the 846

National Academy of Sciences of the United States of America 105: 19744-19749. 847

Pylypenko O, Attanda W, Gauquelin C, Lahmani M, Coulibaly D, Baron B, Hoos S, Titus MA, 848

England P, Houdusse AM. 2013. Structural basis of myosin V Rab GTPase-dependent cargo 849

recognition. Proceedings of the National Academy of Sciences of the United States of 850

America 110(51): 20443-20448. 851

Reddy AS, Day IS. 2001. Kinesins in the Arabidopsis genome: a comparative analysis among 852

eukaryotes. BMC Genomics 2(1): 2. 853

Ridge RW, Uozumi Y, Plazinski J, Hurley UA, Williamson RE. 1999. Developmental transitions and 854

dynamics of the cortical ER of Arabidopsis cells seen with green fluorescent protein. Plant 855

and Cell Physiology 40(12): 1253-1261. 856

Runions J, Brach T, Kuhner S, Hawes C. 2006. Photoactivation of GFP reveals protein dynamics 857

within the endoplasmic reticulum membrane. Journal of Experimental Botany 57(1): 43-50. 858

Ryan JM, Nebenführ A. 2018. Update on myosin motors: molecular mechanisms and physiological 859

functions. 176(1): 119-127. 860

Sakai T, Kagawa T, Kasahara M, Swartz TE, Christie JM, Briggs WR, Wada M, Okada K. 2001. 861

Arabidopsis nph1 and npl1: Blue light receptors that mediate both phototropism and 862

chloroplast relocation. Proceedings of the National Academy of Sciences of the United States 863

of America 98(12): 6969. 864

Sampathkumar A, Lindeboom JJ, Debolt S, Gutierrez R, Ehrhardt DW, Ketelaar T, Persson S. 2011. 865

Live cell imaging reveals structural associations between the actin and microtubule 866

cytoskeleton in Arabidopsis. Plant Cell 23(6): 2302-2313. 867

Sattarzadeh A, Schmelzer E, Hanson MR. 2011. Analysis of organelle targeting by DIL domains of the 868

Arabidopsis myosin XI family. Frontiers in Plant Science 2: 72. 869

Sattarzadeh A, Schmelzer E, Hanson MR. 2013. Arabidopsis myosin XI sub-domains homologous to 870

the yeast myo2p organelle inheritance sub-domain target subcellular structures in plant 871

cells. Frontiers in Plant Science 4: 407. 872

Schattat M, Barton K, Baudisch B, Klosgen RB, Mathur J. 2011. Plastid stromule branching coincides 873

with contiguous endoplasmic reticulum dynamics. Plant Physiology 155: 1667-1677. 874

Shimmen T. 2007. The sliding theory of cytoplasmic streaming: fifty years of progress. Journal of 875

Plant Research 120(1): 31-43. 876

Sinclair AM, Trobacher CP, Mathur N, Greenwood JS, Mathur J. 2009. Peroxule extension over ER-877

defined paths constitutes a rapid subcellular response to hydroxyl stress. Plant Journal 59: 878

231-242. 879

Sparkes I. 2016. Using optical tweezers to characterize physical tethers at membrane contact sites: 880

grab it, pull it, set it free? Frontiers in Cell and Developmental Biology 4:22 881

Sparkes I, Runions J, Hawes C, Griffing L. 2009. Movement and remodeling of the Endoplasmic 882

Reticulum in nondividing cells of tobacco leaves. Plant Cell 21(12): 3937-3949. 883

Page 24 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

25

Sparkes I, White RR, Coles B, Botchway SW, Ward A. 2018. Using optical tweezers combined with 884

total internal reflection microscopy to study interactions between the ER and Golgi in plant 885

cells. Methods in Molecular Biology 1691: 167-178. 886

Sparkes IA, Ketelaar T, Ruijter NC, Hawes C. 2009. Grab a Golgi: laser trapping of Golgi bodies 887

reveals in vivo interactions with the endoplasmic reticulum. Traffic 10(5): 567-571. 888

Sparkes IA, Teanby NA, Hawes C. 2008. Truncated myosin XI tail fusions inhibit peroxisome, Golgi, 889

and mitochondrial movement in tobacco leaf epidermal cells: a genetic tool for the next 890

generation. Journal of Experimental Botany 59(9): 2499-2512. 891

Stefano G, Renna L, Brandizzi F. 2014. The endoplasmic reticulum exerts control over organelle 892

streaming during cell expansion. Journal of Cell Science 127: 947-953. 893

Steffens A, Jaegle B, Tresch A, Hulskamp M, Jakoby M. 2014. Processing-body movement in 894

Arabidopsis depends on an interaction between myosins and DECAPPING PROTEIN1. Plant 895

Physiology 164(4): 1879-1892. 896

Stephan O, Cottier S, Fahlén S, Montes-Rodriguez A, Sun J, Eklund DM, Klahre U, Kost B. 2014. 897

RISAP is a TGN-associated RAC5 effector regulating membrane traffic during polar cell 898

growth in tobacco. The Plant Cell 26(11): 4426-4447. 899

Suetsugu N, Yamada N, Kagawa T, Yonekura H, Uyeda TQP, Kadota A, Wada M. 2010. Two kinesin-900

like proteins mediate actin-based chloroplast movement in Arabidopsis thaliana. 901

Proceedings of the National Academy of Sciences of the United States of America 107(19): 902

8860-8865. 903

Sun T, Li S, Ren H. 2017. OsFH15, a class I formin, interacts with microfilaments and microtubules to 904

regulate grain size via affecting cell expansion in rice. Scientific Reports 7(1): 6538. 905

Tamura K, Iwabuchi K, Fukao Y, Kondo M, Okamoto K, Ueda H, Nishimura M, Hara-Nishimura I. 906

2013. Myosin XI-i links the nuclear membrane to the cytoskeleton to control nuclear 907

movement and shape in Arabidopsis. Current Biology 23(18): 1776-1781. 908

Tang F, Kauffman EJ, Novak JL, Nau JJ, Catlett NL, Weisman LS. 2003. Regulated degradation of a 909

class V myosin receptor directs movement of the yeast vacuole. Nature 422(6927): 87-92. 910

Thazar-Poulot N, Miquel M, Fobis-Loisy I, Gaude T. 2015. Peroxisome extensions deliver the 911

Arabidopsis SDP1 lipase to oil bodies. Proceedings of the National Academy of Sciences of 912

the United States of America 112(13): 4158-4163. 913

Tominaga M, Ito K. 2015. The molecular mechanism and physiological role of cytoplasmic streaming. 914

Current Opinion in Plant Biology 27: 104-110. 915

Tominaga M, Kimura A, Yokota E, Haraguchi T, Shimmen T, Yamamoto K, Nakano A, Ito K. 2013. 916

Cytoplasmic streaming velocity as a plant size determinant. Developmental Cell 27(3): 345-917

352. 918

Tsuboi H, Suetsugu N, Kawai-Toyooka H, Wada M. 2007. Phototropins and Neochrome1 mediate 919

nuclear movement in the fern Adiantum capillus-veneris. Plant and Cell Physiology 48(6): 920

892-896. 921

Ueda H, Yokota E, Kutsuna N, Shimada T, Tamura K, Shimmen T, Hasezawa S, Dolja VV, Hara-922

Nishimura I. 2010. Myosin-dependent endoplasmic reticulum motility and F-actin 923

organization in plant cells. Proceedings of the National Academy of Sciences of the United 924

States of America 107(15): 6894-6899. 925

Van Gestel K, Köhler RH, Verbelen JP. 2002. Plant mitochondria move on F-actin, but their 926

positioning in the cortical cytoplasm depends on both F-actin and microtubules. Journal of 927

Experimental Botany 53(369): 659-667. 928

Wada M. 2013. Chloroplast movement. Plant Science 210(Supplement C): 177-182. 929

Wada M. 2016. Chloroplast and nuclear photorelocation movements. Proceedings of the Japan 930

Academy. Series B, Physical and Biological Sciences 92(9): 387-411. 931

Wang J, Zhang Y, Wu J, Meng L, Ren H. 2013. AtFH16, an Arabidopsis type II Formin, binds and 932

bundles both microfilaments and microtubules, and preferentially binds to microtubules. 933

Journal of Integrative Plant Biology 55(11): 1002-1015. 934

Page 25 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

26

Wang P, Hawes C, Hussey PJ. 2017. Plant endoplasmic reticulum-plasma membrane contact sites. 935

Trends in Plant Science 22(4): 289-297. 936

Wang P, Hawkins TJ, Richardson C, Cummins I, Deeks MJ, Sparkes I, Hawes C, Hussey PJ. 2014. The 937

plant cytoskeleton, NET3C, and VAP27 mediate the link between the plasma membrane and 938

endoplasmic reticulum. Current Biology 24(12): 1397-1405. 939

Wang P, Hussey PJ. 2017. NETWORKED 3B: a novel protein in the actin cytoskeleton-endoplasmic 940

reticulum interaction. Journal of Experimental Botany. 68(7):1441-1450 941

Wang P, Hussey PJ. 2015. Interactions between plant endomembrane systems and the actin 942

cytoskeleton. Frontiers in Plant Science 6: 422. 943

Wang P, Richardson C, Hawkins TJ, Sparkes I, Hawes C, Hussey PJ. 2016. Plant VAP27 proteins: 944

domain characterization, intracellular localization and role in plant development. New 945

Phytologist 210(4): 1311-1326. 946

Wei L, Zhang W, Liu Z, Li Y. 2009. AtKinesin-13A is located on Golgi-associated vesicle and involved 947

in vesicle formation/budding in Arabidopsis root-cap peripheral cells. BMC Plant Biology 948

9(1): 138. 949

Woodhouse FG, Goldstein RE. 2013. Cytoplasmic streaming in plant cells emerges naturally by 950

microfilament self-organization. Proceedings of the National Academy of Science U S A. 951

110(35):14132-7. 952

Wu S-Z, Bezanilla M. 2014. Myosin VIII associates with microtubule ends and together with actin 953

plays a role in guiding plant cell division. eLife 3: e03498. 954

Yan Q, Sun W, Kujala P, Lotfi Y, Vida TA, Bean AJ. 2005. CART: an Hrs/actinin-4/BERP/myosin V 955

protein complex required for efficient receptor recycling. Molecular Biology of the Cell 16(5): 956

2470-2482. 957

958

Figure legends 959

960

Figure 1. Time lapse images showing ER and Golgi dynamics in A. thaliana leaf epidermal cell. 961

The position of Golgi (magenta) and ER (green) at timepoints 0s (a), 15s (b) and 30s (c) are indicated. 962

Blue (e, i), yellow (f, j) and magenta (g,k) panels indicate images of ER (i, j, k) and Golgi (e,f,g) taken 963

at timepoints of 0, 15 and 30 seconds respectively. Overlays of all the timepoints for ER (l) and Golgi 964

(h) indicate regions of slow movement (white) versus highly dynamic movement. White arrow (h) 965

indicates a static Golgi body. White box in panel h is magnified in panel d. Magenta outline of Golgi 966

bodies (d) indicates 0s position and subsequent movement is shown by coloured line tracks . Scale 967

bar 5 µm except for panel (d) which is 1µm Image capture rate was 0.4 frames per second. 968

969

Figure 2. Schematic highlighting myosin recruitment and organelle movement 970

Diagram depicting XI-I recruitment and interaction with components at the outer and inner nuclear 971

envelope (a). Scheme adapted from Tamura et al. (2013). 972

Possible roles for the MyoB/MadA-B decorated vesicles in myosin recruitment and organelle 973

dynamics (b,c,d). MyoB/MadA-B proteins could be located on organelle membranes and mediate 974

the recruitment of specific myosins to the organelle membrane. Said organelle then moves actively 975

Page 26 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

27

along the actin cytoskeleton (b). MyoB vesicles have been shown to move faster than any other 976

organelle. Therefore, as suggested by Peremyslov et al. (2015), MyoB/MadA-B decorated vesicles 977

could be the only structures actively moving on the actin through myosin motors, and able to 978

generate cytoplasmic streaming. Other organelles possibly follow the flow passively (c). MyoB 979

decorated vesicles might be located in proximity, and interact, with certain sub-domains of known 980

organelles, dragging the organelle along the actin filaments in a myosin dependent way (d). Given 981

the size of MyoB and myosin gene families, models are not necessarily mutually exclusive and so a 982

more complex picture of coexistence of several of these options could occur (b, c, d). 983

984

985

Figure 3. Peroxisome-chloroplast interaction in plants 986

Peroxisome (p) juxtaposed to a chloroplast (cp) are depicted prior to optical trapping (a). Peroxisome 987

is trapped and pulled away from the chloroplast (b) inducing a peroxule (white arrowhead) which 988

appears to maintain contact between the two organelles. Upon releasing the peroxisome from the 989

trap (c) it ‘springs’ back towards its original position (a) along with a resultant retraction in the 990

peroxule. Peroxules therefore appear to be membrane manifestations of membrane contact sites 991

between the two organelles. Scale bar 5µm. 992

993

Figure 4. Models proposed for the coordinated action of myosin dependent processes and 994

organelle tethering on subsequent organelle movement and positioning. 995

Organelles may not be tethered and will therefore recruit myosins and move independently from 996

one another in the cell cytoplasm (a). Note, this links to Figure 2. In the case where organelle A is 997

tethered to organelle B (red and black hooks) movement could be controlled via the mechanisms 998

depicted in panels b-d. Organelle A’s movement is myosin dependent which in turn drags tethered 999

organelle B; organelle B is therefore ‘piggybacking’ on organelle A (b). In (c) and (d) both tethered 1000

organelles recruit specific myosins on their membrane and the direction in which the organelles will 1001

move depends on the polarity of actin, and on the resulting forces derived by myosins moving along 1002

the actin filaments. Here, coordinated regulation of myosins on organelle A and B is required to 1003

allow simultaneous movement of both organelles to maintain tethering (c). Tethering between 1004

organelles may involve multiple organelles rather than just two organelles (d). Here, similar to that 1005

depicted in (c), movement is coordinated between three organelles rather than just two. There is 1006

speculation that the ER may act as a bridge between certain organelles. If the rate of movement of 1007

tethered organelles is not coordinated (whereby one moves faster than the other or they move in 1008

opposite directions, as indicated by red arrows in panels c and d), then contacts maybe maintained 1009

Page 27 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

28

through membrane extensions such as peroxules, stromules and matrixules. Note, upper actin 1010

filament in c and d reflects potential for either orientation of filament polarity, with arrows for 1011

organelle movement referring to plus end directed motion. 1012

1013

1014

1015

1016

Page 28 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

Figure 1. Time lapse images showing ER and Golgi dynamics in A. thaliana leaf epidermal cell. The position of Golgi (magenta) and ER (green) at timepoints 0s (a), 15s (b) and 30s (c) are indicated. Blue

(e, i), yellow (f, j) and magenta (g,k) panels indicate images of ER (i, j, k) and Golgi (e,f,g) taken at timepoints of 0, 15 and 30 seconds respectively. Overlays of all the timepoints for ER (l) and Golgi (h)

indicate regions of slow movement (white) versus highly dynamic movement. White arrow (h) indicates a static Golgi body. White box in panel h is magnified in panel d. Magenta outline of Golgi bodies (d) indicates 0s position and subsequent movement is shown by coloured line tracks . Scale bar 5 µm except for panel (d)

which is 1µm Image capture rate was 0.4 frames per second.

199x155mm (300 x 300 DPI)

Page 29 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

Figure 2. Schematic highlighting myosin recruitment and organelle movement Diagram depicting XI-I recruitment and interaction with components at the outer and inner nuclear envelope

(a). Scheme adapted from Tamura et al. (2013).

Possible roles for the MyoB/MadA-B decorated vesicles in myosin recruitment and organelle dynamics (b,c,d). MyoB/MadA-B proteins could be located on organelle membranes and mediate the recruitment of

specific myosins to the organelle membrane. Said organelle then moves actively along the actin cytoskeleton (b). MyoB vesicles have been shown to move faster than any other organelle. Therefore, as suggested by Peremyslov et al. (2015), MyoB/MadA-B decorated vesicles could be the only structures

actively moving on the actin through myosin motors, and able to generate cytoplasmic streaming. Other organelles possibly follow the flow passively (c). MyoB decorated vesicles might be located in proximity, and interact, with certain sub-domains of known organelles, dragging the organelle along the actin filaments in a myosin dependent way (d). Given the size of MyoB and myosin gene families, models are not necessarily

mutually exclusive and so a more complex picture of coexistence of several of these options could occur (b, c, d).

199x76mm (300 x 300 DPI)

Page 30 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

Figure 3. Peroxisome-chloroplast interaction in plants Peroxisome (p) juxtaposed to a chloroplast (cp) are depicted prior to optical trapping (a). Peroxisome is

trapped and pulled away from the chloroplast (b) inducing a peroxule (white arrowhead) which appears to

maintain contact between the two organelles. Upon releasing the peroxisome from the trap (c) it ‘springs’ back towards its original position (a) along with a resultant retraction in the peroxule. Peroxules therefore appear to be membrane manifestations of membrane contact sites between the two organelles. Scale bar

5microns.

150x74mm (300 x 300 DPI)

Page 31 of 32

Manuscript submitted to New Phytologist for review

For Peer Review

Figure 4. Models proposed for the coordinated action of myosin dependent processes and organelle tethering on subsequent organelle movement and positioning.

Organelles may not be tethered and will therefore recruit myosins and move independently from one

another in the cell cytoplasm (a). Note, this links to Figure 2. In the case where organelle A is tethered to organelle B (red and black hooks) movement could be controlled via the mechanisms depicted in panels b-d.

Organelle A’s movement is myosin dependent which in turn drags tethered organelle B; organelle B is therefore ‘piggybacking’ on organelle A (b). In (c) and (d) both tethered organelles recruit specific myosins on their membrane and the direction in which the organelles will move depends on the polarity of actin, and on the resulting forces derived by myosins moving along the actin filaments. Here, coordinated regulation of

myosins on organelle A and B is required to allow simultaneous movement of both organelles to maintain tethering (c). Tethering between organelles may involve multiple organelles rather than just two organelles (d). Here, similar to that depicted in (c), movement is coordinated between three organelles rather than just

two. There is speculation that the ER may act as a bridge between certain organelles. If the rate of movement of tethered organelles is not coordinated (whereby one moves faster than the other or they move

in opposite directions, as indicated by red arrows in panels c and d), then contacts maybe maintained through membrane extensions such as peroxules, stromules and matrixules. Note, upper actin filament in c

and d reflects potential for either orientation of filament polarity, with arrows for organelle movement referring to plus end directed motion.

244x210mm (300 x 300 DPI)

Page 32 of 32

Manuscript submitted to New Phytologist for review