The discovery that myosins XI contribute to the growth of plant cells in both Arabidopsis
and moss initiated a quest for an underlying mechanism (Peremyslov et al., 2010
; Vidali et al., 2010
). Here we propose that this mechanism is based on myosin-driven traffic through the secretory pathway to the plasma membrane, and provide first experimental support for this model. We also posit that vesicle transport is required for homeostasis in mature cells.
It stands to reason that the bulk of the cellular myosin “motor pool” is engaged in transporting cargoes that are most important for cell growth and function, but defining such cargoes has been an elusive goal. Since 2001, studies have reported, in addition to dispersal in the cytosol, myosin XI association with mitochondria, plastids, peroxisomes, ER, Golgi, nuclear envelope, plasma membrane, unidentified organelles, and vesicles (Liu et al., 2001
; Hashimoto et al., 2005
; Li and Nebenfuhr, 2007
; Sparkes et al., 2008
; Sattarzadeh et al., 2011
). The disparate conclusions drawn from these studies are difficult to reconcile; perhaps, the only semblance of consensus in this field was apparent association of myosins with larger organelles such as mitochondria, peroxisomes, and Golgi.
In contrast, this study demonstrated the predominant association of a functional myosin XI-K:YFP with endomembrane vesicles transported along F-actin bundles. The first line of evidence was provided by imaging, which showed XI-K:YFP primarily decorating vesicular bodies rather than larger organelles (Figure A). Quantitative analysis confirmed that less than 7% of the myosin XI-K:YFP was associated with the Golgi (Table S1
in Supplementary Material).
Because ~90% of the XI-K:YFP co-localized with a subset of the ER (Figure D; Table S1
in Supplementary Material), it could be proposed that most of this myosin is engaged in driving ER flow. However, ER is not necessarily the principal myosin XI-K cargo. The concentration of both ER membranes and myosin along F-actin bundles (Figures A,B,D) may imply either true physical interaction, or independent association of each with F-actin. The fine structure of the ER sheets and strands compared to XI-K:YFP-decorated bead-like structures (Figures D,E) appears to support the latter possibility.
A striking pattern of myosin localization was revealed in growing root hairs, where the bulk of XI-K:YFP was localized to the cell tip (Figures A,C), from which F-actin bundles are excluded (Baluska et al., 2000
; Peremyslov et al., 2010
). This polarized localization was similar to that of the secretory vesicle markers SCAMP2 and RabA4b, consistent with extensive, myosin-driven vesicle delivery to the tip. The polarization of the myosin and vesicular markers was lost in mature root hairs (Figures C–E).
To determine the role of myosins XI in the trafficking of SCAMP2-YFP-tagged secretory vesicles, we assessed marker dynamics in root hairs of the 3KO mutant in which myosin XI-K and two other highly expressed myosins were inactivated. Instead of the typical rapid transport, shorter mutant hairs exhibited both reduced vesicle movement and slower recovery of fluorescent signal at the tip following photobleaching (Figure G; Movies S6
in Supplementary Material). These data support a role for myosin XI-K in driving the long-distance vesicle transport toward the tip, and potentially the short-distance transport of vesicles to the plasma membrane. In addition, these data provide a second line of evidence for vesicles being a myosin XI-K cargo. It should be noted, however, that our data do not distinguish between a direct or indirect role for myosin in the transport of SCAMP2-YFP-tagged vesicles. It is equally feasible that these vesicles are directly bound to myosin XI-K, or that their transport occurs via cytosolic flow generated from direct myosin transport of an alternative vesicle type.
Interestingly, moss myosin XI is also concentrated in the tips of growing protonemal cells as shown in an elegant recent work by Vidali et al. (2010
). The evolutionary roots of a role for myosin in polarized growth appear to be very deep, given myosin V involvement in polarization of budding yeast cells and the ancient origin of the myosin V/XI class (Bretcher, 2003
; Richards and Cavalier-Smith, 2005
The third line of evidence compatible with vesicle transport as the primary myosin XI-K function comes from membrane fractionation analysis. Unexpectedly, virtually all of the myosin XI-K was associated with membranes (Figure A). Therefore, most of this myosin is in active, cargo-bound form, making an important distinction with vertebrate myosin Va, which adopts an inactive conformation upon cargo detachment (Krementsov et al., 2004
). A tight engagement of myosin with the endomembranes may in part explain the more extensive cytoplasmic dynamics in plant cells compared to that in other eukaryotes.
Perhaps the most pressing question posed by this study is the identity of the vesicles transported by myosin. On the one hand, the high degree of the overlap between XI-K:YFP and the ER suggests that the myosin-decorated vesicles are ER-derived. On the other hand, this result does not exclude myosin association with post-Golgi secretory vesicles, given that the distributions of the vesicle markers YFP-RabA4b and SCAMP2 overlap with that of the myosin XI-K. Furthermore, fractionation of root extracts showed clear separation between ER and myosin peaks (Figures D,E) and correspondence of the myosin peak with those of exocyst component, likely to be associated with exocytic vesicles (Hála et al., 2008
Two working models could be proposed to account for myosin-driven vesicle trafficking in plant cells. One model posits that vesicles are mobilized via interaction between vesicle type-specific myosin receptors and cognate myosins XI. Among the strongest candidate receptors are Rabs, which interact with myosins V in yeast and vertebrates, and are thought to be important determinants of vesicle type identity (Woollard and Moore, 2008
). The 57 Rabs encoded by the Arabidopsis
genome provide a vast potential resource for regulation of myosin-dependent vesicle transport. It is important to emphasize that although transport of SCAMP2 secretory vesicles requires myosin, only a fraction of these vesicles co-localizes with myosin. The most parsimonious hypothesis explaining these data posits that the myosin-associated vesicle pool is composed of multiple types of vesicles. In other words, a novel class of motile vesicles, comprised of subpopulations of vesicle types defined by vesicle-specific markers (such as Rabs or SCAMP2), appears to be a distinct possibility. A critical proof of this hypothesis could be provided by identification of the vesicular receptors responsible for direct myosin binding, therefore defining this novel class of vesicles.
An alternative model proposes that only one or a few vesicle types are the bona fide myosin cargoes, whereas other vesicles and organelles move passively, with cytosolic flow. Distinguishing between these models will also require identification of the myosin receptors, as well as the primary myosin cargoes. Correspondingly, characterization of myosin receptors, as well as the composition of the associated motile, myosin-decorated vesicles, will be a current focus of research in the plant myosin field.
Another question prompted by this work are the discrepancies between this and previous studies regarding myosin localization. Several of these studies used immunocytochemical approach and antibodies of variable, not always well defined specificity (Liu et al., 2001
; Wang and Pesacreta, 2004
; Hashimoto et al., 2005
; Romagnoli et al., 2007
; Yokota et al., 2009
). The outcomes of corresponding analyses varied from cytosolic co-localization with molecular chaperone TCP-1α to mitochondria to peroxisomes to ER, and are often mutually exclusive. On the other hand, many of the recent studies involved transient expression of the various, fluorophore-tagged myosin fragments rather than full-length proteins (Li and Nebenfuhr, 2007
; Reisen and Hanson, 2007
; Sparkes et al., 2008
; Avisar et al., 2009
; Natesan et al., 2009
; Sattarzadeh et al., 2009
). Even within the same experiment, this approach could result in distinct localization patterns among different cells, likely due to variation in protein accumulation levels (Li and Nebenfuhr, 2007
; Sattarzadeh et al., 2011
). Even more alarmingly, localization patterns of these myosin fragments do not correspond to functional effects caused by their expression (Avisar et al., 2012
). These patterns were further affected by the domain structure of the expressed myosin fragments, which may in turn determine the protein’s ability to interact with appropriate cargoes. Because our work uses a full-length, functional myosin whose expression is driven by its native regulatory elements, the authenticity of the resulting localization pattern is much more certain. This notion is further supported by very similar localization patterns described for a full-size, functional, fluorophore-tagged myosin XI in Arabidopsis
root hairs (this work) and protonemal moss cells that share the polar cell growth mechanism (Vidali et al., 2010
In conclusion, we propose that the principal function of myosin XI-K is trafficking of vesicles through the endomembrane system, including delivery of secretory vesicles to areas of cell growth. We also suggest that the myosin XI-K-associated vesicles encompass subpopulations of the distinct types of vesicles mobilized via attachment of myosins to cognate vesicular receptors. This model is in accord with the data presented herein, and with previous genetic analysis demonstrating arrest of polarized root hair elongation and reduction in the diffuse cell growth in myosin-deficient plants (Peremyslov et al., 2010
). In mature cells, myosin-driven vesicle transport could be required for cell homeostasis, e.g., delivery of plasma membrane proteins. In agreement with this, myosins XI were recently implicated in the steady-state transport of certain integral membrane proteins (Amari et al., 2011
). The approach described here for the myosin XI-K provides a clear path to investigating localization patterns and functions of the other flowering plant myosins.