Xyloglucan constitutes most of the hemicellulose in eudicot primary cell walls and functions in cell wall structure and mechanics. Although Arabidopsis (
Arabidopsis thaliana)
xxt1 xxt2 mutants lacking detectable xyloglucan are viable, they display growth defects that are suggestive of alterations in wall integrity. To probe the mechanisms underlying these defects, we analyzed cellulose arrangement, microtubule patterning and dynamics, microtubule- and wall-integrity-related gene expression, and cellulose biosynthesis in
xxt1 xxt2 plants. We found that cellulose is highly aligned in
xxt1 xxt2 cell walls, that its three-dimensional distribution is altered, and that microtubule patterning and stability are aberrant in etiolated
xxt1 xxt2 hypocotyls. We also found that the expression levels of microtubule-associated genes, such as
MAP70-5 and
CLASP, and receptor genes, such as
HERK1 and
WAK1, were changed in
xxt1 xxt2 plants and that cellulose synthase motility is reduced in
xxt1 xxt2 cells, corresponding with a reduction in cellulose content. Our results indicate that loss of xyloglucan affects both the stability of the microtubule cytoskeleton and the production and patterning of cellulose in primary cell walls. These findings establish, to our knowledge, new links between wall integrity, cytoskeletal dynamics, and wall synthesis in the regulation of plant morphogenesis.The primary walls of growing plant cells are largely constructed of cellulose and noncellulosic matrix polysaccharides that include hemicelluloses and pectins (
Carpita and Gibeaut, 1993;
Somerville et al., 2004;
Cosgrove, 2005). Xyloglucan (
XyG) is the most abundant hemicellulose in the primary walls of eudicots and is composed of a β-1,4-glucan backbone with side chains containing Xyl, Gal, and Fuc (
Park and Cosgrove, 2015).
XyG is synthesized in the Golgi apparatus before being secreted to the apoplast, and its biosynthesis requires several glycosyltransferases, including β-1,4-glucosyltransferase, α-1,6-xylosyltransferase, β-1,2-galactosyltransferase, and α-1,2-fucosyltransferase activities (
Zabotina, 2012). Arabidopsis (
Arabidopsis thaliana) XYLOGLUCAN XYLOSYLTRANSFERASE1 (XXT1) and XXT2 display xylosyltransferase activity in vitro (
Faik et al., 2002;
Cavalier and Keegstra, 2006), and strikingly, no
XyG is detectable in the walls of
xxt1 xxt2 double mutants (
Cavalier et al., 2008;
Park and Cosgrove, 2012a), suggesting that the activity of XXT1 and XXT2 are required for
XyG synthesis, delivery, and/or stability.Much attention has been paid to the interactions between cellulose and
XyG over the past 40 years. Currently, there are several hypotheses concerning the nature of these interactions (
Park and Cosgrove, 2015). One possibility is that
XyGs bind directly to cellulose microfibrils (
CMFs). Recent data indicating that crystalline cellulose cores are surrounded with hemicelluloses support this hypothesis (
Dick-Pérez et al., 2011). It is also possible that
XyG acts as a spacer-molecule to prevent
CMFs from aggregating in cell walls (
Anderson et al., 2010) or as an adapter to link cellulose with other cell wall components, such as pectin (
Cosgrove, 2005;
Cavalier et al., 2008).
XyG can be covalently linked to pectin (
Thompson and Fry, 2000;
Popper and Fry, 2005,
2008), and NMR data demonstrate that pectins and cellulose might interact to a greater extent than
XyG and cellulose in native walls (
Dick-Pérez et al., 2011). Alternative models exist for how
XyG-cellulose interactions influence primary wall architecture and mechanics. One such model posits that
XyG chains act as load-bearing tethers that bind to
CMFs in primary cell walls to form a cellulose-
XyG network (
Carpita and Gibeaut, 1993;
Pauly et al., 1999;
Somerville et al., 2004;
Cosgrove, 2005). However, results have been accumulating against this tethered network model, leading to an alternative model in which
CMFs make direct contact, in some cases mediated by a monolayer of xyloglucan, at limited cell wall sites dubbed “biomechanical hotspots,” which are envisioned as the key sites of cell wall loosening during cell growth (
Park and Cosgrove, 2012a;
Wang et al., 2013;
Park and Cosgrove, 2015). Further molecular, biochemical, and microscopy experiments are required to help distinguish which aspects of the load-bearing, spacer/plasticizer, and/or hotspot models most accurately describe the functions of
XyG in primary walls.Cortical microtubules (
MTs) direct CMF deposition by guiding cellulose synthase complexes in the plasma membrane (
Baskin et al., 2004;
Paredez et al., 2006;
Emons et al., 2007;
Sánchez-Rodriguez et al., 2012), and the patterned deposition of cellulose in the wall in turn can help determine plant cell anisotropic growth and morphogenesis (
Baskin, 2005). Disruption of cortical
MTs by oryzalin, a
MT-depolymerizing drug, alters the alignment of
CMFs, suggesting that
MTs contribute to
CMF organization (
Baskin et al., 2004). CELLULOSE SYNTHASE (CESA) genes, including CESA1, CESA3, and CESA6, are required for normal CMF synthesis in primary cell walls (
Kohorn et al., 2006;
Desprez et al., 2007), and accessory proteins such as COBRA function in cellulose production (
Lally et al., 2001). Live-cell imaging from double-labeled YFP-CESA6; CFP-ALPHA-1 TUBULIN (TUA1) Arabidopsis seedlings provides direct evidence that cortical
MTs determine the trajectories of cellulose synthesis complexes (
CSCs) and patterns of cellulose deposition (
Paredez et al., 2006). Additionally,
MT organization affects the rotation of cellulose synthase trajectories in the epidermal cells of Arabidopsis hypocotyls (
Chan et al., 2010). Recently, additional evidence for direct guidance of
CSCs by
MTs has been provided by the identification of CSI1/POM2, which binds to both
MTs and CESAs (
Bringmann et al., 2012;
Li et al., 2012). MICROTUBULE ORGANIZATION1 (MOR1) is essential for cortical
MT organization (
Whittington et al., 2001), but disruption of cortical
MTs in the
mor1 mutant does not greatly affect
CMF organization (
Sugimoto et al., 2003), and oryzalin treatment does not abolish
CSC motility (
Paredez et al., 2006).Conversely, the organization of cortical
MTs can be affected by cellulose synthesis. Treatment with isoxaben, a cellulose synthesis inhibitor, results in disorganized cortical
MTs in tobacco cells, suggesting that inhibition of cellulose synthesis affects
MT organization (
Fisher and Cyr, 1998), and treatment with 2,6-dichlorobenzonitrile, another cellulose synthesis inhibitor, alters MT organization in
mor1 plants (
Himmelspach et al., 2003). Cortical
MT orientation in Arabidopsis roots is also altered in two cellulose synthesis-deficient mutants,
CESA652-isx and
kor1-3, suggesting that
CSC activity can affect
MT arrays (
Paredez et al., 2008). Together, these results point to a bidirectional relationship between cellulose synthesis/patterning and
MT organization.
MTs influence plant organ morphology, but the detailed mechanisms by which they do so are incompletely understood. The dynamics and stability of cortical
MTs are also affected by
MT-associated proteins (
MAPs). MAP18 is a
MT destabilizing protein that depolymerizes
MTs (
Wang et al., 2007), MAP65-1 functions as a
MT crosslinker, and MAP70-1 functions in
MT assembly (
Korolev et al., 2005;
Lucas et al., 2011). MAP70-5 stabilizes existing
MTs to maintain their length, and its overexpression induces right-handed helical growth (
Korolev et al., 2007); likewise, MAP20 overexpression results in helical cell twisting (
Rajangam et al., 2008). CLASP promotes microtubule stability, and its mutant is hypersensitive to microtubule-destabilizing drug oryzalin (
Ambrose et al., 2007). KATANIN1 (KTN1) is a
MT-severing protein that can sever
MTs into short fragments and promote the formation of thick
MT bundles that ultimately depolymerize (
Stoppin-Mellet et al., 2006), and loss of
KTN1 function results in reduced responses to mechanical stress (
Uyttewaal et al., 2012). In general, cortical
MT orientation responds to mechanical signals and can be altered by applying force directly to the shoot apical meristem (
Hamant et al., 2008). The application of external mechanical pressure to Arabidopsis leaves also triggers
MT bundling (
Jacques et al., 2013). Kinesins, including KINESIN-13A (KIN-13A) and FRAGILE FIBER1 (FRA1), have been implicated in cell wall synthesis (
Cheung and Wu, 2011;
Fujikura et al., 2014). The identification of cell wall receptors and sensors is beginning to reveal how plant cell walls sense and respond to external signals (
Humphrey et al., 2007;
Ringli, 2010); some of them, such as FEI1, FEI2, THESEUS1 (THE1), FERONIA (FER), HERCULES RECEPTOR KINASE1 (HERK1), WALL ASSOCIATED KINASE1 (WAK1), WAK2, and WAK4, have been characterized (
Lally et al., 2001;
Decreux and Messiaen, 2005;
Kohorn et al., 2006;
Xu et al., 2008;
Guo et al., 2009;
Cheung and Wu, 2011). However, the relationships between wall integrity, cytoskeletal dynamics, and wall synthesis have not yet been fully elucidated.In this study, we analyzed
CMF patterning,
MT patterning and dynamics, and cellulose biosynthesis in the Arabidopsis
xxt1 xxt2 double mutant that lacks detectable
XyG and displays altered growth (
Cavalier et al., 2008;
Park and Cosgrove, 2012a). To investigate whether and how
XyG deficiency affects the organization of
CMFs and cortical
MTs, we observed CMF patterning in
xxt1 xxt2 mutants and Col (wild-type) controls using atomic force microscopy (
AFM), field emission scanning electron microscopy (
FESEM), transmission electron microscopy (
TEM), and confocal microscopy (
Hodick and Kutschera, 1992;
Derbyshire et al., 2007;
Anderson et al., 2010;
Zhang et al., 2014). We also generated transgenic Col and
xxt1 xxt2 lines expressing GFP-MAP4 (
Marc et al., 1998) and GFP-CESA3 (
Desprez et al., 2007), and analyzed
MT arrays and cellulose synthesis using live-cell imaging. Our results show that the organization of
CMFs is altered, that
MTs in
xxt1 xxt2 mutants are aberrantly organized and are more sensitive to external mechanical pressure and the
MT-depolymerizing drug oryzalin, and that cellulose synthase motility and cellulose content are decreased in
xxt1 xxt2 mutants. Furthermore, real-time quantitative RT-PCR measurements indicate that the enhanced sensitivity of cortical
MTs to mechanical stress and oryzalin in
xxt1 xxt2 plants might be due to altered expression of
MT-stabilizing and wall receptor genes. Together, these data provide insights into the connections between the functions of
XyG in wall assembly, the mechanical integrity of the cell wall, cytoskeleton-mediated cellular responses to deficiencies in wall biosynthesis, and cell and tissue morphogenesis.
相似文献