TY - JOUR
AU - Newbigin, Edward J.
AB - Abstract The elongation (elo) mutants of barley (Hordeum vulgare cv 'Himalaya') are a class of dwarf plants with defects affecting cell expansion. The phenotypes of mutants in three of the elo loci (elo1, elo2 and elo3) are recessive to the wild-type allele, and the mutations at elo-4 and elo-5 are semi-dominant. Allelism tests showed that elo1, elo2 and elo3 were at separate loci, and mapping data indicated that elo-5 was possibly allelic to either elo1 or elo2. A phenotype common to all elo mutants was the presence of short, radially swollen cells on the leaf epidermis, indicating a defect in longitudinal cell expansion. In three of the mutants, elo1, elo3 and elo5, this was accompanied by a twisting growth habit. Two of the mutations, elo2 and elo-5, affected cell division, with aberrant periclinal cell division resulting in the formation of increased cell layers in the leaf epidermis of elo2 and elo-5 homozygotes and in the aleurone layer of elo2 grains. Misplaced anticlinal divisions also occurred in the elo-5 leaf epidermis. Leaf cell walls of all elo lines contained less cellulose than the wild- type, and the cortical microtubules in elongating root epidermal cells in some elo lines were more randomly oriented than in the wild-type, consistent with the presence of radially swollen cells. We discuss possible functions for the Elo genes in primary cell wall synthesis. Introduction Cell expansion is fundamental to plant growth and morphogenesis, and requires the co-ordination of positional, hormonal and genetic cues to tailor the growth and development of cells to their precise functional requirements (Hülskamp et al. 1998). Several factors have been identified that influence the way in which plant cells expand. Especially important is the plant cell wall, which controls growth at a cellular level by influencing the cell's shape and size. Cell expansion requires both loosening of the cell wall, and synthesis and insertion of new cell wall material into the wall matrix (Taiz 1984, Cosgrove 1997). Plant cell walls are formed from a complex of polysaccharides, proteins and phenolic compounds that are progressively incorporated into or removed from the wall as the cell grows (Carpita and McCann 2000). Two main types of primary cell wall have been identified that differ in their structural constituents: dicotyledons, gymnosperms and most monocotyledons possess type I cell walls, while grasses and their close relatives (commelinoid monocotyledons) possess type II cell walls (Bacic et al. 1988, Smith and Harris 1999, Carpita and McCann 2000). Cellulose microfibrils are the chief load-bearing component of both type I and type II cell walls; however, significant differences exist in the types and quantities of cross-linking glycans, pectins and structural proteins. In type I walls, microfibrils are cross-linked primarily by xyloglucans (XGs), and the cellulose–XG scaffold is embedded in a pectic gel matrix, consisting of homogalacturonans (HGs) and the rhamnogalacturonans I and II (RGI and RGII), which controls wall porosity (Carpita and Gibeaut 1993). Cellulose microfibrils in type II walls are cross-linked by heteroxylans, such as glucuronoarabinoxylans (GAXs), and mixed linkage’ (1→3), (1→4)-β-d-glucans (MLGs). Lesser amounts of pectin, XGs and structural proteins are present in type II cell walls than are in type I walls (Bacic et al. 1988, Carpita and Gibeaut 1993). Genetic studies of plants with aberrant cell expansion phenotypes, such as dwarfing and radial swollen cells, have led to the identification of genes for polysaccharide synthases and hydrolases (Arioli et al. 1998, Fagard et al. 2000, Zuo et al. 2000), as well as a number of novel proteins of unknown function that are implicated in cell wall synthesis (Schindelman et al. 2001, Pagant et al. 2002). Most of these studies, however, have been carried out on species with a type I cell wall, such as Arabidopsis. While these studies have resulted in considerable progress in our understanding of the process of cell wall expansion, the numerous differences in wall structure and composition between type I and type II walls mean that the details of this process are likely to be very different in the agriculturally important grasses. We are using a class of dwarf mutants of barley (Hordeum vulgare L.) to study the process of cell expansion in a species with type II cell walls (Bacic et al. 1988, Smith and Harris 1999). The elongation (elo) mutants were first described by Chandler and Robertson (1999) in a screen for dwarf mutants of barley. These authors identified three classes of dwarf mutants based on plant responses to gibberellic acid (GA3). The gibberellin (GA)-responsive dwarf (grd) and GA-sensitivity (gse) classes consisted of dwarf mutants that responded to low and high concentrations of GA3, respectively. The grd mutants are now known to be GA biosynthetic mutants (Helliwell et al. 2001), and the gse1 mutants have mutations in the GA receptor (Chandler et al. 2008). The elo mutants comprised the third class of dwarf mutants, and initially mutants in just two loci, elo1 and elo2, were identified The elo mutants were distinct from the grd and gse mutants because elo seedlings did not respond to GA3 at any concentration. Furthermore, plants grew no taller when the elo mutation was moved into a slender1 background, a mutation that would normally give a constitutive GA response phenotype (Chandler and Robertson 1999). The elo mutants are, however, still able to respond to GA3 under certain conditions, as treating seedlings with the GA biosynthesis inhibitor tetcyclasis resulted in further dwarfing that could be reversed by applying exogenous GA3. Because the elo mutants could respond to GA3, some other factor must have been responsible for the inhibition of cell elongation in these plants (Chandler and Robertson 1999). However, none of the other plant hormones tested (brassinolide, IAA and kinetin) could overcome dwarfing in elo seedlings (Chandler and Robertson 1999). Here we extended the study of Chandler and Robertson (1999) to include three additional dwarf lines (elo3, elo-4 and elo-5). The five elo lines are phenotypically similar to each other and share features such as the presence of bulging and distorted cells on the leaf epidermis. Analysis of elo mutant morphology and cell wall composition, and cortical microtubule (MT) orientation in elo3 and elo-5, together suggests that the elo mutants define genes whose products are required for primary cell wall synthesis and/or deposition in a type II primary cell wall. Results The barley elo mutants Representative lines from the initial collection of 'elongation' (elo) mutants were intercrossed, and three genetic loci (elo1, 2 and 3) were defined. Mutants at each of these loci have a recessive phenotype. When intercrossed, mutants in the same locus produce dwarf F1 plants. Intercrosses between the three recessive elo mutants produced F1 plants that were tall, although sometimes not as tall as the wild-type. In the F2 generations, however, the fact that there was a proportion of fully tall plants observed in each case confirmed that the mutants are in different loci. It is likely that the slightly diminished growth of the F1 plants was due to smaller grains produced by dwarf mutants, and to heterozygosity at two loci affecting growth. There were also two mutants, designated elo-4 and elo-5, that could only be maintained as segregating populations, since the homozygotes did not produce grain. Because they were phenotypically distinct from mutants at elo1, 2 and 3, and from each other, they were assumed to represent new genetic loci, but crossing studies have not been attempted due to a lack of markers and unpredictable phenotypes of the double mutants. F2 mapping populations were generated to determine genetic locations for four of the five elo mutants (elo1–elo3 and elo-5). These populations were produced by crossing an elo plant, which is in the 'Himalaya' background, with the cultivar 'Sloop'. Bulk segregant analysis (Michelmore et al. 1991, Collard et al. 2005) using simple sequence repeat (SSR) markers was performed to identify markers potentially linked to the elo genes. These markers were amplified from F2 populations, scored and used to calculate the locations of the elo genes. elo1, elo2 and elo-5 were all on the long arm of chromosome 1H, distal to the SSR marker Bmag382. The elo genes were 23.5 cM (elo1), 33.3 cM (elo2) and 17.9 cM (elo-5) from Bmag382. elo3 was mapped to the long arm of chromosome 3H, 10.4 cM away from and on the centromeric side of SSR markers Bmag225 and GBM1034, and 30.4 cM away from GBM1233. Phenotypes of elo seedlings Three weeks after germination, elo seedlings were dwarfed compared with wild-type barley, with the degree of dwarfing varying between lines (Fig. 1). Phenotypes of mutants in three of the elo loci (elo1–elo3) were recessive and two (elo-4 and elo-5) were semi-dominant, with homozygotes exhibiting a more severe phenotype than heterozygotes (Fig. 1). Shoots were significantly shorter in all elo lines compared with the wild-type, and roots were significantly shorter in all but one of the lines (elo1). Homozygous elo-5 plants were the most severely stunted, with shoot and root lengths reduced to around 10% of those of the wild-type (Fig. 1i). Heterozygous elo-4 and elo-5 seedlings had root and shoot lengths that were between those of their elo homozygous siblings and the wild-type. All above-ground organs measured were reduced in length compared with the wild-type (Fig. 1j), suggesting that the elo mutations caused dwarfing of all plant organs, rather than affecting plant height alone. Leaf blades and sheaths were shorter in all elo mutants, as were coleoptiles (Fig. 1j); however, coleoptile length was generally less severely affected than length of the leaf blade or sheath. Fig. 1 View largeDownload slide Dwarfed phenotype of the elo mutants. Three-week-old wild-type (a), elo1 (b), elo2 (c), elo3 (d), elo-4 homozygote (e), elo-4 heterozygote (f), elo-5 homozygote (g) and elo-5 heterozygote (h) seedlings are shown. Insets in (b), (d) and (h) show 5-day-old seedlings, with twisting evident. Scale bars = 3 cm. (i–j) Measurements of above-ground and below-ground organs in 3-week-old seedlings. (i) Lengths of shoots (white bars) and roots (gray bars) of the elo mutants and wild-type seedlings. (j) Lengths of coleoptiles (grey), L1 leaf blades (white) and L1 leaf sheaths (hatched). Measurements are represented as a percentage of wild-type height/length for 3-week-old seedlings ± SD (n ≥7). Fig. 1 View largeDownload slide Dwarfed phenotype of the elo mutants. Three-week-old wild-type (a), elo1 (b), elo2 (c), elo3 (d), elo-4 homozygote (e), elo-4 heterozygote (f), elo-5 homozygote (g) and elo-5 heterozygote (h) seedlings are shown. Insets in (b), (d) and (h) show 5-day-old seedlings, with twisting evident. Scale bars = 3 cm. (i–j) Measurements of above-ground and below-ground organs in 3-week-old seedlings. (i) Lengths of shoots (white bars) and roots (gray bars) of the elo mutants and wild-type seedlings. (j) Lengths of coleoptiles (grey), L1 leaf blades (white) and L1 leaf sheaths (hatched). Measurements are represented as a percentage of wild-type height/length for 3-week-old seedlings ± SD (n ≥7). In addition to being dwarfed, the leaves and coleoptiles of elo1, elo3 and elo-5 heterozygotes developed a right-handed or anticlockwise twist. This was particularly evident in seedlings <1 week old (Fig. 1b, d, h insets). The twisting phenotype was most pronounced in elo-5 heterozygotes, which had more rotations about the growth axis than elo3 and elo1 seedlings of the same age. Twisting was least evident in elo1 seedlings. Although heterozygous elo-5 seedlings displayed a twisting phenotype, twisting was not evident in homozygous elo-5 seedlings. Twisting was not evident in mature elo1 or elo3 plants, but was evident in mature elo-5 heterozygotes, where the awns of floral spikes twisted around each other (data not shown). Cell expansion and division in elo leaves In addition to being dwarfed, the elo mutants displayed a number of other noteworthy features, the most obvious being the slightly wrinkled or lumpy appearance of the leaf epidermes of elo1, elo3 and elo-4 homozygotes and of elo-5 heterozygotes. Replicas of the abaxial surfaces of elo leaves (Fig. 2a–h, upper panels) revealed that the arrangement of cells into distinct files was, for the most part, the same in elo leaves as in the wild-type. Epidermal cells on the abaxial surface of a wild-type barley leaf blade form distinct files that are longitudinal to the direction of growth and parallel to the underlying veins (Metcalfe 1960; Fig. 2). In a wild-type leaf, files can be divided into those lying over veins, termed costal cell files (Fig. 2, 'c'), and those lying between veins, termed intercostal files (Fig. 2, 'ic'). Files of alternating trichomes and sclerenchyma cells (Fig. 2, arrows) occur directly above veins (Wenzel et al. 1997). Files located in intercostal zones consist of alternating stomata and interstomatal cells (Metcalfe 1960, Becraft 1999). Fig. 2 View largeDownload slide Leaf morphology of wild-type and elo mutant leaf blades. Scanning electron micrographs (a–h, upper panels) of replicas taken from the abaxial surface of fully expanded L1 blades approximately one-third of the distance from the blade base (ligule). Micrographs are oriented with the leaf tip towards the right of the figure. Lower panels (a–h) are transverse cross-sections of leaf blades stained with toluidine blue O. (a) Wild- type; (b) elo1; (c) elo2; (d) elo3; (e) elo-4 homozygote; (f) elo-4 heterozygote; (g) elo-5 homozygote; (h) elo-5 heterozygote. Arrows indicate files of cells containing trichomes, and open arrowheads indicate stomata. Brackets delineate files of costal (c) and intercostal (ic) cells. The filled arrowhead in (f) marks a ruptured cell. Phloem (ph), xylem (xy) and parenchymatous bundle sheath (pbs) tissues are indicated in (a). Asterisks in (h) indicate regions of aberrant cell division in the elo-5 homozygote leaf. Aberrant periclinal divisions (arrows) on developing adaxial surfaces of elo-5 homozygote (i) and heterozygote (j) leaves are indicated by arrows. Scale bar for SEM images = 200 μm; scale bar for transverse sections = 100 μm. Fig. 2 View largeDownload slide Leaf morphology of wild-type and elo mutant leaf blades. Scanning electron micrographs (a–h, upper panels) of replicas taken from the abaxial surface of fully expanded L1 blades approximately one-third of the distance from the blade base (ligule). Micrographs are oriented with the leaf tip towards the right of the figure. Lower panels (a–h) are transverse cross-sections of leaf blades stained with toluidine blue O. (a) Wild- type; (b) elo1; (c) elo2; (d) elo3; (e) elo-4 homozygote; (f) elo-4 heterozygote; (g) elo-5 homozygote; (h) elo-5 heterozygote. Arrows indicate files of cells containing trichomes, and open arrowheads indicate stomata. Brackets delineate files of costal (c) and intercostal (ic) cells. The filled arrowhead in (f) marks a ruptured cell. Phloem (ph), xylem (xy) and parenchymatous bundle sheath (pbs) tissues are indicated in (a). Asterisks in (h) indicate regions of aberrant cell division in the elo-5 homozygote leaf. Aberrant periclinal divisions (arrows) on developing adaxial surfaces of elo-5 homozygote (i) and heterozygote (j) leaves are indicated by arrows. Scale bar for SEM images = 200 μm; scale bar for transverse sections = 100 μm. Although the organization of the elo abaxial epidermis appeared similar to that of the wild-type, individual cells showed varying degrees of aberrant expansion (Fig. 2b–h, upper panels). Cells in all files were considerably shorter than their wild-type equivalents, with many having undergone substantial lateral expansion, especially the intercostal cells. Costal cells, in contrast, were rarely more than twice the width of wild-type cells. Trichomes (Fig. 2, arrows) and stomata (Fig. 2, open arrowheads) were observed on all elo leaf surfaces. Trichomes were similar to the wild-type, but sclerenchyma cells were shorter and, in the case of elo1, noticeably wider (Fig. 2b). Interstomatal cells in all elo lines were shorter than those of the wild-type, as were the guard and subsidiary cells that comprise the stomatal complex. The lateral swelling of cells in adjacent files would often obscure stomata. This was especially evident in the elo-5 homozygote (Fig. 2g). Ruptured epidermal cells were also occasionally observed. An example of this is seen in the elo-4 heterozygote (Fig. 2f, closed arrowhead). In elo1 (Fig. 2b), elo3 (Fig. 2d), elo-4 (Fig. 2e and f) and elo-5 heterozygotes (Fig. 2h), swollen cells were widest around their mid-point and tapered at each end. However, not all cells were swollen, with affected cells often in clusters along the length of the leaf and interspersed with clusters of less affected cells. This resulted in the characteristic wrinkled appearance of elo1 and elo3 homozygote, and elo-4 and elo-5 heterozygote leaf blades. Transverse sections of elo leaf blades showed aberrantly swollen cells on the adaxial and abaxial epidermes (Fig. 2b–h, lower panels). In elo1 (Fig. 2b), costal cells on both surfaces had approximately the same diameter as wild-type cells, but intercostal cells were much larger. In elo3 (Fig. 2d), elo-4 homozygotes (Fig. 2e) and heterozygotes (Fig. 2f), and elo-5 heterozygotes (Fig. 2h), costal and intercostal cells on the adaxial surface and intercostal cells on the abaxial surface were radially swollen. In elo2 (Fig. 2c) and homozygous elo-5 leaves (Fig. 2g), cells in all files on both surfaces were swollen. To quantify these epidermal phenotypes, measurements were made of the length (Fig. 3a) and width (Fig. 3b) of intercostal cells (bulliform cells) on the leaf adaxial surface. Except for elo-4 heterozygotes, bulliform cells of elo mutant lines were significantly shorter and wider than wild-type cells (P < 0.05). As expected, the greatest reduction in cell length was seen in elo-5 homozygotes, where the average bulliform cell was only 25% the length of a wild-type cell (Fig. 3a). These cells were, however, almost twice the width of wild-type cells (Fig. 3b). Scanning electron microscopy (SEM) images and cell measurements indicate that cells on either surface of an elo-5 homozygote leaf undergo almost isodiametric expansion. Fig. 3 View largeDownload slide Measurement of cell length (a) and width (b) of intercostal cells on the adaxial leaf blade surface (bulliform cells) and the number of intercostal cells on the adaxial (c) and abaxial (d) leaf surfaces are also shown. Asterisks in (k) and (l) indicate values significantly different from the wild-type (P < 0.05). Error bars in (i) and (j) indicate the SD, and in (k) and (l) indicate the 95% confidence intervals. Fig. 3 View largeDownload slide Measurement of cell length (a) and width (b) of intercostal cells on the adaxial leaf blade surface (bulliform cells) and the number of intercostal cells on the adaxial (c) and abaxial (d) leaf surfaces are also shown. Asterisks in (k) and (l) indicate values significantly different from the wild-type (P < 0.05). Error bars in (i) and (j) indicate the SD, and in (k) and (l) indicate the 95% confidence intervals. Mesophyll cells (Fig. 2a, 'm') and vascular bundle cells of elo leaves were either the same size as corresponding wild-type cells or smaller (Fig. 2, lower panels). In particular, mesophyll cells of homozygous and heterozygous elo-4 and elo-5 leaves were smaller than those of the wild-type (Fig. 2e–h). Vascular bundles in the elo-5 homozygote (Fig. 2g) also had a slightly smaller diameter than those of the wild-type due to fewer xylem (Fig. 2a, 'x') and phloem (Fig. 2a, 'ph') cells, and the mestome sheath just inside of the parenchymatous bundle sheath (Fig. 2a, 'pbs') was less regular than that of the wild-type in shape and cell size. The organization of most elo leaves was similar to that of the wild-type, except for elo2 (Fig. 2c) and homozygous elo-5 leaves (Fig. 2g) where transverse sections revealed areas of aberrant periclinal cell divisions in the adaxial epidermis (Fig. 2c, g, asterisks), resulting in regions where the epidermis was more than one cell layer thick. In elo-5 homozygotes, these regions occurred in both costal and intercostal cells of the adaxial epidermis, but did not occur in the abaxial epidermis. Epidermal thickenings could sometimes be detected in replicas of the adaxial surface of elo-5 homozygote leaves (data not shown). Such growths were not apparent in heterozygous elo-5 leaves (Fig. 2h). In addition to aberrant periclinal divisions, incorrectly oriented anticlinal divisions were observed in developing leaf surfaces of elo-5 homozygotes and heterozygotes (Fig. 2i and j, arrows). Oblique cell divisions appeared to be restricted to certain cell files, but were obscured by subsequent cell expansion in mature regions of developing leaves and in fully expanded leaves (Fig. 2g, h). To determine whether reduced cell number as well as cell length contributed to dwarfism in the elo mutants, measurements of intercostal cells on both the adaxial and abaxial leaf surfaces were used to calculate the number of cells along the length of the leaf blade (Fig. 3c, d). Some elo lines (elo1, elo3, and elo-5 heterozygotes) had numbers of cells on either side of the leaf blade similar to those of the wild-type, while other lines had either more or fewer cells on one or other surface. elo2 leaves had a similar number of cells to the wild-type on the abaxial surface but had 40% fewer cells on the adaxial surface (Fig. 3c, d). Homozygous elo-4 leaves had the same number of cells as the wild-type on the adaxial surface and about 50% more cells on the abaxial surface. The reverse was true for the elo-4 heterozygote where there were 50% more cells on the adaxial surface (Fig. 3c, d). In elo-5, heterozygous plants showed no significant difference in cell number on either leaf surface, yet leaves of the homozygous plants had 50% fewer cells on the adaxial leaf surface and 30% more cells on the abaxial leaf surface. Root morphology of the elo mutants The external morphology of roots of 4-day-old wild-type and elo seedlings was examined (Fig. 4, upper panels). Roots of elo1 (Fig. 4b) and heterozygous elo-4 plants (Fig. 4f) appeared similar in diameter to those of the wild-type. Roots of elo2, elo3, homozygous elo-4, and both homozygous and heterozygous elo-5 were wider than those of the wild-type above the zone of differentiation, which is marked by emergent root hairs (Fig. 4c–e, g, h). The distance from the root tip to the point of root hair emergence (Fig. 4, brackets) was also shorter in the elo roots than in the wild-type. This was especially noticeable in roots of elo-5 homozygotes (Fig. 4g), where the distance from tip to root hair emergence was only 25% of the wild-type distance (Fig. 4a). Root hairs were present in all elo lines, with no obvious change in length or morphology. Fig. 4 View largeDownload slide Root morphology of the elo mutants. Whole roots (upper panels) and transverse sections (lower panels) from 4-day-old seedlings are shown. (a) Wild-type; (b) elo1; (c) elo2; (d) elo3; (e) elo-4 homozygote; (f) elo-4 heterozygote; (g) elo-5 homozygote; (h) elo-5 heterozygote. Dashed lines represent the approximate locations of the transverse sections. Scale bar for whole roots = 2 mm; scale bar for transverse sections = 100 μm. Fig. 4 View largeDownload slide Root morphology of the elo mutants. Whole roots (upper panels) and transverse sections (lower panels) from 4-day-old seedlings are shown. (a) Wild-type; (b) elo1; (c) elo2; (d) elo3; (e) elo-4 homozygote; (f) elo-4 heterozygote; (g) elo-5 homozygote; (h) elo-5 heterozygote. Dashed lines represent the approximate locations of the transverse sections. Scale bar for whole roots = 2 mm; scale bar for transverse sections = 100 μm. To observe differences in their internal anatomy, roots were sectioned transversely at a point just before the point of root hair emergence (Fig. 4, lower panels). The radial organization of elo roots was the same as that of the wild- type, with the central stele (Fig. 4a, 's'), consisting primarily of vascular tissue, surrounded by 3–4 cell layers of ground tissue comprising the root cortex (Fig. 4a, 'c'), and a single layer of epidermal cells (Fig. 4a, 'ep'). Increased root diameter in some elo lines could be attributed to larger cells. Cortical cells were much larger than those of the wild-type in roots of elo2, elo3 and homozygotes of elo-4 and elo-5 (Fig. 4c–e and g). In elo1, elo2 and heterozygotes of elo-4 and elo-5 (Fig. 4b, c, f, h), cells in the stele were similar in diameter to those of the wild-type, but the vascular cells in elo3, elo-4 and elo-5 homozygotes were wider (Fig. 3d, e, g). In elo-5 homozygotes, epidermal cells were less regular in shape than in wild-type cells (Fig. 4g). No twisting was observed in any of the elo roots, and root hairs in the elo mutants were similar to wild-type root hairs. elo grain morphology The external appearance of mature wild-type and elo grains was examined and the 100 grain weights were compared. Homozygous elo-4 and elo-5 lines do not produce grains, so grain weights for these lines were obtained by germinating half-grains of the relevant heterozygote and observing the seedling phenotype to determine the genotype of the remaining half-grain. Apart from a <10% reduction in weight, grains of most elo lines were indistinguishable in appearance from the wild-type (data not shown). Grains of elo2 were wrinkled and noticeably smaller than those of the wild-type (Fig. 5a, b), and elo2 grain weight was 60% that of the wild-type. Fig. 5 View largeDownload slide Grain morphology of the elo mutants. Whole grains of the wild- type (a) and elo2 (b) are shown. Transverse sections through the aleurone layer are also shown (c and d). Seed coat (sc), aleurone (al) and starchy endosperm (se) are indicated in (c). The arrow in (b) indicates regions of ectopic aleurone in elo2. Scale bar for whole grains = 2 cm; scale bar for transverse sections = 100 μm. Fig. 5 View largeDownload slide Grain morphology of the elo mutants. Whole grains of the wild- type (a) and elo2 (b) are shown. Transverse sections through the aleurone layer are also shown (c and d). Seed coat (sc), aleurone (al) and starchy endosperm (se) are indicated in (c). The arrow in (b) indicates regions of ectopic aleurone in elo2. Scale bar for whole grains = 2 cm; scale bar for transverse sections = 100 μm. A longitudinal cross-section through a mature wild-type grain shows the starchy endosperm surrounded by an aleurone layer 3–4 cells thick (Fig. 5c). Aleurone cells were mostly isodiametric to cuboid in shape, with a dense cytoplasm and a thick cell wall. Similar sections through the grains of elo1, elo3, homozygous and heterozygous elo-4 and homozygous and heterozygous elo-5 showed no differences compared with the wild-type in the appearance of either the aleurone layer or the starchy endosperm. Although the aleurone layer in elo2 grains consisted of a continuous layer of cells, its thickness varied from as few as one cell to 6–7 cells (Fig. 5d). Areas of thicker aleurone formed projections into the starchy endosperm and lie beneath wrinkles or depressions in the seed coat. In addition to uneven thickness, elo2 aleurone cells were also less regular in size and shape than wild-type cells (Fig. 5d). There was no obvious difference in the starchy endosperm of elo2. Cell wall composition in elo leaves To determine whether the phenotypic changes seen in the elo mutants were accompanied by changes in cell wall composition, the polysaccharide composition of walls from L1 leaf blades of 3-week-old seedlings was determined (Table 1; Supplementary Tables S1, S2). Linkage analysis of native and carboxyl-reduced cell wall preparations were used to calculate the polysaccharide compositions of wild-type and elo leaf cell walls based on known structures of wall polysaccharides (Table 1; Kato et al. 1981, Shea et al. 1989, Sims and Bacic 1995). Table 1  Cell wall polysaccharide composition of wild-type and elo leaf cell wallsa Polysaccharide  Wild-type  Mutant line       elo1  elo2  elo3  elo-4 (hom)  elo-4 (het)  elo-5 (hom)  elo-5 (het)  GAX  25.7  32.2  32.5  32.9  33.0  30.5  40.4  31.5  MLG  0.0  0.3  0.0  0.2  0.2  0.2  0.0  0.0  XG  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  Pectinsb  7.7  13.2  13.9  8.9  8.7  5.1  11.9  4.8  Mannan  0.0  0.7  1.3  1.0  1.4  1.4  2.2  1.5  Other  5.3  5.1  3.5  4.8  5.5  4.2  3.2  5.7  Cellulose  61.3  48.5  48.8  52.2  51.1  58.6  42.3  56.5  Cellulose (% WT)c  100.0  79.2  79.6  85.2  83.4  95.6  69.0  92.2  Polysaccharide  Wild-type  Mutant line       elo1  elo2  elo3  elo-4 (hom)  elo-4 (het)  elo-5 (hom)  elo-5 (het)  GAX  25.7  32.2  32.5  32.9  33.0  30.5  40.4  31.5  MLG  0.0  0.3  0.0  0.2  0.2  0.2  0.0  0.0  XG  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  Pectinsb  7.7  13.2  13.9  8.9  8.7  5.1  11.9  4.8  Mannan  0.0  0.7  1.3  1.0  1.4  1.4  2.2  1.5  Other  5.3  5.1  3.5  4.8  5.5  4.2  3.2  5.7  Cellulose  61.3  48.5  48.8  52.2  51.1  58.6  42.3  56.5  Cellulose (% WT)c  100.0  79.2  79.6  85.2  83.4  95.6  69.0  92.2  aValues are Mol.% of total cell wall polysaccharides and are the average of duplicate determinations. bLeaf pectins include neutral and acidic monosaccharides comprising type I and II arabinogalactans, arabinans, HG and RGI. cCellulose content represented as a percentage of the wild-type value. View Large Cellulose was the most abundant polymer in wild-type leaf cell walls, comprising about 61.3% of total cell wall polysaccharides. Less cellulose was found in the cell walls of all elo leaf blades, with the largest reduction (31%) being in the elo-5 homozygote (Table 1). However, the cellulose contents of heterozygous elo-4 and elo-5 leaves were only marginally lower than those of the wild-type (4.4 and 7.8%, respectively). elo leaf cell walls also had more GAX than wild-type walls, with the greatest increase being in elo-5 homozygotes (40.4% compared with 25.7% in the wild-type). The pectin content in most elo leaf cell walls was also higher, although 34–38% less pectin was found in heterozygous elo-4 and elo-5 leaf cell walls (Table 1). Other minor polysaccharides included MLG in elo1, elo3 and elo-4 (homozygous and heterozygous), and heteromannan, which was absent in the wild-type but present in small amounts in elo leaves. Around 5% of the sugars detected in leaf cell wall could not be assigned to a polysaccharide (Table 1, 'other'). Cortical microtubule arrays in elongating roots Some features of the elo mutants, such as organ twisting and radial cell expansion, are also features of several Arabidopsis MT mutants (Sedbrook and Kaloriti 2008). To see whether the MT-disrupting drugs taxol and oryzalin could phenocopy the elo mutations, wild-type barley seeds were germinated in the presence of these drugs. However, no organ twisting or radially swollen cells were observed in the seedlings grown from the treated seeds (data not shown). Since poor drug uptake could also result in a failure of taxol or oryzalin to phenocopy the elo mutations, we also compared the cortical MT arrays in elongating epidermal cells of wild-type, elo3 and homozygous and heterozygous elo-5 roots. In both wild-type and elo roots, dividing cells were visible at the root tip, elongating cells were observed behind the tip in the elongation zone, and cell elongation ceased shortly behind the site of root hair emergence. In the elongation zone of wild-type roots, cortical MTs had the transverse orientation characteristic of elongating cells (Fig. 6a), an orientation that was maintained as cells progressed through this zone (Fig. 6b and c). MT orientation was more random in cells that had left the elongation zone (Fig. 6d). Fig. 6 View largeDownload slide Cortical microtubule arrays in epidermal root cells of the wild-type (a–d), elo3 (e–h) and elo-5 homozygotes (i–l) and heterozygotes (m–p). Confocal images in each series represent progress through the root elongation zone from less mature cells (a, e, i, m) to fully expanded cells (d, h, l, p). The root tip in each image is towards the left of the figure. Arrows in e, f and m indicate microtubules lying obliquely to the growth axis, and asterisks in j, k, l and m indicate aberrant cell divisions. Scale bars = 5 μm. Fig. 6 View largeDownload slide Cortical microtubule arrays in epidermal root cells of the wild-type (a–d), elo3 (e–h) and elo-5 homozygotes (i–l) and heterozygotes (m–p). Confocal images in each series represent progress through the root elongation zone from less mature cells (a, e, i, m) to fully expanded cells (d, h, l, p). The root tip in each image is towards the left of the figure. Arrows in e, f and m indicate microtubules lying obliquely to the growth axis, and asterisks in j, k, l and m indicate aberrant cell divisions. Scale bars = 5 μm. In elo3 roots, the orientation of cortical MTs in cells at the start of the elongation zone was similar to that in the wild- type, although some MTs were oblique to the growth axis (Fig. 6e, arrows). Differences in MT orientation became more apparent as cells progressed through the elongation zone. In some cells the cortical MTs were shorter and less uniformly transverse (Fig. 6f), and in other cells the MTs were longer and arranged into a left-handed helix (Fig. 6g). Cortical MTs in mature elo3 epidermal cells were similar to those in the wild-type (Fig. 6h). The cortical MT arrays in homozygous (Fig. 6i) and heterozygous (Fig. 6m) elo-5 cells at the beginning of the elongation zone were mostly transverse, although some of the MTs in heterozygous elo-5 cells were oblique (Fig. 6m, arrows). Cells with an oblique anticlinal cell wall were also observed. These cells were more common in elo-5 homozygous roots than in elo-5 heterozygous roots (marked with asterisks in Fig. 6j–m). In cells with an oblique anticlinal wall, MTs were either obliquely (Fig. 6j) or randomly oriented (Fig. 6k). Cortical MT arrays in fully elongated elo-5 homozygote root cells resembled those of mature wild-type cells (Fig. 6l). MTs in heterozygous elo-5 roots were uniformly oriented within individual cells and generally formed a left-handed helix about the cell axis (Fig. 6n, o). MTs in mature heterozygous elo-5 cells were similar to those of the wild- type (Fig. 6p). The orientation of cortical MTs to the growth axis for wild-type, elo3 and homozygous and heterozygous elo-5 roots was measured in at least 100 cells from the elongation zones, and the distribution of the angles was graphed (Fig. 7). In wild-type cells, MT angles were tightly centred around 90°, with the angle of 73% of MTs being between 80 and 100° (Fig. 7a). MTs in the elo mutants were distributed across a wider range of angles and the distribution was skewed to the right in elo3 (Fig.7b) and in elo-5 homozygotes (Fig. 7c). In elo3 and in el-o5 homozygotes and heterozygotes, the number of cells with an MT angle between 80 and 100° was 23, 56 and 64%, respectively (Fig. 7). Since the cortical MTs were measured on the outermost surface of epidermal cells, the bias towards angles >90° is indicative of MTs forming left-handed helices. Fig. 7 View largeDownload slide Cortical microtubule angles for individual cells showing the distribution of angles observed in the wild-type (a), elo3 (b), elo-5 homozygotes (c) and elo-5 heterozygotes (d). The dotted line indicates 90°. Fig. 7 View largeDownload slide Cortical microtubule angles for individual cells showing the distribution of angles observed in the wild-type (a), elo3 (b), elo-5 homozygotes (c) and elo-5 heterozygotes (d). The dotted line indicates 90°. Discussion The elo mutants are a class of barley dwarfs initially described as being impaired in the process of cell expansion (Chandler and Robertson 1999). Here we have shown that apart from varying degrees of dwarfing, a characteristic feature of all elo mutants is the radial swelling of leaf epidermal cells and root cortical cells. In some lines, twisting of above-ground tissues and aberrant cell divisions are also seen. As the primary motivation for studying the elo mutants was to understand the process of cell expansion in a species with a type II primary cell wall, a comparison of these mutant phenotypes with similar mutant phenotypes in Arabidopsis, a species with a type I cell wall, led us to examine the potential role of cell wall or cytoskeletal defects in causing the elo phenotype. The radial swelling initially suggested that the elo mutations affect cell wall synthesis or deposition in primary rather than secondary cell walls. When cellulose synthesis in the primary cell walls of Arabidopsis is disrupted by mutation, radial swelling is observed, predominantly in actively expanding cells (Arioli et al. 1998, Burton et al. 2000, Fagard et al. 2000, Sato et al. 2001). These mutations are often (but not exclusively) in genes for cellulose synthase (CESA) isoforms expressed predominantly in elongating tissue (Arioli et al. 1998, Delmer et al. 2001, Fagard et al. 2000). Studies of the effects of the cellulose synthesis inhibitor, 2,6-dichlorobenzonitrile (DCB), on primary wall synthesis in excised oat internodes (Montague 1995) suggest that the radial swelling phenotype may be a characteristic of reduced primary cell wall cellulose synthesis that is not restricted to species with type I walls. While mutants defective in cellulose synthesis in secondary walls have been described in other grasses (Kokubo et al. 1989, Yeo et al. 1995, Tanaka et al. 2003), to date no mutants with defects in cellulose synthesis in type II primary cell walls have been reported. Consistent with a proposed role in cellulose synthesis, all elo mutants had lower levels of cellulose in their leaves than did the wild-type, with elo-5 homozygotes showing the greatest reduction (Table 1). However, the reductions in cellulose content seen in most elo mutants were small in comparison with the phenotypic severity of the mutation and with the reductions seen in known cellulose-deficient mutants of Arabidopsis. For instance, elo-4 homozygote seedlings were severely dwarfed, being only about 20% the size of a wild-type seedling, although leaf cellulose content was reduced by only 10%. In contrast, cellulose-deficient mutants of Arabidopsis typically have 40–60% of the wild-type cellulose content (Sato et al. 2001), with the temperature-sensitive mutant rsw2 growing normally at the permissive temperature despite having 25% less cellulose than a wild-type plant (Peng et al. 2000). In addition to reductions in cellulose content, leaves of some elo mutants had a greater proportion of non-cellulosic polysaccharides, such as GAX and pectin (Table 1). Changes to the amounts of non-cellulosic polysaccharides are usually explained in terms of the ability of these polysaccharides to compensate for the reduction in cellulose content. Studies of plants with a type I cell wall commonly report an increase in pectin levels accompanying a decrease in cellulose levels (Shedletzky et al. 1990, Shedletzky et al. 1992, Nickle and Meinke 1998, Sabba et al. 1999, Burton et al. 2000, Fagard et al. 2000, His et al. 2001, Encina et al. 2002). Although fewer studies of this sort have been performed on plants with pectin-poor type II cell walls, it is noteworthy that barley suspension-cultured cells adapted to grow in DCB showed no accompanying increase in pectin levels in cells with 40–70% less cellulose in their walls (Shedletzky et al. 1992, Yulia 2006). The map locations of four elo genes were determined. Three of the genes, elo1, elo2 and elo-5, were on the long arm of chromosome 1H, and the fourth gene, elo3, was mapped to the long arm of chromosome 3H. It is possible that elo-5 represents a semi-dominant allele of either elo1 or elo2. Large collections of barley mutants exist (United States Department of Agriculture–Agriculture Research Service 2005), and >300 mutants and numerous genes and quantitative trait loci (QTLs) have been mapped to the barley genome (Franckowiak 1997, Matthews et al. 2003). No QTLs obviously associated with cell expansion or cell wall composition have been reported in the vicinity of the map locations for the elo genes. A number of mutants on 1H and 3H have been described with phenotypes similar to the elo phenotypes (Franckowiak 1997), including dwarfism (Dahleen et al. 2005) and leaf twisting (Fedak et al. 1972, Hayashi et al. 1984); however, there is limited information from which to draw adequate comparisons between any of these mutants and the elo mutants, and, ultimately, crosses to assess allelism with the elo genes are required. Eight members of the barley CESA gene family have been identified and their map locations determined. HvCESA4 is on the long arm of chromosome 1H, and either HvCESA5 or HvCESA7 maps to a region close to elo3, on the long arm of chromosome 3H (Burton et al. 2004). However, HvCESA4, HvCESA5 and HvCESA7 are all secondary CESA genes, with highest expression in tissues undergoing secondary cell wall synthesis, such as stems and mature regions of roots (Burton et al. 2004). The elo phenotypes more closely resemble plants with primary cell wall defects affecting cell expansion. The primary CESA genes, HvCESA1, HvCESA2 and HvCESA6, are located on chromosomes 2H, 5H and 6H, respectively. However, as there is at least one additional CESA gene in barley (Burton et al. 2004), it remains formally possible that one of the Elo genes corresponds to an unidentified HvCESA involved in primary wall synthesis. Although there are no known barley mutants with mutations in a CESA gene, a recent study used virus-induced gene silencing (VIGS) to silence barley CESA gene expression (Held et al. 2008). This study did not report any elo-like phenotypes in barley leaves with reduced CESA expression, suggesting it is unlikely that the elo lines arose due to a mutation in a barley primary wall CESA. Unlike cellulose, MLG is specific to type II cell walls, and in vegetative cells is usually detected only in cells that are actively expanding (Smith and Harris 1999). MLG levels decline in vegetative cells that have stopped expanding, and only low levels are present in mature organs. Thus, only trace amounts of MLG were present in the fully expanded barley leaves used for cell wall analysis (Table 1). Members of the CELLULOSE SYNTHASE-LIKE F (CSLF) family are implicated in MLG synthesis (Burton et al., 2008) and, of the barley CSLF genes, the nearest to an elo locus is HvCSLF9, which is on the short arm of chromosome 1H (Burton et al. 2008). Other barley CSLF genes are located on chromosomes 2H, 5H and 7H. Moreover, VIGS of the CSLF genes expressed in barley leaves reduced MLG levels but did not result in cell expansion phenotypes (Held et al. 2008). Thus it is unlikely that any of the elo mutations are in a CSLF gene. Mutations in a glycosylphosphatidylinositol (GPI)-anchored protein can disrupt both primary and secondary cell wall biosynthesis in plants. The Arabidopsis cobra mutant has reduced cell wall cellulose content and isodiametrically expanded cells (Schindelman et al. 2001). The COB protein is thought to be involved in cellulose microfibril deposition in primary cell walls (Roudier et al. 2005), and mutation in an ortholog of COB, AtCobL4, specifically affects secondary wall formation in Arabidopsis (Brown et al. 2005). Two related genes, Brittle Culm-1 in rice (Li et al. 2003) and Brittle Stalk 2 in maize (Ching et al. 2006), have been identified, both of which affect secondary cell wall synthesis. No COB family member which affects primary wall synthesis in grasses has been identified. Some of the elo mutations affect cell division as well as cell expansion. Both homozygous and heterozygous elo-4 had more cells than wild-type leaves on one or other surface of the leaf (Fig. 2). In the case of elo-4 heterozygotes, this suggests that some degree of compensation was occurring, whereby leaves with roughly the same morphology and size were produced with more but shorter cells. However, the effects of elo mutations on cell division were most obvious in the extra cell divisions seen in the epidermis of homozygous elo-5 leaves and in the grain aleurone layer and leaf epidermis of elo2 (Figs. 2, 5). Similar phenotypes are also seen in a number of maize mutants with extra cell divisions in the leaf epidermis: extra cell layers1 (xcl1; Kessler et al. 2002), warty-1 (wty-1; Reynolds et al. 1998), crinkly4 (cr4; Becraft et al. 1996), defective kernel1 (dek1; Becraft et al. 2002) and supernumerary aleurone layers1 (sal1; Shen et al. 2003). Indeed, it is possible that Elo2 may be functionally related to XCL1, CR4, DEK1 or SAL1, as mutations in these genes are phenotypically similar, producing dwarfed plants with variable numbers of cell layers in the leaf epidermis and grain aleurone layer. Effects of the xcl1, cr4, dek1 and sal1 mutations on the cellulose content of maize leaves have not been assessed. A helical growth habit was seen in elo1, elo3 and heterozygous elo-5 plants (Fig. 1), a phenotype not previously reported in a cell wall mutant. In all three elo lines, the helix was right-handed. Since there are currently no examples of primary cell wall mutants in plants with type II cell walls with which the elo mutants can be compared, it is possible that defects in type II primary walls alter the growth habit of these plants to bring about the twisted growth of organs. The helical growth phenotype may also result from the effect of elo mutations on the cytoskeleton. Left-handed growth habit is common in Arabidopsis mutants with defects in cortical MT organization, such as mor1 (Whittington et al. 2001), lefty1 and 2 (Thitamadee et al. 2002, Abe et al. 2004), and propyzamide hypersensitive1 (phs1; Naoi and Hashimoto 2004). The MT-specific drugs taxol and propyzamide can also induce left-handed helical growth (Furutani et al. 2000, Wasteneys 2000); however, taxol was unable to phenocopy the twisting growth habit of the elo lines. Right-handed growth, as seen in elo1, elo3 and heterozygous elo-5, has been observed in the Arabidopsis spiral1 (spr1; Nakajima et al. 2004) and wave-dampened2 (wvd2; Yuen et al. 2003) mutants that have defects in MT organization, as well as in plants with mutated tubulin subunits (Ishida and Hashimoto 2007, Ishida et al. 2007). It is possible that the Elo genes may play a role in cytoskeletal organization and that the cell wall defects observed are a downstream consequence of altered cytoskeletal organization or function. Disruption of cortical MT arrays can lead to altered cellulose microfibril orientation (Hogetsu and Shibaoka 1978, Robinson and Quader 1982 Burk and Ye 2002, Eleftheriou et al. 2003, Baskin et al. 2004). A recently proposed model, the microfibril-length-regulation’ model, describes how altered cortical MT organization can adversely affect the stability of the cellulose synthase rosette complex, resulting in shortened microfibrils that are unable to provide the wall with the structural integrity needed to resist radial cell expansion (Wasteneys 2004). Analysis of the cellulose microfibril orientation in elo3 and elo-5 may provide additional evidence that the altered cortical MT arrays have a direct impact on the cell wall deposition and that wall integrity in these lines is compromised as a consequence. Although cell wall analysis is not routinely investigated in cytoskeletal mutants, it is possible that many of the mutants described thus far also have defects in wall composition. Studies have shown that mutations affecting the cortical MT array can result in cell wall compositional changes, as seen in the Arabidopsis fragile fibre2 (fra2) mutant (Burk et al. 2001). Indeed, disruption of cortical MT arrays can also cause radial swelling (Baskin et al. 2004), and is often accompanied by a loss of organization of the cellulose microfibrils (Hogetsu and Shibaoka 1978, Giddings and Staehelin 1988, Burk et al. 2001, Burk and Ye 2002). Recent evidence has revealed that CESA complexes move along linear paths in the plasma membrane in line with underlying cortical MTs (Paradez et al. 2006). However, there are numerous examples where co-alignment of cortical MTs and cellulose microfibrils does not occur (Takeda and Shibaoka 1981, Inada and Shimmen 2000, Baskin 2001, Himmelspach et al. 2003), and CESA complexes are not entirely dependent on MTs for directional movement across the plasma membrane (Paradez et al. 2006). The precise relationship between cortical MT orientation and cell wall integrity is still to be determined (Wasteneys and Fujita 2006, Emons et al. 2008, Lloyd and Chan 2008), although it remains a distinct possibility that the elo mutations affect cytoskeletal processes that in turn alter wall composition and structural integrity. Alterations to cell division also indicate that the Elo genes might not be directly involved in cell wall synthesis, but could be involved in cytoskeletal processes. Two mutants, elo1 and elo3, had no detectable defects in cell division; however, elo2, elo-4 and elo-5 had changes to rates of cell division and/or changes to patterns of cell division. Both elo2 and elo-5 had occasional aberrant cell division patterns in the leaf epidermis (elo2 and elo-5) and seed aleurone layer (elo2). This resulted in regions of the epidermis that were more than one cell layer thick and regions of elo2 aleurone up to seven cell layers thick. Although cell plate formation is sometimes affected in cell wall mutants, such as kor (Zuo et al. 2000, Lane et al. 2001) and cytokinesis-defective1 (cyt1; Nickle and Meinke 1998), these defects usually involve the formation of incomplete cell walls, rather than incorrectly positioned cell walls, as was the case for elo2 and homozygous and heterozygous elo-5 seedlings. Incorrect division planes are more reminiscent of defects in cytoskeletal function. For example, the maize tangled1 mutant has altered planes of cell division as a result of defects in a microtubule-associated protein (MAP; Cleary and Smith 1998, Smith et al. 2001). The compositional differences in type I and type II cell walls suggest that expansion and subsequent rigidification of the cell wall during cell growth and maturation are regulated differently in grass and non-grass species. In order to gain a balanced understanding of the process of cell expansion in plants, and how the cell wall and cytoskeleton play a role in regulating cell expansion, study of plants with both types of cell wall organization is necessary. The phenotypes of the elo mutants presented in this study indicate that while some effects of altered cell wall composition on cell expansion are common to plants with type I and type II cell walls, others are unique to plants with type II cell walls. The elo mutants therefore represent a useful resource for experiments aimed at furthering our understanding of cell expansion in plants with type II cell walls per se, as well as for experiments aimed at identifying the Elo genes themselves. Materials and Methods Plant material and growth conditions All of the elo lines described are from a sodium azide-mutagenized H. vulgare cv. 'Himalaya' population described by Chandler and Robertson (1999). All lines have been through three back-crossing generations prior to the studies undertaken here. For growth measurements, leaf microscopy and cell wall analysis, plants were grown in soil with a temperature regime of 18°C day/13°C night and a 16 h photoperiod. The first fully expanded leaf (L1) was used for analysis. For microscopy of root tissue, grains were germinated between two pieces of moist filter paper for 4 d at room temperature. Mapping The elo mutants (H. vulgare 'Himalaya') were crossed with H. vulgare 'Sloop', and the F1 plants of these crosses were allowed to self-pollinate to generate the F2 mapping populations. Genomic DNA from the elo mapping populations was extracted from one fully expanded second or third leaf blade, according to the methods of Paris and Carter (2000). SSR marker screens were performed to detect polymorphisms between 'Himalaya' and 'Sloop' parental DNA. Bulked DNA from 12 individuals of each phenotype (dwarf, tall and semi-dwarf) was also used to detect linkage to markers being screened. Linked markers were confirmed by amplification from the individuals that comprised each bulk. PCR amplification of SSR markers from 50 ng of genomic DNA was carried out using the methods described in Hayden et al. (2008). Formamide loading dye (3 μl) was added to PCR products and reaction tubes were heated for at least 5 min at 9°C to denature products. PCR products (1 μl) were run on Gel-Scan real-time gel electrophoresis systems (Gel-Scan 2000 and Gel-Scan 3000, Corbett Life Sciences, Sydney, NSW, Australia), alongside GeneScan-400HD molecular weight standards (Applied Biosystems, Melbourne, Vic, Australia). SSR markers were scored as A ('Sloop'), B ('Himalaya') or H (heterozygous). The phenotype of the F2 individuals was used as an additional marker. In recessive lines (elo1–elo3), dwarfs were scored as B and talls were scored as H, while in the semi-dominant line (elo-5) dwarfs were scored as B, semi-dwarfs as H and talls as A. A spreadsheet containing marker scores was imported into Map Manager QTX (Manly et al. 2001) and linkage groups were calculated at P = 0.001 [roughly equal to a likelihood of odds (LOD) score of 3] using the Kosambi map function. Linkage data from Map Manager QTX were used to draw linkage maps in MapChart (Voorrips 2002). Chromosome maps for comparisons with linkage groups were also drawn in MapChart, using map distances (in cM) available on the Scottish Crop Research Institute (SCRI) Plant Bioinformatics Group website (Barley SSRs 1.0; http://germinate.scri.ac.uk/ssr/barley_s.html). Light microscopy Root and leaf samples were fixed in 2.5% (v/v) glutaraldehyde in 0.05 M sodium cacodylate buffer (pH 7.2) for 16 h at room temperature, dehydrated in an ethanol series and embedded in LR White resin (London Resin Company Reading, Berkshire, UK). Blocks were sectioned on a Leica Ultracut R microtome (Leica Microsystems, North Ryde, NSW, Australia) with glass knives, and 1 μm sections were stained for 1 min at 60°C with 0.1% toluidine blue O in phosphate buffer (pH 6.8). For light microscopy of grains, homozygous and heterozygous grains of elo-4 and elo-5 were identified by bisecting grains of these two lines transversely to obtain halves with and without embryos. The embryo-containing half-grains were surface sterilized for 10 min in a solution containing 30% (v/v) sodium hypochlorite and 1% Triton X-100, rinsed in sterile water (5×), placed on to half-strength MS medium (Murashige and Skoog 1962) containing 1% agar and allowed to germinate in the dark at room temperature. Grain genotypes could be verified after 3–4 d, and embryo-less half-grains corresponding to homozygous and heterozygous embryos were embedded for light microscopy. Grains to be embedded were soaked in water for 1 h, bisected longitudinally and fixed in 70% ethanol. Tissue was then embedded in PARAPLAST paraffin wax (Oxford Labware, St Louis MO, USA) using a LX120 Tissue Processor (Hacker Instruments & Industries, Winnsboro, SC, USA), according to the methods of Gordon (1990). Sections (8 μm) were cut using a Microm HM350 rotary microtome (Carl Zeiss, North Ryde, NSW, Australia), dewaxed in a series of histolene, ethanol and water, and stained for 1 min at 60°C with 0.1% toluidine blue O in phosphate buffer (pH 6.8). Stained sections were photographed using a Leica DC300 F digital camera mounted on an Olympus BHS microscope (Olympus Australia, Mt Waverley, Vic, Australia). Whole roots of 4-day-old seedlings were photographed using a Wild Heerbrugg (Gais, Switzerland) dissecting microscope and camera (Photoautomat MPS55). Scanning electron microscopy and cell measurements Resin replicas of the adaxial and abaxial leaf surfaces were prepared from the first emergent leaf (L1) of 2-week-old barley seedlings. Replicas were made using dental impression material (3M Express, 3M, Australia, Pymble, NSW, Australia) and Spurr's resin as previously described (Wenzel et al. 1997). Replicas were mounted on SEM stubs with silver conductive paint (ProSciTech, Kirwan, Qld, Australia), then sputter coated with gold using an Edwards S150B gold sputter coater (Edwards, Wilmington, DE, USA) and imaged in a Philips XL30 FEG field emission scanning electron microscope (FESEM; Philips Systems Australasia, North Ryde, NSW, Australia) using 2 kV of accelerating voltage. SEM images of roots and developing leaves were taken using the cryo-SEM facilities at LaTrobe University (Bundoora, Vic, Australia). Four-day-old leaves and roots were mounted onto a cryo-stage with Cryo-Gel (SPI Supplies, West Chester, PA, USA), frozen in liquid N2 cooled to –210°C under a vacuum, and inserted into the preparation chamber of a Joel Cryo-Scanning Electron Microscope (Joel Australasia, Frenchs Forest, NSW, Australia). Samples were coated with 3 nm gold–palladium before being viewed at a voltage of 5 kV using an LEI detector. Cell width and length measurements of a minimum of 30 cells per replica from various points along the leaf were made using software provided with the FESEM (Philips FEI XL, v5.70). Length and width were defined as being parallel to and perpendicular to the direction of leaf growth, respectively. Cell wall isolation Cell walls were isolated from fully expanded L1 blades of 3-week-old barley seedlings. Plants were placed in the dark 24 h prior to cell wall isolation to deplete starch reserves. Tissue was collected, freeze-dried, and ground to a fine powder with two 3 mm ball bearings for 1 min (30 vibrations s–1) using a Retsch Mixer Mill MM300 (MEP Instruments, North Ryde, NSW, Australia). Wall material was washed sequentially with cold 80% ethanol (5×), 100% methanol (2×), 100% acetone (2×) and 100% n-pentane (2×), and allowed to air dry in a moisture-free container. The method of Smith and Harris (1999) was used to remove starch from the crude wall preparations using porcine pancreatic α-amylase (Sigma, A-6255; Sigma-Aldrich Sydney, NSW, Australia). Wall material was then washed with cold absolute ethanol (4×), and twice with 100% methanol. Cell walls were resuspended in 100% methanol at 2 mg of cell wall ml–1 and stored at –20°C until required. Cell wall compositional analysis For wall analysis, permethylated alditol acetates were prepared using NaOH and methyl iodide as previously described (Sims and Bacic 1995). The permethylated sugars were separated using the splitless mode of injection on a Hewlett Packard HP 6890 series GC-MS (Agilent Technologies, Forest Hill, Vic, Australia) on a high polarity, bonded phase BPX70 capillary column (SGE, Ringwood, Vic, Australia) using helium as the carrier gas with a flow rate of 1 ml min–1. Column temperature started at 170°C for 2 min, then increased at a rate of 3°C min–1 to 240°C. Cell wall uronic acid composition was determined by carboxyl reduction as described by Kim and Carpita (1992) and modified by Sims and Bacic (1995). Uronic acid composition and degree of methyl-esterification was calculated from the GC-MS chromatograms of carboxyl-reduced, permethylated alditol acetates according to methods previously described (Sims and Bacic 1995). The contributions of these uronic acid-containing-peaks to total wall monosaccharides were then determined by comparison of 4-Glc in carboxyl-reduced and non-carboxyl-reduced permethylated alditol acetates. Cell wall polysaccharide composition was determined from the GC-MS chromatograms of the permethylated alditol acetates by the identification of peaks and calculation of areas under all peaks corresponding to monosaccharides. Total monosaccharide linkage compositions were calculated as an average molar percentage of the total amount of sugars detected in each chromatogram. Polysaccharide composition of the cell walls was calculated based on the structures of well-characterized wall polysaccharides from Nicotiana plumbaginifolia (Sims and Bacic 1995) and Daucus carota (Shea et al. 1989). Estimations of the levels of XG and MLG in type II cell walls were based on information from Sims et al (2000) and Kato et al. (1981), respectively. Microtubule labeling MT labeling was performed using the method of Wenzel et al. (2000), with modifications. Root tissue was fixed in 6% (v/v) formaldehyde in PHEM/DMSO buffer [60 mM PIPES, 25 mM N-2-hydroxyethyl PIPES, 10 mM EGTA, 4 mM MgSO4, 5% (v/v) dimethylsulfoxide, pH 6.9] for 1 h at room temperature. After rinsing in PHEM/DMSO buffer (3×5 min), tissue was digested for 20 min at room temperature with a freshly prepared enzyme mixture containing 2% (w/v) driselase (Sigma, St Louis, MO, USA) and 1% (w/v) pectolyase (Kikkoman Chiba, Japan) in water. Samples were briefly washed in PHEM/DMSO buffer, then transferred to a solution of 1% (v/v) Triton X-100 in PHEM/DMSO buffer for 2.5 h to increase membrane permeability. Samples were rinsed in phosphate-buffered saline (PBS; pH 7.2, 6×5 min), incubated in mouse anti-α-tubulin IgG [1:500 dilution in PBS with 1% bovine serum albumin (BSA); Ab Serotec, Kidlington, UK] for 1 h at 37°C and overnight at 4°C. Samples were washed in PBS (6×5 min) and incubated in Alexa Fluor® 488-conjugated goat anti-mouse monoclonal antibody (1:500 dilution in PBS with 1% BSA; Molecular Probes, Eugene, OR, USA) for 2 h at 37°C in the dark. Samples were washed with PBS (6×5 min) prior to being mounted in 0.1% DABCO antifade (1,4-diazobicyclo[2,2,2]-octane; Sigma-Aldrich, Sydney, NSW, Australia) in PBS and glycerol (1: 1). Mounted tissue was kept in the dark at 4°C prior to viewing with a Leica Inverted TCS Confocal Microscope (Leica Microsystems, North Ryde, NSW, Australia). Assessment of microtubule orientation The MT angle was assessed essentially as in Furutani et al. (2000). Images of representative cells were collected at the beginning, middle and end of the root elongation zone. Images were also taken of cells that had stopped elongating after reaching their final length. Images of at least 100 immunolabeled epidermal cells from the elongation zone of wild-type, elo3, and elo-5 homozygotes and heterozygotes were used to determine the orientation of cortical MT arrays. The angle between the longitudinal anticlinal wall of each cell and MTs adjacent to the outer epidermal wall was measured. For each cell, measurements were taken at the apical, middle and basal regions of the cell using Image-Pro Express software (Media Cybernetics, Bethesda, MD, USA) and averaged to give an MT angle for the individual cell. Funding The Grains Research and Development Corporation (GRDC); the University of Melbourne (Melbourne Research Scholarship to D.L.). Acknowledgments We thank Dr Bruce Abaloz (University of Melbourne, Australia) for preparation of grain sections. Crosses were kindly performed by Juan Juttner and Ursula Langridge-Reimold at the University of Adelaide. The 400+ SSR markers used for mapping of the elo genes were kindly provided by Dr. Ken Chalmers and Professor Peter Langridge at the Molecular Plant Breeding Cooperative Research Centre and the Australian Centre for Plant Functional Genomics (University of Adelaide, Waite Campus, Australia). References Abe T,  Thitamadee S,  Hashimoto T.  Microtubule defects and cell morphogenesis in the lefty1lefty2 tubulin mutant of Arabidopsis thaliana.,  Plant Cell Physiol. ,  2004, vol.  45 (pg.  211- 220) Google Scholar CrossRef Search ADS PubMed  Arioli T,  Peng L,  Betzner AS,  Burn J,  Wittke W,  Herth W, et al.  Molecular analysis of cellulose biosynthesis in Arabidopsis.,  Science ,  1998, vol.  279 (pg.  717- 720) Google Scholar CrossRef Search ADS PubMed  Bacic A,  Harris PJ,  Stone BA.  Priess J.  Structure and function of plant cell walls,  The Biochemistry of Plants.  ,  1988 New York Academic Press(pg.  297- 371) Baskin TI.  On the alignment of cellulose microfibrils by cortical microtubules: a review and a model.,  Protoplasma ,  2001, vol.  215 (pg.  150- 171) Google Scholar CrossRef Search ADS PubMed  Baskin TI,  Beemster GTS,  Judy-March JE,  Marga F.  Disorganization of cortical microtubules stimulates tangential expansion and reduces the uniformity of cellulose microfibril alignment among cells in the root of Arabidopsis.,  Plant Physiol. ,  2004, vol.  135 (pg.  2279- 2290) Google Scholar CrossRef Search ADS PubMed  Becraft PW.  Development of the leaf epidermis.,  Curr. Top. Dev. Biol. ,  1999, vol.  45 (pg.  1- 40) Google Scholar PubMed  Becraft PW,  Li K,  Dey N,  Asuncion-Crabb Y.  The maize dek1 gene functions in embryonic pattern formation and cell fate specification.,  Development ,  2002, vol.  129 (pg.  5217- 5225) Google Scholar PubMed  Becraft PW,  Stinard PS,  McCarty DR.  CRINKLY4: a TNFR-like receptor kinase involved in maize epidermal differentiation.,  Science ,  1996, vol.  273 (pg.  1406- 1409) Google Scholar CrossRef Search ADS PubMed  Brown DM,  Zeef LAH,  Ellis J,  Goodacre R,  Turner SR.  Identification of novel genes in Arabidopsis involved in secondary cell wall formation using expression profiling and reverse genetics.,  Plant Cell ,  2005, vol.  17 (pg.  2281- 2295) Google Scholar CrossRef Search ADS PubMed  Burk DH,  Liu B,  Zhong R,  Morrison WH,  Ye Z.-H.  A katanin-like protein regulated normal cell wall biosynthesis and cell elongation.,  Plant Cell ,  2001, vol.  13 (pg.  807- 827) Google Scholar CrossRef Search ADS PubMed  Burk DH,  Ye Z.-H.  Alteration of oriented deposition of cellulose microfibrils by mutation of a katanin-like microtubule-severing protein.,  Plant Cell ,  2002, vol.  14 (pg.  2145- 2160) Google Scholar CrossRef Search ADS PubMed  Burton RA,  Gibeaut DM,  Bacic A,  Findlay K,  Roberts K,  Hamilton A, et al.  Virus-induced silencing of a plant cellulose synthase gene.,  Plant Cell ,  2000, vol.  12 (pg.  691- 706) Google Scholar CrossRef Search ADS PubMed  Burton RA,  Jobling SA,  Harvey AJ,  Shirley NJ,  Mather DE,  Bacic A, et al.  The genetics and transcriptional profiles of the cellulose synthase-like HvCslF gene family in barley.,  Plant Physiol. ,  2008, vol.  146 (pg.  1821- 1833) Google Scholar CrossRef Search ADS PubMed  Burton RA,  Shirley NJ,  King BJ,  Harvey AJ,  Fincher GB.  The CesA gene family of barley. Quantitative analysis of transcripts reveals two groups of co-expressed genes.,  Plant Physiol. ,  2004, vol.  134 (pg.  224- 236) Google Scholar CrossRef Search ADS PubMed  Carpita NC,  Gibeaut DM.  Structural models of primary cell walls of flowering plants: consistency of molecular structure with the physical properties of the walls during growth.,  Plant J. ,  1993, vol.  3 (pg.  1- 30) Google Scholar CrossRef Search ADS PubMed  Carpita N,  McCann M.  Buchanan B,  Gruissem W,  Jones R. ,  The cell wall. In Biochemistry & Molecular Biology of Plants ,  2000 Rockville, MD American Society of Plant Physiologists(pg.  52- 108) Chandler PM,  Robertson M.  Gibberellin dose–response curves and the characterization of dwarf mutants of barley.,  Plant Physiol. ,  1999, vol.  120 (pg.  623- 632) Google Scholar CrossRef Search ADS PubMed  Chandler PM,  Harding CA,  Ashton AR,  Mulcair MD,  Dixon NE,  Mander LN.  Characterization of gibberellin receptor mutants of barley (Hordeum vulgare L.).,  Molecular Plant ,  2008, vol.  1 (pg.  285- 294) Google Scholar CrossRef Search ADS PubMed  Ching A,  Dhugga KS,  Appenzeller L,  Meeley R,  Bourett TM,  Howard RJ, et al.  Brittle stalk 2 encodes a putative glycosylphosphatidylinositol-anchored protein that affects mechanical strength of maize tissues by altering the composition and structure of secondary cell walls.,  Planta ,  2006, vol.  224 (pg.  1174- 1184) Google Scholar CrossRef Search ADS PubMed  Cleary AL,  Smith LG.  The tangled1 gene is required for spatial control of cytoskeletal arrays associated with cell division during maize leaf development.,  Plant Cell ,  1998, vol.  10 (pg.  1875- 1888) Google Scholar CrossRef Search ADS PubMed  Collard BCY,  Jahufer MZZ,  Brouwer JB,  Pang ECK.  An introduction to markers, quantitative trait loci (QTL) mapping and marker-assisted selection for crop improvement: the basic concepts.,  Euphytica ,  2005, vol.  142 (pg.  169- 196) Google Scholar CrossRef Search ADS   Cosgrove DJ.  Assembly and enlargement of the primary cell wall in plants.,  Annu. Rev. Cell Dev. Biol. ,  1997, vol.  13 (pg.  171- 201) Google Scholar CrossRef Search ADS PubMed  Dahleen LS,  Vander Wal LJ,  Franckowiak JD.  Characterization and molecular mapping of genes determining semidwarfism in barley.,  J. Hered. ,  2005, vol.  96 (pg.  654- 662) Google Scholar CrossRef Search ADS PubMed  Delmer DP,  Eshed R,  Hogan P,  Doblin M,  Jacob-Wilk D,  Peng L, et al.  Expression analysis of the ten cellulose synthase (CesA) genes of Arabidopsis thaliana.,  Plant Biol. ,  2001, vol.  2001 pg.  114  Eleftheriou EP,  Karagiannidou T,  Hudak J,  Moustakas M.  Developmental malformations of the protophloem sieve-element cell wall in oryzalin-treated roots of wheat.,  Biologia ,  2003, vol.  58 (pg.  833- 841) Emons AMC,  Höfte H,  Mulder BM.  Microtubules and cellulose microfibrils: how intimate is their relationship?,  Trends Plant Sci. ,  2008, vol.  12 pg.  279  Google Scholar CrossRef Search ADS   Encina A,  Sevillano JM,  Acebes JL,  Alvarez J.  Cell wall modifications of bean (Phaseolus vulgaris) cell suspensions during habituation and dehabituation to dichlobenil.,  Physiol. Plant. ,  2002, vol.  114 (pg.  182- 191) Google Scholar CrossRef Search ADS PubMed  Fagard M,  Desnos T,  Desprez T,  Goubet F,  Refregier G,  Mouille G, et al.  PROCUSTE1 encodes a cellulose synthase required for normal cell elongation specifically in roots and dark-grown hypocotyls of Arabidopsis.,  Plant Cell ,  2000, vol.  12 (pg.  2409- 2423) Google Scholar CrossRef Search ADS PubMed  Fedak G,  Tsuchiya T,  Helgason SB.  Use of monotelotrisomics for linkage mapping in barley.,  Can. J. Genet. Cytol. ,  1972, vol.  14 (pg.  949- 957) Google Scholar CrossRef Search ADS   Franckowiak JD.  Revised linkage maps for morphological markers in barley, Hordeum vulgare.,  Barley Genet. Newslett. ,  1997, vol.  26 (pg.  9- 21) Furutani I,  Watanabe Y,  Prieto R,  Masukawa M,  Suzuki K,  Naoi K, et al.  The SPIRAL genes are required for directional control of cell elongation in Arabidopsis thaliana.,  Development ,  2000, vol.  127 (pg.  4443- 4453) Google Scholar PubMed  Giddings THJr,  Staehelin LA.  Spatial relationship between microtubules and plasma membrane rosettes during the deposition of primary wall microfibrils in Closterium sp.,  Planta ,  1988, vol.  173 (pg.  22- 30) Google Scholar CrossRef Search ADS PubMed  Gordon KC.  Bancroft JD,  Stevens A.  Tissue processing,  Theory and Practice of Histological Techniques. ,  1990 Edinburgh Churchill Livingstone(pg.  43- 59) Hayashi J,  Konishi T,  Moriya I,  Takahashi R.  Inheritance and linkage studies in barley. VI. Ten mutants genes located on chromosomes 1 to 7, except 3.,  Ber. Ohara Ins. Landw. Biol., Okayama Univ. ,  1984, vol.  18 (pg.  227- 250) Hayden MJ,  Nguyen TM,  Waterman A,  McMichael GL,  Chalmers KJ.  Application of multiplex-ready PCR for fluorescence-based SSR genotyping in barley and wheat.,  Molecular Breeding ,  2008, vol.  21 (pg.  271- 281) Google Scholar CrossRef Search ADS   Held MA,  Penning B,  Brandt AS,  Kessans SA,  Yong W,  Scofield SR, et al.  Small-interfering RNAs from natural antisense transcripts derived from a cellulose synthase gene modulate cell wall biosynthesis in barley.,  Proc. Natl Acad. Sci. USA ,  2008, vol.  105 (pg.  20534- 20539) Google Scholar CrossRef Search ADS   Helliwell CA,  Chandler PM,  Poole A,  Dennis ES,  Peacock WJ.  The CYP88A cytochrome P450, ent-kaurenoic acid oxidase, catalyzes three steps of the gibberellin biosynthesis pathway.,  Proc. Natl Acad. Sci. USA ,  2001, vol.  98 (pg.  2065- 2070) Google Scholar CrossRef Search ADS   Himmelspach R,  Williamson RE,  Wasteneys GO.  Cellulose microfibril alignment recovers from DCB-induced disruption despite microtubule disorganization.,  Plant J. ,  2003, vol.  36 (pg.  565- 575) Google Scholar CrossRef Search ADS PubMed  His I,  Driouich A,  Nicol F,  Jauneau A,  Höfte H.  Altered pectin composition in primary cell walls of korrigan, a dwarf mutant of Arabidopsis deficient in a membrane-bound endo-1,4-β-glucanase.,  Planta ,  2001, vol.  212 (pg.  348- 358) Google Scholar CrossRef Search ADS PubMed  Hogetsu T,  Shibaoka H.  Effects of colchicine on cell shape and on microfibril arrangement in the cell wall of Closterium acerosum.,  Planta ,  1978, vol.  140 (pg.  15- 18) Google Scholar CrossRef Search ADS PubMed  Hülskamp M,  Folkers U,  Grini PE.  Cell morphogenesis in Arabidopsis.,  BioEssays ,  1998, vol.  20 (pg.  20- 29) Google Scholar CrossRef Search ADS PubMed  Inada S,  Shimmen T.  Regulation of elongation growth by gibberellin in root segments of Lemna minor.,  Plant Cell Physiol. ,  2000, vol.  41 (pg.  932- 939) Google Scholar CrossRef Search ADS PubMed  Ishida T,  Hashimoto T.  An Arabidopsis thaliana mutant with conditional root-skewing phenotype.,  J Plant Res. ,  2007, vol.  120 (pg.  635- 640) Google Scholar CrossRef Search ADS PubMed  Ishida T,  Kaneko Y,  Iwano M,  Hashimoto T.  Helical microtubule arrays in a collection of twisting mutants of Arabidopsis thaliana.,  Proc. Natl Acad. Sci. USA ,  2007, vol.  104 (pg.  8544- 8549) Google Scholar CrossRef Search ADS   Kato Y,  Iki K,  Matsuda K.  Cell-wall polysaccharides of immature barley plants. I. Isolation and characterization of a β-d-glucan.,  Agric. Biol. Chem. ,  1981, vol.  45 (pg.  2745- 2753) Kessler S,  Sriki S,  Sinha N.  Xcl1 causes delayed oblique periclinal divisions in developing maize leaves, leading to cellular differentiation by lineage instead of position.,  Development ,  2002, vol.  129 (pg.  1859- 1869) Google Scholar PubMed  Kim JB,  Carpita NC.  Changes in esterification of the uronic acid groups of cell wall polysaccharides during elongation of maize coleoptiles.,  Plant Physiol. ,  1992, vol.  98 (pg.  646- 653) Google Scholar CrossRef Search ADS PubMed  Kokubo A,  Kuraishi S,  Sakurai N.  Culm strength in barley: correlation among maximum bending stress, cell wall dimensions, and cellulose content.,  Plant Physiol. ,  1989, vol.  91 (pg.  876- 882) Google Scholar CrossRef Search ADS PubMed  Lane DR,  Wiedemeier A,  Peng L,  Höfte H,  Vernhettes S,  Desprez T, et al.  Temperature-sensitive alleles of RSW2 link KORRIGAN endo-1,4-β-glucanase to cellulose synthesis and cytokinesis in Arabidopsis.,  Plant Physiol. ,  2001, vol.  126 (pg.  278- 288) Google Scholar CrossRef Search ADS PubMed  Li Y,  Qian Q,  Zhou Y,  Yan M,  Sun L,  Zhang M, et al.  BRITTLE CULM1, which encodes a COBRA-like protein, affects the mechanical properties of rice plants.,  Plant Cell ,  2003, vol.  15 (pg.  2020- 2031) Google Scholar CrossRef Search ADS PubMed  Lloyd C,  Chan J.  The parallel lives of microtubules and cellulose microfibrils.,  Curr. Opin. Plant Biol. ,  2008, vol.  11 (pg.  641- 646) Google Scholar CrossRef Search ADS PubMed  Manly KF,  Cudmore RHJr,  Meer JM.  Map Manager QTX, cross-platform software for genetic mapping.,  Mamm. Genome ,  2001, vol.  12 (pg.  930- 932) Google Scholar CrossRef Search ADS PubMed  Matthews DE,  Carollo VL,  Lazo GR,  Anderson OD.  GrainGenes, the genome database for small-grain crops.,  Nucleic Acids Res. ,  2003, vol.  31 (pg.  183- 186) Google Scholar CrossRef Search ADS PubMed  Metcalfe CR. ,  Anatomy of the Monocotyledons. I. Gramineae. ,  1960 Oxford Clarendon Press Michelmore RW,  Paran T,  Kesseli RV.  Identification of markers linked to disease-resistance genes by bulked segregant analysis: a rapid method to detect markers in specific genomic regions by using segregating populations.,  Proc. Natl Acad. Sci. USA ,  1991, vol.  88 (pg.  9828- 9832) Google Scholar CrossRef Search ADS   Montague MJ.  Gibberellic acid promotes growth and cell wall synthesis in Avena internodes regardless of the orientation of cell expansion.,  Physiol. Plant. ,  1995, vol.  94 (pg.  7- 18) Google Scholar CrossRef Search ADS   Murashige T,  Skoog F.  A revised medium for rapid growth and bioassays with tobacco cultures.,  Physiol. Plant. ,  1962, vol.  15 (pg.  473- 479) Google Scholar CrossRef Search ADS   Nakajima K,  Furutani I,  Tachimoto H,  Matsubara H,  Hashimoto T.  SPIRAL1 encodes a plant-specific microtubule-localized protein required for directional control of rapidly expanding Arabidopsis cells.,  Plant Cell ,  2004, vol.  16 (pg.  1178- 1190) Google Scholar CrossRef Search ADS PubMed  Naoi K,  Hashimoto T.  A semidominant mutation in an Arabidopsis mitogen-activated protein kinase phosphate-like gene compromises cortical microtubule organization.,  Plant Cell ,  2004, vol.  16 (pg.  1841- 1853) Google Scholar CrossRef Search ADS PubMed  Nickle TC,  Meinke DW.  A cytokinesis-defective mutant of Arabidopsis (cyt1) characterized by embryonic lethality, incomplete cell walls, and excessive callose accumulation.,  Plant J. ,  1998, vol.  15 (pg.  321- 332) Google Scholar CrossRef Search ADS PubMed  Pagant S,  Bichet A,  Sugimoto K,  Lerouxel O,  Desprez T,  McCann M, et al.  KOBITO1 encodes a novel plasma membrane protein necessary for normal synthesis of cellulose during cell expansion in Arabidopsis.,  Plant Cell ,  2002, vol.  14 (pg.  2001- 2013) Google Scholar CrossRef Search ADS PubMed  Paradez AR,  Somerville CR,  Ehrhardt DW.  Visualisation of cellulose synthase demonstrates functional association with microtubules.,  Science ,  2006, vol.  312 (pg.  1491- 1495) Google Scholar CrossRef Search ADS PubMed  Paris M,  Carter M.  Cereal DNA: a rapid high-throughput extraction method for marker assisted selection.,  Plant Mol. Biol. Rep. ,  2000, vol.  18 (pg.  357- 360) Google Scholar CrossRef Search ADS   Peng L,  Hocart CH,  Redmond JW,  Williamson RE.  Fractionation of carbohydrates in Arabidopsis root cell walls shows that three radial swelling loci are specifically involved in cellulose production.,  Planta ,  2000, vol.  211 (pg.  406- 414) Google Scholar CrossRef Search ADS PubMed  Reynolds JO,  Eisses JF,  Sylvester AW.  Balancing division and expansion during maize leaf morphogenesis: analysis of the mutant, warty-1.,  Development ,  1998, vol.  125 (pg.  259- 268) Google Scholar PubMed  Robinson DG,  Quader H.  Lloyd CW.  The microtubule-microfibril syndrome,  The Cytoskeleton in Plant Growth and Development. ,  1982 London Academic Press(pg.  109- 126) Roudier F,  Fernandez AG,  Fujita M,  Himmelspach R,  Borner GHH,  Schindelman G, et al.  COBRA, an Arabidopsis extracellular glycosyl-phosphatidyl inositol-anchored protein, specifically controls highly anisotropic expansion through its involvement in cellulose microfibril orientation.,  Plant Cell ,  2005, vol.  17 (pg.  1749- 1763) Google Scholar CrossRef Search ADS PubMed  Sabba RP,  Durso NA,  Vaughn KC.  Structural and immunocytochemical characterization of the walls of dichlobenil-habituated BY-2 tobacco cells.,  Int. J. Plant Sci. ,  1999, vol.  160 (pg.  275- 290) Google Scholar CrossRef Search ADS   Sato S,  Kato T,  Kakegawa K,  Ishii T,  Liu Y.-G,  Awano T, et al.  Role of the putative membrane-bound endo-1,4-β-glucanase KORRIGAN in cell elongation and cellulose synthesis in Arabidopsis thaliana.,  Plant Cell Physiol. ,  2001, vol.  42 (pg.  251- 263) Google Scholar CrossRef Search ADS PubMed  Schindelman G,  Morikami A,  Jung J,  Baskin TI,  Carpita NC,  Derbyshire P, et al.  COBRA encodes a putative GPI-anchored protein, which is polarly localized and necessary for oriented cell expansion in Arabidopsis.,  Genes Dev. ,  2001, vol.  15 (pg.  1115- 1127) Google Scholar CrossRef Search ADS PubMed  Sedbrook JC,  Kaloriti D.  Microtubules, MAPs and plant directional cell expansion.,  Trends Plant Sci. ,  2008, vol.  13 (pg.  303- 310) Google Scholar CrossRef Search ADS PubMed  Shea EM,  Gibeaut DM,  Carpita NC.  Structural analysis of the cell walls regenerated by carrot protoplasts.,  Planta ,  1989, vol.  179 (pg.  293- 308) Google Scholar CrossRef Search ADS PubMed  Shedletzky E,  Shmuel M,  Delmer DP,  Lamport DTA.  Adaptation and growth of tomato cells on the herbicide 2,6-dichlorobenzonitrile leads to production of unique cell walls virtually lacking a cellulose–xyloglucan network.,  Plant Physiol. ,  1990, vol.  94 (pg.  980- 987) Google Scholar CrossRef Search ADS PubMed  Shedletzky E,  Shmuel M,  Trainin T,  Kalman S,  Delmer D.  Cell wall structure in cells adapted to growth on the cellulose-synthesis inhibitor 2,6-dichlorobenzonitrile.,  Plant Physiol. ,  1992, vol.  100 (pg.  120- 130) Google Scholar CrossRef Search ADS PubMed  Shen B,  Li C,  Min Z,  Meeley RB,  Tarczynski MC,  Olsen O.-A.  sal1 determines the number of aleurone cell layers in maize endosperm and encodes a class E vacuolar sorting protein.,  Proc. Natl Acad. Sci. USA ,  2003, vol.  100 (pg.  6552- 6557) Google Scholar CrossRef Search ADS   Sims IM,  Bacic A.  Extracellular polysaccharides from suspension cultures of Nicotiana plumbaginifolia.,  Phytochemistry ,  1995, vol.  38 (pg.  1397- 1405) Google Scholar CrossRef Search ADS   Smith BG,  Harris PJ.  The polysaccharide composition of Poales cell walls: Poaceae cell walls are not unique.,  Biochem. System. Ecol. ,  1999, vol.  27 (pg.  33- 53) Google Scholar CrossRef Search ADS   Smith LG,  Gerttula SM,  Han S,  Levy J.  TANGLED1: a microtubule binding protein required for the spatial control of cytokinesis in maize.,  J. Cell Biol. ,  2001, vol.  152 (pg.  231- 236) Google Scholar CrossRef Search ADS PubMed  Taiz L.  Plant cell expansion: regulation of cell wall mechanical properties.,  Annu. Rev. Plant Physiol. ,  1984, vol.  35 (pg.  585- 657) Google Scholar CrossRef Search ADS   Takeda K,  Shibaoka H.  Effects of gibberellin and colchicine on microfibril arrangement in epidermal cell walls of Vigna angularis Ohwi et Ohashi epicotyls.,  Planta ,  1981, vol.  151 (pg.  393- 398) Google Scholar CrossRef Search ADS PubMed  Tanaka K,  Murata K,  Yamazaki M,  Onosato K,  Miyao A,  Hirochika H.  Three distinct rice cellulose synthase catalytic subunit genes required for cellulose synthesis in the secondary wall.,  Plant Physiol. ,  2003, vol.  133 (pg.  73- 83) Google Scholar CrossRef Search ADS PubMed  Thitamadee S,  Tuchihara K,  Hashimoto T.  Microtubule basis for left-handed helical growth in Arabidopsis.,  Nature ,  2002, vol.  417 (pg.  193- 196) Google Scholar CrossRef Search ADS PubMed  United States Department of Agriculture–Agriculture Research Service,  National Small Grains Collection ,  2005  www.ars.usda.gov. Voorrips RE.  MapChart: software for the graphical presentation of linkage maps and QTLs.,  J. Hered. ,  2002, vol.  93 (pg.  77- 78) Google Scholar CrossRef Search ADS PubMed  Wasteneys GO.  The cytoskeleton and growth polarity.,  Curr. Opin. Plant Biol. ,  2000, vol.  3 (pg.  503- 511) Google Scholar CrossRef Search ADS PubMed  Wasteneys GO.  Progress in understanding the role of microtubules in plant cells.,  Curr. Opin. Plant Biol. ,  2004, vol.  7 (pg.  651- 660) Google Scholar CrossRef Search ADS PubMed  Wasteneys GO,  Fujita M.  Establishing and maintaining axial growth: wall mechanical properties and the cytoskeleton.,  J. Plant Res. ,  2006, vol.  119 (pg.  5- 10) Google Scholar CrossRef Search ADS PubMed  Wenzel CL,  Chandler PM,  Cunningham RB,  Passioura JB.  Characterization of the leaf epidermis of barley (Hordeum vulgare L. 'Himalaya').,  Ann. Bot. ,  1997, vol.  79 (pg.  41- 46) Google Scholar CrossRef Search ADS   Whittington AT,  Vugrek O,  Wei KJ,  Hasenbein NG,  Sugimoto K,  Rashbrooke MC, et al.  MOR1 is essential for organizing cortical microtubules in plants.,  Nature ,  2001, vol.  411 (pg.  610- 613) Google Scholar CrossRef Search ADS PubMed  Yeo U.-D,  Soh W.-Y,  Tasaka H,  Sakurai N,  Kuraishi S,  Takeda K.  Cell wall polysaccharides of callus and suspension-cultured cells from three cellulose-less mutants of barley (Hordeum vulgare L.).,  Plant Cell Physiol. ,  1995, vol.  36 (pg.  931- 936) Google Scholar CrossRef Search ADS   Yuen CYL,  Pearlman RS,  Silo-suh L,  Hilson P,  Carroll KL,  Masson PH.  WVD2 and WDL1 modulate helical organ growth and anisotropic cell expansion in Arabidopsis.,  Plant Physiol. ,  2003, vol.  131 (pg.  493- 506) Google Scholar CrossRef Search ADS PubMed  Yulia D.  Characterisation of DCB-habituated suspension cultured cells of barley (Hordeum vulgare L. cv Schooner),  PhD thesis ,  2006 Melbourne, Australia School of Botany, University of Melbourne Zuo J,  Niu Q.-W,  Nishizawa N,  Wu Y,  Kost B,  Chua N.-H.  KORRIGAN, an Arabidopsis endo-1,4-β-glucanase, localizes to the cell plate by polarized targeting and is essential for cytokinesis.,  Plant Cell ,  2000, vol.  12 (pg.  1137- 1152) Google Scholar PubMed  Abbreviations: Abbreviations: BSA bovine serum albumin CESA cellulose synthase CSLF CELLULOSE SYNTHASE-LIKE F DCB 2,6-dichlorobenzonitrile GA gibberellin GA3 gibberellic acid GAX glucuronoarabinoxylan HG homogalacturonan MLG mixed-linkage (1→3), (1→4)-β-d-glucan MT microtubule PBS phosphate-buffered saline RG rhamnogalacturonan SEM scanning electron microscopy SSR simple sequence repeat VIGS virus-induced gene silencing XG xyloglucan © The Author 2009. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists. All rights reserved. For permissions, please email: journals.permissions@oxfordjournals.org
TI - Aberrant Cell Expansion in the elongation Mutants of Barley
JF - Plant and Cell Physiology
DO - 10.1093/pcp/pcp015
DA - 2009-01-30
UR - https://www.deepdyve.com/lp/oxford-university-press/aberrant-cell-expansion-in-the-elongation-mutants-of-barley-a4fZVqpFDB
SP - 554
EP - 571
VL - 50
IS - 3
DP - DeepDyve
ER -