Knockout of the AtCESA2 gene affects microtubule orientation and causes abnormal cell expansion in Arabidopsis

Abstract

Complete cellulose synthesis is required to form functional cell walls and to facilitate proper cell expansion during plant growth. AtCESA2 is a member of the cellulose synthase A family in Arabidopsis (Arabidopsis thaliana) that participates in cell wall formation. By analysis of transgenic seedlings, we demonstrated that AtCESA2 was expressed in all organs, except root hairs. The atcesa2 mutant was devoid of AtCESA2 expression, leading to the stunted growth of hypocotyls in seedlings and greatly reduced seed production in mature plants. These observations were attributed to alterations in cell size as a result of reduced cellulose synthesis in the mutant. The orientation of microtubules was also altered in the atcesa2 mutant, which was clearly observed in hypocotyls and petioles. Complementary expression of AtCESA2 in atcesa2 could rescue the mutant phenotypes. Together, we conclude that disruption of cellulose synthesis results in altered orientation of microtubules and eventually leads to abnormal plant growth. We also demonstrated that the zinc finger-like domain of AtCESA2 could homodimerize, possibly contributing to rosette assemblies of cellulose synthase A within plasma membranes.

Figure 1.

Phenotype of the atcesa2 mutant. A and B, Ten-day-old seedlings grown on Murashige and Skoog medium. A, Wild type. B, atcesa2 (Ler background). C, Six-day-old dark-grown wild-type (left) and atcesa2 (right) seedlings (two seedlings each) grown in the dark. D and E, Seven-day-old wild-type (D) and atcesa2 (E) seedlings grown vertically on Murashige and Skoog medium in 1.5% agar. F and G, Adult plants grown for 6 weeks (F) and 8 weeks (G) in a greenhouse. F, Wild type (left), atcesa2 (Ler background; bottom right), details of atcesa2 (top right). G, Mature wild type (left) and atcesa2 (right) grown under fluent air conditions. Scale bars = 2 mm (A–E), 1 cm (F and G).

Figure 2.

AtCESA2 was knocked out in the atcesa2 mutant. A, The Ds element was inserted 143 bp upstream of the ATG of AtCESA2. B, Northern blot detected AtCESA2 gene expression in the atcesa2 mutant, as well as in wild type using an AtCESA2 gene-specific probe. C, Complementation of the atcesa2 mutant. a, atcesa2 mutant. b, atcesa2 mutant (pCambia 1301) transgenic vector control. c, atcesa2 mutant (pCambia 1301-AtCESA2). d, Wild-type control. D, Cellulose content of the atcesa2 mutant and wild-type control. Cellulose content was reduced in the mutant leaves rather than in wild-type leaves, but there was almost no change in roots. [See online article for color version of this figure.]

Figure 3.

Histochemical analysis of GUS expression in transgenic Arabidopsis plants carrying the AtCESA2 promoter∷GUS fusion constructs. A, Seedling showing GUS expression in the cotyledon and vascular bundles as well as roots. B, GUS expression in roots. C, Details of GUS expression in root tip region. D, No clear GUS expression was detected in root hairs. E, Flower showing GUS expression in the stamen and stigma. F, Leaf of mature plant showing GUS expression in the same part of vascular bundles. G, Silique showing GUS expression at abscission zone.

Figure 4.

The atcesa2 mutation caused defective cell expansion and elongation. Scanning electron microscopy (SEM; A–H) of 6-d-old wild-type (A) and atcesa2 (B) seedlings. C and D, Epidermal cells of hypocotyl of wild type (C) and atcesa2 (D). E and F, Flower of wild type (E) and atcesa2 (F). G and H, Epidermal cells of stamen filament of wild type (G) and atcesa2 (H). Scale bars = 500 μm (A and B, E, and F), 100 μm (C and D, G and H). Bar chart of the length (I) and epidermal cell number (J) of hypocotyl and stamen filament of wild type and atcesa2. Data represent mean ± se for measurements of 10 seedlings and stamen filaments. Hypocotyls and stamen filaments of wild type are longer than the atcesa2 mutants (I), whereas the epidermal cell numbers of hypocotyls and stamen filaments of wild type are almost the same as atcesa2 mutants (J).

Figure 5.

Toluidine blue-stained sections of Arabidopsis 14-d-old seedling. Roots (A and B), leaves (C–F), and vascular bundles (G and H). Wild type (A, C, E, and G) and atcesa2 mutant (B, D, F, and H). Scale bars = 25 μm (A and B), 50 μm (C and D), 5 μm (E and F), and 50 μm (G and H). [See online article for color version of this figure.]

Figure 6.

Ultrastructure analysis of leaves by transmission electron microscopy. Leaves were taken from 14-d-old light-grown seedlings. Wild type (A and B) and atcesa2 mutant (C–F). Black arrowheads, normal cell walls; white arrowheads, defective cell walls. Scale bars = 5 μm (A and C) and 1 μm (B, D, E, and F). Quantitative analysis of pectin content (G). Each analysis was conducted by three repeat reactions (n = 3).

Figure 7.

Visualization of cortical microtubules in hypocotyls (A–D), petioles (E–H), and roots (I–L) of epidermal cells by confocal microscopy. A, C, E, G, I, and K, View of whole hypocotyls (A and C), petioles (E and G), and root (I and K) with MAP4-GFP expression. A, E, and I, Wild type. C, G, and K, atcesa2 mutants. B, D, F, H, J, and L, View of microtubule orientation in epidermal cells of hypocotyls (B and D), petioles (F and H), and roots (J and L) with MAP4-GFP expression. B, F, and J, Wild type. D, H, and L, atcesa2 mutants. M and N, View of microtubule orientation in epidermal cells of wild-type (M) and mutant (N) hypocotyls growing in the dark for 3 d with immunofluorescence microscopy with a mouse anti-β-tubulin monoclonal antibody. Scale bars = 100 μm (A, C, E, and G), 50 μm (I and K), 10 μm (B, D, F, H, M, and N), and 5 μm (J and E).

Figure 8.

The dimerizing domain of AtCESA2. A, Alignment of the CESA members containing zinc finger domains. Black triangle indicates the conserved Cys within the zinc finger domain. B, Interactions between AtCESA2 and itself in yeast. Schematic representation of the AtCESA2 N-terminal fragment, zinc finger domain, and whole protein (top). Yeast transformed with plasmids containing the GAL-4 binding domain (pGBKT7) and the GAL-4 activating domain (pGADT7); fusion proteins were grown on synthetic complete medium minus Leu and Trp (His+) and minus Leu, Trp, and His, plus 5 mm 3-amino 1,2,4-triazole (5 mm 3-AT). β-Galactosidase activity was assayed according to the standard protocol provided by CLONTECH. [See online article for color version of this figure.]