Impact of water deficiency on leaf cuticle lipids and gene expression networks in cotton (Gossypium hirsutum L.)

Abstract

Background: Water deficit (WD) has serious effect on the productivity of crops. Formation of cuticular layer with increased content of wax and cutin on leaf surfaces is closely related to drought tolerance. Identification of drought tolerance associated wax components and cutin monomers and the genes responsible for their biosynthesis is essential for understanding the physiological and genetic mechanisms underlying drought tolerance and improving crop drought resistance.

Result: In this study, we conducted comparative phenotypic and transcriptomic analyses of two Gossypium hirsutum varieties that are tolerant (XL22) or sensitive (XL17) to drought stress. XL17 consumed more water than XL22, particularly under the WD conditions. WD significantly induced accumulation of most major wax components (C29 and C31 alkanes) and cutin monomers (palmitic acid and stearic acid) in leaves of both XL22 and XL17, although accumulation of the major cutin monomers, i.e., polyunsaturated linolenic acid (C18:3n-3) and linoleic acid (C18:2n-6), were significantly repressed by WD in both XL22 and XL17. According to the results of transcriptome analysis, although many genes and their related pathways were commonly induced or repressed by WD in both XL22 and XL17, WD-induced differentially expressed genes specific to XL22 or XL17 were also evident. Among the genes that were commonly induced by WD were the GhCER1 genes involved in biosynthesis of alkanes, consistent with the observation of enhanced accumulation of alkanes in cotton leaves under the WD conditions. Interestingly, under the WD conditions, several GhCYP86 genes, which encode enzymes catalyzing the omega-hydroxylation of fatty acids and were identified to be the hub genes of one of the co-expression gene modules, showed a different expression pattern between XL22 and XL17 that was in agreement with the WD-induced changes of the content of hydroxyacids or fatty alcohols in these two varieties.

Conclusion: The results contribute to our comprehending the physiological and genetic mechanisms underlying drought tolerance and provide possible solutions for the difference of drought resistance of different cotton varieties.

Keywords: Co-expression gene network; Cutin monomers; Differentially expressed genes; Gossypium hirsutum; Transcriptomic analysis; Wax components.

Fig. 1

The effect of WD on the physiological and morphological characteristics of cotton variety tolerant (XL22) or sensitive (XL17) to drought. A Comparison of plant height. B Comparison of leaf area. C the relative water content of leaves. SS, seedling stage; BS, bud stage. WW, well-watered; WD, water deficit. Error bars are standard errors. Values represent the means ± SE, n = 3. Different letters above the bars indicate statistically different from each other as determined by the Student’s t test: p < 0.05

Fig. 2

Comparison of the leaf stomatal density and stomatal aperture of cotton variety tolerant (XL22) or sensitive (XL17) to drought under the WW andWD conditions. A Leaf stomatal density of the two cotton varieties B Leaf stomatal aperture of the two cotton varieties. C Leaf impressions showing leaf stomatal apertere at two stages under WW and WD conditions on abaxial and adaxial surfaces. a and b. XL22 leaf at the seedling stage under WW and WD condition, respectively. c and d. XL22 leaf at the bud stage under WW and WD condition, respectively. e and f. XL17 leaf at the seedling stage under WW and WD condition, respectively. g and h. XL17 leaf at the bud stage under WW and WD condition, respectively. i, j, k and l, the adaxial surface of XL22 leaf corresponds to a, b, c and d. m, n, o and p, the adaxial surface of XL17 leaf corresponds to e, f, g and h. SS-22, XL22 leaves at seedling stage; SS-17, XL17 leaves at seedling stage; BS-22, XL22 leaves at bud stage; BS-17, XL17 leaves at bud stage. Error bars are standard errors. Values represent the means ± SE, n = 3. The data in Fig. 2A and B were analyzed by analysis of variance (ANOVA) and mean comparison between treatments was performed based on Duncan's multiple range method at 5% level. For Fig. 2A and B, different letters above the bars indicate statistically different (p < 0.05)

Fig. 3

The effect of WD on the accumulation of wax constituents in XL22 and XL17 leaves. A and C The main wax constituents in XL22 leaves at the seedling stage and bud stage, respectively. B and D The main wax constituents in XL17 leaves at the seedling stage and bud stage, respectively. C27 ALK, heptacosane; C28 ALK, octacosane; C29 ALK, nonacosane; C30 ALK, triacontane; C31 ALK, hentriacontane; C33 ALK, tritridecane. C16:0, hexadecanoic acid; C18:0, octadecanoic acid; C18:1n-9, 9-octadecenoic acid; C18:2n-6, 9,12-octadecadienoic acid. C28-OH, octacosanol; C30-OH, triacontanol; C32-OH, dotriacontanol. WW, well-watered; WD, water deficit. Error bars are standard errors. Values represent the means ± SE, n = 3. Asterisks denote significant difference as determined by the Student’s t test: *p < 0.05. XL22/SS, XL22 leaves at seedling stage; XL17/SS, XL17 leaves at seedling stage; XL22/BS, XL22 leaves at bud stage; XL17/BS, XL17 leaves at bud stage

Fig. 4

The effect of WD on the accumulation of cutin monomers in XL22 and XL17 leaves. A and C The main cutin monomers in XL22 leaves at the seedling stage and bud stage, respectively. B and D The main cutin monomers in XL17 leaves at the seedling stage and bud stage, respectively. C16:0, hexadecanoic acid; C18:0, octadecanoic acid; C18:1n-9, 9-octadecenoic acid; C18:2n-6, 9,12-octadecadienoic acid. C18:3n-3, 9,12,15-octadecatrienoic acid. C16:0 DCA, hexadecane-1,16-dioic acid; C18:0 DCA, octadecanedioic acid. C26-OH, hexacosanol; C28-OH, octacosanol. WW, well-watered; WD, water deficit. Error bars are standard errors. Values represent the means ± SE, n = 3. Asterisks denote significant difference as determined by the Student’s t test: *p < 0.05. XL22/SS, XL22 leaves at seedling stage; XL17/SS, XL17 leaves at seedling stage; XL22/BS, XL22 leaves at bud stage; XL17/BS, XL17 leaves at bud stage

Fig. 5

The number and distribution of DEGs under the WW and WD conditions. A The number of the up-regulated or down-regulated DEGs (WD vs WW) in XL22 and XL17. B Venn diagram showing overlapping and unique DEGs in different varieties and developmental stages. C The distribution of DEGs with different fold changes. SS-22, XL22 leaves at seedling stage; SS-17, XL17 leaves at seedling stage; BS-22, XL22 leaves at bud stage; BS-17, XL17 leaves at bud stage

Fig. 6

Gene ontology classification of DEGs between the well-watered and water deficit XL22 and XL17 plants. A Venn diagram showing the number of enriched GO terms overlapping or specific to each developmental stage of the two cotton varieties. B Enriched GO terms at the seedling stage and the bud stage of XL22 and XL17. C Enriched GO terms at the seedling stage and the bud stage of XL17. SS-22, the seedling stage of XL22; SS-17, the seedling stage of XL17; BS-22, the bud stage of XL22; BS-17, the bud stage of XL17

Fig. 7

Expression profiles of the DEGs involved in biosynthesis of fatty acids, wax and cutinin XL22 and XL17. Left panel, the biosynthesis pathways of fatty acids, wax and cutin. Right panel, heatmap (the log2(FPKM value)) showing the expression patterns of the major genes of the biosynthesis pathways of fatty acids, wax and cutin. GhCYP86, Cytochrome P450 86A gene; GhSAD, stearoyl-ACP desaturase gene; GhFAD2, fatty acid desaturase 2 gene; GhFAD3, fatty acid desaturase 3 gene; GhPXG, peroxygenase gene; GhCER1, aldehyde decarbonylase gene; GhFAR, fatty acyl reductase gene; GhALDH, aldehyde dehydrogenase; C16:0-ACP, palmitic acid; C18:0-ACP, stearic acid; C18:1-ACP, oleic acid; C18:2-ACP, linoleic acid; C18:3-ACP, α-linolenic acid. S/W: well-watered plants at the seedling stage, S/D: water deficit plants at the seedling stage, B/W: well-watered plants at the bud stage, B/D: water deficit plants at the bud stage

Fig. 8

WGCNA of DEGs between the well-watered and WD plants of XL22 and XL17. A Hierarchical clustering tree of co-expression modules was analyzed by WGCNA and DEGs was divided into 13 modules that were defined by different color. B The correlation between modules and traits. Each row represents a module and each column corresponds to trait. The number in the rectangular box represents the correlation coefficient and corresponding p-value. Red represents the positive correlation and blue represents the negative correlation between the module and the trait. TA: well-watered XL22 at the seedling stage; TB: water deficit XL22 at the seedling stage; TC: well-watered XL17 at the seedling stage; TD: water deficit XL17 at the seedling stage; TE: well-watered XL22 at the bud; TF: water deficit XL22 at the bud stage; TG: well-watered XL17 at the bud stage; TH: water deficit XL17 at the bud stage

Fig. 9

Co-expression network analysis. A Co-expression network analysis results of the black module. B Co-expression network analysis results of the turquoise module. Red circles represent the hub genes. Circles size and color represent the degree. Line size represents the weight