Comparative fiber property and transcriptome analyses reveal key genes potentially related to high fiber strength in cotton (Gossypium hirsutum L.) line MD52ne

Abstract

Background: Individual fiber strength is an important quality attribute that greatly influences the strength of the yarn spun from cotton fibers. Fiber strength is usually measured from bundles of fibers due to the difficulty of reliably measuring strength from individual cotton fibers. However, bundle fiber strength (BFS) is not always correlated with yarn strength since it is affected by multiple fiber properties involved in fiber-to-fiber interactions within a bundle in addition to the individual fiber strength. Molecular mechanisms responsible for regulating individual fiber strength remain unknown. Gossypium hirsutum near isogenic lines (NILs), MD52ne and MD90ne showing variations in BFS provide an opportunity for dissecting the regulatory mechanisms involved in individual fiber strength.

Results: Comprehensive fiber property analyses of the NILs revealed that the superior bundle strength of MD52ne fibers resulted from high individual fiber strength with minor contributions from greater fiber length. Comparative transcriptome analyses of the NILs showed that the superior bundle strength of MD52ne fibers was potentially related to two signaling pathways: one is ethylene and the interconnected phytohormonal pathways that are involved in cotton fiber elongation, and the other is receptor-like kinases (RLKs) signaling pathways that are involved in maintaining cell wall integrity. Multiple RLKs were differentially expressed in MD52ne fibers and localized in genomic regions encompassing the strength quantitative trait loci (QTLs). Several candidate genes involved in crystalline cellulose assembly were also up-regulated in MD52ne fibers while the secondary cell wall was produced.

Conclusion: Comparative phenotypic and transcriptomic analyses revealed differential expressions of the genes involved in crystalline cellulose assembly, ethylene and RLK signaling pathways between the MD52ne and MD90ne developing fibers. Ethylene and its phytohormonal network might promote the elongation of MD52ne fibers and indirectly contribute to the bundle strength by potentially improving fiber-to-fiber interactions. RLKs that were suggested to mediate a coordination of cell elongation and SCW biosynthesis in other plants might be candidate genes for regulating cotton fiber cell wall assembly and strength.

Fig. 1

Fiber property analyses of NILs and their F₂ progeny. a Frequency distribution of bundle fiber strength in an F₂ population from a cross between MD90ne and MD52ne. Average bundle strength value of each F₂ progeny was obtained from five replicates by HVI. Minimum, median (M), and maximum bundle fiber strengths were 31.99, 37.22, and 42.66 g/tex among 384 F₂ progenies. b Comparisons of breaking force distributions of individual fibers from MD52ne and MD90ne fibers. Breaking forces (cN) of 303 individual fibers were determined from mature MD52ne and MD90ne fibers having similar fiber maturity with an identical MIC value (4.931). A 13 mm length gauge was used for breaking fiber with Favimat instrument

Fig. 2

Comparisons of fiber properties from developing MD52ne and MD90ne fibers. Developing fibers at nine different developmental time points (10, 13, 15, 17, 20, 24, 28, 33, and 37 DPA) and full developed fibers at 44 DPA were collected from NILs grown side by side in replicated fields in New Orleans, LA in 2013. Asterisks next to the time points denote statistical significance. a Fiber length. Average lengths of developing fibers at different DPAs were calculated from 30 replicates. b Crystallinity. Average crystallinity (CI _IR) was determined from six replicates of ATR-FTIR spectra. c Maturity. Average fiber maturity (M _IR) was determined from six replicates using ATR-FTIR. d Fineness. Three hundred fibers of 15 mm length were used for each replicate of gravimetric fineness analyses. Average fineness values were calculated from three replicates. e Bundle fiber strength. Average bundle fiber strength of developing fibers was measured from three replicates by Stelometer

Fig. 3

Comparisons of fiber maturity ratios of mature fibers between MD52ne and MD90ne measured by image analysis microscopy. Circularity (θ = 4πA/P²) representing the degree of fiber cell wall thickness was calculated from average wall area (A) excluding lumen and perimeter (P) of cross-sections of 300 fibers. The maturity ratios between MD52ne (0.948 ± 0.010) and MD90ne (0.971 ± 0.033) calculated from the circularities showed no significant variation (p-value, 0.519). A scale bar represents 10 μm. MD52ne, MD90ne

Fig. 4

Summary of RNA-seq analysis comparing MD52ne and MD90ne. a Venn diagrams representing differentially expressed genes (DEGs) at two developmental stages (15 and 20 DPA) in MD52ne and MD90ne fibers. b Comparisons of up- or down-regulated DEGs at 15 and 20 DPA in MD52ne fibers. c GO analysis. Singular enrichment analysis was used to identify GO categories that were commonly found at both 15 and 20 DPA developing fibers from MD52ne. The color and numbers adjacent to the GO identifier represent p-values

Fig. 5

Differential expressions of receptor-like kinases (RLKs) in developing MD52ne fibers at 15 DPA. Differential expressions of RLKs in different classes were generated by MapMan. Red and purple represent up- and down- regulations, respectively. The RLKs contain three domains, including extracellular domain, transmembrane (TM), and kinase domain in a cytoplasmic side. C-lectin, RLKs with lectin-like motifs; Crinkly4-like, RLKs with crinkly4-like domains; DUF26, domain of unknown function 26; Extensin, RLK with extensin motif; L-lectin, RLKs with lectin-binding domains; LRK 10-like, RLK gene on Lr10 locus; LRR, leucine-rich repeats; LysM, RLKs with lysine motif; PERK-like, proline-rich extensin-like kinase; RKF3-like, receptor-like kinase in flowers 3; S-locus, RLK with S-domain; Thaumatin, RLK-like thaumatin protein; WAK, wall-associated kinase

Fig. 6

RT-qPCR validation of selected genes related to cell wall activity during fiber development. Three biological replications and three technical replications at four fiber development stages (10, 15, and 24 DPA) were used for RT-qPCR analyses. ACO4, 1-aminocyclopropane-1-carboxylate oxidase 4, or ethylene forming enzyme (Gh_A07G0774); EBF1, EIN3-binding F box protein 1 (Gh_A05G0085); LRR RLK, Leucine rich repeat receptor-like kinase (Gh_D08G0203); MBY46 transcription factor (Gh_A09G1074); CesA4, Cellulose synthase catalytic subunit A4 (Gh_A08G0421); COL5,CONSTANS like 5 (Gh_D08G0923); CHI/CTL, Basic-Chitinase/Chitinase-like (Gh_D06G0439); L18e/L15, Ribosomal L18e/L15 protein (Gh_D02G1619); FAD-linked oxidase (Gh_D02G1214)

Fig. 7

Alignments of DEGs in the physical map of Gossypium hirsutum TM-1 genome and the QTL regions related to bundle fiber strength (A) and UHML fiber length (B). Corresponding genes were highly abundant in MD52ne fibers than MD90ne. Genetic map locations [49] are shown in centiMorgans (cM) and physical locations are shown in base pairs (bp). Red markers were linked with the QTLs. a DEGs in A03 QTL. Gh_A03G1085 (MADS-box transcription factor), Gh_A03G1136 (Response regulator 11), Gh_A03G1184 and Gh_D02G1619 (Ribosomal protein L18e/L15), Gh_A03G1233 (trehalose phosphatase/synthase 9), Gh_A03G1269 (LRR RLK), Gh_A03G1278 (membrane lipoprotein), Gh_A03G1332 (NAC 83), Gh_A03G1432 (XET 9). b DEGs in D08 QTL. Gh_D08G0203 (LRR RLK), Gh_D08G0923 (COL5), Gh_D08G0986 (LRR RLK), Gh_D08G1062 (Protein kinase), Gh_D08G1172 (NAC), Gh_D08G1309 (Sus 3), Gh_D08G1361 (Germin like protein 10), Gh_D08G1424 (WRKY)

Fig. 8

Proposed model for the mechanism responsible for high fiber bundle strength and individual fiber breaking force in MD52ne. Abbreviated names included in this model are: LRR RLK, Leucine rich repeat receptor-like kinase; PCW, primary cell wall; SCW, secondary cell wall; CME, cellulose micrfibrils; ACO, 1-aminocyclopropane-1-carboxylate oxidase; EBF, EIN3-binding F box protein; AUX/IAA, AUX/IAA transcription factor; MYB, MYB transcription factor: WRKY, WRKY transcription factor; NAC, NAC transcription factor; Exp, Expensin; LTP, Lipid transfer protein; PME, Pectin methylesterase; XET, Xyloglucan endotransglucosylase; Cobra, Cobra like protein; CHI/CTL, Chitinase/Chitinase-like protein; ROS, Reactive oxygen species