Correlation Analysis of Stem Hardness Traits with Fiber and Yield Related Traits in Core Collections of Gossypium hirsutum

Background Stem hardness is one of the major inuencing factors for plant architecture in upland cotton (Gossypium hirsutum L.). Evaluating hardness phenotypic traits is very important for the selection elite lines for resistant to lodging in Gossypium hirsutum L. Cotton breeder are interested in using diverse genotypes to enhance bre quality and high- yield. The research for hardness and its relation with ber quality and yield were very few. This study was designed to nd the relationship of stem hardness traits with ber quality and yield contributing traits of upland cotton. Results Experiments were carried out to measure the bending, acupuncture and compression properties of stem from a collection of upland cotton genotypes, comprising 237 accessions. The results showed that the genotypic difference in stem hardness were highly signicant among the genotypes, and the stem hardness traits (BL, BU, AL, AU, CL and CU) have a positive association with ber quality traits and yield related traits. In descriptive statistics result bending (BL, BU) have maximum coecient of variance and trait ber length and ber strength have less coecient of variance among the genotypes. Principal component analysis (PCA) reduced quantitative characters into nine principal components. The rst nine principal components (PC) with Eigen values >1 explained 0.86% of variation among 237 accessions of cotton crop. Both 2017& 2018, PCA results indicated that BL, BU, FL, FE and LI variables contributed their variability in PC1 and BU, AU, CU, FD, LP and FWPB have shown their variability in PC2. Conclusion We describe here, to the best of our knowledge, the systematic study of the mechanism involved in the regulation of enhancing ber quality and yield by stem bending strength, acupuncture and compression properties of Gossypium hirsutum crop. and reported varied values for yield components and plant height. The varieties of upland cotton were also evaluated by et al. and it was found that the plant height was associated positively with the seed cotton yield and bolls per plant. The positive correlation between plant yield seed was observed by et and and Haiping (2006), and their research showed that plant height contributed 70% of the total variability in seed cotton yield. Therefore, it is concluded that in cotton crop, height of plant is desirable if no lodging occurred.


Introduction
Cotton is one of the most important cash crops and the only major ber crop in the world. The contribution of cotton to the total ber used worldwide is about 35 percent (Zhang et al. 2014). Upland cotton (Gossypium hirsutum L.) is the largest cultivated species of cotton, occupying more than 90 percent of the world cotton cultivated area, which re ects widespread adaptability and high yield production characteristics (Wendel, 1989;Chen et al. 2007). Gossypium hirsutum L. is allotetraploid (2n = 4x = 52), and is composed of two ancestral genomes that are designated as At from Gossypium arboreum and Dt from Gossypium raimondii (Al-Ghazi et al. 2009). Due to long-term natural selection and arti cial breeding, a number of cotton germplasm resources for sustainable genetic improvement have been created under varied climatic and cultivating conditions. In the National Gene Bank for Cotton, China, 7,712 G. hirsutum accessions are present. All these accessions were collected from cotton producing countries around the world since 1865 when the United States introduced upland cotton ( Dai et al. 2016). In order to e ciently use these resources, various efforts have been made to investigate and evaluate cotton diversity (Fang et  Yield and quality of the produce are the most important factors for all crops (Fang et al. 2017a). Stem hardness is basic characteristic in the plant architecture of cotton, which is not well studied. Stem hardness may have a relationship with yield and ber quality. The physical characteristics contributing to the strength of stem are the bending force, puncture force and compression force. The stem bending force is the force at which the trunk bends or breaks under a particular load. The basal portion of the culm internode plays a crucial role in ensuring that the plant remains upright (Peng et al. 2014). The greater carbohydrate accumulation in the base stem could increase the force need to bend the stem (Ishimaru et al. 2008). Stem thickness is a biological indicator for green or dry biomass. The strength (force, stress) and energy requirements are the compressive properties. Therefore, the selection of genotypes with increased stem strength is a useful eld indicator (Beeck et al. 2006). Compression properties of stem depend on species, variety, stalk structure, stalk diameter, maturity, moisture and cell structure (Persson 1987). A physical quantitative measurement may enhance selection effectiveness and boost genetic gain, such as penetrometer sorghum measurement (Pedersen & toy 1999). Therefore, greater understanding of these parameters provide a theoretical basis to enhance the physical strength of the stem and basal part of the culm internode, with the aim of obtaining a higher yield and good ber quality of cotton.
Amorphous brils, lignin, and pectin, present in the cell wall, are also known to enhance the strength and hardness of the stem (Mohsenin 1986). Lignin or cellulose generally determines physical strength, as a low content of lignin or cellulose causes a brittle culm (Tanaka et al. 2003). In wheat plant the mechanical strength of the stem provided by cellulose and lignin to the lodging resistance plants (Cai et al. 2019). The selection of elevated stalk strength and resistance to the corn border increases the elements of cell walls in the breeding programme ) . Cotton has a high biomass output and a high cellulose and lignin proportion. In mature cotton bers the secondary cell wall (SCW) includes over 90% cellulose and it differs from all other known species of plant. By contrast, typical SCWs contain 40-50% cellulose in dicotyledonous stem xylem (Huang et al. 2016). The knowledge of stem hardness therefore allows the cell wall to be modi ed to improve ber quality and quantity, because plant cell wall has a close association with mechanical and biochemical strength of stem parameters.
Principal component analysis (PCA) has been used extensively in the plant sciences for variable reduction and genotype grouping. This is the most prevalent statistical multivariate method used in environmental studies (Tahri et al. 2005;Yongming et al. 2006). PCA commonly used in the analysis of the relationships between observed variables and in the extraction of a small number of autonomous factors (major component) (Tokalıoglu S et al. 2006). It commences with the correlation matrix, it describes the dispersion of the original variables and extracts eigenvalues and eigenvectors (Astel et al. 2008). Eigenvector is a list of coe cients that multiply the original correlated variants to obtain new uncorrelated (orthogonal) principal components that are linearly weighted combinations of the original variables. The number of correlated variables can be reduced to a smaller set of orthogonal factors, which allows the interpretation of a speci ed multidimensional system by showing correlations between the original variables. The analysis of the correlations also re ects a related response of a given character and also provides a good index for predicting the corresponding change in one character to the extent of the proportional change in the other. PCA was used by Kamara et al. (2003) to identify maize (Zea mays L.) traits which accounted for the majority of variance in the data. Granati et al. (2003) have used PCA to investigate the relationship among Lathyrus accessions. Žáková and Benková (2006)  Some studies have been conducted on stem strength behaviors of different plants; however, for stem hardness characteristics of cotton stalk, no data is reported. The present research therefore seeks to establish a relationship between stem hardness and yield characteristics and quality characteristics in Gossypium hirsutum. However, yield is a complicated, multicomponent controlled character. Stem hardness components are less sensitive than yield per se to the environmental changes and are therefore comparatively more likely to improve. Once the nature and extent of relations among these component characteristics and yield are understood, effectiveness of choice in the segregated generation will improve. Therefore, the present research was carried out to assess PCA and correlations of signi cant Gossypium hirsutum characteristics.

Cotton Accessions
From a set of 7,362 G. hirsutum accessions, preserved at the China National Gene Bank, Cotton Research Institute, Chinese Academy of Agriculture Sciences, Anyang, Henan, 237 cotton genotypes were selected. These accessions have various geographical origins including China, the United States, the former Soviet Union, Australia, Brazil, Pakistan, Mexico, Chad, Uganda and Sudan, which are the world's largest cotton-growing areas.

Planting and phenotyping
Phenotyping of stem hardness-related features was performed during the normal cotton growing season (mid-April to late-October) at Cotton Research Institute, Anyang, Henan, China (Yellow River command area) for two years i.e., 2017 and 2018. Coordinates of the location are E 114.07° and N 35.85 °, longitude and latitude respectively. All accessions (237) were planted in randomized complete block design with three replicates in the experimental eld. Each entry plot had a dimension of 7m × 3m and row-to-row and plant-to-plant distance was 30cm and 76cm respectively. Field management practices were conducted according to the local management scheme. The scoring standards for phenotypic traits in both years were identical. Six stem hardness traits and 14 agronomic traits were characterized.

Sample preparation for stem hardness traits
The stems were cut and separated from the branches after harvesting the cotton genotypes. Stem samples were air dried for two months in the lab. At the time of hardness testing, the air-dried cotton stem had low humidity content. The stem was equally divided into two parts for the preparation of test samples: Upper and lower ((Additional le 2: Figure S1a).

Stem Hardness Traits
For each replicate, three plant were selected to test the hardness of the stem. These characteristics were Breaking force of the Upper part (BU), breaking point of lower part (BL), Compression force of upper and lower part (CU and CL), and acupuncture force of the upper and lower part (AU and AL). The YYD-1 SS testing system (TOP Instrument Co., Zhejiang, China) was used to measure all hardness characteristics of 15 cm segment from lower, and upper part of stem (Additional le 2: Figure S2b). The tester was set perpendicular to the culm at the middle, under gradual loading, and the breaking force was measured when the culm was pushed to breaking point. The maximum force in Mega Newtons needed in order to break, puncture & compress the center of the two segments of the stem (upper and lower) was recorded.

Agronomic traits
Days to rst ower opening, FD (days) were calculated from the date of sowing to the day when rst owers bloomed on 50 percent of the plants in each plot. Plant height (PH) recorded from the base of plant above ground to the tip of the plant. Ten consecutive plants were selected for plant height in each plot. From each accession, 30 naturally opened bolls were harvested randomly to calculate boll weight (BW) in grams and to gin the ber. The Seed Index (SI) was calculated after counting and weighing 100 cotton seeds. Fiber samples were separately weighed to calculate the Lint percentage (LP) and Fiber weight per boll (FWPB) in grams. The Lint index (LI) was calculated based on SI and LP data.
Fiber samples were examined in the Cotton Quality Test Center in Anhui, China for ber-quality characteristics using a high-volume instrument (HFT9000). Data on the ber length (FL, mm), ber strength (cN/tex), micronaire value (Mic, µg/inch), elongation percentage (EP, %), Length uniformity (LU, %), Spinning Consistency Index (SCI) recorded. Average of the three replicates in the same year is de ned to be phenotypic information per accession.

Statistical analysis
For the evaluation of phenotypic traits statistics, Minitab 18 and R were used. The primary impacts of the experimental variables and their relationships were analyzed by the analysis of variances (ANOVA). The signi cance level for ANOVA was set at p ≤ 0.05. R software (package "corrplot") was used for calculating and plotting correlation. Principal component analysis was performed using Minitab 18.

Stem hardness variations among the genotypes
The ANOVA given in shows Tab. 1. that genotypic difference in stem hardness were highly signi cant for traits like bending (BL and BU), and compression CU (p > 0.05). Basic descriptive statistics (mean, standard deviation, minimum, maximum and coe cient of variance) of all the genotypes for morphological, yield and ber traits were studied (Additional le 1: Table S1) It was observed that maximum coe cient of   Table 2

Stem hardness correlation with ber quality traits
The result of the 2018 correlation of stem hardness indicated that bending lower (BL) has a positive association with ber length, micronaire value, uniformity percentage, bre elongation, spinning consistency index and days to owering (Fig. 1).

Discussion
In the last decade, there has been great progress in developing new cotton genotypes for better ber quality and higher yield. The stem associated characteristics such as bending, acupuncture and compression may be used to determine yield and quality of the ber. One reason for in uencing crop quality and yield is plant height (Tang JH 2007). The ber quality parameters on which textile processing and the quality of the item rely, are ber strength, length therefore premium pricing are charged for these quality features (Hussain K 2010).
Our breeding program goal for G. hirsutum was to identify high-yield genotypes, some agronomic features that are easily evaluated and linked with these characteristics could be used as markers (Biyun Chen et al. 2014). In this study, we observed that the bending, compression and acupuncture related to stem hardness have a positive and substantial correlation with the ber length, spinning consistency and owering times. May (2002) speculated that enhanced ber strength could demand more energy; therefore, higher strength genotypes produce fewer than lower strength lines. Pettigrew (2001Pettigrew ( & 2008 reported that an increase in light and temperature also increased the strength, the difference was however not enough to cause a yield penalty. Our ndings showed a positive association of length uniformity, micronaire values with stem hardness characteristics. Fiber neness was positively associated with ber length and ber strength by Killi et al. (2005). There was a negative association of ber neness with a ber strength and ber uniformity ratio. The ber strength showed positive correlation with ber uniformity. Mature cotton bers are approximately 95% cellulose with other polysaccharides such as arabinose, galactose and xylose (Meinert and Delmer 1977) and pectin (Meinert and Delmer, 1977;Wang et al. 2010). These are important for determining ber strength by joining cellulose brils. A direct correlation between cellulose molecular weight and ber strength was reported by Timpa and Ramey (1994). Though the metabolic cost of these polysaccharides is higher (Amthor 2010), a higher metabolic cost, unless transport of complex polysaccharides was an issue, seems unlikely to be a yield drain for such small fractions of the ber. These fundamental reasons for negative association yield and ber strength should be investigated further. Fiber diameter reduced from bottom to top, possibly because cell wall thickness decreased.
Similarly, this can be explained by the fact that development in cell walls depends upon the accumulation of metabolism products (cellulose, hemicellulose, lignin, waxes etc.) that rises with maturity (C. Ververis et al. 2003). The major requirement for increasing rice grain yield is to enhance the physical strength of the culm in order to enhance the breaking-type lodging resistance (Hirano et al. 2014). It has therefore been concluded that stem-related characteristics like bending stress have determined the morphology and the quality of the culm, such as cellulose, lignin, pectin inside the cell wall, which have a direct relationship to high yield and crop quality.
The Results also revealed that the boll weight following the bolls per plant had positive effect on seed cotton yield. Therefore, it is concluded that boll weight is an important yield component and should be kept in mind while breeding for seed cotton yield.

Conclusions
It may be concluded from the present study that cotton ber quality and yield can be improved by selecting types having high strength of stem.
Stem hardness related traits bending, acupuncture and compression show positive association with ber and yield related traits. Also enhancing the stem strength has proven to be an effective approach to decrease stem lodging risk. Because stem lodging is a persistent problem to decrease yield. Thus during future breeding programs these parameters also kept in mind during selection, as they were the major attributes of the cotton quality and yield. Recurrent selection could be followed to accumulate genes for the said traits in any population. In addition, the phenotypic data for stem hardness may be used in our subsequent genome-wide association studies for G. hirsutum.

Declarations
This work was supported by funding from the National Key Technology R&D Program, the Ministry of Science and Technology (2016YFD0100306 2016YFD0100203), the National Natural Science Foundation of China (grants 31671746).

Availability of data and materials
No other data related to this study is available at this time.
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.