Non Technical Summary
Cotton is a cornerstone of American agriculture. While most people associate cotton with textiles, the plant also produces seeds that are often overlooked. Cottonseeds are a valuable resource, particularly for the oil they contain. Cottonseed oil is naturally rich in heart-healthy fats and vitamin E, has a neutral flavor, and a high smoke point, making it ideal for cooking and food processing. Despite these advantages, cottonseed oil remains underutilized compared to more familiar oils like soybean, canola, and olive oil.The underuse of cottonseed oil represents a missed opportunity not just for consumers, but for farmers and the broader agricultural economy. In 2023, the United States produced approximately 3.8 million tons of cottonseed, valued at over $4.6 billion. As global demand for cottonseed oil is projected to grow steadily--reaching an estimated $6 billion within the next decade--there is a clear and timely opportunity to expand its market. Doing so would allow cotton producers to gain more value from each harvest without requiring additional land, water, or fertilizer.Promising cotton lines that produce seeds with improved oil content and composition have been identified through the plant breeding program of the Department of Plant and Soil Science in Texas Tech University. These varieties can be used in breeding programs to develop improved cottonseed chemistry that meets the growing consumer demand for healthier, more sustainable food options, while also supporting the needs of food manufacturers who require oils with specific processing qualities.However, before these improved cotton varieties can be widely adopted in breeding purposes, it is essential that we understand how consistent their beneficial traits are across different growing conditions. This project will evaluate these cotton lines in a range of environments to assess how stable and reliable their oil characteristics remain under varying conditions. The primary objective is to determine the extent to which differences in oil quality and quantity are driven by the plant's genetic makeup, how much is influenced by environmental factors such as moisture availability, and how these two elements interact to shape the chemical content and composition of cottonseed oil. Gaining this understanding is critical, as it will guide the development of more effective breeding strategies aimed at enhancing the desired traits in cottonseed oil. With this knowledge, breeders can make informed decisions that could lead to steady genetic improvements and consistently better oil quality--without negatively affecting the overall health, resilience, or productivity of the cotton crop.

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
201	1710	1081	100%

Knowledge Area
201 - Plant Genome, Genetics, and Genetic Mechanisms;

Subject Of Investigation
1710 - Upland cotton;

Field Of Science
1081 - Breeding;

Goals / Objectives
The primarygoal of the proposed research is to improve the genetic potential in breeding for cottonseed oil content and composition towards meeting the demands of end-users for a more nutritious and healthy edible oils with a wide range of food processing applications. To this end, the relative contributions of genotype, environment and genotype x environment effects on the observed variations in cottonseed oil content of select landraces and oil composition of RILs with variable fatty acid will be assessed in multi-environment trials. Partitioning of the relative contributions of genotype and environment factorials on phenotypic outcomes will allow the estimation of the true genotypic value of the germplasm which can be exploited towards the development and release of improved cultivars. Thegoal of the project will be achieved through the following specific objectives:1. Conduct multi-environment trials for six recombinant inbred lines with variable fatty acid profiles, two conventional cultivars used to develop the RILs and nine landraces representing a wide range of oil content. The cotton genotypes will be grown under two irrigation regimes (i.e. normal and deficit irrigation (i.e. 80% irrigation) in two locations (Texas and North Carolina) for two successive years (2025 and 2026).2. Measure oil content and composition of cottonseeds generated from the experimental materials grown in multi-environments (i.e. irrigation treatment x location x year) using the Proton Nuclear Magnetic Resonance (1H NMR) method.3.Evaluate fiber yield and quality of the experimental materials grown in multi-environments (i.e. irrigation treatment x location x year) using High Volume Instrument (HVI; Uster Technologies 1000 systems) and Advanced Fiber Information System (AFIS).4.Assess the relative contribution of genotype, environment and genotype x environment interactions on the phenotypic performances of the experimental materials using a set of statistical tools for standard and multivariate analysis.

Project Methods
Cotton genotypes composed of landraces with variable seed oil content (21-30%) and recombinant inbred lines (RILs) with variable fatty acid profiles including the two cultivars used to develop the RILs will be used in the study. Seeds of the RILs have 17.30-19.60% palmitic acid, 1.90-3.00% stearic acid, 17.70-24.00% oleic acid and 58.20-61.80% linoleic acid, as well as oil content of 13-15%.Prior to field trials, cottonseed oil content and composition of the fifteen genotypes will be determined using Proton Nuclear Magnetic Resonance (1H NMR). Previous analysis has used NMR to establish oil content and gas chromatography to quantify the fatty acid composition of the RILs, whereas the landraces have not been evaluated for fatty acid composition. Cottonseed analysis using 1H NMR will establish a comparative profile of the seed oil chemistry of all experimental materials based on just one protocol for oil content and composition analysis.The experimental materials will be grown in Quaker Farm in Lubbock, Texas and Field Station in Rocky Mount, North Carolina during the cotton growing seasons of 2026 and 2027. Each genotype will be grown in triplicate rows arranged in split-plots within a randomized complete block design. Split plots will be used to accommodate irrigation treatments (i.e. normal and 80% water deficit) that are difficult to randomize under field conditions. The inclusion of deficit irrigation as a treatment within field locations will allow analysis of the stability of phenotypic outcomes for the RILs which have demonstrated drought tolerance under field conditions and the landraces which, by nature, are supposed to maintain productivity under sub-optimum environments.Fifteen plants per genotype per irrigation treatment will be hand-planted at a 25 cm spacing. Three seeds per hill will be sown to ensure enough plants for phenotypic evaluation. At the 4-true leaf stage, the plots will be thinned to one plant per hill. The remaining plant will be grown to full maturity following standard field management practices in each location. At the reproductive stage, all flowers at the candle stage will be bagged to ensure self-pollination. At plant maturity, all mature bolls will be hand-harvested from each replicated treatment to recover fiber and fuzzy cottonseeds for fiber and cottonseed oil analysis, respectively. Samples from North Carolina will be sent to Lubbock, Texas for ginning using a tabletop 10-saw gin and downstream analysis for fiber and cottonseed oil chemistry.Rainfall and temperature data during the 2-year trial in Lubbock, Texas and Rocky Mount, North Carolina will be obtained from the National Weather Service of the National Oceanic and Atmospheric Administration. Soil chemical and physical properties in the experimental fields including electrical conductivity, pH, water capacity, bulk density, texture and organic matter content will be sourced from the USDA National Resources Conservation Service.To ensure that variations in the fatty acid composition of cottonseed oil, particularly that of the RILs have no negative effects on fiber properties, correlation between seed oil chemistry and fiber yield and quality will be determined. Fiber yield of the cotton genotypes grown during the 2-year field trials will be evaluated based on boll count as outlined by the Mississippi State Extension Office. Briefly, the average number of harvestable bolls per row foot will be calculated from replicated 10-feet rows of each genotype per treatment. The average boll weight for the field will then be calculated from the weight of 50 hand-picked bolls representing all boll sizes on the plant. Using standard calculations for expected turnouts of 33-35%, the number of bolls required to produce a 480-pound bale of cotton will be determined using data on the average number of harvestable bolls and boll weight. Yield in bales per acre can be calculated by dividing the number of bolls per row foot counted by the number of required bolls to produce a bale of cotton.Following ginning, the fibers will be tested for quality parameters using High Volume Instrument (HVI; Uster Technologies 1000 systems) and Advanced Fiber Information System (AFIS). HVI testing provides replicated measures of fiber micronaire, length, strength and color, whereas AFIS provides replicated measures of length, maturity ratio, fineness, neps, and trash. Both HVI and AFIS testing of fibers generated from the 2-year trials will be outsourced to the Fiber and Biopolymer Research Institute in Lubbock, Texas.The content and proportion of the four major fatty acids (i.e., palmitic, stearic, oleic and linoleic acids) in cottonseed oil will be determined using 1H NMR. Pure oil will be extracted from the samples by total lipid fraction. Briefly, 5 g of each sample will be homogenized with 100 mL of chloroform-methanol (2:1 v/v). The obtained mixture will be centrifuged (10 min, 1540 g), and filtered. Distilled water (25 mL) will be added to the filtrate, and the resulting mixture will be shaken and again centrifuged (10 min, 1540 g). The organic phase will be separated and dried with anhydrous sodium sulfate. The solvent will be removed in a rotary evaporator and then under a gentle stream of nitrogen to prevent lipid oxidation. Next, 500 MHz 1H-NMR spectra will be obtained under ambient conditions on Varian Unity INOVA 500, using deuterated chloroform both as a solvent and an internal reference standard. The data will be collected using an autosampler, and processed and analyzed using the MestReNova software (MestReNova, version 12.0.0-20080).The multi-environmental trial analysis or 'metan' package integrated in the R Statistical software v4.1.2 will be used to process and analyze data from the multi-location trials. The 'metan' package integrates the necessary tools to perform all the steps in analyzing multi-environment trial data including data checking and filtering, manipulation, analysis and visualization. These include statistical tools for analysis of variance (ANOVA), principal component analysis (PCA), correlation analysis, genotype plus genotype x environment (GGE) biplot analysis, additive main effects and multiplicative interaction (AMMI). ANOVA based on random effects model will be used to determine variability in the cottonseed oil content and composition, as well as fiber yield and quality of the fifteen genotypes due genotype, environment, replication within environment, genotype x environment effects and residual error. PCA will be applied to examine the significance of genotype x environment interaction based on the proportional or disproportional responses of genotypes across environments. Based on PCA, GGE biplots will be generated to allow visual examination of the relationship between genotype and genotype x environment effects on target traits and identify stable genotypes in the tested environments. In conjunction with GGE, AMMI will also be applied to assess the magnitude of genotype, environment and genotype x environment interaction based on ANOVA and PCA results. Correlation analysis to establish the relationship between cottonseed oil and fiber properties will also be run in 'metan'. The heritability of the variable fatty acid profiles in the RILs will also be estimated based on calculated phenotypic, genotypic and environmental variances.