Selection in cowpea genotypes for nutritional traits

HIGHLIGHTS: Cowpea genotypes show variability in nutritional traits. Cowpea genotypes with high nutrient concentration are indicated for genetic improvement. Cowpea genotypes have favorable chemical properties and nutritional value for cultivar development. ABSTRACT Cowpea is a worldwide consumed legume due to its high nutrient concentrations. Selecting genotypes with high nutrient concentrations can contribute to developing biofortified cultivars. This study aimed to evaluate the nutritional potential and indicate genotypes of Vigna unguiculata for genetic improvement based on nutritional traits. Forty-three cowpea genotypes belonging to the Active Germplasm Bank of the Federal University of Ceará were used in the study. The analyses of dry matter, ash, ether extract, proteins, and minerals (phosphorus, potassium, calcium, magnesium, sulfur, sodium, copper, iron, zinc, manganese, and boron) were performed in triplicate. The data were subjected to analysis of variance, means comparison test, genetic parameters, nutritional quality indexes, and the sum of ranks. Genetic variances predominated concerning environmental variation for the nutrient concentrations of cowpea can be transmitted to future generations. The CE-0151, CE-0189, CE-0207, CE-0228, CE-0248, CE-0542, CE-0685, CE-0686, CE-0796, CE-0997, and CE-1002 genotypes are indicated for selection for continuing the biofortified cowpea breeding program.


Introduction
Cowpea (Vigna unguiculata L. Walp) is a legume species native to Africa and belonging to the family Fabaceae; showing high concentrations of proteins, fats, carbohydrates, minerals, vitamins (Rengadu et al., 2020), and starch, which can be used to produce gels for industrial use (Oyeyinka et al., 2020).This species is one of the most important vegetable protein sources in developing countries and is mainly grown in Western and Central Africa, Latin America, and Southeast Asia (Nardi & Ozcan, 2022).
The varieties of cowpea differ in terms of their morphology and proximate quality, which are fundamental aspects of breeding programs for the species (Gerrano et al., 2022).These variations can be explored to benefit human health by fighting hidden hunger (Silva et al., 2021).
Hidden hunger (lack of nutrients) is a reality in several developing countries, affecting around two billion people worldwide (Loureiro et al., 2018).Cowpea breeding programs are necessary to improve the nutritional grain quality in cowpea cultivars and assist in hunger reduction (Lovato et al., 2018).
Legumes have been consumed in various ways for thousands of years worldwide.These plants contain various nutritional components, such as proteins, minerals, and vitamins essential to human health (Kumar & Pandey, 2020).Therefore, breeding programs aimed at increasing the concentrations of these components can reduce the risk of several diseases.
Biofortification can be performed by conventional breeding methods (Buratto & Moda-Cirino, 2017).Thus, the selection of cowpea genotypes is essential to identify germplasms with high nutritional potential, which can be used in breeding programs to develop biofortified cultivars.In this scenario, genetic parameter estimates are reliable indicators for improving genetic traits through selection (Kumar et al., 2015).Furthermore, selection indexes also assist in efficiently predicting selection gains by establishing a linear combination of several traits, thus enabling efficient simultaneous selection (Cruz, 2016).Therefore, this study aimed to evaluate the nutritional potential and indicate genotypes of Vigna unguiculata for genetic improvement based on nutritional traits.

Material and Methods
The experiment was conducted in the Horticulture Sector of the Department of the Agricultural Sciences Center of the Federal University of Ceará (CCA/UFC), Pici Campus (3º44'24.4"S, 38º34'32.0"W, and altitude of 19.5 meters) in Fortaleza, Ceará, Brazil.The experiment was conducted under rainfed conditions.During the experimental period, from February to May 2020, the cumulative rainfall was 1,111.8mm, whereas the mean temperature was 27.3 ºC.
The experimental design was completely randomized, with 43 treatments (39 genotypes and four commercial cultivars) and three replicates.The treatments consisted of 43 cowpea genotypes (V.unguiculata) and four commercial cultivars (BRS Juruá, BRS Tumucumaque, BRS Guariba, and BRS Aracê).BRS Juruá and BRS Tumucumaque are biofortified commercial varieties belonging to the Active Germplasm Bank (BAG) of the Plant Science Department of CCA/UFC, according to Table 1.
Table 1.Common name, origin, class, and subclass of cowpea genotypes The total experimental area had 214 m², containing ten plants per plot.The soil of the area was classified as Ultisols (United States, 2014), which corresponds to the Argissolo Vermelho Distrófico in the Brazilian soil classification system (EMBRAPA, 2018).The area was conventionally prepared by plowing and harrowing.Fertilization was performed based on soil analysis (Table 2) and considering the crop requirements at sowing with the use of phosphorus (super simple phosphate) and potassium (potassium chloride) and at top dressing (15 days after the emergence) with nitrogen (Urea).The spacing was 1.0 m between rows and 0.50 m between plants in the row.Three seeds were sown per hole, and the plants were thinned to two plants per hole 15 days after sowing.
The crop management practices for weed control consisted of manual hoeing in the seedling emergence period and close to flowering.Insecticides were applied to control pests during plant development (20 and 40 days after emergence, Decis ® 25 EC).The pods were collected and threshed during harvest, depending on the cycle of each genotype (Pessoa et al., 2022), and the seeds were selected for nutrient analysis.
The nutrient analyses were performed by washing the raw, dry grains with distilled water and placing them in a forced-air oven at 60 °C for 48 hours.After this procedure, the grains were ground in an electric coffee grinder to obtain a powder used in the analyses.The powder was stored in hermetically sealed polyethylene bags and kept under refrigeration (4 ºC) while it was used in the analyses.The chemical analyses were performed in triplicate.
The analyses of dry matter, ash, ether extract, and protein were performed at the Laboratory of Food Analysis and Animal Nutrition of the Center of Agricultural Sciences of the Federal University of Paraíba, following the recommendations of the Association of Official Analytical Chemists (AOAC, 2022).
Mineral characterization (phosphorus, potassium, calcium, magnesium, sulfur, sodium, copper, iron, zinc, manganese, and boron) was performed at the Laboratory of Soils of Embrapa Agroindústria Tropical, according to the methodologies described by Miyazawa et al. (2009).
The NQI (Nutritional Quality Index) was calculated according to Carvalho et al. (2012) and Pereira (2013), with slight modifications.This index was calculated by considering the arbitrarily determined weights of four for protein, three for the minerals Fe (iron) and Zn (zinc), and two for Ca (calcium) and Mg (magnesium).The arbitrary weight attributed to each component was multiplied by the difference calculated between each value of the respective component and their overall mean, followed by the algebraic sum of each term.The result of this sum was divided by the sum of weights according to the following Eq.1: relationship between genetic and environmental variation coefficients were also calculated.

Results and Discussion
Genetic variability is essential for selecting individuals for pre-improvement and generation advancement (Carvalho et al., 2021).From this perspective, the present study revealed significant differences between genotypes for most traits evaluated, except dry matter (Table 3), indicating variability between genotypes for the grain nutrient concentrations, enabling the increase in micro and macronutrient levels through conventional breeding.Freitas et al. ( 2022) also reported cowpea variability for iron, zinc, and protein.
The heritability values were high, above 70% for most traits (Table 3).The highest mean values were observed for ash (96.22%), ether extract (93.68%), proteins (98.54%), phosphorus (85.25%), calcium (88.40%), sodium (91.50%), zinc (92.01%), manganese (94.13%), and boron (94.85%).These data will serve to support the choice of genotypes for selection.The heritability coefficient defines the proportion of the selection differential that will be transmitted to the following generation, enabling the selection of more promising genotypes for desired traits (Leite et al., 2015), i.e., the higher this coefficient, the higher the success of selection for a given trait (Marialva et al., 2019).
Knowledge about genetic parameters, such as heritability, is required in breeding programs aimed at obtaining new bean cultivars with high grain nutrient concentrations (Buratto & Moda-Cirino, 2017), as reported in the present study.
The sum of ranks index of Mulamba & Mock (1978) was determined by considering all nutrients according to the following expression: where: Ii -index of the ith genotype.Ʃ Ka -sum of the weight assigned to each variable.r ik -rank of progeny i for trait k.All analyses were performed using the software Genes (Cruz, 2016).3. Summary of the analysis of variance: mean squares, heritability (h 2 %), and the relationship between the genetic and environmental coefficients (CVg/CVe) for cowpea nutrient descriptors (1) (2) The relationship between the coefficients of genetic and environmental variation (CVg/CVe) was higher than one for most traits (Table 2), e.g., ash (2.61), ether extract (2.22), proteins (5.61), phosphorus (1.38), calcium (1.59), magnesium (1.04), sodium (1.89), zinc (1.95), manganese (2.31), and boron (2.48), corroborating the highest heritability values and favoring selection since the highest contribution to the next generation is of genetic origin.The relationship between the coefficients of genetic and environmental variation is used to quantify the genetic variability available in the population when determining its potential for breeding purposes (Araújo et al., 2014).
The Scott-Knott test allowed clustering genotypes into two to 11 distinct groups, varying according to the trait analyzed and showing that the genotypes had varying nutrient concentrations (Table 4).Similar to this study, Melo et al. (2017) reported that the chemical and nutrient compositions of cowpea vary among cultivars, assisting the selection and breeding of accessions with high nutrient concentrations.
The trait with the highest data variation corresponded to proteins, forming 11 different groups with mean values ranging from 17.14 (CE-0248) to 33.88 (CE-0686), showing the highest variability among the studied genotypes.In that regard, genotypes CE-0686 (33.88), CE-0165 (31.30), and CE-1002 (30.78) are recommended for selection for showing the highest protein concentrations (Table 4).Developing cultivars with high protein concentrations is one of the main objectives when breeding vegetable species such as cowpea (Frota et al., 2017).
Seven distinct groups were formed based on the nutrient manganese, with the CE genotypes 0206, 0689, 0313, 0199, 0244, 0686, 0243, 0253, 0997, 0207, 0248, 0685, 0964, 0164, 0070, 0114, and 978 showing the highest mean values (Table 4).CE-0689 also showed the highest ash, phosphorus, and potassium concentrations among these genotypes.The presence of these nutrients in cowpea makes this crop potentially important in the human diet from a nutritional perspective (López-Morales et al., 2020) since genotypes with high concentrations of macro Continued on the next page Table 4. Mean nutrient values in cowpea genotypes and micronutrients have the potential to be used against malnutrition, a problem that affects many people.
Five groups were formed only by boron, with genotypes CE-0151, CE-0958, CE-0228, and CE-0978 showing the highest performances.Several cowpea biofortification studies aim to increase zinc and iron concentrations (Melo et al., 2017;Kumar & Dhaliwal, 2021).However, selecting genotypes with high concentrations of various elements (Fe, Zn, Mn, for example) is essential for breeding programs since it will be possible to develop a cultivar rich in several nutrients, including boron.
The ether extract (CE-0964 and CE-0164), calcium (CE-0114), sodium (CE-0068, CE-0207, and CE-0398), and zinc (CE-0151, CE-0957, CE-0165, CE-0997, CE-0958, CE-0337, CE-0796, and CE-0925) showed for different groups, with some genotypes showing the highest mean values (Table 4).Haider et al. (2021) reported variability among genotypes of Vigna radiata for the grain Zn concentrations, ranging from 15 to 45 mg kg -1 .However, the genotypes of the present study showed higher values, highlighting the potential to initiate a breeding program aimed at biofortification.In addition to zinc, the ether extract, calcium, and sodium also showed variability.Since the environmental conditions were the same for all plants in the present study, the variability observed among the genotypes is inherent to genetic aspects, as observed in Table 3.
The lowest data variation was observed in the concentrations of potassium, magnesium, copper, and iron (Table 3).Genotype CE-0796 showed a higher iron concentration than AH -Ash; EE -Ether extract; PB -Proteins; P -Phosphorus; Ca -Calcium; Mg -Magnesium.S -Sulfur; Na -Sodium; Cu -Copper; K -Potassium; Fe -Iron; Zn -Zinc; Mn -Manganese; B -Boron.Means followed by the same letter in the column belong to the same group by the Scott-Knott clustering algorithm at p ≤ 0.05 Continued on the next page The data were subjected to analysis of variance.The means were grouped by the Scott-Knott clustering algorithm at p ≤ 0.05.The estimates of heritability, genetic variance, and the *, ** -Significant at p ≤ 0.05 and 0.01 by the F-test, respectively; ns -Non-significant; DM -Dry matter; AH -Ash, EE -Ether extract; PB -Proteins; P -Phosphorus, Ca -Calcium; Mg -Magnesium; S -Sulfur, Na -Sodium; Cu -Copper; K -Potassium; Fe -Iron; Zn -Zinc; Mn -Manganese; and, B -Boron Table

Table 2 .
Chemical attributes of the soil of the experimental area OM -Organic matter; BS -Base saturation; CEC -Cation exchange capacity