**3. Determination of grain quality profile and phytochemical content of elite lines**

#### **3.1 Introduction**

Adaptation to drought is a result of biochemical adaptation to water stress leading to a change in chlorophyll content, production of antioxidant scavenging enzymes, increase in proline content, production of secondary metabolites such as; alkaloids, terpenes, flavonoids, mevalonic acid, shikimic acid among others [4, 5]. Drought like other environmental changes can bring about marked differences in the defense chemistry of the plant. These qualitative traits (secondary metabolites) also control the aroma, taste and acceptability of products and can be integrated into the breeding pipeline at the priority setting and trait discovery stage. Variations in phytochemicals could be used to broaden the genetic base in the three current gene pools; food, feed, fodder through introgression to generate desirable genetic complexes (linkage groups). From breeder point of view, these phytochemicals can be grouped as valuable or useful and sometimes negative hampering the application of germplasm in a breeding program.

#### **3.2 Materials and methods**

The study evaluated the hypothesis that alkaloid content reduce in advanced generations as result of selection. They are however associated with serious side effects on products at high levels. This was investigated among the 20 breeding materials in study 2. The genotypes included; (7 cultivars, 4 progeny lines, 8 Varieties and 1 Landrace populations) making minicore-germplasm. There are from different gene pools and breeding history. Biochemical analysis was carried out at the National Crops Resources Research Institute (NACRRI) Bio-Nutrition Laboratory in 2019. Principles and methods of biometrical designs were applied according to the protocol developed by [16]. Absorbance was read at wavelength 470 nm. The variation with respect to chemical composition in genotypes was attributed to genetic differences. Canonical correlation analysis was used between sets of independent variables for data interpretation and cluster analysis for grouping genotypes.

#### **3.3 Variability for biochemical contents in grain sorghum**

The multivariate analysis utilized all the variations of traits in generic way to group genotypes with similar sets of traits and quantify the importance of various traits in

grouping/clustering genotypes**.** Results revealed statistically significant (P < 0.001) differences among the sorghum accessions profiled for biochemical contents, demonstrating the influence of genotype with respect to checks (**Table 6**)**.** The subset of sorghum minicore germplasm has been categorized into four clusters. This could be the influence of physiological and biochemical processes in modifying the plant in response to abiotic and biotic stresses [4, 17]. Cluster 1 in red (**Figure 5**), with the largest number of genotypes displayed more scope for selection against water stress as supported by positive and significant correlations (**Figure 6**). The coefficient of genotypic variance was above 80% indicative of substantial genetic diversity and prospects for improvement through chemical selection (**Table 6**). Low levels of phytochemicals of less than 4 mg/100 g were consistent as all released varieties were clustered together with checks supporting the hypothesis (**Figure 5**). Polyphenols were positively correlated among themselves but negatively associated to the levels of carbohydrates. The level of tannins was important in establishing groups and contributed a lot to the total variability among the accessions. Landraces and their derived lines and hybrids clustered together, hence exploitation of this material require a lot of chemical selection due to probably strong linkages with the wild (i.e., NPT 10, NPT 5 and NPT 9). Improvement of genetic gain for these plant chemical defense compounds might be possible through hybridization.


#### **Table 6.**

*Variability for biochemical contents in grain sorghum in Serere in 2017 based on observed values of absorption spectrophotometer.*

*DOI: http://dx.doi.org/10.5772/intechopen.112322 Accelerating Breeding for Drought Tolerance in Sorghum (*Sorghum bicolor*): An Integrated…*

**Figure 6.**

*Correlation among variables in grain sorghum in Serere in 2017 with their corresponding coefficient values and probabilities potential.*

#### **3.4 Correlated response to selection and indirect selection**

A high level of carbohydrates was negatively correlated to the levels of polyphenols. Therefore, selection for large seed size of ≥3 gms/100 seed could reduce the concentration of polyphenols relative to the increase in water and carbohydrate (starch) content in the seed. In nature, the association among desirable traits can be negative as for the case of maize, (e.g., increasing grain yield is associated with lower protein content [18]. The variation for biochemical traits was represented

by the two-dimensional scatter diagram that accounted for 70% of the variance. Genotypes; NPT15 (IESV92034DLSEL2) and NPT 19 (IESV23007DL), were plotted in the upper right quadrat. Meanwhile genotypes; NPT 5 (IESV24029SH), NPT 9 (IESV24029SH × ICSB 479) and NPT10 (KAK-7780) are intermediate occupying the lower right quadrant (**Figure 7**). The differences could reflect breeding, selection history and complex interrelationships between ecological factors important for parental selections for multiple traits [14, 15]. Diversity among genotypes has been categorized into groups of similar characteristics that can be used for designing optimized crossing strategies. Released varieties clustered together support the argument that selection and hybridization among themselves have taken place (**Figure 5**). The key traits that are most likely to improve the rate of genetic gains for grain quality traits are levels of tannins and carbohydrates since they are negatively correlated.

#### **3.5 Sectional conclusion and recommendations**

Chemical factors such as high carbohydrate content, less intensive color and tannin content are good attributes for better quality products**.** This study showed the value of exploiting the information in correlated traits that will contribute toward improving the accuracy of breeding values of the products such as bread and malt quality. Accurate selection of sorghum breeding lines can accelerate annual genetic gain for these correlated traits when used to generate an optimized crossing design

**Figure 7.** *Scatter plot showing the best selections on the upper right hand.*

#### *DOI: http://dx.doi.org/10.5772/intechopen.112322 Accelerating Breeding for Drought Tolerance in Sorghum (*Sorghum bicolor*): An Integrated…*

(study 4); where there is a high and positive correlation between secondary traits such as color and target traits such as tannins and phenolic, then greater selection intensities can be applied to the secondary trait during screening in big populations. Positive results will be expected when the information is integrated with the pedigrees during breeding. Furthermore innovative research is important into processes that mitigate ant nutritional factors while enhancing bio-availability of proteins, amylose starch among the high tannin genotypes in the utilization of such materials in feed, food and beer value chains.
