Diversity analysis of South African sorghum genotypes using agronomic traits, SSR markers and protein content and amino acid composition.

Abstract

Sorghum is an important food security crop and ranks fifth after wheat, rice, maize, and barley in total area of production globally. In South Africa, sorghum is mainly grown for food and livestock feed both by small-holder and large-scale farmers. Diverse sorghum genotypes are grown in South Africa, which have not been fully characterized using agronomic, molecular or protein markers for breeding or strategic conservation. There is also little knowledge of farmers’ views and perception of the constraints affecting sorghum production, and their trait preferences, information that is needed to direct sorghum breeding programmes. The objectives of the study were to: (i) determine farmers production constraints and preferences of sorghum varieties in the Limpopo Province in South Africa; (ii) assess the level of genetic diversity present among South African sorghum genotypes using agro-morphological traits; (iii) compare random amplified polymorphic DNA (RAPD) and simple sequence repeat (SSR) markers and high resolution melt (HRM) analyses to determine genetic variation among selected sorghum genotypes; (iv) assess the genetic diversity present among South African sorghum genotypes using genetic distances as measured by SSR markers; and (v) determine genetic diversity of selected South African sorghum genotypes grown in two different agro-ecologies, especially for protein content and amino acid composition, and to select candidate lines for subsequent breeding and conservation. A participatory rural appraisal (PRA) study was conducted in two selected districts of Limpopo province to determine sorghum production constraints and variety preference involving 311 respondent farmers. The PRA data was collected using semi-structured questionnaires, focus group discussion, matrix ranking and transect walks. The main constraints affecting sorghum production were bird damage (53.08 %), storage pests (weevils) (50.05%), parasitic weeds (35.00%), drought (35.25%) and postharvest diseases (30.75%). Good taste, high yields, resistance to bird damage, insect pests (weevils), and diseases, early maturity and drought tolerance were farmers-preferred traits of sorghum varieties in the study areas. Ninety eight diverse South African sorghum genotypes were characterized using agro-morphological traits. Principal component analysis revealed that the three most important components contributed 38.9%, 30.96% and 18.13% to the total variation. The traits that

contributed most to the variation were plant height, seed weight and panicle weight. A dendrogram was constructed using the Unweighted Pair Group Method with Arithmetic Mean, and this grouped the genotypes into three major clusters. The grouping of the genotypes was independent of the source or place of origin. The genotypes MP 4277, EC 2934, KZ 5097, FS 4909, and LP 4303 were phenotypically identified as the most diverse. The best lines with quantitative and qualitative attributes were MP 4276, NW 5430, 05-Potch-167 and EC 3217 across the locations. Eight selected sorghum genotypes were used to compare high resolution melt (HRM) analysis with random amplified polymorphic DNA (RAPD) and simple sequence repeat (SSR) analyses. DNA was extracted using the CTAB extraction method. The template DNA was amplified, using three RAPD and SSR primers for each sample. Both markers revealed variation among the sorghum genotypes, with a moderate correlation between the RAPD and SSR results. The genotypes were further subjected to high resolution melt (HRM) analysis, which showed considerable variation between the genotypes. There was a high level of correspondence between the clustering of genotypes when using SSR markers or HRM analysis. One hundred and three sorghum genotypes collected from various South African provinces by the Department of Plant Genetic Resources, the African Centre for Crop Improvement (ACCI) and the Agricultural Research Council-Grain Crops Institute (ARC-GCI) were genotyped using 30 polymorphic SSR markers. The SSR analysis revealed extensive variation among the sorghum genotypes. The genotypes Macia-SA and AS4 had the lowest dissimilarity index, whereas POTCH-115 and MP 2048 showed the highest dissimilarity index. The size and number of alleles ranged from 90 to 294, and 2 to 15, respectively. The polymorphic information content (PIC) varied between 0.02 and 0.84. The heterozygosity data points ranged between 0.02 and 0.85, with the genetic distances ranging between 0 and 8.4. Fifty nine selected sorghum genotypes were grown at two locations, Makhathini and Ukulinga Research Farm, Pietermaritzburg. These were analysed for crude protein content using near-infrared spectroscopy (NIR). The genotypes that had a high protein content at both locations were AS4 (15.07%), Maseka-a-swere (15.13%), AS19 (15.22%), Macia-SA (15.31%), AS16 M1 (15.57%) and Mammopane (16.18%). Nineteen sorghum genotypes with high crude protein content were selected and analysed for their amino acid composition. The genotypes with high lysine content were KZ 5246 (2.27%), AS17 (2.25%), Manthate (2.16%) and LP 1481 (2.11%). Lines identified with high leucine levels were LP 1948 (14.3%), FS 4905 (14.3%), MP 4154 (14.26%) and LP 1481 (14.25%). The genotype AS16cyc was the best candidate for high phenylalanine content (5.99%). Overall, the study found considerable levels of genetic variability among South African sorghum germplasm using agro-morphological, SSR markers and protein content and amino acid levels. The selected lines should be useful for future breeding programmes. Knowledge of the genetic diversity can be used to direct efforts to conserve the diversity of sorghum germplasm present in South Africa.