Assessment of cup quality, morphological, biochemical and molecular diversity of Coffea arabica L. genotypes of Ethiopia
Bekele, Yigzaw Dessalegn
MetadataShow full item record
The genetic diversity of C. arabica genotypes collected from the northwestern and southwestern parts of Ethiopia was evaluated using morphological, biochemical and molecular (AFLP and SSR) markers. The objectives of the study were to (1) estimate the level of genetic variation among currently grown C. arabica genotypes, (2) estimate the level of associations among agro-morphological characters, coffee quality traits and green bean biochemical compounds, (3) compare the level of genetic variation between coffee genotypes collected from the northwestern and southwestern parts of Ethiopia and (4) compare the efficiency of AFLP and SSR markers in detecting genetic variation in C. arabica. Results of diversity analysis using nine qualitative and 18 quantitative morphological characters indicated the presence of substantial variability among evaluated coffee genotypes. Genotypes were significantly different for all 18 quantitative morphological characters. Higher Shannon-Weaver diversity indices ranging from 0.401 to 0.989 were recorded for nine qualitative morphological characters. Cluster analysis using qualitative, quantitative and a combination of qualitative and quantitative characters classified genotypes into two groups. Qualitative characters failed to independently distinguish seven of the genotypes. Principal component analysis grouped 18 quantitative characters into 15 principal components and the first four explained 82.3% of the entire variability among genotypes. Average green bean yield per tree and bean length were important for the variation explained in three of the first four principal components. Correlation analysis performed among 18 agro-morphological characters indicated positive associations between average green bean yield per tree and all other characters. Average green bean yield per tree had statistically significant correlations with percentage of bearing primary branches per tree, bean weight, canopy and trunk diameters, tree height, bean length and thickness, internode lengths of orthotropic and primary branches and fruit and petiole lengths. Broad sense heritability varied from 38% for bean thickness to 94% for bean weight and number of secondary branches per tree. Average green bean yield per tree, canopy diameter, tree height and bean weight had higher genetic coefficients of variation and broad sense heritabilities. These characters were positively and significantly correlated, allowing simultaneous improvement of these traits. Genotypes were evaluated for variability in four cup quality traits (acidity, body, flavour and overall standard) and four green bean physical characters (bean shape, size, uniformity and weight). Considerable variation was observed among genotypes both for cup quality as well as green bean physical characters. Of all the genotypes, AD0691 and AD1691 had both desirable cup quality and green bean physical characters. Cluster analysis performed using cup quality and green bean physical characters, classified genotypes into two groups. Genotypes were not clustered according to collection regions. All cup quality traits were positively and significantly correlated and correlation coefficients ranged from 0.69 to 0.93. Positive correlations were observed among desirable cup quality and green bean physical characters indicating that cup quality and green bean physical characters can be improved simultaneously. Genotypes were evaluated for green bean caffeine, chlorogenic acids, sucrose and trigonelline content variability. Analysis of variance indicated significant differences among genotypes for all biochemical compounds. Green bean caffeine, chlorogenic acids, sucrose and trigonelline contents ranged between 0.91-1.32%, 2.34-4.67%, 5.30-8.98% and 1.04-1.71%, respectively on dry matter basis. Cluster analysis classified coffee genotypes into eight groups and genotypes were not clustered according to collection regions. Coffee genotypes collected from northwestern Ethiopia were more diverse in green bean caffeine, chlorogenic acids and sucrose contents compared to southwestern Ethiopia genotypes. On average, southwestern Ethiopia coffee genotypes had higher green bean caffeine, chlorogenic acids, sucrose and trigonelline contents compared to northwestern Ethiopia genotypes. Green bean caffeine content showed negative and statistically significant associations with all desirable cup quality attributes and correlation coefficients ranged from -0.305 to -0.407. Desirable cup quality traits and low green bean caffeine content can be selected simultaneously. Green bean sucrose content showed positive associations with all desirable cup quality and green bean physical characters. Higher green bean sucrose content, desirable cup quality as well as green bean physical characters can be improved together. The genetic diversity of coffee genotypes was investigated using 10 AFLP primer combinations. Results revealed differences among AFLP primer combinations in detecting genetic variation among genotypes. Of the 10 primer combinations EcoRI-ACA/MseI-CAA, EcoRI-ACC/MseI-CAG, EcoRI-ACT/MseI-CAG and EcoRI-AAC/ MseI-CAA were superior in detecting genetic variation. Almost all primer combinations were positively correlated in estimating pair-wise genetic similarity coefficients, indicating the similarity of primer combinations in detecting genetic variation among genotypes. All genotypes were independently distinguished and pair-wise genetic similarity coefficients ranged from 0.851 to 0.982 with an average of 0.915. Ten genotypes had 18 specific AFLP markers which could be utilised for genotype fingerprinting. AD1291 and AD1491 were the most dissimilar while AD3591 and AD3991 were the most similar genotypes. Genotypes were classified into two groups using UPGMA method of cluster analysis. Genotypes did not cluster according to collection regions. Results unveiled the presence of genetic variation among genotypes collected in each region and the presence of close genetic similarity among some coffee genotypes collected from different regions. The genetic diversity among coffee genotypes from southwestern Ethiopia was higher compared to those from northwestern Ethiopia. Genetic diversity analysis performed using six SSR primer pairs indicated the presence of considerable genetic variation among some of the genotypes collected from northwestern and southwestern Ethiopia. Twenty polymorphic SSR markers were amplified of which two were specific to genotypes AD1491 and AD2991. Using 20 polymorphic SSR markers, 64.3% of the genotypes were independently distinguished. Pair-wise genetic distances ranged from 0.286 to 1.000. The lowest pair-wise genetic similarity coefficient was recorded between AD0591 and AD1491 as well as AD1491 and AD4591. Genotypes from southwestern Ethiopia were more diverse compared to those from northwestern Ethiopia. Molecular markers differ in the amount of information generated per PCR reaction. The efficiency of 10 AFLP primer combinations and six SSR primer pairs in detecting genetic variation was compared using 28 C. arabica genotypes. AFLP primer combinations amplified 220 and SSR primer pairs 20 polymorphic alleles among evaluated genotypes. AFLP markers independently distinguished all evaluated coffee genotypes whereas SSR markers distinguished 64.3% of the genotypes. Moreover, 18 genotype specific AFLP markers compared to two SSR genotype specific markers were identified. The assay efficiency index of AFLP markers (22) was far superior to that of SSR markers (3.3). AFLP markers surpassed SSR markers in detecting genetic variation among evaluated arabica coffee genotypes. Results of this study indicated the presence of good congruence between the two genetic markers in estimating pair-wise genetic similarity coefficients among genotypes.