Genetic and Phenotypic diversity of cultivated Robusta coffee (Coffea canephora Pierre) in Uganda and effect of environmental factors on quality
Cultivated Coffea canephora in Uganda has low productivity with unknown quality trait variability. At present in Uganda, improvement of quality aspects of the Coffea canephora that would directly translate to increased incomes is hindered by the lack of information on the extent of the genetic and phenotypic diversity of the cultivated C. canephora and the effects of environmental influence on the coffee quality. The objective of this work was to evaluate the genetic and phenotypic diversity of cultivated C. canephora and assess the effects of the environmental influence on quality traits. Genotypic variation, morphological and green bean physical characteristics, biochemical compounds and organoleptic cup attributes present in the local landraces were evaluated. The effects of environmental factors such as soil elements and texture, locations, altitude, rainfall and temperature at different bean development stages were also assessed. In the experiments reported here, a total of 265 Robusta coffee genotypes collected from farmer fields in 13 traditional growing districts were analyzed using 32 Simple Sequence Repeats and were compared with 84 on station germplasm collections. SSRs efficiency and reliability was reflected by the low values for quality index and average minimum variance, detection of abundant and rare alleles to the extent of differentiation exemplified by allelic drift. Three genetic diversity groups with a mean genetic distance of 0.60 were different from those from Western and Central Africa possibly due to isolation or controls were few. Genetic diversity values were not antagonistic as all relationships were positive. High genetic variability among cultivated Robusta coffee was demonstrated by the high Polymorphic Information Content (PIC) of 0.55 to 0.93 with a mean of 0.75 for 18 selected markers. The 54.05% within individual variability implied that genetic selection for desirable traits should be based on individual plants. Fairly polymorphic change over generations was revealed by the genetic differentiation value of 0.17 and gene flow mean value of 0.7±O.09 that caused exchange of one migrant between populations in each generation. The germplasm collection had fewer alleles, effective alleles, PIC, allele frequency and rare alleles as compared to accessions from farmer fields reflected limited variability among the few genotypes in the collection. Research controlled hybrid crosses and commercial varieties though genetically and morphologically, had limited genetic variability as each type confined to single groups. Evaluation of morphological characters for 476 Ugandan farm Robusta coffee accessions showed that farmers cultivate predominantly 88% "uganda" and "erecta" landraces with improved types constituting 12%. Though 'uganda' and 'erecta' types were not genetically defined, were morphologically close and significantly different. The 206 Robusta coffee green bean physical characters evaluated using International Coffee Organization (ICO) criteria showed that 61.27% were of average screen size 15 and 19% larger beans. The 16 accessions caffeine content analyzed using both Near Infra Red Spectroscopy (NIRS) and High Performance Liquid Chromatography (HPLC) methods were insignificantly different making NIRS technique better option because is fast and economically viable. Ugandan Robusta coffee biochemical compounds were aggregated independently from that of CIRAD data base possibly because it was CIRAD's first encounter with Ugandan accessions and the possibly high species biochemical compound variability. The high fat (10.44 to 15.94 % dry matter) and chlorogenic acid (10.88 to 15.64 % dry matter) with a wide range of sucrose content (2.48-7.34% dm) indicated that Ugandan Robusta coffees are diverse with high fragrance, aroma and flavour. Other than potential to serve many markets, quality improvement and conservation purposes, the 4 multivariate groups were significantly different for dry matter and caffeine, implying it possible to select for dense beans with low caffeine to reduce bitterness in coffee brew. Genotypes in the three major genetic diversity groups distributed differently in the four multivariate green bean biochemical compounds diversity groups most likely due to the influence of environment on the latter. Lack of genetic diversity relationships with locations where they were grown meant that trait variations were strongly influenced by genotypes. Evaluating roasted coffee green beans using the sensory technique developed by The Coffee Quality Institute of America (CQIA) alongside Uganda Coffee Development Authority, showed that the evaluator organoleptic cup trait rating was significantly different implying that individuals perceive cup taste differently. Positive correlation among organoleptic cup attributes implied that they complimented each other instead of antagonizing. Cup balance regression coefficient of 0.90 rated it as most important in overall cup evaluation. About 84% of Ugandan Robusta coffee brew was premium reflecting high quality. Acidity was least rated as average while other cup attribute rating ranged from premium to specialty grade pointing out that Robusta coffee can have as good cup as Arabica coffee with medium acidity. Fragrance, aroma and flavour attributes could easily be selected from the 4 multivariate groups because because the attributes were significantly different. The present study has estimated optimum expression of different phenotypic quality traits at 30 to 35 years at a height of 1200 m.a.s.L The genetic diversity and multivariate variability derived from this study offer numerous opportunities to select for breeding materials to develop diverse flavours that serve different markets, enrich existing germplasm collection, design further desirable trait studies and advice policy makers and coffee industry stakeholders. Cultivating at 1201-1300 m above sea level, use of "uganda" and "erecta" landraces, selecting trees of 31-40 years are some of the genotypic and environmental management practices that could be used to improve cup acidity. Soil, leaf and green bean essential elements were detected and quantified using Energy Disperse X-Ray Florescence. Calcium, manganese, iron, zinc and lead mean content were higher in the soil and lowest in the green beans implying that not all the soil elements were taken up by plant. Potassium content in coffee was highest in the green bean because of its involvement in the metabolism and translocation of compounds such as oil, proteins and carbohydrates that influence cup quality. Although calcium is known to influence fruit ripening and quality, its content was highest in the leaves possibly because it's a major constituent of cell walls and its content, involved in crucial functions such as cell division, root and leaf development, various enzymatic functions and transport carbohydrates. Multivariate analysis of the effects of phenotype and environmental factors on quality traits revealed none of the variables was explained 100% by a single or a combination of phenotypic traits, elements, rainfall and temperature bean development stages suggesting that this trait is complex to select. Soil, leaf and bean elements explained most variability in organoleptic cup attributes (6.3 to 22.1%), bean physical characters (24.3 to 50.9%) and morphological characters (13.9 to 44.4%), possibly because elements are responsible for plant functions, vegetative development and constitute substances in cells based on genotypes. The highest chemical compound variability of 3.7 to 35.5% that was explained by rainfall and temperature bean development stages may have been a requirement for moderating plant function and facilitating access and movement of elements from the soil and in the plant. Most desirable bean size was average screen size 15 which more sucrose and less dry matter, caffeine and trigonelline content with better aftertaste, acidity, flavour. This thesis evaluated cultivated C. canephora from traditional growing areas for genetic and phenotypic variability as well as influence of some environmental factors on quality traits. Information that has been derived is important not only in the identification of needed future research goals but the analysis of such diversity also goes a long way to contributing to an improved national economy and to raising the living standards of small-scale farmers of Uganda. Although all methods that have been used to assess quality traits are important, in case of meager resources and limited time, genetic diversity studies on quality can be complemented with fast, easy to use NIRS technique validated by organoleptic cup assessment which determines acceptability.