Abstract:

Late blight, caused by Phytophthora infestans, is a major disease limiting potato yield and
productivity in Uganda, especially in the highlands, and accounts for up to 70% of the yield losses
and sometimes-even results in a total crop destruction. Breeding for host plant resistance is a
sustainable approach to late blight control and management in Uganda,as other measures are
expensive, ineffective or
varieties is imperative to address the changing weather patterns and crop adaptation in different
agro-ecologies.The overall goal of this study was to contribute to food security in Uganda through
developing high yielding and early maturing potato genotypes that are resistant to the late blight
disease. The specific objectives of the study were to: (1) document farmers’ knowledge of preferred
traits in potential new varieties and their perspectives on late blight prevalence and severity, (2)
phenotypically characterize potato genotypes in Uganda, (3) assess the genetic diversity among
potato genotypes using SSR markers, (4) determine the yield response of potato genotypes to late
blight disease in the tropical highlands of Uganda, (5) determine the combining ability effects for
yield, yield related traits, resistance to Phytophthora infestansand early maturity in potato
genotypes.
A participatory rural appraisal (PRA) involving 577 individual farmers showed that farmers’practices
in potato production and disease management varied greatly. Most of the respondents used farm
saved seed, while only 2% from Eastern and South-western Uganda obtained seed from research
stations.The major pests were aphids and cutworms, while late blight and bacterial wilt were the
prominent diseases. Commonly grown varieties were Rwangume and Victoria. The most preferred
attributes in new varieties were high yield, resistance to late blight, early maturity and
marketability,in that order. Late blight has been experienced by 98% of the farmers and 96% of
these reported to have used fungicides to manage the disease.
Phenotypic characterization of 48 potato genotypes in Kachwekano and Karengyere research
stations showed significant sites differences(P < 0.01) among the genotype performances for all
measured parameters. The cluster analysis using15 phenotypic traits grouped genotypes into three
major clustersand the similarity distance ranged from 0.5 to 1.0. The mean tuber yield for the two
sites was 29.8 t ha’1 and tuber yield was higher in Kachwekano than Karengyere. The best yielding
genotypes were 396038.105 (54.5 t ha'1) and NAKPOT5 (50.9 t ha'1). Fifty two percent of the
genotypes were high yielding (30 t ha'1 and above) and the most stable genotypes in terms of tuber
yield were Rutuku, 395112.32, 395017.14 and 393220.54.
Forty-eight genotypes, including advanced clones from population B3C2 of the International Potato
Centre, commercial and farmers’ varieties, were evaluated under two environments for two seasons
to determine their reaction to late blight in an 8 x 6 alpha lattice design with three replications.
Genotypes showed significant differences in yield and resistance to blight. The most resistant
genotypes were 395077.12 and 392657.8 with disease severity of 12% and 14%, respectively. The
mean tuber yield under late blight infection was19.8 t ha'1 and the best yielding genotype across
sites was 395112.32 (35.6 t ha'1). The following genotypes; 395112.32, 391919.3, 393220.54.
393077.54, 396038.107. 392657.8, Kinigi, 395014.17, NKRN59.58, NKRK19.17 and 395011.2 had
high yield and high to medium resistance to Phytophthora infestansand thus were identified as
promising parents for subsequent crosses.
In addition, 20 selected tetrapioid potato genotypes were characterized using 16 SSR markers to
determine the pattern and level of genetic diversity amongst themto identify suitable parents for
breeding purposes. The microsatellites showed considerable variation among genotypes and the 16
primer pairs amplified 64 alleles. The number of polymorphic alleles per locus ranged from 2 to 8,
polymorphic information content (PIC) values from 0.0948 to 0.7832, while heterozygosity values
ranged from 0.0997 to 0.805.A dendrogram was constructed using UPGMA clustering algorithm
based on SSR data matrices, and this grouped the potato clones into three major clusters with a
genetic distance of 1 .Oto 5.7.
Combining ability effects for resistance to late blight, yield and yield related traits were determined
using 12 potato genotypeswere crossed in a North Carolina mating design ll(NCD ll)in two sets of
six parents each to generate 18 families. Both additive and non-additive gene actioncontrolled yield
and late blight resistance in potato. However, additive gene action was predominant over nonadditive
for total tuber weight and late blight resistance. Broad-sense heritability estimates were
0.78 for total tuber weight and 0.68 for relative area under disease progress curve (rAUDPC).
Parents Kinigi, 392657.8, 396034.103, 396038.107, 395011.2, NKRK19.17, NKRN59.58 and
395017.14 had good general combining ability (GCA) effects for both late blight disease resistance
and yield related traits. Crosses 392657.8 x 395017.14 and 396038.107 x NKRN59.58 had the
highest specific combining ability (SCA) effects for all the yield related traits, while families Kinigi x
NKRK19.17and 392657.8 x NKRN59.41 had the lowest SCA effects for rAUDPC.This study showed
some evidence of maternal effects for rAUDPC (1.45) and ATW (1.56), although these were not
significant at P5O.O5. The selected parents and families were the best candidates to develop
improved potato varieties that combined both high yield and resistance to late blight.These will be
subjected to further clonal evaluationbefore possible release.
Combining ability effects for yield and yield related traits and earliness were also determined in a
different set of 12 genotypes divided into two sets of six parents each and crossed in a North Carolina mating design II (NCD ll)mating scheme to generate 18 families. Additive gene action
predominantly controlled days to flowering, total number of tubers, total tuber weight and average
tuber weight. Broad-sense heritability estimates were 0.70 for total tuber weight and 0.78 for days to
flowering. Mean total tuber yield was 9.3 t ha’1, while the average number of days to flowering was
54. Parents Rwangume, 396038.107, 395011.2 and NKRK19.17 had desirable GCA effects for the
number of days to flowering. For yield and related traits, parents 396038.107, 393077.54,
Rwangume, NKRK19.17, Kimuri, and 392657.8 had desirable GCA effects. The selected parents
had desirable attributes for high yield and early maturity and families will be subjected to further
clonal evaluation and selection.
Overall, this study documented the potato production constraints, farmers’ perceptions on late blight
management and varietal attributes. It also identified potato genotypes with desirable combining
ability for tuber yield, late blight resistance and early maturity. The nature of gene action controlling
yield, late blight disease resistance and early maturity was also determined. The predominant
control of additive genetic effects on most of the tested traits implies that advances in breeding can
be made through population improvement methods and selection. Nevertheless, for traits with
significant non- additive genetic effects, a hybridization breeding strategy could be employed to
develop hybrids. In general, clones with these traits can be selected and evaluated further for
eventual release as varieties in Uganda.