Abstract:
Effective conservation and utilization of genetic resources of tea are essential for
sustainability and increasing productivity of this valuable export plantation crop. No systemic
assessments of genetic variability in tea have been carried out in Sri Lanka except for a few
preliminary morphological or isozyme based studies. In the present study, DNA markers,
Randomly Amplified Polymorphic DNAs (RAPDs) and Amplified Fragment Length
Polymorphisms (AFLP) were applied to elucidate genetic relatedness of Sri Lankan tea
(Camellia sinensis L.) genetic resources.
Twenty RAPD primers that produced consistent amplifications in pretesting were selected
and used to generate 223 polymorphic loci. A considerable range of genetic diversity (genetic
distances ranging between 0.17- 0.58) was observed with an average genetic distance of 0.37.
In both ‘TRI cultivars’ and ‘estate cultivars’ not much diversity was found between (0.38) or
within two groups (0.35 & 0.33 respectively). Two phenotypically distinct cultivars,
‘Yabukita’ and ‘China’ varied significantly from both ‘TRI cultivars’ and ‘estate cultivars’.
Furthermore, ‘TRI 777’, ‘TRI 2016’ and ‘TRI 4006’ cultivars demonstrated a higher degree
of genetic divergence. ‘TRI 777’ showed closer relationship to ‘estate cultivars’ (0.39) than to
‘TRI cultivars’ (0.42). Cultivars ‘TRI 2016’, ‘TRI 4006’, ‘TRI 777’, ‘China’ and ‘Yabukita’
have been identified as progenitors that could be very effective in generating heterosis in the
breeding programmes. The clear separation of ‘TRI cultivars’ and ‘estate cultivars’ into two
distinct clusters in the dendrogram, confirmed most of their origins. The sub clusters within
the main clusters clearly conformed parentages and historical pedigree records of tea
germplasm accessions studied.
In the AFLP study, as revealed in the RAPD study, ‘estate cultivars’ were separated into two
separate groups with HS10A and MT 18 separating from the rest. Distinct accession
‘Yabukita’, TRI 777, TRI 4071 and TRI 4079 also illustrated a closer affinity to ‘Assam’ and
‘China’ hybrids. Furthermore, in the AFLP study too, most of the ‘TRI developed cultivars’
clustered together confirming their similar origin of the ancestral type ASM 4/10. The
clustering pattern of accessions TRI 4071, TRI 4079, TRI 777, TRI 3072, TRI 4052 and TRI
2043 was in agreement of their historical pedigrees as well as the findings of the present
RAPD study.
vii
Hence results generated from the AFLP study confirmed the results of RAPD study and
pedigree data. Again, the sub clusters within the main clusters reflected the parentages and
nicely confirmed historical pedigree data.
The major limitation for the application of DNA markers in tea is unavailability of teaspecific
SSR primers. In the next part of this study, isolation and characterization of genomic
SSR primers were attempted. In conventional method, cultivar TRI 2023 was used to
construct a X zap genomic library. Sequences generated from the genomic clones consisted
with SSR motifs (GAA)s and (CA)s. In the enriched method, cultivar TRI 2023 was used for
the development of (CA/GT)n and (CTT/GAA)n enriched genomic libraries. Out of the
sequences generated, 945 (84%) were mined for SSRs, where 339 (36%) were identified as
containing SSR-positive sequences. Two hundred and seventy five of above 945 sequences
consisted of perfect microsatellite repeat motifs (29.1%) and 252 (26.7%) were imperfect
repeats. A total of 625 SSRs (331 perfect and 294 imperfect) were detected. Majority of the
isolated perfect repeats were di- nucleotides (67%) followed by tri-nucleotides (27%).
(CA/GT)n and (CTT/GAA)n were predominant in the genomic libraries. A total of 305 (157
perfect and 148 imperfect) genomic-SSR primer pairs (PPs) were developed.
Construction of detailed genetic linkage maps is a fundamental first step towards application
of marker assisted breeding (MAB) in cultivar improvement programs. Although a practical
genetic map is not available for tea, the potential of maker assisted selection is clear and
strong. Thus, an attempt was made to construct a genetic linkage map of tea. Fi full-sib
population was developed according to a two-way pseudo-testcross strategy. Two parents and
six Fi progenies were first screened with 384 genomic and EST- SSR primer pairs, and 216
polymorphic primers were identified. Finally, 104 of the EST and 86 of the genomic SSR
primers were used for construction of two genetic linkage maps. The male parent (TRI 2023)
map was constructed with 83 EST- SSR and 97 genomic SSR loci. A total of 180 markers
were grouped in 16 linkage groups (LGs) covering a total length of 1,227 cM with an average
marker density of 6.8 cM/ marker. In the female parent (TRI 2043) map, 146 markers were
clustered in 15 LGs spanning 1,018 cM with an average distance of 7.0 cM between adjacent
markers. Fifteen LGs were consistent with the basic chromosome numbers of tea