Molecular Plant Breeding 2016, Vol.7, No.11, 1
-
16
13
India which is probably a source of the first peppers
supplied to Eritrea and 4) Mexico which is considered
the centre of diversity in addition to improved
varieties developed in AVRDC. Due to variation
observed within sample collections during the survey
and seed collection as well as morphological
characterization, three individual plants from each
seed collection was applied as a sampling strategy for
genotyping. However, due to seed viability constraint,
in some cases only one or two plants from each were
used, therefore a total of 407 individual plants were
genotyped
Farmer seed samples were collected from individual
farmers in 24 sub-regions of 4 administrative regions
of Eritrea (Table 7). Villages from where seed was
collected in each sub-region are in close distance
ranging from 4 to 10 Km except in Gindae where the
villages are in three distinct areas; Midland, lowland
and coastal areas. Most farmers are small scale
farmers who grow pepper in close proximity to each
other in lands usually less than one hectare and as
small as 0.015 ha. The collected seed materials are not
named varieties or named landraces but kind of
heirlooms maintained by farmers through selecting the
best pods or plants and transferred from generation to
the next. These are exchanged among farmers within
the village, sub-region and beyond. Under such
conditions genotypes has an opportunity to interbreed
and create a common gene pool leading to formation
of populations. Thus samples collected from each
sub-region are considered as a population. The term
farmer variety will be used for describing each seed
sample collected from an individual farmer. Breeding
lines of each NARI and HAC (Table 7) are a result of
mass selection from local seed in two independent
breeding programs, germplasm of each of the two
institutions has its own common ancestor, therefore
each represents a population. Detailed information of
each collected seed sample is provided in the
supplementary tables 1, 2 and 3.
4.2 DNA extraction
The Mace et al. (2003) protocol with some minor
modifications was used for DNA extraction. Seventy
mg of leaf samples stored at -80oC was cut into small
pieces and placed in 12 x 8 strip tubes and macerated
in the presence of DNA extraction buffer (100 mM
Tris-HCl [pH 8], 1.4 M NaCl, 20 mM EDTA, CTAB
[2-3% w/v], DDT [0.03-3% v/v]) using a
Geno/Grinder 2010 (Spex Sampleprep), followed by
chloroform-isoamylalcohol (24:1) extraction, DNA
precipitation, purification and resuspension in TE.The
amount and the quality of genomic DNA were
determined using a Nanodrop (Thermoscientific) and
1% agarose gel electrophoresis.
4.3 Genotyping
Gradient PCR was used to optimize conditions for the
44 fluorescent-labelled SSR markers obtained from
AVRDC. A gradient of 11 temperatures from 50 to 60
oC was tested on two samples. All markers amplified
SSR loci except five that were therefore excluded
from the study. Of the remaining 39 markers, only 36
markers were used for amplification. The PCR
conditions were initial denaturation at 95oC for 5
minutes followed by 35 cycles of 94oC for 30 seconds,
annealing for 1 minute (Table 8) and extension at 72
oC for 1 minute and final extension at 72 oC for 20
min. PCR products were examined on 2% agarose
gels. Sets of 4 markers each with a different colour
labels, were co-loaded and genotyped using the
Genetic analyzer ABI 3730, (Applied Biosystems).
Alleles were called using GeneMapper version 4.1
(Applied Biosystems). Out of the 36 markers eight
were excluded because of unreliable peaks
(AVRDC-PP17 and AVRDC-PP68) or having more
than 10% missing data (AVRDC-PP83, AVRDC-PP86,
AVRDC-PP117, AVRDC-PP135, AVRDC-PP137 and
AVRDC-PP160) while the remaining 28 markers
proceeded to statistical analysis of (Table 8).
4.4 Data Analysis
The 28 markers were analyzed with different software.
PowerMarker (Liu, 2001-2004) was used to determine
gene
diversity,
heterozygosity,
polymorphic
information content, number of alleles in each marker
and allele frequency. Genetic distance, Analysis of
molecular variance (AMOVA), correlation, genetic
dissimilarity and Number of effective emmegrants
was calculated using GeneAlex (Peakall and Smouse
2006, 2012). Darwin (Perrier and Jacquemoud-Collet,
2006) was used for clustering the genotypes in a
dendrogram or tree and for running factor analysis
based on five axis. STRUCTURE version 2.3.4
(Pritchard et al, 2012) was used for a model-based
clustering for inferring population structure using
genotype data. The program was used for
