Bioagro 29(3): 153-162.

2017

MOLECULAR MARKER-BASED CHARACTERIZATION OF

ECUADORIAN DRY FOREST TAMARIND PLUS TREES
Leidy Sarmiento1, Iris Pérez-Almeida1, Byron Díaz1, Hugo Álvarez2 and William Viera2

ABSTRACT
To improve the potential of tamarind as an economically valued domesticated species it is important to characterize its variability
in Ecuador for breeding purposes. Our aim was to investigate the genetic diversity of 32 tamarind plus trees using inter-simple
sequence repeat (ISSR) markers. Eighty four loci were examined using 12 markers, with a mean number of 4.42 loci per primer; 8
loci (9.52 %) were monomorphic and 76 (90.48 %) polymorphic, revealing genetic variability among the individuals.
Polymorphic information content (PIC) values varied from 0.29 (ISSR_808) to 0.93 (ISSR_HB12), whereas the marker index
ranged from to 26.4 (ISSR_814) to 62.5 (ISSR_17899A). Primers ISSR_HB11, ISSR_836, ISSR_842, ISSR_848, ISSR_860,
ISSR_17899A and ISSR_17899B were useful to discriminate the grouping of the accessions according to their PIC values. Ward
cluster analysis grouped accessions into two major groups with five subgroups with 46 % similarity according to Jaccard distance.
The genotypes from Loja, Manabí and Guayas provinces were grouped in the first cluster; while only individuals from Manabí
located in the other group, indicating major diversity in the latter province. Genotypes T1-ECUM-001 and T1-ECUM-002
presented 76 % similarity, while T1-ECUM-008, T1-ECUM-010, T1-ECUM-012, T1-ECUM-017 and T1-ECUM-018 shared
60 %. All materials from Loja grouped with 65 % similarity. Other genotypes clustered with similarity of 54 %. The cophenetic
correlation coefficient (0.634) showed a good fit between the data matrix and the dendrogram results. A reasonable degree of
diversity was found among tamarind genotypes potentially useful to select plus trees for clonal propagation as well as to identify
diverse parents for hybridization programs.
Additional key words: Fruit tree breeding, ISSR, molecular markers, Tamarindus indica

RESUMEN
Caracterización molecular de árboles élite de tamarindo del bosque seco ecuatoriano
Para aumentar el potencial del tamarindo como especie domesticada con valor económico es importante caracterizar la
variabilidad en Ecuador con propósitos de mejoramiento. Nuestro objetivo fue investigar la diversidad genética de 32 árboles élite
de tamarindo utilizando marcadores de secuencia inter-simple repetida (ISSR). Se examinaron 84 loci con 12 marcadores, con un
número promedio de 4,42 loci per primer; 8 loci (9,52 %) fueron monomórficos y 76 (90,48 %) polimórficos, revelando
variabilidad genética entre individuos. El contenido de información polimórfica (PIC) osciló entre 0,29 (ISSR_808) y 0,93
(ISSR_HB12), mientras que el índice de marcador fluctuó entre 26,4 (ISSR_814) y 62,5 (ISSR_17899A). ISSR_HB11,
ISSR_836, ISSR_842, ISSR_848, ISSR_860, ISSR_17899A e ISSR_17899B fueron útiles para discriminar accesiones según sus
CIPs. El análisis de conglomerados de Ward formó dos grupos principales y cinco subgrupos con 46 % de similitud según la
distancia de Jaccard. Genotipos de Loja, Manabí y Guayas se aglomeraron en un grupo; mientras que sólo accesiones de Manabí
quedaron en el otro, indicando mayor diversidad en la última provincia. Los genotipos T1-ECUM-001 y T1-ECUM-002
presentaron 76 % similitud, mientras T1-ECUM-008, T1-ECUM-010, T1-ECUM-012, T1-ECUM-017 y T1-ECUM-018
compartieron 60 %. Todos los materiales de Loja se agruparon con 65 % de similitud. Otros genotipos se concentraron con
similitud de 54 %. El coeficiente de correlación cofenética (0,634) mostró buen ajuste entre la matriz de datos y los resultados del
dendrograma. Se encontró un grado razonable de diversidad entre los genotipos de tamarindo potencialmente útil para seleccionar
árboles élite para propagación clonal así como para identificar progenitores diversos para programas de hibridación.
Palabras clave adicionales: ISSR, marcadores moleculares, mejoramiento de frutales, Tamarindus indica

INTRODUCTION donous perennial tree with a wide geographical

distribution in the subtropics and semi-arid
Tamarind (Tamarindus indica L.) is a dicotyle- tropics. It is native to areas throughout Africa and

Recibido: Diciembre 16, 2016 Aceptado: Mayo 31, 2017

1
Dpto. Nacional de Biotecnología, Estación Experimental Litoral Sur, Instituto Nacional de Investigaciones Agropecuarias
(INIAP). Guayas, Ecuador. e-mail: [email protected]; [email protected] (corresponding author)
2
Programa Nacional de Fruticultura, Estación Experimental Portoviejo, Estación Experimental Santa Catalina,
INIAP. Ecuador. e-mail: [email protected]; [email protected]

153
154
Volumen 29 (2017) BIOAGRO N° 3

Southern Asia (Tapia et al., 2012), although the Identification of cultivar and estimation of genetic
precise origin of this species is a subject of diversity using phenotypic markers have several
controversy (Diallo et al., 2008). It was introduced limitations, especially in perennial crops
into America during the 16th century and now (Purushotham et al., 2008). However, molecular
grows widely in tropical and subtropical areas. In diversity using DNA and protein-based molecular
Ecuador, this tree species is cultivated in areas markers are more reliable and unaffected by
where production systems include scattered trees, environmental factors (Dhanraj et al., 2002).
mainly for local consumption. Establishment of core collections based on field
The tree has an important role in local evaluation and molecular variation shown by
economies, supplements the local diet, it is used in accessions could be obviously advantageous.
traditional and modern therapies by 80 % of the However a clear and detailed assessment of
world’s population in Africa, Asia and Latin molecular diversity in tamarind is not currently
America, and it is used as a laxative and purgative available (Gangaprasad et al., 2013), although
with minimal side effects (El-Siddig et al., 2006). there have been some efforts to characterize
Furthermore, it also exhibits antibacterial, several tamarind populations with molecular
antifungal, and antioxidant properties (Graf et al., techniques such as RAPD (Diallo et al., 2007;
2016), and is used as a construction material, and Gangaprasad et al., 2013; Kumar et al., 2015) and
for fuel and fodder (Tapia et al., 2012). AFLP (Algabal et al., 2011).
Pharmaceutical companies have invested money In the present study, ISSR have been used for a
and time in developing natural products extracted deeper molecular analysis of genotypes because of
from this tree to generate remedies that are the advantages of this technique over SSR and
affordable (Doughari, 2006). RAPD (Reddy et al., 2002). The ISSR markers are
Despite its commercial importance worldwide, highly polymorphic and represent a simple,
this multi-purpose tree has been little investigated reproducible, efficient and quick method that
(Algabal et al., 2011), although it was identified as combines most of the advantages of microsatellites
one of the top ten agroforestry tree species to be (SSRs) and amplified fragment length
prioritized for crop diversification programs and polymorphism (AFLP) to the universality of random
development in sub-Saharan Africa in efforts to amplified polymorphic DNA (RAPD). ISSR
enhance the conservation and utilization of genetic markers have high reproducibility possibly due to
species (Gunasena & Hughes, 2000). the use of longer primers (16-25mers) as compared
Tamarind has a relatively long generation time to RAPD primers (10 mers) (Reddy et al., 2002).
and reproduces primarily by outcrossing, so any This research was conducted to estimate
conventional breeding approaches would require genetic diversity and to assess relationships
considerable investment in time and money. among 32 accessions of tamarind using ISSR
Although it is one of the oldest domesticated markers for the basis of a breeding program in
crops, little is known about its genetic Ecuador.
characteristics and population biology. Available
knowledge focuses on developing efficient in situ MATERIALS AND METHODS
conservation and genetic improvement strategies
(Fandohan et al., 2010). Two key elements for Plant material. The experimental material (Table
cultivar development are the identification of 1) comprised 32 tamarind accessions
‘‘plus trees’’ in natural populations and their (geographically distinct) that were collected from
propagation by vegetative techniques (Leakey & January to December 2015 in three provinces of
Page, 2006). Tamarind is mostly self-sown or Ecuador (Guayas, Manabi and Loja), in dry forest
sown with seeds of unknown parentage, which areas with less than 500 mm rain per year,
results in wide variation among seedling progenies temperatures from 20 to 25 °C, high luminosity,
(El-Siddig et al., 2006). and low nutrient soils.
Characterization of tamarind trees has been Young and healthy leaves were harvested
mainly limited to descriptions of morphological individually in the field, tagged, submerged in a
and agronomic traits, which are known to be solution of polyvinylpyrrolidon (PVP) 1 %, placed
deeply affected by environmental factors. in paper envelopes and transported to the
 155
Sarmiento et al. Molecular characterization of Ecuadorian tamarind trees

Biotechnology Department, Instituto Nacional de Experimental Litoral Sur, Guayas Province, for
Investigaciones Agropecuarias (INIAP), Estacion DNA extraction.

Table 1. Codes and collecting site of the tamarind plus accessions analyzed in this research
Location Coordinates Elevation
Accession code Owner´s name
Site County Province Longitude Latitude (masl)1
TI-ECUM-001 Sócrates Quimis Joa Jipijapa Manabí 26°36´2.0" 01°05´49.1"
TI-ECUM-002 26°36´2.0" 01°05´48.7" 79.25
TI-ECUM-003 26°36´1.6" 01°05´48.3"
TI-ECUM-004 Marco Zambrano Cantagallo Manabí 80°48´27.3" 01°17´01.2"
105.77
TI-ECUM-005 80°43´27.8" 01°17´00.7"
TI-ECUM-006 Viterbo Navarrete Macías El Cady Manabí 80°24´19.6" 01°07´04.5" 62.79
TI-ECUM-007 Fradil Parraga Macías Maconta Portoviejo Manabí 80°21´16.6" 01°02´17.8"
83.21
TI-ECUM-008 80°21´16.8" 01°02´19.4"
TI-ECUM-009 Vicenta Ramírez Macías Tabacales Rocafuerte Manabí 80°26´43.2" 00°56´32.2" 36.58
TI-ECUM-010 Calixto Ruiz Valdez Manabí 80°26´31.8" 00°56´52.6" 45.42
TI-ECUM-011 Bosco Giler Parraga El Cardón Manabí 80°23´62.5" 00°54´55.7" 44.20
TI-ECUM-012 Honorato Navia Navia La Balsita Manabí 80°23´38.5" 01°00´25.8" 44.81
TI-ECUM-013 José Roque Cevallos La Horma Manabí 80°23´44.9" 00°54´41.2" 78.33
TI-ECUM-014 José Zamora Arteaga Las Flores Manabí 80°21´22.4" 00°55´37.5" 104.85
TI-ECUM-015 Ulbio Muentes Zambrano Zapatón Manabí 80°28´22.9" 00°53´14.7" 21.03
TI-ECUM-016 80°28´22.5" 00°53´14.3" 24.99
TI-ECUM-017 Manuel Zambrano Cristo Rey Sucre Manabí 35.97
80°29´39.1" 00°49´05.4"
Figueroa
TI-ECUM-018 Geravides Lucas Herrera El Blanco Manabí 80°29´45.9" 00°49´02.1" 26.52
TI-ECUM-019 Eduardo Castro Choez Costa Rica Portoviejo Manabí 80°27´43.0" 00°59´54.8" 28.35
TI-ECUM-020 Verónica Pinargote El Retiro Manabí 39.62
80°28´37.5" 00°59´31.6"
Vergara
TI-ECUM-021 INIAP-E.E. Portoviejo Lodana Santa Ana Manabí 80°23´16.8" 01°10´13.6" 73.15
TI-ECUM-022 Diocles Pico Barrezueta Manabí 80°38´3.99" 01°19´96.6"
65.53
TI-ECUM-023 80°38´3.54" 01°19´96.2"
TI-ECUM-024 Gloria Coloma Garofalo Mate Manabí 80°33´23.0" 01°22´87.5" 96.01
TI-ECUM-025 Marcela Ortega Zambrano Los Tillales 24 de Mayo Manabí 80°25´3.39" 01°15´0.86" 113.69
TI-ECUM-026 José Delgado Varela El Guarango Rocafuerte Manabí 80°24´14.3" 00°53´4.39" 43.28
TI-ECUG-027 Ana Decimaviya Valle de la VirgenPedro Carbo Guayas 80°11´46.2" 01°44´33.9" 77.42
TI-ECUM-028 Domingo Moran Macías Guale Paján Manabí 80°12´29.3" 01°40´50.4" 110.34
TI-ECUL-029 INIAP Garza Real Zapotillo Loja 80°13´58.3" 04°18´24.3" 236
TI-ECUL-030 INIAP Garza Real Zapotillo Loja 80°13´58.0" 04°18´24.0" 236
TI-ECUL-031 INIAP Garza Real Zapotillo Loja 80°13´57.7" 04°17´58.5 233
TI-ECUL-032 INIAP Garza Real Zapotillo Loja 80°13´17.1" 04°17´59.6" 232
1
Meters above sea level

DNA extraction. DNA was extracted following shaken. Following cell lysis, 1 mL
the method reported by Khanuja et al. (1999) with chloroform:isoamyl alcohol (CIA) (24:1) was
modifications. Briefly, 120 mg fresh leaf tissue added mixing by inversion for 15 min and
were ground in a mortar, transferred to a 2 mL centrifuged at 5900 g for 10 min. The supernatant
microcentrifuge tube containing 1 mL extraction was transferred to a fresh 1.5 mL tube, adding 500
buffer (100 mM Tris, HCl pH 8.0; 1.5 M NaCl; µL 5M NaCl and 0.6 volume cold isopropanol,
25 mM EDTA pH 8.0; 2.5 % CTAB; 1 % PVP; mixed by inversion and placed at -20 °C for 1 h.
0.2 % 2β-mercaptoetanol), mixing by inversion. Samples were centrifuged at 15616 g for 10 min,
The sample was incubated at 60 °C for 1.5 h and pellet washed with ethanol at 80 %, dried for 20
156
Volumen 29 (2017) BIOAGRO N° 3
min and resuspended in 170 µL high salt TE (Tris of sterile double-distilled water.
HCl 10 mM, EDTA 1 mM, NaCl 1 M), followed DNA quantification and quality. DNA
by RNase treatment, adding 2 µL RNase A (10 quantification was estimated in a Quantus
mg·µL-1) and keeping at 37 °C for 30 min. After Fluorometer (Promega) using the Quant-iT assay
that, 100 µL CIA was added, mixing well and then kit developed by ThermoFisher Scientific. The
centrifuged at 15,000 g. Supernatant was quality was determined by running a 1 % agarose
transferred to a tube where 100 µL of cold ethanol gel.
at 100 % was added and centrifuged at 15616 g PCR amplification. A total of 19 ISSR primers
for 10 min. The pellet was washed with 80 % (Table 2) were selected to characterize the
ethanol, dried for 20 min and dissolved in 50 µL tamarind accessions.

Table 2. ISSR primers used for the comparative study of genetic variability of 32 plus trees of
Tamarindus indica L.
PRIMER Oligonucleotide sequence 5’to 3` Range (bp) Tm (°C) TBN PBN MBN POL PIC MI SE
HB12 CACCACCACGC 750-1400 48.9 3 1 2 33.33 0.93 30.9 8.1
HB11 * GTGTGTGTGTGTCC 500-2800 44.3 8 4 4 50.00 0.86 42.9 6.3
17899A * CACACACACACAAG 650-1600 52.9 4 4 0 100.00 0.63 62.5 11.8
815 CTC TCT CTC TCT CTC TG 1000-1600 50.3 2 2 0 100.00 0.59 58.6 11.3
842 * GAG AGA GAG AGA GAG AYG 400-1200 47.6 4 4 0 100.00 0.59 58.6 11.3
17899B * CACACACACACAGG 600-2200 50.3 6 6 0 100.00 0.58 58.3 9.6
860 * TGT GTG TGT GTG TGT GRA 700-2800 44.3 8 7 1 87.50 0.56 48.9 8.5
17898A CACACACACACAAC 400-1650 50.3 6 6 0 100.00 0.55 54.7 19.1
836 * AGA GAG AGA GAG AGA GYA 500-1400 44.3 3 3 0 100.00 0.53 53.1 18.2
848 * CAC ACA CAC ACA CAC ARG 160-2000 51.7 3 3 0 100.00 0.51 51.0 21.3
812 GAG AGA GAG AGA GAG AA 800-1750 44.3 4 4 0 100.00 0.43 43.0 21.1
844ª * CTCTCTCTCTCTCTCTAC 1500-4000 45.8 4 4 0 100.00 0.42 42.2 15.4
HB9 GTGTGTGTGTGTGG 1150-1500 47.6 2 2 0 100.00 0.41 40.6 27.6
873 GAC AGA CAG ACA GAC A 650-850 52.9 2 2 0 100.00 0.36 35.9 27.3
814 * CTC TCT CTC TCT CTC TA 350-1750 44.6 4 3 1 75.00 0.35 26.4 10.0
835 * AGA GAG AGA GAG AGA GYC 300-1500 52.9 4 4 0 100.00 0.34 33.6 12.9
807 * AGA GAG AGA GAG AGA GT 800-2800 44.6 6 6 0 100.00 0.33 32.8 5.9
844B CTCTCTCTCTCTCTCTGC 1200-2000 51.7 3 3 0 100.00 0.32 32.3 17.3
808 * AGA GAG AGA GAG AGA GC 600-2500 48.9 8 8 0 100.00 0.29 28.9 8.5
Total 350 -4000 84 76 8
Mean 4.42 0.50 44
Tm = annealing temperature; TBN = Total band number; PBN = Polymorphic band number; MBN = Monomorphic band
number; POL = Polymorphism percentage; PIC = Polymorphic information content; MI = Marker Index; SE = Standard
Error; IUPAC 1-letter code abbreviations for mixed oligo bases: R= A + G; Y = C + T.
Primers with asterisks were used for statistical analysis.

Polymerase chain reactions were carried out in Master Cycler 230 AG model). The amplification
a 20 µL volume, in a tube containing 2 µL 10X profile was kept for initial denaturing at 94° C for
buffer (200 mM Tris-HCl pH 8.4, 500 mM KCl); 5 min followed by 40 cycles of denaturation
50 mM MgCl2 1 µL; 10 mM dNTPs 0.8 µL; 0.4 at 94 °C for 30 s; primer annealing at
µL 0.2 µM ISSR primer; 0.12 µL Taq polymerase recommended temperature for 1 min; extension at
5U· µL-1 (Invitrogen), and 3.2 µL of template 72 °C for 2 min; and a final extension at 72 °C for
DNA (5ng·µL-1). PCR amplifications were 7 min.
performed using a thermocycler (Eppendorf The amplification products were mixed with
 157
Sarmiento et al. Molecular characterization of Ecuadorian tamarind trees
2.5 µL of 10X loading dye (0.25 %) bromophenol method (Ward, 1963) to develop the dendrogram.
blue. PCR products were resolved by To estimate congruency between the dendrogram
electrophoresis on 1.5 % (w/v) agarose gels using and the data, a cophenetic correlation coefficient
1X TAE buffer (40 mM Tris–acetate, pH 8, 1 M was calculated.
EDTA), at 100V for 85 min, followed by staining RESULTS
with 15 ppm ethidium bromide and photographed.
The molecular marker (1 kb, Invitrogen) was The banding pattern of each marker was
loaded in the last lane. examined visually. It was considered that primers
Scoring of bands and statistical analysis. Gel ISSR_873, ISSR_815 and ISSR_17898A gave
electrophoresis DNA profiles of tamarind little information; ISSR_812, ISSR_HB12,
accessions, amplified by each ISSR primer, ISSR_844B and ISSR_HB9 were only just
were used to generate a band presence (1) informative; but ISSR_HB11 showed to be an
and absence (0) matrix. Each marker was excellent primer due to its banding pattern
reviewed by assessing its banding pattern i.e. the across the population. The rest of the markers
number of yielded bands and presence or absence yielded intermediate information according to
in the studied accessions. The number of the evaluation criteria established for this
polymorphic or monomorphic loci, their research.
percentage, standard deviation and experimental Only 12 of 19 primers (Table 2) were
error were calculated. considered informative for the statistical analysis
The polymorphic information content (PIC) because of their consistency giving reproducible
was calculated according to Roldan-Ruiz et al. and good quality banding patterns; those were
(2000) where PICi (polymorphic information ISSR_807; ISSR_814; ISSR_836; ISSR_860;
content of marker ‘i’) = 2fi (1−fi); fi is the ISSR_HB11; ISSR_808; ISSR_844A; ISSR_835;
frequency of the amplified allele (band present), ISSR_17899A; ISSR_17899B; ISSR_848; and
and 1−fi is the frequency of the null allele. The ISSR_842.
PIC value ranges from zero for monomorphic This set of ISSR primers generated 84 loci,
markers to 0.5 for markers that are present in with a mean number of 4.42 loci per primer,
50 % of the plants and absent in the other 50 %. ranging from 8 (ISSR_HB11, ISSR_808 and
This content provides an estimate of the ISSR_860) to 1 (ISSR_HB12) (Table 2). Of the
discriminatory power or whether a locus or observed loci, 8 (9.52 %) were monomorphic and
loci is informative, taking into account not 76 (90.48 %) polymorphic, revealing high genetic
only the expressed number of alleles but their variability between the individuals. PIC values
relative frequencies. varied from 0.29 (ISSR_808) to 0.93
The value of each marker represents the (ISSR_HB12), with an average of 0.50, whereas
probability of finding this marker in one of two MI ranged from 26.4 (ISSR_814) to 62.5
different states (present or absent) in two plants (ISSR_17899A). Primers ISSR_HB11, ISSR_836,
drawn at random from the population. Marker ISSR_842, ISSR_848, ISSR_860, ISSR_17899A
index (MI), calculated as the product of the and ISSR_17899B were useful to discriminate the
polymorphism percentage and the PIC, is used to grouping of the accessions according to their PIC
estimate the overall utility of each marker system value above mean (0.50).
and was calculated according to Sorkheh et al. A typical polymorphic ISSR fingerprint, using
(2007). the ISSR_808 marker, is shown in Figure 1.
The Jaccard genetic similarity coefficient Genotype TI-ECUM-016 amplified only with a
was calculated for the data matrix using few primers, therefore it was not considered for
InfoStat version 2011 (Universidad Nacional de further statistical analysis, and the study was
Córdoba, Argentina). The generated data was maintained with 31 genotypes.
used to estimate genetic similarity for pairwise A moderate degree of genetic diversity was
accessions based on Jaccard similarity obtained with the Jaccard similarity coefficient.
coefficient. A similarity matrix was constructed The plus trees formed five groups (Figure 2) with
and subjected to cluster analysis following Ward’s an average of 46 % similarity, although two
158
Volumen 29 (2017) BIOAGRO N° 3
subgroups clustered together and the other three ECUM-026, T1-ECUG-027 and T1-ECUM-028,
subgroups shared another division. Genotypes T1- clustered in the largest group with an average
ECUM-001 and T1-ECUM-002 presented 76 % degree of similarity of 54 %. Other genotypes
similarity, while T1-ECUM-008, T1-ECUM-010, grouped with a similarity of 65 %.
T1-ECUM-012, T1-ECUM-017 and T1-ECUM- Using the Jaccard distance method, the
018 shared 60 %. All materials selected at Loja cophenetic correlation coefficient (0.634) was
Province grouped together with 65 % similarity. obtained and showed a reasonable fit between the
Genotypes T1-ECUM-015, T1-ECUM-019, T1- data matrix and the dendrogram results. No
ECUM-020, T1-ECUM-021, T1-ECUM-022, T1- distortion was caused by the conglomerate
ECUM-023, T1-ECUM-024, T1-ECUM-025, T1- method.

Figure 1. ISSR 808 marker showing typical polymorphic fingerprint amplification of 32 Tamarindus
indica genotypes in an agarose 1.5 % gel. M =1kb Invitrogen ladder

DISCUSSION distances and seed transported by farmers in long

distances, these two factors being important for
In this research, the analyzed tamarind the genetic variability process (Zetina et al.,
accessions showed intermediate genetic 2012). The relatively high percentage of
variability, with an average of 46 % similarity polymorphism observed could also be related to
among selected individuals, and moderate level of the natural pollination method of this species
polymorphism indicating that a wide and diverse which expands the genetic base. The tamarind
genetic base existed among the tamarind plus trees populations analyzed could be considered as
genotypes of the three Provinces of Ecuador, isolated, which favors genetic variability.
which could be explained due to the cross Figure 2 shows the formation of two major
pollinating nature of the species. We hypothesized groups. The tamarind trees from Loja, Manabí and
that there is genetic flux among the studied Guayas were grouped in the first cluster, while
tamarind populations because they are conformed only individuals from Manabí were located in the
by dispersed individuals (each one genetically other group, indicating that there is major
different), thus their cross-pollination generates diversity of this fruit tree in this Province.
greater variability and genetic recombination. In Extending the analysis, five subgroups were
addition, the distance among the majority of observed for the molecular characterization of the
sampling sites is relatively short (about 200 km 31 accessions of tamarind, with an ultrametric
between Guayas and Manabí), making possible distance of 2.06 units, whereas six groups were
pollen exchange by vectors (insects) in close formed using the Ward algorithm with the
 159
Sarmiento et al. Molecular characterization of Ecuadorian tamarind trees
phenotypic information from the morphological more specific differences for the formation of
characterization (data not shown). Therefore, groups. It seems that molecular characterization is
morphological analyses yielded clusters that did related mainly to the origin of the material, thus
not completely account for the genetic similarity with their geographical distribution, while the
found among the accessions when using molecular grouping by morphological traits could be
markers. Morphological descriptions may be influenced by external factors (agronomic and
confounded with environmental variation and environmental) that affect the phenotypic
be prone to subjective evaluations. These expression of the plant. However, different
results are comparative with those from morphotypes were noted within phenotypic
Cervera et al. (2001) who reported that characterizations, and thus it is recommended to
molecular and morphological characterization carry out crosses among individuals which
cannot be directly related, i.e. physical showed contrasting traits related to pulp
characteristics and molecular data are not percentage, number of fruit per bunch, number of
usually associated. This is evidenced by the seeds per seedcase, and fruit weight, characters
fact that individuals within groups are that are important for breeders of this fruit tree
different for both cases. The markers used in according to Diallo et al. (2008), since fruit from
this study are dominant, semi-random and all the 31 evaluated accessions are harvested for
cannot determine heterozygosity, inferring human consumption.

Distance:

Figure 2. Dendrogram based on ISSR molecular data from 31 Tamarindus indica plus trees using
Jaccard coefficient and Ward grouping. Cophenetic correlation coefficient = 0.634

The information generated can be used to tamarind populations or genotypes and have
suggest selected plus trees for clonal propagation reported genetic variability and diversity.
as well as to identify diverse parents for However, ISSR markers are highly polymorphic
hybridization programs. and therefore useful for studies of genetic
As stated before, several studies using diversity (Reddy et al., 2002), and also variation
RAPD markers (Diallo et al., 2007; Gangaprasad between and within populations can be compared
et al., 2013; Kumar et al., 2015) and AFLP using this type of marker (Qian et al., 2001). In
markers (Algabal et al., 2011) have characterized addition, higher polymorphism has been detected
160
Volumen 29 (2017) BIOAGRO N° 3
using ISSRs than any other technique (Virk et al., Suef University Journal of Basic and Applied
2000). Sciences 2(1): 120-127.
In several cases genotypes did not cluster
according to their site of collection, which was 2. Algabal, A.Q.A.Y., N. Papanna and L. Simon.
attributed to their highly cross-pollinating nature, 2011. Amplified fragment length
small distribution area and that most tamarind polymorphism marker-based genetic diversity
genotypes are grown from seed. Genotypes which in tamarind (Tamarindus indica L.).
were morphologically closely related were found International Journal of Fruit Science 11(1):
to be unrelated at the molecular level (Kumar et 1-16.
al., 2015).
A breeding program should be based on the 3. Algabal, A.Q.A.Y., N. Papanna and L. Simon.
2010. Estimation of genetic variability in
results of both morphological and molecular
tamarind (Tamarindus indica L.) using RAPD
characterization; nonetheless, choosing parental
markers. International Journal of Plant
plants based on the molecular results obtained in
Breeding 5(1): 10-16.
this study should be done by selecting individuals
with higher genetic distances and corroborating 4. Cervera, M.T., I. Rodriguez, J.A. Cabezas, J.
the phenotypic traits of the parents in the field to Chávez, J.M. Martínez-Zapater and F. Cabello.
avoid hybridizations between materials that 2001. Morphological and molecular
genetically are different but express the same characterization of grapevine accessions
phenotype. known as albillo. American Journal of Enology
Tree breeding begins through the application of and Viticulture 52: 127-135.
genetic principles basically directed towards
modifying the heredity of tree populations to meet 5. Dhanraj, A.L., E.V.V.B. Rao, K.R.M. Swamy,
the needs of the farmers (Gunasena & Hughes, M.G. Bhat, T. Prasad and S.N. Sondur. 2002.
2000). Determination of genetic variation is Using RAPDs to assess the diversity in Indian
important to plant breeders for development of a cashew (Anacardium occidentale L.)
high yielding variety (Kumar et al., 2015), and germplasm. Journal of Horticultural Science
therefore this research contributes to generate and Biotechnology 77: 41-47.
knowledge about the diversity of tamarind trees in
Ecuador in order to determine future advances in 6. Diallo, B.O., H.I. Joly, D. Mckey, M.
fruit breeding. Hossaert-Mckey and M.H. Chevallier. 2007.
Genetic diversity of Tamarindus indica
CONCLUSION populations: Any clues on the origin from its
current distribution? African Journal of
The tamarind genotypes assessed in this Biotechnology 6(7): 853-860.
research showed a reasonable degree of genetic
diversity that can be used as a basis for 7. Diallo, B.O., D. Mckey, M. Chevallier, H.I.
hybridization breeding programs. Joly and M. Hossaert-Mckey. 2008. Breeding
system and pollination biology of the
ACKNOWLEDGEMENTS semidomesticated fruit tree, Tamarindus indica
L. (Leguminosae: Caesalpinioideae):
The authors wish to thank INIAP´s Project Implications for fruit production, selective
“Fortalecimiento Institucional-Frutales” for funding breeding, and conservation of genetic
this research. resources. African Journal of Biotechnology
7(22): 4068-4075.
LITERATURE CITED
8. Doughari, J.H. 2006. Antimicrobial Activity of
1. Abdel-Hameed, U.K., U.I. El-Magly, I.F. Ishak Tamarindus indica Linn. Tropical Journal of
and M.E. Tantawy. 2013. A contribution to the Pharmaceutical Research 5(2): 597-603.
specification of Caesalpinioideae (L) based on
9. El-Siddig, K., H.P.M. Gunasena, B.A.
morphological and molecular criteria. Beni-
 161
Sarmiento et al. Molecular characterization of Ecuadorian tamarind trees
Prasad, D.K.N.G. Pushpakumara, K.V.R. 17. Nandini, R., T.R. Singh and G.N. Dhanapal.
Ramana, P. Vijayanand and J.T. Williams. 2011. Morphometric and molecular diversity
2006. Tamarind Tamarindus indica L. studies in Tamarind (Tamarindus indica Linn.)
Southampton Centre for Underutilised Crops, population. Journal of Plant Molecular Biology
Southampton, UK, 198 p. & Biotechnology 2(2): 26-33.

10. Fandohan, A.B, A.E. Assogbadjo, R.L. 18. Purushotham, B., P. Narayanaswamy, L.
Glele, B. Sinsin and P. Van Damme. 2010. Simon, S. Shyamalamma, H. Mahabaleshwar
Impact of habitat type on the conservation and B. Jayapalogwdu. 2008. Genetic
status of tamarind (Tamarindus indica L.) relationship between cultivars of areca
populations in the W National Park of Benin. nut (Areca catechu L.) determined by
Fruits 65(1): 1-9. RAPD. The Asian and Australasian Journal
of Plant Science and Biotechnology 2(1):
11. Gangaprasad, S., R. Rajkumar, R.L. 31-35.
Ravikumar, M.H. Savita, K. Krishnamurthy
and S. Hittalamani. 2013. Genetic diversity 19. Qian, W., S. Ge and D.Y. Hong. 2001. Genetic
analysis in tamarind (Tamarindus indica L.). variation within and among populations of a
Journal of Spices and Aromatic Crops 22(1): wild rice Oryza granulata from China detected
55-61. by RAPD and ISSR markers. Theoretical and
Applied Genetics 102: 440-449.
12. Graf, B., P. Rojas and M. Baldeón. 2016.
Discovering the Pharmacological Potential of 20. Reddy, M.P., N. Sarla and E.A. Siddiq. 2002.
Ecuadorian Market Plants using a Screens-to- Inter simple sequence repeat (ISSR)
Nature Participatory Approach. Journal of polymorphism and its application in plant
Biodiversity, Bioprospecting & Development breeding. Euphytica 128(1): 9-17.
3(1): 1-9.
21. Roldan-Ruiz, I., J. Dendauw, E. Van
13. Gunasena, H.P.M. and A. Hughes. 2000. Bockstaele, A. Depicker and M. De Loose.
Tamarind – Tamarindus indica L. Fruits for 2000. AFLP markers reveal high polymorphic
the Future 1. International Center for rates in ryegrasses (Lolium spp.). Molecular
Underutilised Crops (ICUC), Southampton, Breeding 6(2): 125-134.
UK. 169 p.
22. Sorkheh, K., B. Shiran, T.M. Gradziel, B.K.
14. Khanuja, S.P.S., A.K. Shasany, M.P. Darokar Epperson, P. Martínez-Gómez and E. Asadi.
and S. Kumar. 1999. Rapid isolation of DNA 2007. Amplified fragment length
from dry and fresh samples of plants producing polymorphism as a tool for molecular
large amounts of secondary metabolites and characterization of almond germplasm: genetic
essential oils. Plant Molecular Biology diversity among cultivated genotypes and
Reporter 17(1): 1-7. related wild species of almond, and its
relationships with agronomic traits. Euphytica
15. Kumar, M., V. Ponnuswami, C. Rajamanickam 156(3): 327-344.
and T.L. Preethi. 2015. Assessment of genetic
diversity in tamarind (Tamarindus indica L.) 23. Tapia-Pastrana, F., P. Mercador-Ruaro and S.
using random amplified polymorphic DNA Gómez-Acevedo. 2012. Contribución a la
markers. SAARC Journal of Agriculture 13(1): citogenética de Tamarindus indica
27-36. (Leguminosae: Caesalpinioideae). Acta
Botánica Mexicana 98: 99-110.
16. Leakey, R. R. B. and T. Page. 2006. The
24. Virk, P.S., J. Zhu, H.J. Newbury, G.J. Bryan,
‘ideotype concept’ and its application to the
M.T. Jackson and B.V. Ford-Lloyd. 2000.
selection of ‘AFTP’ cultivars. Forest Trees
Effectiveness of different classes of molecular
Livelihood 16: 5-16.
marker for classifying and revealing variation
162
Volumen 29 (2017) BIOAGRO N° 3
in rice (Oryza sativa) germplasm. Euphytica 26. Zetina, R., A.L. Del Ángel, E.N. Bezerra and
112: 275-284. M. Rebolledo 2012. Producción sustentable de
Tamarindo en el centro de Veracruz. Libro
25. Ward, J. H. 1963. Hierarchic grouping to Técnico No. 31. Editorial Alfa y Omega S.A.
optimize an objective function. Journal of Instituto Nacional de Investigaciones
American Statistical Association 58(301): 236- Forestales, Agrícolas y Pecuarias. Veracruz,
239. México. 174 p.