African Journal of Agricultural Research Vol. 6(28), pp. 6126-6131, 26 November, 2011 Available online at http://www.academicjournals.org/AJAR DOI: 10.5897/AJAR11.1237 ISSN 1991-637X ©2011 Academic Journals
Full Length Research Paper
Molecular study of a diallel chilli with amplified fragment length polymorphism (AFLP) markers Guillermo Castañon-Najera1*, Netzahualcóyotl Mayek-Pérez2, Régulo Ruiz-Salazar2 and Araceli C. Garcia3 1
División Académica de Ciencias Biológicas (DACBiológicas), Universidad Juárez Autónoma de Tabasco (UJAT), Road Villahermosa-Cárdenas, Km 0.5 Entronque Bosques de Saloya, Villahermosa, Centro, Tab., México. 2 Centro de Biotecnología Genómica (CBG), Instituto Politécnico Nacional (IPN), Boulevard del Maestro S/N, Col. Narciso Mendoza, Reynosa, Tams, México. 3 George Washington Carver School, 1341 5th street Yuma, Az 85364, USA. Accepted 20 October, 2011
Molecular markers, such as Amplified fragment length polymorphisms (AFLP) were used to study genetic diversity in four chilli varieties. Two wild chilli’s (Amashito and Ojo de cangrejo: Capsicum annuum var. glabriusculum) and two growers (Jalapeño and Habanero: C. annuum and Capsicum chinense Jacq.), were used as parents in a diallel cross, with no reciprocal crosses, to produce six F1 hybrid combinations. Crosses between parents were made by hand in the greenhouse at the División Académica de Ciencias Biológicas (DACBiológicas) of the Universidad Juárez Autónoma de Tabasco (UJAT), in Villahermosa, Tabasco, Mexico. The four AFLP markers used amplified 256 bands, of which 62 (24.2%) bands were polymorphic. Dice’s similarity index was used to determine the genetic distances between parents and their offspring. Cluster analyses based on genetic distances separated the parents and crosses belonging of C. annuum from those of Capsicum chinense, confirming the genomic differences between the two species. The cophenetic correlation coefficient (r) value between the similarity matrix and the original matrix to determine if the original data were well represented in the cluster analysis, under 1000 permutations, was 0.87, which is within the range of good. The results of this research are useful for use in a breeding program chilli. It has been shown that the more genetic distance is between the potential parents, the progeny of both present greater heterosis. Key words: Genetic distances, PCR, Capsicum, diallel crosses, Chilli species.
INTRODUCTION Chilli (Capsicum spp.) is a major vegetable crop considering the high economic income and the employment that it generates in México. Although chilli has a great social importance in Mexico after the year 2000 during the hybrid production of chilli for specific important traits, such as shelf-life started (Martínez et al., 2005), together with studies of heterosis effects on chilli production and quality (Pérez-Grajales et al., 2009), as
*Corresponding author. E-mail:
[email protected] or
[email protected]. Tel/Fax: 055-993-3-5443-08.
well as the contents of capsaicinoids, vitamin C during fruit development (Cruz-Perez et al., 2007). Additionally, researches on the estimation of genetic distances between improving lines of green chilli using RAPD markers and their relationship with heterosis were performed (García et al., 2002). However, it is necessary to identify the heterotic groups of chilli that may exist between and within species, through the application of molecular techniques. According to Wilches (2004), and thanks to advances in molecular biology, it is possible to locate and incorporate only the gene or genes that have the desirable novelty trait, saving a lot of time in the production of better varieties, and considering that, evaluation of hybrids for heterosis or combining abilities in the field is expensive and time-consuming (Sant et al.,
Castañon-Najera et al.
6127
Table 1. Sequences of oligonucleotide adapters and primers used in the characterisation of chilli parents and crosses with AFLP markers.
Adapters
EcoRI
5’-CTCGTAGACTGCGTACC-3’ 3’-CTGACGCATGGTTAA-5’
MseI
5’-GACGATGAGTCCTGAG-3’ 3’-TACTCAGGACTCAT-5’
EcoRI MseI
5‘-AGACTGCGTACCAATTC/A-3’ + A 5'-GACGATGAGTCCTGAGTAA/A-3' + C
M1
EcoRI MseI
5’-AGACTGCGTACCAATTC-3’ + AAG 5'-GACGATGAGTCCTGAGTAA-3' + CAG
M2
EcoRI MseI
5’-AGACTGCGTACCAATTC-3’ + ACG 5’-GACGATGAGTCCTGAGTAA-3’+CAG
M3
EcoRI MseI
5’-AGACTGCGTACCAATTC-3’ + AAG 5’-GACGATGAGTCCTGAGTAA-3’ + CAA
M4
EcoRI MseI
5’-AGACTGCGTACCAATTC-3’ + ACG 5’-GACGATGAGTCCTGAGTAA-3’ + CAA
M5
EcoRI MseI
5’-AGACTGCGTACCAATTC-3’ + ACT 5’-GACGATGAGTCCTGAGTAA-3’ + CTG
M6
EcoRI MseI
5’-AGACTGCGTACCAATTC-3’ + AGG 5’-GACGATGAGTCCTGAGTAA-3’ + CTG
Pre-selective amplification
Selective amplification
1999). Several studies on heterosis and general combinatory ability have used molecular techniques using anther culturing in maize (Xin et al., 2010), sunflower (Gvozdenovic et al., 2009), wheat (Sun et al., 1997), tobacco (Conner et al., 1998), rice (Quimio and Zapata, 1990) and chilli (Sant et al., 1999). As described Geleta et al. (2004), advances in genome research have generated interest in predicting hybrid performance using molecular markers in crop breeding programmes. The amplified fragment length polymorphisms (AFLPs) is one of the fastest and cheapest molecular marker methodologies based on the selective amplification of fragments of DNA that were fragmented with restriction enzymes using the PCR (polymerase chain reaction) (Vos et al., 1995). AFLP marker segregates as dominant marker. The aim of this work was to determine whether AFLP marker analysis is useful to estimate genetic distances between parents and crosses in a diallel set.
MATERIALS AND METHODS Plant material The chilli varieties included in the study were (1) Jalapeño, (2) Amashito, (3) Ojo de cangrejo and (4) Habanero. Varieties 1 and 4 are cultivated and belong to Caspicum annuum and Capsicum chinense Jacq., while varieties 2 and 3 are wild and belong to C.
annuum var. glabriusculum. They were used as parents of diallel crosses without reciprocal crosses, and were done by hand in the greenhouse at the División Académica de Ciencias Biológicas (DACBiológicas), Universidad Juárez Autónoma de Tabasco (UJAT), in Villahermosa, Tabasco, Mexico during Spring 2009. DACBiológicas is located at a mean altitude above sea level of 10 0 0 m, at 19 32'03'' N, 99 07'46'' W, in a hot humid Am(f)''(i')g climate (García, 1970), with 2500-3000 mm of rainfall and an average annual temperature of 24-25°C. The plant parents were s elected considering different traits of the fruit and plant morphology.
AFLP reaction Young leaves were collected from five plants of each material (parents and crosses), and three 0.5 g individual tissue samples were obtained from the leaves. The samples were placed in liquid nitrogen (-196°C) in order to extract the genomic DNA ( Dellaporta et al., 1983). The concentration of gDNA extracted was quantified in a Jenway 6305® UV/vis spectrophotometer at an absorbance of 260 nm, and its quality was verified after electrophoresis in 5 µg of DNA and 1% agarose gel (p v-1) with TAE buffer (40 mM trisacetate, pH 7.6, 1 mM Na2.EDTA), for 2 h at 80 V and stained with ethidium bromide (0.5 mg mL-1). AFLP (amplified fragment length polymorphism) type molecular markers were used for the genetic analyses of parents and their crosses. Six combinations of molecular markers (MM) were evaluated (Table 1) in a preliminary test. The combinations M1 and M3 did not give a good amplification, for which reason only the other four combinations (M2, M4, M5, M6) were used. The AFLP amplified fragments were separated by electrophoresis in 6.5% polyacrylamide gel (Vos et al., 1995), using the commercial IRDye Flourescent AFLP® Kit for large plant
6128
Afr. J. Agric. Res.
Table 2. Combination of markers, total number of bands and percentage of polymorphic bands detected in chilli parents and crosses.
Combination of markers EcoRI + ACG / MseI + CAG EcoRI + ACG / MseI + CAA EcoRI + ACT / MseI + CTG EcoRI + AGG / MseI + CTG Total Average
Code M2 M4 M5 M6
Total 56 77 60 63 256 64
Genome Analysis (Li-COR®). The separation of the amplified fragments was made in a semiautomatic sequencing IR system (model 4200-029; LICOR®; Lincoln, NE, USA). For the AFLP analysis, the clear bands were visually scored as either present (1) or absent (0) for parents and crosses, resulting in one binary data matrix. Each marker was treated as an independent unit character.
Statistical analyses Each AFLP fragment in the binary matrix was considered as a locus, with each band corresponding to an allele and the absence of a band corresponding to an alternative allele. The determination of genetic distances was carried out according to Dice’s method (Nei and Li, 1971) and was estimated as:
2a ( 2a + b + c ) where a=total number of bands common in parents i and j (1,1), b=total number of bands common in parent i but not in parent j (1,0), c= total number of bands common in parent j but not in parent i (0,1), d= total number of bands absent in both parents i and j (0,0) (Anónimo, 2003, Cornide et al., 2002). Dice’s coefficient counts the percentage of shared bands between two individuals, and is comparable to Jaccard´s coefficient. However, Dice´s coefficient gives more importance where bands are present in both, laying emphasis on the similarity between individuals rather than on their dissimilarity (Bonin, 2007). This method considers that the absence of one band has less biological importance, and the coefficient is thus significant with respect to the similarity of the DNA. Dice’s method is the only coefficient that has biological significance as it expresses the probability that a band in one individual may be found in another individual, and it considers the relationship of the number of bands that coincide in two individuals with respect to the total number of bands (average of bands in an individual). The cluster analysis dendrogram was obtained with the unweighted pair group method, with an arithmetic average (UPGMA) based on Nei’s (Dice) genetic distance values, to compare the relationship between parents and crosses. The confidence limits placed on the dendrogram using the bootstrapping programme with 1000 bootstrap replicates were generated with the Freetree and Treeview programmes (Page, 1996). The cophenetic correlation coefficient (r) that compares the similarity matrix and the original matrix to determine whether the original data were well represented in the cluster analysis, under 1000 permutations, was calculated with the aid of the computer programme NTSYSpc 2.02 (Nuñez, 2011).
Number of bands Polymorphic Monomorphic 13 43 22 55 19 41 8 55 62 194 15.5 48.5
Polymorphism (%) 23.21 28.57 31.67 12.70
RESULTS The four markers used in this study amplified 256 bands (Table 2), of which 62 (24.22%) were polymorphic and 194 were monomorphic (75.78%). The primer EcoRI + ACG/Msel + CAA amplified the greater number of fragments (77), while the primer EcoRI + ACG/Msel + CAG amplified less fragments (56). The primer EcoRI + AGG/Msel + CTG generated the smaller polymorphism (12.70%) and the EcoRI + ACT/Msel + CTG generated the greater polymorphism (31.67%). These markers generated a similar total number of bands (60 to 63 bands each). Genetic distance values were calculated with Nei’s similarity index. The values estimated were sufficiently large. The range (Table 3) of data for parents and offspring was the smallest at 0.563, the one presented by parent 1 with a 2×4 cross, and the highest value was 0.883, which corresponded to parent 4 and the combination 1×4, indicating that the parents and combinations genome was not well preserved. The genetic distances based on molecular markers, determined for the parents (data not presented), indicated that the Amashito and Ojo de cangrejo obtained the greatest distance value (0.81), making these the more similar genotypes, and the Habanero the most distant of the four parents (0.63). The dendrogram (Figure 1) for the genetic relationship between parents and their crosses formed four groups, each with different subgroups. Group 1 was formed by the crosses 1×2, 1×3 and parent 1, group 2 by parents 2 and 3 and their respective cross (2×3), group 3 by parent 4 (Habanero) and the 2×4 and 3×4 crosses, and an independent group was formed by the 1×4 cross (Jalapeño × Habanero). The aforementioned may be observed in Figure 2 where the parents that belong to annuum were grouped separately from the Habanero, which is chinense Jacq. Figure 2 also shows that the four varieties of parents clustered into one major group with two varieties (Amashito and Ojo de cangrejo) of similar small fruit size, and left the Jalapeño and Habanero apart.
Castañon-Najera et al.
6129
Table 3. Genetic distances (GD) between crosses and parents estimated with Dice´s coefficient.
1 1 12 13 14 2 23 24 3 35 4
12 0.731
13 0.605 0.643
14 0.616 0.709 0.712
2 0.760 0.787 0.658 0.684
23 0.581 0.632 0.654 0.774 0.584
24 0.563 0.652 0.639 0.772 0.625 0.734
3 0.674 0.814 0.716 0.731 0.731 0.629 0.678
34 0.635 0.623 0.570 0.681 0.651 0.624 0.733 0.606
4 0.643 0.715 0.667 0.883 0.684 0.815 0.796 0.710 0.615
1 (Jalapeño), 1 × 2 (Jalapeño x Amashito), 1 × 3 (Jalapeño × Ojo de cangrejo), 1 × 4 (Jalapeño × Habanero), 2 (Amashito), 2 × 3 (Amashito × Ojo de cangrejo), 2 × 4 (Amashito × Habanero), 3 (Ojo de cangrejo), 3 × 4 (Ojo de cangrejo × Habanero), 4 (Habanero).
12 100 13
60
1 40
2 100 23
67 76
3 24 93
100
34
59
4 14 0
Figure 1. UPGMA dendrogram of genetic distance relationships based on AFLP markers between chilli parents and F1 crosses: 1 (Jalapeño), 2 (Amashito), 3 (Ojo de cangrejo), 4 (Habanero), 1×2 (Jalapeño x Amashito), 1×3 (Jalapeño × Ojo de cangrejo), 1×4 (Jalapeño × Habanero), 2× 3 (Amashito x Ojo de cangrejo), 2× 4 (Amashito × Habanero), 3×4 (Ojo de cangrejo x Habanero).
In general, the viability given to each group by the robust analysis is accepted (60% for group 1, 67% for group 2, 59% for group 3). The cophenetic correlation value (r) between the similarity matrix and the original matrix data was r=0.87, which is “good” according to Nuñez (2011) who established the value r as: 0.9≤r very good, 0.8≤r
