Vegetal material

Two cultivated species and five wild amaranth species were studied, represented by 13 national amaranth accessions, previously identified morphologically and coming from the INIFAP Genetic Resources program, Valley of Mexico Experimental Field located in Coatlinchan, Texcoco, México (Table 1). The collections and the sampling strategies that were followed for them were determined based on the mandate of the Germplasm Bank and the objectives of the collection. From each accession, 5 individuals were evaluated for the genetic diversity analysis.

DNA purification

The DNA extraction was done according to the protocol of ^{De la Cruz et al. (1997)}, with some modifications. Young leaves (0.3 g) were taken and macerated in liquid nitrogen. The macerate was transferred to a 1.5 mL micro tube (Eppendorf) containing 600 μL of extraction buffer (20 mL 1 M Tris-HCL, pH 8, 20 mL 0.5 M EDTA, pH 8, 20 mL 5 M NaCl; μL β-Mercaptoethanol, 40 mL of sodium dodecyl sulphate 20%) and incubated at 65 °C for 10 min in a Standard Heatblock of VWR^® Scientific Products, with occasional inversion of the tubes. Then, 200 μL of 5M potassium acetate was added, mixed by inversion and incubated for 30 min at 0 °C.

It was centrifuged at 15 000x g for 10 min at room temperature, and the supernatant was transferred to another tube containing 700 μL cold isopropanol (-20 °C). It was mixed by inversion and incubated at -20 °C for 30 min, then centrifuged for 5 min at 10 000 x g at room temperature. The supernatant was removed and the precipitate was dissolved in 200 μL of dilute solution (50 mM Tris-HCl, 10 mM EDTA-Na2, pH 8), stirred gently until the precipitate was well diluted. 2 μL of RNase A was added and incubated at 37 °C for 1 h in a Felisa^® incubator.

Then 20 μL of 3M sodium acetate plus 200 μL of isopropanol was added, mixed by inversion and allowed to precipitate at -20 °C for 2 h. It was centrifuged at 10 000 × g for 5 min at room temperature. The supernatant was removed and the precipitate was washed with 300 μL of 70% ethanol. The pellet was dried and dissolved in TE buffer (10 mM Tris-HCl, 1 mM EDTA-Na2, pH 8) and finally stored at 4 °C. The concentration of the DNA was determined using a Genesys 10 UV Scanning^® spectrophotometer (Thermo Scientific) and the quality thereof was observed by electrophoresis in a 0.8% agarose gel. The DNA was used in subsequent PCR reactions.

PCR reaction

The DNA concentration of each sample was adjusted to 10 ng∙μL^-1 and 10 primers previously selected were used for the amplification (Table 2). The amplification reactions were performed in a volume of 25 μL, which included 3.2 μL of sterile double distilled water, 10 μL of dNTPs (500 μM), 2.5 μL of 10X buffer (Tris-HCl 750 mM, pH 8.8; (NH4)2S04 200 mM; Tween 20 at 1% (v/v)); 2 μL MgCl2 (25 μM); 3 μL of initiator at a concentration of 10 pM; 0.3 μL of Taq DNA polymerase enzyme at a concentration of 5U μL^-1 and 4 μL of genomic DNA at a concentration of 10 ng μL^-1. The PCR reactions were performed in a Techne^® 28 TC-512 thermal cycler. The reaction conditions were: an initial denaturation step at 94 °C for 3 min, 30 cycles with denaturation at 94 °C for 1 min, alignment at 50 °C for 1 min, and extension at 72 °C for 2 min and finally a final extension cycle at 72 °C for 10 min.

The amplified fragments were separated by electrophoresis in agarose gels at a concentration of 1.2% (w/v) with TAE buffer (40 mM Tris-acetate, pH 7.6, 1 mM Na2 EDTA), from 1 h to 120 V. For the sagging of the samples, 12 μL of the PCR reaction product were used adding 1 μL of charge buffer (0.1% bromophenol blue, 0.1% xylen cyanol, 30% Ficoll and 1% SDS). For reference, the 1kb DNA ladder plus marker (GeneRuler) was used. The gels were stained with ethidium bromide (0.5 mg mL^-1) for 5 min and the excess dye was eliminated by rinsing in water for 5 min and documented using the Kodak^® High Perfomance Ultraviolet Transilluminator EDAS-290, under UV light.

Data analysis

The materials were compared based on similarities and differences in banding patterns, assigning a value of 1 to the presence of a band and 0 to the absence of it, assuming that bands of equal molecular weight are identical between populations and species, building a Basic Data Matrix (MBD). With the use of the POPGENE 32 program (^{Yeh et al., 1999}), the following parameters were estimated: percentage of polymorphic loci for the species and accessions considered as populations, the Nei genetic diversity coefficient, the Shannon information index, the GST parameter or coefficient of genetic differentiation between species and between populations, and the Nm or number of effective migrants, as well as the matrix of distances and genetic identities of ^{Nei (1973)}.

In addition, with the software NTSYS-pc version 2.2 (^{Rohlf, 2000}) the dendrogram generated in the POPGENE 32 program was visualized. Finally, the present variation among the species and populations was quantified by the molecular analysis of variance (Amova) using the GenAlEx 6.5 program (^{Peakall and Smouse, 2006}) with 999 permutations.

Genetic diversity in 13 amaranth populations

In Figure 2, the identity dendrogram is shown where the genetic relationships between thirteen amaranth populations corresponding to seven species are established and using the Nei coefficient. At an identity coefficient of 0.718 the formation of two groups is observed.

Figure 2 Dendrogram of genetic relationships between populations and amaranth species, constructed using the Nei identity coefficient. 

The group I was divided into four subgroups. The first sub-group included the cultivated species A. hypochondriacus and A. cruentus, where the most identical materials were those belonging to the species of A. cruentus (92CRU, 42 CRU), followed by the materials of A. hypochondriacus (89HYP, 27HYP). The most distant materials within this sub-group were A. hypochondriacus (89HYP) and A. cruentus (42CRU). The second sub-group was formed with accessions of A. hybridus and A. retroflexus. The third sub-group included materials from A. powellii (49POW) and A. retroflexus (38RETR), while the fourth sub-group only integrated a material from A. powellii (50POW).

The group II included A. palmerii dioic amaranths and the monoecious material with the presence of spines of A. spinosus, where the most similar accessions are those belonging to A. palmerii (27PALM and 35PALM), while the most distant were 27PALM of A. palmerii and 47SPIN of A. spinosus. The above results suggest that the cultivated amaranths A. hypochondriacus and A. cruentus are closely related to each other and then moderately related to the wild amaranths A. hybridus, A. retroflexus and A. powellii, in the order indicated, and less related to the species A. palmerii and A. spinosus.

With regard to wild amaranths, it was observed that the genetically most related species were A. hybridus with A. retroflexus and A. powellii, and A. palmerii with A. spinosus. Among these, the least related were A. palmerii and A. spinosus.

In Table 3, the indices of diversity calculated within 13 amaranth populations are shown. It is observed that the majority of the populations have low polymorphic loci percentages, with an average of 13.18% and a range that goes from 3.25% for the accession 26RETR, belonging to the species A. retroflexus to 30.52% of the accession 27PALM belonging to the species A. palmerii.

The Shannon indexes (I) and genetic diversity of Nei (H) (Table 3) were also low for the populations. The Shannon indexes had an average of 0.077, with a range of 0.0168 for the accession 26RETR of A. retroflexus and 0.1808 for the accession 27PALM of A. palmerii, while the index of diversity of Nei (H) showed an average of 0.054, with a range of 0.011 for the accession 26RETR of A. retroflexus and 0.124 for the accession 27PALM of A. palmerii.

For the number of alleles per locus (NA) and the effective number of alleles per locus (NE) the range of values between the populations was between 1 032 to 1 305 and 1 019 to 1 225, respectively. The analysis of genetic diversity among the thirteen populations showed that the percentage of polymorphic loci (LP %) was 99.35%. The coefficient of genetic differentiation between populations was high (GST= 0.82), and indicates that approximately 82% of the variation detected can be attributed to genetic differences between populations. The rest (18%) represents genetic diversity within the populations. Based on the coefficient of total genetic differentiation between populations (GST) the estimated gene flow level (Nm) was 0.11. This indicates that there is less than one migrant individual per generation among the populations, which also explains the high level of differentiation detected among them.

Genetic diversity in seven amaranth species

The genetically closest species were the wild A. hybridus and A. retroflexus with a coefficient of 0.93, followed by A. powellii with a coefficient of 0.87. The cultivated species were genetically close to each other, although related to the wild species mentioned above. A. hypochondriacus was related to A. cruentus to a coefficient of genetic identity of Nei equal to 0.86. In addition, the related A. palmerii and A. spinosus wild species were far removed from the cultivated amaranths A. hypochondriacus and A. cruentus, as well as the wild amaranths A. hybridus, A. retroflexus and A. powellii.

With regard to the genetic diversity indices calculated within species, Table 4 shows that the percentage of polymorphic loci was quite low for the seven species evaluated, with an average of 24.02% and a range of values that goes of 7.79% for A. spinosus and 44.16% for A. palmerii.

The Shannon (I) and genetic diversity indexes of Nei (H) were also low (Table 4), the Shannon indexes showed an average of 0.132, with a range of values ranging from 0.042 for the species A. spinosus and of 0.235 for the species A. palmerii, while the index of diversity of Nei (H) showed an average of 0.089, with a range of 0.029 and 0.157.

For the number of alleles per locus (NA) and the effective number of alleles per locus (NE), the range of values for the evaluated species was between 1 078 to 1 442 and 1 051 to 1.27, respectively.

The analysis of genetic diversity among the species indicated a percentage of polymorphic loci (% LP) equal to 99.35%. The coefficient of genetic differentiation between the species was high (GST= 0.7), and indicates that approximately 70% of the genetic variation detected can be attributed to genetic differences between the species. The rest (30%) represents genetic diversity within the species. Based on the coefficient of total genetic differentiation between species (GST) the estimated gene flow level (Nm) was 0.21. This suggests that there is less than one individual migrant per generation among species, which also explains the high level of differentiation detected among them.

The genetically closest species were A. hybridus and A. retroflexus with an identity value of 0.927, while the most distant species were A. spinosus and A. hybridus with a value of 0.534.

Analysis of molecular variance (Amova)

The results of the molecular variance analysis (Amova) indicated significant genetic differences (p≤ 0.001) among the populations studied (Table 5). A genetic variability value equal to 79% was detected among the populations, while the remaining 21% is within the populations. When considering species, it was found that between them 68% of the total genetic variation is present and only 32% is within the species.

When comparing the cultivated materials against the wild ones it was found that the genetic variation present within both types of materials is very low with a value of 21%, while the variation among them was very high (79%), which indicates the existence of genetic variation susceptible to use in breeding programs for cultivated amaranths.

Also, it was found that genetic variation among cultivated amaranth populations was 72%, while only 28% is within them, which also demonstrates the high degree of genetic differentiation present between A. hypochondriacus and A. cruentus

When only wild materials were compared, 66% of the genetic variation was found among the populations and only 34% within them. And finally, when making the comparison between the wild materials dioecious against the monoecious, the results indicated that the genetic variation present between the populations is 46% and the rest (54%) is within the populations.