Characterization of DNA polymorphisms in Caryocar brasiliense ( Camb . ) in populations with and without thorn at the endocarp by RAPD markers

Caryocar brasiliense (pequi), is one of the main species at the biome of the Brazilian savannah due to its use in culinary, popular medicine, industry in general, and iron and steel industry. At São José do Xingu (MT), a tree of C. brasiliense without thorn at the endocarp was found, which enables the improvement of C. brasiliense not only for consumption but also to the high appreciation it already has. To detect the existing differences between the pequi with and without the thorn at the endocarp, RADP markers were used. The generated polymorphisms were cloned and sequenced in order to identify the sequences that are responsible for the fenotypical alteration. It was observed that the pequi without thorn is genetically isolated from the other populations of pequi with thorn at the endocarp, proving that this characteristic is related to the genetic divergence of the species. Analysis in BLASTn evidenced the similarity of the Dof1 genes of Zea mays to its gene of phosphinotricin acetyl transferase. In the analysis of BLASTx, the similarity was verified to the proteins responsible for the deficiency in ferric reductase 4, and catalase.


INTRODUCTION
Caryocar brasiliense Camb.(Caryocaraceae), is one of the species that have been highlighted in the biome of the Brazilian savannah (Fig. 1).It is a tree of great importance in the North of Minas, where its fruit harvest is of great relevance for the feeding of the hinterland inhabitants, besides constituting an income source (Araujo 1995, Alencar 2000) (Figs.2a and 2b).
This species has been highlighting itself by the use of its fruits (mesocarp) in the human feeding in many regions through the preparations by cooking typical dishes and preparing spices, oils and sugar sweetened beverages (Barradas 1972, Corner 1976, Almeida and Silva 1994, Lopes et al. 2003, Santos et al. 2007).It is considered as a source of A and C vitamins, thiamine, pro-teins and mineral salts, presenting elevated carotenoid tenors.It is also used in oils extraction for the manufacturing of cosmetics (Almeida et al. 1998).In the popular medicine, it is used for the treatment of respiratory and ophtalmic problems, against bronchitis, colds, flus, and in tumors control (Almeida and Silva 1994).It is considered as an aphrodisiac; and its leaves are adstringent, also stimulating the bile production (Almeida and Silva 1994, Ribeiro 1996, Brandão et al. 2002, Kerr et al. 2007).The nut, besides being used in tannery, is tinctorial (Brandão et al. 2002).Its wood has excellent quality and high resistance (Almeida and Silva 1994), and it is a source of coal to iron and steel industries (Ribeiro 1996).
The exploration of the pequi tree is based on fruit harvesting, which characterizes an action of extraction and, as the trend has been to extract the most from it, its possible extinction in the future can be speculated about.The pequi region is spread over almost 2,000 municipalities, with about 40,000 collectors to sell its fruits or the "seed kernels" to buyers or resellers.The small pharmacies and shamans of the villages and cities of these regions require about 3,000 people to remove the thorns left by the seed kernels at the palate of unaware eaters.This characteristic is the main flaw that eliminates the pequi from being cultivated at home and of being considered as market fruit (R. Gribel, unpublished data, Kerr et al. 2007).
At São José do Xingu, Kerr et al. (2007) interestingly found a plant with an endocarp without seed kernels, producing about 500 fruits on 2004 and only 30 on 2005, and being the "seed kernels" fleshy, a little sweeter than the common ones and much better than the others due to the absence of thorns (Figs.3a and 3b).This fact became a hope for us to have, from 4 to 8 years from now, a distribution of pequi seedlings without thorn at the mesocarp.This "mutation" presents all the characteristics to turn pequi into a market fruit, as much for the poor populations as for the regular markets of national and foreign fruits.It improves pequi not only for a bigger consumption, but also for home harvesting.
Many polymorphisms among pequi populations by using micro-satellites were described.However, the RAPD (Williams et al. 1990) will be used in this paper once it discriminates well the existing polymorphisms among the individuals without thorn from the populations with thorn at the endocarp.
The aim of this paper was to identify standards of polymorphisms by means of RAPD markers, among pe-qui populations with the prickly endocarp and the individuals without thorns.

GENOMIC DNA EXTRACTION
The genomic DNA was extracted by the CTAB method with modifications according to Doyle and Doyle (1990) by using "bulks" of the 20 individuals of the populations with and without thorn, separately, with about 1g of vegetable tissue.
When homogeneizing the genetic material, the tubes containing the DNA were maintained at the freezer at -20 • C, being unfreezed only before the use.
The amount of extracted DNA was estimated in spectrophotometer, model GBC-UV/VIS911A (SONY), by reading of absorbance at 260nm.
Genomic DNA bands, separated by electrophoresis in agarose gel 1.0% (p/v), were used as indicators of integrity of the extracted DNA.The quantified and evaluated sample was diluted in milli-Q water for the concentration of work of 5 ng/μL, and maintained in freezer at -20 • C.

RAPD REACTION (RANDOM AMPLIFIED POLYMORPHISM DNA)
In order to evaluate the genetic diversity of the accesses, samples of DNA C. brasiliense leaves with and without thorn were amplified through PCR (Polymerase Chain Reaction) for the obtainment of RAPD markers.The reactions of amplification were made in a volume of 20 μL, with 2 μL of Taq 10X cap, 3 μL dNTP (4 mM), 1 μL MgCl2 (2.5 mM), 1 μL of polymerase Taq (1U/mL), 1 μL of primer (10 pmoles), 1 μL of genomic DNA (5 ng) and 11.8 μL of milli-Q water.77 random primers were used, out of which 53 were selected for presenting better standards of bands (Table I).
The amplifications were done in thermocycler (MJ Research, Inc., model PTC-100), programmed for 40 cycles of 1 minute at 94 • C, 1 minute at 35 • C, and 2 minutes at 72 • C.After the 40 cycles, an extension stage was done with 8 cycles of 20 seconds at 40 • C, 7 minutes at 72 • C, and reduction for 4 • C.After the amplification, in each sample, 2 μL of the bromophenol blue mixture (0.25% p/v) and glycerol (60% v/v) were added in water.15 μL of each sample was applied in agarose gel 1.5%, colored with Ethidium Bromide (0.5 μg/mL) and immersed in TBE 0.5X cap (Tris-Borate 90 mM, EDTA 1 mM).The electrophoretic separation was of 3 hours at 120 volts, after what the gels were visualized and photographed in Image Mater TM VDS Software (Pharmacia Bioscience), and the polymorphic bands were cut under ultraviolet light.

DATA ANALYSIS
The generated RAPD markers were converted into a matrix of binary data, from which the genetic distances among the different bulks were estimated.The matrix of genetic distance was done for the analysis of genetic diversity by means of the dendogram.
Duplicates of each gel were done and the bands in both gels were analyzed.
The matrix generated by the STATISTICA 4.5A (1993) program was used for the calculation of the ge-netic distances and the grouping analysis.The first was calculated by the method of Percentage of Disagreement, which is given by the formula: N AB /N T , where N AB is the total number of polymorphic bands among the compared genotypes, and N T is the total number of generated bands at the process.
The analysis of groups or clusters was made by the non-ponderated method of grouping to pairs using arithmetic averages (UPGMA -"Unweighted pair -group method using arithmetic average"), which aggrupates individuals according to their similarity.

PURIFICATION OF THE BANDS
The unique polymorphic bands visualized at the bulks were cut from the gel with a scalpel and put in a microtube of 1.5 ml.The purification occurred by the elution of the DNA in a microtube with three microholes and glass cotton at the bottom, and centrifugation for 5 minutes at 5,000 rpm.1/10 of the volume of ammonium acetate 7.5 M, 2 volumes of cold ethanol 100%, and 10 μL of glycogen (20 mg/mL) to the centrifuged liquid, for overnight precipitation at -20 • C. The solution was centrifuged for 15 minutes at 13,000 rpm, rejecting the supernatant and the precipitate washed with ethanol 70%, centrifuged again at 13,000 rpm during 5 minutes and re-suspended in 15 μL of milli-Q water.

CLONING OF THE AMPLIFIED POLYMORPHIC FRAGMENTS
The cloning of inserts of interest was done with the pGEM -T Easy Vector Systems I (Promega) kit, following the manufacturer's recommendations.

PURIFICATION OF THE PLASMID CONTAINING THE INSERT
The extraction of the DNA was done according to the Perfectprep Plasmid Mini (EPPENDORF) kit, following the manufacturer's protocol.

REACTION OF SEQUENCING IN MICROTUBES
Positive clones were sequenced in both directions in MegaBaceTM 1,000 (Molecular Dynamics, Amersham Life Sciences), using the dideoxi sequencing method with universal primers M-13 forward and M-13 reverse, according to the manufacturer's recommendations.

ANALYSIS OF THE IDENTITY IN GENETIC BANKS
An analysis of the identity of sequenced fragments was done using the GenBank data base.The BLAST program (Altschul et al. 1990) was used for the alignment of the sequences obtained from the pequi with others already deposited at the specific bank of plants.

RESULTS AND DISCUSSION
The DNA standard obtained for the samples was of good quality, which favors the RAPD reactions.
The generated bands at the RAPD enabled the analysis of the genetic variability among the five studied populations (Figs.4a and 4b).
The observed markers were as much for the presence of bands in all populations with thorn and their respective absence in populations without thorn, as for the presence of bands in population without thorn and absence in other populations with thorn at the endocarp.
The data obtained based on the 53 primers generated 326 fragments of RAPD, which were submitted to the analysis of cluster generating the dendogram of genetic divergence among the populations (Fig. 5).
The populations with smaller genetic divergences are from Minas and Tocantins, presenting a difference of 25% among their genotypes.Minas and Goiás presented a divergence of 30%, whereas these three regions with the Xingu region (population with thorn) presented a genetic divergence of about 37%.The pequi populations without thorn from the Xingu region was genetically isolated from the others, presenting a divergence of about 53%.
The molecular characterization of the genotypes is a tool of great importance for the programs of genetic improvement, once crossbreeds among individuals with a greater genetic diversity may contribute to the amplification in segregating populations, thus enabling a greater genetic gain (Mesmer et al. 1993).
Despite the high genetic similarity among the analyzed populations, it is suggested that the characteristic presence or absence of thorns at the endocarp is related to the genetic divergence of the populations.They may be caused by a punctual mutation or even by chromossomic differences between the individual without thorn and the population with thorn at the mesocarp.
The selection of bands for the purification was based on the polymorphisms detected on the five studied populations.The cloning and sequences of the fragments generated information about genetic differences (mutations) or conserved regions (monomorphisms) among individuals selected for the absence and presence of thorn at the endocarp.The sequences of the clones were deposited at the GenBank (Accesses numbers ET052997 until ET053017).
The analysis of BLAST of nucleotides (Table II) demonstrated that, in the population with thorn at the endocarp, the ET053010 clone was shown as similar to the Dof1 gene of Zea mays, whereas in the population without thorn, the ET053000 clone was similar to the phosphinotricin acetyl transferase gene of Z. mays.
The DOF1 gene encompasses a group of transcription factors in plants, with a sequence of well-conserved aminoacids (Yanagisawa 1996) containing the AAAG sequence that forms a dominium in zinc finger, which is essential for the bond to DNA, this sequence is present at the polymorphism that was found and related to this characteristic in pequi.The DOF proteins are responsible for processes of expression of the regulation of luminosity, regulation of answers to stress, answer to the expression of phytohormones and control of the transcriptional tissue-specific activities (Baumann et al. 1999, Chen et al. 1996, Cavalar et al. 2007, Kang and Singh 2000, Kisu et al. 1998, Mena et al. 1998, Vicente-Carbajosa et al. 1997, Yanagisawa 2000, Yanagisawa and Sheen 1998, Zhang et al. 1995).In the corn, the DOF1 protein activates the transcription of the responsible promoters by the luminosity signal, including C4 plants with promoter of the photosynthetic gene of the phosphoenolpyruvate carboxylase (Yanagisawa 1996, Yanagisawa andIzui 1993).The super expression of DOF1 in Arabidopsis thaliana induces the genes involved at the carbon metabolism and increases the fixation of nitrogen under limiting conditions (Yanagisawa et al. 2004).The phosphinotricin acetyl transferase gene, with which the pequi sequence without thorn demonstrated similarity, catalyses the phosphinotricin acetylation, an analogous of glutamate, a herbicide that exercises the phytotoxic effect on the plants, inhibiting the glutamine synthetase (Hérout et al. 2005, Hoagland 1999, Gill et al. 2001).

An Acad Bras Cienc
The reason for the DOF1 to appear in the populations with thorn and the phosphinotricin acetyl transferase to express in the pequi without thorn is unknown.The effect that these genes exercise in the pequi is unknown.The genomic markers developed in this paper will be able to contribute for an identification of the genetic causes of the absence of thorn in the endocarp in the restricted population of São José do Xingu.It is possible to search the metabolical way of these codified proteins in these genes, and their function in the obstruction of the appearance of thorn in the pequi.The analysis of BLAST of proteins (Table III) demonstrated similarities of the ET053002 and ET052006 clones to the CPFSTY protein (deficiency in ferric reductase 4) of Z. mays in the pequi without thorn.The ET053011 and ET053012 clones presented similarity to the catalase protein of Arabidopsis thaliana in the pequi with thorn.
The plants with deficiency in iron induce the plasmatic membrane to link to the ferric reductase, which transfers electrons from the intracellular NADH or NADPH for Fe 3+ in the rhizosphere (Buckhout et al. 1989, Sijmons et al. 1984).Fe 2+ ions are subsequently liberated and conducted to the cytoplasm through transport proteins, inducing the acidification of roots, stimulated by the deficiency in iron (Kochian 1991, Fox et al. 1996, Cohen et al. 1997).Plants that grow in soils poor in iron are selected for adaptative mechanisms that increase the availability of iron in the rhizosphere (Inskeep and Bloom 1987).The adaptative answers of the roots differ according to the species.Non-gramineous dicotiledones and monocotiledones induce morphological and physiological differentiations of roots (Marschner and An Acad Bras Cienc (2010) 82 (3) Romheld 1994, Schmidt 1999, Orturk et al. 2007).The increase in the capacity of the roots to reduce iron chelants by the induction of ferric reductase in the plasmatic membrane is another answer for the deficiency of iron; the reduction of Fe 3+ in Fe 2+ in the surface of roots is an obligatory process to the acquisition of iron in plants (Marschner andRomheld 1994, Robinson et al. 1999).The acidification of the rhizosphere contributes to the increase of solubility of iron available to plants (Marschner et al. 1986).Catalase is one of the antioxidant enzymes that degradate the oxygenated water (Scandalios et al. 1997).It performs an important role in the defense of the plant, aging and senescence.The catalase activity is influenced by acetylsalicylic acid and nitric oxide (Yang and Poovaiah 2002, Paterson et al. 1984).The balance between catalase and super-oxide dismutase is crucial in the determination of the level of O 2 and H 2 O 2 in the cell.The process of photo-oxidation during abiotic stress is common in plants, being able to cause cellular destruction and liberating reactive species of the oxygen such as the oxygenated water.(Foyer 1996).The super-reduction of the photosystem II (PSII) occurs when the assimilation of carbon is repressed during the environmental stress, and when the luminosity flow is elevated.In these states of excess of photons, the PSII becomes progressively reduced and leads the plant to an oxidative stress, producing free and possibly super-oxide oxygen (Krause 1994, Osmond andGrace 1995).In this sense, catalase acts with an antioxidant enzyme in the plants.The role of the protein responsible for the deficiency of ferric reductase and catalase is not known in Caryocar brasiliense.It can be suggested that the deficiency of iron in the roots leads the population without thorns to an environmental stress, repressing the production of thorns at the endocarp, whereas in the populations with thorn the presence of catalase prevents the oxidative stress and a repression of the thorns production does not occur.

CONCLUSIONS
From the analyzed primers of RAPD, it could be observed that there is a genetic divergence among the populations of pequi with and without thorn at the endocarp, and that the absence of thorn in the pequi might in part be due to differences in the metabolism route.However, biochemical and genetic studies are needed to confirm these mechanisms.

ACKNOWLEDGMENTS
This study is part of the Doctoral thesis of the first author and received financial aid of Universidade Federal de Uberlândia, Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES).

TABLE I Primers used for the PCR reactions for the obtainment of RAPD markers.
An Acad Bras Cienc (2010)82(3) 784 LUCIANA N. LONDE et al.