SELECTION OF PEACH GENOTYPES FOR BIOCHEMICAL FRUIT QUALITY CHARACTERIZATION1

ABSTRACT Peach is much appreciated by consumers and its popularity is mainly related with organoleptic characteristics. However, with emergence of concepts of functional foods (health promoters), there is high interest to study and to quantify the biochemical components of fruits. The aim of this work was to perform the biochemical characterization of peach genotypes, evaluating the genetic diversity and selecting those with desirable biochemical qualities for use as parents in future breeding programs. The experiment was carried out at the Laboratory of Plant Physiology UTFPR Campus of Dois Vizinhos, PR (Brazil), with fruits from 26 and 29 peach genotypes (Prunus persica) in the 2009/2010 and 2010/2011 crop years, respectively. The experimental design was entirely randomized, considering each genotype as treatment, using four replicates and four fruits per plot. Total and reducing sugars, total proteins, amino acids, total phenols, anthocyanins, flavonoids and phenylalanine ammonia-lyase enzyme activity (PAL) in fruits were evaluated. According to the results of two crop years, ‘Cascata 967’, ‘Conserva 985’, ‘Kampai’, ‘Tropic Snow’ and ‘Cascata 1055’ were selected as those with the highest levels of these compounds.


INTRODUCTION
Peach is much appreciated by consumers, and its popularity stems mainly from its pleasant taste, nutritional characteristics, appearance and practicality for fresh consumption (FISHER et al., 2016), being also widely used for canning, jellies and sweets.
In the last years, peach production has presented significant growth (TOMAZ et al., 2016). According to FAO (2008) data, Brazil is the 13 th largest world's peach producer.
The southern region of Brazil is responsible for more than half of the country's peach production (IBRAF, 2007), with potential for expansion, as there is a growing market for its consumption, since the population has searched for foods that in addition to nourishing, can offer compounds that provide health benefits (SANTANIN and AMAYA, 2007).
Thus, in the last few years, there has been a greater interest in studying and quantifying phenolic metabolites present in fruits and vegetables due to their health-promoting properties (GIL et al., 2002). The beneficial effects observed by the consumption of phytochemical compounds include the protection of individuals against the risks of genetic and environmental harm (ANGELIS, 2001).
Phenolic metabolites, including flavonoids, are composed of flavones, flavonols, catechins and anthocyanins, acting against free radicals, allergies, inflammations, ulcers, viruses and hepatotoxic tumors. They also protect against the oxidation of LDL-cholesterol (GERMAN and DILLARD, 2000).
Functional phenolic compounds present in fruits provide visual quality to them, making them more attractive to consumers, since they act as pigments. Another important function of these phytochemical compounds is that they are generally used by the plant in defense against diseases and pests.
As the phytochemical composition of fruits varies greatly among genotypes and are important quality attributes, it should be considered in peach breeding programs, since fruits with low sensory quality may present some desirable biochemical characteristics.
Faced with this growing productive potential and the need for fruits with additional qualities, it is necessary to identify and select genotypes with better biochemical characteristics for use as parents within breeding programs and / or as cultivars for use in commercial orchards.
For this, genetic divergence studies are important and necessary, which would provide parameters necessary for the identification of the most favorable parents to obtain segregating populations in hybridization programs, which allows the selection of superior genotypes and, consequently, genetically improved populations (COSTA et al., 2006).
The aim of this work was to perform the biochemical characterization of peach genotypes, evaluating the genetic diversity and selecting those with desirable biochemical qualities for use as parents in breeding programs.
SELECTION OF PEACH GENOTYPES FOR BIOCHEMICAL FRUIT... Management practices were performed according to general recommendations for the crop.
After harvested, fruits were transported to the laboratory for biochemical analyses [total and reducing sugars, total proteins, amino acids, total phenols, phenylalanine ammonia-lyase enzyme (PAL), anthocyanins and flavonoids]. Subsequently, samples were stored in freezer at -20 o C until evaluations were performed. Biochemical analyses were performed with epidermis tissues and fruit pulp.
For the quantification of total sugars, reducing sugars and proteins, samples were prepared, which were composed of approximately 1 g of each fruit. These samples were macerated in mortar along with 10 mL of 0.2 M phosphate buffer (pH 7.5) and then placed in eppendorf tubes and taken to refrigerated centrifugation at 4°C for 10 minutes at 12,000 rpm (14,700 xg), thus obtaining the extract supernatant.
The concentration of total sugars of peach tissues was determined by the phenol-sulfuric method described by Dubois et al. (1956). Samples were read at 490 nm in spectrophotometer (UV-SP2000-Spectrum). The concentration of total sugars was obtained by standard glucose curve (100 μg L -1 ).
For the quantification of reducing sugars of fruits, the dinitrosalicylate (DNS) method was used (MILLER, 1959). Samples were read at 540 nm in spectrophotometer model UV-SP2000-Spectrum. The concentration of reducing sugars of samples from each genotype was calculated as a function of a standard glucose curve.
For the determination of total proteins, the Bradford test (1976) was used. Reading was performed in spectrophotometer model UV-SP2000-Spectrum at 630 nm, with bovine serum albumin as standard.
The determination of total amino acids of peach genotypes was obtained by macerating samples with masses between 0.3 and 0.5 g in mortar with 5 mL of sulfosalicylic acid. Centrifugation was performed for 15 minutes at 6,000 rpm (7,350 xg) at 5°C in spectrophotometer (UV-SP2000-Spectrum). About 2 mL of supernatant extract were collected, adding 2 mL of acetic acid and 2 mL of acetic ninhydrin, leaving in a water bath for one hour at 100 ºC. Samples were then cooled on ice. Samples were read in spectrophotometer at 520 nm. The concentration of amino acids was estimated by a standard proline curve.
The quantification of total phenolic compounds of fruits was carried out in two stages, according to method adapted from Bieleski and Turner (1966). The first stage was composed of the extraction of total phenols from the pulp and epidermis of fruits through the maceration of approximately 1 g of fruit in a mortar, adding 4 mL of methanol, chloroform and water solution (MCA) at the ratio of 6:2.5: 1.5 (v / v), followed by centrifugation at 6.000 rpm (7,350 xg) for 20 min, and all supernatant was collected. Subsequently, a further extraction of the remaining residue was performed, adding 4 mL of MCA, centrifuging again at 6.000 rpm (7.350 xg) for 20 min and mixing the supernatant extract to the former, thereby obtaining the MCA extract. This extract was added of 1 mL of chloroform and 1.5 mL of distilled water, centrifuging at 6.000 rpm (7.350 xg) for 15 min to separate the phases. The second stage comprised the determination of total phenols performed by the adapted method of Jennings (1991). Quantification of phenols was performed through a standard curve using tyrosine and the result was expressed as mg total phenols.g -1 of fresh tissue.
For the quantification of the phenylalanine ammonia-lyase enzyme (PAL), the methodology adapted from Rodrigues et al. (2006) was used. The PAL activity was evaluated based on the difference in absorbance resulting from the conversion of phenylalanine into trans-cinnamic acid (HYODO et al., 1978).
For the determination of the anthocyanin and flavonoid content of fruits, the methodology described by Lees and Francis (1972) was used.
Biochemical data were submitted to analysis of variance and the means were compared by the Scott & Knott test (a = 0.05).
As a selection criterion, the selection of 20% of the evaluated genotypes was adopted, which presented the highest frequency of superiority in the evaluated characteristics in both crop years (PATERNIAN; MIRANDA FILHO, 1987). In order to calculate the frequency, genotypes were ranked in each of the 8 variables analyzed, from the 1 st to the 26 th or 29 th (2009/2010 crop year or 2010/2011 crop year, respectively), classified by means of the variables in descending order. The ranking position of each genotype in each of the variables was summed up. With the result, genotypes were classified in ascending order. For each evaluation cycle, genotypes ranked among the top 20% were

RESULTS AND DISCUSSION
The results of analyses performed for PAL enzyme in the 2009/2010 crop year were not significant among genotypes analyzed. However, all other biochemical variables (total and reducing sugars, total proteins, amino acids, total phenols, anthocyanins and flavonoids) analyzed in this crop year were significant (Table 1). On the other hand, in the second crop year (2010/2011), all variables evaluated were significant, demonstrating, that genotypes differed in biochemical characteristics ( Table 2).
The concentration of flavonoids and reducing sugars (2009/2010 crop year) and PAL enzyme activity (2010/2011 crop year) present in fruits, although being significant through the F test, indicated grouping of genotypes into a single group by the Scott & Knott test.
The contents of total sugars of fruits grouped genotypes in two groups in both crop years (Tables 1  and 2 (Table 1).
In the second crop year, 13 genotypes were grouped according to their higher content of total sugars (209.62 to 301.44 mg.g -1 ), being formed by  Table 2).
In the second crop year (2010/2011), all genotypes were grouped into a single group for reducing sugars (Table 2).
When comparing the results of both crop years for the content of total and reducing sugars, it was observed that there was a decrease in the concentrations of the second crop year in relation to the first one (Tables 2 and 1, respectively), which may possibly be related to the higher yield of the 2010/2011 crop year, since, with greater number of fruits, the number of drains per source increases. In addition, there was an increase in precipitation at harvest time in the second crop year (351.3 mm) in relation to the first one (190.6 mm), which may also be related to the decrease in sugar concentrations.
The sugar content of fruits is important because it is indicative of quality, and sweeter fruits are more accepted by consumers, especially for fresh consumption, emphasizing that peach breeding programs have emphasized the importance of flavor in the selection of new cultivars (CRISOSTO, CRISOSTO, 2005).
Analyzing the content of amino acids of fruits, two groups were formed in each crop year analyzed (2009/2010 and 2010/2011) (Tables 1 and  2  It was verified that fruits of the 2009/2010 crop year presented higher protein concentrations than those obtained in the 2010/2011 crop year. When analyzing peach genotypes for the protein content, the formation of two groups was observed in both crop years analyzed (Table 1 and (Table 2).
Thus, genotypes 'Conserva 985', 'Conserva 681', 'Conserva 871' and 'Âmbar' remained in both crop years grouped among the genotypes with the highest protein content, showing regularity in the production of this fruit component in subsequent crop years, conferring greater reliability for their selection considering this variable.
In the 2009/2010 crop year, the PAL activity did not present significant differences among genotypes. It may have occurred due to the higher protein content found in this crop year, when compared to the contents evaluated in the next one, since for the determination of the activity of this enzyme, the protein content is used.
However, in the second crop year analyzed, the F test detected significant differences among genotypes, but the Scott & Knott test grouped them into a single group, as previously mentioned.
Plants have structural and biochemical apparatus that compose their mechanism of defense against the action of biotic, abiotic and physical agents (AGRIOS, 1998). One of the key enzymes in the defense of plants is PAL, and when they are being attacked by some aggressor (pests or diseases) that trigger the defense process, this enzyme can be quantified in greater activity.
The PAL enzyme acts on the removal of the ammonia group from the aromatic amino acid phenylalanine, transforming it into trans-cinnamic acid, which is precursor of phenolic compounds. Thus, it is important to estimate the activity of this enzyme for the selection of genotypes to be used in breeding programs, since when it is found in greater activity, fruits with greater possibility of defenses against pathogens will be produced, being determinant in the resistance of plants to diseases.
The selection of genotypes with higher contents of phenolic compounds is important because these compounds present functional characteristics (ANJO, 2004), and, thus, meet the emerging trend DOI 10.1590/0100-29452017 669 Jaboticabal -SP K. C. FABIANE et al.
of the market, which seeks foods that can act in the prevention and cure of diseases (SENTANIN and AMAYA, 2007).
It is also possible that these compounds act by increasing the levels of resistance of fruits to brown rot, since these appear to be related to the higher accumulations of phenolic compounds in pulp and epidermis (BYRDE and WILLETS, 1977;GRADZIEL et al., 1998).
For the concentration of total phenols, genotypes were grouped into three distinct groups in both crop years ( Table 1).
The second crop year (2010/2011) also presented five genotypes individualized due to their higher contents of total phenols, but only genotypes 'Cascata 1070' and 'Conserva 681' were repeated, the other three genotypes grouped by the highest contents of total phenols in this crop year were 'Conserva 985', 'Cascata 587' and 'Conserva 1223', the latter being evaluated only in this second crop year. This group had contents of total phenols ranging from 0.57 to 0.88 mg.g -1 .
Peach and nectarine shells contain higher amounts of phenolic compounds, anthocyanins and flavonoids than pulp tissues; however, no significant differences were found between genotypes of white and yellow pulp (GIL et al., 2002), which allows a single selection criterion for both.
As for the content of flavonoids present in fruits of genotypes analyzed in the 2009/2010 crop year, there was no separation of genotypes in different groups (Table 1) (Table 2).
Anthocyanins are a group of phenolic compounds belonging to the class of flavonoids. This pigment, which reflects the red light, in addition to providing functional qualities to fruits due to its antioxidant potential, adds desirable characteristics, since fruits with high levels of anthocyanins present a very attractive visual aspect.
Concerning the content of anthocyanins, genotypes in both crop years were grouped into two groups (Tables 1 and 2 Table 2).
Considering that the process of genotype selection in the breeding program considers the preferences of the consumer market, the characteristic of fruit coloring becomes of paramount importance.
In the past, peach fruits were accepted by the consumer market for their visual and sensory characteristics, and today there is a tendency for the functional characteristics they present, which makes the evaluation of the biochemical characteristics important for breeding programs.
Based on the criteria adopted, five genotypes were selected in the 2009/2010 crop year because they presented the best biochemical characteristics, namely 'Kampai', 'Cascata 1055', 'Conserva 871', 'Rubimel' and 'Âmbar'. In the second crop year (2010/2011), six genotypes were selected based on their desired biochemical characteristics: 'Conserva 985', 'Cascata 967', 'Cascata 1065', Conserva 1216', 'Conserva 1223 However, as there was no convergence in the selection of genotypes in both crop years, the average of crop years was adopted as the selection method with the aim of selecting genotypes with lower oscillation for biochemical characteristics. Thus, the five best genotypes for the biochemical characteristics, based on the evaluations made during the two crop years, were 'Cascata 967', 'Conserva 985', 'Kampai', 'Tropic Snow' and 'Cascata 1055'. However, genotypes 'Cascata 1065', Conserva 1216' and 'Conserva 1223should be evaluated in subsequent crop years in order to verify if the superior biochemical characteristics presented in a single crop year are maintained during the next ones or if they were conditioned by the environment in the crop year analyzed.