Determination of organic acids and carbohydrates in ‘Salustiana’ orange fruit from different rootstocks

https://doi.org/10.1590/1981-6723.32918. Abstract This study aimed to determine carbohydrates and organic acids in fruit from ‘Salustiana’ orange tree [ Citrus sinensis (L.) Osb.] grafted onto different rootstocks using the HPLC technique, as well as to evaluate their physicochemical properties. As rootstocks, we tested ‘Rangpur’ lime, ‘Cleopatra’ mandarin, ‘Sunki’ mandarin, ‘Swingle’ citrumelo, and ‘C-13’ citrange. Fully mature fruit was first characterized according to their physicochemical parameters as fruit mass, longitudinal and equatorial diameters, peel color, juice yield, soluble solids, titratable acidity, and ratio. Then, freshly squeezed juice was used to determine and to quantify organic acids and carbohydrates by the HPLC system. All analyses were performed in triplicate complete randomized with five treatments (rootstocks). Fruit from all evaluated rootstocks presented adequate physicochemical attributes, meeting the quality standards of the citrus industry. ‘C-13’ citrange induced in a production of large-sized fruit whereas ‘Rangpur’ lime promoted higher soluble solids content in its juice. Significant differences were not observed between the treatments with regards to organic acids and carbohydrates concentrations. Citric and ascorbic acids were identified and quantified in the juice samples. The sugars identified were sucrose, glucose, and fructose. Organic acids and carbohydrates concentrations are consistent with those reported in the literature for citrus juice, providing information about nutrition and quality of ‘Salustiana’ sweet orange produced onto different


Introduction
Brazil leads the worldwide production of oranges, which ensured an annual production of 17.3 million tons in 2016, responsible for 25% of the total production (Food and Agriculture Organization Corporate Statistical Database, 2018). About 80% of its production goes to the industry, mainly to process frozen concentrated orange juice (FCOJ) for exportation and the remaining (20%) is destined to supply the Brazilian fresh market. For this reason, the production of orange cultivars intended for the fresh and processing purpose has increased, because of the advantage of dual decision making and flexibility to the citrus growers.
By this context, the sweet orange cv. Salustiana [Citrus sinensis (L.) Osb.] has excellent performance for both purposes, but mainly for in natura consumption because its fruit is seedless, very taste (Castle & Baldwin, 2011) and has suitable physicochemical characteristics according to the fruit quality standards (Organisation for Economic Co-operation and Development, 2010; Companhia de Entrepostos e Armazéns Gerais de São Paulo, 2011). Nevertheless, physicochemical quality may appear divergent in their values according to the rootstock used for citrus cultivation.
Some authors have reported this influence on fruit quality in several citrus cultivars (Cantuarias-Avilés et al., 2010;Gonzatto et al., 2011;Legua et al., 2011;Silva et al., 2013;Tazima et al., 2014;Bacar et al., 2017). Similarly, Castle (2010) confirms that rootstocks are highly related to many fruit attributes such as fruit size, and juice quality in which sugars and acids are affected, being a major factor for growers of fresh and processing fruit. In this way, it is important to assay new alternatives of rootstocks to diversify citrus groves and verify their influence on fruit quality.
As far as the physicochemical quality of citrus is important for commercialization and consumption, the description of their compounds profile is required, because the general composition and antioxidant properties of foodstuff are critical to determine consumer acceptance and preference. Carbohydrates and organic acids are among the major compounds present in citrus fruit, and their nature and concentration largely affect taste characteristic and organoleptic quality (Kelebek et al., 2009). organoleptic properties and stability of fruit juices, being an important index to determine fruit authenticity (Kelebek et al., 2009;Scherer et al., 2012). However, their profile and concentration in citrus fruit are related to sugar content and are dependent on citrus varieties, soil, climate, and stress conditions (Jones, 1998;Lado et al., 2014). In general, citric, ascorbic, and malic acids are the most abundant organic acids found in sweet orange. According to Scherer et al. (2012) citric and malic acids are commonly used as acidulants to enhance beverages and ascorbic acid as an antioxidant.
By this context, this study aimed to determine carbohydrates and organic acids in fruit from 'Salustiana' orange tree grafted onto different rootstocks using HPLC techniques, as well as, to analyze other physicochemical characteristics.

Physicochemical analysis
The physical characteristics of fruit were evaluated, determining the fruit mass (g) with a semi-analytic scale (M505 Bel, Piracicaba, Brazil). A digital caliper (ABS Mitutoyo, Kawasaki, Japan) measured the longitudinal diameter (mm) and equatorial diameter (mm). Peel color was obtained by the Hue angle (h o ) means for each sample, using mean values for red-green (a*) and yellow-blue (b*) measured with a digital colorimeter (CR-10 Konica Minolta, Tokyo, Japan), considering three measure points in the equatorial surface of each fruit. Hue angle was determined according to McGuire (1992). Then, the juice yield was obtained by the ratio between juice mass and fruit mass (Tazima et al., 2014).
Fruit samples were also characterized according to their chemical profile. A digital refractometer with automatic temperature compensation (DR301-95 Krüss Optronic, Hamburg, Germany) was used to determine soluble solids (SS) contents, the result was expressed as °Brix. Titratable acidity (TA) was assessed by titration with a standard 0.1 N NaOH solution in a semi-automatic titrator, the results were expressed in Braz percentage of citric acid (Gaithersburg & Horwitz, 1990). The ratio was determined by the relation between soluble solids (SS) and titratable acidity (TA).

Chromatographic analysis
For HPLC analysis, 5 ml of freshly squeezed juice mixed with 15 ml of ultra-pure water (Milli-Q) were centrifuged at 50000 rpm in a centrifuge (CP100WX Hitachi, Tokyo, Japan) for 10 min, the supernatant was then filtered through 0.22 μm pore size membrane filters (Sartorius Stedim Biotech, Goettingen, Germany).
The determination of organic acids and carbohydrates were performed by HPLC instrumental system (LC-20 A Shimadzu, Kyoto, Japan) equipped with a high-pressure pump system (LC-20AT), the automatic injector (SIL-20AC HT), refractive index detector (RID-10A), photodiode array (SPD-M20A), column oven (CTO-20A) and control module (CBM-20A). Data collection and integration of chromatographic peaks were computed processed by the LC Solutions software (Shimadzu, Kyoto, Japan).

Organic acids
The centrifuged and filtered samples were vialed and injected into the HPLC system. The injection volume used was 20 μL. Organic acids were analyzed onto a Phenomenex C18 column (250 × 4.6 mm, particle size 5 μm) (Phenomenex, Torrance, USA). The mobile phase consisted of 25 mM sodium phosphate buffer solution, and pH adjusted to 2.4 with phosphoric acid at a flow of 1.0 mL min -1 . The column was kept at 30 o C with refractive index detection (RID -10A).

Carbohydrates
Samples of 2 mL of orange juice were used to determine carbohydrates, previously centrifuged, and filtered. Solid-phase extraction was performed passing samples through a Sep-Pak C18 500 mg cartridge after activated with 2 mL of methanol and 4 mL of ultra-pure water (Milli-Q). The eluate was collected and filtered through a 0.22 μm membrane filter (Sartorius Stedim Biotech, Goettingen, Germany). The HPLC system received 20 μm of the sample. Sugars were analyzed using a Shim-pack CLC-NH2 (M) column (250 × 4.6 mm, particle size 5 μm) (Shimadzu, Kyoto, Japan) kept at 30 °C (CTO-10AS VP). The used mobile phase was a mixture of acetonitrile: water 75:25 (v/v) at a flow of 1.0 mL min -1 , according to Pauli et al. (2011).

Statistical analysis
All analyses were performed in triplicate in a complete randomized design with five treatments (fruit from different rootstocks). Data were processed by the R 3.4.1 version (The R Foundation for Statistical Computing, Vienna) and assayed according to the experiment design, using analysis of variance followed by Tukey's test (p ≤ 0.05).

Physicochemical analysis
Fruit from 'Salustiana' sweet orange was first characterized according to its physical characteristics (Table 1 and Figure 1). The results showed that 'C-13' citrange rootstock induced production of large-sized fruit, presenting fruit mass of 222 g, longitudinal and equatorial diameters of 72 and 77 mm, respectively. These values were significantly greater than those noticed in oranges from 'Cleopatra' mandarin, which resulted in fruit of small caliper with 163 g per fruit and longitudinal and equatorial diameter of 65 and 68 mm, respectively.  128 to 139 g with diameters between 64 and 68 mm, depending on the crop season. On the other hand, our findings were related to those found by Santos et al. (2010), characterizing eight seedless varieties of citrus produced in Viçosa (Brazil), in which 'Salustiana' orange fruit ranked 160 g with diameters between 68 and 70 mm. This was similar to other authors who assessed the influence of rootstocks under different citrus cultivars as 'Valencia' orange (Auler et al., 2008), 'Okitsu' Satsuma mandarin (Tazima et al., 2013), and 'Jaffa' orange (Bacar et al., 2017). As observed in the present study, larger and heavier fruit may present greater demand in the market since Brazilian consumers have a preference for large size fruit (Pio et al., 2001).
Concerning peel color, the tested rootstocks did not influence this aspect. Flavedo color was based on the Hue angle that demonstrates its changes, values closest to 90 h o indicates a high yellowish content in the peel, as observed for this cultivar which ranged from 81.6 to 84.9 h o according to the rootstock. This color change occurs due to the chlorophyll degradation and synthesis of carotenoids, responsible for the yellow or orange color of the flavedo (Ladaniya, 2008;Santos et al., 2010). In general, all treatments presented fruit with adequate peel color based on the international and domestic standards of table citrus (Organisation for Economic Co-operation and Development, 2010; Companhia de Entrepostos e Armazéns Gerais de São Paulo, 2011). Table 2 shows the juice yield and chemical composition of 'Salustiana' orange juice. We observed a significant difference between juice yield and soluble solids contents parameters. 'Swingle' citrumelo and 'Sunki' mandarin rootstocks induced higher juice content for this cultivar, as reported by Santos et al. (2010) in fruit from 'Salustiana' orange and 'Swingle' citrumelo combination. In addition, the evaluated treatments were superior in juice yield than those observed by Bini et al. (2009), 39%; and also, higher than the minimum juice content required by the Organisation for Economic Co-operation and Development (2010) and the Companhia de Entrepostos e Armazéns Gerais de São Paulo (2011), 35 and 45% respectively.  Stenzel et al. (2006), who reported an early maturation in fruit from 'Folha Murcha' sweet orange tree onto 'Rangpur' lime, comparing to other rootstocks. The range between our treatments was 9 to 11 o Brix, indicating a suitable quality for harvest since fruit should present at least 9 o Brix as soluble solids content in its juice (Pereira et al., 2006). As soluble solids content, titratable acidity can also be used to predict the maturation degree of fruit. These Braz indices were similar among the treatments ranging from 0.68 to 0.77 and considered adequate according to Pereira et al. (2006). Because fully mature fruit of sweet oranges must present titratable acidity, expressed as citric acid, between 0.5 and 1.0%. The ratio (soluble solids:acidity), which indicates the fruit maturation index, was similar for all tested treatments (rootstocks) varying from 13.1 to 14.7. These indices are close to those reported by Bini et al. (2009) and Castle & Baldwin (2011) for the same cultivar, 11.5 and 12.8 respectively.
Environmental conditions and crop managements are among the major factors that influence fruit quality during its growth. For example, high temperatures contribute to the chlorophyll degradation and accumulation of soluble solids in the pulp (Ladaniya, 2008;Santos et al., 2010), while water availability regulates juice content and manages fruit production. Despite some differences, the evaluated rootstocks induced adequate physicochemical parameters for 'Salustiana' sweet oranges, meeting the minimum requirement of the international and domestic fresh market (Organisation for Economic Co-operation and Development, 2010; Companhia de Entrepostos e Armazéns Gerais de São Paulo, 2011). Table 3 describes the determination and quantification of organic acids. We did not observe significant differences among the treatments. Organic acids showed a total average of 737 mg 100 mL -1 of the juice of 'Salustiana' sweet orange, in which citric acid was the major organic acid identified in the juice ranging from 602.4 to 735.1 mg 100 mL -1 of orange juice, depending on the rootstock. These findings agree to Kelebek et al. (2009), Nour et al. (2010, Kafkas et al. (2011), Selli (2011) andScherer et al. (2012), who quantified citric acid as the major organic acid in sweet orange juices [C. sinensis (L.) Osb.]. Table 3. Organic acids composition in fruit from 'Salustiana' sweet orange tree grafted onto five rootstocks.

Rootstocks
Organic acids (mg 100 mL -1 of orange juice) Mean ± SD Non-significant at 0.05 level; a Total of organic acids expressed as citric acid equivalent. Kelebek et al. (2009) found 1266 mg of citric acid in 100 mL of orange juice produced in Turkey, close to those reported by Nour et al. (2010) notifying a concentration of 1392 mg 100 mL -1 in Romania. Contrasting these values, Scherer et al. (2012) reported concentration of citric acid between 517 and 529 mg 100 mL -1 of orange juice produced in Brazil, close to the values obtained in the present study (602 to 735 mg 100 mL -1 ) indicating that the concentration of organic acid, like other components, is dependent of the local environment (Lado et al., 2014). Now, ascorbic acid levels ranged from 56.2 to 60.9 mg 100 mL -1 of orange juice, similar to the values quantified by Kelebek et al. (2009) andNour et al. (2010) for sweet orange, 49.0 and 63.6 mg 100 mL -1 respectively. But higher than those noticed by Meléndez-Martínez et al. (2007) who described an average of ascorbic acid equal to 39.1 mg 100 mL -1 in 'Valencia' orange juice produced in Spain; and Legua et al. (2014) describing concentrations of 27 mg 100 g -1 of fresh weight of fruit from 'Clemenules' mandarin tree grafted onto 'Cleopatra' mandarin and 'Swingle' citrumelo cultivated in Portugal.
However, the level of ascorbic acid appears to be variable in citrus fruit due to the variability between cultivars and species. Fruit of 'Mexerica do Rio' produced in Viçosa (Brazil) presented high amount of ascorbic acid in its juice, about 115 mg 100 g -1 of juice (Silva et al., 2011); superior to the level mentioned above and those reported by Machado et al. (2017) for fruit from 'Tahiti' acid lime trees grafted on 12 rootstocks in Jaíba (Brazil), which ranged from 27 to 33 mg 100 g -1 of juice, being among the evaluated rootstock the 'Rangpur' lime, 'Swingle' citrumelo, 'Cleopatra' mandarin, and 'Sunki' mandarin. It indicates that the quantification of ascorbic acid is dependent on the cultivar associated with the managements and local environment. Thus, it is essential knowing the concentration of this organic acid in orange, because it is an important antioxidant and its concentration may indicate the quality of citrus juice (Kelebek et al., 2009;Legua et al., 2014), being the main biologically active form of vitamin C and can eliminate several free radicals (Davey et al., 2000;Scherer et al. 2012).

Sugar composition in orange juice
Despite the differences verified in the total soluble solids content of the evaluated treatments (Table 2), we did not notice significant differences among them for the concentration of carbohydrates (Table 4). There were three major sugar components in 'Salustiana' orange juice samples: sucrose, glucose, and fructose, just as described in previous studies regarding different citrus varieties (Cancalon, 1994;Kelebek et al., 2009;Kafkas et al., 2011;Kelebek & Selli, 2011). According to Silva et al. (1999) and Lee & Coates (2000), these carbohydrates represent more than 80% of the total soluble solids content presented in orange juice. The ratio between these compounds is generally 2:1:1, sucrose: glucose: fructose respectively. Similar ratio was observed in our study for 'Salustiana' orange juice. It is considered typical for commercial orange juices according to Zhang & Ritenour (2016) and is consistent to those described in other reports for orange cultivars (Lee & Coates, 2000;Kelebek et al., 2009;Kelebek & Selli, 2011;Kafkas et al., 2011).
The glucose:fructose ratio for the samples was also performed because it allows us to determine the quality and authenticity of orange juices (Kelebek & Selli, 2011). According to these authors, this ratio should present indices between 0.85 and 1.00 since orange fruit contains similar quantities of fructose and glucose, or even a slight increase in fructose, just as observed in our study for 'Salustiana' orange juice and for juice produced from different cultivars in the United States (Zhang & Ritenour, 2016). Thus, orange juice must present this pattern (0.85 -1.00) as a sugar constituent for the citrus industry, if not, an adulteration should be considered (Lee & Coates, 2000).
The glucose:fructose ratio ranged from 0.85 to 0.87 in 'Salustina' orange samples. These values were between those described by Zhang & Ritenour (2016) for 286 samples of orange juice produced in four States of the United States (0.8-0.9). They are also close to those observed by Kelebek & Selli, (2011) for 'Dortyol' oranges in Turkey (0.94), and lower than the maximum values required by the European Union legislation (1.03). It attends the orange juice requirement with regards to quality, authenticity, and identity because most of all produced juice in Brazil, as frozen concentrated orange juice (FCOJ), is exported to these countries (Lado et al., 2014;Kelebek & Selli, 2011).
In summary, despite the similarity among the evaluated treatments, the concentration, and composition of organic acids and carbohydrates may differ according to the citrus variety, soil-climate conditions, season, stress condition, fruit maturity, juice processing, crop and postharvest management as described in the literature (Cancalon, 1994;Jones, 1998;Rekha et al., 2012;Lado et al., 2014). However, the tested rootstocks do not appear to influence the organic acids and carbohydrates contents in 'Salustiana' sweet oranges.

Conclusions
'Salustiana' oranges produced on all evaluated rootstocks presents adequate physicochemical parameters, attending the quality standards for the citrus industry. 'C-13' citrange induces production of larger and heavier fruit. Significant differences were not observed between the treatments regarding organic acid and carbohydrate concentrations. As organic acids, citric and ascorbic acids were identified in the samples. The sugar composition consisted of sucrose, glucose, and fructose (1:1:2). The results regarding sugars and organic acids concentration characterize the evaluated cultivar, providing information about nutrition and the quality of fruit produced onto different rootstocks.