Cashew (Anacardiumoccidentale) used was from the Costa Grande region located in the state of Guerrero. The cultivation area is at the south, southwest, and west of the country, as well as at the north in the Tierra Caliente region and part of Michoacán state. Cashew fruit must be harvested at an optimum state of maturation to have the nut in the best condition. The March-June 2015 harvest was transported under refrigeration conditions (4°C) to avoid fast over-mature or deterioration and was stored in the laboratory at -18°C until analysis. The sample preparation was as follows:

a. Pulp: cashew apple was cut into small slices and macerated. The flesh was obtained on a Whatman No. 40 filter paper and was used to obtain the juice, which determined the total phenol and antioxidant properties.

b. Nut peel (coarse): the thick skin of the nut was subjected to thermal treatment at 100°C for 1 hour in an electric oven to extract the kernel (the thick shell was not studied because it is inedible). The nut peel was considered an untreated sample.

c. Husk: the thin crust covering the kernel was separated manually using tip tweezers and then was macerated to obtain a fine powder that was also packaged and stored at 4°C for analysis.

d. Nut: the nut without the thin shell was macerated until a powder was obtained that was packaged in a flask and stored in refrigeration at 4°C to determine phenolic compounds and antioxidant activity.

e. Toasted Nut: the toasting process employed an electric oven, in which the whole nut was placed in a tray and exposed at 133°C for two hours, followed by storage until subsequent analysis.

Physicochemical parameters: the pH measurement of the pulp of cashew apple was performed according to standard procedures29. The acidity percentage was determined using the flesh of cashew apple according to standard procedures expressing the results as % of citric acid for being the one of greater proportion in the samples30. Soluble solids (°Brix) were determined using the juice of the pseudofruit31. The density (g/ml) of cashew apple juice was determined according to standard procedures32. The humidity (%) of cashew apple was assessed by applying the technique of Bidwell Sterling33 using standard procedures34.

Samples of cashew apple, nut, commercial nut and husk were previously milled and macerated. Next, 2 g of cashew apple, 2 g of nut, 2 g of commercial nut and 0.5 g of husk were taken and placed into flasks separately. The samples were then subjected to extraction in 100 % methanol (10 ml) for 24 h in the dark and room temperature, followed by filtration to obtain the extract.

The method used was described previously35: 200 μL of the extract was mixed with 0.1 ml of Folin-Ciocalteu reagent for 3 minutes at room temperature, followed by the addition of 2 mL of sodium carbonate (7.5 %) and 2.8 mL of water. The sample was incubated for 1 h in the dark, followed by measurement in a Spectrophotometer at a wavelength of 725 nm. A curve of gallic acid was used, and the results were expressed as milligram equivalents of gallic acid (GAE) per 100 grams of sample fresh and dry.

A modified previously reported method36 was used to determine the antioxidant properties. The reducing power of the samples was determined by the method described by Brand-Williams et al. (1995),37 for the inhibition of the stable radical 2,2-diphenyl-1-picrylhydrazyl (DPPH). The assay was performed by mixing 0.1 mL of methanolic extract from the samples with 3.9 ml of a methanol solution of DPPH (0.005 g DPPH/10 ml of methanol). After 30 minutes in the dark, the absorbance at 515 nm was measured using 100% methanol as a blank and methanol 80% as a control treated with DPPH. The percentage of inhibition with respect to the initial absorbance obtained for the methanolic solution of DPPH is expressed as micromoles of TEAC (antioxidant activity equivalent of TROLOX) per 100 g of fresh sample, based on the TEAC type curve.

The raw material fruit was washed at the select maturation stage of both varieties. The fruit was divided into quarters and 500 g of fresh fruit, 400 g of sugar and 600 g of water were mixed, using citric acid as a preservative, and the marmalade was obtained using an artisanal method as follows: 3 g of the white skin of a lemon was added to 50 mL of water, followed by heating in a microwave oven for one minute (0.17 % citric acid). This hot citric acid mixture was used in the marmalade.

The previous preparation of water, citric acid and fruit was heated at 50 °C until 65 °Bx was obtained. After cooling 50 g samples in jars, each underwent pasteurization (100°C for 30 min) to obtain a shelf life of 45 days stored under refrigeration conditions38.

The sensory evaluation was carried out using 50 untrained judges performing discriminative and affective tests. A sample comprised 30 ml at 25°C. Discriminative analysis was performed using a comparison by pairs: two samples, (1) marmalade of cashew and (2) mango marmalade of a commercial brand, were evaluated for their characteristics-flavor, acidity, color intensity, aroma and like or dislike perception. A descriptive test applied using a scale of attributes, in which two samples were used: (1) marmalade of cashew and (2) marmalade of mango of a commercial brand39.

Statistical analysis of the results was carried out with the MINITAB 17.0 Software to obtain the mean ± standard deviation of the data (n = 3), and the multiple comparisons of Tukey test to determine significant difference between the samples.

(Table 1) shows the results obtained for the physicochemical parameters of the red and yellow cashew varieties.

A significant difference in pH values was observed between red and yellow cashew apple, while °Brix, density and acidity did not show it. Studies40 showed pH values of 4.14 and 4.15 for the red and yellow cashew varieties, respectively, similar to that obtained in this study. Some studies reported the pH of the false fruit, varying from 4 to 5.5. Guerrero et al. (2008)25, performed the physicochemical characterization, of the nut and false fruit of Anacardiumoccidentale L. (merey) under dry conditions and obtained for the harvested product pH values 6.47±0.02 and 3.57±0.07 for the nut and false fruit, respectively. The soluble solids obtained from both varieties (11.55 °Bx) were in the range reported for samples of Anacardiumoccidentale L. of Brazil and oscillated between 6 and 13 ºBx41,42. However, Chávez (2009)40 obtained pH values of fruits from El Salvador of 8.5 and 10 °Bx for the red and yellow cashew varieties, respectively. Guerrero et al. (2008)25 reported a higher value (17°Bx) for false fruit. An increased concentration of total soluble solids (monosaccharide) exists because of polysaccharide hydrolysis (homopolysaccharide) during maturation41,43. Significant differences (with a level of confidence of 95%) were not observed in the density values between both varieties. Cashew red presented a higher percentage of acidity than that reported by Chavez (2009)40 at 0.36 % total acidity, but the variety of cashew used was not specified. Guerrero et al. (2008)25 reported a higher value of acidity in fruits of A. occidentale L. under dry conditions (1.42 %) than that observed under conditions of irrigation (0.32 %)44. False fruit pulp presented a lower acidity than that reported for samples harvested in Tanzania, Costa Rica, India and Brazil41,42. Rodríguez (2011)44 expressed values of acidity of 0.185 ±0.01% for malic acid of the juice of cashew fruit.

The values of humidity for cashew red and yellow were similar to those reported by Chávez (2009)40 who attributed the differences to the conditions of cashew harvest, mature stage, and origin. The physicochemical characteristics of the fruits vary depending on their maturity, handling, and origin (climatic conditions and soil, among others). Water supply is an essential influence of the physiological development of the fruit (preservation of nutrients) and determines the quality and quantity of the harvested product45. For the nut sample, only the humidity generated a value of 12.84 %. This value has not been reported.

The results of the total phenolic compounds obtained from extracts of different samples of cashew apple and nut are presented in Table 2.

Cashew red pulp showed a superior value of the phenol content to that of yellow cashew pulp. Rufino (2010)46 determined the content of phenols in non-traditional fruits from Brazil, reporting a value of 118±3.7 mg GAE/100 g wb for cashew (Cajú). Fruits such as Bacuri, Carnaúba, Gurguri, Jaboticaba have a wide range of values of 23.8±0.7 to 549±22.2 mg AG/100 g. A high variability in the content of phenolic compounds of cashew is shown in the literature because of their origin, soil difference cyclic conditions, seasonal growth, postharvest, storage and extraction methods of polyphenols47. Plants can activate polyphenol synthesis as a response to stress conditions such as injuries, pathogens, and low nutrient compounds48.

On the other hand, when comparing the content of total phenols obtained for the red cashew with that for other fruits, the fruit of cashew had a value close to red fruits, such as strawberry (205.4±2.1 mg GAE/100 g wb), pomegranate (328.7±2.5 mg GAE/100 g wb51, and was higher than that reported for saladette tomato (92.31 mg GAE/100 g wb)52.

The untreated nut sample showed a higher content of phenolic compounds. However, in the nut toasted at 133°C for two hours, the TPC was decreased. Ogunlade et al. (2014)49 and Chandrasekara50reported values of 204.95±1.4 and 114±0.47 (defatted meal) mg GAE/100 g db, respectively, for the nut of Anacardumoccidentale. The value of 528±1.0 was published for the nut treated at 130°C for 33 min50. This difference could be because, in the toasted process used in the present work, the samples were exposed to higher temperatures and more extended periods (133°C; 2 hrs), causing structural changes, such as isomerization and a reduced content of phenolic compounds53,54. Other causes of these differences could be because, in the present study, defatted samples were used that could modify the general composition53. However, Chandrasekara and Shahidi50 reported that, during heat treatment, the reaction between reducing sugars and amino acids, known as the Maillard reaction, can occur, leading to the formation of various byproducts, intermediate and brown pigments (melanoidins), which may contribute to the TPC, flavor, antioxidant activity and color of food. The value of the TPC of nut obtained in the present work (174.19±20.98 mg GAE/100 g db) was higher than that reported for almond, hazelnut and Macadamia nut at 114, 111 and 87 mg GAE/100 g (wb), respectively, but inferior to that reported for cashew nut (nut of India), 381 mg GAE/100 g (wb)55 (Figure 2).

A high value was obtained for the TPC of the husk of cashew nut (74.30 mg GAE/g) compared with those of the kernel and false fruit. Other authors have reported a value of 269.05±9.77 GAE mg/g in a defatted meal50. The samples of cashew nut and skin had proteins and sugar, and the formation of MRPs during roasting is possible. Thus, the roasting condition, as well as the type of nut, affects the TPC values of the extracts of cashew skins50. Studies have reported high amounts of husk or cuticle of different nuts. Rosales et al. (2014)54 declared a value of 132.9 mg GAE/g for the peanut cuticle Virginia variety. The high amounts of TPC obtained in the husk sample of cashew compared with the fruit are because the husk is the external part of the nut and protects against stress conditions, synthesizing the highest amount of phenols in this structure56. Francisco & Resurrección (2009)57 reported that phenolic compounds in the outer layer of plants as shell, cuticle, and sheath are present in a high concentration to protect all internal materials as cotyledons. Some phenolic acids, although they are united with a covalent bond of polymers, are not soluble in water and other components of the cellular wall as arabinoxylanes58. Thermal treatment could set free antioxidants of low molecular weight that could be sub-units of polymers of high molecular weight59. Studying the skin or husk of nuts is important because several studies have reported that the husk has healing properties such as hypoglycemic, hypoglycemic, antihypertensive, astringent, antihelminthic, and anti-inflammatory properties56.

(Table 2) shows the results of the antioxidant activity of the extracts of false cashew fruit and treated and untreated nut. Red cashew presented a higher antioxidant capacity at 77.65 μmol Trolox/g and was superior to that of the kernel. Additionally, red cashew had a higher content of TPC. The differences between red and yellow cashew can be attributed to the type and amounts of phenolic compounds present in each because they contain tannins, polyphenols, and carotenes55. Some studies have reported that cashew has different characteristics and amounts of phenolic acids, such as ferulic, gallic, ellagic, p-coumaric and syringic acids, as well as flavonoids, such as myricetin, catechin, epicatechin, and epigalocatequine59,60. Ogunlade et al. (2014)49 reported a value between 50 to 53.21 μmol Trolox/100 g for Anacardiumoccidentale nut but they used a FRAP method.

Comparing values of the antioxidant capacity obtained for cashew with that of a common fruit, such as red apple (Malus sylvestris), which contains 0.259 μmol TE/g59,or with an exotic fruit of a higher consumption, such as soursop, which contains 2.88 μmol TE/g61, cashew presented a higher antioxidant capacity than these fruits. It is essential to exploit the benefits of cashew apple and use it as a raw material in functional food processing.

(Table 2) presents the values of the nut extracts. The nut studied had 38.52±2.9 μmol TE/g sample, and the commercial nut value was lower at 2.48 μmol TE/g sample. These results could be due to processes such as toasting and salting that are used commercially; high-temperature processes and high salt content can produce structural changes in phenolic compounds and generate structures with a lower antioxidant capacity and storage time54,62. Chandrasekara50reported a higher value of 58.14±2.84 (μmols Trolox/g db, defatted meal) for the toasted nut (130°C, 33 min) of Anacardumoccidentale than the one obtained in the present study and may be the difference was because of the treatment time. The husk possesses a higher antioxidant capacity than other samples and is because it is the structure that protects the seed and is where the antioxidant compounds are synthesized in a greater proportion54.

Antioxidants isolated from foods and beverages, such as the cashew apple and its processedproducts, may be superior to non-natural products15,51.

By relating the phenolic compounds against the antioxidant capacity of the fruit, nut and husk by means of a linear regression. The coefficient of determination provides a value of 0.9725 for red cashew, 0.9383 for yellow cashew, 0.9857 for nut and 0.9241 for husk; these values indicate a good fit of the data towards the linear, which represents that there is a directly proportional relationship between the phenolic compounds and the antioxidant capacity.

(Figure 3) shows the pH and acidity of marmalade for 6 weeks. The pH value decreased at the end of the period, showing a value of 3.68; this value is a warranty of a good shelf life of the final product. This value is in the range reported for the pH of marmalades (3.25-3.75) to ensure microbiological safety38. The final acidity level was 0.35% (presenting 22.2 % of reduction), and, according to standards, is acceptable until a value of 1%. Those values have been reported in other marmalades such as mango-flavored marmalade with an acidity value of 0.5% or orange-flavored marmalade with a final acidity value of 1%; the acidity value of cashew marmalade is in the range of that for jams of other more common fruits38. This low reduction might be due to the process because of the addition of sugar and water, and the general change in composition. The thermal process could cause structural changes that caused the difference in the acidity63.

Cashew is considered an acid fruit, giving the confidence of a final product with good quality38. The acidity of prepared marmalades (0.35 %) presented a significant difference in the acidity of fresh fruit (0.43 g citric acid/100 ml). In jams prepared with fruits, a lower value of acidity and low pH values give the advantage of the best shelf life; it must be considered that the addition of citric acid forms a gel, avoiding sugar crystallization and acting as a preservative agent64.

Rodríguez (2011)44 prepared a juice of cashew pulp and presented a pH value of 4.4, which was similar to that reported by other authors65,66. The acidity value, determined as the malic acid percentage, was 0.177 %. This value is outside the range of established norms in Brazil67 for the juice of false fruit of cashew, which sets values between 0.22 and 0.52 %.

(Figure 4) shows the °Bx and phenol variation for six weeks. The °Bx data follow a linear trend and a slope almost at 0, obtaining a final value of 63.41°Bx, and an acceptable amount of marmalades according to the norms31. The behavior observed of phenolic compounds showed a slight increase in the initial value of 106.25 mg GAE/mL to 114 mg GAE/mL (6.8 % of growth). The cause may be the slight changes in temperature and light during the storage time64,67. The °Brix value did not change during this period. Vera (2012)68 formulated marmalades of other fruits, such as the apricot, and evaluated the physicochemical properties during storage (6 weeks) to guarantee the shelf life of the product by 12 months under storage conditions at room temperature (37°C). De Oliveira et al. (2015)69 also prepared strawberry marmalade, and evaluating the content of phenolic compounds during 60 days of storage showed a decreased of 235.84±10.15 mg EAG/100 g to 226.57±12.42 mg EAG/100 g, similar behavior to that in the present work.

Those results could suggest that some structural changes in the phenolic compounds can occur after 30 days. Klopotek et al. (2005)70 argued that the processing and storage conditions caused co-pigmentation and auto-association of molecules, reducing the number of hydroxyl groups available to react in the Folin-Ciocalteu colorimetric assay, apparently decreasing the concentration of total phenols. The release of monomers of polymeric polyphenols also showed the highest values of the content of total phenols using the Folin-Ciocalteu assay71. Thus, it could explain the apparent increase in the phenolic compounds during the last week of storage of marmalade elaborated in this work. Thus, marmalade processing is an alternative to preserve the properties of bioactive compounds present in cashew fruit72.

Diverse products derived from cashew have been developed as beverages. Rodríguez (2011)44 developed a juice with cashew apple and reported a total phenolic concentration of 2321 mg GAE/L, lower than the value reported by Melo et al. (2006)73in cashew juice of 2952.5 mg EAG/L, but higher that the value reported in the juice of common fruits, such as orange, which contain values of 620.7 to 630.5 mEq GAE/L74. For the above, the false fruit of cashew represents a good source of antioxidants (106.25 mg GAE/mL) compared with fruits of regular consumption.

A discriminative and descriptive test was applied using 20 untrained judges. The first test used comparison pairs to find differences between marmalade formulated for this work and a commercial brand. According to panelist perception, cashew marmalade presented a less acidic flavor than the mango marmalade of the retail brand. The results matched the physicochemical parameters, and no significant differences were found in the acidity content during the period of storage.

The panelists observed a similar color of marmalade of cashew (color characteristic) and mango. Regarding flavor, cashew marmalade was a slightly astringent; this result also matched those obtained for the physicochemical parameters (pH and acidity). According to the panelists, mango marmalade was more astringent than cashew marmalade. The cause may be that commercial jams used the highest amount of citric acid as a preservative; in cashew marmalade, the citric acid was obtained from a natural source and was at a lower concentration.

The second test consisted of describing attributes of flavor using a graphic scale; the objective was to evaluate the characteristics of cashew marmalade compared with mango marmalade of a commercial brand. The chart scale of taste used grading extremes: low astringent to extremely astringent. Cashew marmalade was evaluated as having a low astringent flavor, while mango marmalade had a high astringent flavor; the difference might be due to the high amount of preservative (citric acid). On the other hand, fruits such as cashew and mango have a content of different polyphenols, such as phenolic acids, which tend to contribute to the astringency and acidity in foods75,76.

Cashew marmalade presented a higher °Bx value (63.41) than that of mango marmalade (30.67 °Bx)38. According to sensorial analysis, cashew marmalade belongs to the group of jams, jellies, and marmalades, according to the values of Brix degrees established in the norm of CODEX75,76. The final product presented a good appearance at 45 days of storage and refrigeration.