Espacios. Vol. 36 (Nº 16) Año 2015. Pág. 20

Potential of wild food plants from the semi-arid region of northeast Brasil: chemical approach ethnoguided

Potencial de plantas alimentícias silvestres da região semiárida do nordeste do Brasil: abordagem química etnoguiada

Edna Maria Ferreira CHAVES 1; Jurandy Nascimento SILVA 2; Alessandro LIMA 3; Ulysses Paulino ALBUQUERQUE 4; Roseli Farias Melo BARROS 5

Recibido: 05/05/15 • Aprobado: 13/06/2015

Contenido

1. Introduction

2. Materials and Methods

3. Results

4. Discussion

5. Conclusions

References

1. Introduction

Wild food plants are classified as non-timber forest products (NTFPs) (Guinand, Lemessa,  2001) and are important for the subsistence for many peoples throughout the world (Heubach et. al. 2011). Studies have shown that although many of these plants, referred to as "bush plants" (plantas do mato), are used for food, others have fallen into disuse or are under-used (Rapoport, Ladio, 1999; Nascimento et. al. 2012, 2013). 

In recent decades there has been an increasing interest around the world in the investigation of native regional food plants traditionally used by local communities for their nutritional value (Fernández, 2009) and anti-oxidant potential (Silva et. al. 2012), in a search for alternative ways to guarantee food and nutritional security (Gundc, El, 2003). In Brazil, however, considering its continental dimensions of approximately 8,500,000 km2 (Ab'Sáber, 1974) and the richness of its flora of over 56,000 species (Giulietti et. al. 2005), there are few ethnobotanical investigations which have as their objective the clarification of the food potential of native plants (Albuquerque 2010; Araújo, Lopes, 2011; Morais, Silva, 2011; Nascimento et. al. 2011, 2012, 2013).

The objectives of this study were to analyse the centesimal composition; quantify levels of ascorbic acid, anthocyanins, carotenoids, flavonoids and phenolics; and evaluate the in vitro anti-oxidant activity in six wild species used for food in the municipalities of Buriti dos Montes and Cocal, in Piauí state, in the semi-arid region of northeast Brazil.

This study offers on the one hand a significant contribution to food programs which aim to improve the food and nutritional security of rural communities, making possible the inclusion of native wild foods in diets. On the other hand, the study contributes to adding value to local products by revealing specific properties which can be exploited by agro-industry and the food technology industry.

2. Materials and Methods

2.1. Delimitation and characterization of the study area

An ethnobotanical survey was carried out in four rural communities: Bebedouro and Oiticica, in the municipality of Buriti dos Montes, and Itapecuru and Pinga, in the municipality of Cocal, both located in Piauí state, in the semi-arid region of northeast Brazil. The municipality of Buriti dos Montes has an area of 2,652.103 km², situated at 05º18'43"S e 41º05'52"W, 250 km from Teresina, the state capital. The municipality of Cocal has an area of 1,269 km², located at 03°24'53,9"S 41°40'03,9"W, 273 km from Teresina (IBGE, 2008). The mean annual temperatures  and mean annual precipitation of Buriti dos Montes and Cocal are, respectively, 24.4°C and  26.6°C and 1,100.8 mm and 1,168.4 mm. According to Köppen's classification, the study areas belong to the Aw' – Tropical class, with rainfall maxima in the summer (Medeiros, 2004).

The estimated population of the two municipalities is respectively 7,977 and 26,044 inhabitants (IBGE, 2010). In Buriti dos Montes the communities of Bebedouro and Oiticica have respectively, a total of 18 and 15 family units comprising approximately 77 and 45 inhabitants. In Cocal the communities of Itapecuru and Pinga the observed values were, respectively, 49 and 11 family units, comprising 182 and 42 inhabitants.

After approval and certification by the Research Ethics Committee (CAAE nº 02773212.0.0000.5214) meetings were held to present the objectives of the project and how it would be carried out. The Statement of Informed Consent Form (ICF) was read, and after was signed by the person to be interviewed. One individual was interviewed in each family unit.  The interviews were carried out using semi-structured questionnaires (Albuquerque et. al. 2014).

Selection of the species and sampling

Criteria used to select the species to be characterized chemically and assayed for in vitro anti-oxidant activity: 1. Cited as a food plant by 100% of informants in at least one of the studied municipalities; 2. Available for collection in the years 2012 and/or 2013, the period when botanical collections were made in sufficient quantity and quality for analysis. 3. Published information on their nutritional composition and/or anti-oxidant activity lacking or very scarce. The selected species are: coroa-de-frade (Melocactus zehntneri (Britton & Rose) Luetzelb.), jacarandá (Swartzia flaemingii Raddi), macambira (Bromelia laciniosa Mart. ex Schult. & Schult. f.), jatobá (Hymenaea martiana Hayne), taturapé (Randia armata (Sw.) DC.) and taturubá (Pouteria macrophylla (Lam.) Eyma) (Table 1).

Using the methods of Rufino et al. (2010), samples were collected of the plant parts used as food by the interviewees.

Table 1. Species selected for chemical analysis and anti-oxidant activity in an ethnobotanical survey in the communities of Bebedouro and Oiticica, municipality of Buriti dos Montes and Itapecuru and Pinga, municipality of Cocal, Piauí, northeast Brazil.

Specimens collected in: 1Buriti dos Montes and Cocal; 2Bebedouro and Oiticica, Buriti dos Montes; 3Oiticica; 4Itapecuru and Pinga, Cocal. *Sembereba: local term for the extract from manually maceration of fruits in water.

2.3. Sample analysis

2.3.1. Determination of pH, total titratable acidity (TTA) and total dissolved solids (TDS)

The analyses of pH and total titratable acidity followed AOAC (2005). Respectively, the methods used a potentiometer with a previously calibrated bench pH meter and a calorimeter with volumetric titration. For the analysis of total dissolved solids (in °Brix) a portable refractometer was used.

2.3.2. Centesimal composition

The analysis of moisture level, fixed mineral residue (ash) and ether extract (lipids) followed IAL (2008). To measure moisture level, 2 g of the sample were dried in an oven at 105 °C to a constant weight. To determine ash content, 5 g of the sample were incinerated in a muffle furnace at 550 °C until turning into white or pale grey ash. To obtain the ether extract, a Soxhlet extractor was used and an oven at 105 °C to constant weight. For protein analysis, the quantity of nitrogen was determined by the Kjeldahl digestion process with modifications (IAL, 2008). Carbohydrates (including fibres) were measured by the difference between 100 and the sum of percentages of protein, lipids, ash and moisture (AOAC, 2005). The total energy value (TEV) was determined by the summing the calories (Kcal) provided by total carbohydrates, total lipids and proteins, multiplied by their values in grams, according to the Atwater system (MERRILL; WATT, 1973), which considers 4 Kcal, 9 Kcal and 4 Kcal respectively (IAL, 2008). The quantities of carbohydrates, ash, proteins, lipids, moisture and total energy value were compared with those from food plants used conventionally in northeast Brazil and referenced in the pertinent literature and/or the Brazilian Table of Food Composition (TACO, 2011).

2.3.3. Bioactive compounds

To quantify the ascorbic acid, the standard method no. 43.065 of AOAC (1984) was used, as modified by Benassi e Antunes (1988). 2 g of sample were homogenized in 50 mL of a 2% oxalic acid solution at ambient temperature for two minutes. A 20 mL aliquot of this preparation was taken and diluted in a further 50 mL of 2% oxalic acid solution. After filtering the diluted extract, a 10 mL aliquot was used for the quantitative determination of ascorbic acid using oxidative titration with 2.6 dichlorophenolindophenol 0.01%. The result was expressed in mg of ascorbic acid/100 g of sample.

To determine the carotenoids, the method of Rodriguez-Amaya (2001) was used; 5 g of pulp and 2 g of acid-washed Celite® were added to an Erlenmeyer flask containing 20 mL of acetone. The contents were homogenized in an orbital shaker at 10 rpm. Filtration was by vacuum in a Büchner funnel using C41 quantitative filter paper and the sample was washed with acetone until the extract was colorless. The filtrate was transferred to a separation funnel to which were added 30 mL of petroleum ether and 100 mL of distilled water. The subnatant, the water and acetone were almost entirely discarded. Another 100 mL of distilled water was added to the separation funnel and the process repeated. This cycle of exhaustive washing was repeated four times. The solution obtained of petroleum ether and pigments was then measured using a spectrophotometer at 450 nm using petroleum ether to calibrate the instrument at zero. The total carotenoid content, expressed in µg of β-carotene.g-1 of sample was determined using the formula: Abs × 50 mL × 106 / 100 × E1% 1 cm × P, where Abs = absorbance of the sample, E1% 1 cm = 2592 (coefficient of absorption), P = sample weight.

For anthocyanins and flavonoids the method of Francis (1982) was used. A quantity of 0.5 g of the sample was weighed in an Erlenmeyer flask protected from light, adding 10 mL of a previously prepared (85:15) solution of 95% ethanol + 1.5N HCl. After homogenization, the content was transferred without filtering to a 25 mL volumetric flask and the volume compared with 95% ethanol + 1.5N HCl (85:15). The material was kept refrigerated and at rest for 12 h in darkness. Measurement was them carried out with a digital spectrophotometer, model SP-220, at a wavelength of 535 nm. The total anthocyanin level was then obtained using the formula: Absorbance × dilution factor / 98.2. The dilution factor was obtained by dividing the sample weight (0.5 g) by the dilution volume (25 mL). For the analysis of flavonoids the same process was carried out as for anthocyanins, changing only the reading band on the spectrophotometer to 374 nm.

2.3.4. Preparation of extracts of the species for analysis of total phenolics and in vitro anti-oxidant activity

The extracts were prepared following Rufino et al. (2010) with modifications. The extraction was carried out in triplicate in Falcon tubes, adding to each tube 2 g of dried sample, and 10 mL of an 80/20 solution of methanol/water. After addition of the solvent, the tubes were shaken in a vortex mixer and left to stand for extraction at room temperature (approx. 25 ºC) in the dark, for 24h. The tubes were then centrifuged at 4,000 rpm for 10 minutes and the supernatant liquid obtained was vacuum-filtered in a Büchner funnel with C40 quantitative filter paper, yielding the material considered as the extract. This extract was then stored in amber flasks below zero at approximately -20 ºC for later analysis.

2.3.5. Level of total phenolic compounds

To determine total phenolics, the method of Swain e Hills (1959) was followed. After 24h, 0.5 mL of the methanol/water extract was taken and 8 mL of distilled water and 0.5 mL of Folin-Ciocalteu reagent were added. The solution was homogenized and after 3 minutes 1 mL of saturated NaCO3 was added. After standing for 1h, the absorbance measurements were carried out using a spectrophotometer at 720 nm. Gallic acid was used as a standard at concentrations of 150, 300, 450, 600 and 750 mg/L to construct a calibration curve. Using the straight-line graph obtained, the level of total phenolics was calculated expressed in mg of gallic acid/100g of dried sample.

2.3.6. In vitro anti-oxidant activity

2.3.7. Method of capture of DPPH· (2,2-diphenyl-1-picrylhydrazyl) radicals

For the analysis, 1.5 mL of a methanol solution of DPPH• (6x10–5M) and an aliquot of 0.5 mL of the samples were added together in different concentrations for each extract. Measurements were made with a spectrophotometer at 517 nm, after 2, 5, 10, 20 and 30 minutes from the start of the reaction. The fall in optical density of the samples and of the gallic acid standard were correlated with the control, establishing the percentage decoloration of the DPPH• radical according to the formula: % protection = [(Abscontrol – Abssample) / Abs control] x 100.  After establishing the reaction equilibrium it was possible to calculate the concentration of extract necessary to reduce the DPPH• radical by 50% (EC50 value) (Brand-Williams et al., 1995).

2.3.8. Method of capture of ABTS•+ [2,2'-azino-bis(3-ethylbenzothiazoline-6-sulphonic acid)] radicals

The ABTS•+  radical was generated from the reaction of 7 mM of ABTS with 2.45 mM of potassium persulphate, the reaction mix being maintained at room temperature and in darkness for 12h. After this period the solution was diluted in absolute ethanol until a solution with absorbance of 0.70 (± 0.01) was obtained.  40 μL of sample (diluted in ethanol) were added to 1,960 μL of the solution containing the radical, and the absorbance was measured with a spectrophotometer at 734 nm after 30 minutes reaction time. The fall in optical density of the samples was correlated with the control (only the radical), establishing the percentage discoloration of the ABTS•+ radical (RE et al., 1999).

2.4. Statistical analysis

All the results were expressed as means (M) ± standard deviation (SD) of three independent experiments. The tests for determination of ascorbic acid, total anthocyanins, total carotenoids, total phenolics, flavonoids and in vitro  anti-oxidant activity (ABTS and DPPH) were compared using Analysis of Variance (ANOVA) and tested using Tukey's range  test. Prisma 5.0 (GraphPad Prisma) software was used for computation and a 5% significance level was adopted (p < 0.05).

3. Results

3.1. Determination of pH, total dissolved solids (TDS), total titratable acidity (TTA) and the relation of total dissolved solids/total titratable acidity (TDS/TTA)

The data resulting from the physico-chemical analysis of the species samples are presented in Table 2. Only R. armata had fruits with a pH equal to 4.5, which classifies them as acidic and thus less susceptible to attack by microorganisms. The fruits of the other species had larger values, in the case of S. flaemingii attaining a pH very close to 7.0, thus chemically basic and more vulnerable to contamination, implying the need for care in using them as food. As regards TSD, the samples varied between 3.13 a 32.93%. Pouteria macrophylla showed the highest value which, added to the fact that the fruits of this plant have 44.98 g of carbohydrate per 100 g of the edible part (Table 2), explains why 100% of interviewees qualified this as a sweet fruit. Among the other species it is noteworthy that S. flaemingii was cited by 63% of interviewees as having a sweet flavour when eaten raw. As regards TTA, we observed that 100% of the samples had values above 1.5% and 50% had values above 3%, percentages considered high for fruits and which resulted in low values for the TSD/TTA ratio for R. armata, P. macrophylla and S. flaemingii, indicating that consumption when raw may not provide good palatability.

Table 2. Total titratable acid, pH, and total dissolved solids, centesimal composition (g/100g) of carbohydrates, ash, proteins, lipids, moisture e and total energy value     of analysed parts of wild food plants selected in an ethnobotanical survey in the communities of Bebedouro and Oiticica, municipality of Buriti dos Montes and Itapecuru and Pinga, municipality of Cocal, Piauí, northeast Brazil.

Legend: M:mean; SD: standard deviation; TDS: total dissolved solids; ºBrix: brix level; TTA: total titratable acid; NA: sample not analysed; U: moisture; C: ash; P: total proteins; L: total lipids; Cb: total carbohydrates; TEV: total energy value; M: mean; SD: standard deviation; Kcal: kilocalories.

3.2. Centesimal composition

3.2.1. Carbohydrates, ash, proteins, lipids, moisture and total energy value

The observed values of these parameters are shown in Table 2. Regarding carbohydrate content (including fibre), H. martiana had a value above 50% with 77.89 g/100 g of pulp, and P. macrophylla approached this percentage with 44.98 g/100 g. The other species varied between 12.83 and 27.66 g/100 g, except M. zehntneri which had only 2.17 g/100 g of the edible part. 66.66% of samples had an ash content of less than 1 g/100 g. However, H. martiana differed in a value of 4.39 g/100 g of sample.

In 50% of species we observed protein content which varied between 1 and 1.5 g/100 g (M. zehntneri, R. armata e S. flaemingii) and in the others (B. laciniosa, H. martiana and P. macrophylla) content values were lower than 1 g/100 g. Among the species studied, lipid content was highest in H. martiana at 2.38 g/100 g of pulp, while R. armata and S. flaemingii showed lower levels (0.33 g/100 g each). Most samples (83.33%) showed moisture content above 50%, the highest being M. zehntneri with 95 g/100 g of the edible part and the lowest H. martiana with only 14.57 g/100 g of pulp. In 66.66% of the species studied, total energy value was above 100 kcal/100 g of sample, with the species in the descending order H. martiana, P. macrophylla, S. flaemingiiand B. laciniosa. That in spite of the low levels of proteins and lipids, the higher carbohydrate levels mean that the majority of studied species have significant potential to contribute to the daily calorie requirements of children and adults in these communities (Table 3) and that H. martiana could be an important source of minerals, and M. zehntneri of water.

Table 2. Total titratable acid, pH, and total dissolved solids, centesimal composition (g/100g) of carbohydrates, ash, proteins, lipids, moisture e and total energy value     of analysed parts of wild food plants selected in an ethnobotanical survey in the communities of Bebedouro and Oiticica, municipality of Buriti dos Montes and Itapecuru and Pinga, municipality of Cocal, Piauí, northeast Brazil.

Legend: M:mean; SD: standard deviation; TDS: total dissolved solids; ºBrix: brix level; TTA: total titratable acid; NA: sample not analysed; U: moisture; C: ash; P: total proteins; L: total lipids; Cb: total carbohydrates; TEV: total energy value; M: mean; SD: standard deviation; Kcal: kilocalories.

3.2.2. Bioactive compounds: ascorbic acid (vitamin C), anthocyanin, carotenoids, total phenolics and flavonoids

The results for the above bioactive compounds are presented in Table 3. The ascorbic acid content of the fruits varied from 3.55 mg/100 g (S. flaemingii) to 8.76 mg/100 g (R. armata). The fruits of H. martiana and P. macrophylla were statistically similar (p ˂ 0.05), with intermediate values. The lowest levels found amongst the samples analysed were the cladodes of M. zehntneri and the leaves of B. laciniosa.

The highest content levels of carotenoids and total phenolics were observed in R. armata. The second highest and statistically similar carotenoid levels (p ˂ 0.05) were observed in H. martiana, M. zehntneri and S. flaemingii and for total phenolics in B. laciniosa and S. flaemingii. The highest anthocyanin content was found in H. martiana, while B. laciniosa and S. flaemingii had intermediate and statistically similar levels (p ˂ 0.05), and the lowest levels were recorded in P. macrophylla and R. armata. Flavonoid content occurred in the following decreasing order: B. laciniosa, R. armata, P. macrophylla and H. martiana, the last two being statistically similar (p ˂ 0.05).

3.3. In vitro antioxidant activity

Pouteria macrophylla and H. martiana showed respectively the highest and second highest capacity to reduce free radicals ABTS•+ and DPPH· (maximum anti-oxidant activity in a period of 30 minutes). Using the values of anti-oxidant capacity equivalent to TROLOX (TEAC) (Table 4) and the anti-oxidant activity at 50% of effective concentration (EC50) (Table 5), the results showed that the secondary metabolites of the two species were effective in reducing the free radicals ABTS•+ e DPPH·. This result implies that such species are rich in anti-oxidant substances which are detectable by these methods. Randia armata showed a TEAC near to that found in H. martiana (Table 4), however these were statistically different. The other species proved not to be good sources of in vitro anti-oxidant activity in the presence of the free radicals ABTS•+ e DPPH· and all the studied species were statistically different (Tables 4, 5).

Table  4. TEAC value (Equivalent Anti-oxidant Capacity to TROLOX) by the ABTS method in a 80/20 methanol/water extract of wild food plants selected in an ethnobotanical survey in the communities of Bebedouro and Oiticica, municipality of Buriti dos Montes and Itapecuru and Pinga, municipality of Cocal, Piauí, northeast Brazil.

Legend: M: mean; SD: standard deviation, ABTS•+ [2,2'-azino-bis(3-ethylbenzothiazoline-6-sulphonic acid)],  a,b,c,d,e,f Means, in the same column, followed by different letters differ statistically from one another at 5% significance level (p ˂ 0.05)

-----

Table 5. Concentration interval, linear coefficient (r2), protection percentage and anti-oxidant activity (EC50 in μg/mL) in a 80/20 methanol/water extract of wild food plants selected in an ethnobotanical survey in the communities of Bebedouro and Oiticica, municipality of Buriti dos Montes and Itapecuru and Pinga, municipality of Cocal, Piauí, northeast Brazil, using the free radical DPPH·

Legend: M: mean; SD: standard deviation, DPPH· (2,2-diphenyl-1-picrylhydrazyl), a,b,c,d,e,f Means, in the same column, followed by different letters differ statistically from one another at 5% significance level (p ˂ 0.05)

4. Discussion

4.1. Bromelia laciniosa

This species was shown to be a good source of flavonoids (Table 3), which are compounds known to have beneficial anti-oxidant activity against free radicals and are considered important in retarding or preventing the oxidation of food products (Celli et al., 2011). The level observed was higher than that cited by Rufino et al. (2010), in mg/100 g, for bacuri (Platonia insignis Mart.= 16.9) and murici (Byrsonima dealbata Griseb. = 13.8).

This species is also noteworthy for its carbohydrate level (Table 2) and because it remains available during prolonged droughts, thus providing a food resource of great importance as a complement for the daily calorie requirement of local families. Its calorific value (Table 2) is higher than that of cooked sweet potato (Ipomoea batatas L. = 77 kcal/100 g (TACO, 2011), which are foods regarded as good sources of calories. Nascimento et al. (2012) analyzed Encholirium spectabile Mart., from the same plant family as B. laciniosa with a similar life form and economic use, and they found calorific values (124 kcal/g) similar to those of the latter species analysed in the present study. These authors stated that E. spectabile is one of the main emergency foods used by the communities they studied, as we observed in the present study for B. laciniosa. 

We thus consider that B. laciniosa has potential for use in the food industry, either for products (biscuits, cakes, cuscuz, breads, porridge base).

4.2. Hymenaea martiana

This species has a high calorific value (Table 3), higher than those of species which are common food sources in Brazil and known as energy-rich as, for example, cooked black and carioquinha beans (Phaseolus vulgaris L.),with respectively 322 kcal/100 g and 326 kcal/100 g (TACO, 2011). Its calorific value is also higher than that of caatinga fruits, being exceeded only by coco-catolé (Syagrus cearensis Noblick = 393,67 kcal/100 g), of the nine native species of Paraíba and/or Pernambuco analysed by Nascimento et al. (2011). According to FAO (2001), a child of 10 years and an adult have calorific needs of approximately 2,000 kcal e 3,100 kcal per day, respectively, so that 100 g of the fruit pulp of this species would be sufficient to meet 16.80% of the daily calorific requirement of a child and 10.84% of that of an adult.

After analysing the content and lipid fractions of another species of Hymenaea (Hymenaea courbaril L.), Dias et. al. (2013) noted the volume and quality of the lipid fractions of the aril and seed for use in food. This result indicates the possibility of similar levels in H. martiana and suggests further investigation.

The ash content (Table 2) is also noteworthy, being similar to raw kidney beans (feijão-roxo) (Phaseolus vulgaris L. = 4,0 g/100 g) TACO (2011), which is considered to be a good source of minerals for the daily food of families in the semi-arid region of Brazil (Mesquita et. al. 2007); the anthocyanin content (Table 3) is higher than that of guava (Psidium guajava L. = 2,7 mg/100 g) (Kuskoski et. al. 2006) and the total phenolic content (Table 3) is higher than that of the pequi aril (Caryocar brasiliense Camb. = 209 mg/100 g) (TACO, 2011). The fruits of P. guajava and C. brasiliense are indicated by research as good sources of anti-oxidant substances (Roesler et. al. 2008; Corrêa et. al. 2011).

Of the species we analysed for efficiency in inhibiting the free radicals ABTS•+ and  DPPH· in in vitro screening(Tables 4, 5), H. martiana showed a higher TEAC than that of species routinely consumed in the semi-arid region of Brazil and indicated as good sources of bioactive compounds, such as araticu-do-mato (Rollinia sylvatica A. St.-Hil. = 3,85)  (Pereira et. al. 2013) and caju (Anacardium occidentale L. = 11,2) (Rufino et. al. 2010). 

The data analysed here reveal that H. martiana has the potential to complement the calorie requirements of communities in the semi-arid region of Brazil and to provide protection against the action of free radicals.

4.3. Melocactus zehntneri

This species has a high moisture content (Table 2), greater than that reported by Nascimento et. al. (2011) for the cladodes of mandacaru (Cereus jamacaru L. = 86,28 mg/100 g) and facheiro also of the family Cactaceae and also used as sources of water and food in the semi-arid region of northeast Brazil.

The regional custom of making cocadas, sweet desserts and raw sugar with the addition of the juice of sugar cane stems or industrially produced sugar (Saccharum officinarum L.) (Silva et. al. 2005) makes this species very important to these rural  populations in providing necessary calories, particularly in times of food shortage.

4.4. Randia armata

This species showed the most diversity in those bioactive constituents that according to Sakihama et. al. (2002), are important food constituents which combat free radicals and play an active metabolic role in maintaining human health. The main characteristic of this species was the high level of total phenolics (Table 3), much higher than that found in species widely used as food in Brazil, such and the native and endemic araçá (Psidium cattleianum Sabine = 4,439 mg EAG/2129 g), recorded as rich in phenolic compounds by Mccook-Russell et. al. (2012). According to Simões et. al. (2003), phenolics are compounds which contribute to the flavour, odour and colour of plant organs, including the fruits, making them more attractive to their consumers, and species with these organoleptic qualities have the potential for use as natural food flavouring substances and colorants. These properties make the species interesting for the food industry.

Among the species analysed in this study, R. armata shows the highest level of total carotenoids, expressed as β-carotene (Table 3). According to Pereira et. al. (2013), this carotenoid is considered the main precursor of vitamin A and the second-most effective anti-oxidant in sequestering free radicals after lycopene. According to Liu et. al. (2013), the presence of β-carotene gives the fruit their attractive yellow colour which attracts their consumers and this is a characteristic which favours the acceptance of a food product in the market.

In addition to this, R. armata had the second highest flavonoid level (Table 3), only lower than Psidium schenckianuma (7.90 mg/100 g), Sideroxylon obtusifolium (47.21 mg/100 g) and Tacinga inamoena (25.53 mg/100 g) when compared to the caatinga species analysed by Nascimento et. al. (2011, 2012). It has higher ascorbic acid (vitamin C) (Table 3) than the much appreciated Caryocar brasiliensis raw = 8,3 mg/100 g (TACO, 2011), and is classified as acid (Table 2) according to Baruffaldi e Oliveira (1998), which is a desirable characteristic for fruits from the viewpoint of the food industry.

4.5. Pouteria macrophylla

This species showed the greatest effect in inhibiting free radicals of ABTS•+ and DPPH· in the in vitro screening(Tables 4, 5), with TEAC to all the species analysed by Almeida et. al. (2011). In regard to EC50, P. macrophylla was observed to be richer in anti-oxidants than the species analysed by Rufino et. al. (2010).

This species was also notable for the level of total dissolved solids (Table 2), which are components that are important in fruits because they influence thermophysical, chemical and biological properties that determine the period for harvesting and storage and influence the degree of sweetness (Costa et. al., 2004). These factors are important for palatability and acceptance of the fruits as a food to be eaten raw, as for example a table fruit (Gomes, Vale, 2013). The content in °Brix was higher than that of fruits of caatinga species analysed by Nascimento et. al. (2011). It was also higher than the values cited by Souza et. al. (2012) for fruits of the Brazilian cerrado.

According to Chitarra, Chitarra (2005), species with high levels of total dissolved solids have potential for the manufacture of juices, wines, preserves and sweet desserts. According to Quave, Pieroni (2014), this characteristic allows the diversification of use types and an improved exploitation of wild foods.

This species also has a considerable calorific value (Table 3), sufficient for 10.03% of the daily requirement for a child of 10 years and 6.47% of an adult (FAO, 2001). This species can thus be considered an important food resource to complement daily calorie needs in the Brazilian semi-arid northeast region.

4.6. Swartzia flaemingii

This species showed the highest pH value (Table 2) among those analysed in this study, and is classified as barely acid according to Baruffaldi, Oliveira (1998). The pH value found can be considered high for wild fruits since it is higher than all the values reported by Nascimento et. al. (2011, 2012) for wild species. According to Mcglynn (1992), fruits with high levels of pH require additional care in pre-preparation, preparation and storage phases in order to guarantee the hygienic conditions necessary for food security.

Swartzia flaemingii is also notable for its total dissolved solids (TDS) (Table 2), having the second highest level in this study. Total titratable acid (TTA) was low and consequently its TSD/TTA value was the highest among the species analysed. This parameter is considered important for the determination of indices of quality and indispensable for the determination of the flavour and aroma of fruits (Gomes, Vale, 2013).

Although S. flaemingii cannot be considered a fruit of high calorific value (Table 3), according to the interviewees it formed part of the food diet in the past during periods of severe food shortage. It provides more calories (kcal/100 g) than many tropical fruits analysed by Silva et. al. (2008), such as araçá (Psidium araca Raddi = 37.09) and araticum (Annona crassiflora Mart = 90.47). In our opinion new research efforts are needed to investigate the food potential of this species.

5. Conclusions

By using analysis of nutritional and physico-chemical parameters and biological and in vitro anti-oxidant activity, this study has shown the potential of six food plant species currently used by resident populations of the semi-arid region of northeast Brazil as complements to the daily food diet of local families and for technological exploitation. This potential is even clearer when the results we obtained are compared with those of species studied by other authors.

The species B. laciniosa, H. martiana, P. macrophylla and S. flaemingii demonstrated great potential, especially in relation to high carbohydrate content and consequently calorie content, making them suitable as complements to the daily calorie requirement of these communities and for the development of technological food products. It was observed, for example, that H. martiana can readily make up for unavailability of the traditional feijão (Phaseolus vulgaris L.), which represents one of the main energy sources in the studied communities, and that B. laciniosa, available in the dry season, has more energy than the sweet potato (Ipomoea batatas L.), which is a much appreciated food in the region.

As regards bioactive compounds, we concluded that although some are only present in small quantities, for example ascorbic acid, these species should be considered important sources of these compounds since they are available in periods of food shortage (B. laciniosa, Melocactus zehntneri and S. flaemingii) or because they are habitually used as food in the region and are appreciated by adults and children (H. martiana, P. macrophylla and R. armata), often being the only sources of these substances accessible to these communities.

As regards in vitro anti-oxidant activity, it was observed that H. martiana and P. macrophylla showed notable capacity to reduce the free radicals ABTS•+ e DPPH·, suggesting that these species are rich in anti-oxidant substances and thus capable of defending the human body from damage caused by free radicals.

It was observed that all the species analysed in this study contained nutrients and bioactive compounds which justify their use for food by the studied communities and gave access to substances which are indispensable for healthy metabolism and health preservation. For this reason we conclude that they are species with the potential to be included in regional programs aimed at food and nutritional security and the prevention of the harmful effects of free radicals on human health.

We believe that this study provides a valuable addition to knowledge of the food potential of plants of the Brazilian semi-arid region which can effectively contribute to nutritional needs and health preservation of rural communities. However, it should be noted that other investigations are needed into the technological development of food products, more effective exploitation of these resources by lengthening their shelf life, and the addition of value to these regional foodstuffs.

References

AB'SÁBER, A. N. (1974); "O domínio morfoclimático semi-árido das caatingas brasileiras", Geomorfologia, 43, 1-39.

ALBUQUERQUE, U. P et. al. (2010); Caatinga: Biodiversidade e Qualidade de Vida. Bauru: Ed. NUPEEA, 113p.

ALBUQUERQUE, U. P. et. al. (2014); Methods and Techniques Used to Collect Ethnobiological. In: ALBUQUERQUE, U. P.; Cruz da Cunha, L. V. F.; Lucena, R. F. P. (Eds.). Methods and Techniques in Ethnobiology and Ethnoecology, New York: Springer Science, 480p.

AOAC. Official Methods of Analysis of AOAC International. AOAC International, Gaithersburg: Maryland, 2005.

ARAÚJO, F. R.; Lopes, M. A. (2011); "Diversity of use and local knowledge of palms (Arecaceae) in eastern Amazonia". Biodiversity and Conservation, 21(2), 487-501.

BARUFFALDI, R.; Oliveira, M. N. (1998); Fatores que condicionam a estabilidade de alimentos. In: Fundamentos de tecnologia de alimentos. São Paulo: Atheneu, p.13-25.

BENASSI, M. T.; Antunes, A. J. (1988); "A comparison of metaphosphoric and oxalic acids as extractants solutions for the determination of vitamin C in selected vegetables", Brazilian Archives of Biology and Technology, 31, 507-513.

BRAND-WILLIAMS, W.; Cuvilier, M. E.; Berset, C. (1995); "Use of a Free Radical Method to Evaluate Antioxidant Activity", Lebensmittel-Wissenschaft und-Technologie, 30, 25-30.

CELLI, G. B.; Pereira-Netto, A. B.; Beta, T. (2011); "Comparative analysis of total phenolic content, antioxidant activity, and flavonoids profile of fruits from two varieties of Brazilian cherry (Eugenia uniflora L.) throughout the fruit developmental stages", Food Research International, 44(8), 2442-2451.

CHITARRA, M. I. F.; Chitarra, A. B. (2005); Pós-colheita de frutos e hortaliças: filisologia e manuseio. Lavras: Ed. ESAL/FAEPE, 783p.

CORRÊA, L. C. et. al. (2011); "Antioxidant content in guava (Psidium guajava) and araçá (Psidium spp.) germplasm from different Brazilian regions", Plant Genetic Resources, 9(3), 384-391.

COSTA, W. S. et. al. (2004); "Influência da concentração de sólidos solúveis totais no sinal fotoacústico de polpa de manga", Revista Brasileira de Produtos Agroindustriais, 6(2), 141-147.

DE SOUZA, V. R. et. al. (2012); "Determination of bioactive compounds, antioxidant activity and chemical composition of Cerrado Brazilian fruits", Food Chemistry, 134(1), 381-386.

DIAS, L. S.; Luzia, D. M. M.; Jorge, N. (2013); "Physicochemical and bioactive properties of Hymenaea courbaril L . pulp and seed lipid fraction", Industrial Crops & Products, 49, 610-618.

FERNÁNDEZ, C. C. (2009); Plantas comestibles de centroamérica. Santo Domingo de Heredia, Costa Rica: Ed. Instituto Nacional de Biodiversidad, 360p.

FRANCIS, F. J. (1982); Analysis of anthocyanins, in: MARKAKIS, P. (Ed.). Anthocyanins as Food Colors. London: Ed. Academic Press, p.181-206.

GERALDO, R. R. C. et. al. (2003); "Distribuição da hipovitaminose A no Brasil nas últimas quatro décadas : ingestão alimentar , sinais clínicos e dados bioquímicos", Revista de Nutrição de Campinas, 169(4), 443-460.

GIULIETTI, A. N. A. M. et. al. (2005); "Biodiversidade e conservação das plantas no Brasil", Megadiversidade, 1(1), 52-61.

GOMES, S. O.; Vale, E. M. (2013); "Quality assessment of physics and chemistry of cajuí (Anacardium spp.) Mid-North Region", Revista Geintec, 3(3), 139-145.

GUINAND, Y.; Lemessa, D. (2001); The Potential of Indigenous Wild Foods. Diani, Kenya: Ed. USAID/OFDA,  46p.

GUNDUC, N.; El, S. N. (2003); "Assessing antioxidant activities of phenolic compounds of common Turkish food and drinks on in vitro low-density lipoprotein oxidation", Journal of food science, 68(8), 2591-2595.

HEUBACH, K. et. al. (2011); "The economic importance of non-timber forest products (NTFPs) for livelihood maintenance of rural west African communities: a case study from northern Benin". Ecological Economics, 70(11), 1991-2001.

IAL. Instituto Adolfo Lutz. (2008); Métodos físico-químicos para Análise de Alimentos. São Paulo: Ed. IAL, 1020p.

IBGE (Instituto Brasileiro de Geografia e Estatística). Divisão Territorial do Brasil e Limites Territoriais 2008. [Acesso em 08 jul 2014]. Disponível em: http:www//geoftp.ibge.gov.br/organizacao_territorial/divisao_territorial/apresentacao/divisao_territorial_do_brasil_e_limites_territoriais2.pdf.

IBGE (Instituto Brasileiro de Geografia e Estatística). Cidades 2010.  [Acesso em 10 jul 2014]. Disponível em: http://www.ibge.gov.br/cidadesat/topwindow.htm.

KUSKOSKI, E. M. et. al. (2006); "Frutos tropicais silvestres e polpas de frutas congeladas: atividade antioxidante, polifenóis e antocianinas", Ciência Rural, 36(4), 1283-1287.

LIU, F.-X. et. al. (2013); "Physico-chemical and antioxidant properties of four mango (Mangifera indica L.) cultivars in China", Food chemistry, 138(1), 396-405.

MCCOOK-RUSSELL, K. P. et. al. (2012); "Nutritional and nutraceutical comparison of Jamaican Psidium cattleianum (strawberry guava) and Psidium guajava (common guava) fruits", Food chemistry, 134(2), 1069-1073.

MCGLYNN, W. (1992); The Importance of Food pH in Commercial Canning. Oklahoma: Ed. Division of Agricultura Sciences and Natural Resources, 8p.

MEDEIROS, R. M. (2004); Estudo Agrometeorológico para o Estado do Piauí. Teresina: Ed. SEMA,  119p.

MENDES, B. V. (1994); Uso e conservação da biodiversidade no semi-árido. Brasilia: Ed. Ministério da Integração Nacioanal, 151p.

MERRILL, A. L.; Watt, B.K. (1973); Energy Value of Foods: Basis and Derivation. United States of America: Ed. Department of Agriculture, 74p.

MESQUITA, F. R. et. al. (2007); "Linhagens de feijão (Phaseolus vulgaris L.): composição química e digestibilidade protéica", Ceência Agrotecnica, 31(4), 1114-1121.

MORAIS, F. F.; Silva, C. J. (2011); "Etnoecologia de plantas nativas na comunidade de Estirão Comprido, Pantanal Matogrossense, Brasil", Revista de Ciências Agro-Ambientais, 9(1), 13-30.

NASCIMENTO, V. T. et. al. (2013); "Knowledge and use of wild food plants in areas of dry seasonal forests in Brazil", Ecology of Food and Nutrition, 52(4), 317-343.

NASCIMENTO, V. T. et. al. (2011); "Chemical characterization of native wild plants of dry seasonal forests of the semi-arid region of northeastern Brazil", Food Research International, 44(7), 2112-2119.

NASCIMENTO, V. T. et. al. (2012); "Famine Foods of Brazil's Seasonal Dry Forests: Ethnobotanical and Nutritional Aspects", Economic Botany, 66(1), 22-34.

PEREIRA, M. C. et. al. (2013); "Characterization, bioactive compounds and antioxidant potential of three Brazilian fruits", Journal of Food Composition and Analysis, 29(1), 19-24.

POTT, A.; Pott, V. J.; Bueno Sobrinho, A. A. (2004); Plantas úteis à sobrevivencia no Pantanal. Corumbá: Ed. SIMPAN, 73p.

QUAVE, C. L.; Pieroni, A. (2014); "Fermented foods for food security and food sovereignty in the Balkans: a case study of the Gorani People of Northeastern Albania", BioOne Research Evolved, 34, 28-43.

RAPOPORT, E.; Ladio, A. (1999); "El uso de plantas silvestres comestibles en una población suburbana del noroeste de la Patagonia". Parodiana, 11(1), 49-62.

RE, R. et. al. (1999); "Antioxidant activity applying an improved ABTS radical caion descolarization assay", Free Radical Biology & Medicine, 26, 1231-1237.

RODRIGUEZ-AMAYA, D. B. (2001); A Guide to Carotenoid Analysis in Foods. Washington: ILST Press, 64p.

ROESLER, R. et. al. (2008); "Antioxidant activity of Caryocar brasiliense (pequi) and characterization of components by electrospray ionization mass spectrometry", Food Chemistry, 110, 711-717.

RUFINO, M. D. S. M. et. al. (2010); "Bioactive compounds and antioxidant capacities of 18 non-traditional tropical fruits from Brazil", Food Chemistry, 121(40), 996-1002.

SAKIHAMA, Y. et. al. (2002); "Plant phenolic antioxidant and prooxidant activities: phenolics-induced oxidative damage mediated by metals in plants", Toxicology, 177, 67-80.

SANTO, F. S. E.; MacieL, J. R.; Siqueira Filho, J. A. (2012); "Impacto da herbivoria por caprinos sobre as populações naturais de Bromelia laciniosa Mart. ex. Schult. f. (Bromeliaceae)", Revista Árvore, 36(1) 143-149.

SILVA, A. S. et. al. (2005); "Avaliação da composição físico-química da coroa-de-frade". Revista de Biologia e Ciência da Terra, 5(2) 1-8.

SILVA, B. A. et. al. (2012); "Antioxidant capacity and phenolics of Pouteria macrophylla, an under-utilized fruit from Brazilian Amazon", European Food Research and Technology, 234, 761-768.

SILVA, M. R. et. al. (2008); Caracterização química de frutos nativos do cerrado, Ciência Rural, 38(6) 1790-1793.

SIMÕES, C. M. O. et. al. (2003); Farmacologia: da Planta ao Medicamento. Porto Alegre: Ed. UFRGS, 1102p.

SOUZA, V. R. et. al. (2012); "Determination of bioactive compounds, antioxidant activity and chemical composition of Cerrado Brazilian fruits", Food Chemistry, 134, 381-386.

SWAIN, T. et. al. (1959); "The phenolic constituents of Punnus domestica I-quantitative analysis of phenolic constituents", Journal of the Science of Food and Agriculture, 19, 63-68.

TACO. (2011); Tabela Brasileira de Composição de AlimentosCampinas: Ed. NEPA/UNICAMP, 2011. 162p.

1. Bióloga e Nutricionista, Profa. do Núcleo de Hospitalidade, Lazer e Produção Alimentícia, Instituto Federal de Educação, Ciência e Tecnologia do Piauí (IFPI)/Doutoranda em Desenvolvimento e Meio Ambiente, Universidade Federal do Piauí (DDMA/UFPI). E-mail: emfchaves@gmail.com;

2. Químico, Técnico do Laboratório de Análise de Alimentos do IFPI;

3. Nutricionista,  Prof. Dr. do Núcleo de Hospitalidade, Lazer e Produção Alimentícia do IFPI;

4. Biólogo, Prof. Dr. do Departamento de Biologia/Laboratório de Etnobiologia Aplicada e Teórica (LEA) da Universidade Federal Rural de Pernambuco (UFRPE);

5. Bióloga, Profa. Dra. do Departamento de Biologia/DDMA/Herbário Graziela Barroso da UFPI.