Antioxidant and prebiotic effects of a beverage composed by tropical fruits and yacon in alloxan-induced diabetic rats

The Onternational Scientific Association for Probiotics and Prebiotics (OSAPP) defines a prebiotic as “[...] a substrate that is selectively utilized by host microorganisms conferring a health benefit” (Gibson et al., 2017, p. 493). Besides they are naturally present in plant sources like chicory, onion and garlic (Farias et al., 2019; Mohanty et al., 2018; Quigley, 2019), the yacon (Smallanthus sonchifolius) – a Andean tuberous root – is one of the most abundant source of fructooligosaccharides (FDS) which exhibit prebiotic properties (Al-Sheraji et al., 2013).

Recently, studies have shown that the yacon possess different biological effects, among immunomodulation (Delgado et al., 2012), antimicrobial (Djansivu et al., 2011) and antioxidant (Sousa et al., 2015b;Campos et al., 2012). On addition, studies have reported that the ingestion of yacon reduces glycemia and increases the concentration of insulin in the plasma (Park et al., 2009;Aybar et al., 2001;Satoh et al., 2013) and reduces the glycemia of diabetic rats . Moreover, in clinical assays, the yacon have also to be associated with subjective improvements in satiety and reductions in post prandial glucose and insulin concentrations (Adriano et al., 2019;Silva et al., 2017;Kellow et al., 2014 -see also the references listed).
The American Diabetes Association (2010, p. S62) defines diabetes mellitus as "[...] a group of metabolic diseases characterized by hyperglycemia resulting from defects in insulin secretion, insulin action, or both". High blood glucose concentrations promote auto-oxidation of glucose to form free radicals. The generation of free radicals beyond the scavenging abilities of endogenous antioxidant defenses results in macro-and microvascular dysfunction and polyneuropathy (Bajaj & Khan, 2012). Cells have evolved highly complex enzymatic (e.g., catalase) and non-enzymatic antioxidant systems (e.g., ascorbic acid and polyphenols), which work synergistically, and in combination with each other, to protect the body against free radical-induced damage. There are several lines of evidence to suggest that antioxidant defenses may be lower in diabetes (Karunakaran & Park, 2013). On this sense, antioxidants are effective in reducing diabetic complications and may be beneficial either by ingestion of natural antioxidants (Bajaj & Khan, 2012).
Several tropical fruits have been reported in literature by their functional activities. Camu-camu (Myrciaria dubia McVaugh) is native to the Amazon region, and is considered the richest natural source of vitamin C in Brazil (Cunha-Santos et al., 2019;Chirinos et al., 2010), followed by acerola (Malpighia emarginata D.C.), a fruit recognized by their antioxidant properties due the phenolic and vitamin C composition (Belwal et al., 2018;Rosso & Mercadante, 2007). Cashew-apple (Anacardium occidentale L.) and yellow

Antioxidant and prebiotic effects of a beverage composed by tropical fruits and yacon
in alloxan-induced diabetic rats mombim (Spondias mombim L.) belongs to the Anacardiacea family, and are important fruits in the Northeast region of Brazil, rich in bioactive compounds and responsible for a sweet and sour taste (Rufino et al., 2010). Furthermore, acai (Euterpe oleracea Mart.) is one of the most popular fruit from Amazon and widely used in the world due to recognized antioxidant properties (Yamaguchi et al., 2015). Pereira et al. (2015) evaluated the synergistic, antagonistic and additive effects of different tropical fruits (camu-camu, cashew-apple, acerola, yellow mombim and acai), with aim to obtain a beverage with high levels of bioactive compounds. Furthermore, the beverage was evaluated in vivo, and shown to be effective in endogenous antioxidant defense of health Wistar rats, altering the catalase (CAT) and glutathione peroxidase (GPx) activities, and decreasing the lipid peroxidation in liver and serum in the animals . Moreover, the antimutagenic and antiproliferative activities were also demonstrated by Carvalho-  using health Swiss mice.
Considering the positive effects in the antioxidant status of this tropical fruit beverage in health models, and the importance of the antioxidants compounds and FDS for diabetes, the aim of this work was to evaluate the prebiotic and antioxidant effects of a beverage composed by tropical fruits (rich in antioxidant compounds) and yacon (source of FDS), in alloxan-induced diabetic rats.

Chemical and reagents
The reagents used were potassium persulphate from Acrós Drganics and ferrous sulphate from Vetec TM . HPLC grade water was prepared from distilled water using a Milli-Q system (Millipore Lab., Bedford, MA, USA). All other chemicals were purchased from Sigma Aldrich Canada Ltd. (Dakville, DN, Canada).

Beverage formulations
The blend formulation was obtained in a statistical planning as reported previously , composed by 10% acerola, 5% acai, 5% yellow-mombim, 5% cashew apple, 5% camu-camu, 20% pineapple and 50% water. However, in the present experiment, the water was replaced by yacon extract in different concentrations to provide the prebiotic characteristic of the beverage (due to its FDS constituents). For the prebiotic beverage, a two-level (2 2 ) central composite design (CCD) was employed to optimize the values of sensorial acceptance (dependent variable) using the concentration of yacon (50 to 70%) and concentration of sweetener (stevia, 0.01 to 0.10%) as independent variables. All the experiments were carried out in order to have a random selection method to minimize the effect of unexplained variability in the responses due to systematic errors. The analysis of variance (ANDVA) was applied to validate the model, and the regression coefficients were then used to generate response surfaces. The P < 0.05 was considered to be statistically significant. All results were performed using Statistica 7.0.

Sensorial analysis
The sensory evaluation of acceptance was carried out with fifty untrained panelists (63% female, 82% aged 18-45 years), as suggested by Meilgaard et al. (2006) in the laboratory test, using nine-point structured hedonic scales (1I: 'disliked extremely' to 9I: 'liked extremely') (Souza et al., 2019). The order of presentation of the samples followed balanced order. This sensory test procedure was approved by the Research Ethics Committee of the Ceará State University, under protocol number 11044529-5.

Prebiotic beverage
The optimized prebiotic beverage was prepared by mixing the blend formulation (50%), yacon extract (50%) and 0.07% of stevioside. An Armfield FT74 tubular heat exchanger unit was used to pasteurize the prebiotic beverage after preparation. The beverage, placed in a feed tank, was pumped through the heat exchanger to achieve the treatment condition (85 °C for 90 s). After heating, the samples were cooled in an ice/water bath, packed, and then stored under -18 ± 2 °C until needed for chemical analyses. For the in vivo assay, the beverage was lyophilized immediately after thermal processing and stored at -18 ± 2 °C prior to use.

Physico-chemical analyses
The color was performed in a Minolta Colorimeter (Model CR-400, Konica Minolta Sensing, Onc., Dsaka, Japan), with results based on three color coordinatesI: L* (whiteness or brightness/darkness), a* (redness/greenness), and b* (yellowness/blueness). The acidity (% citric acid) was determined according to the method of Onstituto Adolfo Lutz (2008), while the pH values were obtained using a pH meter (Hanna Onstruments, Romania) (Association of Dfficial Analytical Chemists, 2005). The soluble solids content (°Brix) was determined using a refractometer at 20.0 ± 0.5 °C, as recommended by Association of Dfficial Analytical Chemists (2005). All analyses were carried out in triplicate.

Microbiological analyses
The presence of total coliform and Escherichia coli in the prebiotic beverage was evaluated according to the Feng et al. (2013). Mesophilic aerobic, mold and yeast counts, and the safety microbial parameters Salmonella spp. according to the Andrews et al. (2016). Analyses were carried out according to the methodology described by FDA's Bacteriological Analytical Manual, and the results were expressed as colony-forming units per milliliter (cfu/mL) of product. These analyses were carried out in triplicate.

Total Antioxidant Capacity (TAC), Total Polyphenols (TP) and Fructooligosaccharides content (FOS)
The total antioxidant capacity was measured by the ABTS, FRAP and DPPH methods. For antioxidant extraction, the procedure developed by Larrauri et al. (1997) was used. The samples were extracted sequentially with 4 mL of methanol/water (50I:50, v/v) at 25 °C for 1 h, centrifuged at 25 °C, 400 g for 15 min, and the supernatant was recovered. Then, 4 mL of acetone/water (70I:30, v/v) were added to the residue at 25 °C, extracted for 60 min, and centrifuged at the same condition. Methanolic and acetonic extracts were combined, and added to 10 mL with distilled water.
The free radical scavenging activity was determined by the DPPH method (Brand-Williams et al., 1995); the ABTS·+ assay was based on a method developed by Miller et al. (1993) and, for the FRAP assay, the procedure described by Benzie & Strain (1996). All the methods were used with the modifications suggested by Rufino et al. (2010). The results of DPPH method were expressed as the concentration of antioxidant required to reduce the original amount of free radicals by 50% (EC 50 ) and the values were expressed as g prebiotic beverage per g of DPPH.
For the ABTS and FRAP assay, the results were expressed as µM Trolox and µM Fe 2 SD 4 /g of prebiotic beverage, respectively. Thus, the total polyphenols (TP) was determined by the Folin-Ciocalteu method (Dbanda et al., 1997) and the results were expressed as mg GAE (gallic acid equivalent)/100 g of prebiotic beverage. The fructooligosaccharides were determined as described by Horwitz et al. (2010), and the results expressed as g FDS/100 g of prebiotic beverage.

Animals and experimental design
The present animal study was carried out following the guidelines in the use of animals for experimental purpose after due approval from the Research Ethics Committee on the use of animals at the Federal University of Alfenas/UNOFAL-MG (Protocol no. 528/2013). Male rats weighing 90 ± 5 g were kept under standard laboratory conditions of temperature (22 ± 2 °C), relative humidity (52 ± 5%), and 12 h light-dark cycle. Diet and water were provided ad libitum. For diabetes induction, overnight fasted rats received intraperitoneal injection (i.p.) of alloxan (120 mg/kg), followed by an aqueous solution of 10% sucrose for 24 h. After four days, the glycemia was measured and all rats that presented glucose levels higher than 200 mg/dL were considered positive diabetic induced. The lyophilized prebiotic beverage and saline solution (NaCl 0.9%, for the control groups) were administered daily by gavage (0.5 mL/100 g body weight) for 30 days. Rats (n = 30, six animals for group) were divided into the following groupsI: (1) negative control (healthy animals receiving saline solution); (2) positive control (diabetic animals receiving saline solution); (3) diabetic animals receiving 100 mg of prebiotic beverage per kg body weight; (4) diabetic animals receiving 200 mg of prebiotic beverage per kg body weight and (5) diabetic animals receiving 400 mg of prebiotic beverage per kg body weight.
Nutritional parameters such as weight gain and food consumption were collected three times a week. The blood glucose levels were measured nine times. At the end of the experiment, rats were killed, and blood was collected in heparinised tubes (20 U heparin/mL blood) and centrifuged at 1900 g for 10 min. The separated plasma was stored at -80 °C until processed. An aliquot from caecal content was used for microbial analyses and the liver was perfused with saline solution (0.9% w/v), collected and immediately frozen at -80 °C.

Catalase activity
Catalase activity (CAT) was measured in liver tissue, and the methodology used is in according to Aebi (1984). Briefly, the activity was assayed at 25 °C by a method based on the disappearance of 10 mM H 2 D 2 . The decomposition of H 2 D 2 by CAT contained in the samples a first-order kinetic and changes in absorbance were measured after addition of H 2 D 2 , and then at 10 s intervals, in a total of 60 s.

Lactobacilli analysis of caecal material
To measure the prebiotic effect of the beverage, one gram of caecal material was transferred into a sterile tube and mixed with 9 mL of sterile saline phosphate solution (PBS, Sigma Aldrich) and then serially diluted (from 10 -1 to 10 -7 ) in a Rogosa Agar SL medium (Becton Dickinson, USA). Oncubation was performed at 35 °C under anaerobic conditions using the anaerobic jar with Anaerocult A (Merck, Darmstadt Germany). After 48 h incubation, the number of colonies was recorded as log 10 cfu/g of wet sample.

Statistical analyses
The results of the optimization of the beverage were analyzed using the Statistic 7.0 (P < 0.05). For the in vivo assays, the GraphPad Prism 4.0 for Windows (San Diego, CA, USA) was used. A one-way analysis of variance (ANDVA) and Tukey test (P < 0.05) were applied to the results.

Optimization of the beverage formulation
The Table 1 shows the results obtained for the sensorial acceptance for all formulations tested. The values obtained ranged of 4.5 to 7.0, which corresponded to "dislike slightly" to "like moderately". Considering the effect of each parameter tested (sweetener or yacon extract concentration, in %), only the sweetener presented a significant effect (P < 0.05), in both linear and quadratic parameters.
Analysis of variance for the dependent variables indicated that the response surface model developed was adequate (R 2 = 82.57). The coefficient of determination (R 2 ) value is quite high, indicating that a high proportion of variability was explained by the data and that the RSM model was adequate (Table 2). Although F-value was higher than Ftab (Fcalc = 18.9; Ftab = 4.46), the lack of fit, which measures the fitness of the model, resulted in a significant F-value. However, the pure error is approximately zero, and for this reason the Fcalc presents high value, and consequently, is higher than the Ftab (Fcalc = 44.4; Ftab = 19.33). Analysis of variance (ANDVA) was performed considering only the statistically significant (P < 0.05) variables, and a valid model was defined by Equation 1I: . . -.
According to Figure 1, the maximal sensorial acceptance of the beverage was obtained using 0.07% of sweetener. Considering that the yacon extract did not present statistical significance (P > 0.05) in the range tested (50 to 70%), the cost of yacon and the legislation (that allow the use of 50% of water), the condition defined by the authors as 50% of yacon extract and 0.07% of sweetener.
The total phenolic compounds (126.83 ± 9.48 mg GAE/100 g) and ascorbic acid contents (171.64 ± 7.31 mg/100 g) were also determined in the beverage. These values are in accordance with the reports of Pereira et al. (2015), using similar composition of tropical fruits, however, without yacon. The values obtained by the authors were 103.01 ± 5.96 mg GAE/100 g and 155.46 ± 2.07 mg/100 g for polyphenols and ascorbic acid, respectively. On addition, the total antioxidant activity of the prebiotic beverage were 10.57 ± 0.50 µM Trolox; 33.45 ± 0.41 µM Fe 2 SD 4 and 866.36± 25.30/g of prebiotic beverage for ABTS, FRAP and DPPH method, respectively. These values are considerable high, and show the antioxidant potential of the beverage, which contributed to their functional properties.
As mentioned before, the yacon were added into beverage due the high FDS content. Considering this, the FDS was measured in the product. The values obtained were 4.64 ± 0.13 of FDS/200 mL that represents a portion of beverage. This value is about two times higher than that required by Brazilian claim for products with functional properties (Brasil, 2019), considering a minimum of 2.5 g of FDS for portion of a prebiotic beverage. On this case, the permitted claim is "FDS as prebiotics contribute to a balance/equilibrium of the intestinal flora. Their consumption should be associated with a balanced diet and a healthy life-style".

In vivo study
Non-significant differences (P > 0.05) among the groups that received the prebiotic beverages and their effects on diet consumption when compared to positive control (diabetic animals who received saline solution) are presented in Figure 2. Besides the diet consumption of all diabetic animals was higher than the health animals (negative control), there are no statistical difference (P > 0.05) in the weight gain in all treatments studied (see Figure 2).
Several papers report the potential of yacon in the glycemia control for diabetic animals or humans (Park et al., 2009;Aybar et al., 2001;Dionisio et al., 2015;Kellow et al., 2014). Unfortunately, the beverage showed no statistical effect (P > 0.05)  on the glycemia of the diabetic animals in all concentrations tested (see Figure 3).
On the present experiment, was observed a decrease, although with no statistical significance (P > 0.05), in the catalase activity (CAT) in the liver of diabetic and non-diabetic animals. Moreover, the prebiotic beverage diet for 30 days caused a significant increase in the antioxidant enzyme activity in a dose-dependent manner. As mentioned before, the hyperglycemia is the source of most of the deleterious effects usually associated with Diabetes mellitus, and there is considerable evidence suggesting that chronic elevation of blood glucose concentrations promote auto-oxidation of glucose, and consequently leads to the generation of reactive oxygen species (RDS) (Karunakaran & Park, 2013;Ceretta et al., 2012). RDS scavenging abilities of endogenous antioxidant defenses -such as catalase, that can be can compensate by exogenous antioxidants (Bajaj & Khan, 2012). Considering this, we suggested prebiotic beverage antioxidants (phenolic, ascorbic acid and others) can protect the body against free radical-induced damage, being effective in reducing diabetic complications, as mentioned by Bajaj & Khan (2012) and Karunakaran & Park (2013).
The results from bacteriological analysis of caecal material are presented in Figure 3. Concentrations (log cfu/ g wet sample) of lactobacilli were significantly higher (P < 0.05) in samples obtained from animals treated with the prebiotic beverage compared to the positive (diabetic animals) or negative (health animals) control groups, in a dose-dependent manner. The prebiotic effect of yacon are supported by previous studies, showing that FDS from yacon is efficiently metabolized by lactobacilli in vitro Figure 2. Effect of 30 day administration of prebiotic beverage on alloxan-induced diabetic rat on the following parameters (A) body weight (g); (B) diet consumption (g) and (C) glucose (mg.dL -1 ).

Conclusion
On conclusion, the present investigation showed that the prebiotic beverage of yacon and tropical fruits possess considerable concentration of fructooligosaccharides and antioxidant compounds, that could promoted the increase the catalase activity on liver and growth of lactobacilli in the caecal material of alloxan-induced diabetic rats.