Chemical composition and antioxidant capacity of coffee pulp

Coffee pulp is produced in large quantities, and its disposal can pollute the environment. However, the nutritional value and antioxidant content of coffee pulp make it a good option for animal feed. Therefore, the objective of this study was to determine the chemical composition, the phenolic compounds and the antioxidant capacity of coffee pulp using fresh (FCP), ensiled (ECP), and ensiled and sun-dried (EDCP) coffee pulp. The study design was completely randomized with three treatments (n=4). Dry matter (DM), crude protein (CP), ash, acid detergent fiber (ADF), neutral detergent fiber (NDF), lignin, phenolic compounds and antioxidant capacity were determined. Data were analyzed by analysis of variance, and means were compared with the Tukey test. The percentage of CP, NDF and ADF was higher in ECP and EDCP than in FCP. There were no changes in lignin content. Ensiling and sun drying did not decrease (P>0.05) caffeine or tannins. No differences were found in caffeic acid (2.031±2.873, 5.103±0.391, 4.913±0.018 mg g-1 DM in FCP, ECP, EDCP, respectively). The concentration of chlorogenic acid (FCP 2.593±1.756; ECP 5.368±0.422; EDCP 4.875±0.678 mg g-1) increased with the ensiling process, but it was not affected by sun drying. Ethanol decreased in ECP and EDCP relative to FPC (FPC 15.88±0.97%; ECP 7.04±1.16%; EDCP 0.00%); however, antioxidant capacity was not affected. It was concluded that the ensiling process and sun drying affected the nutritional value of coffee pulp but not its antioxidant capacity.


INTRODUCTION
After harvest, coffee berries need to be processed before their use and commercialization.This process can be done using two methods, i.e., the dry and wet methods (Berecha;Lemmessa;Wakjira, 2011).In the wet industrial process, coffee pulp is the main by-product and represents 55% of the whole berry (Murthy;Naidu, 2012).Ripe coffee fruits are harvested, and the skin and pulp are removed from the coffee beans (Duangjai et al., 2016).Most of the coffee pulp generated during wet processing is deposited directly into huge waste disposal sites or river streams without undergoing any treatment (Geremu;Bekele;Sualeh, 2016).The coffee pulp has been used to produce bioethanol (Menezes et al., 2014), biogas (Corro et al., 2013), compost (Bernas, 2011), substrate for mushrooms (Salmones;Mata;Waliszewki, 2005) and animal feed (Salinas-Rios et al., 2015).As a byproduct, coffee pulp contains associated bioactive compounds, such as polyphenols with good antioxidant properties, which endow additional health benefits (Geremu;Bekele;Sualeh, 2016), and anthocyanins that have potential applications as natural food colorants (Murthy;Naidu, 2012).The polyphenols can prevent or decrease the peroxidation of fatty acids, reduce oxidative stress in animals at critical physiological stages (Salinas et al., 2014), and increase the shelf life of animal meat when it is included in their diets (Salinas-Ríos et al., 2014).Ensiling coffee pulp is a viable process, and some chemical composition studies have showed that, compared with fresh coffee pulp, silage has higher crude protein content and lower values of nitrogenfree extracts and tannin; additionally, the antioxidant capacity of coffee pulp remains unchanged during the ensiling process (Salazar et al., 2009;Salinas et al., 2014).
The objective of this study was to assess the chemical characteristics, presence of antioxidant compounds and antioxidant capacity of coffee pulp, including fresh pulp, pulp ensiled with 5% molasses, and ensiled and sundried pulp.

Location of the study area and establishment of the silo
Coffee pulp was obtained from a de-pulping plant that uses the wet method to remove the berry from the coffee (Coffea arabica) bean.The de-pulping plant is in the municipality of Huatusco, Veracruz, Mexico (19º 09' N and 96º 58' W, 1344 m altitude).The coffee pulp was drained for 12 h to eliminate any water it absorbed during the depulping process (fresh coffee pulp, T1: FCP).Afterwards, the pulp was then placed in four plastic 1100-L capacity containers, and 5% molasses was added for a 140-day fermentation process to obtain ensiled coffee pulp (T2: ECP).Immediately, part of the ensiled coffee pulp was sun dried for eight days to reduce water content (ensiled sundried coffee pulp, T3: EDCP).
At ensiling, samples were collected from each container in a similar manner.When the silos were opened, samples were taken from the upper, middle and lower parts and mixed to obtain a single sample for future analysis.
Analyses performed on fresh, ensiled, and ensiled and sundried coffee pulp FCP, ECP, and EDCP were analyzed for dry matter (DM), crude protein (CP), ether extract (EE) and ash (AOAC, 1990).Additionally, coffee pulp samples were analyzed for neutral detergent fiber (NDF), acid detergent fiber (ADF) and lignin (Van Soest;Robertson;Lewis, 1991).Ethanol content was determined following the technique of Davies and Chace (1969).Five mL of HPLC grade H 2 O was added to 200 mg of the sample from each of the treatments (i.e., FCP, ECP, and EDCP) to quantify the phenolic acids and caffeine.The mixture was shaken in a vortex and spent 30 min in a shaker (Heidolph), and then the sample was centrifuged at 4500 rpm for 10 min.Two milliliters of the supernatant was filtered through a nylon acrodisc filter (0.45 µm diameter) and deposited into amber vials (2 mL).Phenolic acids were determined using a high precision liquid chromatograph (Agilent Technologies, Model 1100 with a photodiode array detector and a Nucleosil column Model Nautilus <18 100-5, 125 × 4 mm), with HPLC grade H 2 O in channel A; furthermore, the pH was 2.5 after being adjusted using trifluoroacetic acid and acetonitrile in channel B. The gradient was as follows: min 10, A: 85%, B: 15%; min 20, A: 65%, B: 35%; and min 23, A: 65%, B: 35%.The flow rate was 1 mL per min at 30ºC, 20 µL of the sample was injected, and the reading was taken at 280 nm.The phenolic acid standards were Sigma (e.g.,gallic,chlorogenic,syringic,vanillic,2,caffeic,2,. Caffeine was quantified by isocratic analyses using water and HPLC grade acetonitrile in a 75:25 proportion with a Nucleosil column (similar to that used for phenolic acid determination), into which 10 µL of the sample was injected at a flow rate of 0.8 mL per min at 25 ºC, and the reading occurred at 273 nm.Caffeine (Merk) was used at concentrations between 2 and 12 µg to generate a standard curve with 5 points.
Tannins were analyzed in three phases: 1) 10 mL of 70% acetone was added to a 25 mg sample of coffee pulp, which was placed in an ultrasound bath (Brasonic model 220) for 10 min, spent 5 min in repose and underwent 10 min of sonification.Samples were immediately centrifuged at 5000 rpm for 10 min.The supernatant was collected and put on ice.2) 1 mL of distilled H 2 O was added to 100 mg polyvinyl pyrrolidone (PVPP) and 1 mL of the extract with tannins.This mixture was shaken and stored at 4 ºC for 15 min, shaken for 1 min and centrifuged at 3000 rpm for 5 min.The supernatant was collected and put on ice.3) A standard curve was obtained using 0, 2, 4, 6, 8 and 10 μg solutions of tannic acid prepared from a mother solution (0.1 mg mL -1 ), distilled H 2 O (i.e., 500, 480, 460, 440, 420, and 400 µL), folin reagent (i.e., 250 µL) and 20% sodium carbonate (1.250 mL).The mixture was measured using a spectrophotometer (Agilent Technologies Model 1100) at an absorbance wavelength of 725 nm (Makkar, 2003).

Antioxidant capacity
Samples of extracts from FCP, ECP and EDCP were obtained.With these pulp samples, the antioxidant capacity was determined by the ferric reducing antioxidant power (FRAP) method (Benzie;Strain, 1999).

Statistical analysis
The experimental design was completely randomized with three treatments (n=4).The procedure PROC GLM (SAS, Cary, NC, USA, 2002) was used, and means were compared with the Tukey test (P<0.05).All variables were subjected to a normality test using the Univariate procedure.

Chemical composition
Table 1 shows the chemical composition of the treatments.Dry matter content was affected (P<0.05) by the ensiling and drying processes, and the lowest value was observed in ECP.The DM value (22.95±0.49%) in FCP was similar to the value of 23.20% reported by another study (Elias, 1978), as was the value found in ECP (18.85±0.96%),which was close to the 15.40% obtained in a previous study (Bautista et al., 2005).
The concentration of crude protein (CP) tended to increase (P<0.05) in FCP (10.85±0.17%),ECP (13.10±0.26%)and EDCP (13.24±0.13%);however, ECP and EDCP were not significantly different (P>0.05).Similar results have been reported, with increments in CP from FCP (9.12%) to ECP (13.14%) and from 11.60% in FCP to 12.00% in ECP (Cardona et al., 2002).The ensiling process increased the percentage of CP, probably due to the associated decrease in soluble carbohydrates.Protein content in the pulp was not affected by sun drying.
The values of NDF and ADF increased (P<0.05) in EDCP, although there were no differences (P>0.05) in NDF between FCP and ECP.ADF, however, was affected (P<0.05) by the ensiling process.This increase might have been caused by the soluble carbohydrates that mixed with the effluents produced during the ensiling process; these soluble carbohydrates were then lost when the ensiled coffee pulp was sun dried.Therefore, the fiber content increased, and the percentages of NDF, ADF, and lignin were higher (Salinas et al., 2014).
Although the percentage of lignin was similar (P>0.05) in FCP (31.51±0.24%),ECP (30.97%±0.94)and EDCP (31.26±0.98%), the resulting values were higher than those published by other authors (Oliveira et al., 2007).The high content of lignin in the silage of this experiment may have been due to the presence of coffee plant residues in the silage, the age at harvest, the environmental conditions and the variety of coffee used.Ash content had significantly different (P<0.05)values among FCP (7.40±0.01%),ECP (8.76±0.32%),and EDCP (10.82±0.41%).The increase could be due to soil contamination where the coffee pulp was sun dried or due to the addition of molasses (Salinas-Rios et al., 2014).Nevertheless, these values coincided with those obtained by other authors (Moreau et al., 2003;Cardona et al., 2002), i.e., 7.73 and 7.90%, respectively, in FCP and 9.55% and 10.80%, respectively, in ECP; furthermore, these values were higher than the value obtained for sundried coffee pulp, 6.60% (Figueroa;Mendoza, 2010).

Caffeine, tannins and ethanol
Table 2 shows the results for caffeine, tannins, and ethanol.An increasing trend in caffeine content was observed in FCP, ECP,and EDCP (18.60±0.31,25.49±2.88 and 30.30±8.23 mg g -1 DM, respectively); however, the values were not significantly different (P>0.05).The ensiling and drying processes did not modify the initial levels of caffeine.However, even though caffeine is a factor that may limit the use of coffee pulp as animal feed, no adverse effects have been observed in productive ruminant variables that were fed fresh or dehydrated coffee pulp that were provided in different proportions of the diet, e.g., (Salinas-Rios et al., 2015) reported 0.05% of caffeine in ensiled coffee pulp and studied diets that included 8% and 16% of coffee pulp.
Tannin content was similar (P>0.05)among the three treatments, although there was an increase from FCP to ECP and a decrease in EDCP (3.50±0.016,4.49±0.023and 1.18±0.984mg g -1 DM, respectively).The lack of differences may have been because the fermentation process in the silo did not modify the tannins; thus, their concentrations were maintained (Noriega;Silva;García, 2009).
Differences (P<0.05) were observed in the quantity of ethanol in FCP and ECP (15.88±0.97 and 7.04±1.16g 100 -1 DM, respectively), but no ethanol was detected in EDCP.The ethanol present in ECP was higher than that reported by authors who asserted that they obtained between 2 and 2.5 g ethanol after fermenting fresh coffee pulp for 48 h at 28 ºC (Murthy;Naidu, 2010).The ethanol measured in our study was within the acceptable range of 10 to 30 g kg -1 DM for silage (Kung;Shaver, 2001).Yeasts, gram-negative and gram-positive bacteria, and filamentous fungi were present in high numbers during the different stages of the de-pulping process (Schwan;Wheals, 2003).During coffee fermentation, some pectinolytic microorganisms were associated with the degradation of the pulp and mucilage (which is rich in polysaccharides), producing alcohols and acids and other metabolic compounds (Reis et al., 2014), which might have contributed to the production of ethanol in FCP as FCP: fresh coffee pulp, ECP: coffee pulp ensiled with 5% molasses, EDCP: coffee pulp ensiled and sun dried a, b: different letters in the same row indicate significant differences.SEM: standard error of the mean a consequence of the time elapsed from the de-pulping process and the ensiling of coffee pulp (i.e., 18 hours).

Phenolic acids
Eight phenolic acids were identified (Table 2).Ferulic, caffeic and chlorogenic acids were found in greater proportions, and they increased during the ensiling process but decreased when silage was sun dried.Differences (P<0.05) were found only for chlorogenic acid; although, the amount of chlorogenic acid was similar in ECP and EDCP (P>0.05).Syringic, vanillic and p-cumaric acids were found in lower concentrations.Only chlorogenic, p-hydroxybenzoic, syringic and gallic acids were different (P<0.05).
Chlorogenic acid was not the most abundant, as has been reported in previous studies (Murthy;Naidu, 2012).The quantity of chlorogenic acid varies with the degree of maturation (Aerts;Baumann, 1994), the species and the other factors associated with coffee quality, such as the altitude and the presence or absence of shade, as well as resistance to some diseases (Humphrey;Macrae, 1987).However, chlorogenic acid has been identified as a potent inhibitor of lipid peroxidation in vitro and has shown possible anticarcinogenic activity in vivo (Kasai et al., 2000); furthermore, chlorogenic acid was one of the three most abundant phenolic acids in FCP, ECP, and EDCP.

Antioxidant capacity
Antioxidant capacity was not affected (P>0.05) by ensiling or sun drying (FCP: 2769.9±88.61;ECP: 2594.7±100.82;EDCP: 2486.3±306.16nmol Trolox -1 mL) (Table 2), which supports the findings reported in previous studies (Salinas et al., 2014) for fresh and ensiled coffee pulp.The antioxidant capacity of fermented coffee pulp extracts can be attributed to the presence of phenolic substances (Palomino et al., 2015).However, the mechanisms that affect antioxidant activity during fermentation are extremely varied, and studies have shown that the ensiling process may cause considerable reductions in polyphenol concentrations (Weinberg et al., 2008).Our study found no differences in certain phenolic acids, which may have derived from a microbial hydrolysis reaction that increased the quantity of phenolic compounds; moreover, fermentation induces the structural breakdown of plant cell walls, leading to the liberation or synthesis of various antioxidant compounds (Hur et al., 2014).

CONCLUSIONS
Ensiling increased the percentages of CP, ADF, NDF and ash without affecting the content of lignin and EE.The highest concentrations of CP, ADF, NDF, ash and lignin were observed in the ensiled sun-dried coffee pulp.The p-hydroxybenzoic and syringic acids decreased with the ensiling process and with sun drying; however, chlorogenic acid increased.The gallic acid increased with fermentation, but it was imperceptible in the sun-dried pulp.The ferulic, caffeic, vanillic and p-cumaric acids were not affected.The antioxidant capacity did not vary among the fresh, ensiled or dried ensiled pulp.

Table 1 :
Chemical composition of coffee pulp.
FCP: fresh coffee pulp, ECP: coffee pulp ensiled with 5% molasses, EDCP: coffee pulp ensiled and sun dried.ND: not determined .a,b,c Different letters in the same row indicate significant differences.SEM: standard error of the mean.

Table 2 :
Caffeine, tannin, antioxidant compounds and antioxidant capacity (FRAP) in the coffee pulp.