Functional and technological potential of arabica coffee oils

The purpose of the study was to evaluate the antioxidant activity, rheological behaviour, oxidative stability, and antibacterial potential of coffee oils (Coffea arabica L.). The extraction process took place from green and roasted beans, by cold pressing, and filtration via filtering card. The experimental design consisted of five treatments: R100 (100% roasted oil); R75G25 (75% roasted oil and 25% green oil); R50G50 (50% roasted oil and 50% green oil); R25G75 (25% roasted oil and 75% green oil), and G100 (100% green oil). The treatment R75G25 showed a higher content of total phenolic compounds and higher DPPH and ABTS Research, Society and Development, v. 9, n. 9, e700997702, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i9.7702 3 anti-radical efficiency. Regarding rheological behaviour, all coffee oils can be characterized as Newtonian fluids because the shear stress and strain rate varied linearly, with the line intersecting at zero. The treatments R75G25 and R25G75 showed a longer oxidation induction time, in the oxidative stability analysis, in addition to antibacterial activity having been verified for all oil samples. Besides that, the investigated green and roasted coffee oils are sources of fatty acids, including from the omega 3, 6, and 9 classes. Therefore, the use of arabica coffee oil mixtures as natural preservatives in food can be considered a promising alternative for the partial replacement of chemical additives in food matrices and in cosmetic formulations allowing the development of innovative products.

anti-radical efficiency. Regarding rheological behaviour, all coffee oils can be characterized as Newtonian fluids because the shear stress and strain rate varied linearly, with the line intersecting at zero. The treatments R75G25 and R25G75 showed a longer oxidation induction time, in the oxidative stability analysis, in addition to antibacterial activity having been verified for all oil samples. Besides that, the investigated green and roasted coffee oils are sources of fatty acids, including from the omega 3, 6, and 9 classes. Therefore, the use of arabica coffee oil mixtures as natural preservatives in food can be considered a promising alternative for the partial replacement of chemical additives in food matrices and in cosmetic formulations allowing the development of innovative products. Keywords: Arabica coffee; Fluids; Industry; Healthiness; Fatty acid profile.

Introduction
The increase in the world population has driven the demand for improving the quality, quantity, and diversity of oil sources (Prates-Valério, Celayeta & Cren, 2019). This led to the popularization of the consumption of vegetable oils associated with the search and development of new sources (Senger et al., 2017) aiming its beneficial effects on human health and high consumer acceptance (Vieira, Souza, Rodrigues & Sousa, 2019). In this context, studies with bioactive lipids from easily available plant sources have been conducted in order to analyse their quality, stability, and nutritional characteristics (Hashempour-baltork, Torbati, Azadmard-damirchi & Savage, 2016;Ramadan & Wahdan, 2012).
The use of mixtures of vegetable oils can be suggested as a simple way to improve nutritional properties and oxidative stability (Hashempour-baltork et al., 2016). Combinations of canola and palm oils showed that tocopherols and tocotrienols were comparatively more stable in the mixture when compared to pure oils, besides, different antioxidant compounds can synergize and enhance the nutritional and technological effects (Mba, Dumont & Ngadi, 2017). The study of Ramadan and Wahdan (2012) presented the functional, stability, and antiradical DPPH · activity effects in mixtures of corn oils with black cumin and coriander and predicted the commercial exploitation of mixtures of corn and coriander oils. In this sense, it is proposed that the mixture of coffee oils from green and roasted beans is a promising alternative to meet the demands of some industrial segments.
Also, sterols, tocopherols, and diterpenes received broader attention due to their anticarcinogenic properties  and protection against aflatoxin B1-induced genotoxicity (Huber, Scharf, Peter & Schulte-hermann, 2002). Considering the presence of pentacyclic diterpenes, caveol, and cafestol in green and roasted coffee beans as fatty acid esters, these bioactivities can also be expected in this food (Kurzrock & Speer, 2001).
For robusta coffee, the lipids levels are in the range of 9.0-13.0% and 11.0-16.0% in green and roasted beans, respectively (Lago, 2001). On the other hand, arabica coffee has better quality when compared to robusta coffee. In this sense, the objective of this study was to evaluate, in an unprecedented way, the antioxidant activity, the rheological behaviour, the oxidative stability, and the antibacterial potential of coffee oils (Coffea arabica L.) from green and roasted beans. Research, Society and Development, v. 9, n. 9, e700997702, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i9.7702 6

Coffee oils preparation and experimental design
The oils of green and roasted coffee were provided by Cooperativa Regional de Cafeicultores em Guaxupé -COOXUPÉ, located in Minas Gerais, Brazil. The process of preparation of the oil samples occurred from the green and roasted Coffea arabica L. grains by cold pressing and subsequent filtration through the filter cart. The samples were stored in commercial oil packaging until the moment of analysis. The experiment was conducted in a completely randomized design -CRD, as shown in Table 1.

Determination of the fatty acid profile
The esterification of fatty acids was carried out according to the methodology described by Hartman and Lago (1973). Subsequently, the fatty acid profile was determined by gas chromatography on a Shimatzu CG 2010 chromatograph (Agilent Technologies Inc., Palo Alto, CA, USA), equipped with flame ionization detector, Split injection with a 1:100 ratio and SPTM-2560 Supelco capillary column, 100 m  0.25 mm  0.20 μm (Supelco Inc., Bellefonte, PA, USA). The initial temperature of the column was 140 °C, maintained for 5 min, changing to 240 °C, maintained for 30 min, for a total of 60 min. The injector and detector were kept at a temperature of 260 °C and helium was used as the carrier gas. After the analysis was completed, the peaks obtained were integrated.

Total phenolic content and antioxidant activity analyses
The coffee oils extraction was realized according to the methodology described by Larrauri, Rupérez, and Saura-Calixto (1997) with adaptations. About 1 g of the sample from each treatment was homogenized with 80 mL of methanol 50%. After an hour, in a dark environment, the tubes were centrifuged for 15 minutes at 21,952 g. Then, after a filtration step was performed, the volume was completed to 100 mL and the extracts were stored in a freezer (-18 ºC) until analyses.
The total phenolic content of each treatment was evaluated using the Folin Ciocalteau methodology, according to Waterhouse (2003). Aliquots of 0.5 mL of the sample extracts were added in test tubes, in combination with 2.5 mL of the Folin Ciocalteau reagent 10% and 2 mL of sodium carbonate 4%, with readings taken at 750 nm after two hours incubation. The standard used was gallic acid and the results were expressed in mg gallic acid equivalent (GAE) 100 g -1 sample.
For the determination of antioxidant activity by the DPPH method, the methodology described by Rufino et al. (2007a) was employed. The results were expressed as EC50 in g of sample per g of DPPH. Meanwhile, the determination of antioxidant activity by the ABTS + method was carried out according to the methodology described by Rufino et al. (2007b), and the results were expressed in µM Trolox equivalent (TE) g -1 sample.

Rheological behavior
The determination of the viscosity of the treatments containing different proportions of green and roasted coffee oil (Table 1)  Flow curves were applied to the fluids through three continuous ramps (increasing, decreasing, and increasing ramp), with the shear rate varying between 0.1 to 300 s -1 , for 2 min Research, Society and Development, v. 9, n. 9, e700997702, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i9.7702 8 for each curve (Mitschka, 1982). Newtonian model (Equation A) was adjusted to the rheological data obtained from the second rising ramp, which represents steady-state flow.

(Equation A)
where τ is the shear stress (Pa); is the shear rate (s -1 ); µ is the Newtonian viscosity (Pa s).
The model was adjusted using the software Statistical Analysis System -SAS ® University Edition.

Oxidative stability
The oxidative stability of each treatment was estimated by measuring the time of

Antibacterial activity by disc-diffusion method
The strains Staphylococcus aureus ATCC 25923, Escherichia coli CDC 055, Salmonella enteritidis S64, Cronobacter sakazakii ATCC 29004, and Listeria monocytogenes ATCC 19117 were kindly provided by the Enterobacteria Laboratory -LABENT of the Osvaldo Cruz Foundation -FIOCRUZ, Rio de Janeiro, Brazil. The stock cultures were stored in freezing medium (15 mL of glycerol; 0.5 g of bacteriological peptone; 0.3 g of yeast extract; 0.5 g of NaCl; 100 mL of distilled water, and pH 7.0).
The cultures were thawed and reactivated by inoculating 50 µL aliquots in tubes containing 10 mL of BHI broth (Brain Heart Infusion -Himedia, India) and incubated at 37 ºC for 24 hours. The inoculum was standardized according to growth curves previously determined for a population of 1.0 x 10 8 CFU mL -1 using spectrophotometric measurements (600 nm).
One hundred µL aliquots of each respective strain were spread in BHI agar, separately.
Subsequently, a set of three 0.8 mm diameter test disks were placed in the agar medium Research, Society and Development, v. 9, n. 9, e700997702, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i9.7702 9 previously delimited on the plates. Ten µL of the antibiotic chloramphenicol, as the negative control, and the oils of green and roasted coffee, diluted in Tween 80 at 0.5%, were added on the top of each test disks. The plates were incubated at 37 ºC for 24 hours and then analysed (EUCAST, 2020).

Statistical analysis
The analysis of variance of the data was performed using the software Statistical Analysis System -SAS ® University Edition. The differences between the mean values obtained were assessed using the Duncan means test at a level of 5% probability (p < 0.05). Table 2 shows the percentage of fatty acids identified in green coffee oil, in which the values of individual components ranged from 0.05 to 46.77%. Most of the lipids in foods have linear fatty acids with an even number of carbons (Ordóñez et al., 2007). Oliveira et al. (2014) studied the characterization of green arabica coffee oil and found palmitic (32%), linoleic (38.3%), and oleic (12.8%) acids in greater concentration, which corroborates with the results obtained in the present study.
The high content of poly and monounsaturated fatty acids found in the present study for both green and roasted arabica coffee oils, in particular Ω3, Ω6, and Ω9, are relevant due to impacts on human health, including anticarcinogenic (Cavin et al., 2002) and antioxidants effects (Lee, Choi & Jeong, 2007). Also, due to the roasting of coffee beans, the content of linoleic acid (Ω6) is increased (Garrett et al., 2013), a fact that is confirmed by the results obtained in the present study. Therefore, the mixture of both oil types can improve the general functional and technological potential when compared to the pure oils.

Total phenolic content and antioxidant activity
The results of the analyses of total phenolic compounds, total antioxidant activity by DPPH • free radical scavenging and total antioxidant activity by ABTS •+ free radical scavenging is shown in Table 4. Development, v. 9, n. 9, e700997702, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i9.7702  For the analysis of total phenolic compounds, it was observed that the values in the mixtures of oils (R50G50, R75G25, and R25G75) were higher than that of roasted coffee oil (R100) and green coffee oil (G100), in which the treatment R75G25 showed the best result.
Considering the results found, it is possible to suggest the roasting influence on the phenolic content of the coffee oils, since it is known that this process affects the composition of phenolic compounds through Maillard reactions, for example. The compounds in question are formed by the thermal degradation of carbohydrates, chlorogenic acids, and lignins. In addition, the time and temperature ranges used in the roasting process influence the oil chemical composition (Araújo, 2012).
The results found in the present study occurred due to the presence of bioactive phenolic compounds present in coffee, which are characterized by their antioxidant action.
Among these compounds, chlorogenic acids stand out, in which coffee representing the main source of chlorogenic acids in human nutrition (Rodrigues & Bragagnolo, 2013).
The DPPH method is based on the capture of the DPPH • radical (2,2-diphenyl-1picryl-hydrazil) by antioxidants, producing a decrease in absorbance at 515 nm. In this study, the final result corresponds to the amount of sample needed to decrease the initial concentration of the DPPH • radical by 50%. Thus, the lower the EC50 value, the greater the antioxidant activity of the sample. Thus, all treatments demonstrated DPPH • radical scavenging ability (Table 4), but the treatment R75G25 differed statistically from the other treatments, with the highest antiradical efficiency value. The antiradical efficiency of a Development, v. 9, n. 9, e700997702, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i9.7702 14 compound is directly proportional to its activity as an antioxidant (Chen, Bertin & Froldi, 2013).
Another methodology used in this study to measure the antioxidant activity of coffee oils was that based on the capture of the radical ABTS •+ (3-ethylbenzothiazoline-6-sulfonic acid) which can be generated through a chemical, electrochemical, or enzymatic reaction (Lucas-González, Viuda-Martos, Pérez Álvarez & Fernández-López, 2018). The treatment R75G25 showed the highest antioxidant activity value for this method when compared to the other treatments. The structure of the radicals in question is planar, which makes them react more easily with reducers, such as ABTS •+ either by transferring an electron or hydrogen.
Another important factor is that ABTS •+ can be dissolved in both aqueous and organic solvents, and its antioxidant activity can be measured considering the hydrophilic and lipophilic nature of the antioxidants in the sample, unlike the DPPH method, which is solubilized only in organic medium (Kuskoski, E. M;Asuero, A. G.;Troncoso, A. M.;Mancini-Filho & J. Feet, 2009). Despite that, the treatment R75G25 showed the highest antioxidant activity, for both methods employed in this study, as well as the highest total phenolic content. In this sense, these findings suggest a correlation between these parameters, in which the phenolic compounds can be indicated as an important bioactive group with antioxidant effect present in the coffee oils.

Rheological behavior
Information on the rheology of products is essential to design equipment and industrial pipes, control the quality of raw materials and finished products, in addition to contributing to the prediction of the shelf life of products (Rao, 2014).
In this study, the Newton's Law was adequate to describe the behaviour of all treatments since they presented high values of determination coefficient (0.98 < R 2 < 0.99).
Coffee oils can be characterized as Newtonian fluids, as can be seen in Figure 1, since the shear stress and shear rate varied linearly, with the line intersecting at zero. The rheological properties of these fluids are independent of a previous history of shear and depend only on the composition and temperature (Irgens, 2014). In Table 5 are present the values of the determined rheological parameter, the Newtonian viscosity. The temperatures of 20, 30, and 40 ºC were chosen to be studied in the rheological tests because they are in the temperature range in which food processing containing vegetable oil are carried out as well as the sensory analysis of new products that will be placed on the market. Oliveira et al. (2009) reported that the viscosity of fluid tends to decrease with increasing temperature, a fact that was observed in the resent study. Development, v. 9, n. 9, e700997702, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i9.7702  The decrease in viscosity with an increase in temperature is attributed to the increase in intermolecular distances caused during heating, which reduces the forces of attraction between molecules (Ahammed, Asirvatham & Wongwises, 2016). In this sense, the viscosity is directly proportional to the force of attraction between the molecules. As the temperature increases, this attraction force decreases, also decreasing the viscosity (Brunetti, 2008). Thus, the information that coffee oils behave like Newtonian fluids is relevant to the planning and adequate sizing of pumps and pipes, in order to guarantee the correct functioning of an industry that processes this raw material, generating a high-quality final product.

Oxidative stability
Among the oxidative stability tests, the Rancimat method makes possible to measure the oxidation index of lipids through volatile acids formed during oxidation (Aktar & Adal, Research, Society and Development, v. 9, n. 9, e700997702, 2020 (CC BY 4  Regarding the total amount of unsaturated fatty acids, green coffee oil showed 56.61% and roasted coffee oil presented 64.2%, a fact that may have contributed to the treatments R100 (A), R50G50 (C), and R75G25 (D) had a shorter oxidation induction time. It is known that vegetable oils that have a high fatty acid index with two or three unsaturations have lower oxidative stability due to the presence of allylic or bile-allylic carbons (Pratt, Mills & Porter, 2003). Probably, the most relevant factor for the rapid oxidation of the aforementioned treatments is the coffee roasting process, since the green and roasted coffee oils were from the same batch. In addition, because coffee oils are crude, it is believed that vacuum drying can provide higher oxidation stability.

Antibacterial activity
The sensibility test applied for the verification of coffee oils samples antibacterial activity showed that no inhibition halos were present for all five pathogens studied using disc diffusion tests. On the other hand, it was observed that after the removal of the filter paper with coffee oil samples from the agar, the area where the disc embedded was lying down did not show microbial growth for any species. These evidences imply that all coffee oil treatments exhibit an inhibition of bacterial growth by contact, which is an interesting outcome in the bioactive packing context. As reported by Sanla-Ead, Jangchud, Chonhenchob and Suppakul (2012), the plant-based oils used in in vitro tests failed to exhibit a clear inhibitory zone probably because of low water solubility of the oils, however, the application in antimicrobial packaging film or coating could be considered due to the inhibitory effect presented by the oils when in contact with a surface.
The mechanism of action of antimicrobial agents occurs through one of the following factors: reaction with the cell membrane causing increased permeability and loss of cellular constituents; inactivation of enzyme systems or essential enzymes, including those involved in the energy production process and synthesis of structural components; and/or destruction or functional inactivation of genetic material (Kim, Cornell & Preston, 1995).

Final Considerations
The green and roasted arabica coffee oils investigated in this study are sources of fatty acids, mainly the omega 3, 6, and 9 class, which are health promoters. An increase in the functionality of the oils was observed when they were mixed, with a high functional potential of the treatment R75G25, evidenced by the total phenolic content and antioxidant activity values. Also, all the tested oils behaved like Newtonian fluids at all temperatures to which they were subjected to rheological tests. Regarding oxidative stability, treatments G100 and R25G75 showed longer oxidation induction times, and the antimicrobial sensitivity tests demonstrated inhibition of bacterial strains in the area of contact with all the coffee oils samples. Therefore, the use of mixtures of arabica coffee oils as natural preservatives in food can be considered a promising alternative for the partial replacement of chemical additives in