Antioxidant and Angiotensin-Converting Enzyme Inhibitory Activity of Eucalyptus camaldulensis and Litsea glaucescens Infusions Fermented with Kombucha Consortium

In recent years, the consumption of herbal infusions around the world has increased due to their benefi cial health eff ects. These beverages are prepared by placing a small amount of the selected plant material in freshly boiled water, allowing the preparation to steep for a short period of time (1). Although herbal infusions do not have any particular nutritional value, they represent an important source of bioactive compounds such as polyphenols. It has been shown that these compounds can act by diverse mechanisms providing signifi cant protection against chronic diseases (2). For example, the consumption of some herbal polyphenols with antioxidant activity may regulate hypertension through inhibition of the angiotensin-converting enzyme (ACE), a key component in the renin-angiotensin aldosterone system which regulates blood pressure (3). ISSN 1330-9862 preliminary communication


Introduction
In recent years, the consumption of herbal infusions around the world has increased due to their benefi cial health eff ects.These beverages are prepared by placing a small amount of the selected plant material in freshly boiled water, allowing the preparation to steep for a short period of time (1).Although herbal infusions do not have any particular nutritional value, they represent an impor-tant source of bioactive compounds such as polyphenols.It has been shown that these compounds can act by diverse mechanisms providing signifi cant protection against chronic diseases (2).For example, the consumption of some herbal polyphenols with antioxidant activity may regulate hypertension through inhibition of the angiotensin-converting enzyme (ACE), a key component in the renin-angiotensin aldosterone system which regulates blood pressure (3).
Recently, several research studies have focused on enhancing the benefi cial health potential of herbal infusions.For example, Aloulou et al. (4) reported that fermentation process with kombucha enhanced the phenolic content, antioxidant activity and α-amylase inhibitory activity in fi ve commonly consumed teas.Similar results were observed by Velićanski et al. (5) in fermented lemon balm (Melissa offi cinalis L.) infusion; they reported an enhancement of phenolic compounds such as rosmarinic, caff eic and ferulic acids (1.3-, 1.9-and 4.6-fold higher, respectively), as well as major antioxidant activity against DPPH radicals than in unfermented infusions.
The benefi cial eff ects of these fermented beverages are att ributed to the presence of restructured polyphenols, gluconic acid, glucuronic acid, lactic acid, vitamins, amino acids, antibiotics and a variety of micronutrients produced during fermentation (6).
The fermentation process involves the activity of yeasts that ferment glucose and fructose to ethanol, which is then oxidized to acetic acid by acetic acid bacteria (AAB).The main source of carbon in this process is sucrose.The sugar is hydrolyzed by the enzyme invertase from the yeast present in the kombucha consortium, producing ethanol via the metabolic pathway of glycolysis, with a preference for fructose as the substrate.Subsequently, AAB convert glucose and ethanol into gluconic and acetic acids (7).
In occidental countries, the consumption of tea prepared from Camellia sinensis and analogues, such as fermented beverages, is not that common; on the contrary, the consumption of infusions made with plant materials typical of each region predominates.In this same regard, in a previous study we analyzed phenolic composition and anti-infl ammatory eff ect of diff erent herbal infusions (8), including Eucalyptus camaldulensis and Litsea glaucescens.Results showed anti-infl ammatory activity through the positive modulation of COX-2, TNFα, NFκB, and IL-8 in HT-29 cells, especially of L. glaucescens.Even though both plants have diff erent traditional uses (e.g.Eucalyptus camaldulensis is consumed for decongestant, antiseptic, anti-infl ammatory or antibiotic, among other purposes, whereas Litsea glaucescens infusion is traditionally consumed for its bactericidal, antiseptic, expectorant and anti-infl ammatory properties), both of them contain bioactive compounds with antioxidant activity such as polyphenols (8,9).
Although these plant materials are commonly consumed as infusions, the fermentation process may or may not enhance their positive eff ects.This is unknown and can be poorly related with black tea, despite the fact that fermentation conditions are similar.Therefore, the aim of this work is to compare chemical composition, consumer acceptance, antioxidant and angiotensin-converting enzyme inhibitory activities of E. camaldulensis and L. glaucescens in traditional infusions and fermented beverages prepared with the kombucha consortium.

Plant materials
Representative samples of Eucalyptus camaldulensis and Litsea glaucescens were collected in Ciudad de Duran-go, Durango, Mexico, and positively identifi ed by botanist Socorro González-Elizondo.Voucher specimens (nos.44008 and 44007, respectively) were deposited at the Herbarium of the Interdisciplinary Research Centre for Comprehensive Regional Development, Durango Unit (CI-IDIR-IPN, Durango, Mexico).

Preparation of herbal infusions and fermented beverages
Both materials were dried at room temperature in shade, ground (1.13 mm) and stored in the dark for later determinations.Infusions (1 %, by mass per volume) were prepared by adding 200 mL of freshly boiled water to 2 g of dried ground sample.Infusions were left to stand for 10 min and fi ltered.
The fermentation of infusions was performed using the conditions reported by Vázquez-Cabral et al. (7).The infusions were cooled at room temperature, and then sweetened by dissolving 10 g/L of sucrose.In small bioreactors, 1 L of sweetened infusions was inoculated and acidifi ed with 2.5 g of inoculum (biofi lm with a symbiotic culture of bacteria and yeasts) and 100 mL of black tea vinegar, respectively.The bioreactors were covered with cheesecloth, and the fermentation at (25±1) °C was monitored for seven days.On the last fermentation day, the biomass was removed from the beverage by fi ltration.
For preparation of black tea vinegar, inoculum was grown at 25 °C in sweetened (sucrose, 10 %) black tea for 20 days.Freshly cultured kombucha was used for further subcultures or for fresh fermentation batches.

Physicochemical evaluation
The pH was measured using a digital pH meter (pH meter sension™1; Hach, Loveland, CO, USA) calibrated at pH=4, 7 and 10 at 25 °C.Colour characteristics of infusions were evaluated by measuring their colour coordinates with a colourimeter (chroma meter CR-400/410; Konica Minolta, Osaka, Japan).Colour intensity measurements were expressed as L* (lightness), a* (red/green colour), and b* (yellow/blue colour) values.

Consumer preference tests
The consumer preference evaluation of both, infusions and fermented beverages was performed by 15 untrained panellists who declared drinking a cup of any herbal infusion at least once a week.The panellists ranked the samples according to the following scale: 1=very unpleasant to 9=very pleasant.The samples were coded with random three digit numbers.Black tea was used as a positive control.

Sugar content in fermented beverages
Sample analysis was carried out with an Acquity ultra-performance liquid chromatography (UPLC) system (Waters Corp., Milford, MA, USA) coupled with an evaporative light scatt ering detector (ELSD; Waters Corp., Wexford, Ireland).The LC system consisted of a sample manager (5 °C) and a binary solvent manager.The column used to determine fructose, glucose and sucrose was an Acquity UPLC BEH Amide, 100 mm×2.1 mm, 1.7 μm (Waters Corp., Wexford), operated at 50 °C.The elution profi le used two solvents (both J.T.Baker, Mexico City, Mexico): acetonitrile/water (80:20) with 0.1 % NH 4 OH (solvent A), and acetonitrile/water (30:70) with 0.1 % NH 4 OH (solvent B) and an initial gradient of 5-60 % B in 5 min, isocratic by 1 min, reset and equilibrated for 5.5 min.The fl ow rate was 0.25 mL/min.The UPLC control and data processing were performed using the Masslinx (Waters Corp., Milford) soft ware.

Ultra-performance liquid chromatography coupled with tandem mass spectrometry analysis
Sample analysis was carried out with an Acquity UPLC system (Waters Corp., Milford) coupled with a tandem Xevo TQ-S triple quadrupole mass spectrometer (Waters Corp., Wexford).The LC system consisted of a sample manager (20 °C) and a binary solvent manager.The column used to determine fl avan-3-ols was an Acquity UPLC BEH C18, 50 mm×2.1 mm, 1.7 μm (Waters Corp., Wexford) operated at 40 °C.The elution profi le included two solvents (both J.T.Baker): acidifi ed water with 1 % formic acid (solvent A) and methanol LC-MS grade (solvent B): initial 98 % A, 0-2 min, 68 % A, 2-3.8 min, 55 % A, 3.8-4.5 min, 45 % A, 4.5-6.0min, 5 % B (linear gradient) for column washing and subsequently 6.0-9.5 min, 98 % A for column stabilization.Multiple reaction monitoring (MRM) data were collected from 0 to 9.5 min.Negative ionization mode was used for MS assays.Electron spray ionization (ESI) conditions were as follows: capillary voltage 2.85 kV, desolvation temperature 500 °C, source temperature 150 °C, desolvation and cone gas 794 and 151 L/h, respectively, and collision gas 0.14 mL/min.For identifi cation and quantifi cation, a multiple reaction monitoring mode with standards was used.Rutin (20 ng/μL) was employed to check the stability of the ionization effi ciency of mass spectrometer and a mixture of phenolic compound standards (20 ng/μL; Sigma Corp., St. Louis, MO, USA) for monitoring retention time and m/z values.The UPLC and tandem Xevo TQ-S triple quadrupole mass spectrometer control and data were processed using MassLinx v. 4.1 (Waters Corp., Milford) soft ware.

Radical scavenging and chain-breaking assay
Antioxidant capacity of herbal infusions and fermented beverages was evaluated using the stable radical 2,2 --diphenyl-1-picrylhydrazyl (DPPH˙; Sigma Corp.) method described by Brand-Williams et al. (11).The percentage of scavenging or quenching radicals, Q, was calculated using the following equation: /1/ where A 0 is the absorbance of the blank, and A c is the absorbance of the mixture of radical and sample at diff erent concentrations.For the calculation of the half maximal inhibitory concentration (IC 50 , in μg/μL), the logarithm of the concentration against the inhibition percentage was plott ed.
For chain-breaking kinetics, 3 mL of 3•10 -5 mol/L of DPPH in methanol solution was used.The reaction began with the addition of 10 μL of herbal infusion or fermented beverage at previously determined IC 50 (μg of total content of analyzed fl avonoids per μL).The DPPH bleaching was monitored spectrophotometrically at 515 nm with a spectrophotometer (model 50-BIO; Varian Cary ® , Melbourne, Australia) at 25 °C for at least 60 min (12).The following equation was used to obtain the reaction rate, k: where A is the absorbance at increasing time (t), and A 0 is the initial absorbance.Results on dry mass basis were expressed as both IC 50 and chain-breaking activity (k/mg).Catechin was used as positive control.

Lipid peroxidation assay
Lipid peroxidation was determined according to Jozwik et al. (13) with several modifi cations.Inhibition of the formation of thiobarbituric acid reactive substances (TBARS), a lipid oxidation product, was quantifi ed with a spectrophotometer (model 50-BIO; Varian Cary ® ), using healthy human plasma donated by the blood bank of the Public General Hospital (City of Durango, Durango, Mexico).Inhibition of plasma oxidation was evaluated using Fen ton´s reagent with the addition of phosphate buff er (150 μL of 500 mM H 2 O 2 and 150 μL of 100 mM FeCl 3 for 3 h at 37 °C; Caledon Laboratories Ltd, Georgetown, ON, Canada).Catalyzed plasma (600 μL) oxidation was evaluated for each herbal infusion and fermented beverage samples (100 μL) at several concentrations using phosphate buff er (pH=7.4) to reach the fi nal volume (1 mL).Prior to oxidation phase, a thiobarbituric acid (TBA) reagent (MP Biomedicals, LCC, Illkirch, France) was prepared consisting of 40.5 mL of 20 % acetic acid buff ered at pH=3.5 with 1 M NaOH, 13.2 mL of 8.2 % SDS (Sigma Corp.), 40.5 mL of 0.8 % TBA, and made up to 100 mL with double distilled water.Samples (1 mL) were mixed with 4.0 mL of TBA reagent, incubated at 96 °C for 80 min, and cooled down on ice.Then, they were mixed with n-butanol (5 mL) and centrifuged at 3000×g for 15 min.The ability of the samples to inhibit plasma oxidation was determined by the absorbance decrease at 532 nm.For the calculation of the IC 50 (μg/μL), the logarithm of the concentration against the inhibition percentage was plott ed.Catechin standard (Sigma Corp.) was included as positive control.

Nitric oxide scavenging assay
Infusions and fermented beverages were screened for antioxidant capacity potential using the nitric oxide scavenging assay described by Balakrishnan et al. (14).The absorbance of the chromophores formed during the diazotization of nitrite with sulphanilamide, subsequently coupled with naphthylethylenediamine dichloride (Griess reagent, Sigma Corp.) was read at 546 nm (spectrophotome- ter model 50-BIO; Varian Cary ® ).Catechin was used as a positive control and results were expressed as IC 50 .

Angiotensin-converting enzyme inhibitory activity
The angiotensin-converting enzyme (ACE) inhibitory activity of infusions and fermented beverages was screened using the methodology described by Actis-Go rett a et al. (15).The activity of ACE from rabbit lung (0.05 U/mg of protein; Sigma Corp.) was evaluated following the hydrolysis of hippuryl-l-histidyl-l-leucine to hippuric acid.The formed hippuric acid was separated and quantifi ed with Waters HPLC system consisting of Waters 717 plus autosampler and Waters 600 pump (Waters Corp., Milford).Chromatographic separations were performed with a Supelcosil LC-18-DB column, 150 mm×4.6 mm, 5 μm i.d.(Waters Corp., Wexford), and a mobile phase composed of 0.1 % (by volume) trifl uoroacetic acid in a volume ratio of H 2 O and acetonitrile of 75:25 (J.T.Baker).The fl ow rate used to achieve a retention time of 2.9 min was 1 mL/min.The hippuric acid detection was carried out at 228 nm (Waters 2487 UV spec trophotometric detector; Waters Corp., Milford).Commer cial hippuric acid was used as the standard (Sigma Corp.).For the inhibition of ACE activity by infusions and fermented beverages, a mixture of 10 μL of ACE (0.05 U/mg of protein) in 50 mM HCl-Tris, 300 mM NaCl (pH=8.3),and diff erent concentrations of samples to reach the volume of 90 μL was pre-incubated for 5 min at 37 °C.A volume of 10 μL of 3 mM hippuryl-l--histidyl-l-leucine (1 mM fi nal concentration; Sigma Corp.) was added to the above mixture and incubated for 30 min at 37 °C.The reaction was stopped by placing the sample in a water bath at 100 °C for 5 min.For the calculation of the IC 50 (μg/μL) the logarithm of the concentration against the inhibition activity percentage was plott ed.Captopril (Sigma Corp.) was used as a positive control.

Statistical analysis
Data were expressed as mean value±standard error (SE).Statistical signifi cance was determined by one-way variance analysis (ANOVA) (p<0.05)followed by the Tukey´s test, where p<0.05 was considered signifi cant.Statistical analysis was made using JMP v. 5.0.1 soft ware (SAS Institute, Cary, NC, USA).

Results and Discussion
Results of physicochemical characterization are shown in Table 1.It was observed that the pH of fermented beverages decreased by 2 units compared with the unfermented infusions.The overall decrease in pH is att ributable to the increased concentration of organic acids produced during fermentation by bacteria and yeasts from the tea fungus.In concordance, Velićanski et al. (16) reported decreased values of pH (between 2 and 3 units) in analogues of kombucha at the same fermentation times.
On the other hand, several authors reported changes in colour as a consequence of fermentation.For example, a progressive lightening of colour in fermented teas has been reported (17,18).However, in this work we did not observe signifi cant diff erences between infusions and fermented beverages (Table 1).This could be att ributed to the nature of plant materials and their chemical composition.Contrary to the species studied in this work, epigallocatechin-3-gallate (ECGC) is the main polyphenol found in green tea (about 59 % of total catechins) (19).In fermentation EGCG is the major precursor for other compounds such as theafl avins (20,21).These and other compounds, including bisfl avanols and thearubigins are mainly responsible for the colour change in kombucha beverages (22).
During fermentation, the sucrose is hydrolyzed to glucose and fructose by yeast.Glucose is mainly converted into gluconic acid via the pentose phosphate pathway by acetic acid bacteria, whereas fructose is metabolized to acetic acid, and in minor proportion to gluconic acid (23).The results of sucrose consumption can be observed in Table 1.In both fermented beverages, the kombucha consortium utilized more than 70 % of sucrose for fermentation process.This is consistent with the results of residual glucose and fructose (Table 1), suggesting an active process of fermentation in these herbal infusions (Table 1).
Plant infusions such as E. camaldulensis are traditionally consumed due to their pleasant smell and taste; however, the changes that occur during fermentation process can alter the consumer preference despite their nutraceutical potential.For this reason, we compared the acceptability of infusions and fermented beverages.For this eval-  1.
Both herbal infusions have similar consumer preferences, but 1 point lower than the black tea infusion.Among fermented beverages, E. camaldulensis had lower acceptability, whereas L. glaucescens had similar results to the control.It has been reported that biotransformation of several compounds produced during fermentation increases the preference by consumers compared with certain herbal infusions (7).Nevertheless, our results have not shown any diff erences in the acceptability between herbal infusions and fermented beverages; even when a minor decrease (1 point on hedonic scale) for fermented E. camaldulensis beverage was observed.This could be att ributed to the formation of other compounds that can infl uence fl avour; however, more chemical studies are needed.

Phenolic profi le and concentrations of herbal infusions and fermented beverages
Results of phenolic profi le of infusions and fermented beverages are shown in Fig. 1.Compounds such as gallocatechin, epigallocatechin, gallocatechin gallate, catechin, rutin, kaempferol and quercetin were identifi ed in both beverages.The most abundant detected compound was gallocatechin gallate (GCG), being higher in L. glau cescens infusions and in fermented E. camaldulensis beverage.
Chromatogram results (Fig. 1) and concentrations (Table 2) of each identifi ed compound demonstrated that the fermentation infl uenced the polyphenols present in the herbal infusions of each plant diff erently.For instance, in fermented L. glaucescens beverage, the concentration of GCG decreased signifi cantly (p<0.05;1.8-fold), whereas in fermented E. camaldulensis beverage, the concentration increased (1.8-fold), both compared with their unfermented infusions.A similar trend was observed by Jayabalan et al. (18,24), who reported an increased concentration of total phenolics of kombucha beverages aft er the fermentation.This could be att ributed to the degradation of polyphenol complexes, as a consequence of the increased acidity during fermentation and to the enzymes produced by the kombucha consortium (25).
On the other hand, the decreased concentration of phenolic compounds observed in fermented L. glaucescens beverage could be explained by the partial oxidation of polyphenols to form polymerized compounds with higher molecular mass (16), and thus lower concentration of these compounds.

Antioxidant capacity of beverages
Results of DPPH radical scavenging activity showed that both herbal infusions had antioxidant activity, especially that of E. camaldulensis, with its IC 50 closer to the positive control (Table 3).Despite the chain-breaking DPPH activity results, the fermentation process induced a decrease in the concentration required to stabilize 50 % of DPPH radical (1.2-to 1.7-fold than unfermented infusions).In concordance with our results, Jayabalan et al.
(24) reported an enhancement in DPPH scavenging activity in fermented beverages with respect to unfermented tea.Previous studies with kombucha beverages att ributed the antioxidant activity to the polyphenols and catechins, with a statistically signifi cant correlation between the phenolic content and the DPPH scavenging capacity (R=0.7180;24).However, in our work a lower correlation by Spearman`s test (p<0.05)was observed (R=0.3162).This could be explained by the existence of a partial oxidation of polyphenols, forming macromolecular compounds, and therefore decreasing their relative concentration.However, these compounds could still provide a noteworthy radical scavenging activity as we demonstrated in this work.
In order to complement the analysis of antioxidant activity of plant infusions and fermented beverages, we evaluated the inhibition of lipid peroxidation in pre-oxidized healthy human serum.Results showed that both herbal infusions have antioxidant capacity with IC 50 values close to the positive control (catechin).Improvement of antioxidant capacity of fermented beverages was not observed, showing similar IC 50 values to the unfermented infusions.Jayabalan et al. (18) reported that the antioxidant activity is extremely dependent on the used conditions and the analyzed substrates or products; therefore, not all methods give the same response.
Authors such as Arnao (26) reviewed limitations of antioxidant activity assays, suggesting that the activity must be measured using more than one method, includ- ing primary and secondary oxidation products, and using tests that measure specifi c substrates or products.Therefore, we used a third method: the nitric oxide (NO) scavenging assay.Nitric oxide reacts with superoxide (O 2 .-) to form the peroxynitrite anion, which is a potentially strong oxidant as the decomposition of this molecule produces hydroxyl radical and nitrogen dioxide, contributing signifi cantly to oxidative cell damage (21).The results of the NO assay showed that infusions had weaker eff ect on NO scavenging activity than catechin, used as a positive control (Table 3).However, the antioxidant activity increased with the fermentation (49 % of L. glaucescens and 14 % of E. camaldulensis), unlike the DPPH test results, where we did not fi nd any statistically signifi cant correlation between the phenolic content and antioxidant activity.A high correlation (R=0.8000) between polyphenol content and NO scavenging response by Spearman's test (p<0.05)was observed.The same was observed by Jayabalan et al.
(24), reporting that during fermentation, polyphenol structural modifi cations occurred, such as glycosylation, resulting in bett er scavenging performance on nitrogen radicals.

Antihypertensive activity in vitro
Finally, the antihypertensive potential of fermented beverages was evaluated in vitro through the inhibition of the angiotensin-converting enzyme (ACE) activity.ACE is a carboxypeptidase, and it participates in regulating blood pressure by converting an inactive form of the decapeptide angiotensin I to a potent vasopressor octapeptide, angiotensin II.Therapeutic ACE inhibitors represent an important and common class of pharmaceuticals for hypertension control (27).Results of ACE inhibition activity demonstrate that herbal infusions inhibit the activity of this enzyme (Table 3).L. glaucescens infusion had the highest eff ect, being similar to the positive control (captopril, main pharmaceutical used to control blood pressure).On the other hand, fermented beverages did not enhance the ACE inhibitory potential of herbal infusions; on the contrary, fermentation process reduced the positive eff ect of infusions (1.8-fold of E. camaldulensis and 15.1-fold of L. glaucescens).However, the IC 50 of fermented beverages was still lower than of the positive control.Hur et al. (28) proposed that the increase or decrease of bioactivity of phenolic compounds produced by fermentation may be infl uenced by various factors, including microorganism species, pH, temperature, solvent, water content, fermentation time, kind of plant material and aerobic conditions.Thus, the variations of results could be att ributed to these factors and not only to the fi nal conversion of bioactive compounds.

Conclusions
The present study has demonstrated that fermentation of Litsea glaucescens and Eucalyptus camaldulensis infusions with kombucha consortium modifi es their concentration of phenolic compounds, and their antioxidant and antihypertensive activities.Fermented beverages exhibited increased free radical scavenging activities, but this response depends on the used medium, substrate and mainly on the plant material used for infusion prepara-tion.Additionally, it is interesting to note that E. camaldulensis and L. glaucescens can be considered as natural sources of biocompounds with antihypertensive potential (especially the latt er) either as infusions or fermented beverages.Therefore, they can be considered for further investigations, especially for in vivo assays.

Table 1 .
Physicochemical evaluation and consumer acceptance of herbal infusions and fermented beverages of Litsea glaucescens and Eucalyptus camaldulensis tea infusion and fermented beverage were included as positive controls, resulting in 5 and 4 score points on a hedonic scale for consumer preference, respectively.Results for E. camaldulensis and L. glaucescens infusions and fermented beverages are shown in Table b Values are means of duplicate determinations±standard error.Diff erent lett ers in each row indicate statistical diff erence (p≤0.05) by Tukey's test.N.D.=not determined uation black

Table 2 .
Ultra-performance liquid chromatography (UPLC) coupled with tandem mass spectrometry (MS/MS).MS data and concentration of major phenolic compounds found in infusion and fermented products dValues are expressed as means±standard error (N=2).Values in a row followed by diff erent lett ers for each infusion and fermented beverage are signifi cantly diff erent (p<0.05) by Tukey's test.t R =retention time, N.D.=not detected