Generation of Volatile Compounds from Carotenoids of Dunaliella bardawil Algae by Water Bath Heating and Microwave Irradiation

The volatile compounds formed by thermal degradation of the carotenoids present in Dunaliella bardawil were investigated by microwave irradiation (MW) and water bath heating (WB) in different conditions of temperature and time using central composite design. Volatiles extraction by solid phase microextraction (SPME) was optimized and performed at 40 °C for 15 minutes, and those in greater amount were quantified by a validated method of gas chromatography coupled to a quadrupole mass spectrometer (GC-qMS). Employing WB, 10, 12 and 120 ng mL of β-cyclocitral, α-ionone e β-ionone, respectively, were obtained between 60 and 87 °C for 30-75 min, while by MW, 5, 5, 50 ng mL were obtained between 75-107 °C for 1.5-2.8 min. Considering the shorter time of MW, it can be concluded that if the time necessary to obtain the best yield by WB is employed in multiple MW cycles, an amount 10 times greater of those compounds would be obtained by MW than by WB. The results suggest a new biotechnology application for the carotenoids of the microalgae of the Dunaliella genus.


Introduction
Microalgae are photosynthetic microorganisms widely distributed in marine, freshwater and terrestrial ecosystems. 1,24][5][6][7] The production of microalgae supplies the cosmetic and food supplement industries, and boosts the development of natural food products, an emerging and promising field for industrial application. 8Microalgae have properties typical of higher plants (efficient oxygenic photosynthesis and simplicity of nutritional requirements), but also biotechnological attributes of microbial cells (rapid growth in culture medium and capacity to accumulate or secrete metabolites of interest), 4 supporting large scale applications.
Among the microalgae used to obtain carotenoids, two species of the genus Dunaliella, D. bardawil and D. salina stand out for the production of β-carotene in their chloroplasts. 9The accumulation capacity of β-carotene in Dunaliella is influenced by stress factors in culture, such as high light intensity and salt concentration, low temperatures and/or nitrate deficiency. 10][13][14] Herrero et al. 15 evaluated the chemical composition and antimicrobial activity of the pressurized extract of D. salina.Norisoprenoids such as α-ionone, β-ionone and β-cyclocitral, besides neophytadiene and phytol were observed, as well as antimicrobial activity associated to volatiles.Previously, Larrouche et al. 16 showed that β-ionone inhibits the respiratory chain, preventing microbial oxygen consumption.
8][19][20] The oxidation of β-carotene results in epoxides, apocarotenals and apocarotenones, which in later stages of oxidation produce lighter carbonyl compounds such as norisoprenoids. 21owever, the production of aroma compounds from plants is a low-yield process, and the main source is from synthesis. 22,23On the other hand, biotechnological processes Vol. 27, No. 8, 2016   usually involve less environment damage than chemical processes, and can favor stereospecific flavors.
A preliminary evidence that Dunaliella might be useful in biotechnological processes for aroma production was obtained by Donadio et al. 24 investigating the sea salt known as "Fleur de sel", the solar salt of Saint-Armel (Brittany, France), a valuable spice in international cuisine.The authors suggested that, in addition to its pale orange color influenced by the carotenoid content of the sea algae, the flavor of the salt could be correlated to the norisoprenoids formed from the Dunalliela carotenoids.
Microwave irradiation is widely known as an alternative tool to conventional methods of synthesis and extraction, for instance, microwave-assisted extraction (MAE) can substitute Soxhlet extraction, 25 ultrasound and simple heating reflux.In food processing, microwave irradiation has been successfully applied in pasteurization, decontamination and disinfection. 26,27hus, the aim of this study was to investigate the formation of volatile compounds from the controlled degradation of carotenoids present in D. bardawil microalgae by microwave irradiation versus water bath heating.

Cultivation conditions of Dunaliella bardawil
Microalgae D. bardawil (UTEX LB 2538, University of Texas) cultivation was performed using the culture medium developed by Shaish et al. 28 at pH 8.5 and agitation at 180 rpm in an orbital shaker table at 25 °C (± 3).Cultivation was carried out in two stages, in order to obtain a biomass rich in carotenoids, according to previous studies of our group (data not shown).To maximize cell growth, cultivation was carried out with a constant light intensity of 150 µE m -2 s -1 (fluorescent lamps of 59 W).After the culture achieved the stationary growth phase (6 × 10 -5 cells mL -1 ), it was centrifuged at 3500 rpm, at 15 °C for 15 min.The supernatant was discarded and the green cell pellet (GP) resuspended in 300 mL of culture medium (without NaNO 3 ).In the second stage performed to promote the accumulation of β-carotene, the culture remained for 72 h under stirring and light intensity of 400 µE m -2 s -1 .Then, aliquots of 10 mL of biomass rich in carotenoids (BC1) were stored in 15 mL tubes with Shaish medium in a freezer until the moment of analysis, when the biomass was thawed and centrifuged at 3500 rpm, at 15 °C for 15 min.The supernatant (Shaish medium) was discarded and the orange cell pellet (OP) resuspended in 10 mL of sterile distilled water, and then transferred to 40 mL vials with cover and silicone septum.The biomass rich in carotenoids with distilled water (BC2) was used to evaluate the controlled degradation of carotenoids in water bath and microwave controlled conditions.

Solid phase microextraction (SPME)
The experimental conditions of solid phase microextraction (SPME) were optimized to monitor the production of volatile compounds in the biomass (BC2) process.A central composite factorial design with five levels was employed (Table 1).The volatile compounds were trapped on a divinylbenzene/carboxen/ polydimethylsiloxane (DVB/CAR/PDMS) SPME fiber and analyzed by gas chromatography coupled to a quadrupole mass spectrometer (GC-qMS).
Statistica software version 7 (2004) was used to process the analyses results.Each experiment was performed in triplicate.

Quantification of β-cyclocitral and ionones
The parameters evaluated in the determination of the concentration of the major volatiles, were: selectivity, linearity, range, limit of detection (LOD) and quantification (LOQ), accuracy and precision (repeatability), in accordance to Instituto Nacional de Metrologia, Normalização e Qualidade Industrial (INMETRO). 31The analytical curves of β-cyclocitral and α-ionone were built by using aqueous solutions fortified with standards.
To determine selectivity, matrix effect was investigated for β-cyclocitral and α-ionone by comparing their curves in solution (water and ethanol) and matrices (cell suspension of D. bardawil added with β-cyclocitral and α-ionone, in both solvents) and the slopes of two regression lines. 32After determining the matrix effect, linearity was verified by the presence of outliers, by Grubb's test, and homoscedasticity by Cochran's test.The determination coefficient and the residual plot were then obtained.
From a stock solution of β-cyclocitral and α-ionone (36.7 and 35.7 µg mL -1 , respectively), aqueous standard solutions were prepared with the following concentrations: 1, 3, 5, 10, 15 ng mL -1 ; 1, 30, 60, 90, 150 ng mL -1 , respectively.Each solution was analyzed by HS-SPME/GC-qMS.First, the cell suspension of D. bardawil in water was placed in a 40 mL vial and standard solutions added (final volume of 10 mL).Then, SPME fiber was manually exposed in the vial headspace under the conditions determined by the central composite factorial design.All measurements were made in three replicates, each replicate being the analysis of a different aliquot of each aqueous solution.
Accuracy was determined through the recovery of β-cyclocitral and α-ionone and precision through repeatability.LOD and LOQ were obtained by injecting β-cyclocitral and α-ionone until a signal to noise ratio equal to 3 and 10, respectively.

Quantification of carotenoids
Total carotenoid content was measured in the β-carotenerich biomass (BC2).The biomass was extracted with ethanol (95%) and spectrophotometrically measured by the method of Lichtenthaler. 33Carotenoid profile was obtained by HPLC-UV (460 nm) after separation in a Nova Pak C-18 analytical column according to Brotas and Plante-Cuny 34 related to standards.

Controlled degradation of carotenoids from D. bardawil by water bath heating and microwave irradiation
In order to evaluate the influence of time and temperature on the degradation of the carotenoids present in D. bardawil biomass (BC2), a central composite design with two variables was used, in which five levels of temperature were employed for both water bath heating (WB) and microwave irradiation (MW) processes (Table 2).Each experiment was made in triplicate, using 10 mL of sample BC2.Statistica software version 7 (2004) was used for data analysis.
A considerable increase of the absolute areas of ionones occurred when SPME was carried out up to 60 °C and 40 min.For β-cyclocitral, the increase was observed between 50-75 °C and 35-50 min.With the aid of standards, Nonier et al. 35 showed that from room temperature up to 40 °C (40 min), the concentration of β-carotene suffered a reduction of only 2.5%, while at 70 °C (20 min) it was completely degraded.Therefore, the increase in αand β-ionone areas with time and temperature does not necessarily indicate that a more effective extraction occurred, as suggested for β-cyclocitral (Supplementary Information section).In fact, higher temperatures may promote the degradation of carotenoids in D. bardawil.
Based on this, softer conditions for SPME extraction were used.At 30 °C/30 min and 40 °C/15 min, the GC profile was equivalent for αand β-ionone, suggesting that the degradation of carotenoids in D. bardawill biomass had not yet started.So, the last condition of 40 °C for 15 min was chosen due to the shorter time.

Norisoprenoids quantification
As no quantitative method for the determination of norisoprenoids obtained from carotenoids of D. bardawil has been described in literature, the parameters of validation were defined as shown in Table 3. Evaluation of the matrix effect was performed by comparing the slope of the straight line and, as t calculated was greater than t tabulated for β-cyclocitral (Table 3), 32 a matrix effect was accepted for this compound, and extrapolated for α-ionone.
The analytical curves and residue profile for β-cyclocitral and α-ionone are shown in the Supplementary Information section (Figure S3).The method was successfully implemented, allowing more accurate data for the experiments aimed at the production of norisoprenoids.
Optimization of volatiles formation from D. bardawil biomass (BC2) by water bath heating and microwave irradiation HPLC-UV of the carotenoids present in D. bardawil biomass showed that 90% was composed by β-carotene and 10% by lutein (Supplementary Information section, Figure S4) and total carotenoids in BC2 biomass was of 15 µg 10 mL -1 .The controlled degradation of carotenoids and the formation of β-cyclocitral, αand β-ionones were evaluated in a controlled water bath heating (WB) and by microwave irradiation (MW) (Table 2, Experimental section).SPME-GC-qMS (40 °C for 15 min) was performed, as previously discussed.
The influence of temperature and incubation time on the response variable (concentration of β-cyclocitral, α-ionone and β-ionone from BC2 biomass), using MW and WB, is shown in the contour plots of Figure 1 (response surface graphs are displayed in the Supplementary Information section, Figure S5).The concentration (ng mL -1 ) of β-cyclocitral, α-ionone and β-ionone at the central point is shown in the Supplementary Information section, Table S1.
In SPME from D. bardawil (BC2) WB treatment, the response variable (concentration) was greater for β-cyclocitral when temperature was between 60-87 °C for 30-70 min; for α-ionone, it was between 65-85 °C for 30-55 min and for β-ionone, between 60-62 °C for 45-75 min, leading to the production of 10, 12 and 120 ng mL -1 of β-cyclocitral, α-ionone and β-ionone, respectively.For MW, the response variable (concentration) was greater for β-cyclocitral between 92-107 °C for 2.3-2.8min, while for α-ionone and β-ionone the best conditions were between 75-85 °C for 1.5-1.8min and All results were obtained from triplicate injections; cal: calculated; tab: tabulated; LOQ: limit of quantification; LOD: limit of detection.75-88 °C between 2.0-3.5 min, respectively, achieving 5, 5 and 50 ng mL -1 of β-cyclocitral, α-ionone and β-ionone, respectively.Previous study 36 on the production of norisoprenoids from carotenes in potato chips revealed contents of the same order of magnitude.Related to the carotenoid content, WB process imparts a complete degradation at temperatures between 75-95 °C for 50-85 min, while in MW, the complete degradation occurs between 80-95 °C for 1.7-2.5 min.Kanasawud and Crouzet 37 studied the kinetic degradation of β-carotene in water under sonication and saturated by oxygen.Twenty-one volatiles were observed, dihydroactinolide included, postulated to be formed from β-ionone, as well as other volatiles.A diversity of works investigated the radical oxidation of carotenoids at different conditions.More recently, Dongho et al. 38 investigated carotenoid degradation caused by sunlight exposure and heating conditions in crude palm oil, revealing a significant decrease of carotenoids in accelerated heating.Interestingly, they also observed a high increase of peroxide value related to lipid oxidation, where the reactive oxygen species (ROS) produced in the process could also impact carotenoid degradation.In fact, ROS react with the polyene chain of carotenoids to generate apocarotenals and, later, norisoprenoids. 21s D. bardawil is rich in tri-and diacylglicerols, 39 both direct heating and ROS mechanisms may contribute to norisoprenoid production.
For the time variable (minutes), WB process suggests 45 to 55 min for the simultaneous production of the three compounds.For MW, it was not possible to observe a common area for the production of the three compounds, but only for β-cyclocitral and β-ionone (2.3-2.8 min).
It is also interesting to observe that MW process can lead to a greater yield of volatiles when the time is the same as in the WB process.As an example, for β-cyclocitral at the center point, 1.5 min were necessary for MW and 45 min for WB (Table S1), to give 4.46 ± 0.34 and 12.83 ± 0.11 ng mL -1 , respectively.Therefore, 30 cycles of 1.5 min in MW would yield 10 times more β-cyclocitral at the central point conditions in both methods.

Conclusions
To the best of our knowledge, this is the first work using carotenoids from D. bardawil microalgae to produce volatile compounds, employing microwave irradiation and water bath controlled heating.This proposal suggests a sustainable production or enrichment of volatiles commonly present in food products, since water and mild conditions of time and temperature were employed.

Table 1 .
Experimental design matrix for SPME

Table 2 .
Matrix of experiments for water bath heating (WB) and microwave irradiation (MW) controlled degradation of D. bardawil biomass (BC2)