Recovery of bioactive substances from rowanberry pomace by consecutive extraction with supercritical carbon dioxide and pressurized solvents

Journal Pre-proof Recovery of bioactive substances from rowanberry pomace by consecutive extraction with supercritical carbon dioxide and pressurized solvents ˙ Paulius Kraujalis, Laura Tamkute, ˙ Dalia Ramune˙ Bobinaite, ˙ Pranas Viˇskelis, Petras Rimantas Venskutonis Urbonaviˇciene,

PII:

S1226-086X(20)30056-3

DOI:

https://doi.org/10.1016/j.jiec.2020.01.036

Reference:

JIEC 4956

To appear in:

Journal of Industrial and Engineering Chemistry

Received Date:

25 September 2019

Revised Date:

6 January 2020

Accepted Date:

30 January 2020

Please cite this article as: Bobinaite˙ R, Kraujalis P, Tamkute˙ L, Urbonaviˇciene˙ D, Viˇskelis P, Venskutonis PR, Recovery of bioactive substances from rowanberry pomace by consecutive extraction with supercritical carbon dioxide and pressurized solvents, Journal of Industrial and Engineering Chemistry (2020), doi: https://doi.org/10.1016/j.jiec.2020.01.036

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier.

Recovery of bioactive substances from rowanberry pomace by consecutive extraction with supercritical carbon dioxide and pressurized solvents

Ramunė Bobinaitėa, Paulius Kraujalisa, Laura Tamkutėa, Dalia Urbonavičienėb, Pranas Viškelisb, Petras Rimantas Venskutonisa Department of Food Science and Technology, Kaunas University of Technology, Radvilėnų, pl.19,

of

a

b

ro

Kaunas, LT-50254, Lithuania

Biochemistry and Technology Laboratory, Institute of Horticulture LAMMC, Kaunas st. 30, Babtai,

Jo

ur na

lP

re

-p

Kaunas distr., LT-54333, Lithuania

1

ro

of

Graphical abstract

Y = 4.8%

SFE-CO2

Protein = 10 %

β-Carotene recovery up to 53% C18:2n6c = 59 %

re

Sugars = 7 %

-p

Rowanberry pomace

ur na

lP

Crude fiber = 20 %

C18:1n9c = 27%

Sequential PLE Total Y = 33.1 %

(ActEtOHH2O)

TPC = 44-109 mg/GAE g Proanthocyanidins up to 11 mg/ ECE g

Jo

Extraction residue

Strong antioxidant capacity

Extraction residue Protein = 17 % Sugars (nd) Crude fiber = 43 %

2

Highlights

Supercritical CO2 and pressurised liquid extractions applied to rowanberry pomace Supercritical CO2 extraction process was optimized by response surface methodology Pressure and temperature largely affected the yield in supercritical CO2 extraction Total carotenoid and -carotene concentrations and their recoveries were monitored

of

Hydrophilic extracts were strong antioxidants, lipophilic rich in unsaturated fatty acids

ro

Abstract

Rowanberry (Sorbus aucuparia L.) pomace was consecutively extracted with supercritical carbon

-p

dioxide (SFE-CO2) and pressurized solvents (PLE) of increasing polarity (acetone, ethanol and water).

re

SFE-CO2 parameters were optimized using central composite design (CCD) and response surface methodology (RSM) in order to obtain the highest lipophilic extract yield. The highest extract yield

lP

(4.80%) was obtained at 45 MPa pressure, 60 °C temperature and 180 min extraction time. The changes in SFE-CO2 parameters substantially influenced carotenoids content in the extracts; the recovery of total

ur na

carotenoids was up to 49.7% of the amount determined by hexane extraction. Linoleic (59%), oleic (27%) and palmitic (9%) fatty acids were dominating in the extracted oil. PLE of SFE-CO2 residue yielded polyphenol-rich extracts (the total extraction yield was 33.1%) with strong antioxidant capacity. Rowanberry pomace should be regarded as a potential source of functional ingredients for food and other

Jo

relevant uses.

Keywords: rowanberry pomace, supercritical CO2 extraction, pressurized liquid extraction, carotenoids, antioxidant capacity 1. Introduction

3

Fruits and vegetables are globally recognized as a part of healthy diet due to the high content of various bioactive phytochemicals and other nutrients [1,2]. Berries are particularly rich in some classes of health beneficial phytochemicals and their production and consumption is regularly increasing [3]. Many berry species may be consumed as fresh fruits; however, due to a very short shelf life, large parts of berry harvests are processed into various more stable during storage products, resulting in generation of large amounts of by-products containing valuable nutrients, including health beneficial bioactive

of

compounds [4]. Nowadays the major part of such by-products are used rather inefficiently (e.g. for animal feed and composting) or even discarded as a waste to landfills. Consequently, development of

ro

more effective ways for processing and application of berry by-products is an important task for the

-p

society.

One of the most promising strategies of utilization of berry by-products is converting it to various

re

high added value natural ingredients, which might increase nutritional value of foods, particularly by providing health beneficial functionality, or substitute some synthetic additives, which are widely used

lP

in foods nowadays instead of increasing negative consumer attitude. Nutraceuticals, cosmetics and pharmaceuticals are other promising areas for the application of plant phytochemicals. These aspects

ur na

have been important factors in expanding research on various horticultural plants [5] as well as their byproducts [4,6-8]. However, despite the increasing number of publications on valorization of agroindustrial by-products, including berry pomace [9-11], many fruit and berry species still remain under

Jo

investigated.

Rowanberry (Sorbus aucuparia L., Rosaceae family), also known as mountain ash, is one of them.

The fruits are small (6-9 mm in diameter), round-shaped and of bright orange-red color. In different countries they have been traditionally used for their diuretic, anti-inflammatory, vasoprotective properties, for the treatment of various gastrointestinal disorders and respiratory tract-related ailments [12] as well as a source of vitamins, especially vitamin C, which is present at approx. ~490 mg/kg [13]. 4

Rowanberry extracts were also reported to possess antioxidant activity [14] and bacteriostatic effect [15]. Caffeoylquinic acids (chlorogenic and neo-chlorogenic), contributing up to 80% of the total phenolics were reported as the main phenolic constituents in wild and cultivated rowanberries [15]. Other hydroxycinnamic acids, their derivatives, and flavonols such as quercetin, rutin, hyperoside and isoquercetin glycosides were also identified in rowanberries [16,17]. Anthocyanins are mainly found in the cultivated rowanberry species, whereas in wild rowanberries

of

these compounds are present in low amounts, typically ≤1% of the total phenolics [15]. The seeds of mountain ash contain approx. 20% of oil [18]. In general, seed oils of various berries have the unique

ro

composition, consisting mainly of unsaturated fatty acids, and high content of lipophilic antioxidants

ingredients and/or cosmetic products [11,18,19].

-p

(tocopherols, carotenoids), which make berry seeds valuable raw materials for functional food

re

Fresh rowanberries usually are not consumed as a food due to their astringent taste; therefore, berries are commonly processed into juice, purees, jams and sugar confections. The reports on rowanberry

lP

processing by-products are rather scarce: the seeds or pomace were extracted with organic solvents [14,18] and supercritical carbon dioxide (SFE-CO2) [19]; however, the processes were not optimized

ur na

and more systematic and comprehensive studies have not been performed until now. On the other hand, the benefits of combined or separate use of SFE-CO2 and pressurized liquid extraction (PLE), enabling efficient and selective isolation of valuable substances, were previously demonstrated for the recovery

Jo

of phenolic compounds from black mulberry leaves [20], essential oil and phytochemicals from medical plants [21,22], oil and phenolic compounds from citrus by-products [8] and antioxidants from spent coffee grounds [23]. Therefore, the aim of this study was development and evaluation of a more effective processing scheme of rowanberry pomace based on biorefining concept, which is focused on obtaining several valuable products and remarkably reducing the amounts of waste. For this purpose, a multistep extraction 5

process was applied. It includes SFE-CO2 at optimized parameters for obtaining the highest yield of lipophilic substances in general and recovery of carotenoids from rowanberry pomace in particular; and application of PLE for further isolation of the higher polarity substances from the SFE-CO2 residue. The products obtained were evaluated by determining their chemical composition and antioxidant capacity using in vitro assays. It is expected that the results of this study will provide important data for the valorization of rowanberry pomace as a source of natural, valuable food and non-food products and will

of

contribute to the sustainability in agro-food sector.

ro

2. Materials and methods

-p

2.1. Rowanberry pomace and chemicals

Frozen rowanberry pomace was kindly donated by the joint-stock company Melyne (Babtai, Kaunas

re

distr., Lithuania). Defrosted pomace was freeze-dried in a freeze dryer F100 (Frozen in Time Ltd., York,

lP

England) and ground in a Retsch GM 200 mill (Haan, Germany), using sieve ring with 0.2 mm trapezoidal perforation. The moisture content of ground pomace was 4.9±0.19%.

ur na

Carbon dioxide (99.9%) was from Gaschema (Jonava, Lithuania), agricultural origin ethanol from MV group (Kaunas, Lithuania). HPLC-grade solvents (hexane, methanol, methyl-tert-butyl ether, tetrahydrofuran), β-carotene standard (≥93%), gallic acid, BHT, p-(dimethylamino)cinnamaldehyde (DMAC), 2,2’-azinobis(3-ethylbenzothiazoline-6-sulfonic acid)diammonium salt (ABTS),2,2-diphenyl-

Jo

1-picrylhydrazyl hydrate stable radical (DPPH•, 95%), (-)-epicatechin were purchased from SigmaAldrich (Steinheim, Germany). A Supelco 37 component fatty acid methyl esters (FAME Mix), were purchased from Sigma-Aldrich, (Taufkirchen, Germany), Folin-Ciocalteu phenol reagent - from Fluka Chemie (Buchs, Switzerland), 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox) from Acros Organics (Geel, Belgium). All other reagents were of analytical grade.

6

2.2. Extraction processes 2.2.1. Supercritical fluid extraction (SFE-CO2) SFE-CO2 was carried out by a modified procedure of Kraujalis and Venskutonis [24] in a Helix extractor (Applied Separation, Allentown, PA, USA) using 20 g of ground rowanberry pomace. Each sample was loaded into a 50 cm3 extractor vessel (14  320 mm) between two layers of cotton wool. The flow rate of CO2 was controlled manually by the micro-metering valve (back-pressure regulator). The

of

volume of consumed CO2 was measured by a ball float rotameter and a digital mass flowmeter in standard

ro

liters per minute (SL/min) at standard state (P100 kPa, T20 C, 0.0018 g/mL). The extracts were collected in amber-coloured glass bottles.

-p

Extraction time (180 min) and flow rate of CO2 (2 SL/min) were constant for all experiments. These

re

parameters were selected based on the previous kinetic studies. Pressure and temperature ranges of 2545 MPa and 40-60 °C (Table 1), respectively, were selected for optimization using Response Surface

lP

Methodology (RSM). A static time of 10 min was included in to the total extraction time. Calculated solvent/feed ratio was 0.648/0.02=32.4.

ur na

For the evaluation of gradual recovery of carotenoids in lipophilic extracts SFE-CO2 was up scaled using a 500 cm3 extractor vessel at selected pressure and temperature values, whereas extraction time was prolonged to 360 min and CO2 flow rate was kept 2 SL/min. For comparison, the yield of lipophilic fraction was also determined by Soxhlet extraction with n-

Jo

hexane using 5 g of dried rowanberry pomace and 5g SFE-CO2 residue. The solvent after Soxhlet extraction was removed in a rotary vacuum evaporator at 40 °C.

2.2.2. Pressurized liquid extraction (PLE)

Rowanberry pomace residue left after optimized SFE-CO2 was further extracted in an accelerated 7

solvent extractor ASE-350 (Dionex, Sunnyvale, CA, USA). The sample (11 g) was mixed with diatomaceous earth and placed in a cell with stainless steel frits and cellulose filters on both sides. Extraction parameters were selected based on the previously performed PLE optimization for raspberry pomace [11]; temperature 70 °C, pressure 10.3 MPa, 5 min pre-heating and three 5 min cycles of static extraction (15 min in total), load cell flush volume 100% and 100 s purge time with N 2. Extraction was performed sequentially using increased polarity solvents, acetone, ethanol and water. Acetone and

of

ethanol were evaporated in a rotary vacuum evaporator and then the extracts were finally dried under gentle nitrogen stream. Water extract was freeze-dried. The dry extracts were kept in tight sealed

-p

2.3. Analysis of carotenoids

ro

containers in a freezer (−18 °C) until analyses.

re

For HPLC analysis, 0.1 g of extract was dissolved in 10 mL of hexane/tetrahydrofuran mixture (4:1 v/v) containing 1% BHT, then filtered through a 0.45 mm polyvinylidene fluoride (PVDF) syringe filter

lP

(Millipore, Burlington, MA) and analyzed on a Waters HPLC system consisting of 2695 liquid separation module, UV-Visible detector UV-Vis2489 (Waters Corporation, Milford, MA), and equipped with a RP-

ur na

C30 column, (5μm, 4.6250 mm, YMCTM Europe, Dinslaken, Germany) connected to a C30 guard column (5 μm, 104.0 mm, YMC Europe, Dinslaken, Germany). The flow rate was 0.65 mL/min, column temperature 22 C and -carotene was detected at 450 nm. The mobile phase consisted of

Jo

methanol (solvent A) and methyl-tert-butyl ether (solvent B). The samples were injected at 1% B (held 1 min), and the gradient then changed to 100% B (1-90 min) and again to 1% B in 5 min (held 5 min). For quantification a calibration curve was produced using an authentic all-trans--carotene standard (concentration range was from 0.1 to 5.0 mg/100 mL). The coefficient of determination (R2) of the calibration curve was 0.996, with a limit of detection (LOD) 0.030 and limit of quantitation (LOQ) 0.068 mM. The total carotenoid content was calculated in all-trans--carotene equivalents. 8

2.4. Analysis of fatty acid composition and peroxide value FAMEs were prepared by transesterification of triacylglycerols present in SFE-CO2 and Soxhlet extracts using BF3 as a catalyst [25] and analyzed on a HRGC 5300 with a flame ionization detector (Mega Series, Carlo Erba Instruments, Milan Italy). Separation of FAMEs was performed on a silica capillary column SPTM-2560, 100 m, i.d. 0.25 mm, df 0.20 μm (Supelco, Bellafonte, PA, USA) using temperature program from 80 to 240 °C at 4 °C/min. The injector temperature was 220 °C and detector’s

of

temperature 240 °C, split ratio was 100:1. Helium was used as a carrier gas at a flow rate of 20 cm3/s.

ro

The FAMEs were identified by comparing retention times with those in the standard mixture, while their content was expressed as weight percentage of each fatty acid (the means of 3 replicate determinations)

re

CO2 was measured according to EN ISO 3960 [26].

-p

in relation to the total fatty acid content. Peroxide value of lipophilic extract obtained at optimized SFE-

lP

2.5. Determination of antioxidant potential and total proanthocyanidins Three well-known in vitro assays were used for assessing antioxidant potential of extracts. The DPPH• scavenging capacity was determined by the method of Brand-Williams et al. [27] with slight

ur na

modifications fully described elsewhere [28]. The ABTS•+ scavenging capacity was determined by the method of Re et al. [29] as described by Dobravalskytė et al. [7]. Briefly, 2 mL of ABTS•+ solution were mixed with 20 μL extract in a 1 cm path length cuvette. The reaction mixture was kept at room

Jo

temperature in the dark for 30 min, and the absorbance was read at 734 nm. Total phenolic content (TPC), which is also based on a single electron transfer reaction and therefore

may be considered as an indicator of antioxidant potential, was measured with Folin–Ciocalteu reagent [30]. The test cuvettes were filled with 1.0 mL of prepared extract and mixed with 5.0 mL diluted in distilled water (1/10, v/v) Folin-Ciocalteu’s phenol reagent and 4.0 mL of Na2CO3 (7.5%). The absorbance was read at 765 nm after 60 min incubation in the darkness. The DPPH• and ABTS•+ 9

scavenging values are expressed as μg Trolox equivalents in g extract (μgTE/g); TPC values are expressed in mg of gallic acid equivalents in 1 g extract (mg GAE/g). Total proanthocyanidins were determined spectrophotometrically using 0.1% DMAC solution, which was prepared with hydrochloric acid in ethanol (1:9, v/v) [31]. Then appropriately diluted extracts (10 μL) were added to 3 mL of DMAC solution. The resulting mixtures were allowed to stand at room temperature for 5 min, and absorbance was measured at 640 nm. Total proanthocyanidin content in the

ro

2.6. Determination of crude fiber, crude protein, sugars and dry matter

of

extracts was expressed in mg of (-)-epicatechin equivalents in 1 g extract (mg ECE/g).

-p

The crude fiber content in rowanberry pomace residues left after SFE-CO2 and PLE was measured by an Ancom200 Fiber Analyzer (ANKOM Technology, Macedon NY, USA) according to the

re

manufacturer method. Crude protein was determined by the Kjeldahl method (N×6.25) [32]. D-fructose, D-glucose, and sucrose contents were determined using Megazyme Assay Kit (K-SUFRG

lP

04/18 Megazyme Assay Procedure) and Genesys-10 UV/Vis spectrophotometer (Thermo Spectronic, Rochester, USA). Dry matter content was determined gravimetrically after a forced air convention drying

ur na

at 105°C to a constant weight.

2.10. Statistical analysis

Jo

RSM using central composite design (CCD) was employed to determine optimal SFE-CO2 pressure (P) and temperature (T) for maximal extract yield. Data was analysed and the model was established using Design-Expert 7.0 software (Stat–Ease Inc., Minneapolis, MN, USA). The number of experiments was calculated by the formula N= (2f+2f+c); where f and c – the number of factorial and center points, respectively. The complete design for SFE-CO2 consisted of 13 experimental runs, including 4 factorial, 4 axial and 5 centre points. The data of CCD was fitted with a second order polynomial equation:𝑌 = 10

𝛽0 + ∑4𝑖=1 𝛽𝑖 𝑋𝑖 + ∑4𝑖=1 𝛽𝑖𝑖 𝑋𝑖2 + ∑𝑖 ∑𝑗=𝑖+1 𝛽𝑖𝑗 𝑋𝑖 𝑋𝑗 ; where Y is the predicted response; β0 is a constant; βi, βii, βij are the coefficients for linearity; Xi and Xj are independent variables. Statistical significance of the model and variables was determined at 5% probability level (p<0.05). The adequacy of the model was determined by evaluating the ‘lack of fit’ coefficient and the Fisher test value (F-value) obtained from the analysis of variance. Extractions at every point were performed in triplicate and in random order.

of

The mean values and standard deviations (SD) of the experimental data were calculated using MS Excel (Microsoft, JAV). One-way analysis of the variance (ANOVA) followed by a single-step multiple

ro

comparison Turkey’s HSD test to compare the means that showed significant difference (p ≤ 0.05) were

3. Results and discussion

ur na

3.1.1. Total extraction yield

lP

3.1. Optimization of SFE-CO2 parameters

re

-p

performed using STATISTICA 12 (StatSoft, Inc., Tulsa, OK, USA) package.

CO2, as a solvent, has a number of advantages (nontoxic, nonflammable, inexpensive and yields high purity extracts), thus it has been successfully explored for extraction of substances for food, cosmetic and pharmaceutical applications. The ability of CO2 to dissolve various substances highly depends on

Jo

process parameters. Therefore, SFE-CO2 conditions should be optimized for every individual raw material. In this study RSM-CCD was used to optimize the impact of two independent variables, namely temperature (T) and pressure (P) for obtaining the highest yield of lipophilic extract. Experimentally obtained SFE-CO2 extract yields from rowanberry pomace varied from 4.12 to 4.80 g/100g (dw) and it was lower than in conventional Soxhlet extraction (6.15±0.37g/100 g dw) by 33– 22% (Table 2). The

11

SFE-CO2 residue obtained at optimal conditions was re-extracted in a Soxhlet apparatus resulting in additional 1.28±0.11% recovery of lipophilic extract. The model fitting and the effects of independent variables were evaluated by ANOVA (Table 3). The analysis of the quadratic regression model showed that the model was significant (p<0.1) and the “lack of fit” was not significant (p>0.05) relative to the pure error. The four effects of the model indicated that they are significantly different (p<0.05) from zero at the 95.0% confidence level. The model showed that

of

the factor with the largest effect on extraction yield was P (p<0.0003) with F value of 43.30, followed by T (p<0.0004; F=41.38). The effects of second-order termsT2 (p<0.0008) and P2 (p<0.0102) were lower

ro

with F values of 31.40 and 12.13, respectively.

-p

The adequacy of the model described by determination coefficient R2 (0.95) indicates reasonable fit of the model to the experimental data. Model analysis also showed reasonable agreement between the

re

adjusted and predicted coefficients of determination (R): 0.91 and 0.76, respectively. The results of CCD

lP

can be expressed by the following second-order polynomial regression model equation (1): (Yield) = 4.28 + 0.17×T + 0.17 × P – 0.08 × T × P + 0.16 × T2 + 0.10 × P2 (1)

ur na

3D response surface plot showing the effect of extraction temperature, pressure and their interactions on extract yields are presented in Fig. 1. The lowest pressure (25 MPa) and temperature (40 °C) gave the lowest yield; it increased by raising temperature to 60 °C at the constant pressure. This may be explained

Jo

by the increase of CO2 vapor pressure at higher temperature, resulting in lower viscosity of extracts. Figure 1 also shows that increasing pressure from 25 MPa to 45 MPa results in higher extract yields. This may be explained by the increase of CO2 density by increasing pressure at lower temperature, resulting in a higher diffusivity and solvating power [11,24]. In our study, solvating power of CO2 at the highest applied pressure (45 MPa) and temperature (60 °C) was the highest resulting in the maximum experimental yield (4.80 g/100 g dw). From this point of view, some inverse effects of pressure and 12

temperature may be observed. For instance, increasing temperature from 40 to 60 °C at the lowest 25 MPa pressure increase extract yield by 0.57%; the same temperature increase at 45 MPa increased extract yield only by 0.25%. Increasing pressure from 25 to 45 MPa at 40 °C increased extract yield by 0.44%; the same pressure increase at 60 °C increased extract yield only by 0.11%. These findings indicate that rising of extraction pressure at lower temperature has bigger impact on extraction yield increase than at higher temperature.

of

Based on literature findings, pressure is the mostly reported parameter that influences extraction yields in SFE-CO2of solid biomass, including pomace and seeds [11,24,33]. The results of current

ro

research are in agreement with previously reported observations concerning SFE-CO2 of plant materials.

-p

Predicted values of rowanberry lipophilic extract yield were calculated using a second order polynomial equation (1) and compared with experimental values (Table 2). Predicted values from the fitted equation

re

and obtained values were in a good agreement (Fig. 2). The optimal conditions suggested of the model for obtaining the highest extract yield were: P= 45 MPa; T= 60 °C. It may be noted that in this case the

lP

model suggested maximal values of variables as optimal parameters; therefore from the mathematical point of view optimization may be considered as valid only for the selected parameter range. On the

ur na

other hand numerous studies demonstrated that further increase in pressure provides negligible change in extract yield, whereas higher temperatures may cause degradation and isomerization of carotenoids [34], compromising their bioactivity.

Jo

In the current study the lipophilic extracts yields obtained from rowanberry pomace after SFE-CO2 at optimized conditions and Soxhlet extraction were on average 3.2 times lower than previously reported for rowanberry seeds (19.9% dw) [18]. It suggests that the seeds contribute the majority of lipids extracted from the whole rowanberry pomace. To the best of our knowledge there are no reports on SFE-CO2 yields of rowanberry pomace. However, SFE-CO2 has been applied previously for pomaces of raspberries [11], guelder-rose berries 13

[35] and sea buckthorn [9] with the maximal yields more than two times higher (≈14.6% dw) compared to the results obtained in the present work for rowanberry. This might be due to the relatively low seed content in rowanberry pomace (7.3 - 2.6-fold lower) compared to pomaces of other berries, including chokeberry [36]. On other hand, the reported SFE-CO2 yield obtained from chokeberry pomace was significantly lower (2.95% dw); higher extraction yields from chokeberry pomace (up to 7.08% dw) were obtained only with the use of co-solvent [10].

of

Although, the modeling of extraction kinetics was beyond the scope of this study, the extraction time was additionally checked by the preliminary evaluation of the extraction kinetics at the optimal SFE-

ro

CO2 parameters (Fig. 3). It can be observed that the kinetic profile of the experiments fitted with the

-p

previously developed models [37, 38]. Sovová [39] separated 3 stages of the SFE kinetics curves; all of them may be clearly seen in Fig 3. The extraction rate was high in the beginning of SFE-CO2 (stage of

re

constant extraction rate); 68% of the yield was reached during 15 min. Convection is the main mass transfer mechanism from the solid to the fluid phase during the first stage. Afterwards the extraction rate

lP

decreased due to the gradual exhaustion of the free solute, and finally entered the falling extraction rate phase; 96% of the extractable mass had been recovered after 90 min. When all the easily extractable

ur na

solute is exhausted, the diffusion-controlled stage begins where the extraction rate can be seen as an almost-straight line (90 - 240 min). The sufficient time for the recovery of bulk lipophilic components was found to be between 90 and 120 min, just before the third (diffusion-controlled) extraction stage,

Jo

when 96% and 98% of the extractable mass had already been recovered.

3.1.2. Recovery of carotenoids

The extraction yield is only one of the indicators for the assessing SFE-CO2 efficiency. Another important indicator, especially when extracting ingredients for nutraceuticals and functional foods, is the 14

amount of target bioactive compounds recovered from the raw materials. In this regard, carotenoids are important bioactive compounds present in the lipophilic extract obtained from rowanberry pomace. The content of total carotenoids and -carotene recovered from dried rowanberry pomace by Soxhlet extraction was 78.91± 3.50 mg/100 g dw (1283.8 mg/100 g extract) and 38.69 ± 1.43 mg/ 100g dw (629.1 mg/100 g extract), respectively. In this study, gradual recovery of -carotene and total carotenoids in lipophilic extracts was measured

of

at optimal SFE-CO2 conditions (P=45 MPa; T=60 °C). Furthermore, the recovery of carotenoids was also tested at 2 sets of min/max temperature and pressure values, i.e. P= 25 MPa; T= 60 °C, and P= 45

ro

MPa; T= 40 °C. For obtaining larger amounts of test samples for carotenoids determination, the

-p

extractions were up scaled using a 0.5 L extractor and prolonged extraction time (360 min); flow rate was kept constant 2 SL/min.

re

Depending on the extraction time, at optimal SFE-CO2 conditions for maximal extraction yield (P= 45 MPa; T= 60 °C, time 360 min), the total carotenoid content in the extracts varied from 512.4 to 1913.9

lP

mg/100 g and -carotene content from 282.4 to 976.8mg/100 g, whereas the total carotenoid and carotene recovery as compared with Soxhlet extraction constituted 49.7% and 52.5%, respectively (Table

ur na

4). It should be noted, that at optimal SFE-CO2 conditions when higher capacity extractor (500 cm3) was used the lipophilic extract yield was slightly lower (4.62%) compared to the yield obtained in a smaller (50 cm3) extractor (4.80%). It also had an effect on the recovery results. Therefore, in case of 4.80%

Jo

extraction yield the calculated carotenoid recovery would be slightly higher, 51.7% and 54.6% for the total carotenoid and -carotene, respectively. It is also interesting noting that the concentration of -carotene and total carotenoid in the lipophilic

extract gradually increased with increased extraction time (Table 4). The highest concentration of carotenoids was measured in the extracts collected between 240-300 and 300-360 min of extraction, whereas the highest lipophilic extract yield was obtained between 30-180 min of extraction. The lowest 15

content of total carotenoid and -carotene was measured in the extract collected during the first 30 min of extraction (Table 4). It indicates that transfer of carotenoids in SFE-CO2 proceeds more slowly than other lipophilic constituents, e.g. triacylglycerols. The experiment carried out at lower extraction temperature (P=45 MPa; T=40 °C, t=360 min) resulted in the lipophilic extract yield of 4.61 g/100 g dw, whereas the extraction carried out at lower extraction pressure (P=25 MPa; T=60 °C, t=360 min) resulted in the lowest lipophilic extract yield of

of

4.33 g/100 g dw. The total carotenoid content extracted from rowanberry pomace at 40 °C was 30.3±1.01 mg/100 g dw and -carotene content –15.9±0.52 mg/100 g dw, whereas the total carotenoid and -

ro

carotene content extracted from pomace at 60 °C was 27.8±0.84 mg/100 g dw and 14.7±0.44 mg/100 g

with the optimal parameters for the highest yield.

-p

dw, respectively. It means that both parameter sets resulted in remarkably lower recoveries comparing

re

The total carotenoid recovery at the lowest pressure 25 MPa and the highest temperature 60 °C was

lP

the lowest (35.2%), whereas reducing temperature to 40 °C at the highest pressure (45 MPa) results in the increase of total carotenoid recovery to 38.4% (Fig. 4). It shows that pressure is more significant factor influencing the recovery of carotenoids (Fig. 4). Previously, Prado et al. [34] concluded that in

ur na

SFE-CO2, pressure is the most important factor influencing recovery of carotenoids, since carotenoids are large molecules with a low vapor pressure; these considerations are in agreement with our findings. The data obtained in this study suggest that optimal SFE-CO2 conditions for the highest extraction

Jo

yield are also suitable for good carotenoid and β-carotene recovery in the extract. Rowanberry fruits are rich source of carotenoids, a natural highly conjugated class of pigments. However, the importance of carotenoids in food goes beyond their coloring properties. Scientific evidence suggests that carotenoidrich foods significantly reduce the risk of various chronic diseases [40]. The total carotenoid and βcarotene content in rowanberries can vary in a wide range from 3.9 to 265.9 mg/ 100 g dw and from 2.3 to 126.2 mg/ 100 g dw, respectively [16]. 16

In our study, in the investigated extracts β-carotene comprised from 44 to 61% of the total carotenoids present (52% on average). The data obtained in this study indicate that SFE-CO2 parameters influence the recovery of β-carotene and total carotenoid from rowanberry pomace. In some cases higher concentration of bioactive substances could minimize or eliminate the need for further downstream unit operations that are used to concentrate and/or purify target compounds. However, recoveries of low polarity β-carotene and total carotenoids achieved in our study were average, ≤ 52.5%. Previously, when

of

equilibrium solubility of β-carotene and co-solutes (α-tocopherol, palmitic acid and capsaicin) in CO2 at supercritical conditions have been studied, it was found that β-carotene shows the lowest solubility

ro

values for any system tested [41].

-p

It can be hypothesized that composition of plant matrix may also influence the recovery of lipophilic extract including carotenoids. In our study, when SFE-CO2 extraction residue of rowanberry pomace

re

obtained at optimal conditions was re-extracted in a Soxhlet apparatus additional 1.28% of lipophilic extract was recovered, which was still rich in total carotenoid 28.14±1.14 mg/100 g dw (2198.1±88.92

lP

mg/100 g extract) and β-carotene 12.35±0.51 mg/100 g dw (965.2±39.61 mg/100 g extract). It shows that even prolonged SFE-CO2 at optimized conditions was not sufficiently efficient to fully recover all

ur na

lipophilic compounds and carotenoids.

In some previously reported studies the recoveries of carotenoids by SFE-CO2 from different plant materials were higher than obtained in our study. For example extremely high contents of carotenoids

Jo

were recovered from carrot peels, paprika and pumpkin (86, 86 and 76%, respectively) [33,42,43]. On other hand the majority of the research studies on SFE-CO2 of carotenoid-rich plant materials reported recoveries of carotenoids no greater than 32-37% [33]. Furthermore, in substantial number of studies (including some with the highest recoveries achieved) co-solvent (usually ethanol) was used and was reported to be an important factor influencing the recovery of carotenoids by SFE-CO2 [33, 43, 44].

17

Contrary, Mezzomo and Ferreira [44] in their review article concluded that in case of extracting nonpolar compounds such as carotenoids the use of ethanol as a co-solvent is not the best option for increasing the solubility of carotenoids in CO2. In addition, the use of ethanol may have similar disadvantages as conventional extraction since it has to be removed from the extract obtained, which normally involves heat. Thermal treatment may compromise the stability and bioactivity of extracted

of

compounds.

ro

3.2. Pressurized liquid extraction (PLE)

The chemical nature of compositional elements of berry pomace is very different. Lipids and proteins

-p

are mostly concentrated in berry seeds, whereas remaining pomace fraction is rich in polyphenolics, carbohydrates and fiber [9,11].

re

In this study sequential PLE extraction with acetone (PLE-Act), ethanol (PLE-EtOH) and water

lP

(PLE-H2O) was applied for isolation of polar compounds from the rowanberry pomace residue remaining after the optimized SFE-CO2. The yields of PLE-Act, PLE-EtOH and PLE-H2O at 70 °C were 11.9, 16.0

ur na

and 5.1%, respectively (Table 5). Recalculation to the initial mass of pomace gave slightly lower values, 11.3, 15.3 and 4.9% respectively; thus, the total extraction yield of SFE-CO2 at optimal conditions and sequential PLE with increased polarity solvents was 36.3%. To the best of our knowledge there are no reports on the application of PLE to rowanberry pomace.

Jo

PLE with 50% EtOH of raspberry pomace residue after SFE-CO2 resulted in extraction yield of 25.2% [11]. High PLE yields were obtained from the SFE-CO2 residue of black chokeberry pomace, 17.9 and 22.7%, for extraction with acetone and ethanol respectively [10]. However, in the study of Grunovaitė et al. [10] the solvents were used separately, but not sequentially, like in our study. PLE-EtOH of sea buckthorn residue after SFE-CO2 resulted in a considerably lower extraction yield of 15.8% [9]. Sequential re-extraction of guelder-rose berry residues after SFE-CO2 by pressurized acetone, water and 18

ethanol yielded 9.9, 20.8 and 16.8% of extracts when recalculated for the initial mass of pomace [35]. Thus, the sum of PLE extracts isolated from V. opulus pomace (47.5%) was somewhat higher than it was obtained in our study from rowanberry pomace (31.5%). It can be concluded that extraction yields obtained by PLE depends not only on extraction conditions but also on the composition and structure of extracted raw material.

ro

3.3.1. Fatty acid composition and peroxide value of lipophilic extracts

of

3.3. Evaluation of extract composition and properties

-p

Seed triacylglycerols constitute the main part of berry pomace lipophilic substances; therefore, fatty acid composition is one of the main characteristics of SFE-CO2 extract (Table 6). It may be observed

re

that fatty acid composition of hexane extract was almost similar although statistical data handling indicates significant differences for 3 fatty acids. This is in agreement with previously published results

lP

reporting that there are no major differences in the fatty acid profiles of plant oils extracted by either SFE or n-hexane [45]. In addition, quantitatively minor long chain fatty acids, cis-11-eicosenoic (C20:1)

ur na

and eicosapentaenoic (C20:5n3), were detected exclusively in the lipophilic extract obtained using Soxhlet method, while eicosanoic (C20:0) acid was found only in the extract obtained by SFE-CO2. Linoleic (18:2n6c) and oleic (C18:1n9c) fatty acids were the major ones in the extracts and constituted 86% of the total quantified fatty acids. The third major fatty acid was palmitic (9% of the

Jo

total fatty acids), whereas the content of other fatty acids was below 1% except for stearic acid, which constituted 1.49 and 1.58% in the extract obtained by Soxhlet and SFE-CO2, respectively. It may be noted that the content of α-linolenic acid, which is one of the major fatty acids in some berry seeds [19], was less than 1% and this is in agreement with the previously reported data showing that this acid is almost absent in the seed oil of Sorbus aucuparia species [18,19]. Although the content of omega-3 fatty acids is considered as an important indicator of nutritional oil 19

quality, high content of n-6 fatty acids together with high content of carotenoids in the lipophilic fraction suggests that it might have a potential as cosmetic agent for skin care products [46]. Moreover, health claim that linoleic acid contributes to the maintenance of normal blood cholesterol is approved by the EC Regulation [47]. Consequently, lipophilic rowanberry pomace extracts may be considered as promising products for food and cosmetic applications. Oxidative stability is an important factor that determines the potential uses of the lipophilic extracts

of

(oils) obtained. It mainly depends on the fatty acid composition and the presence of antioxidants. Peroxide value of fresh SFE-CO2 extract obtained from rowanberry pomace was quite high 15.7±0.46

ro

meq/kg. For comparison, the reported peroxide value of raspberry pomace extract obtained by SFE-CO2

-p

was significantly lower, 0.62 meq/kg [11]. However, significantly higher peroxide values were reported for cold-pressed seed oils of various berries, including raspberries, from 8.8 to 10.6 meq/kg [48] and

re

from 0.6 to 44 meq/kg [49]. These values indicate a wide range of oxidation levels of berry seed oils and lipophilic fractions separated from berry pomaces. In general, the acceptable peroxide value for refined

lP

vegetable oils is bellow 10 meq/kg. On other hand, cold-pressed virgin olive oils are characterized by increased peroxide values (up to 20 meq/kg) as a result of protein, mineral, and other impurities being

ur na

left after oil extraction, which favor oxidation processes [48].

3.3.2 Antioxidant properties of extracts

Jo

Determination of the in vitro antioxidant activity of isolated extracts is widely used and may be considered as an important tool for preliminary evaluation of their biological activity. The antioxidant capacity values of isolated extracts are presented in Table 5. The TPC and in vitro antioxidant capacity values of lipophilic extracts and hydrophilic extract obtained by PLE indicate that the majority of antioxidatively active phenolic compounds after SFE-CO2 remained in the pomace residue (Table 5). The TPC, DPPH• and ABTS•+ scavenging capacity values of 20

extracts obtained by PLE from rowanberry pomace residue were significantly higher than that of hexane and SFE-CO2 extracts. Similar observations were reported by other scientists who noted that lipophilic extracts obtained from berry pomace possess significantly lower antioxidant activity compared to hydrophilic extracts [10,11]. Sequential extraction of berry pomace with different polarity solvents enables to obtain several fractions with different properties and composition. In our study, PLE extracts obtained with acetone and

of

ethanol possessed almost identical TPC values, 44.1 and 44.9 mg/GAE g, respectively, whereas TPC of PLE-H2O extract was more than two times higher, 108.7 mg/GAE g (Table 5). Similarly, the DPPH• and

ro

ABTS•+ scavenging capacity values of extract obtained by PLE-H2O were also up to 7 and 4 times higher

-p

compared to respective values measured for extracts obtained using acetone and ethanol (Table 5). Significantly higher antioxidant capacity of PLE-H2O extract obtained in our study can partly be

re

explained by its high content of proanthocyanidins (condensed tannins) (10.7 mg/CE g). The total proanthocyanidins content in PLE-Act extract was approx. 11 times lower and in PLE-EtOH extract

lP

almost 4 times lower (Table 5). Similarly to the results obtained in our study, when SFE-CO2 residue of V. opulus berry pomace was sequentially extracted with pressurized acetone, ethanol and water, the

ur na

extract obtained using water also demonstrated the highest TPC value and antioxidant capacity [35]. The TPC of hydrophilic extract obtained from rowanberry pomace residue was higher compared to previously reported results for extract of raspberry pomace isolated by PLE using methanol (26.3-40.0

Jo

mg/GAE g) [11] and extracts isolated from SFE-CO2 extraction residue of black chokeberry pomace by PLE using acetone (35.4 mg/GAE g) [10]. In general, the results of the study demonstrate that PLE is an effective method for the isolation of different polarity, antioxidant-rich fractions from rowanberry pomace.

3.4. Chemical composition of pomace residue after extractions 21

The basic chemical composition of rowanberry pomace before and after the extractions is given in Table 7. Previously reported fat and protein content of dried rowanberry pomace was 3.97 and 7.09g/100 g dw, respectively [36], which is somewhat lower than determined in the current study (6.15 and 10.04% dw, respectively). The differences can be attributed not only to different raw material, i.e. berry genotype, but also different technological treatments during juice pressing.

of

After SFE-CO2 extraction (removal of lipophilic fraction) the content of crude proteins, crude fiber and sugars in the pomace residue was slightly higher than in the initial material, showing that after SFE-

ro

CO2 the majority of macronutrients such as proteins and sugars present in the rowanberry pomace, remain

-p

in the residual mass. These findings are in agreement with previously reported data for SFE-CO2 of carrot peels [33]; however, it was observed that the use of increased addition of co-solvent (EtOH) led to higher

re

transfer of proteins and carbohydrates into the extract [33]. Crude protein content in the pomace residue after PLE was significantly higher (by 73%) compared to the pomace before extractions.

lP

PLE extraction of pomace remaining after SFE-CO2 resulted in total reduction of sugars within the extraction residue i.e., high polarity solvents (ethanol and water) extracted very polar compounds

ur na

(saccharose, glucose and fructose) and presumably some other carbohydrates, together with phenolic antioxidants present in the rowanberry pomace. The PLE pomace residue also contained significantly higher amount of crude fiber, consisting mainly of cellulose, pentosans, insoluble lignin and other minor

Jo

insoluble components. The crude fiber content after PLE was approximately two times higher compared to initial content found in ground rowanberry pomace. It should be noted that in the current study only part of the dietary fiber was evaluated (crude fiber), whereas previously reported total dietary fiber content of dried rowanberry pomace was one of the highest among investigated berry pomaces, approx. 660g/kg dw [36]. Therefore, it may be reasonably expected that the total dietary fiber content of the pomace after extraction procedures would also be higher. 22

A number of applications could be further investigated for the solid residues left after extraction processes. Pomace residue after PLE could be used as a starting material for the extraction of proteins, whereas fiber rich fraction can be further utilized in different food products, i.e. bread, cookies, extruded snacks [50]. In addition, berry pomace residue left after PLE can be further transformed via fermentation to further increase the recovery of soluble constituents and extraction yields [9].

of

4. Conclusion

SFE-CO2 and PLE were demonstrated to be effective methods for isolation of carotenoid rich

ro

(lipophilic) and antioxidant-rich (hydrophilic) fractions from rowanberry pomace. Application of RSM

-p

enabled to optimize SFE-CO2 parameters for obtaining the highest lipophilic extract yield, which at 45 MPa pressure and 60 °C temperature and 180 min extraction time reached was 4.80%.

re

The recovery of total carotenoids by SFE-CO2 was up to 49.7%. The main carotenoid in the lipophilic extracts obtained from rowanberry pomace was β-carotene, which comprised on average 52% of the total

lP

carotenoid content. The changes in SFE-CO2 extraction parameters influenced total carotenoid and βcarotene content in lipophilic extracts, thus enabling to increase the concentration of these compounds

ur na

in the final product.

Consecutive PLE extraction of pomace residue left after SFE-CO2 enabled to recover 33.07% of polyphenol-rich extract possessing strong antioxidant capacity. Consequently, rowanberry pomace

Jo

should be regarded as potential source of functional ingredients for food and non-food uses and, in addition, developed biorefining process may increase the sustainability of juice production from this type of fruits.

Declaration of interests The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. 23

Acknowledgements

The research Project No. 09.3.3-LMT-K-712-02-0121 was funded under the European Social Fund measure „Strengthening the Skills and Capacities of Public Sector Researchers for Engaging in High

Jo

ur na

lP

re

-p

ro

of

Level R&D Activities“ and administered by the Research Council of Lithuania.

24

References

[1] N. Ersoy, M. Gudogdu, S. Ercisli, Folia Hortic. 17 (2017) 179–186. https://doi.org/10.1515/fhort2017-0017 [2] S. Serce, S. Ercisli, M. Sengul, K. Gunduz, E. Orhan, Pharmacogn. Mag. 6 (2010) 9–12. doi: 10.4103/0973-1296.59960 [3] K. Gunduz, H. Ozbay, Turk. J Agric. For. 42 (2018) 145–153. doi: 10.3906/tar-1706-65

of

[4] N.A. Sagar, S. Pareek, S. Sharma, E.M. Yahia, M.G. Lobo, Compr. Rev. Food Sci. Food Saf. 17

ro

(2018) 512–531. https://doi.org/10.1111/1541-4337.12330

[5] A. Altemimi, N. Lakhssassi, A. Baharlouei, D.G. Watson, D.A. Lightfoot, Phytochemicals:

[6]

D.

Urbonaviciene,

P.

Viskelis,

LWT-Food

Sci.

Technol.

85

(2017),

517–523.

lP

https://doi.org/10.1016/j.lwt.2017.03.034

re

(2017) 42. https://doi.org/10.3390/plants6040042

-p

Extraction, Isolation, and identification of bioactive compounds from plant extracts, Plants (Basel). 6

[7] D. Dobravalskytė, P.R. Venskutonis, T. Talou, B. Zebib, O. Merah, O. Ragažinskienė, Rec. Nat.

ur na

'Prod. 7 (2013) 201–209. https://hdl.handle.net/20.500.12259/49791 [8] J. Ndayishimiye, D.J. Lim, B.S. Chun, J. Ind. Eng. Chem. 57 (2018) 339–48. 10.1016/j.jiec.2017.08.041

[9] V. Kitrytė, D. Povilaitis, V. Kraujalienė, V. Šulniūtė, A. Pukalskas, R. Venskutonis, LWT-Food Sci.

Jo

Technol. 85 (2017) 534–538. https://doi.org/10.1016/j.lwt.2017.02.041 [10] L. Grunovaitė, M. Pukalskienė, A. Pukalskas, P.R. Venskutonis, J. Funct. Foods 24 (2016) 85–96. https://doi.org/10.1016/j.jff.2016.03.018 [11] N. Kryževičiūtė, P.

Kraujalis, R. Venskutonis, J.

Supercrit. Fluids 108 (2016) 61–68.

https://doi.org/10.1016/j.supflu.2015.10.025

25

[12] A.N. Shikov, O.N. Pozharitskaya, V.G. Makarov, H. Wagner, R. Verpoorte, M. Heinrich, J. Ethnopharmacol. 154 (2014) 481–536. https://doi.org/10.1016/j.jep.2014.04.007 [13] S.H. Häkkinen, S.O. Kärenlampi, I.M. Heinonen, H.M. Mykkänen, A.R. Törrönen, , J. Agric. Food Chem. 47 (1999) 2274–2279. https://doi.org/10.1021/jf9811065 [14] S.E. Fomenko, N.F. Kushnerova, V.G. Sprygin, E.S. Drugova, T.V. Momot, Russ. J. Bioorg. Chem. 42 (2016) 764–769. https://doi.org/10.1134/S1068162016070074

of

[15] P. Kylli, L. Nohynek, R. Puupponen-Pimiä, B. Westerlund-Wikström, G. McDougall, D. Stewart,

ro

M. Heinonen, J. Agric. Food Chem. 58 (2010) 11985–11992. https://doi.org/10.1021/jf102739v [16] M. Mikulic-Petkovsek, A. Slatnar, F. Stampar, R. Veberic, Food Chem. 135 (2012) 2138–2146.

-p

https://doi.org/10.1016/j.foodchem.2012.06.115

[17] K. Zymone, L. Raudone, R. Raudonis, M. Marksa, L. Ivanauskas, V. Janulis, Molecules. 23 (2018)

re

2593. https://doi.org/10.3390/molecules23102593

lP

[18] A. Johansson, P. Laakso, H. Kallio, Z. Lebensm. Unters. Forsch. 204 (1997) 300–307. https://doi.org/10.1007/s002170050

ur na

[19] B. Yang, M. Ahotupa, P. Määttä, K. Kallio, Food Res. Int. 44 (2011) 2009–2017. https://doi.org/10.1016/j.foodres.2011.02.025 [20] N. Nastić, I. Borrás-Linares, J. Lozano-Sánchez, J. Švarc-Gajić, A. Segura-Carretero, J. Ind. Eng. Chem. 68 (2018) 282–92. https://doi.org/10.1016/j.jiec.2018.07.055

Jo

[21] J.N. Moon, A.T. Getachew, A.S.M.T. Haque, P.S. Saravana, Y.J. Cho, D. Nkurunziza, B.S. Chun, J. Ind. Eng. Chem. 69 (2019) 217–24. https://doi.org/10.1016/j.jiec.2018.09.019 [22] W.J. Kim, B. Veriansyah, Y.W. Lee, J. Kim, J.D. Kim, J. Ind. Eng. Chem. 16(3) (2010) 425–30. https://doi.org/10.1016/j.jiec.2009.08.008 [23] B.M. Pedras, M. Nascimento, I. Sá-Nogueira, P. Simões, A. Paiva, S. Barreiros, J. Ind. Eng. Chem. 72 (2019) 453–456. https://doi.org/10.1016/j.jiec.2019.01.001 26

[24]

P.

Kraujalis,

P.R.

Venskutonis,

J.

Supercrit.

Fluids.

73

(2013)

80–86.

https://doi.org/10.1016/j.supflu.2012.11.009 [25] EN ISO 12966, Animal and Vegetable Fats and Oils – Gas chromatography of fatty acid methyl Esters, 2015. https://www.iso.org/standard/63503.html [26] EN ISO 3960, Animal and Vegetable Fats and Oils – Determination of peroxide value – Iodometric (visual) endpoint determination, 2017. https://www.iso.org/standard/71268.html

of

[27] W. Brand-Williams, M.E. Cuvelier, C. Berset, LWT − Food Sci. Technol. 28 (1995) 25–30. https://doi.org/10.1016/S0023-6438(95)80008-5

ro

[28] R. Bobinaitė, P. Viškelis, A. Šarkinas, P.R. Venskutonis, Phytochemical composition, antioxidant

342. https://doi.org/10.1080/19476337.2013.766265

-p

and antimicrobial properties of raspberry fruit, pulp, and marc extracts. CyTA J. Food 11 (2013) 334–

re

[29] R. Re, N. Pellegrini, A. Proteggente, A. Pannala, M. Yang, C. Rice-Evans, Free Radic. Biol. Med.

[30]

K.

Slinkard,

V.L.

lP

26 (1999) 12311237. https://doi.org/10.1016/S0891-5849(98)00315-3 Singleton,

Am.

J.

Enol.

Vitic.

28

(1977)

49–55.

https://www.ajevonline.org/content/28/1/49

ur na

[31] M. Heil, B. Baumann, C. Andary, K.E. Linsenmair, D. McKey D, Naturwissenschaften, 89 (2002) 519–524. https://doi.org/10.1007/s00114-002-0366-3 [32] AOAC Official Method 990.03, Protein (Crude) in Animal Feed, Combustion Method, in Official

Jo

Methods of Analysis of AOAC International, 18th edition, AOAC International, Gaithersburg, MD, 2006. Chapter 4, pp. 30-31. [33] M.A. Lima, D. Charalampopoulos, A. Chatzifragkou, J. Supercrit. Fluids 133 (2018) 94–102. https://doi.org/10.1016/j.supflu.2017.09.028 [34] J.M. Prado, P.C. Veggi, M.A.A. Meireles, Curr. Anal. Chem. 10 (2014) 29–66. DOI: 10.2174/1573411011410010005 27

[35] P. Kraujalis, V. Kraujalienė, R. Kazernavičiūtė, P. R. Venskutonis, J. Supercrit. Fluids 122 (2017) 99–108. https://doi.org/10.1016/j.supflu.2016.12.008 [36] A. Reißner, S. Al‐Hamimi, A. Quiles, C. Schmidt, S. Struck, I. Hernando, C. Turner, H. Rohm, J. Sci. Food Agric. (2018). https://doi.org/10.1002/jsfa.9302 [37]

M.M.

Esquivel,

M.G.

Bernardo-gil,

M.B.

King,

Fluids

16

(1999)

43–58.

https://doi.org/10.1016/S0896-8446(99)00014-5

of

[38] J. Martínez, A.R. Monteiro, P.T.V. Rosa, M.O.M. Marques, M.A.A. Meireles, Ind. Eng. Chem. Res. 105 (2003) 1057–1063. https://doi.org/10.1021/ie020694f

ro

[39] H. Sovová, J. Supercrit. Fluids 129 (2017) 3–8. https://doi.org/10.1016/j.supflu.2017.02.014

-p

[40] J. Fiedor, K. Burda, Nutrients 6 (2014) 466–488. https://doi.org/10.3390/nu6020466 [41] A. Zúñiga-Moreno, M. G. Arenas-Quevedo, O. Elizalde-Solis, J. Supercrit. Fluids 148 (2019) 66–

[42]

A.

Ambrogi,

D.A.

Cardarelli,

re

73, https://doi.org/10.1016/j.supflu.2019.03.005 R.

Eggers,

J.

Food

Sci.

67

(2002)

3236–

lP

3241. https://doi.org/10.1111/j.1365-2621.2002.tb09572.x

[43] X. Shi, H. Wu, J. Shi, S.J. Xue, D. Wang, W. Wang, A. Cheng, Z. Gong, X. Chen, C. Wang, LWT-

[44]

N.

ur na

Food Sci. Technol. 51 (2013) 433–440. https://doi.org/10.1016/j.lwt.2012.11.003 Mezzomo,

S.R.S.

Ferreira,

J.

Chem.

(2016),

Article

ID

3164312.

http://dx.doi.org/10.1155/2016/3164312 A.A.

Mariod,

M.

Bertrand,

I.

Maznah,

J.

Oil

Fat

Ind.

88

(2011)

931–

Jo

[45]

935. https://doi.org/10.1007/s11746-010-1754-z [46] A.R. Vaughn, A.K. Clark, R.K. Sivamani, V.Y. Shi, Am. J. Clin. Dermatol, 19 (2018): 103–117. https://doi.org/10.1007/s40257-017-0301-1

28

[47] Commission Regulation (EU) No 432/2012 of 16 May 2012 establishing a list of permitted health claims made on foods, other than those referring to the reduction of disease risk and to children’s development and health, Text with EEA relevance OJ L 136, 25.5.2012, pp. 1–40. [48] M. Pieszka, W. Migdał, R. Gąsior, M. Rudzińska, D. Bederska-Łojewska, M. Pieszka, P. Szczurek, J. Chem. (2015) Article ID 659541.https://doi.org/10.1155/2015/659541 [49] V. Van Hoed, N. De Clercq, C. Echim, M. Andjelkovic, E. Leber, K. Dewettinck, V. Roland, J.

of

Food Lipids 16 (2009) 33–49. https://doi.org/10.1111/j.1745-4522.2009.01130.x [50] H. Rohm, C. Brennan, C. Turner, E. Günther, G. Campbell, I. Hernando, S. Struck, V. Kontogiorgos,

Jo

ur na

lP

re

-p

ro

Foods 4 (2015) 690–697. https://doi.org/10.3390/foods4040690

29

Figure captions

Figure 1. Response surface plot of SFE-CO2 showing the effect of independent variables on extraction yield. Figure 2. Comparison of predicted values versus actual values for the lipophilic extract yield. Figure 3. Kinetics of rowanberry pomace SFE-CO2 process at optimal extraction parameters (45MPa,

of

60°C).

Jo

ur na

lP

re

-p

ro

Figure 4. Kinetics of total carotenoid recovery at different SFE-CO2 conditions.

30

of

ro

-p

re

lP

ur na

Jo Fig. 1.

31

of

ro

-p

re

lP

ur na

Jo Fig. 2.

32

4 3 2 1

of

Extraction yield, g/100 g dw

5

0

15 30 45 60 75 90 105 120 135 150 165 180 195 210 225 240 255

Extraction time, min

Jo

ur na

lP

re

-p

Fig. 3.

ro

0

33

45 MPa 60 °C

25 MPa 60 °C

45 MPa 40 °C

50 40 30 20 10 0 0

30

60

90

120

150

180

210

240

270

300

330

360

ro

Extraction time, min

of

Total carotenoid recovery, %

60

Jo

ur na

lP

re

-p

Fig. 4.

34

Tables

Table 1 Factors selected as independent variables for optimization of SFE-CO2 parameters. Coded levels

Extraction temperature (T) Extraction pressure (P)

-α

-1

0a

1

+α

°C

36

40

50

60

64

MPa

20.9

25.0

35.0

45.0

49.1

Levels -α and +α are axial star points representing extreme values for each analytical factor (selected by the software Design-

Jo

ur na

lP

re

-p

Expert), while analytical factors values at the levels -1 and +1 are chose.

ro

a

Units

of

Independent variables

35

Table 2.

T,

P,

°C

MPa

Total lipophilic extract yield

Total lipophilic

(g/100 g dw)

extract recovery, %

Predicted

Experimental

40

25.0

4.11

4.12± 0.12

67.0± 1.98

2

60

25.0

4.61

4.69± 0.13

76.4± 2.19

3

40

45.0

4.62

4.55± 0.05

74.0± 0.82

4

60

45.0

4.80

4.80± 0.12

78.1± 1.95

5

36

35.0

4.36

4.40± 0.10

6

64

35.0

4.84

4.78± 0.05

7

50

20.9

4.23

4.17± 0.08

67.9± 1.25

8

50

49.1

4.72

4.77± 0.11

77.5± 1.85

9

50

35.0

4.28

4.30± 0.10

69.9± 1.64

10

50

35.0

4.28

4.38± 0.09

71.2± 1.49

11

50

35.0

4.28

4.22± 0.10

68.7± 1.67

12

50

35.0

4.28

4.23± 0.03

68.8± 0.42

13

50

35.0

4.28

4.26± 0.05

69.3± 0.87

-p 71.7± 1.66 77.8± 0.82

re

lP

ur na

ro

1

of

Experiment No.

Fully coded CCD and results obtained for extract yield and total lipophilic extract recoverya.

a

Values are represented as a

Jo

mean ± standard deviation (n3).

36

Table 3.

df

MS

F-value

p-value

Model

0.71

5

0.14

25.68

<0.0002*

Temperature (T)

0.23

1

0.23

41.38

0.0004*

Pressure (P)

0.24

1

0.24

43.30

0.0003*

TP

0.026

1

0.026

4.60

0.0692**

T2

0.17

1

0.17

31.40

0.0008*

P2

0.067

1

0.067

12.13

0.0102*

Residual

0.039

7

0.0056

Lack of fit

0.022

3

0.0074

1.74

0.2961**

Pure error

0.017

4

0.0042

Total SS

0.75

12

-p

ro

SS

of

Analysis of variance of the regression parameters of SFE-CO2 yield from rowanberry pomace.

*: significant effect; **: not significant effect; F: Fisher value; MS: mean square;

Jo

ur na

lP

re

SS: sum of squares.

37

Table 4. SFE-CO2 extract yields, β-carotene content and total carotenoid content and recovery in the extracts obtained at optimal extraction parameters (45 MPa, 60 °C, t=360 min) a. β-carotene content,

Total carotenoid content

Carotenoid recovery

min

g/100 g dw

mg/100 g extract

mg/100 g extract

%

10

0.28 ± 0.01bc

282.4 ± 19.6a

512.4 ± 35.52a

2.96 ± 0.12bc

20

0.36 ± 0.01d

334.1 ± 21.1ab

625.4 ± 39.41ab

3.87 ± 0.18c

30

0.35 ± 0.00d

301.4 ± 12.0ab

560.2 ± 22.38ab

3.76 ± 0.10c

60

0.85 ± 0.03f

363.9 ± 34.8b

683.1 ± 65.37b

9.16 ± 0.71d

120

1.40 ± 0.02g

329.5 ± 20.0ab

663.0 ± 40.27b

15.03 ± 0.71e

180

0.76 ± 0.04e

511.1 ± 17.7c

1005.0 ± 34.73c

8.19 ± 0.34d

240

0.32 ± 0.01cd

769.3 ± 36.1d

1563.6 ± 73.31d

3.43 ± 0.30bc

300

0.23 ± 0.00b

953.1 ± 24.6e

360

0.07 ± 0.01a

976.8 ± 28.6e

Total

4.62 ± 0.14

433.4 ± 23.28

-p

ro

of

Lipophilic extract yield

2.52 ± 0.15b

1913.9 ± 56.02e

0.79 ± 0.05a

848.5 ± 45.27

49.71 ± 2.65

re

1902.4 ± 49.08e

Values presented as means (n=3) ± standard deviation. Different letters within the same column indicate significant

Jo

ur na

differences between means (p≤ 0.05).

lP

a

Time,

38

Table 5 The yields, total proanthocyanidins content and antioxidant properties of rowanberry pomace extracts a. Yield

Proanthocyanidins

TPC

DPPH•

ABTS•+

g/100 g dw

mg/ ECE g

mg/ GAE g

µmol TE/g

µmol TE/g

extract

extract

extract

extract

Extracts 6.15 ± 0.24b

nd

3.01 ± 0.15a

9.43 ± 0.19a

11.10 ± 0.24a

SFE-CO2

4.80 ± 0.09a

nd

2.90 ± 0.05a

6.72 ± 0.31a

9.22 ± 0.36a

PLE-Actb

11.90 ± 0.76c

0.97 ± 0.05a

44.13 ± 0.75b

87.70 ± 1.08b

333.27 ± 4.58b

PLE-EtOHb

16.04 ± 0.28d

2.98 ± 0.15b

44.86 ± 1.24b

129.47 ± 2.73c

389.80 ± 1.60b

PLE-H2Ob

5.13 ± 0.11ab

10.72± 0.21c

108.7 ± 4.44c

582.88 ± 32.86d

1306.20 ± 51.40c

Values presented as means (n=3) ± standard deviation. Different letters within the same column indicate significant

ro

a

of

Soxhlet (Hex)

differences between means (p≤ 0.05).

-p

PLE of residue left after SFE-CO2.

Jo

ur na

lP

re

b

39

Table 6 Fatty acid composition of rowanberry pomace extracts obtained by Soxhlet extraction and SFE-CO2a. Soxhlet

Content, %

Myristic, C14:0

0.36 ± 0.03a

0.35 ± 0.08a

2

Palmitic, C16:0

9.00 ± 0.30a

9.29 ± 0.99a

3

Palmitoleic, C16:1

0.25 ± 0.01a

0.27 ± 0.01a

4

Stearic, C18:0

1.49 ± 0.04a

1.58 ± 0.05a

5

Oleic, C18:1n9c

25.25 ± 0.13b

26.98 ± 0.42a

6

Linoleic, C18:2n6c

61.18 ± 0.13a

58.88 ± 0.41b

7

α-Linolenic, C18:3n3

0.88 ± 0.00b

0.96 ±0.00a

8

Eicosanoic, C20:0

−

0.48 ±0.21a

9

cis-11-Eicosenoic, C20:1

0.23 ± 0.00a

−

10

Eicosapentaenoic, C20:5n3

0.47 ± 0.09a

−

11

Behenic, C22:0

0.42 ± 0.06a

0.38 ± 0.10a

re

-p

ro

1

Values presented as means (n=3) ± standard deviation. Different letters within the same line indicate significant differences

Jo

ur na

between means (p≤ 0.05).

lP

a

Fatty acid

of

No.

SFE-CO2

40

Table 7 Chemical composition of rowanberry pomace and pomace residues after the SFE-CO2 and PLEa. Protein Rowanberry pomace

Crude fiber

%

Saccharose

Glucose

Fructose

%

Before extraction

10.04 ± 0.45a

0.50 ± 0.02a

3.29±0.04a

2.50± 0.05a

19.57 ± 0.50a

Residual after SFE-CO2

11.84 ± 0.10b

0.54± 0.04a

3.45±0.06b

2.72± 0.00b

22.14 ± 0.67b

Residual after PLE

17.34 ±0.54c

nd

nd

nd

42.64 ± 1.92c

Values presented as means (n=3) ± standard deviation. Different letters within the same column indicate significant

of

a

Sugars, %

Jo

ur na

lP

re

-p

ro

differences between means (p≤ 0.05).

41