The present invention relates to a method for producing an antioxidant composition from potatoes and antioxidant compositions obtainable thereby.

BACKGROUND TO THE INVENTION

Potatoes (Solatium tuberosum) constitute an important staple food. Potatoes are consumed directly but, increasingly, processed potato products such as French fries, chips, mashed potato and crisps represent the majority of consumption. Potato processing may result in 10-25 % of raw product being discharged as waste. Such waste potato by-products include potato peel and potato rasp, a semi-solid material of the potato tuber that remains after removing the starch and most of the liquid (potato fruit juice).

Potato peel is known to contain phenolic acids including hydroxycinnamic acids and hydroxybenzoic acids such as chlorogenic acid, caffeic acid, p-coumaric acid, ferulic acid, protocatechuic acid, gallic acid, salicylic acid and vanillic acid. These phenolic acids have antioxidant and antibacterial properties. Accordingly, attempts have been made to extract phenolic acids from potato peel in order to produce natural antioxidant compositions. Solvent extraction techniques have been employed, such as using methanol with a degree of success. It is an aim of the present invention to provide improved methods for producing antioxidant compositions from potatoes.

SUMMARY OF THE INVENTION

The present invention provides a method for producing an antioxidant composition from potatoes, which method comprises (a) processing the potatoes to generate potato by-product comprising potato peel or potato rasp by (i) steaming potato tubers and removing the peel, or (ii) isolating the rasp from processed potatoes and treating the rasp with a polysaccharide hydrolytic enzyme; and (b) treating the potato by-product with aqueous ethanol to form an extract comprising the antioxidant composition. It has surprisingly been found that when potato by-product such as potato peel or potato rasp is pre-treated prior to extraction with aqueous ethanol, yields of antioxidants are significantly increased. Without wishing to be bound by theory, it is thought that pre-treatment of the peel by steaming the potato tubers or pre-treatment of the rasp with a polysaccharide hydrolytic enzyme, weakens the plant cell wall matrix and thereby enhances the release of antioxidants such as phenolic acids.

The aqueous ethanol used to treat the potato by-product typically has an ethanol concentration in the range of from 20 to 80 % by weight. Advantageously, the ethanol concentration in the aqueous ethanol is in the range of from 40 to 80 % by weight, more advantageously 45 to 75 % by weight, preferably from 55 to 65 % by weight, more preferably 55 to 60 % by weight where the extraction efficiency for phenolic acids is at its greatest.

The amount of aqueous ethanol used to treat the potato by-product is typically at least 10ml aqueous ethanol/g (dry solids) potato by-product. The yield of phenolic acids tends to increase as the ratio of aqueous ethanol to potato by-product increases. Advantageously, the amount of aqueous ethanol used to treat the potato by-product is at least 20ml aqueous ethanol/g (dry solids) potato by-product, preferably at least 25ml aqueous ethanol/g (dry solids) potato byproduct. Amounts up to 30ml aqueous ethanol/g (dry solids) by-product were found to be useful according to the invention although higher amounts may also be used.

In one arrangement, the aqueous ethanol is acidified aqueous ethanol. An acid, such as hydrochloric acid for example in an amount of 1 % by weight, may be added to the aqueous ethanol. Advantageously, the pH of the acidified aqueous ethanol is in the range 1 to 2, such as 1.2 to 1.3. It is found that at acidic pH the yield of phenolic acids may be increased.

The temperature at which the potato by-product is treated with the aqueous ethanol is not found to be particularly critical. Temperatures in the range of from 20°C to 60°C were found to be effective although other temperatures may be used.

The step of treating the potato by-product to form the extract may be carried out by simply mixing the aqueous ethanol with the potato by-product to enable release of antioxidants by leaching or maceration. This may be typically carried out in a vessel with a magnetic stirrer. Typical times employed may be from 30 minutes to 1 hour. It is also possible to intensify the extraction conditions. In one arrangement, microwave assisted extraction may be employed by subjecting the potato by-product to microwaves of a suitable frequency. Useful conditions include power set at 100 W for 1 to 3 minutes. Ultrasound assisted extraction may also be employed. The potato by-product is subjected to ultrasound for example using a sonifier typically set at 40 % power for continuous treatment for 30 minutes.

Generally, intensification of the treatment of the potato by-product with aqueous ethanol increases the yield of phenolic acids.

The step (a) of processing the potatoes typically has as its purpose to product potato-based products for human consumption such as potato flakes which may be used to form mashed potato or cut pieces which may be used to make fries or chips. Ordinarily, the potato byproduct described herein would constitute a waste product from such processes.

Once the potatoes have been processed to generate potato by-product, the by-product may be treated directly with the aqueous ethanol as a fresh sample or may be stored and treated later. For example, the potato by-product may be freeze-dried and may be stored in frozen form until required. The frozen potato by-product may be treated directly with aqueous ethanol or may be thawed before use. The potato by-product may be comminuted after freeze-drying to increase the surface area before treating with the aqueous ethanol. Comminution may be effected by grinding the freeze-dried potato by-product. Advantageously, the potato byproduct comprises potato peel. Potato peel has been found to contain a higher concentration of antioxidant compounds such as phenolic acids as compared with potato rasp. When the potatoes are processed to generate the potato peel by steaming potato tubers, the peel may be removed by subjecting the surface of the tubers to pressurised water impact. Typically, the softened surface of the tuber is flushed with high speed water flow to release the outer layer. Peel material may be collected as a sludge.

The potato peel may be treated with a polysaccharide hydrolytic enzyme prior to treatment with aqueous ethanol. As with the rasp, this enzymatic treatment increases the yield of antioxidants such as phenolic acids. The polysaccharide hydrolytic enzyme may be selected from cellulases, hemicellulases, pectinases, xylanases, glucanases, arabinases, polygalacturonases, pectin esterases and mixtures thereof. Such enzymes are commercially available. For example, MethaPlus L100 is a mixture of cellulase, xylanase and beta- glucanase. Pectinex Ultra contains a mixture of pectinases where pol ygal acturonase is the main activity. Viscozyme L contains a mixture of arabinase, cellulase, beta-glucanase, hemicellulase and xylanase.

The antioxidant composition obtainable from the method according to the invention typically comprises chlorogenic acid as a major component and caffeic acid as a minor component when the potato by-product is potato peel. When the potato by-product is rasp, caffeic acid is a major component and chlorogenic acid a minor component. Thus, the antioxidant composition comprises phenolic acids including chlorogenic acid and/or caffeic acid. The extract comprising the antioxidant composition may be further concentrated. The concentrated antioxidant composition may comprise the extract in an amount of at least 25mg GAE/g extract, for example in the range 25 mg to 40mg GAE/g extract, preferably at least 30mg GAE/g extract. It is found that as the concentration of extract increases, the ability to inhibit oxidative processes such as rancification increases.

Such products may be used as a natural ingredient having antioxidant or radical scavenging activity, for example in foodstuffs and other products.

There is therefore provided use of the antioxidant composition as an antioxidant additive for a product, wherein the composition provides at least 50 μg GAE/g product on a dry solids basis. The composition may provide up to 80 μg GAE/g or up to 100 μg GAE/g product on a dry solids basis.

DETAILED DESCRIPTION

The invention will now be described in further detail by way of example only, with reference to the accompanying drawings, in which:

Figure 1 shows extraction yield of phenolics from freeze-dried peel and flesh of hand-peeled potato; Figure 2 shows the effect of ethanol concentration on extracted phenolics from homogenized whole red-skinned potato;

Figure 3 shows the effect of varying wet sample: solvent ratios on the extraction efficiency of phenolics from red-skinned potato;

Figure 4A shows an effect plot of factors tested in 3 ^3"1 design of experiment for ethanolic extraction of phenolics from freeze-dried potato rasp;

Figure 4B shows a contour plot of the extractions from Figure 4A, modelled from multilinear regression of total phenolic content yield data. The dashed line indicates an arbitrary minimum extraction yield of 4000 g GAE (gallic acid equivalents) / gDS;

Figure 5 shows the effect of a two-stage extraction in 60 % ethanol of total phenolics from two preparations of freeze-dried potato rasp;

Figure 6 shows a comparison of total phenolic extracts from freeze-dried potato rasp using regular and acidified (1 % HC1) ethanol;

Figure 7 shows yield of total phenolics extracted in 60 % ethanol from freeze-dried preparations of potato rasp and steamed potato peel;

Figure 8 shows enzyme pre-treatment of freeze-dried peel and rasp using three hydrolytic enzyme preparations to enhance extraction yield of TPC ^g GAE/g DS);

Figure 9 shows extraction yield of phenolic compounds from potato material using various methods for yield enhancement compared to standard extraction conditions;

Figure 10 shows scale-up of standard extraction conditions using freeze-dried and wet peel;

Figure 11 shows identification of the major phenolic compounds, chlorogenic acid (CQA) and caffeic acid (CA) in ethanolic extracts of rasp and steamed peel by HPLC-UV analysis (top) and by HPLC-MS (bottom, MRM); and Figure 12 shows the effect of various doses of concentrated potato peel extract (EPPE) on inhibiting the rancification of sunflower oil under accelerated oxidation.

MATERIALS AND METHODS

Potato material

Potato peel (PP) was obtained fresh from the processing line for flakes and fries at HOFF SA, Gj0vik. The peeling process is in form of steaming the tuber then the softened surface is flushed with high-speed waterflow to release the outer layer. The material is collected as a sludge and pumped continuously into an open holding tank. Samples were collected consecutive times over several months and stored in plastic bags of 0.5 litre, immediately frozen at -20°C.

Potato rasp (PR) was obtained was obtained fresh from the processing line (conveyor screw) for starch at HOFF SA, Brumunddal. Rasp is the semi-solid material of the tuber that remains after removing the starch and most of the liquid (potato fruit juice). Rasp samples were collected during the fall campaign and stored at -20°C.

For comparison and initial labscale testing red and yellow skin potato were obtained from the local market. Whole tubers or the peel from tubers were diced and homogenized. To arrest browning by polyphenoloxidase citric acid was added to 0.4 % (v/v) final concentration.

Chemicals

As extraction solvent was used absolute ethanol (EtOH) lab grade (100 %, Kemetyl Norge AS) or rectified ethanol distillate (ca 80 %), an intermediate fraction from the alcohol distillery at HOFF SA, Gj0vik).

Fine chemicals for analysis were obtained from Sigma Aldrich (St.Louis, MO); DPPH (D8132), Folin Ciocalteu phenol reagent 2N (F9252), caffeic acid (C0625), chlorogenic acid (C3878), ferulic acid (127031 1), gallic acid (G7384), p-coumaric acid (C9008) and quercetin dehydrate (Q0125).

Enzyme preparations used containing cellulase, hemicellulase and/or pectinase activities: Viscozyme L and Pectinex Ultra SP-L (both products of Novozymes, Bagsvaerd, Denmark) and MethaPlus L100 (DSM, Heerlen, the Netherlands).

Material preparation

Aliquots of frozen peel and rasp were freeze dried (Christ Alpha 1 -4 freeze dryer) and the dried material ground to powder and sifted through a 0.5 mm mesh. The dried material was stored cold.

Extraction conditions tested

Test material was used either in its freeze dried form or in thawed wet form. Water-ethanol mixtures (20-100 %) were prepared using alcometer suitable for the ethanol strength tested.

Leaching (maceration MAC): Solvent leaching was tested in 30 ml extraction volume in closed vessels under magnetic stirring. Aqueous ethanol was added either to dry material at ratio between 1 : 10 to 1:30 (3.0-1.0 g per 30 ml) or to wet material at ratio between 1 : 1 to 1 :5 ( 15.0- 6.0 g per 30 ml). Incubation temperatures were tested between 20 to 60°C and between 30 minutes to 1 hour. In one experiment the EtOH -water mixture was acidified with 1 % HC1.

Microwave assisted extraction (MAE): In MAE, the 30 ml extraction contained 60 % EtOH added to dried potato material at 1 :30 (1.0 g). Reactions were carried out in a sealed 50 ml Teflon vessel (Bohlender GmbH, Grunsfeld, Germany). The power was set at 100W for 1, 2 and 3 minutes. The treated content was cooled before further handling. The inner lining of the reaction vessel was thoroughly cleaned between extractions.

Ultrasound assisted extraction (UAE): A 30 ml extraction mixture of 60 % EtOH and dried potato material at 1 :30 (1.0 g) was treated with ultrasound from a Branson 250-450 Sonifier (Danbury, CT) set at 40 % power and a continuous duty cycle for 30 minutes. Due to some solvent evaporation during the open treatment, the extraction volume was brought back to 30 ml after cooling.

Enzyme assisted extraction (EAE): Dried potato powder ( 1.0 g) was suspended in 50 niM citrate buffer pH 4.8-5.0. Three commercially available fibre degrading enzymes (containing cellulase, hemicellulase and/or pectinase activities) were added to a final dosage of 1.0 % (v/v). The reactions were incubated in closed vessels under shaking at 50°C for 30 minutes. Following the enzyme pretreatment, 45 ml of 96 % EtOH was added to obtain final concentration of 60 % EtOH in a reaction volume of 75 ml. The leaching process was under stirring for 60 minutes at room temperature.

Post-extraction treatments

After extraction the mixtures were clarified by centrifugation at 5,000 g (Beckman Coulter Allegra 25R Centrifuge), and the supernatant carefully collected. Remaining particles were removed by filtration through Whatman filter paper in a Buchner funnel (70 mm) under vacuum. Filtrates were collected in 50 ml plastic tubes and stored at -20°C before further analysis.

Analytical methods

Total phenolic content (TPC): The content of phenolic compounds in the extracts was determined using their reaction with the Folin-Ciocalteu reagent (Waterhouse, 2001). As phenolic standard gallic acid was used and TPC noted in microgram ^g) gallic acid equivalents (GAE).

Antioxidant activity (DPPH): The antioxidant activity of extracts was assayed by their ability to quench radical formation from the substrate a,a-diphenyl-P-picrylhydrazyl (Siatka & Kasparova, 2010). The antioxidant effect was observed as the reduction in absorbance at 517 nm due to inhibition of the radical formation. Inhibition was calculated according to the formulae:

% DPPH inhibition = [Abs (Ctrl) -Abs (Sample)]/ Abs (Ctrl) xlOO %. The negative control was assayed with 60 % EtOH instead of unknown sample. For reference were used caffeic acid, chlorogenic acid, ferulic acid, p-coumaric acid and quercetin.

Identification of phenolic acids determination by HPLC: To assess the composition of extracts especially with respect to the most frequently reported hydroxycinnamic acids (HCA), analysis by HPLC-UV (Ultimate 3000, Thermo Scientific) was employed. The sample separation was carried out on a KINETEX C18 column (Phenomenex) and monitoring at 330 nm. The mobile phase was water with 0.1 % trifluoracetic acid (TFA) and 5 % acetonitrile (ACN). Identification of peaks in the primary analysis was attempted by comparing the retention time to known standards. These peaks were later confirmed in a secondary, more refined analysis carried out by an independent contract lab (VITAS AS, Oslo) using LC-MS and HPLC-UV with diode array.

Statistical methods

ANOVA (analysis of variance) was performed on TPC results at p< 0.05 to determine significant differences between extraction conditions and sample materials.

In factorial experimental designs (Design of Experiments, DoE), multilinear regression (MLR) was employed in the modelling and to detect main and interaction effects of tested variables, using the MODDE 1 1 (Umetrics AB, Umea, Sweden) statistical package. Significance of effects was tested by ANOVA (p<0.05).

RESULTS

Comparing TPC in flesh and peel

Red skinned tubers were peeled manually at a thickness of ca 1 mm. Peel and flesh fractions were then freeze-dried separately to obtain a dry powder. One gram of freeze-dried potato material was subjected to 1 hr extraction (maceration) at 60 °C in 40 ml 60 % aqueous EtOH. Following separation and clarification, the filtrates were analyzed by the TPC assay. Figure 1 shows the extraction yield of phenolics from freeze dried peel and flesh of hand-peeled potato. The peel fraction (4.7 mg /g DW) contained fourfold more of the phenolic substances than the flesh fraction ( 1.2 mg/g DW). Yellow skinned potatoes were also tested the same way and showed the same distribution (data not shown).

Screening of extraction conditions using fresh tubers

Based on various reports into alcoholic extractions of plant materials, factors of influence to the extraction yield are temperature, solvent strength (or concentration) and the sample mass to solvent ratio. Effect of solvent strength (% EtOH) and of the mixing ratio were screened in individual experiments (one factor at a time; OF AT). Extractions at room temperature were performed with whole homogenized tuber (red skin). For a range of water-EtOH mixtures (20, 40, 60, 75 and 100% EtOH) a mixing ratio of 1 part wet sample (10 ml) to 4 part solvent (40 ml) was used, and extractions made in parallel, with stirring. Following separation and clarification the TPC content of filtrates were assayed in parallel as μg GAE /ml. Figure 2 shows the effect of ethanol concentration on TPC for each of the two extraction series from homogenized whole red skinned potato, bars indicating standard error (STDV) in assayed TPC. Although there was some difference between the extraction series, both showed maximum extracted phenolics and with the least error at 60 % EtOH. This strength was chosen for next experiments.

Homogenized red skinned tuber (10 ml) was mixed at various ratios (1 : 1, 1 :2, 1 :3, 1 :4 and 1 :5) with 60 % EtOH and extracted under stirring for one hour at room temperature. The extractions were set up with successively increasing total volumes (20 to 60 ml), in two parallel series. The clarified filtrates were assayed in parallel and the total phenolic content extracted as function of increasing solvent volume (Figure 3). The intra-series error of TPC assayed was less than the error between each series, thus, the result is shown as the mean of both extraction series. The extraction yield from 10 ml of tuber homogenate approaches 500 μg GAE and above a mixing ratio 1 :4 only minor increase is observed.

Screening of extraction conditions using potato rasp by-product

The effect of temperature was tested together with the previous parameters EtOH strength and mass-to-solvent ratio in a factorial design, varying all three factors at the same time, so-called Design of Experiments, but keeping extraction time at 1 hr. The design was a fractional factorial of a 3 ^3-1 design consisting of 12 experiments ((3x3) + 3 center points = 12)) as depicted in Table 1. Extraction temperature was tested at 20, 40 and 60 °C and EtOH at 55, 60, 65 %. For this set of experiments, freeze-dried potato rasp from the starch production was used. Based on the estimated dry matter content (ca 13 %) of fresh rasp a mass of freeze dried rasp material was used corresponding to the mid and upper range tested with whole tuber (1 :2-5). Rasp (1.0 g) to solvent was tested at ratio 1 : 10, 1 :20 and 1 :30 keeping the solvent volume constant.

Figure 4(A) shows an effect plot of factors tested in a 3 ^Λ (3-1) DoE for ethanolic extraction of total phenolics yield from freeze-dried potato rasp; solvent-to-dry solids (SlvDS), solvent strength (%EtOH) and temperature. Figure 4 (B) shows a predictive contour plot of the same extractions, modelled by MLR of yield data. The line designated Min indicates an arbitrary minimum extraction yield (4000 μg GAE / g DS).

The main effects of three factors are shown in figure 4, where the largest effect was observed for increasing volume of solvent, up to 30 ml pr gram dry material, whereas solvent strength had a slight negative effect, being more efficient at 55 % EtOH. Temperature had no significant effect on the extraction yield over the range 20 to 60 °C. Figure 4 illustrates a response surface at 20°C modelled from the screening where the more intense shading indicates the area of highest yield of extracted phenolics can be expected. It was concluded that for simple extraction by maceration, solvent concentrations 55 to 60 % EtOH and solids to solvent ratio above 1 :25 provided best conditions, even at room temperature.

In a further refinement with freeze-dried potato rasp the effect of two-stage separation (Figure 5), and the use of acidified ethanol (Figure 6) was tested. Figure 5 shows the effect of two- stage extraction (n=2) in 60 % EtOH of total phenolics from two preparations ( 1 & 2) of freeze- dried potato rasp. Figure 6 shows a comparison of total phenolic extraction (n=2) from freeze- dried potato rasp using regular and acidified (1% HCI) ethanol. In two-stage extraction, where the residual pellet (2.5 g) after filtration was re-suspended in 30 ml 60 % EtOH and extracted again, on the average 85 % of TPC was extracted in the first stage. Acidified EtOH (1 % HCI) increased the TPC yield by 30 % compared to regular 60 % EtOH. Extraction of total phenolics from steamed potato peel

The extraction yields obtained from whole tuber and freeze-dried potato rasp were all in the range 3-5 mg GAE / g DS. However, the initial test comparing hand-made peel and tuber flesh indicated a higher content of phenolics in the peel fraction. Therefore, an industrial peel byproduct fraction was identified from the flakes and fries manufacturing at the HOFF Gj0vik- plant, where heat through steaming and hydraulic abrasion was applied to the tuber. In the subsequent extractions, this potato peel fraction was used applying the same set of extraction conditions investigated for whole tuber and rasp. The steamed peel was prepared as a freeze- dried powder as described above.

A comparison was made using standard extraction conditions (60 % EtOH, 1:30 mass:solvent, 20 °C, lhr) of freeze-dried preparations of rasp and steamed peel (n=2) (Figure 7). The yield of phenolic (mg GAE /g DS) in the peel fraction was then 3.5-fold higher (12.4) compared to the rasp (3.5).

Enzyme pretreatment

Since the potato processing by-products rasp and steamed peel mainly consist of plant cell wall fibers and complex polysaccarides (cellulose, hemicellulose, pectin), a partial degradation from polysaccharide hydrolytic enzymes (cellulases, hemicellulases, pectinases) was applied as a pretreatment. It was assumed this treatment should accommodate for enhanced penetration of the extracting solvent and a faster release of embedded phenolics. Three commercially available enzyme preparations were used, known for containing various hydrolytic activities; MethaPlus LI 00, Pectinex Ultra and Viscozyme. Enzyme dosage was 1.0 % (v/v) in 50 mM citrate buffer pH 4.8-5.0, and incubation under mixing for 30 minutes at 50°C. To the digestate 40 ml 100 % EtOH was added to give a total volume of 70 ml of 60 % EtOH. Extraction then followed standard conditions.

Figure 8 shows enzyme pretreatment of freeze-dried peel and rasp using three hydrolytic enzyme preparations (n=l), to enhance extraction yield of TPC ^g GAE / g DS). The extraction yield of TPC increased significantly by the enzyme pretreatment for both for rasp and for steamed peel (Figure 8) compared to untreated material (Figure 7), 89 % and 18 % respectively. Although Viscozyme showed the highest outcome on rasp, the difference between the enzyme preparations was not significant (F 2,2 = 9,21, p < 0.05) in their ability to enhance the yield.

Intensifications of the standard phenolics extraction

In addition to the standard extraction conditions a set of process intensifications were investigated; ultrasound assisted extraction (UAE), microwave-assisted extraction (MAE). The same set of amplification methods was previously tested with rasp (data not shown), and based on the outcome of these the settings were applied to the steamed peel fraction. A comprehensive comparison of the extraction methods on the yield of phenolic compounds, including enzyme pretreatment was then made encompassing whole tuber homogenate, hand-made-peel, rasp and steamed peel (Figure 9).

Figure 9 shows extraction yield (n=2) of phenolic compounds from potato material using various methods for yield enhancement compared to standard extraction (Std.) conditions. UAE, ultra sound; MAE, microwave; Enz, enzyme pretreatment. The following conditions apply: reaction volume was 30 ml for standard extraction, UAE and MAE, but for steamed peels the volume was increased to 40 ml to better facilitate stirring. UAE was at 40% power and continues pulse for 30 minutes; MAE was at 100 W and 3 minutes. Enzymatic pretreatment was performed with MethaPlus L100 (1.0 % v/v) and only for steamed peel and rasp. All extractions were made in duplicate and analyzed for TPC.

It is evident that steamed potato peel provided the highest yielding source of phenolic compounds of all tested materials. The intensifications gave some variation in yield ultrasound being more effective. Most significant increase from standard conditions was with the enzyme pretreatment, corresponding to the enhancements described above (Figure 8). However, the more striking result is the large difference observed between hand-made peel and the steamed peel, the latter being 2-3 times higher in yield. This signifies the positive influence short-term heat treatment apparently has on weakening the plant cell wall matrix and thereby enhance the release of phenolics. Enzyme pretreatment accelerate this process further, but is not crucial for the yield when UAE and MAE intensifications are compared to standard extraction. Scale up of extraction using wet steamed peel

To validate the yield of phenolics from steamed peel a bench scale up of the standard extraction conditions was performed. Anticipating that at large scale the by-product would be subject to extraction directly, wet peel was used in these experiments. From multiple samples of peel obtained over 3 months the dry substance (DS) varied between 8 and 16 % (mean 12.0 ± 2.6, n = 6). The DS of peel material used was ca 15 %, thus the corresponding weight of 1.0 g dry material was 6.7 g. The water content was assumed to be included as part of the aqueous phase of the solvent, and the added solvent strength adjusted accordingly.

Figure 10 shows scale up of standard extraction conditions using freeze-dried (lx) and wet peel (5x and 25x). Scaling up was done in two steps of 5x increase of material and reaction volume, i.e. 30, 150 and 750 ml. To 33.0 g wet peel (5.0 g DS) holding 28.1 ml water was added 32.0 ml water and 90.0 ml of 100% EtOH to obtain a final volume of 150.0 ml. To 167.0 g wet peel (25.0 g DS) holding 142.0 ml water was added 158.0 ml water and 450.0 ml 100 % EtOH to obtain a final volume of 750.0 ml. The scaled up extractions were performed under same conditions as for the 30 ml volume with freeze-dried material, and in glass vessels of similar geometry. Due to lack of material, the 750ml reaction was without duplicate. Following multiple centrifugations and filtration, the recovered filtrates were analyzed for TPC (Figure 10). The yields (12.2 mg/ g DS) were comparable throughout the range of scale-up tested, showing no significant difference.

Radical scavenging activity of extracts

To assess the antioxidant acti vity of the extracts, their ability to quench or scavenge radicals in a photometric assay was assayed using DPPH. DPPH is a stable radical absorbing at 765 nm. Upon reaction with a scavenging compound like phenolics it loses colour, which is observed as a function of time. The antioxidant (AOX) activity is calculated as % inhibition. Assaying various amounts of the tested extract, it is possible to obtain the amount sufficient for 50 % inhibition (IC50) of DPPH from a plot of μg GAE added versus percent assay inhibition. The lower the value of IC50 the more potent is its AOX activity. In Table 2 are given the AOX activity of extracts of steamed peel and of potato rasp obtained through the standard and enhanced extraction methods. The IC50 values for peel is throughout 50 - 70 % less than that of the rasp, indicating the higher content of scavenging compounds, i.e. phenolics and thus 2-3 times more active. This implies that peel is a more potent source of natural antioxidants, and values are comparable to synthetic antioxidants and even better than those reported for potato peel in the literature.

Identification of phenolic compounds

The characterization of phenolic content in typical extracts was obtained by HPLC separation and detection by UV and MS. The scientific literature reports 3-4 different hydroxycinnamic acids in potato flesh and peel; chlorogenic acid (CQA), caffeic acid (CA), ferulic acid (FA) and para-coumaric acid (pCA). Under the current experimentation, only CQA and CA were targeted for identification, as these probable were the most prominent phenolics. This assumption does not exclude other phenolics to be present in the material and in the extracts. Figure 11 shows identification of major phenolic compounds chlorogenic acid (CQA) and caffeic acid (CA) in ethanolic extracts of rasp and steamed peel, by HPLC-UV analysis (top) and by HPLC-MS (bottom, MRM). Reference standards are shown in top panel. These high- resolution chromatograms show that in rasp CA was a major species (peak B); peak A was also noticeable but unidentified as were peaks C, D, E, F. No significant peak of CQA was identified in rasp. Steamed peel contained CQA as the one major species (peak B), whereas peak C identical to CA was only minor; peaks A and D unidentified. From the analysis it was evident that extracts of steamed potato peel was notably high in CQA, while in potato rasp multiple types of phenolics occurred, where CA appeared to be in high amount.

Assessing content of total glvcoalkaloids in extracts

A preliminary assessment has been carried out on the content of total glykoalkaloids (TGA = solanin + chakonin). However, it was observed that the ethanol extraction process also co- extracted these non-desirable substances, but pending more accurate analysis, TGA levels are not reported here. Preparation of concentrated extract

Concentrated preparations of the potato peel extracts obtained from the scale up experiments (150 and 750 ml), were made by vacuum evaporation of the alcoholic solvent (Rotavapor, BUchi Labortechnik AG, Flawil, Switserland). Extracts (0.41 ± 0.03 mg GAE /ml) were concentrated 60- 100 times at 40-50 °C to a syrup of greenish hue and no smell. Combination of batches resulted in a syrup called ΈΡΡΕ', having TPC content of ca 32 mg GAE/ g syrup.

Testing of concentrated potato peel extract as antioxidant in sunflower oil

The evaporated potato peel extract (EPPE) was used in experiments testing its potential of inhibiting oxidation in an oily product. Sunflower oil was selected as test matrix because of its frequent use in in food applications (frying etc.). An accelerated thermal oxidation protocol was employed observing the resulting oil stability (Samarin et al. 2012; Vieira and Regitano- d'Arce 2001). For reference were used BHA as synthetic antioxidant, and a commercial preparation of Rosemary extract as natural antioxidant.

Samples (200 g) of sunflower oil were prepared in triplicate with and without additives. Additives were EPPE at concentrations 200, 1600 and 2400 mg/kg oil (ppm), Rosemary extract at 1600 ppm, and BHA at 200 ppm; total of 6 series including negative control (no additive). Tween-20 (500 μΐ in 1 ml water) was used as emulgator of EPPE and the mixture sonicated with the oil. The glass beakers with oil were incubated in dark at 63 °C for 21 days, while aliquots of 40 ml were withdrawn at start and days 6, 12, 14, 16, and 21. The capacity of the additives to inhibit fatty acid oxidation was evaluated by assaying the primary oxidation parameters peroxide value (PV) and anisidine value (AV), where the sum of the two results in the Totox value.

Figure 12 shows inhibition of accelerated oxidation at 63 °C in sunflower oil as change in Totox-value (n=3), when added butylated hydroxyanisol (BHA), Rosemary extract (RE), or concentrated potato peel extract (EPPE). The Figure shows the outcome of a 3- week experiment of controlled accelerated oxidation as the change in Totox value. The antioxidant effect of increasing the EPPE dosage to the oil, i.e. delaying the increase in Totox, showed a significant linear negative correlation for the range 200 to 2400 ppm. At 1600 ppm and higher the antioxidant effect of EPPE was comparable to Rosemary extract and even better than for synthetic BHA at 200 ppm. Thus, the example indicates that an effective inhibitory dosage of 1600 ppm EPPE (1600 mg / kg product) having a TPC content of 32 mg GAE, will correspond to a final concentration in the product of 0.051 mg GAE / g product (i.e. approx. 50 μg / g) on a dry solids basis.

CONCLUSIONS

A one-stage ethanolic extraction proved most efficient method to stable yields of total phenolics (i.e. HCA). Critical process variables were screened and initially optimized by use of Design of Experiments. The peel (fries/flakes) appeared to be the best source of HCA ( 12- 13 mg TP / g DS). Enzyme pretreatment increased yield by 30 %, and UAE by 1 1 %. Total phenolics in rasp was only 1/3-1/4 that of peel. While steamed peel predominantly contains CQA, rasp was higher in CA. Further increase in yields can be anticipated by a combination of enzyme pretreatment, acidified ethanol and by two-stage extraction. The significant capacity of a concentrated ethanolic extract to inhibit rancification in sunflower oil has been demonstrated, and the effect is comparable to both that of synthetic and commercial natural antioxidants.

References

Samarin, A. M., Poorazarang, H., Hematyar, N., & Elhamirad, A. (2012). Phenolics in potato peels: Extraction and utilization as natural antioxidants. World Applied Sciences Journal, 18(2), 191-195.

Siatka, T., & Kasparova, M. (2010). Seasonal Variation in Total Phenolic and Flavonoid Contents and DPPH Scavenging Activity of Bellis perennis L. Flowers. Molecules, 15(12), 9450.

Vieira, T. M. F. S., & Regitano-d'Arce, M. A. B. (2001). Canola Oil Thermal Oxidation During Oven Test and Microwave Heating. LWT - Food Science and Technology, 34(4), 215-221. doi: http://dx.doi.org/10.1006/fstl.2000.0735

Waterhouse, A. L. (2001). Determination of Total Phenolics Current Protocols in Food Analytical Chemistry: John Wiley & Sons, Inc.

obtained by the inhibition of DPPH.

standard maceration

Extraction

Without Enzyme Microwave Ultrasound method

pretreat. pretreat. assisted assisted

Steamed peel 44,7 51,1 50,5 55,2

Potato rasp 108,4 162,0 103,2 152,4