Food Chemistry 125 (2011) 1028–1036
`
`Contents lists available at ScienceDirect
`
`Food Chemistry
`
`j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / f o o d c h e m
`
`Extraction and characterisation of lipids from Antarctic krill (Euphausia superba)
`Joseph C. Gigliotti a, Matthew P. Davenport b, Sarah K. Beamer a, Janet C. Tou a, J. Jaczynski a,⇑
`
`a West Virginia University, Division of Animal and Nutritional Sciences, P.O. Box 6108, Morgantown, WV 26506-6108, United States
`b University of Alaska Fairbanks, School of Fisheries and Ocean Sciences, Fishery Industrial Technology Centre, 118 Trident Way, Kodiak, AK 99615, United States
`
`a r t i c l e
`
`i n f o
`
`a b s t r a c t
`
`Article history:
`Received 18 June 2010
`Received in revised form 27 September
`2010
`Accepted 3 October 2010
`
`Keywords:
`Krill oil
`Oil extraction
`Fatty acid profile
`Lipid composition
`Omega-3 polyunsaturated fatty acids
`Nutraceutical food products
`Functional food
`
`There is significant commercial interest in oil extraction from krill because it is rich in omega-3 polyun-
`saturated fatty acids (n-3 PUFA) such as eicosapentaenoic (EPA, 20:5n3) and docosahexaenoic (DHA,
`22:6n3) acids. The objectives were to determine oil extraction efficiency using different solvent systems
`and the composition of extracted oil and spent krill following extraction. Extraction efficiency was the
`highest (P < 0.05) for one-step extraction using freeze-dried krill with 1:12 or 1:30 krill:solvent ratio
`(w:v) compared to Folch, Soxhlet, and conventional two-step extraction. Extracted oils contained pre-
`dominantly phospholipids (20–33%), polar non-phospholipids (64–77%), and minor triglycerides
`(1–3%). Triglycerides contained much less (P < 0.05) total n-3 (4.0%), DHA (1.1%), and EPA (2.3%), but more
`(P < 0.05) saturated FA (38.7%) than phospholipids (total n-3-47.4%, DHA-18.0%, EPA-28.2%, saturated
`FA-23.5%). Antioxidant capacity of krill oil extracted by one-step extraction (9.4–14.2 lmol Trolox Equiv-
`alents/ml oil) was generally similar to antioxidant capacity of krill oil extracted by ethanol (22.9), but
`greater (P < 0.05) than antioxidant capacity of krill oil extracted by acetone (1.2) and Folch method
`(1.5). The spent krill following oil extraction contained protein (72.9–75.8%, dry basis). Based on the
`extraction efficiency and composition of the extracted oil, the one-step extraction using 1:12 krill:solvent
`ratio is recommended.
`
`Ó 2010 Elsevier Ltd. All rights reserved.
`
`1. Introduction
`
`Antarctic krill (Euphausia superba) are small, shrimp-like crusta-
`ceans. Commercial capture is simple because krill form high-
`density surface swarms. Despite their small size, krill likely has
`the largest biomass of any multi-cellular animal species on earth
`(Nicol, James, & Pitcher, 1987). Although it is difficult to accurately
`determine the sustainable biomass for krill harvest, this significant
`resource may be comparable to the biomass of all other aquatic
`species currently harvested. The total annual capture from all fish-
`eries has been approximately 130 million tons (MT) since 2000
`(FAO, 2007). By comparison, krill biomass has been estimated at
`400–1550 MT with a sustainable harvest at 70–200 MT (Suzuki &
`Shibata, 1990). However, newer estimates suggest that the krill
`biomass may be lower (Priddle, Boyd, Whitehouse, Murphy, &
`Croxall, 1998; Smetacek & Nicol, 2005). Nicol and Foster (2003)
`estimated the annual krill capture to be 0.1 MT, making krill an
`underutilized species. However, due to the role that krill play in
`marine ecology, an internationally monitored and governed eco-
`system approach is a necessity for a long-term sustainability of this
`fishery (Everson, 2000; Hureau, 1985; Laws, 1985).
`
`⇑ Corresponding author. Tel.: +1 304 293 1893; fax: +1 304 293 2232.
`
`E-mail address: Jacek.Jaczynski@mail.wvu.edu (J. Jaczynski).
`
`0308-8146/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
`doi:10.1016/j.foodchem.2010.10.013
`
`Page 1
`
`Grantham (1977) reported that krill contains 77.9–83.1% mois-
`ture, 0.4–3.6% lipids, 11.9–15.4% protein, and 2% chitin and glu-
`cides. Saether, Ellingsen, and Mohr (1986) determined that due
`to seasonality lipid content ranges widely from 12–50% (dry basis).
`Lipid content and its composition in krill also depend on species,
`age, and the time between capture and freezing (Kolakowska,
`1991). Kolakowska, Kolakowski, and Szczygielski (1994) reported
`that the n-3 PUFA, EPA and DHA are particularly abundant, which
`was attributed to krill consuming single-cell marine micro-algae.
`However, shellfish are often perceived as high in cholesterol; and
`therefore, reduce its acceptance as food by consumers. Cholesterol
`level in krill is higher than fish, but lower than shrimp (Tou, Jaczyn-
`ski, & Chen, 2007). Also, it is important to emphasise that two-
`thirds of the sterols in shellfish are non-cholesterol sterols, which
`interfere with absorption of dietary cholesterol (Feeley, Criner, &
`Watt, 1972; Vahouny, Connor, Roy, Lin, & Gallo, 1981).
`Despite its potential as a high quality lipid and protein source
`(Bridges, Gigliotti, Altman, Jaczynski, & Tou, 2010; Chen, Tou, &
`Jaczynski, 2009; Gigliotti, Jaczynski, & Tou, 2008; Gigliotti, Smith,
`Jaczynski, & Tou, 2010; Tou et al., 2007), the use of krill as human
`food has been limited (Suzuki & Shibata, 1990). Krill is mainly used
`by reduction fisheries for manufacture of fish feeds due to its high
`astaxanthin content. In addition, encapsulated krill oil is used as a
`dietary supplement with various potential health benefits includ-
`ing protection against cardiovascular disease (CVD) (Bunea, El
`
`NEPN Ex. 2043
`Aker v. Neptune
`IPR2014-00003
`
`

`
`J.C. Gigliotti et al. / Food Chemistry 125 (2011) 1028–1036
`
`1029
`
`Farrah, & Deutsch, 2004; Sampalis, 2007; Sampalis et al., 2003).
`Bunea et al. (2004) attributed some of these benefits to the n-3
`PUFA in krill being mainly associated with phospholipids (PL); un-
`like in fish where the n-3 PUFA are associated with triglycerides
`(TG). The oxidative stability of krill oil has been attributed to its
`antioxidants content, in particular astaxanthin (Suzuki & Shibata,
`1990). Krill oil may be valuable in the development of nutraceuti-
`cal food products (Kassis, Beamer, Matak, Tou, and Jaczynski, 2010;
`Kassis, Drake, Beamer, Matak, & Jaczynski, 2010; Kassis, Gigliotti,
`Beamer, Tou, and Jaczynski, submitted for publication).
`A major hindrance to commercial processing of krill and develop-
`ment of new krill-based food products may be due to high activity of
`krill lipases and proteases (Anheller, Hellgren, Karlstam, & Vincent,
`1989). These enzymes are released immediately upon the demise of
`krill, resulting in autolysis, which leads to a rapid spoilage. The en-
`zymes combined with its small size makes krill processing for hu-
`man food a significant challenge. Another concern is high fluoride
`content in the exoskeleton. However, centrifugation removes fluo-
`ride (Christians & Leinemann, 1983; Karl et al., 1986).
`Krill oil is currently extracted by two-step solvent extraction
`using acetone and ethanol in the first and second step, respectively
`(Beaudoin & Martin, 2004; Sampalis, 2007). However, this extrac-
`tion requires two separate extraction steps and takes a relatively
`long time. In addition, the two-step extraction does not mention
`water removal from krill prior to oil extraction. Water interferes
`with solvent extraction and water removal prior to oil extraction re-
`sults in greatly improved extraction efficiency and less water in the
`extracted oil (Dunford, Temelli, & LeBlanc, 1997; Nilsson, 1996).
`Another process to extract krill oil takes advantage of superctitic-
`al-CO2 entrained with up to 20% ethanol (Bruheim et al., 2008).
`However, this process requires thermal inactivation of lipases at
`over 50 °C prior to oil extraction. Although heat likely inactivates li-
`pases resulting in reduced hydrolysis of ester bonds and conse-
`quently fewer free FA, it simultaneously denatures heat-labile
`krill muscle proteins (Carvajal, Lanier, & Macdonald, 2005).
`Due to structural changes, the recovery of denatured proteins
`would be difficult and even if krill proteins were recovered, the
`proteins would exhibit reduced functionalities (i.e., gel-forming
`ability, extractability, water-holding-capacity, etc.). Bruheim
`et al.’s (2008) process is similar to the two-step solvent extraction
`(Beaudoin & Martin, 2004; Sampalis, 2007), but does not require
`water removal prior to processing. However, freeze-drying of krill
`prior to oil extraction with superctitical-CO2 has been shown to in-
`crease extraction efficiency approximately three times (Yamaguchi
`et al., 1986). The protein remaining in the residual spent krill fol-
`lowing oil extraction can be recovered with techniques such as iso-
`electric solubilisation/precipitation and if protein functionalities
`are retained, this protein could be used in human food products
`contributing to the fuller use of this tremendous resource (Chen
`& Jaczynski, 2007a,b; Chen et al., 2009; Gehring, Gigliotti, Moritz,
`Tou, & Jaczynski, 2010; Jaczynski, 2010; Torres, Chen, Rodrigo-Gar-
`cia, & Jaczynski, 2007).
`It is hypothesised that one-step extraction with acetone:etha-
`nol mixture for 2 h from whole krill will result in high extraction
`efficiency. The objectives were to determine oil extraction effi-
`ciency from whole krill using different solvent systems and charac-
`terise the composition of extracted lipids and residual spent krill
`following oil extraction.
`
`2. Materials and methods
`
`2.1. Sample preparation and oil extraction
`
`Whole frozen Antarctic krill (Euphausia superba) was obtained
`from Krill Canada (Langley, BC, Canada). The krill blocks were
`transported overnight to the West Virginia University food science
`
`Page 2
`
`laboratory in heavily insulated industrial strength boxes filled with
`dry ice. Upon arrival the boxes were immediately stored at 80 °C
`until use. Whole frozen krill was freeze-dried without thawing
`(VirTis Genesis 35SQ Super XL freeze-dryer, Virtis, Gardiner, NY,
`USA), vacuum-packed and stored at 80 °C until processed. A flow
`diagram of oil extraction from krill is shown is Fig. 1.
`In the one-step extraction, oil was extracted from freeze-dried
`krill using 1:1 acetone:ethanol (v:v) solvent mixture (ACS grade
`acetone, Fisher Scientific, Fairlawn, NJ, USA; ACS grade 95% etha-
`nol, Pharmco, Brookfield, CT, USA). The following krill:solvent ra-
`tios were tested 1:6, 1:9, 1:12, and 1:30 (w:v). The weight of the
`initial freeze-dried krill was recorded in order to determine extrac-
`tion efficiency (see below). The 1:30 ratio was used in the present
`study to mimic the 1:6 ratio of whole fresh krill as described by
`Beaudoin and Martin (2004) as well as Sampalis (2007). Fresh
`whole krill is currently used in commercial oil extraction (Beaudo-
`in & Martin, 2004; Bruheim et al., 2008; Sampalis, 2007). However,
`in the present study freeze-dried krill was used (Dunford et al.,
`1997; Yamaguchi et al., 1986). Therefore, 1:30 ratio (freeze-dried
`krill) used in the present study approximately corresponded to
`1:6 ratio (fresh whole krill) during commercial krill oil extraction
`based on the lipid content in relation to the solvent volume.
`Freeze-dried krill was dispersed in the solvent mixture (ace-
`tone:ethanol) by homogenisation for 30 s using a laboratory blen-
`der (model 51BL31, Waring Commercial, Torrington, CT, USA). Oil
`extraction was conducted for 2 h at 4 °C using a continuous shaker
`(model Excella E25R, New Brunswick Scientific, Edison, NJ, USA)
`followed by centrifugation at 10,000g and 4 °C for 20 min (model
`Sorvall RC-5B Refrigerated Superspeed, Kendro Laboratory Prod-
`ucts, Newtown, CT, USA). The supernatant (i.e., extracted krill oil)
`was decanted and air dried at atmospheric pressure. The sediment
`(residual spent krill including protein, shell, etc.) was also dried
`under air at atmospheric pressure and analysed for crude protein
`(Kjeldahl), total fat (Soxhlet), and ash content (see below).
`In the two-step extraction oil was extracted from freeze-dried
`krill using two separate extractions. The weight of the initial
`freeze-dried krill was recorded in order to determine extraction
`efficiency (see below). The freeze-dried krill was first mixed with
`acetone at a 1:6 ratio (krill:acetone, w:v), centrifuged, and then
`the sediment was mixed with ethanol at a 1:6 ratio (krill:ethanol,
`w:v), followed by final centrifugation. Therefore, two separate ex-
`tracts were obtained. Acetone extract was obtained in step 1 and
`ethanol extract in step 2. In step 1, freeze-dried krill was dispersed
`in acetone by homogenisation for 30 s using the laboratory blen-
`der. Oil extraction in step 1 was conducted for 2 h at 4 °C using
`the continuous shaker followed by centrifugation at 10,000g and
`4 °C for 20 min. The supernatant (i.e. acetone extract) was dec-
`anted and air dried at atmospheric pressure; while the sediment
`was subjected to step 2 of the extraction. Step 2 was conducted
`in the same manner as step 1 except ethanol was used instead of
`acetone. Therefore, total extraction time (i.e., step 1 and 2) was
`4 h. Following step 2, final centrifugation at 10,000g and 4 °C for
`20 min was applied. The supernatant (i.e., ethanol extract) was
`decanted and air dried at atmospheric pressure. The sediment
`(residual spent krill including protein, shell, etc.) was analysed
`for crude protein (Kjeldahl), total fat (Soxhlet), and ash content
`(see below).
`Dry krill oils (i.e., krill oil from one-step extraction, acetone ex-
`tract from two-step extraction, and ethanol extract from two-step
`extraction) were clarified in 2:1 chloroform:methanol (v:v) mixture
`(ACS grade chloroform, Fisher Scientific, Fairlawn, NJ, USA; HPLC
`grade methanol, Fisher Scientific, Fairlawn, NJ, USA) with 20 ml of
`10% NaCl in water added to a separation funnel (Folch, Lees, & Slo-
`ane, 1957). This clarification removed any residual water in the krill
`oil samples. Sufficient volume of the chloroform:methanol mixture
`was added until there was no visible separate layers. Following clar-
`NEPN Ex. 2043
`Aker v. Neptune
`IPR2014-00003
`
`

`
`1030
`
`J.C. Gigliotti et al. / Food Chemistry 125 (2011) 1028–1036
`
`Fig. 1. A flow diagram for oil extraction from krill and subsequent analyses of the extracted oil and residual spent krill.
`
`ification, the krill oil samples were air re-dried at atmospheric pres-
`sure. Dry clarified krill oils were weighed and compared to the initial
`weight of the freeze-dried krill subjected to extraction in order to
`determine extraction efficiency (see below).
`For comparison, oil was also extracted from freeze-dried krill
`using the Folch method (Folch et al., 1957) and Soxhlet extraction
`(AOAC, 1995). In Folch method, 3 g of freeze-dried krill were dis-
`persed in 60 ml of the 2:1 chloroform:methanol mixture by
`homogenisation for 30 s using the laboratory blender, followed
`by filtration through Whatmann #40 paper (Whatman Interna-
`tional, Maidstone, UK) in the separation funnel. A 20 ml aliquot
`of 10% NaCl in water was added to the separation funnel and the
`mixture was manually shaken and allowed to stand until the
`phases had completely separated. The bottom organic solvent
`phase was removed and air dried at atmospheric pressure. The dry-
`ing resulted in solvent evaporation yielding krill oil. The weight of
`krill oil was recorded in order to determine extraction efficiency. In
`Soxhlet extraction, 3 g of freeze-dried krill were dispersed in petro-
`leum ether (ACS grade petroleum ether, VWR International,
`Bridgeport, NJ, USA) (AOAC, 1995). The extraction was carried
`out in a standard Soxhlet apparatus for 18 h ( Chen, Nguyen, Sem-
`mens, Beamer, & Jaczynski, 2006, 2007), followed by air drying at
`atmospheric pressure. The drying resulted in evaporation of petro-
`leum ether yielding krill oil. The weight of krill oil was recorded in
`order to determine extraction efficiency.
`
`In preliminary experiments, the above solvent systems were ap-
`plied to extract oil from fresh krill and freeze-dried krill. The
`extraction efficiency was significantly lower (P < 0.05) for all sol-
`vent systems using fresh krill when compared to freeze-dried krill.
`These preliminary experiments confirmed earlier reports by Yam-
`aguchi et al. (1986) and Dunford et al. (1997). Therefore, only
`freeze-dried krill was used in the present study. All oil extractions
`were performed in triplicate (n = 3) using the same batch of freeze-
`dried krill.
`
`2.2. Determination of extraction efficiency
`
`The weight of the initial freeze-dried krill prior to the extraction
`with different solvent systems was recorded. Following extraction,
`the weight of extracted oil was also recorded. The weight of ace-
`tone extract and ethanol extract were combined in order to deter-
`mine an overall extraction efficiency of the two-step extraction.
`The extraction efficiency was determined according to the follow-
`ing equation:
`
`Extraction efficiency
`weight of extracted krill oilðgÞ
`¼
`weight of freeze-dried krill subjected to extractionðgÞ  100
`ð1Þ
`
`Page 3
`
`NEPN Ex. 2043
`Aker v. Neptune
`IPR2014-00003
`
`

`
`J.C. Gigliotti et al. / Food Chemistry 125 (2011) 1028–1036
`
`1031
`
`separately determined for acetone and ethanol extracts. Cholesterol
`content of oil extracted from krill by one-step, two-step, Folch, and
`Soxhlet procedure was determined using a colorimetric assay (EMD
`Chemicals Inc., Darmstadt, Germany). This cholesterol assay relies
`on the oxidation of cholesterol by added cholesterol oxidase and
`yielding H2O2. The resulting H2O2 interacts with a cholesterol probe
`to produce resorufin that is detected spectrophotometrically.
`Krill oil samples were prepared in a cholesterol reaction buffer.
`In a 96 well microplate, 50 ll of the buffered krill oil samples were
`mixed with 50 ll of reaction mix (containing cholesterol oxidase
`and cholesterol probe). Samples were covered and incubated at
`37 °C for 1 h. Absorbance was read at 570 nm using a Spectramax
`Plus microplate reader (Molecular Devices, Sunnyvale, CA). In order
`to determine cholesterol content in krill oil, standard curve was
`constructed using cholesterol standards mixed with 50 ll of the
`reaction mix. The cholesterol content is reported as mean values
`(± standard deviation) of at least three replicates and the mean val-
`ues are expressed as grams of cholesterol per 100 grams of krill oil.
`
`2.6. Antioxidant capacity of extracted krill oil
`
`It has been reported that krill oil is rich in antioxidants, in par-
`ticular astaxanthin (Kolakowska, 1991; Kolakowska et al., 1994;
`Suzuki & Shibata, 1990; Tou et al., 2007). These antioxidants were
`likely included in the polar non-PL class. Therefore, total antioxi-
`dant capacity for the oil extracted from krill by the different extrac-
`tion methods was determined according to a colorimetric
`antioxidant assay (Cayman Chemical Company, Ann Arbor, MI).
`For krill oil extracted with the two-step extraction, total antioxi-
`dant capacity was separately determined for acetone and ethanol
`extracts. This antioxidant assay relies on the ability of endogenous
`antioxidants in krill oil to inhibit the metmyoglobin-mediated oxi-
`dation of ABTS (2,20-Azino-di-[3-ethylbenzthiazoline sulphonate])
`to ABTS+. ABTS and metmyoglobin were added to krill oil samples
`as reagents and the amount of ABTS+ was determined spectropho-
`tometrically. The capacity of the endogenous antioxidants in krill
`oil to prevent ABTS oxidation was compared with that of a
`water-soluble tocopherol analogue,
`trolox (6-Hydroxy-2,5,7,8-
`tetramethylchroman-2-carboxylic acid).
`In a 96 well microplate, 10 ll of krill oil samples were mixed with
`10 ll of metmyoglobin and 150 ll of ABTS. To initiate the reaction,
`40 ll of H2O2 was added to each well. Samples were covered and
`incubated at room temperature on a shaker for 5 min. Absorbance
`was read at 750 nm using a Spectramax Plus microplate reader
`(Molecular Devices, Sunnyvale, CA). In order to determine trolox
`
`Extraction efficiency is expressed as g of extracted oil per 100 g
`of freeze-dried krill subjected to the extraction.
`
`2.3. Determination of lipid classes in extracted krill oil
`
`Thin layer chromatography (TLC) was applied to resolve lipid
`classes of oils extracted by one-step, two-step, Folch, or Soxhlet.
`For krill oil extracted by the two-step extraction, TLC and subse-
`quent densitometry analysis were applied to acetone and ethanol
`extracts. A 10 ml aliquot of extracted krill oil was dissolved in
`1:1 chloroform:methanol (v:v) and loaded onto TLC plates (What-
`man K6F silica plates with 60 A pore sizes, P.J. Cobert Associates, St.
`Louis, MO). The TLC plates were developed using a 80:20:1.5 hex-
`ane:ether:acetic acid solution (v:v:v) as a mobile phase. Once
`developed, plates were air dried for 5 min.
`Plate images were captured using a digital camera interfaced
`with a PC and spot densitometer (Fluorchem 8000, Alpha Innotech
`Corp., San Leandro, CA) using transluminating white light (Alpha
`Innotech Corp., San Leandro, CA). The images were analysed using
`the Fluorchem software (version 1.0, Alpha Innotech Corp., San
`Leandro, CA). Phospholipids (PL) and triglycerides (TG) were iden-
`tified using RF values obtained from triolein (Sigma–Aldrich, St.
`Louis, MO) and soybean lecithin (Fisher Scientific, Fairlawn, NJ)
`standards. Once identified, the bands corresponding to PL, TG,
`and polar non-PL class were scraped from the TLC plates and sus-
`pended in 1:1 chloroform:methanol (v:v) for determination of fatty
`acid profile (FAP) by gas chromatography (GC). The densitometry
`data are reported as mean values (±standard deviation) of at least
`three replicates and the mean values are expressed as percent of
`lipid class in total krill oil.
`
`2.4. Fatty acid profile (FAP) of extracted krill oil
`
`The PL and TG were separately scraped from the TLC plates. The
`FAP of the PL and TG scrapes was determined (Chen, Nguyen, Sem-
`mens, Beamer, & Jaczynski, 2007, 2008a, 2008b; Folch et al., 1957;
`Taskaya, Chen, Beamer, Tou, & Jaczynski, 2009). FA were transme-
`thylated by the addition of 4 ml of 4 g/100 ml H2SO4 in anhydrous
`methanol and heated in a water bath set at 90 °C for 60 min. The
`mixture was saponified by transferring through a Na2SO4-filled
`glass Pasteur pipette and subsequently dried under N2 in a water
`bath set at 60 °C for 60 min. The FA methyl esters (FAME) were
`re-suspended in filtered iso-octane (HPLC grade iso-octane, Fisher
`Scientific, Fairlawn, NJ, USA). The FAME were analysed by a gas
`chromatograph (GC) (model CP-3800, Varian Analytical Instru-
`ments, Walnut Creek, CA, USA) using flame ionisation detector fit-
`ted with a WCOT-fused silica capillary column (50 m length,
`0.25 mm inside diameter; Varian Analytical Instruments, Walnut
`Creek, CA, USA). Injection and detection temperatures were main-
`tained at 220 °C and column temperature was 190 °C. The station-
`ary phase was CP-Silica 88 (Varian Analytical
`Instruments,
`Walnut Creek, CA, USA). Nitrogen was the carrier gas and a split ra-
`tio of 10 to 1 was used. The FA were identified by comparing their
`retention times with those of known standards and references
`(Ackman, 1980). Peak area and the amount of each FA were com-
`puted by an integrator using the Star GC workstation (version 6,
`Varian Analytical Instruments, Walnut Creek, CA, USA). The data
`are reported as mean values (±standard deviation) of at least three
`replicates and the mean values are expressed as percent of a fatty
`acid in total fatty acids.
`
`2.5. Determination of cholesterol content in extracted krill oil
`
`Due to the prevalence of polar non-PL, the TLC and densitometry
`analysis did not allow determination of cholesterol content. For krill
`oil extracted with the two-step extraction, cholesterol content was
`
`Fig. 2. Extraction efficiency* of different solvent systems for extraction of oil from
`freeze-dried whole krill. *Different letters on the top of data bars indicate significant
`differences (Tukey’s test, P < 0.05) between mean values (±SD, n = 3).
`
`Page 4
`
`NEPN Ex. 2043
`Aker v. Neptune
`IPR2014-00003
`
`

`
`1032
`
`J.C. Gigliotti et al. / Food Chemistry 125 (2011) 1028–1036
`
`Fig. 3. Densitometry analysis* of thin layer chromatography (TLC) plates. For two-step extraction, TLC plates with acetone and ethanol extracts were analysed separately.
`Insert: An example TLC plate showing lipid classes of krill oil extracted with one-step extraction using 1:12 krill:solvent ratio (w:v) (triplicate). TG – triglycerides, PL –
`phospholipids, CHOL – cholesterol. *Different letters on the top of data bars indicate significant differences (Tukey’s test, P < 0.05) between mean values (±SD, n = 3) within a
`lipid class.
`
`equivalent (TE) for krill oil, standard curve was constructed using
`trolox standards mixed with 10 ll of metmyoglobin and 150 ll of
`ABTS. The antioxidant capacity is reported as mean values (±stan-
`dard deviation) of at least three replicates and the mean values are
`expressed as lmols of TE per ml of krill oil.
`
`2.7. Proximate analysis of the residual spent krill following oil
`extraction
`
`To determine proximate composition (i.e., crude protein, total
`fat, and ash content) on dry basis, the residual spent krill was ana-
`lysed for moisture content. A sample (2 g) was placed on an alumin-
`ium dish (Fisher Scientific, Fairlawn, NJ, USA), spread evenly across
`the dish and oven dried (105 °C for 24 h) (AOAC, 1995). The crude
`protein, ash content, and total fat were determined in the residual
`spent krill following oil extraction with one- and two-step extrac-
`tion only (i.e., sediment in Fig. 1). Crude protein was determined
`by Kjeldahl assay (AOAC, 1995). Ash content was determined by
`incinerating samples in a muffle furnace at 550 °C for 24 h (AOAC,
`1995). Residual fat content was determined according to the Soxhlet
`extraction method (AOAC, 1995). The proximate data are reported
`as mean values (±standard deviation) of at least three replicates
`and the mean values are expressed as grams per 100 g of the residual
`spent krill following oil extraction (dry weight basis).
`
`2.8. Statistical analysis
`
`The oil extraction experiments were performed in triplicate
`(n = 3). For each triplicate, at least three measurements were per-
`formed. One-way independent measures analyses of variance (AN-
`OVA) were used to determine individual differences between
`treatments except for the differences of FA content between TG
`and PL where Student’s t-test was used. Post-hoc analysis was con-
`ducted using Tukey’s test with a significance level of (P < 0.05). AN-
`OVA statistical comparisons were conducted using KYPlot
`(KyensLab, Tokyo, Japan).
`
`3. Results and discussion
`
`3.1. Extraction efficiency
`
`Extraction of krill oil using the two-step procedure resulted in
`similar (P > 0.05) efficiency as the one-step procedure using a
`
`Page 5
`
`krill:solvent ratio of 1:6 (Fig. 2). Extraction efficiency for the one-
`step procedure increased (P < 0.05) as the krill:solvent ratio in-
`creased, with 1:12 and 1:30 ratios having the greatest (P < 0.05)
`efficiencies. There was no (P > 0.05) significant difference in extrac-
`tion efficiency between the one-step procedure using 1:12 and
`1:30 ratio. The krill:solvent ratio is a critical parameter for extrac-
`tion efficiency using the one-step procedure. Likely, if the krill:sol-
`vent ratio were increased in the two-step extraction, the extraction
`efficiency would similarly increase. However, one-step extraction
`is simpler than the two-step procedure; and therefore, one-step
`extraction with acetone:ethanol (1:1, v:v) using 1:12 krill:solvent
`ratio is recommended.
`
`3.2. Lipid classes in extracted krill oil
`
`Major lipid classes of the extracted krill oils were separated
`using TLC (Fig. 3 insert), quantified by densitometry and presented
`as % of lipid class in total krill oil extracted (Fig. 3). The extraction
`procedure did not (P > 0.05) affect TG content in the extracted oils,
`which ranged 1.0–3.2% (Fig. 3). However, extraction procedure had
`an effect (P < 0.05) on PL content, with Folch, Soxhlet, and 1-step
`(1:6 krill:solvent ratio) extractions having the highest (P < 0.05)
`content of PL. The PL content of all extracted oils ranged from
`20.4–32.7% (Fig. 3). A high PL content in krill oil has been described
`previously, with total PL accounting for approximately 40% of krill
`oil (Bottino, 1975). The high PL content makes krill oil unique as
`compared to other dietary lipids. The TG content in fish oils is
`approximately 60% (Tou et al., 2007).
`The major lipid classes in all of the oils extracted from krill were
`the polar non-PL classes (>60%). These classes consist of choles-
`terol, mono- and di-glycerides, and the red pigment primarily asta-
`xanthin. The association of the red pigment with the polar non-PL
`classes made quantifying each class individually with TLC difficult.
`Thus, these classes were combined and accounted for as the polar
`non-PL class. The one-step extraction with krill:solvent ratios
`greater than 1:6 and the acetone fraction from the two-step extrac-
`tion had the highest (P < 0.05) polar non-PL class content. The dif-
`ferences in the polar non-PL class content should be of commercial
`interest because an extraction procedure that yields the oil with
`lower cholesterol content, but higher antioxidant capacity would
`provide the healthiest oil. Therefore, cholesterol content and total
`antioxidant capacity were determined (see Section 3.4 and 3.5,
`respectively).
`
`NEPN Ex. 2043
`Aker v. Neptune
`IPR2014-00003
`
`

`
`J.C. Gigliotti et al. / Food Chemistry 125 (2011) 1028–1036
`
`1033
`
`n6/n3 and saturated FA/unsaturated FA were lower (P < 0.05) in
`PL than TG (Fig. 4A). Oil with a lower n-6/n-3 and saturated
`FA/unsaturated FA is considered healthier. EPA and DHA were pre-
`dominant (P < 0.05) in PL, contributing to much higher (P < 0.05)
`content of total n-3 FA and unsaturated FA in PL than in TG
`(Fig. 4B). EPA and DHA are the bioactive n-3 PUFA associated with
`health benefits such as reduced risk of CVD (cardiovascular
`disease). The TG had a higher (P < 0.05) content of myristic,
`palmitoleic, and oleic acid than PL (Fig. 4C).
`The PL and TG require different digestive enzymes and
`therefore, there may be differences in bioavailability and tissue
`accretion of n-3 PUFA. In turn, this may result in different physio-
`logical and health effects ( Amate, Gil, & Ramirez, 2001, 2002; Mat-
`thews et al., 2002). Therefore, having EPA and DHA esterified as PL
`in krill oil may have significant implications for human health. The
`preferential esterification of EPA and DHA into PL is intriguing and
`has been noted previously (Saether et al., 1986). However, avail-
`able data comparing the benefits of consuming x-3 PUFA as krill
`oil compared to other sources is scarce. Therefore, further studies
`are needed to compare and understand the nutritional value and
`health effects of krill oil versus other sources of n-3 PUFA.
`
`3.4. Cholesterol content in extracted krill oil
`
`The one-step extraction procedures yielding oil with the highest
`(P < 0.05) content of polar non-PL (one-step extractions in Fig. 3)
`did not
`(P < 0.05) result
`in the highest cholesterol content
`(Fig. 5A). The one-step procedures with krill:solvent ratios P1:9
`generally had the lowest cholesterol content. Cholesterol content
`of the two-step, Folch, and Soxhlet procedure was similar to the
`content of the polar non-PL (Fig. 3), suggesting that the cholesterol
`content contributed more to the content of polar non-PL in these
`procedures than the one-step procedures.
`In general, cholesterol content of krill is lower than shrimp, and
`slightly higher than fish (Tou et al., 2007). Krill oil resulted in 18%
`reduction in total serum cholesterol compared to 6% with fish oil in
`hyperlipidemic patients (Bunea et al., 2004). The hypolipidemic ef-
`fects of seafood oils are not fully understood, but are significantly
`influenced by n-3 PUFA. Since krill oil is rich in n-3 PUFA bound
`in PL and is low in cholesterol, the hypolipidemic effects of krill
`oil could be additive.
`
`3.5. Antioxidant capacity of extracted krill oil
`
`The red colour of krill oil is due to the carotenoid astaxanthin, a
`potent antioxidant. Frozen krill contains 3–4 mg of carotenoids/
`100 g and astaxanthin is >80% of the total carotenoids (Yamaguchi
`et al., 1983). By measuring the total antioxidant capacity of the krill
`oil extracted in the present study, the antioxidative effect of asta-
`xanthin was accounted for. In vitro studies have shown that asta-
`xanthin decreases membrane oxidative injury to a greater degree
`than a-tocopherol, an antioxidant commonly used in food products
`(Kurashige, Okimasu, Inoue, & Utsumi, 1990).
`Oils extracted with the one-step procedure using 1:9 and 1:30
`krill:solvent ratios as well as the ethanol extract of the two-step
`procedure had the greatest
`(P < 0.05) antioxidant capacity
`(Fig. 5B). The antioxidant capacities of oils extract