JFS C: Food Chemistry
C.K. BOWER, K.A. HIETALA, A.C.M. OLIVEIRA, AND T.H. WU
ABSTRACT: Smoking of meats and fish is one of the earliest preservation technologies developed by humans. In
this study, the smoking process was evaluated as a method for reducing oxidation of pink salmon (Oncorhynchus
gorbuscha) oils and also maintaining the quality of oil in aged fish prior to oil extraction. Salmon heads that were
subjected to high temperatures (95◦C) during smoking unexpectedly produced oils with fewer products of oxida-
tion than their unprocessed counterparts, as measured by peroxide value (PV), thiobarbituric acid reactive sub-
stances (TBARS), and fatty acids (FA). Higher temperatures and longer smoking times resulted in correspondingly
lower quantities of oxidative products in the oils. Fatty acid methyl ester (FAME) analysis of smoke-processed oils
confirmed that polyunsaturated fatty acids (PUFA) were not being destroyed. Smoke-processing also imparted an-
tioxidant potential to the extracted oils. Even when antioxidants, such as ethoxyquin or butylated hydroxytoluene,
studies supported the antioxidant results, with smoke-processed oils displaying higher levels of α-tocopherol than
raw oils. Results suggest that smoking salmon prior to oil extraction can protect valuable PUFA-rich oils from oxi-
enhance levels of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) in foods.
Keywords: Alaska salmon, fish oil, fishery by-products, lipid oxidation
Bechtel 2003). Wild-caught Alaskan salmon possess high concen-
trations of long-chain n-3 polyunsaturated fatty acids (PUFA) that
may not be present in farmed salmon raised on diets containing
tropicaloilssuch aspalm(Bellandothers 2002).Unsaturated lipids
found in fish oils, have been associated with beneficial health ef-
fects (Hasler 2000), and consequently are appearing more often in
foods. However, lipids with increased unsaturation are sometimes
more susceptible to oxidative degradation (Nawar 1996). Vulner-
ability to lipid oxidation is known to vary among fish species as
well as among different fish components. For example, crude oil
extracted solely from herring heads had a lower stability than oil
from mixed herring by-products (skins, frames, and viscera), which
the authors attributed in part to the lower concentration in herring
heads of α-tocopherols, naturally occurring antioxidants found in
fish tissues (Aidos and others 2002).
Incorporating PUFAs into foods and feeds can decrease the
oxidative stability of the final product (Augustin and Sanguansri
2003), which can result in loss of desirable flavors, colors, aromas,
and nutritive properties, including the destruction of PUFAs and
loss of fat-soluble vitamins. Oxidation of oils can also produce toxic
long-chain PUFAs, resulting in eicosapentaenoic acid (EPA) and
docosahexaenoic acid (DHA), becoming more stable than shorter
laska is responsible for over half the total fish landings in
the United States, about 9% of which is salmon (Crapo and
MS 20080846 Submitted 10/27/2008, Accepted 1/5/2009. Authors Bower,
Hietala, and Wu are with USDA Agricultural Research Service, PO
Box 757200, Fairbanks, AK 99775–7200, U.S.A. Author Oliveira is with
Fishery Industrial Technology Center, Univ. of Alaska Fairbanks, 118 Tri-
dent Way, Kodiak, AK 99615, U.S.A. Direct inquiries to author Bower
chain fatty acids (FA) such as linoleic and linolenic (Miyashita and
others 1993). This unexpected stability appears to be associated
with the conformation of EPA and DHA in aqueous environments
(Kato and others 1992). Oil-in-water emulsions were also found to
provide oxidative stability for fish oils rich in long-chain PUFAs
(Frankel and others 2002). A practical application of this concept
was demonstrated in salad dressing, where emulsified oils were
more effectively protected from oxidation than when straight fish
oil was added (Let and others 2007).
Smoking fish has historically been used as a preservation
method. In addition to imparting flavor and aroma attributes,
smoke can protect foods from oxidation (Pearson and Gillett 1996;
Schwanke and others 1996). Chemicals from wood smoke com-
monly include phenols, organic acids, alcohols, carbonyls, hydro-
carbons, and nitrogen compounds such as nitrous oxide (Pearson
and Gillett 1996). Among these components, phenols have demon-
strated antioxidant capabilities (Pearson and Gillett 1996). Nitrites
are also known to inhibit lipid peroxidation by participating in the
radical-quenching process (Nicolescu and others 2004). The perox-
idation process can also be inhibited by formate anion, which is a
radical scavenger found in smoke flavorings and smoke-processed
meats (Schwanke and others 1996; Coronado and others 2002).
As PUFA-enriched products become increasingly associated
with enhanced nutritional value, new methods for controlling ox-
idative damage will emerge. When fish oil was added directly to pig
feeds, higher levels of lipid oxidation occurred and “fishy” odors
in the pork were detectable (Trout and others 1998). Bacon from
pigsthatreceived adietarysupplement offishmealexperienced in-
creased levels of oxidation (as measured by thiobarbituric acid re-
lipid oxidation was reduced when the pigs also received 200 mg α-
tocopherol per kilogram of feed, and the bacon was processed with
a combination of liquid and wood smoke (Coronado and others
No claim to original US government works
Journal compilation C ?2009 Institute of Food TechnologistsR ?
Further reproduction without permission is prohibited
JOURNAL OF FOOD SCIENCE—Vol. 74, Nr. 3, 2009
Stabilizing oils from smoked salmon (Oncorhynchus gorbuscha). ..
2002). These findings suggest that smoke-processing fatty fish such
rich oils. The objective of this study was to examine the condi-
tions of smoke-processing required to reduce oxidation of salmon
oil during storage.
Materials and Methods
Pink salmon (Oncorhynchus gorbuscha) heads (n = 108, ranging
from 113 to 608 g) were collected from a commercial processor lo-
cated in the city of Kodiak, (Kodiak Island, Alaska, U.S.A.) in Au-
gust 2007, and stored at −20◦C until shipped to Fairbanks (Alaska,
U.S.A.) for processing. Samples were placed in a Bradley Smoker
(Bradley Technologies Canada Inc., Richmond British Columbia,
Canada), and smoked using hickory bisquettes. In one study,
salmon heads were smoked at different temperatures (40, 60, 75,
or 95◦C) for 5 h (n = 9 per treatment). In a 2nd study, salmon
heads (n = 9 per treatment) were smoked at a constant tempera-
ture (95◦C) for different times (1, 2, 3, 4, or 5 h). Smoked salmon
heads were processed using a Tor Rey F12-FS electric meat grinder
(Tor Rey USA Inc., Houston, Tex., U.S.A.) with a 0.32-cm (1/8 inch)
centrifuge tubes for oil extraction (16500 × g; 20 min; 4◦C) using a
Beckman J2-HS centrifuge (Fullerton, Calif., U.S.A.) equipped with
a JA-20 rotor. Oil was removed from each tube, pooled, and placed
into amber glass vials, flushed with nitrogen and stored frozen
(−80◦C). Cooked oils were obtained by homogenizing 10 raw
salmon heads and heating the ground tissue in a 95◦C water bath
for 55 min. Oil was extracted and stored as described previously.
Shelf life of smoked salmon heads and extracted oils
Oils (1 g) extracted from smoked and raw salmon heads were
stored in small (1.5 mL) loosely capped vials at 2 temperatures (4
or 35◦C) for up to 30 d to assess storage stability. Three vials were
removed daily from each treatment group and tested for indicators
of oil quality (FA, PV, TBARS). Intact salmon heads were also stored
up to 6 d at 2 different temperatures (4 or 35◦C) prior to oil ex-
traction to compare storage stability of oils within raw and smoke-
processed salmon heads. Alpha-tocopherol (vitamin E) levels were
heads stored at 4 and 35◦C. Vitamin E values were also determined
for salmon heads smoke-processed (5 h) at different temperatures
(40, 60, 75, or 95◦C).
Three samples from each treatment were analyzed in duplicate
for each of the following analyses. Moisture was determined gravi-
metrically by drying samples for 24 h and measuring water loss
analyzer (Mt. Laurel, N.J., U.S.A.), which multiplies nitrogen values
of bovine serum albumin) to calculate protein values. Lipids were
determined by processing dried samples on a Soxtec Model 2043
(method 991.36, AOAC 1990) using a methylene chloride extraction
to dryness to remove solvent, and then weighed. Ash content was
determined by placing samples into a muffle furnace at 550◦C for
24 h and then weighing the remaining material (method 938.08,
FA were analyzed in triplicate according to Bern´ ardez and others
(2005). In brief, approximately 50 mg of oil were dissolved in 1.5 mL
of cyclohexane and 0.5 mL of 5% (w/v) cupric acetate (adjusted to
pH 6 with pyridine) was then added and vortexed for 30 s. Samples
were centrifuged for 10 min at 2000 × g and the absorbance of the
upper layer (cylcohexane) read at 710 nm. Results were measured
in milligrams per gram oil using oleic acid as the standard.
Lipid peroxide values (PV) were determined in triplicate using
the Intl. Dairy Federation iron-based method (IDF 1991), where
oil (10 mg) was first dissolved in 9.8 mL of a 7:3 chloroform–
methanol mixture, then vortexed prior to the addition of 30% am-
monium thiocyanate solution (50 μL). After vortexing for 2 to 4 s,
50 μL of iron(II) chloride solution were added and vortexed. Sam-
ples were incubated at room temperature for no more than 5 min
and absorbances were immediately read at 500 nm. Results were
expressed in milliequivalents peroxide per kilogram oil using the
equation: PV = (μg Fe3+)/(55.84)(m)/(2), where m is the mass of oil
from a standard curve.
Thiobarbituric acid reactive substances
(1978) with slight modifications by dissolving 50 mg of oil in
3.5 mL cyclohexane and 4.5 mL of 7.5% trichloroacetic acid (TCA)
terfering solubles by acid-precipitating the lipoprotein fractions.
Samples were mixed for 5 min (to allow the secondary lipid oxi-
dation products from the oil to dissolve into the polar layer), and
then centrifuged for 15 min at 1555 × g. The aqueous TCA–TBA
phase was separated from the nonpolar solvent, and incubated at
to form a chromogen, which was detected at 532 nm using a Spec-
troMax Plus microplate spectrophotometer (Molecular Devices,
Union City, Calif., U.S.A.). The intensity of color correlated with the
quantity of TBARS (principally MDA) in the oxidized oil, and was
reported as milligrams per kilogram oil using malonaldehyde-bis
as the standard. Although the smoked-oil samples in this study ac-
quired progressively darker colors with increased smoking times
and temperatures, full-spectrum absorbance scans revealed no in-
terfering compounds at 532 nm.
Fatty acid methyl esters preparation and analysis
Crude salmon oils from each treatment were stored under nitro-
gen at –80◦C until derivatization. Methyl esters were prepared ac-
cording to the procedure of Maxwell and Marmer (1983) using 23:0
as internal standard. Fatty acid methyl esters were separated and
quantified as described by Bechtel and Oliveira (2006). Briefly, an
Agilent Technologies model 6850 gas chromatograph (GC) (Wilm-
ington, Del., U.S.A.) equipped with a flame ionization detector
(FID) and a DB-23 (60 m × 0.25 mm id., 0.25 μm film) capillary
column (Agilent Technologies) was used for separation and quan-
tification of fatty acid methyl esters. Hydrogen was used as carrier
gas at a constant flow of 1 mL/min. Detector and injector were held
at a constant temperature of 275◦C, and the split ratio was 25:1.
The oven programming was 140 to 200◦C at a rate of 2◦C/min,
200 to 220◦C at a rate of 0.5◦C/min, and 220 to 240◦C at a rate
of 10◦C/min for a total run time of about 62 min. An autosam-
pler performed the injections of standards and samples at a con-
stant volume of 1 μL. Data were collected and analyzed using the
GC ChemStation program (Rev.A.08.03 ; Agilent Technologies,
Vol. 74, Nr. 3, 2009—JOURNAL OF FOOD SCIENCE
Stabilizing oils from smoked salmon (Oncorhynchus gorbuscha). ..
thereby offering processors the additional time needed to extract
valuable marine oils.
This research is being performed as part of a larger USDA Agri-
cultural Research Service project designed to convert underutilized
Alaska fish by-products into value-added ingredients and products
(CRIS nr 5341 31410 003 00D).
Aidos I, Masbernat-Martinez S, Luten JB, Boom RM, van der Padt A. 2002. Composi-
tion and stability of herring oil recovered from sorted by-products as compared to
oil from mixed by-products. J Agric Food Chem 50:2818–24.
Aminullah-Bhuiyan AKM, Ratnayake WMN, Ackman RG. 1993. Nutritional compo-
sition of raw and smoked Atlantic mackerel (Scomber scombrus): oil and water-
soluble vitamins. J Food Comp Anal 6:172–84.
Analytik Jena AG Konrad-Zuse-Strasse 1, 07745 Jena Germany.
[AOAC] Association of Official Analytical Chemists. 1990. Official methods of analysis
of the Association of Official Analytical Chemists. 15th ed, Vol. 2. Arlington, Va.:
Augustin MA, Sanguansri L. 2003. Polyunsaturated fatty acids: delivery, innovation
and incorporation into foods. Food Aust 55:294–6.
Bechtel PJ, Oliveira ACM. 2006. Chemical characterization of liver lipids and proteins
from cold water fish species. J Food Sci 71(6):S480–5.
Bell JG, Henderson RJ, Tocher DR, McGhee F, Dick JR, Porter A, Smullen RP, Sar-
gent JR. 2002. Substituting fish oil with crude palm oil in the diet of Atlantic
Salmon (Salmo salar) affects muscle fatty acid composition and hepatic fatty acid
metabolism. J Nutr 132:222–30.
Bern´ ardezM, Pastoriza L, Sampedro G, Herrera JJR, Cabo ML. 2005. Modified method
for the analysis of free fatty acids in fish. J Agric Food Chem 53:1903–6.
Coronado SA, Trout GR, Dunshea FR, Shah NP. 2002. Effect of dietary vitamin E, fish-
meal and wood and liquid smoke on the oxidative stability of bacon during 16
weeks’ frozen storage. Meat Sci 62:51–60.
Crapo C, Bechtel P. 2003. Utilization of Alaska’s seafood processing by-products. In:
Bechtel PJ, editor. Advances in seafood by-products, 2002 conference proceedings.
Fairbanks, Alaska: Alaska Sea Grant College Program, University of Alaska Fair-
banks, p 105–19.
algae oils containing long-chain polyunsaturated fatty acids in bulk and in oil-in-
water emulsions. J Agric Food Chem 50:2094–9.
Hasler CM. 2000. The changing face of functional foods. J Am Coll Nutr 19(5):499S–
Hultin HO. 1994. Oxidation of lipids in seafoods. In: Shahidi F, Botta JR, edi-
tors. Seafoods: chemistry, processing technology and quality. Glasgow, Scotland:
Blackie, Academic & Professional. p 49–74.
[IDF] International Dairy Federation. 1991. IDF standard 74A. Square Vergote 41,
Kato T, Hirukawa T, Namiki K. 1992. Selective terminal olefin oxidation ofn-3 polyun-
saturated fatty acids. Tetrahedron Lett 33:1475–8.
Kobayashi H, Yoshida M, Miyashita K. 2003. Comparative study of the product com-
ponents of lipid oxidation in aqueous and organic systems. Chem Phys Lipids
Lands WEM. 2005. Fish, omega-3 and human health. 2nd ed. Urbana, Ill.: The Ameri-
can Oil Chemists’ Society (AOCS) Press.
Leaf L. 2008. Beyond cholesterol: prevention and treatment of coronary heart disease
with n-3 fatty acids. Historical overview of n-3 fatty acids and coronary heart dis-
ease. Am J Clin Nutr 87(6):1978S–80S.
Let MB, Jacobsen C, Meyer AS. 2007. Ascorbyl palmitate, γ-tocopherol, and EDTA af-
fectlipid oxidation in fishoil enrichedsalad dressingdifferently.J Agric Food Chem
Maxwell RJ, Marmer WN. 1983. Systematic protocol for the accumulation of fatty acid
data from multiple tissue samples: tissue handling, lipid extraction, class separa-
tion, and gas chromatographic analysis. Lipids 18:453–8.
Miyashita K, Nara E, Ota T. 1993. Oxidative stability of polyunsaturated fatty acids in
an aqueous solution. Biosci Biotechnol Biochem 57:1638–40.
Nawar WW. 1996. Lipids. In: Fennema OR, editor. Food chemistry. New York: Marcel
Dekker Inc. p 225–319.
inhibition of lipid peroxidation and radical reactions. Chem Res Toxicol 17:185–96.
Oliveira ACM, Bechtel PJ. 2005. Lipid composition of Alaska pink salmon (On-
corhynchus gorbuscha) and Alaska walleye pollock (Theragra chalcogramma) by-
products. J Aquat Food Prod Tech 14(1) 73–91.
Pearson AM, Gillett TA. 1996. Processed meats. 3rd ed. New York: Chapman and Hall.
Schwanke S, Ikins WG, Kastner C, Brewer MS. 1996. Effect of liquid smoke on lipid
oxidation in a beef model system and restructured roasts. J Food Lipids 3:99–113.
Sk˚ ara T, Sivertsvik M, Birkeland S. 2004. Production of salmon oil from filleting
by-products: effects of storage conditions on lipid oxidation and content of ω-3
polyunsaturated fatty acids. J Food Sci 69(8):E417–21.
Siu GM, Draper HH. 1978. A survey of the malonaldehyde content of retail meats and
fish. J Food Sci 43:1147–9.
Trout GR, Hanrahan B, Dinh J, Chai J. 1998. Incidence of fish-oil fatty acids in Aus-
tralian ham and effects on rancidity development during frozen storage. In: Pro-
ceedings of 44nd International Congress of Meat Science and Technology. August
30–September 4 1998, Barcelona, Spain. p 660–1.
Vol. 74, Nr. 3, 2009—JOURNAL OF FOOD SCIENCE