11_127.pdf

Food Sci. Technol. Res., ++ (+), +,1ῌ+--, ,**/

Milkfish (Chanos chanos Forskaal) Consumption in the Philippines and the Docosahexaenoic Acid Level of the Cooked Fish Cecile Leah P. Tiangson-BAYAGA῍ and Genevieve F. DEVEZA College of Home Economics, University of the Philippines, Diliman, Quezon City Received June ,+, ,**. ; Accepted January ,2, ,**/ Milkfish consumption and the e#ects on the proximate composition and docosahexaenoic acid (DHA) level of three cooking methods (Paksiw, Sinigang and fried) used to prepare milkfish were determined. Total moisture content, crude fat, crude protein and the total ash of the dishes were analyzed. Fatty acid methyl esters (FAMEs) were analyzed through gas chromatography. Results showed that milkfish is eaten once to twice a week by Filipino households. Proximate analysis revealed that Paksiw had the highest moisture content, and Sinigang had the highest crude fat content. Sinigang had a significantly lower crude protein content compared to raw fish, fried fish had the highest crude protein on a wet basis and Paksiw on a dry basis. The DHA level for Paksiw was not significantly di#erent from raw milkfish, but DHA level for Sinigang and fried milkfish were significantly di#erent from the raw sample. Di#erent cooking methods can thus cause changes in the proximate composition and DHA content of milkfish. Keywords : Docosahexaenoic Acid (DHA), Milkfish, proximate composition, fatty acid composition, milkfish consumption

Introduction The nutritional status of the breast-fed infant is highly dependent on the mother. Therefore, if the mother is malnourished, then it is likely that the infant will have a poor nutritional status. Infants acquire the nourishment needed from mother’s milk and it has been established that the breast milk composition is influenced by maternal diet. One of the dietary constituents that has been gaining recognition for good infant nutrition is long chain (,*- and ,,- carbon) polyunsaturated fatty acids (LCPUFAs). Recently, interest has grown in fish and fish products as sources of LCPUFAs, particularly docosahexaenoic acid (DHA), because of its role in brain and retinal development (Makrides et al., +33/ ; Innis et al., +33. ; Carlson et al., +33-). DHA have also been recognized in helping to lower blood pressure, improving antioxidant levels and preventing lipid peroxidation, decreasing the rate platelet aggregation, and reducing plasma lipid concentrations (Frenoux et al., ,**+). A recent study showed that the level of DHA in the breast milk of lactating Filipino women (nῌ+**) was significantly associated with frequency of milkfish (bangus) consumption (Tiangson et al., ,**-). Furthermore, Tiangson (,**-) concluded that the frequency of milkfish consumption among families was not significantly associated with annual income, implying that the consumption of milkfish is not limited to the low-, middle-, or high-income families. Milkfish is consumed in the Philippines, Indonesia, Taiwan, Malaysia, Thailand, Southern Vietnam, Japan, Myanmar, Sri Lanka, Southern India, Iraq, Hawaii and E-mail : [email protected]

Mexico (Bagarinao, +333). Milkfish is considered to be a “fatty” fish because it has more than /ῌ fat in relation to its body weight. However, the fatty acids present in fish are primarily unsaturated, whereas saturated fatty acids are predominant in other meat. The precursor of essential fatty acids like DHA and eicosapentaenoic acid (EPA), a-linolenic acid, is prevalent in fishes (Kinsella, +320). DHA, being a polyunsaturated fatty acid (C,, : 0w--), is highly susceptible to oxidation. Cooking at high temperatures, changes in pH and other factors may cause changes in DHA structure, resulting in decreased levels (Nawar, +33*). It is therefore essential to determine which cooking methods result in minimal DHA loss. The main objective of this study is to determine the e#ects of di#erent conventional cooking methods on DHA levels in milkfish.

Methods Survey An interview was conducted in four areas of the University of the Philippines’ Diliman Community. A random sample of ,,* households was interviewed. The survey questionnaire was designed to obtain demographic information, and assess milkfish consumption and the common cooking method for milkfish. Detailed recipes and cooking methodologies of the respondents were also obtained. The survey was encoded and analyzed using the software Statistical Package for Social Sciences (SPSS). The frequency of the types of fish commonly eaten, common cooking methods for milkfish and the vegetables used for the dishes were noted. Only the top ten most commonly eaten fish, most common ingredients, and most common cooking procedures were considered in the analysis. The

128

top three cooking methods were standardized. Standardization of Recipes The three most common cooking methods for milkfish were Paksiw, milkfish boiled in water and vinegar ; Sinigang milkfish boiled in water with a tamarind soup base ; and fried. These were standardized based on the recipes and methodology described by the respondents of the survey. The amount and kind of each ingredient, cooking time and temperature used per recipe was set such that the resulting product was consistent in flavor, texture and doneness. For Paksiw, the milkfish was first thawed at room temperature for + hour and -* minutes. Then it was sliced into three and iodized salt was added. The thawed milkfish and the other ingredients (i.e., eggplant, onion, ginger, garlic, jalapeno, iodized salt, black pepper, vinegar and water) were placed in a covered pot. The ingredients were allowed to boil for . minutes or until the mixture reached +*+῎. The dish was then allowed to simmer for +* minutes. For Sinigang, the milkfish was thawed for + hour and -* minutes. The scales were removed and the fish sliced into three. In a pot, tomatoes, onions, radish, and salt were boiled for / minutes at 33῎. Then, the fish were added and allowed to simmer for 0 minutes. Swamp cabbage and jalapeno were added and the dish was allowed to simmer again for . minutes, after which the swamp cabbage was removed from the pot and set aside. Tamarind mix was added to the dish and the swamp cabbage was placed back in the pot immediately prior to serving. For the frying treatment, milkfish was thawed for + hour and then sliced into three. Iodized salt was then rubbed all over the fish that was then allowed to stand at room temperature for + hour. In a frying pan, the cooking oil was heated to ,**ῌ,+*῎ before immersing each fish slice. Each slice was cooked for 3 minutes. The fish samples used in the three dishes were acquired from the Lingayen Public Market, Lingayen, Pangasinan. The size of the milkfish used ranged from ,3*ῌ-+* mm and weighed -**ῌ--* grams. The fishes were cultivated in freshwater fishponds and fed using filamentous green algae. Sensory evaluation was conducted to determine acceptability of the dishes. A nine-point hedonic scale was used to rate acceptability. A rating of 1 or above, similar to moderate, meant that the characteristic of the product being examined was acceptable. The fishes were harvested on March ,**, for the sensory evaluation. Sample Preparation Samples were obtained from each of the standardized dishes for chemical analysis. Samples were consisted of the cooked fish meat and other ingredients like the vegetables and broth (for the Paksiw and Sinigang dishes). The cooked fish meat was flaked prior to homogenization with the other ingredients. Aliquots of the resulting homogenates were then subjected to proximate and fatty acid analysis. Proximate Analysis The standardized formulation of Paksiw, Sinigang and fried milkfish together with a raw sample were subjected to proximate analysis. Total moisture content was analyzed using Vacuum Oven

C. BAYAGA & G. DEVEZA

Method (AOAC Method 30*.+2), crude fat was determined using Soxhlet Extraction Method (AOAC Method 3,*.-3 C). The Kjeldahl Method (AOAC Method 3//.*.) was employed to determine crude protein and Dry Ashing (AOAC Method 3,-.*-) was performed to check total ash of the samples. Analyses were performed in triplicate for statistical validity. Identification and exclusion of outliers was determined using a Q-test. Results are expressed using the mean and standard error of the mean (SEM). Analysis of Variance (ANOVA) was used to determine significant di#erence among treatments at a῍*.*/. Duncan’s Multiple Range Test was used to detect which treatment is significantly di#erent from the others. Fatty Acid Analysis The raw and cooked milkfish samples were brought to the Standards and Testing Division of Industrial Technology Development Institute (STD-ITDI) in the Department of Science and Technology (DOST) Complex, Taguig, Metro Manila for fatty acid profiling. The amount of DHA in the samples was determined based on the results of the fatty acid profile. Samples of fresh and used oil in frying were also submitted for analysis. Fat was extracted from raw, Paksiw, Sinigang and fried milkfish using the Soxhlet Method. Fatty acids were liberated and esterified using the AOAC (+33*) Method 303.-- (Boron Trifluoride Method). Fatty acid methyl esters (FAMEs) were analyzed by gas chromatography using a DB-WAX column (-* mῌ*.,/ mm i.d., with *.,/ mm film thickness) in a Shimadzu GC +. A equipped with flame ionization detector. The injection temperature was set at ,/*῎ and initial column temperature was set at ./῎ for , minutes and programmed to rise at +*῎ per minute until the final ,/*῎ was achieved. Detection temperature was set at -**῎. Helium was used as the carrier gas with an initial injection volume of *., ml. FAMEs were identified on the basis of their retention time versus authentic lipid standard obtained from Nuchek Prep Inc (Elysian, MN). The significant di#erences in the DHA contents of the four samples were analyzed using Analysis of Variance (ANOVA) at /ῌ level of significance. Duncan’s Multiple Range Test was also performed to detect which samples are significantly di#erent. Student’s t-test was used to determine whether there were any significant di#erences in the fatty acid profiles of fresh and used cooking oils.

Results and Discussion Survey The respondents (n῍,,*) were classified according to their annual income in order to show whether there are certain di#erences between households falling into di#erent income groups and their attitude towards fish consumption. When the respondents were asked to list the kind of fish they usually eat, milkfish ranked first, followed by tilapia and then by round scad. Figure + shows the top ten most commonly consumed fish as determined by the survey. The respondents were also asked to indicate the three most convenient ways of cooking milkfish. The top

Consumption and DHA Levels in Milkfish (Paper #*.*0,)

Fig. +.

Fig. ,.

Top ten most commonly consumed fish in Filipino households (nῌ,,*)ῌ

Top ten common cooking methods for milkfish in Filipino households (nῌ,,*)ῌ

129

130

C. BAYAGA & G. DEVEZA

three cooking methods were Sinigang, Paksiw and frying. Figure , shows the top ten responses. Table + shows the relationship between frequency at which each household eats milkfish and classification of the household with regards to annual family income. Income was classified into - groups. Group A earns ῏ 2*,*** PHP/year ; group B earns 2*,***ῌ+,*,*** PHP/year ; and, group C earnsῐ +,*,*** PHP/year. Of the ,,* households, //./ῌ belonged to group A ; -*.3ῌ belonged to group B ; and, +-.0ῌ had annual incomes belonging to group C. It can be inferred from the table that the frequencies at which families eat milkfish is generally similar in the total population. The category ῌ+ῌ, times a week” has the highest frequency (n῎+,2), followed by “-ῌ. times a week” (n῎..) and “,ῌ- times a month” (n῎-,). Proximate Analysis The proximate compositions of raw, Paksiw, Sinigang and fried milkfish are summarized in Table , (wet basis) and Table - (dry basis). Results are expressed as mean῍standard error of the mean (SEM). Mai, et al., (+312) reported that when data are expressed as dry weight, the changes indicate real weight changes more accurately. This has the advantage of relating more accurately to the actual gain or loss of lipids and thus aids comparison between sample treatments. The assumption is made that changes in protein, carbohydrates and other components are minor. Since food is

Table +.

consumed in a wet form, its importance in nutrient intake calculations should be acknowledged (Mai, et al.+312). Moisture The moisture content of the milkfish sample was lower than that reported in the Philippine Food Composition Table (FCT) (FNRI, +331). This may be because the samples were harvested during the dry season of March, after the eggs were fertilized in January during the wet or rainy season in the Philippines. In the study of Mustafa and Medeiros (+32/), the moisture content of catfish fillets was lower in samples harvested in August (summer season) than those harvested in December and April. This shows that harvesting season a#ects the moisture content of fishes. Paksiw had the highest moisture content among the four treatments (00.0ῌ), but this was not significantly di#erent from that observed for raw milkfish at /ῌ level of significance. The moisture content of raw milkfish was also not significantly related to Sinigang (0,.-ῌ). Since Paksiw and Sinigang involve boiling in water, no major moisture loss or gain occurred. The moisture content of Paksiw was significantly di#erent from that of Sinigang with the lower moisture content possibly attributed to the proportional increase in the crude fat content. The moisture content of fried milkfish (//.,ῌ) is significantly lower than the other treatments and is comparable to the FCT value of /0.2ῌ. This is because frying

Frequency of milkfish consumption per income group of the Filipino households surveyed (nῌ,,*).

Table ,. Proximate composition on a wet basis for raw, Paksiw, Sinigang and fried milkfish*ῌ

Table -. Proximate composition on a dry basis for raw, Paksiw, Sinigang and fried milkfish*ῌ

131

Consumption and DHA Levels in Milkfish (Paper #*.*0,)

has a dehydrating e#ect on food. Crude Fat There were no significant di#erences in crude fat content between raw (+*.3ῌ) and Paksiw (3.+,ῌ), and between Sinigang (+0.3ῌ) and fried (+0.1ῌ) milkfish at /ῌ levels of confidence. However, the crude fat of raw and Paksiw was significantly di#erent from Sinigang and fried milkfish. On a dry basis, the crude fat content of raw, Paksiw, fried and Sinigang milkfish are not significantly di#erent. However, the crude fat content of Paksiw and Sinigang were significantly di#erent. The crude fat content of Sinigang is also not significantly di#erent from fried milkfish, but fried milkfish was significantly di#erent from Paksiw and raw milkfish. The crude fat in the fried sample (+0.1ῌ) was higher than the value given in the Philippine Food Composition Table (+*..ῌ). The increased crude fat percentage of fried milkfish is due to oil absorption associated with

frying. Since milkfish can be categorized as a fatty fish, it has a greater tendency to absorb oil during cooking compared to non-fatty fish (Mai, et al., +312). The high crude fat content of Sinigang may be attributed to the proportional increase due to lower moisture content compared to Paksiw and raw milkfish. Sinigang also has lower moisture and crude protein content that contributes to a proportional increase in the crude fat content. Crude Protein Sinigang has significantly lower levels of crude protein (+3.*ῌ) compared to raw (,,.0ῌ), Paksiw (,,.0ῌ) and fried milkfish (,/.1ῌ). Sinigang is cooked in more water that may cause leaching out of water-soluble proteins during cooking. The collagen from milkfish is hydrolyzed and this causes gel formation. This can be observed in Sinigang after being kept in cold storage. The protein content of raw, Paksiw and fried milkfish were not significantly di#erent from each other on a wet

Table .. Fatty acid profile of raw, Paksiw, Sigang and fried milkfish as percent weight of total fatty acids.

132

basis. On dry weight basis, the crude protein content of Paksiw (01.0ῌ) and raw milkfish (0-.3ῌ) were not significantly di#erent from each other. The protein content of raw and fried milkfish (/1./ῌ), as well as Sinigang (/*..ῌ) and fried milkfish were also not significant from each other. However, Paksiw and fried milkfish were significantly di#erent from each other and the protein content of Sinigang was significantly di#erent from Paksiw and raw milkfish. This was also due to the proportional increase in crude fat content. Total Ash There were no significant di#erences in the total ash content among the raw, Paksiw, Sinigang and fried milkfish treatments at /ῌ confidence levels of confidence for wet and dry bases. The slight increase in the ash content of fried milkfish can be attributed to the addition of salt directly to the fish prior to frying. The slight decrease in Paksiw and Sinigang samples may be due to the leaching out of minerals during cooking (IFT, +31.). Both cooking methods involve boiling in water. DHA Content of Milkfish Fatty acid profiles of raw, Paksiw, Sinigang and fried milkfish are shown in Table . and are expressed as ῌw/w of total fatty acids. The significant gains or losses of each fatty acid using the three cooking methods were compared to the profile of the raw milkfish sample. Considering the focus of the study, the DHA level of raw milkfish (+.0-ῌ) is lower than that reported by Bautista, del Valle and Orejana (+33+) for freshwater reared milkfish, which was ,.+3ῌ. The DHA level of Paksiw (+.,1ῌ) was not significantly di#erent to the raw milkfish ; those of the Sinigang (*.1-ῌ) and fried milkfish (*./*ῌ) are significantly di#erent to the raw sample. Fish meat absorbs large quantities of oil during cooking. Although the lipids in fish meat are oxidized, the influence of oxidation is limited. Thus, the fatty acid composition of fried fish meat is strongly influenced by the absorption of cooking oil, and the decrease in the DHA content of the fried milkfish is due to the absorption of fried oil. The decrease in DHA, especially in the fried milkfish sample is due to cooking at very high temperatures (+3*ῌ,+*ῌ) that may lead to oxidation. Exposure to air and metal also contributed to oxidation. The significant loss of DHA in Sinigang compared to the raw sample may be attributed to some air exposure during occasional stirring. The flavonoids present in black pepper, which was added to Paksiw, may have served as an antioxidant (Skibsted, Mikkelsen and Bertelsen, +332). However, the DHA levels of the three cooking methods are not significantly di#erent from each other. Other Fatty Acids In Milkfish Bautista, et al. (+33+), studied the fatty acid profile of milkfish grown in brackish water and freshwater. The study showed that the most abundant fatty acid in milkfish grown in freshwater was C+2 : * (+0./.ῌ), followed by C+0 : + (+/.2*ῌ), C+2 : + (+,.30ῌ), C+0 : * (2.2/ῌ) and C+2 : - (1.21ῌ). The DHA level was ,.,ῌ. In the fatty acid profile obtained, the

C. BAYAGA & G. DEVEZA

most predominant was C+2 : + (-+./-ῌ), followed by C+0 : * (-*./.ῌ), C+2 : , (1.1*ῌ). The abundance of fatty acids is almost the same but the quantities of these fatty acids vary. In another study by Cabling et al. (+32,), some di#erences in the fatty acid composition of milkfish were observed. The levels of C+0 : +, C+2 : *, C+2 : +, and C,* : / of the milkfish in their study were higher compared to the values obtained in the present study. Many factors may have contributed to the di#erences and may involve the place of origin. The milkfish used in this study were obtained in Lingayen, Pangasinan Gulf, while that of Cabling’s study were from Laguna Lake. Other factors may also include the spawning and harvesting seasons, fish diet, and moisture content of the fish, which may influence the fat and lipid levels of the milkfish. For the Paksiw sample, the most abundant fatty acids were C+0 (--.0,ῌ) and C+2 : + (-*.1+ῌ). This cooking method brought about significant gains in C+0 : + (,.,2ῌ), C,* : - (+.+1ῌ) fatty acids. However, significant deceases in C+/ (*..2ῌ), C,* : , (+.,+ῌ) and C,* : . (*.0+ῌ) were observed and C,* : - and C,* : / decreased markedly. The most predominant fatty acids in Sinigang are the same as those of Paksiw ; C+0 and C+2 : +. This may be attributed to the similarity of cooking methods of the two recipes. Significant gains in C,* (*./0ῌ), C,, : + (*.-3ῌ) and C,, : , (+.*2ῌ) fatty acids were observed. There were significant decreases in C,* : , (+.0,ῌ), C,* : - (+.,/ῌ) and C ,* : . (*.**ῌ). For the fried milkfish, the most prevalent fatty acids were C2 (+1.,+ῌ), C+, (,-.+,ῌ), C+. (3.3*ῌ), C+0 (+1.03ῌ), and C+2 : + (+-.3/ῌ). The abundance of lauric acid (C+,) was due to the absorption of coconut oil by the milkfish that occurred during frying as lauric acid is the most abundant fatty acid in coconut oil. Significant gains were observed in C2 (+1.,+ῌ), C+* (-.31ῌ), C+, (,-.+,ῌ), C +. (3.3*ῌ), C,* : - (*../ῌ) and C,, : , (*...). Decreases were significant in C+0 (+1.03ῌ), C+2 (,.2/ῌ), C+2 : + (+-.3/ῌ), C+2 : , (+.//ῌ), C,* : + (+.+0ῌ), C,* : , (*./-ῌ), C,* : - (*../ῌ) and C,* : . (*.**ῌ). Decreases in unsaturated and polyunsaturated fatty acids may be attributed to oxidation during cooking at very high temperatures (,**ῌ ,+*῍). Fatty acid profile was also determined in the fresh and used oils. In the Philippines, coconut oil is generally used for frying food. It is marketed as refined, bleached, and scented coconut oil. Fatty acid analysis of the oils was done to determine possible fatty acid transfer between the milkfish sample and the oil used for frying. Using Student’s t-test, the di#erences between the fresh and used oils were determined. Significant di#erences were observed only in gamma linolenic acid (C+2 : -) ; fresh oil had *.+0ῌ and used oil had *.+*ῌ. The di#erence may be attributed to the oxidation of the fatty acid during cooking at high temperatures (,**ῌ,+*῍).

Summary and Recommendations Milkfish is a common food commodity for Filipinos. Among the cooking methods used to prepare it, Paksiw is

Consumption and DHA Levels in Milkfish (Paper #*.*0,)

the method that has the least e#ect on the proximate composition on its wet and dry bases compared to raw fish samples. It also exhibits the least DHA decrease among the three cooking methods. Further studies on the e#ect of di#erent seasons, water type and other cooking methods on milkfish should be undertaken. The amount of each fatty acid per total fat or per gram of sample (mg fatty acid/g sample) would also help in preparing baseline data for recommendations for pregnant and lactating mothers. Measurement of fatty acid profiles, especially DHA, of other fish and seafood would be useful for comparing which fish or seafood would be more beneficial to human beings. Acknowledgement The authors acknowledge with thanks the O$ce of the Vice-Chancellor for Research and Development (OVCRD), University of the Philippines, Diliman for providing the funds for this research and Dr. Victor C. Gavino of the University of Montre ´al, Canada for his comments and suggestions on the research methods.

References Association of O$cial Analytical Chemists. +33*. O$cial Methods of Analysis. +.th ed. Association of O$cial Analytical Chemists. +33/. O$cial Methods of Analysis. +0th ed. Bagarinao, T.U. +333. Ecology and Farming of Milkfish. Iloilo City : SEAFDEC Aquaculture Department. Bautista, MN, del Valle, M. J. and Orejana, F.M. +33+. Lipid and Fatty Acid Composition of Brackish water and Freshwater reared Milkfish (Chanos chanos Forskaal). Aquaculture 30 : ,.+ῌ ,.2. Cabling Jr., F, Takama, K and Zama, K. +32,. Lipids of Milkfish.

133 Bulletin of the Faculty of Fisheries, Hokkaido University. -(+) : -/ῌ.+. Carlson, S.E., Werkman, S.H., Rhodes, P.G., Tolley, E.A. +33-. Visual-acuity Development in Healthy Infants : E#ects of Marine-Oil Supplementation. Am J Clin Nutr. /2 : -/ῌ.,. Frenoux, J.R., Prost, E.D., Belleville, J.L., Prost, J.L. ,**+. A Polyunsaturated Fatty Acid Diet Lowers Blood Pressure and Improves Antioxidant Status in spontaneously Hypertensive Rats. J Nutr. +-+ : -3ῌ./. Hardman, W.E., Reddy Avula, C.P., Fernandes, G., Cameron, I.L., ,**+. Three Percent Dietary Fish Oil Concentrate Increased E$cacy of Doxorubicin Against MDA-MB ,-+ Breast Cancer Xenografts. Clin Cancer Res. 1 : ,*.+ῌ,*.3. Innis, S.M., Nelson, C.M., Rioux, M.F., and King, D. J. +33.. Development of visual acuity in relation to plasma and erythrocyte omega-0 and omega-- fatty acids in healthy term gestation infants. Am J Clin Nutr. 0* : -.1ῌ-/,. Kinsella, J.E. +320. Food Components with potential Therapeutic Benefits : The n-- Polyunsaturated Fatty Acids of Fish Oil. Food Technol. ῌ : 23ῌ31. Makrides, M., Neumann, M., Simmer, K., Pater, J., Gibson, R. +33/. Are Long Chain Polyunsaturated Fatty Acids Essential Nutrients in Infancy? Lancet. -./ (230-) : +.0-ῌ+.00. Nawar. +33*. Lipids in Food Chemistry. Fennema, O.R. (ed.). New York : Marcel Dekker, Inc. Skibsted, L.H., Mikkelsen, A. and Bertelsen, G. +332. Lipid Derived O# Flavor in Meat. Ch. +* in Flavor of Meat, Meat Products and Seafoods. Shahidi F. (ed). London : Blackie Acad and Professional. The Philippine Food Composition Tables. +331. Taguig, Metro Manila : Food and Nutrition Research Institute, Dept. of Science and Technology. Tiangson, C.L.P, Gavino, V.C., Gavino, G, and Panlasigui, L.N. ,**-. Docosahexaenoic acid level of the breast milk of some Filipino women. Int J Fd Sci Nutr. /. : -13ῌ-20.