The regiospecific characteristics of n-3 polyunsaturated fatty acids (PUFAs) in triacylglycerol (TAG) significantly affect the physicochemical and physiological properties of marine fish oils. In this study, the TAG molecular species composition and positional distributions of fatty acids were investigated in three marine fish species rich in omega-3 PUFAs (anchovy, tuna, and salmon). The regiospecific distribution of the fatty acids was measured with the allylmagnesium bromide (AMB) degradation method. The TAG compositions were analyzed with HPLC and the TAG molecular species were identified with APCI/MS. DHA was preferentially distributed at the sn-2 position of TAG, whereas EPA was evenly distributed along the glycerol backbone. The combinations of FAs, DDO, EOP, EPS, DSS, OOS, and PPS were the predominant TAG molecular species, and OOP, DOS, and DPoPo were the characteristic TAG molecules in the anchovy, salmon, and tuna, respectively. These data can be used to distinguish other marine fish species. The TAG composition categorized by TCN and ECN showed well-structured distributions, with double or triple peaks. These findings should greatly extend the use of marine fish oils in food production and may significantly affect the future development of the fish oil industry.
The health benefits of omega-3 polyunsaturated fatty acids (n-3 PUFAs), especially eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), are well documented in the literature. DHA in the retina has been associated with visual outcomes and in the brain with intellectual and behavioral outcomes [
At present, the most common sources of n-3 PUFAs are oils of marine origin. Marine fish oils are rich in n-3 PUFAs, typically containing between 20% and 30%, and DHA and EPA account for more than 80% of these total n-3 PUFAs [
Structure of the triacylglycerol (TAG) molecule.
Furthermore, the characteristic effects of TAGs on digestion, absorption, and transportation are closely associated with the present TAG molecular species and their fatty-acid-binding positions [
To the best of our knowledge, little detailed data is available on the regiospecificity of the FAs in the TAGs in marine fish oils and particularly the structures of TAGs containing PUFAs such as DHA and EPA. Therefore, the main objective of this study was to extend our knowledge of the characteristic positional distributions of TAG FAs and the TAG compositions of marine fish oils rich in n-3 PUFAs. This should improve the exploitation potentials of these valuable and underutilized resources.
Three species of marine fish oil, extracted from the anchovy, tuna, and salmon, were donated by Zhonghai Ocean Technology Co., Ltd (Jiangsu, China). Silica gel GF254 thin-layer chromatography (TLC) plates were purchased from Haiyang Chemicals Co., Ltd (Qingdao, China). The chemicals used for high-performance liquid chromatography (HPLC) were of chromatographic-grade purity and were purchased from J&K Chemical Scientific Co., Ltd (Shanghai, China). Other chemicals were of analytical grade and were purchased from Sinopharm Chemical Reagent Co., Ltd (Beijing, China). Forty fatty acid methyl ester (FAME) standards and allylmagnesium bromide (AMB) were purchased from Sigma-Aldrich (St. Louis, MO, USA).
The FAMEs of the three marine fish TAGs were prepared with the method described by Birch et al. [
The FAs composition was analyzed as described by Li et al. [
The samples were partially degraded with AMB according to the method adapted by Xu et al. [
Sn-2 monoglyceride (sn-2 MAG) was separated with thin-layer chromatography (TLC). TLC plates were coated with 0.4 M boric acid, air-dried overnight, and stored in a desiccator until use. The sn-2 MAG fraction was separated with TLC on boric-acid-impregnated silica gel plates with 100 mL of developing solvent, chloroform: acetone (90:10). The plates were developed twice for 45 min each, with a 10-min drying period. The sn-2 MAG band was removed and extracted three times with 1 mL of diethyl ether. The corresponding bands were identified, scraped off, and extracted three times with diethyl ether.
Sn-2 MAG was methylated with potassium hydroxide with the IUPAC method [
The FAMEs were analyzed with the GC analysis procedure and parameters described in the previous section.
The TAG samples (50 mg) were separated on TLC plates (20 × 20 cm2, 0.25 mm thickness) with a mixture of n-hexane: diethyl ether: acetic acid (80:20:1, v:v:v) for molecular species analysis. The bands corresponding to the TAGs were scraped off and recovered by extraction with n-hexane. The TAG molecular species were analyzed with RP-HPLC-ELSD.
The recovered TAGs (30 mg) were dissolved in 1 mL of n-hexane, and a 10 mL portion of this solution was injected into the HPLC apparatus equipped with an ELSD (Waters, USA). The ELSD was set at 55°C, with a nitrogen nebulizer gas at a flow rate of 1.8 mL/min. Separation was performed on a 250 × 4.6 mm i.d. 5
The TAG molecular species were identified with HPLC-APCI/MS. The analyses were performed in a solvent delivery system coupled to a Micromass ZQ Mass Spectrometer (Waters) fitted with an APCI source, with full-scan acquisition. Data acquisition, processing, and instrument control were managed with the Xcalibur™ software (Thermo Scientific). The instrumental conditions were vaporizer temperature 400°C, capillary voltage 6.0 kV, and corona voltage 40 V. The spectra were obtained over the range of m/z 80–2000, with a scan time of 1.0 s.
At least three (n ≥ 3) samples of each marine fish oil were used for the analysis. All analyses were conducted in triplicate, and the means ± standard deviations were calculated with the Microsoft Office statistical software. One-factor ANOVA and a
The FA compositions of the TAGs from anchovy, salmon, and tuna oils are shown in Table
Major fatty acid compositions of TAGs in marine fish
| | | ||
---|---|---|---|---|
| | % | % | % |
| | 7.88±1.21 | 12.24±0.98 | 9.46±1.06 |
| | 14.74±1.98 | 35.90±2.06 | 20.00±2.05 |
| | 6.96±0.68 | 8.33±0.56 | 11.73±0.99 |
| | 10.51±0.64 | 4.75±0.35 | 3.97±0.08 |
| | 32.92±2.01 | 16.19±1.09 | 22.00±1.14 |
| | 17.00±1.47 | 12.54±0.32 | 10.90±0.58 |
| | 9.99±0.25 | 10.04±0.68 | 21.94±1.39 |
Superscript letters in a row indicate significant differences.
Among the SFAs, the levels of C16:0 (35.90%) were significantly higher in anchovy than in salmon or tuna, whereas the level of C18:0 (10.51%) was highest in salmon. Among the MUFAs, the most abundant FA in salmon was C18:1 (32.92%), whereas C16:1 (11.73%) was most abundant in tuna. Among the PUFAs, the level of DHA (C22:6) was clearly higher in tuna (21.94%) than in the other two species, whereas the EPA content was highest in salmon (17.00%). EPA and DHA accounted for 20%–30% of the total FAs and for > 50% of the total unsaturated FAs. It is well known that ALA acts as the precursor of the longer-chain (LC) n-3 PUFAs and can be converted to EPA or DHA. However, marine fish species are incapable of desaturating and elongating ALA to LC-PUFAs because the activity of their delta-6 desaturase enzyme is low [
The distributions of the major PUFAs, MUFAs, and SFAs in the sn-2 positions of the fish TAGs considered here are shown in Table
sn-2 positional compositions of major fatty acids in TAGs of marine fish
| | | ||||
---|---|---|---|---|---|---|
| sn-2 | %sn-2 | sn-2 | %sn-2 | sn-2 | %sn-2 |
(mol%) | (%) | (mol%) | (%) | (mol%) | (%) | |
| 6.33±0.28 | 31.54±0.39 | 12.79±0.59 | 41.87±0.27 | 3.12±0.25 | 12.86±4.88 |
| 11.37±0.66 | 28.29±1.64 | 27.35±1.27 | 30.54±0.20 | 16.08±0.42 | 31.35±4.33 |
| 4.38±1.12 | 24.74±1.02 | 6.85±0.32 | 32.95±0.21 | 5.06±0.12 | 16.81±0.40 |
| 8.91±0.64 | 33.2±0.08 | 2.69±0.13 | 22.68±0.14 | 3.26±0.27 | 32.03±0.72 |
| 16.15±1.02 | 19.27±0.74 | 8.71±0.40 | 21.58±0.14 | 14.68±0.43 | 26.01±0.54 |
| 17.54±0.01 | 36.31±0.05 | 10.54±0.63 | 33.64±0.28 | 9.56±0.45 | 34.17±4.60 |
| 12.62±1.86 | 50.61±3.07 | 11.59±0.54 | 49.28±0.30 | 25.88±0.49 | 49.00±0.86 |
%sn-2 calculations were based on more than three samples with triplicate measurements per sample.
Different superscript letters in a row indicate significant differences (
In general, the total SFAs and MUFAs are present at the sn-1 and sn-3 positions, but individual SFAs and MUFAs can occur at various sites. In this study, the saturated residues were assembled differently in the marine fish TAGs. For instance, the C18:0 and C16:0 residues occurred nearly equally at the terminal sn-1 and sn-3 sites and in the middle sn-2 position, respectively, with the exception of C18:0 in anchovy. The percentage of C14:0 differed in each fish species. It was mainly esterified in the sn-1 and sn-3 positions (14.16%) in tuna, but with a preference for the sn-2 position (43.25%) in anchovy, whereas in salmon, it was equally distributed across all three positions (34.63% each). In contrast, MUFAs C18:1 and C16:1 predominantly occurred at the terminal sn-1 and sn-3 positions, except for C16:1 in anchovy, which appeared to be equally distributed between sites.
Several approaches to the separation of TAGs have been developed, depending on the characteristics of the fats and oils involved. Chromatographic separation is the most basic way to resolve molecular species and provides information on the combinations of FAs in the TAG molecule. High-temperature GC (HT-GC) and HPLC are the most common chromatographic separation techniques and are frequently used to separate TAG molecules from vegetable oils and animal fats. Because the separation of TAGs by HT-GC is according to the total carbon number (TCN), TAGs with the same TCN cannot be efficiently separated with GC. Thermal instability and oxidation are the major problems that arise during the HT-GC analysis of TAGs containing PUFAs [
The main TAGs in the marine fish oils were successfully separated and large amounts of TAGs were detected as relatively intense peaks on RP-HPLC-ELSD. Approximately 40 TAG species were eluted into different peaks with RP-HPLC and about 30 TAG molecules were identified with APCI-MS (listed in Table
Triacylglyceride compositions of marine fish
| | | | | | |
---|---|---|---|---|---|---|
| 60 | 15 | 30 | 1.09±0.26 | 0.56±0.06 | |
| 66 | 18 | 30 | 0.75±0.04 | ||
| 58 | 11 | 36 | 5.67±1.25 | 2.23±0.16 | |
| 62 | 13 | 36 | 9.70±1.38 | 0.20±0.01 | 4.69±0.89 |
| 60 | 12 | 36 | 1.14±0.41 | 1.12±0.78 | |
| 52 | 7 | 38 | 2.64±0.56 | ||
| 54 | 8 | 38 | 2.46±0.12 | 6.56±1.25 | |
| 50 | 6 | 38 | 8.74±1.11 | ||
| 48 | 5 | 38 | 4.54±0.98 | 3.18±0.38 | 12.26±3.68 |
| 50 | 6 | 38 | 5.29±1.00 | ||
| 56 | 8 | 40 | 6.94±0.44 | 0.32±0.05 | 1.31±0.10 |
| 54 | 7 | 40 | 1.79±0.04 | 0.22±0.01 | 1.47±0.20 |
| 52 | 6 | 40 | 1.03±0.02 | 1.90±0.10 | 1.65±0.68 |
| 58 | 8 | 42 | 4.12±0.89 | 3.75±0.69 | |
| 54 | 6 | 42 | 11.17±2.00 | 4.54±1.02 | |
| 56 | 7 | 42 | 4.51±0.55 | 6.10±1.45 | |
| 52 | 5 | 42 | 1.13±0.33 | ||
| 54 | 6 | 42 | 2.10±0.33 | 1.99±0.69 | 1.03±0.54 |
| 56 | 6 | 44 | 2.78±0.68 | 3.02±0.95 | |
| 58 | 7 | 44 | 10.47±2.34 | ||
| 54 | 5 | 44 | 3.89±0.66 | 2.86±0.58 | 17.39±3.81 |
| 56 | 6 | 44 | 2.70±0.21 | 3.66±0.93 | 7.27±2.11 |
| 56 | 5 | 46 | 2.56±0.34 | ||
| 58 | 6 | 46 | 2.31±0.29 | 11.67±2.51 | 1.16±0.02 |
| 54 | 3 | 48 | 0.54±0.01 | 0.53±0.04 | 1.70±0.12 |
| 52 | 2 | 48 | 3.17±0.58 | 0.61±0.02 | |
| 50 | 1 | 48 | 1.94±0.67 | 1.07±0.07 | 1.00±0.22 |
| 54 | 2 | 50 | 8.50±1.80 | 0.91±0.05 | |
| 50 | 0 | 50 | 3.41±0.39 | 12.18±2.22 | 2.35±0.03 |
| 54 | 1 | 52 | 1.14±0.24 | 1.97±0.32 | 2.11±0.09 |
| 52 | 0 | 52 | 2.92±0.41 | ||
| 54 | 0 | 54 | 2.84±0.54 | 0.80±0.11 |
Fatty acid symbols: M, myristic acid; P, palmitic acid; Po, palmitoleic acid; S, stearic acid; O, oleic acid; E, EPA, eicosapentaenoic acid; D, DHA, docosahexaenoic acid. Values are means ± standard deviation with triplicate measurements per sample (mass %). Different superscript letters in a row indicate significant differences (
Mass spectrometry is a commonly used means of resolving molecular structures and can be used to determine the molecular masses of TAGs. Electrospray ion (ESI) and APCI are soft ionization techniques that generate intact TAG molecular ions and produce fragments that are useful for structural characterization when combined with MS [
Fragment ions of TAG obtained with atmospheric pressure chemical ionization (APCI).
We successfully identified 30–40 major TAG molecular species in these three marine fish oils with HPLC–APCI-MS. Supplementary Tables
As mentioned above, the TAG species were separated with RP-HPLC and identified with APCI-MS. The major TAGs identified in the marine fish oils were expressed as percentages of the total TAGs (Table
In the anchovy, the predominant TAG molecule was EOP (11.17%), followed by DDO (9.70%) and EPoM (8.74%). These top three TAG species accounted for up to 30% of the total TAGs. It is noteworthy that OOP accounted for 3.41% of the total TAG, which was not high and differed from the OOP levels in the other two marine fish oils. In the salmon, the predominant TAG was PPS (12.18%), followed by DSS (11.67%), DOS (10.47%), and OOS (8.50%), which distinguished salmon oil from the oils of the other fish species. These three TAGs accounted for more than one-third of the total TAGs in salmon oil. These data confirm that salmon contains much more stearic and oleic acids than anchovy or tuna. Compared with the other two fish species, the tuna TAGs showed a more complex molecular species composition. The dominant TAG was EPS (17.39%), followed by EMM (12.26%), and these two TAG molecules accounted for nearly 30% of the total TAGs. However, the dominant TAG in the oil from the belly and skeletal muscles of the tuna was PDD (20.8% and 15.8%, respectively), followed by POD (9.7% and 13.0 %, respectively) [
The TAG molecular compositions of the marine fish oils varied quite widely. The characteristic TAG species in the oils are summarized in histograms in Figure
Distributions of TAGs according to total carbon number (TCN) and equivalent carbon number (ECN).
In these three fish species, the TAGs showed well-structured double or triple peaks for TCN. In the anchovy (black), the TAGs were mainly distributed in TCN50, ECN54, and ECN56, and these three groups account for 56.73% of the total TAGs. These three bands principally consisted of EPoM (8.74%), EOP (11.17%), and DPoO (6.94%), which contained EPA, DHA, palmitic acid, palmitoleic acid, and oleic acid. In the salmon (white), the TAGs were enriched in TCN54 and TCN58, which accounted for 28.49% and 25.91% of the total TAGs, respectively. These two groups were chiefly represented by combinations of EPA, DHA, oleic acid, stearic acid, and palmitic acid and consisted of DSS (11.67%), DOS (10.47%), OOS (8.50%), and EOP (4.54%). In the tuna (gray), the TAGs were enriched in TCN48, TCN54, and TCN56. TCN54 was most common, accounting for almost one-third of the total TAGs (31.97%), and mainly consisted of EPS (17.39%) and DPoPo (6.56%). Briefly, TCN54 was the most frequently represented group in the TAGs of the three marine fish oils. The most frequently represented TAGs were predominantly combinations of EPA or DHA with SFAs and MUFAs containing 16 or 18 carbon atoms. Furthermore, specific TAGs were representative of each fish species and could be used to distinguish different marine fish oils, because their relative contents are very significantly higher in one species than in the others: EOP in anchovy, EPS in salmon, and OOS in tuna.
The distribution of TAGs was also characteristic of each marine fish oil. Tuna oil contained a greater proportion of TCN48 TAGs than salmon or anchovy oil, whereas salmon oil was richer in TCN58 TAGs than were the other two marine fish species. The percentage of TCN50 TAGs was higher in both anchovy and salmon oils than in tuna oil, whereas the percentage of TCN56 TAGs was significantly higher in anchovy and tuna than in salmon.
The TAGs also showed well-structured double or triple modal distributions according to ECN. In anchovy (black), the TAGs were mainly ECN36, ECN38, and ECN42, and these three groups accounted for 55.52% of the total TAGs and principally consisted of DDO (9.70%) and EPoM (8.74%). In salmon (white), the TAGs were enriched in ECN44 (19.77%) and ECN50 (20.68%) and chiefly consisted of DOS (10.47%), OOS (8.50%), and PPS (12.18%). In tuna (gray), the TAGs were mainly enriched in ECN44 (27.68%), followed by ECN38 (18.82%). These two groups accounted for nearly 50% of the total TAGs and mainly consisted of EPS (17.39%) and EMM (12.26%). When these marine fish TAGs were compared, anchovy oil had a higher content of ECN36 TAGs than salmon or tuna, whereas salmon had a higher content of ECN50 TAGs than the other two fish species. The percentage of ECN44 TAGs was markedly higher in salmon and tuna than in anchovy, whereas the percentage of ECN38 TAGs was distinctly lower in salmon than in anchovy or tuna. The three marine fish oils were abundant in ECN42 TAGs, which constituted > 10% of all three oils.
Panels A and B in Figure
In this study, the molecular TAG species present in anchovy, tuna, and salmon oils were first separated, identified, and discussed in detail, together with the regiospecific distributions of their FAs. Both the similarities and differences in the TAGs of these marine fish species were identified. DHA was the main esterified FA in the sn-2 position, whereas EPA was evenly incorporated at each position of the TAG molecules. Combinations of DDO, EMM, EOP, EPS, DSS, OOS, and PPS were the predominant TAG species, and OOP, DOS, and DPoPo were the characteristic TAG molecules, and these should be useful in distinguishing the three marine fish species. Management of the data according to TCN and ECN allowed the molecular TAG species to be classified and suggested that the distributions of TAGs display well-structured double or triple peaks. The TCN54 group was enriched in the TAGs and predominantly consisted of combinations of EPA or DHA with SFAs and MUFAs containing 16 or 18 carbon atoms. Therefore, the data management approach to TAG molecules based on TCN and ECN is useful in identifying the main features of different oils and allows both the differences and similarities of the samples to emerge. The lower ECN values (≤ 50) and higher TCN values (≥ 50) indicated that the most abundant TAGs had a high degree of unsaturation. Our findings should greatly extend the utilization of marine fish oils in food production and may significantly affect the future development of the fish oil industry.
The data used to support the findings of this study are available from the corresponding author upon request.
Huijun Zhang and Hui Zhao are co-first authors.
The authors declare that they have no conflicts of interest.
Huijun Zhang and Hui Zhao contributed equally to this work.
Observed ions in the HPLC-APCI-mass spectra of TAGs in positive and negative ion modes. Table 1: observed ions in the HPLC-APCI-mass spectra of major salmon TAGs in positive and negative ion modes. Table 2: observed ions in the HPLC-APCI-mass spectra of major anchovy TAGs in positive and negative ion modes. Table 3: observed ions in the HPLC-APCI-mass spectra of major tuna TAGs in positive and negative ion modes.