<<

PRODUCTION OF OMEGA- 3 CONCENTRATES FROM FISH AND ALGAL AND THEIR STRUCTURAL ASSESSMENT BY NMR

Kralovec, JA 1, Weijie W 1, Barrow CJ 2 and Reyes-Suarez E 1

1Ocean Nutrition Canada Ltd. Dartmouth, Nova Scotia, B2Y 4T6, Canada; 2School of Life and Environmental Science, Faculty of Science and Technology, 1 Pigdons Road, Geelong, Victoria 2127, Australia

1 LC-PUFA, OMEGA-3 vs. OMEGA-6

H3C COOH ααα-linolenic acid (ALA, C18:3, n-3)

H C 3 COOH linoleic acid (C18:2, n-6)

H C 3 COOH arachindonic acid (ARA, C18:4, n-6)

H3C COOH eicosapentaenoic acid (EPA, C20:5, n-3)

COOH H3C docosahexaenoicoic acid (DHA, C22:6, n-3)

 LC-PUFAs are categorized by the position of the 1 st double

bond counting from the CH 3-terminal  Location of the 1 st double bond dictates biological activity

 All double bonds are CH 2-interruped  Omega-3 series starts with α-linolenic acid (ALA,18:3n-3)

 Omega-6 series starts with linoleic acid (LA,18:2n-6) Calder. Schweiz. Zeit. Ernährungmedizin 2009 , 5, 14-19  The most studied omega-3 FA are EPA and DHA 2 PRODUCTION OF EPA/DHA CONCENTRATES

STARTING OMEGA-3 SELECTION

 The richest and the cheapest source of EPA/DHA is fish oil

 Starting oils with right FA profiles and EPA/DHA content ~30%!

 anchovy/sardine (18/12TG, for high level EPA concentrates)  tuna (05/25TG, for high level DHA concentrates)

CORE OMEGA-3 CONCENTRATION O COR 1 TECHNOLOGIES

 Fish oils are in the form of →→→ O COR 2

 Winterization O COR  Cold Solvent Precipitation 3

 Urea Clathration

 Ethanolysis + Fractional Distillation ⇒ Ethyl (EE) Concentrates

 -assisted Concentration ⇒⇒⇒ (TG) Concentrates 3 EE CONCENTRATES

ETHANOLYSIS AND DISTILLATION

Ethanolysis Distillation TG EE EE 30% 30% 60%

Ethanolysis Distillation - TG EE EE

Medium chain or medium Splitting TG molecules lowers the boiling points ! chain EE DHA-TG bp ~ 910 oC, vs. DHA-EE bp ~ 450 oC) DHA residues or DHA-EE EPA residues or EPA-EE backbone or glycerol Popular technology towering over the other options O = CI 1.67 O, output CI, capital investment

Economic advantage of large throughput! 4 CONCENTRATION STRATEGY A ETHANOLYSIS / DISTILLATION / GLYCEROLYSIS

Ethanolysis Distillation - TG EE EE TG

GLYCEROLYSIS EE TG HO FFA RCOO

i RCOOR1 + HO RCOO - R1OH

HO RCOO

CALB R, a mixture of C14-C24 hydrocarbon residues,, primarily those of EPA and/or DHA R1, H or Et; i, CALB, 60-85 C, vacuum

 Immobilized Candida antarctica B (CALB) in Packed Bed Reactor

 The reaction is driven forward by EtOH or water removal under vacuum 5 PACKED BED REACTOR

Trap (condenser) Evaporation tank

Heated stirred tank

Vacuum pump

Distributor

Reactor

Pump Pump

6 MANUFACTURING OF TG CONCENTRATES

 Green technologies (free or immobilized )  Packed bed reactors

Reaction mixture cycling trough PBR

 First to introduce food grade immobilized lipase for EE –TG conversion !

i ii EE FFA TG

o i: CALB-L, EtOH/H 2O, 50 C, vacuum ii: CALB-FPX66, glycerol, 70 oC, vacuum ~ 80 reactions on 1 batch of immobilized. enzyme! 7 CONCENTRATION STRATEGY B

HYDROLYSIS AND EPA (DHA) INSERTION

Hydrolysis EPA/DHA Insertion - TG DG TG

Hydrolysis: Lipo-JD, P buffer pH7.5, 40 oC, 42 h; EPA/DHA insertion: CALB, EPA/DHA source, 70 oC, 24h

TIME COURSE OF 18/12TG HYDROLYSIS

35 DHA in glyceride EPA in glyceride 30 DHA in FFA EPA in FFA 25

20

15 EPA/DHA Content (%) (%) EPA/DHA Content

10 Candida rugosa lipase 5

0 0 10 20 30 40 50 60 8 Total FFA in reaction mixture (%) POSITIONAL DISTRIBUTION (PD) OF EPA/DHA IN TG OILS

It is very likely that the oxidative stability and thus also sensory qualities of EPA/DHA rich TG oils and their TG concentrates correlate with the positional distribution of fatty acid residues, primarily EPA and DHA in TG molecules.

FACTS AND STRATEGY OCOR1 Complexity

 Fish oil contains 30 to 60 different fatty acids  Assuming 30 and taking into account positional distribution, OCOR2 theoretically 27000 different TGs

Chromatography

 Complex mixtures, need for non-aqueous conditions OCOR3

If R 1≠R2, C 2 is chiral Mass spectrometry

 Neutral compounds - poor ionization, difficulty adding reagents to mobile phase for adduct formation

NMR 9  Complex mixtures, limited data in the literature SYNTHESIS OF STANDARDS

O O SYMMETRIC (CH2)14CH3 (CH2)14CH3 O O DHA EDCl, DMAP HO DHA O

CH2Cl2 O O (CH2)14CH3 (CH2)14CH3 O O 1,3-dipalmitate O O ASYMMETRIC (CH ) CH (CH ) CH 2 14 3 2 14 3 O O O O DHA H C(H C) Method A H C(H C) 3 2 14 3 2 14 EDCl, DMAP O O CH Cl 2 2 DHA O HO 1,2-dipalmitate EDCl, DMAP CH2Cl2

HO Method B O DHA O Amberlyst O EDCl, DMAP O HO MeOH, CH2Cl2 CH2Cl2 HO DHA O DHA O 10 Wijesundera et. al. J. Amer. Oil Chem. Soc . 2007 , 84 , 11-21 POSITIONAL DISTRIBUTION OF EPA AND DHA BY NMR

13 C NMR (preferred)

 Better resolution (larger spectral range) than 1H NMR, although lower sensitivity

CARBONYL (C-1) CARBONS (preferred)

 Carbonyls from sn-1,3- and sn-2 chains form clusters of peaks

 Within each cluster, chemical shifts differences between acyl chains occur as a result of the proximity to a neighboring double bond

 Unable to discriminate between sn -1 and sn -3 positions

11 POSITIONAL DISTRIBUTION OF EPA AND DHA BY NMR

Carbonyl region of the 125 MHz 13 C NMR spectrum of a mixture of TG standards of seven fatty acids BASIC FACTS

EPA  Carbonyls from sn -1,3 chains resonate at (sn -1,3) higher chemical shifts than sn -2 chains EPA DHA (sn -2) (sn -1,3)  Within each cluster carbonyls from saturated DHA (sn -2) chains resonate at a higher chemical shift

 Chemical shifts are in a direct correlation with the distance of the carbonyl carbon to sn -1,3 sn -2 the first double bond 173.4 173.2 173.0 172.8 172.6 172.4 ppm  Unsaturation on the γ-position ( ∆4, DHA)

Carbonyl chemical shifts of acyl chains that differ in double bond induces a greater shift on carbonyl chemical group shifts; DHA carbonyls resonate in a unique 173.40 γ to an olefinic 173.20 carbon position, which allows a straightforward 173.00 172.80 assignment 172.60 172.40

Chemical shift (ppm) shift Chemical C-1 (sn-1,3) 172.20 C-1 (sn-2) 172.00 Sat Δ9 Δ8 Δ7 Δ6 Δ5 Δ4 DHA Position of the first double bond relative to the c arbonyl carbon 12 POSITIONAL DISTRIBUTION OF EPA AND DHA BY NMR

INVESTIGATED FISH OILS QUANTITATIVE 13 C NMR

 18/12TG (sardine oil)  Under full-decoupling conditions nOe  Re-esterified 18/12TG build up may differ for carbonyls of sn -1,3 and sn -2 chains  05/25TG (tuna oil)  Sn -1,3/ sn -2 area ratios were measured for EPA and DHA carbonyls as a function of the

relaxation delay for 18/12TG (100mg/0.7mL CDCl 3)

Variation of the carbonyl areas ratio: A(sn-1,3)/A(sn-2) for EPA and A(sn-2)/A(sn- 1,3) for DHA as a function of the relxation delay in a full-decoupled sequence Area ratios for both EPA and 5 4.5 DHA carbonyls remain 4

3.5 DHA (C=O) constant after 12s delay Areas ratio Areas 3 EPA (C=O) 2.5 (working range) 2 1.5 1 1.5 3 4.5 6 7.5 9 10.5 12 13.5 15 16.5 18 19.5 21 22.5 Relaxation delay (s) 13 POSITIONAL DISTRIBUTION OF EPA AND DHA BY NMR

Carbonyl region of the 125 MHz 13 C NMR spectrum of Carbonyl region of the 125 MHz 13 C NMR spectrum of

18/12TG oil (100 mg/ 0.7 mL CDCl 3) Re-esterified 18/12TG oil (100 mg/ 0.7 mL CDCl 3)

Monounsat Sat Monounsat Sat (sn -2) Sat Sat (sn -1,3) (sn -1,3) (sn -1,3) (sn -1,3) (sn -2) Monounsat DHA Monounsat DHA EPA (sn -2) (sn -2) EPA (sn -2) (sn -2) (sn -1,3) EPA (sn -1,3) EPA (sn -2) DHA (sn -2) DHA STA STA (sn -1,3) STA STA (sn -1,3) (sn -1,3) (sn -2) (sn -1,3) (sn -2)

173.4 173.2 173.0 172.8 172.6 172.4 172.2 ppm 173.4 173.2 173.0 172.8 172.6 172.4 172.2 ppm

Carbonyl region of the 125 MHz 13 C NMR spectrum of

0525TG tuna oil (100 mg/ 0.7 mL CDCl 3) Sat (sn -1,3) Monounsat Sat DHA (sn -1,3) (sn -2) (sn -2) DHA (sn -1,3) Monounsat (sn -2) EPA (sn -1,3) EPA (sn -2)

173.6 173.4 173.2 173.0 172.8 172.6 172.4 172.2 ppm 14 POSITIONAL DISTRIBUTION OF EPA AND DHA BY NMR

Molar percentages of individual chains insn -1,3 an sn -2 glycerol RE-ESTERIFICATION OF 18/12TG positions (fish oil TG 18/12)

Sat Ethanolysis % 30 Mono STA - 25 EPA 20 DHA (No concentration of EPA or DHA) 15 10  The distribution of saturated and mono- 5 unsaturated chains as well stearidonic 0 % (sn-1,3)/Total % (sn-2)/Total acid (STA) remains unaffected

Molar percentages of individual chains snin -1,3 andsn -2 glycerol positions  The distribution of EPA and DHA chains (TG 18-12 re-esterified) is affected % Sat 30 Mono STA 25  EPA that was mostly in sn -1,3 is EPA 20 DHA randomized on re-esterification 15 10  DHA that was mostly in sn -2 is 5 located mostly in sn -1,3 on re- 0 % (sn-1,3)/Total % (sn-2)/Total esterification

15 POSITIONAL DISTRIBUTION OF EPA AND DHA BY NMR

05/25TG TUNA OIL

Molar percentages of individual chains snin-1,3 andsn -2 glycerol positions (tuna oil)  Saturated and monounsaturated Sat % 40 Mono chains are preferably located in 35 EPA sn -1,3 positions 30 DHA 25 20 15  EPA is present in equal amounts 10 in both sn -1,3 and sn -2 positions 5 0 % (sn-1,3)/Total % (sn-2)/Total  DHA shows a slight preference for the sn -2 position

16 SUMMARY

 The evidence for the beneficial effects of omega-3 fatty acids FA, particularly EPA and DHA is undisputable.

 Science and innovation are the major drivers of dietary supplement and pharmaconutrient industry.

 To deliver high quality omega-3 products the replacement of traditional chemistry by enzymatic technologies for manufacturing omega-3 based products is becoming more and more important.

 The understanding of the positional distribution (PD) of the FA residues in the oils is of crucial importance since PD might be important for the stability, sensory and biological effects.

 ONC is leading the way in implementation of sophisticated manufacturing technologies and thorough understanding of physiochemical properties of omega-3 fatty acids.

17 THANK YOU !

18