Casimir C. Akoh

Department of Food Science and Technology The University of Georgia, Athens, GA 30602

1 11/10/2015

Introduction Objectives Methodology Results Conclusions

2

3

TFA: Unsaturated fatty acids with carbon-carbon double bonds in trans orientation (Lee et al., 2007)

Stearic acid (C18:0)

Elaidic acid (C18:1t)

Oleic acid (C18:1c)

O

OH

CH3

HH

O

OH

CH3

O

OH

CH3

H

H

The two main sources of TFA are ruminant animals and partial hydrogenation

Partial hydrogenation of oil (PHO) is a common industrial process to solidify oils and increase their oxidative stability

PHO is the primary dietary source of industrially-produced trans fat (10-60%)

Hydrogenation converts liquid oils or soft fats into plastic or hard fats

In the U.S. diet, 80% of TFA consumed is due to partial hydrogenation (Eckel et al., 2007)

The average daily intake of TFA in North America is 3-4 g/person (Craig-Schmidt, 2006)

4

5

Cakes, cookies, crackers, pies, bread, etcAnimal products

Margarine

Fried potatoes

Potato chips, corn chips, popcornHousehold shortening

*Other

(Hunter, 2005)

May impair the metabolism of EFA involved in inflammatory pathways (Mojska, 2003)

Increases LDL-chol, decreases HDL-chol, (Mozaffarian and Clarke, 2009)

Each 2% increase in energy intake from TFA is associated with 23% higher incidence of CHD (Mozaffarian et al., 2006)

Increases weight gain (Axen et al.,2003)

Increases incidence of gallstones (Tsai et al., 2005)

Increases infertility in women (Chavarro et al., 2007)

Alzheimer’s disease in older adults (Morris et al., 2003)

6

www.abouttown.us/.../images/whats_wrong.jpg

Increased risk of CHD Increased diabetes risk Worsening insulin resistance Adverse effects on fetuses and

breastfeeding infants Impaired growth Total chol/HDL-chol and LDL-chol/HDL-

chol ratios

7

FDA, 2015

TFA consumption should be as low as possible (Dietary Guidelines for Americans, 2010; Institute of Medicine, 2002)

Population nutrient intake goal for TFA should be <1% energy (AHA, 2006)

United States Food and Drug Administration (FDA) - all foods containing ≥0.5 g trans fat/serving to be labeled accordingly, effective from January 2006 (Federal Register, 2003)

8

In November 8, 2013 FDA (78 FR 67169) made tentative determination that PHOs could no longer be considered GRAS and are therefore food additives

FDA, HHS final determination that PHOs no longer has GRAS status (June16, 2015) for use in human food, but not animal feed (Docket No. FDA-2013-N-1317)

Industry has until June 18, 2018 to comply and remove trans fats from foods

Conjugated linoleic acid (CLA) does not fit the definition of PHO and is excluded from FDA order

9

Naturally stable oils/fats Trait-enhanced oils Modification of hydrogenation process Interesterification – chemical or enzymatic Structured fats – chemically or

enzymatically prepared Physically blended fats and oils

10 (Hunter, 2005), Akoh, 2015

R1COOR2 + R3COOR4 catalyst R1COOR4+ R3COOR2

11

Chemical

Less costly

Non-specific

Poor control over the

final product

Harsh conditions

Enzymatic

Costlier

Both specific and non-specific

Good control

Milder reactions

(Modified from Marangoni and Rousseau, 1995)

Structured lipids (SLs): Lipids (here, TAG) that have been structurally modified from their natural form by changing the positions of FAs, or the FA profile, or synthesized to yield novel TAGs through chemical or enzymatic processes (Akoh, 2008)

sn-2 sn-2

sn-1

sn-3

[ R3

R4

R1

R2 R1

R2

12

Nutraceuticals Low/reduced fats Cocoa butter alternatives Infant formula trans-Free fats

13

Brands Description Applications Company

Caprenin® 8:0-10:0 (43-45%) and 22:0 (40-54%)

Ingredients for candy bars and confectionery coatings

Procter & Gamble

Benefat® 30–67% SCFA and 33–70% LCFA

Low calorie-chocolate-flavored coatings, baked and dairy products, dressings

Danisco A/S

Neobee® 8:0, 10:0, and LCFA (n-6 and n-3)

Pharmaceutical uses, incorporated in nutritional or medical beverages or in snack bars

Stepan Company

Impact® Randomized high-lauric acid oil and high linoleic acid oil

Pharmaceutical uses targeted for patients who have suffered surgery or cancer

Novartis Nutrition

Captex® Esterification of fractionated coconut or palm kernel oils (mainly caprylic and capric acids) and glycerol

Cinical applications

Abitec Corp.

Crucial® 50% fat source as MCT + marine oil and soybean oil (n-6/n-3, 1.5:1)

Clinical nutrition

Nestlé Nutrition

Betapol® Tripalmitin with UFA using 1,3-specific lipase

First commercially available enzymatically synthesized SL used in infant formula

Loders Croklaan

Vital AF1.2 Cal® Blend of SL (marine oil + MCT), MCT, canola oil, and soybean oil

Therapeutic elemental nutrition used for patients with inflammation and gastrointestinal disorders

Abbott Nutrition

14 (Modified from Akoh and Kim, 2008; Xu, 2005; Pande and Akoh, 2012)

α-Linolenic acid (ALA) (n-3) and Linoleic acid (LA) (n-6)

Western diet n-6:n-3 high: 15-16.7:1 (Simopoulos, 2008)

Blood pressure (Gelenjinse et al., 2002)

Risk of CVD (Calder, 2004)

Inflammation (Calder, 2006)

AHA Dietary Guidelines suggest Americans consume at least two servings of fish per week and include n-3 fatty acid, ALA, in their diet (Kris-Etherton et al., 2003)

15

DIET

C18:2n-6 (LA) C18:3n-3 (ALA)

C18:3n-6 (GLA) C18:4n-3 (SDA)

C20:3n-6 (DGLA) C20:4n-3

C20:4n-6(ARA) C20:5n-3 (EPA)

C22:4n-6 C22:5n-3

C22:5n-6 (DPA) C22:6n-3 (DHA)

C24:5n-3

C24:6n-3

Δ6-desaturase (+1 double bond)

elongase (+2 carbon atoms)

Δ5-desaturase (+1 double bond)

elongase (+2 carbon atoms)

Δ4-desaturase (+1 double bond)

elongase (+2 carbon atoms)

Δ6-desaturase (+1 double bond)

β-oxidation (-2 carbon atoms)

16

Water in oil (w/o) emulsion containing at least 80% fat

Margarines were originally developed in 1869

as an alternative to butter Made with vegetable oils - soybean, corn,

canola, and olive oils No cholesterol Lower SFA than butter Partial hydrogenation generates TFA

Types (Wassell and Young, 2007)

Table margarine Firmer industrial margarine Puff pastry margarine

17

Mean daily intake of TFA per person in the United States is 2-

4% of total food energy (Craig-Schmidt, 2006)

n-3 FAs intake is 0.1–0.2 g/person/day (Kris-Etherton et al., 2002)

Despite several expert committee guidelines, the American diet

is still low in n-3 FAs and high in TFA

18

n-3 FA TFA Chronic Diseases

Optimization of the reaction conditions and blending ratios of substrates for trans-free structured margarine fat synthesis

It is hypothesized that optimization of reaction conditions and blending ratios will result in a model that can be adapted in large scale synthesis of trans-free SL

Characterization of physical and chemical properties of

the SLs and comparison with physical blends It is hypothesized that enzymatically produced SLs will have better physical and

chemical properties than the physical blend and would be more suitable for margarine formulation

Characterization of textural and sensory properties of margarines prepared with selected SLs and comparison with commercial brands

It is hypothesized that the margarines prepared with trans-free SL will have similar or superior textural and sensory properties when compared to commercial brands

19

Characterization of substrates

Small-scale study - RSM

Gram-scale synthesis and characterization Comparison with physical blends

Large-scale synthesis and characterization Comparison with commercial margarine fat

Formulation of margarine Physical and sensory analysis Comparison with commercial brands

20

Cargill Inc.(Minneapolis, MN) Palm stearin (PS) Nonhydrogenated cottonseed oil (CO) Fully hydrogenated cottonseed oil (HCO)

Monsanto Co. (St. Louis, MO) High stearate soybean oil (HSSO) Stearidonic acid soybean oil (SDASO)

Novozymes North America Inc. (Franklinton, NC) Lipozyme® TLIM (Thermomyces lanuginosus lipase, sn-1,3 specific)

Novozym® 435 (Candida antarctica lipase, non-specific)

21

SDA C18:4n-3 Δ6-desaturase product of ALA Dietary SDA increases plasma EPA 3-4 times more efficiently than ALA

(James et al., 2003)

Increases the n-3 index (Harris et al., 2008)

SDASO Genetically modified, land-based sustainable source of n-3 FAs

22 (Hammond et al., 2008)

ALA SDA EPA Bioefficacy Stability

Δ 15

Regular soybean oil Low in SFA High in UFA Principal source of n-3 FA (ALA, ~7%) in the U.S. diet

High stearate soybeans are designed with elevated levels of

stearic acid Regular soybean oil: 4% 18:0 HSSO: 17% 18:0

Increased stability for many types of foods that require solid fat

functionality

23

In 2010-2011, the estimated total production of cottonseed oil in the US was 835 million pounds

Light golden color and bland flavor Crystallizes in the β' crystal form, the desirable

crystal type for margarines Major FAs are linoleic, palmitic, and oleic acids Requires less hydrogenation to achieve the same

degree of hardness compared to other linoleic oils (National Cottonseed Products Association) 24

Solid fraction obtained from fractionation of palm oil by crystallization at controlled temperatures

High proportion of SFAs and TAGs with a high m.p. of

48-50 °C Palmitic acid 49-68%, oleic acid 24-34%, stearic acid

~5% Useful source of fully natural hard fat component for

products such as shortenings, pastry and bakery margarines (American Palm Oil Council)

25

26

Milligram-scale RSM

Optimization

Independent Variables Time - 6, 10, 14, 18, 22 h Temperature - 50, 55, 60, 65 °C Substrate molar ratio - 2, 3, 4, 5 Immobilized enzyme - Lipozyme TLIM, Novozym 435

Dependent Variable Incorporation of stearic acid (mol%) SDA content (mol%)

Gram-scale (SL1, PB1, SL2, PB2) 1 L Stir-batch reactor Short-path distillation

GC-FID Fatty acid profile

Positional analysis

RPHPLC-ELSD TAG molecular

species

NMR Solid fat content

DSC Melting

Crystallization

XRD Polymorphism

Kilogram-scale SL

4 L Stir-batch reactor Short-path distillation

EF Extracted fat from

commercial margarine

GC-FID Fatty acid profile

Positional analysis

RPHPLC-ELSD

TAG mol species

NMR Solid fat content

DSC Melting

Crystallization

XRD Polymorphism

OSI Oxidative stability

NPHPLC-FLD

Tocopherols

Texture analyzer Texture

Rheometer Rheological properties

Polarized Light Microscope

Microstructure

Sensory evaluation Triangle test 27

Margarine Formulation

SLM (SL containing margarine)

RCM (Reformulated commercial margarine)

28

Part 1

High Stearate Soybean Oil (HSSO) and Stearidonic Acid Soybean Oil (SDASO)

29

Exp Temp1 (°C)

Time (h)

SR2 Enz3 Stearic acid inc4 (mol%)

SDA content (mol%)

N1 55 6 2 TLIM5 15.4±0.0 8.5±0.8 N2 65 10 2 TLIM 14.3±0.8 7.8±1.3 N3 50 18 2 TLIM 16.0±0.0 10.0±1.0 N4 60 22 2 TLIM 15.3±1.2 7.1±0.4 N5 50 6 3 TLIM 15.3±0.0 7.1±0.0 N6 65 6 3 TLIM 13.9±2.1 4.1±0.0 N7 55 22 3 TLIM 14.8±1.1 6.5±0.5 N8 50 10 4 TLIM 14.2±1.1 6.8±0.2 N9 65 22 4 TLIM 15.9±0.9 5.3±0.0 N10 55 6 5 TLIM 15.2±0.5 4.2±0.0 N11 60 10 5 TLIM 15.5±1.8 4.1±0.1 N12 65 18 5 TLIM 16.0±0.2 3.4±0.5 N13 50 22 5 TLIM 15.9±1.6 4.3±0.0 N14 50 6 2 N4356 15.6±1.3 4.1±0.8 N15 65 6 2 N435 13.5±1.8 5.5±1.2 N16 50 22 2 N435 14.3±1.1 2.2±0.0 N17 65 22 2 N435 14.5±0.0 7.6±1.0 N18 55 14 3 N435 14.6±1.8 3.9±0.0 N19 60 18 4 N435 16.1±2.0 5.2±0.2 N20 50 6 5 N435 15.3±3.1 2.7±0.2 N21 65 6 5 N435 15.1±1.9 2.2±0.8 N22 50 22 5 N435 16.0±2.1 2.1±0.0 N23 65 22 5 N435 17.9±3.4 2.1±0.7 N24 65 22 5 N435 17.7±1.7 1.9±0.0 N25 65 22 5 N435 18.1±3.8 2.1±0.0 N26 65 22 5 N435 17.9±2.6 2.2±0.0

1 Temp, temperature. 2SR, substrate molar ratio (HSSO:SDASO). 3Enz, enzyme. 4Inc, incorporation. 5TLIM, Lipozyme TLIM (Thermomyces lanuginosus lipase). 6N435, Novozym 435 (Candida antarctica lipase). Each value is the mean of triplicates ± standard deviation.

30

Sub

stra

te m

olar

ratio

S

ubst

rate

mol

ar ra

tio

Sub

stra

te m

olar

ratio

S

ubst

rate

mol

ar ra

tio

Temperature (°C) Temperature (°C)

Temperature (°C) Temperature (°C)

Lipozyme TLIM, 14 h

Novozym 435, 18 h

mol% stearic acid

mol% stearic acid

mol% stearidonic acid

mol% stearidonic acid

31

Stearic acid incorporation = 14.92 + 0.44Time + 0.51SR + 0.67 (SR*SR) + 0.44 (Temp*Time) + 0.43 (Temp*SR) + 0.44 (Time*SR) ± 0.32 (SR*Enz) R2=0.955; Q2=0.754 SDA content = 5.90 - 1.75SR ± 1.11Enz - 1.56 (Time*Time) + 0.59 (Temp*Time) - 0.68 (Temp*SR) ± 0.87 (Temp*Enz) R2=0.932; Q2= 0.558

50 °C 18 h 2:1

TLIM

SL1

50 °C 18 h 2:1

No enzyme

PB1

58 °C 14 h 2:1

N435

SL2

58 °C 14 h 2:1

No enzyme

PB2

32

Fatty acid SDASO1 HSSO2 SL13 PB14 SL25 PB26 14:0 0.1±0.0a 0.1±0.0a nd7 nd nd nd 16:0 12.2±0.1a 10.4±1.5b 15.1±2.9c 15.2±1.3c 15.3±1.3c 14.7±0.8d 16:1n-7 0.1±0.0a 0.1±0.0a 0.1±0.0a 0.1±0.0a 0.1±0.0a 0.1±0.0a 18:0 4.0±0.0a 15.8±0.9b 14.9±1.5cd 14.0±1.7c 15.9±1.2b 15.4±1.1d 18:1n-9c 15.6±0.1a 16.0±1.8a 25.6±3.6b 21.9±2.0c 22.6±2.0c 20.7±2.3c 18:2n-6t 0.2±0.0a 0.5±0.0b nd nd nd nd 18:2n-6c 24.6±2.3a 50.2±3.2b 19.7±2.1c 25.6±3.2ad 27.6±3.1d 33.2±3.8e 20:0 0.4±0.1a 1.1±0.0b 0.9±0.0bc 0.9±0.0bc 0.9±0.0bc 0.8±0.0c 18:3n-6 7.5±1.3a 0.4±0.0b 2.7±0.0c 2.4±0.9c 2.6±0.6c 2.4±0.0c 20:1n-9 0.5±0.0a 0.5±0.0a 0.6±0.0a 0.6±0.0a 0.5±0.0a 0.5±0.0a 18:3n-3 10.1±1.0a 4.3±1.6b 9.6±1.0c 7.3±1.0d 8.8±1.7e 7.5±0.8d 21:0 0.3±0.0a nd 0.1±0.0b 0.1±0.0b 0.1±0.0b nd 18:4n-3 23.5±1.9a nd 10.2±1.1b 7.1±1.1c 8.9±0.9d 7.1±1.0c 22:0 0.3±0.0a 0.6±0.0b 0.7±0.b 0.7±0.0b 0.6±0.0b 0.6±0.0b 20:3n-6 0.1±0.0 nd nd nd nd nd SFA8 17.3 28.0 31.7 30.8 32.8 31.5 UFA9 81.7 71.5 68.6 65.1 71.2 71.5 TFA10 0.2 0.5 0.0 0.0 0.0 0.0

1 SDASO, stearidonic acid-enriched soybean oil. 2HSSO, high stearate soybean oil. 3SL1, Lipozyme TLIM catalyzed structured lipid. 4PB1, corresponding physical blend of SL1. 5SL2, Novozym 435 catalyzed structured lipid. 6PB2, corresponding physical blend of SL2. 7nd, not determined. 8SFA, saturated fatty acids. 9UFA, unsaturated fatty acids. 10TFA, trans fatty acids. Each value is the mean of triplicates ± standard deviation. Values with the same letter in each row are not significantly different at P ≤ 0.05.

33

Fatty acid SDASO1

HSSO2 SL13 PB14 SL25 PB26

14:0 nd7 nd nd nd nd nd 16:0 nd nd 13.1±1.7a nd 15.7±1.0b nd 16:1n-7 nd nd nd nd nd nd 18:0 nd nd 7.9±1.0a nd 11.5±1.8b nd 18:1n-9c 15.5±1.3a 12.6±1.0b 14.8±1.0a 19.2±2.3c 15.1±1.7a 18.3±2.2c 18:2n-6t nd nd nd nd nd nd 18:2n-6c 50.3±4.9a 84.2±5.9b 45.3±4.9c 68.1±5.3d 42.5±3.7c 68.7±5.8d 20:0 nd nd nd nd nd nd 18:3n-6 8.0±1.0a nd 3.5±0.7b 1.9±0.0c 2.4±0.8d 0.9±0.0e 20:1n-9 nd nd nd nd nd nd 18:3n-3 5.9±0.9a 3.0±0.7b 5.2±1.2a 4.4±0.8c 5.2±0.6a 4.7±0.9c 21:0 nd nd nd nd nd nd 18:4n-3 19.8±2.1a nd 9.2±1.8b 5.3±0.4c 7.1±1.2d 5.6±1.0c 22:0 nd nd nd nd nd nd 20:3n-6 nd nd nd nd nd nd 1 SDASO, stearidonic acid-enriched soybean oil. 2HSSO, high stearate soybean oil. 3SL1, Lipozyme TLIM catalyzed structured lipid. 4PB1, corresponding physical blend of SL1. 5SL2, Novozym 435 catalyzed structured lipid. 6PB2, corresponding physical blend of SL2. 7nd, not determined. Each value is the mean of triplicates ± standard deviation. Values with the same letter are not significantly different at P ≤ 0.05.

34

min 5 10 15 20 25 30 35 40 45

mV

StO

St/S

tlLnG

St

LnLn

St

LnSt

LnLn

Ln

StLL

Ln

LnL

LLLn

OLL

n

LnSt

O

StO

O

LLL

OLL

LnG

S

LnLP

LnLn

S

PLL

PLO

min 5 10 15 20 25 30 35 40 45

mV

LLLn

PLL

OPO

OO

O

OSO

SLL

LLO

LnLn

S LL

L

SDASO

HSSO

35

min 0 5 10 15 20 25 30 35 40 45

mV

StSt

St

StLn

St

StLn

Ln

StO

St/L

nGSt

Ln

LnLn

LnLn

L Ln

LnO

/LLL

n St

OO

O

LLn

LLL

LnG

S Ln

LP

OLL

PL

L O

OL

SLL

POL

PPL O

OP

PPO

min 5 10 15 20 25 30 35 40 45

LnLn

Ln

LnLn

L

PPO

O

OP

PPL

POL

PLL

OLL

Ln

LP

LnG

S LL

L O

LLn

StO

O

LnLn

O/L

LLn

PSO

min 5 10 15 20 25 30 35 40 45

StLn

St

StLn

Ln

StO

St/L

nGSt

Ln

LnLn

Ln

GL Ln

LnL

LnLn

O/L

LLn

StO

O

OLL

n LL

L Ln

GS

LnLP

O

LL

PLL

SLL

POL

OO

P

min 5 10 15 20 25 30 35 40 45

LnLn

O/L

LLn

StO

O

LLL

LnG

S

OLL

PL

L

POL

OO

P PP

O

PSO

StSt

St

SL1

PB1

SL2

PB2

36

1

1,5

2

2,5

3

3,5

4

4,5

5

-55 -40 -25 -10 5 20 35 50 65 80

Nor

mal

ized

hea

t flo

w (W

/g)

End

o up

Temperature (°C)

SDASO

SL2 PB1 SL1

HSSO

PB2

14.4 °C

12.6 °C 13.2 °C

1.8 °C

11.2 °C

13.1 °C

0,00

1,00

2,00

3,00

4,00

5,00

6,00

7,00

-55 -40 -25 -10 5 20 35 50 65 80

Nor

mal

ized

hea

t flo

w (W

/g)

Exo

dow

n

Temperature (°C)

SDASO

PB2 SL2 PB1

SL1

HSSO

-7.2 °C

37.1 °C 32.5 °C

28.6 °C 32.6 °C

26.5 °C

37

0

2

4

6

8

10

0 10 20 30 40 50 60

Solid

fat c

onte

nt (

%)

Temperature (°C)

SL1 PB1

SL2 PB2

Sample Polymorphic form

SDASO β′>β

HSSO β′>β

SL1 β′>>>β

PB1 β′>β

SL2 β′>>β

PB2 β′>β

Solid Fat Content SL Polymorphic Form

Based on these results SL1 was selected for margarine production

Fatty acid Total sn-2 sn-1,3 SLa EFb SL EF SL EF

14:0 nd 1.0±1.1 ndc nd nd 1.5±1.1 16:0 14.1±1.1a 24.8±1.9b 12.8±1.3a 23.1±2.0b 14.8±1.3a 25.7±2.3b 16:1n-7 0.1±0.0a 0.1±0.0a nd nd 0.2±0.0a 0.2±0.0a 18:0 15.1±1.2a 4.8±1.0b 8.2±0.9a 4.6±1.1b 18.6±1.9a 4.9±1.2b 18:1n-9c 25.8±2.3a 42.1±3.2b 15.0±1.6a 28.2±2.2b 31.2±2.5a 49.1±4.1b 18:2n-6c 18.9±1.4a 20.3±2.0b 45.8±3.8a 41.3±3.4b 5.5±1.7a 9.8±0.8b 20:0 0.9±0.0a 0.3±0.0b nd nd 1.4±1.0a 0.5±0.0b 18:3n-6 2.6±1.1a 1.8±0.6b 3.8±0.9 nd 2.0±1.2a 2.7±1.0b 20:1n-9 0.6±0.0a 0.5±0.0a nd nd 0.9±0.0a 0.8±0.0a 18:3n-3 9.9±0.6a 3.8±0.6b 5.5±1.2a 2.8±1.0b 12.1±1.7a 4.3±1.4b 21:0 0.1±0.0a 0.1±0.0a nd nd 0.2±0.0a 0.2±0.0a 18:4n-3 10.5±1.0 nd 9.8±1.6 nd 10.9±1.3 nd 22:0 0.7±0.0a 0.4±0.0b nd nd 1.1±0.1a 0.6±0.0b 20:3n-3 nd 0.1±0.0 nd nd nd 0.2±0.0 20:5n-3 nd 0.2±0.0 nd nd nd 0.3±0.0 n-6/n-3 1.1 5.8 3.2 14.8 0.3 2.9

aSL, large scale Lipozyme TLIM catalyzed structured lipid. bEF, extracted fat from commercial margarine. cnd, not determined. Each value is the mean of triplicates ± standard deviation. Values with the same letter in each row within total, sn-2, and sn-1,3 columns separately are not significantly different at P < 0.05. 38

39

TAG species SLa EFb StStSt 1.1±0.4 ndc StLnSt 2.6±0.9 nd StLnLn 2.4±0.5 nd StOSt/LnGSt 2.6±0.2 nd LnLnLn 1.7±0.3a 0.4±0.0b LnLnL 7.7±1.1a 1.4±0.1b LnLnO/LLLn 5.0±1.0a 3.8±0.9b StOO 3.9±1.2 nd LnOO nd 6.2±1.3 OLLn 2.5±0.7 nd LLL 12.4±1.9a 7.4±1.1b LnGS 2.6±0.7a 1.2±0.0b LnLP 3.4±0.9a 5.6±1.2b OLL 10.7±1.6a 7.1±1.4b PLL 12.2±1.7a 4.4±0.5b OOL 1.1±0.0a 4.6±0.8b SLL 1.6±0.1a 5.2±0.4b POL 14.4±1.3a 8.0±0.9b PPL 2.4±0.0a 2.3±0.0a OOO nd 10.3±1.3 POO nd 14.2±1.2 POP nd 12.7±1.0 PPP nd 1.8±0.0 SOL 5.5±0.7 nd PSL 2.4±1.2 nd SOO 0.1±0.0a 2.3±0.3b SOS 1.6±0.1a 1.5±0.4a

aSL, large scale Lipozyme TLIM catalyzed structured lipid. bEF, extracted fat from commercial margarine. cnd, not determined. P is palmitic, S is stearic, O is oleic, L is linoleic, Ln is linolenic, G is γ-linolenic, St is stearidonic acid. Each value is the mean of triplicates ± standard deviation. Values with the same letter in each row are not significantly different at P < 0.05.

40

Tocopherol content OSI at

110 °C α-Ta α-T3b β-T γ-T γ-T3 δ-T δ-T3

SDASOc 82.6±4.9 ndd 1.1±0.9 867.4±8.3 nd 286.9±7.3 nd 6.5±1.1

HSSOe 74.8±3.1 nd nd 755.8±5.2 nd 194.7±9.4 nd 44.2±2.3

Before SPDf 232.2±12.6 nd 1.1±0.4 2379.3±11.7 nd 676.3±9.8 nd -

SLg

(after SPD) 98.3±9.1 nd 0.4±0.0 1126.3±10.8 nd 286.4±9.7 nd 13.8±1.3

EFh 130.1±6.8 82.6±3.3 8.3±1.2 931.6±8.7 106.8±7.4 346.4±8.8 26.4±3.2 15.4±1.0

Each value is the mean of triplicates ± standard deviation aT, tocopherols. bT3, tocotrienols. c SDASO, stearidonic acid soybean oil. dnd, not detected. eHSSO, high stearate soybean oil. fSPD, short-path distillation. gSL, large scale Lipozyme TLIM catalyzed structured lipid. hEF, extracted fat from commercial margarine

41

0,00

1,00

2,00

3,00

4,00

5,00

6,00

-55 -10 35 80

Nor

mal

ized

hea

t flo

w (W

/g)

endo

dow

n

Temperature (°C)

11.3 °C

10.7 °C

SL EF

0,00

0,50

1,00

1,50

2,00

2,50

3,00

3,50

4,00

4,50

-55 -10 35 80

Nor

mal

ized

hea

t flo

w (W

/g)

exo

up

Temperature (°C)

10.9 °C

9.1 °C

Melting Thermograms Crystallization Thermograms

Solid Fat Content

SL EF

SL EF

Sample Polymorphic form

SL β′>>>β

EF β′>>>β

Polymorphic Form

0

5

10

15

20

25

0 20 40 60 80

Sol

id fa

t con

tent

(%)

Temperature (°C)

EF

SL

42

0

400

800

1200

1600

Hardness (g) Adhesiveness (g-s) Cohesiveness (10^-3)

SLM RCM

b

a

a a

a

a

Texture profile analysis of margarine formulated with structured lipid (SLM) and reformulated commercial margarine (RCM). Each value is the mean of triplicates ± standard deviation. Columns with the same letter within each texture attribute are not significantly different at P < 0.05.

SLM RCM

Morphology of fat crystals of margarine formulated with structured lipid (SLM) and reformulated commercial margarine (RCM).

43

0,0 2,0 4,0 6,0 8,0

10,0 12,0

0,0 0,5 1,0 1,5 2,0 2,5

Visc

osity

(kP

a-s)

Shear stress (kPa)

SLM RCM

0 5

10 15 20 25 30 35

0 300 600

Stra

in (%

)

Time (s)

SLM RCM

0 5

10 15 20 25 30 35 40 45

0 5 10

G' G

'' (kP

a)

Frequency (Hz)

SLM-G' SLM-G" RCM-G' RCM-G"

Difference Test (Triangle Test)

No significant difference (P > 0.05) was observed between SLM and RCM

Stress viscometry

Creep analysis

Dynamic analysis

44

0

20

40

60

80

100

0

4

8

12

16

20

1 2 3 4 5 6 7 8 9 10

Yiel

d (%

)

Mol

%

Run number

Total stearic acid

Total SDA

sn-2 Stearic acid

sn-2 SDA

Yield %

Mol% stearic acid incorporation and SDA content (primary y-axis) and yield% (secondary y-axis) of SLs as determining factor of Lipozyme TLIM enzyme reusability

SL1 (50 °C, 18 h, 2:1, TLIM) suitable for soft/liquid margarine 14.9 mol% stearic acid; 10.2 mol% stearidonic acid; no TFA Diverse FAs and TAG species Desirable SFC (<1% at RT) and polymorph (β′) Melting completion temperature 11.2 °C

SL and EF had similar SFA (~31 mol%) and UFA (~68 mol%), but SL had much lower n-6/n-3 ratio (1.1:1) than EF (5.8:1)

SL and EF had similar melting profile (SL10.7 °C and EF 11.3 °C) and β' polymorph (desirable)

SLM was softer and more spreadable than RCM No sensory difference was observed between the two margarines SL contains 10.5 mol% SDA, so SLM (consisting of 80% SL) will have 1.1 g

SDA /serving (13 g). Relative to EPA, conversion efficiency of SDA to EPA is 17-30%, therefore 1.1 g SDA will result in 0.2-0.3 g EPA

Reusability: Yield (~90.6%) unchanged till the ninth run. After the tenth run the yield % decreased to 88.8%

No change in total stearic acid (~15.0 mol%) and total SDA content (~10.2 mol%) After the seventh run sn-2 positional SDA decreased and stearic acid increased

45

46

SL1/PB1 50 °C 20 h 2:1

Novozym 435

SL2/PB2 57 °C 6.5 h 2:1

Lipozyme TLIM

Part 2

Palm Stearin (PS) and High Stearate Soybean Oil (HSSO)

47

fatty acid HSSOa PSb SL1c PB1d SL2e PB2f SLg EFh

C14:0 0.1±0.0 1.4±0.9 0.7±0.0 0.8±0.0 0.9±0.0 0.8±0.3 0.4±0.0 0.1±0.0

C16:0 10.1±1.5 58.1±4.9 31.0±3.3 38.1±3.8 38.5±4.2 39.3±3.3 30.6±3.3 10.2±1.5

C18:0 16.8±0.9 5.2±1.1 12.3±1.8 8.3±1.2 9.5±0.6 8.6±1.0 12.8±1.7 6.1±0.7

C18:1t ndi nd nd nd nd nd nd 18.7±1.9

C18:1n-9c 16.0±1.8 26.9±2.7 20.1±2.1 21.2±3.0 25.5±2.9 22.6±3.2 21.2±2.0 31.6±3.2

C18:2n-6t 0.6±0.0 0.4±0.1 nd nd nd nd nd 0.4±0.0

C18:2n-6c 50.2±3.2 7.6±1.1 34.6±2.9 29.4±3.2 23.8±1.8 26.7±3.2 35.1±2.5 29.4±1.9

C18:3n-3c 4.3±1.6 0.2±0.0 2.1±0.7 1.3±0.1 1.5±0.1 1.3±0.0 2.0±0.2 3.2±0.2

∑SFA 28.5 65.3 44.1 47.6 49.6 48.9 42.6 16.9

∑UFA 71.5 34.4 57.2 52.2 51.2 50.9 58.3 64.3

∑TFA 0.6 0.4 0 0 0 0 0 19.1

aHSSO, high stearate soybean oil. bPS, palm stearin. cSL1, Novozym 435 catalyzed structured lipid. dPB1, corresponding physical blend of SL1. eSL2, Lipozyme TLIM catalyzed structured lipid.fPB2, corresponding physical blend of SL2. gSL, large scale Novozym 435 (SL1) catalyzed structured lipid. hEF, extracted fat from commercial margarine ind, not determined. Each value is the mean of triplicates ± standard deviation

SL2 PB2

HSSO

PS

SL1 PB1

SL2 PB2

SO

O

PS

O

PS

O

SO

O

48

49

0

10

20

30

40

50

60

0 20 40 60

Sol

id fa

t con

tent

(%)

Temperature (°C)

SL1 PB1 SL2 PB2 SL EF

HSSO PS SL1 PB1 SL2 PB2 SL EF

Tmc (°c) 14.4 55.6 45.4 55.1 47.2 55.1 45.6 38.6

Tco (°c) 33.1 29.6 30.7 29.2 30.3 29.2 30.1 17.1

Solid Fat Content

Thermal Behavior

50

0

1000

2000

3000

4000

5000

6000

Hardness (g) Adhesiveness (gs)

Cohesiveness (10^-3)

SLM RCM b

a a a a

a

0

10

20

30

40

50

0 200 400 600

Stra

in %

Time (s)

SLM RCM

0

100

200

300

-0,5 0,5 1,5 2,5

Visc

osity

(kPa

-s)

Shear stress (kPa)

SLM

RCM

0

20

40

0 5 10

G’G

’’ (k

Pa)

Frequency (Hz)

SLM-G' SLM-G" RCM-G' RCM-G"

Creep Analysis

Stress Viscometry

Dynamic Analysis

Difference Test (Triangle Test)

Significant difference (P<0.05) was observed between SLM and RCM

Texture Profile Analysis

Both SLs may be suitable for margarine formulation Novozym 435 synthesized SL (SL1) more suitable as

margarine fat stock because of its desirable FA content, melting temperature, and dominant β' polymorph

SL had higher SFA (42.6 mol%) than EF (16.9 mol%) SL had no TFA whereas EF had 19.1 mol% TFA SLM harder and less spreadable than RCM Significant sensory difference between SLM and RCM

SL suitable for hard margarine

51

52

SL1/PB1 56 °C 6 h 4:1

Novozym 435

SL2/PB2 57 °C 14 h 4:1

Lipozyme TLIM

Part 3

Palm Stearin (PS) and Cottonseed oil (CO)

53

Fatty acid (mol%) C14:0 C16:0 C16:1c C18:0 C18:1t C18:1c C18:2n6t C18:2n6c C18:3n3

SLa

total 1.1±0.1a 46.1±3.3a 0.2±0.0a 5.5±0.6a ndb 24.3±2.1a nd 23.1±1.3a 0.2±0.0a

sn-2 nd 38.2±3.8a nd 3.3±0.5a nd 37.5±3.6a nd 21.4±1.7a nd

EFc

total 0.1±0.0b 10.2±1.5b 0.1±0.0a 6.1±0.7a 18.7±1.9 31.6±3.2b 0.4±0.0 29.4±1.9b 3.2±0.2b

sn-2 nd 6.5± 1.4b nd 5.6±1.5b 15.6±2.7 32.1±3.6b nd 40.3±4.8b nd

Each value is the mean of triplicates ± standard deviation. Values with the same letter in each column within total and sn-2 rows separately are not significantly different at P < 0.05.; aSL, large scale SL1, bnd, not detected cEF extracted fat from commercial margarine

54

COa PSb SL1c PB1d SL2e PB2f

LnOLn ndg nd 1.1±0.9 nd 0.1±0.0 nd LLLn nd nd 1.4±1.0 nd 1.7±0.0 nd LLL 22.7±3.1 nd nd nd nd nd LOL 11.8±1.6 nd nd nd nd nd PLL 36.5±3.7 nd 2.0±0.8 1.8±0.9 2.2±0.0 1.5±0.1 SLL nd nd 0.7±0.0 nd 0.8±0.0 nd OOL 1.7±0.9 nd nd nd nd nd POL 17.6±2.1 2.0±0.6 9.0±1.0 7.3±0.9 15.5±1.1 6.4±0.2 PLP 9.6±1.1 6.5±0.3 11.2±1.1 10.2±1.1 18.0±1.2 9.9±0.7 OOO nd 1.8±0.9 1.5±0.2 0.2±0.0 1.8±0.7 1.3±0.2 POO nd 14.8±1.2 15.5±0.9 14.7±1.0 13.0±1.0 12.8±1.7 POP nd 47.3±3.2 36.8±2.7 49.2±3.5 39.4±2.3 46.8±4.2 PPP nd 26.9±2.9 16.0±2.1 19.7±1.3 15.4±0.8 20.4±2.7 PSO nd nd nd nd nd nd SOO nd nd nd nd nd nd

Each value is the mean of triplicates ± standard deviation. Ln, linolenic, O, oleic, L, linoleic, P, palmitic, S, stearic. aCO, cottonseed oil; bPS, palm stearin; cSL1, Novozym 435 catalyzed structured lipid; dPB1, corresponding physical blend of SL1; eSL2, Lipozyme TLIM catalyzed structured lipid; fPB2, corresponding physical blend of SL2; gnd, not detected

55

0 0,5

1 1,5

2 2,5

3 3,5

4 4,5

5

-55 -10 35 80

Nor

mal

ized

end

othe

rmic

hea

t flo

w

[W/g

]

Temperature [°C]

CO

PB2

SL2

PB1

SL1 PS

6.2 °C -42.4 °C

54.2 °C -8.7 °C

40.5 °C -5.8 °C

51.6 °C -8.7 °C

-6.7 °C 46.2 °C

-8.2 °C 54.8 °C

-1

0

1

2

3

4

5

6

7

8

-55 -10 35 80

Nor

mal

ized

exo

ther

mic

hea

t flo

w

[W/g

]

Temperature [°C]

PB2

SL2

PB1

SL1

PS

CO

31.5 °C

25.8 °C

26.0 °C

31.3 °C

30.5 °C

-5.8 °C -41.8 °C

-12.8°C

-9.9 °C

-9.6 °C

-10.4 °C

-9.6 °C

0,00

1,00

2,00

3,00

4,00

5,00

6,00

7,00

8,00

-55 -10 35 80

Nor

mal

ized

hea

t flo

w [W

/g]

exo

up

Temperature [°C]

SL-melting

SL-cooling

EF-melting

EF-cooling

41.1 °C -4.4 °C

38.6 °C -32.7 °C

31.3 °C -9.8 °C

17.1 °C -35.8 °C

Melting

Cooling

0

10

20

30

40

50

60

70

0 20 40 60

Sol

id fa

t con

tent

[%]

Temperature [°C]

EF

SL

SL1

PB1

SL2

PB2

Solid Fat Content

SFC at 25 °C were lower for SLs (24.8-30.8%) than the corresponding PBs (34.7-39.3%)

Novozym 435-catalyzed SL product had desirable FA profile, physical properties and β′ polymorph

Lipozyme TLIM-catalyzed SL contained 53.6 mol% PA at sn-2 position - suitable for possible use in human milk fat substitutes

Compared to EF (19.1 mol% TFA), the SL had no TFA SLM was harder and less spreadable than RCM No sensory difference was observed SL suitable for possible use as hard/industrial margarine

with high oxidative stability and no TFA

56

57

SL2/PB2 65 °C 6 h 2:1

Novozym 435

SL1/PB1 65 °C 16.5 h

2:1 Lipozyme TLIM

Part 4

Palm Stearin (PS) and Hydrogenated Cottonseed Oil (HCO)

58

Fatty acid C16:0 C18:0 C18:1t C18:1c C18:2n6t C18:2n6c C18:3n3

HCOa 3.2±1.8a 84.7±0.8a ndb 2.2±1.0a nd 12.6±3.4a nd

PSc 58.1±4.9b 5.2±1.1b nd 26.9±2.7b 0.4±0.1a 7.6±1.1b 0.2±0.0a

SL1d 40.7±3.9c 23.3±1.0c nd 16.2±2.1c nd 18.2±1.9c 0.1±0.0a

PB1e 44.8±3.2d 25.2±0.6d nd 15.5±1.7d nd 15.8±2.4d 0.1±0.0a

SL2f 42.4±4.1e 23.1±1.1c nd 16.7±2.5c nd 17.0±3.8e 0.1±0.0a

PB2g 42.8±2.7e 24.6±1.3d nd 17.8±2.8e nd 14.1±2.1f 0.1±0.0a

SLh 40.1±3.3c 23.5±0.6c nd 16.3±2.1c nd 18.1±1.3c 0.2±0.0a

EFi 10.2±1.5f 6.1±0.7e 18.7±1.9 31.6±3.2f 0.4±0.0a 29.4±1.9g 3.2±0.2b

Each value is the mean of triplicates ± standard deviation. Values with the same letter in each column not significantly different at P < 0.05.aHCO, fully hydrogenated cottonseed oil; bnd, not detected; cPS, palm stearin; dSL1, Lipozyme TLIM catalyzed structured lipid; ePB1, corresponding physical blend of SL1; fSL2, Novozym 435 catalyzed structured lipid; gPB2, corresponding physical blend of SL2; hSL, large scale SL1; iEF, extracted fat from commercial margarine

59

HCO PS SL1 PB1 SL2 PB2 SL EF

Tmc (°C) 63.8 54.2 50.9 56.3 52.1 56.2 50.1 38.6

Tco (°C) 47.8 31.5 38.1 38.3 38.1 38.2 38.2 17.1

0

20

40

60

80

100

10 20 30 40 50 60

Sol

id fa

t con

tent

(%)

Temperature (°C)

EF SL SL1 PB1 SL2 PB2

Thermal Behavior

Solid Fat Content

Lipozyme TLIM-catalyzed SL had high melting temperature (50.1 °C), β'-polymorph, and high SFC

Compared to EF (19.1 mol% TFA), the SL had no

trans fat SLM was harder and less spreadable than RCM Significant difference was observed between SLM

and RCM in triangle test SL suitable as hard margarine fat stock-puff pastry

60

All the models developed using response surface methodology had high predictive power

and were used in large-scale syntheses of SLs For the first combination, SL was synthesized at 50 °C, 18 h, 2:1 (HSSO:SDASO) using

Lipozyme TLIM, containing 15.1 mol% stearic acid and 10.5 mol% SDA The margarine formulated with this SL was trans-free, SDA-enriched with desirable taste and texture

for a soft/liquid margarine

For the second combination, desirable SL containing 11.2 mol% stearic acid and no TFA was obtained at 50 °C, 20 h, 2:1 (PS:HSSO) using Novozym 435 lipase This SL was more suitable for stick/hard margarine because of its higher melting completion

temperature (45.4 °C) and SFC

For the third combination, SL was synthesized at 56 °C, 6 h, 4:1 (PS:CO) catalyzed by Novozym 435 The margarine formulated with this SL had high oxidative stability and no TFA and may be suitable

as hard/industrial margarine

In the last combination, SL was synthesized at 65 °C, 16.5 h, 2:1 (PS:HCO) using Lipozyme TLIM The margarine formulated with this SL was trans-free hard margarine that may be used for puff

pastries or baking purposes

This research resulted in the production of trans-free SLs as alternatives to partially hydrogenated fat that can be used to formulate trans-free foods

61

62

Agriculture and Food Research Initiative Competitive Grant no. 2009-65503-05734 from the USDA National Institute of Food and Agriculture

Monsanto for providing the substrates Prof. Alejandro G. Marangoni (University of Guelph) for

his assistance with the solid fat content analysis Dr. Garima Pande – Performed the experiments as PhD

student in my lab

63