Nutritional value of soybean meal: Influence of origin and opportunities for improvement - G.G. Mateos at DSM european poultry symposium

Nutritional value of soybean meal:

Influence of origin and opportunities

for improvement

DSM European Poultry Symposium. Istanbul, November 2013

Gonzalo G. MateosDepartment of Animal Science, UP Madrid

High protein SBM, 47% CPEnergy content, poultry

Institution Year CP (%) AMEn (Mcal/kg)

NRC 1994 48.5 2.44

INRA 2002 47.2 2.32

Japan 2009 47.1 2.47

Fedna 2010 47.5 2.38

CVB 2011 47.5 2.22

Brazil 2011 48.1 2.30

NRC(1994)

INRA(2002)

Japan(2009)

Fedna(2010)

Braz(2011)

CVB(2011)

MOI 10.0 12.4 11.0 12.0 10.8 12.6

CP 48.5 47.2 47.0 47.0 48.1 47.5

EE 1.0 1.5 1.6 1.9 1.5 1.8

Ash 6.3 6.3 6.4 6.2 5.7 6.5

NDF 8.5 8.9 11.1 9.1 14.9 8.2

74.3 76.3 77.1 76.2 81.0 76.6

What is missing?

SBM, 47% CPProximal analyses, %

AMEn content of SBM1

European tables

AMEn = 37.5 x CP + 46.4 x EE + 14.9 x NFE

Digestibility of the CP fraction?

83-93%

How ether extract is determined?

1.0 vs. 2.2% EE content (HCl hydrolisis?)

What is the composition of the NFE fraction?

Sucrose (4 to 9%) vs. pectins vs. lignin1 Jansen, 1989 (kcal/kg)

Objectives

To test the effects of origin of the beans on chemical characteristics and nutritive value of commercial SBM

To verify the need of using different matrixes of chemical composition andnutritive value for SBM according to theorigin of the beans

Samples collected from 2007 to 2013Country of origin (63%) and Europeanports (37%)

Analysis conducted in duplicateProximal, AA, sugars, and mineralsAmino acids (NIR)TIA, KOH, PDI, urease, HDI (AminoRed)

Same lab, same technician

n= 491SBM survey, SpainLaboratory studies

n = 491

EU survey, 2007-2013Origin of the samples

SBM surveyChemical composition1, 88% DM

n CP CF NDF EE

ARG 158 45.5c 4.7b 9.1b 1.7ab

BRA1 148 46.7b 5.5a 10.5a 1.8a

USA2 185 47.2ª 3.8c 7.8c 1.6b

SEM 0.13 0.08 0.14 0.04

P *** *** *** **

1SBM, 48% CP and profat2Mississippi river and East Cost procedure

n Ash P Fe Sucrose Oligo.1

ARG 128 6.6ª 0.66b 109b 6.7b 6.1b

BRA 136 6.3b 0.62c 168a 5.7c 6.0b

USA 168 6.6ª 0.69a 113b 7.3ª 6.6a

SEM 0.04 0.005 4.3 0.08 0.05

P *** *** *** *** ***

SBM surveyMineral and sugars, 88% DM

1Stachyose+raffinose. Brazil higher raffinose content

n = 451

SBM surveyAmino acid profile, % CP

n Lys TSAA Thr Trp

ARG

BRA

USA

SEM

P

141

144

183

6.10b

6.05c

6.16a

0.005

***

2.87ª

2.81b

2.87a

0.005

***

3.93a

3.88b

3.91a

0.002

***

1.37a

1.34c

1.36b

0.001

***

SBM, all origins (n=403)LYS:CP ratio

2,4

2,6

2,8

3,0

3,2

42 44 46 48 50 52CP (%)

LYS (%)

ARG BRA USA

y = 0.062x - 0.04R2 = 0.90 P<0.001

SBM by originCP vs. Lys content

y = 0.059x + 0.105R²= 0.92 P<0.001

2,4

2,6

2,8

3,0

42 44 46 48 50CP (%)

LYS (%)

RSD = ± 0.026CV (%)= 0.91Bias = + 0.0006

ARG (n=114)

y = 0.066x - 0.249R² = 0.89 P<0.001

2,4

2,6

2,8

3,0

3,2

42 44 46 48 50 52CP (%)

LYS (%)

RSD = ± 0.029CV = 1.00%Bias= + 0.0009

y = 0.054x + 0.375R²= 0.93 P<0.001

2,6

2,8

3,0

3,2

44 46 48 50 52CP (%)

LYS (%)

RSD = ±0.023CV = 0.80%Bias = + 0.0007

BRA (n=124)

USA (n=159)

BRA USA

5,8

6,0

6,2

6,4

42 44 46 48 50 52CP (%)

LYS (% CP)

NS ; P= 0.54

5,8

6,0

6,2

6,4

42 44 46 48 50 52CP (%)

LYS (%CP)

r= + 0.21 ; P < 0.05

5,8

6,0

6,2

6,4

44 46 48 50 52CP (%)

LYS (%CP)

r= 0.47 ; P< 0.001

n=397

n=124 n=159

SBM by origin vs. Lys profile

ARGBRAUSA

KOH solubility, %: 75-85PD index, %: 15-30Urease activity, mg N/g: 0.00-0.10TI activity, mg/g: < 2.2-3.5

“Poor agreement between Research Institutions and Industry for suitable

values for protein quality of SBM”

SBM protein qualityIndustry standards

TIA PDI KOH HDI

n (mg/g) (%) (%) AminoRed2

ARG 135 2.5b 16.3b 81.5c 12.5b

BRA 135 2.6b 15.3c 82.6b 15.6a

USA 170 3.1a 19.5a 86.4a 9.0c

SEM 0.05 0.34 0.33 0.38

P *** *** *** ***

SBM protein qualitySurvey

1 Urease < 0.03 mg N/g for all origins (P < 0.01)2 Evonik-Degussa

ItemStorage (d) SEM

(n = 8)0 30 60 90 120

Ureaseactivity1 0.00 0.00 0.00 0.00 0.00 0.00

PDI2 (%) 21.9a 21.5a 20.1b 18.8c 17.7d 0.21

KOH solubi.(%) 83.8 83.6 83.2 83.8 83.8 0.47

TIA3 3.0 3.1 3.0 3.0 3.0 0.072

1mg N/g x min2Linear effect, P < 0.0013mg/g SBM

JAPR, 2013

SBM qualityLength of storage

SBM surveyAminoRed (HDI1)

Year ARG BRA USA

Evonik2 2007200820092010201120122013

14.613.111.913.313.611.812.9

14.312.613.913.715.515.411.6

6.99.77.39.69.53.14.5

UP Madrid n 9612.7b

11315.5a

1509.0c

1Heat Damage Indicator2n > 2,500 to 2011 and > 4,800 in 20133n = 359 (P<0.001), SEM = 0.42 (2007 a 2013)

Soybeans storage Argentina

Soybeans storage Argentina

* Adapted from FEDNA (2010) for SBM, 47%CP

76

78

80

82

84

86

%CP (88% DM)

ARG

USABRA

Anim. FST, 2012

SBMSID (%) of Cys

SBM, all origins1

SID of CP vs. quality indexes

Variable r P

Crude proteinReactive lysKOH solubilityTI activity

+ 0.51+ 0.56+ 0.70+ 0.54

0.050.01

0.0010.01

1 Broiler trial (n = 22 SBM of 3 origins) Frikha et al., 2012

Soybean meal colorHeating and processing

Particle size

Country of origin

n LightnessL*

Rednessa*

Yellownessb*

ARG 87 67.2b 6.5b 25.2b

BRA 95 65.0c 7.6a 24.6c

USA 148 70.7a 6.0c 26.3a

SEM 0.28 0.10 0.14

P *** *** ***

1 Minolta (ground at 1mm)

Soybean meal colour1

All origins1,2 PDI KOH UA TIA

Lightness L* 63*** 52*** 14* 56***

Redness a* - 67*** - 50*** - 23*** -56***

Yellowness b* 12*** 31*** - 18** NS

1 n = 3472 Ground at 1 mm

SBM correlations (r, P <)Protein quality vs. color variables

ARG BRA USA All

n 89 102 156 347

L* 58*** 27** 55*** 56***

a* - 49*** - 26** - 65*** - 56***

b* ns ns - 17* ns

1 Ground at 1 mm

SBM1

TIA vs. color variables (r, P<)

n GMD1 SD

ARG 91 1,363b 472

BRA 92 1,522a 474

USA 85 856c 168

SEM 50.4

P ***1GMD: Geometric mean diameter. 7 Sieves ( m) : 2500, 1250, 660, 330, 80, 40 and residue

Average particle sizeSBM origin

Particle size distribution in SBM byorigin1

0

10

20

30

40

50

2500 1250 630 315 160 80 40 < 40

Perc

enta

geof

tota

l in

each

siev

e

Sieve size(µm)

ARGBRAUSA

1n=268

324 samplesNIRS 5000 (1,100-2,500 nm)

As received

Validation set 1/3 of samples

Training set 2/3 of samples origins

n = 255 n = 69

Soybean mealQualitative analysis (NIRS)

Origin n ARG BRA USA

ARG 20 19 0 1BRA 13 0 13 0USA 36 2 0 34

Soybean meal validation set (NIRS)

Correctly classified = 96%

ARG

BRA USA

12007-2011

Validation SBM surveyMain findings (88% DM)

Argentina Brazil USA

5 y1 12/13 5 y1 12/13 5 y1 12/13

nCPNDFSucroseTotal PFeLys (% CP)

13045.49.46.7

0.6511260.09

2245.97.77.3

0.761166.14

11246.510.55.8

0.6119360.05

2747.110.25.6

0.631716.04

14847.57.87.2

0.6911760.15

2646.08.08.2

0.681016.14

12007-20112Kakade, method (88% DM)

Validation SBM surveyProtein quality

Argentina Brazil USA

5 y1 2012 5 y1 2012 5 y1 2012

n

PDI (%)

KOH sol. (%)

HDI (Evonik)

TIA (mg/g)

112

17.1

82.5

13.2

2.5

17

13.2

75.0

8.9

2.3

109

15.2

83.6

16.2

2.5

27

15.8

77.7

12.7

2.7

148

19.9

87.3

9.7

3.2

20

18.1

82.0

3.9

2.7

SID (%) of SBM in broilers New Zealand, Exp. 1

USA(n = 3)

ARG(n = 4)

BRA(n = 3)

IND(n = 4)

LysMetCysThrAvge AAAMEn1

91.0a

89.5ab

76.3a

83.3a

87.02,146ª

90.9a

89.7a

71.4ab

83.4a

86.62,112ª

89.9a

89.9a

70.6ab

83.1a

86.12,128ª

87.4b

87.2b

62.9b

79.8b

83.21,861b

1Kcal/kg, as fed Ravindran and Bootwalla, 2011

SID (%) of SBM in broilersNew Zealand, Exp. 2

USA(n = 7)

ARG(n = 6)

BRA(n = 3)

IND(n = 4)

CPLysCysMetThrAvge AA

85.787.874.688.582.585.6

84.786.869.887.681.184.3

83.285.167.985.678.482.5

78.280.157.382.873.877.9

Ravindran, 2013

SID (%) of Lys of SBM (range)Broilers (New Zealand, Exp. 2)

USA(n = 7)

ARG(n = 6)

BRA(n = 3)

IND(n = 4)

Total Lys (% CP)

SID Lys (%)

React. Lys (%)

4.946.61

84.490.6

96.4

5.306.39

82.088.5

90.5

5.525.77

82.788.2

93.5

4.796.39

76.383.3

89.7

Ravindran, 2013

HP-SBM had 100 kcal more AMEn than R-SBMDetermined AA content

Same availability for all AAThree ratios Lys:ME for each SBM

Low: 380Medium: 415High: 450

Experimental dietsAssumptions

De Coca et al., 2010

SBM qualityBroiler performance

R-SBM HP-SBMCP, %Total Lys, %Sucrose, %NDF, %TIA, mg/kgKOH sol., %AMEn

46.32.755.8

10.24.8

80.72,200

48.63.046.77.01.8

84.32,300


Broiler performance, 1-10 dMain effects1

ADG (g) FCR (g/g)

SBM typeR-SBMHP-SBMSEM (n = 30)

Lys:ME ratioLowMediumHighSEM (n = 20)

22.3b

23.5a

0.24

21.3c

23.0b

24.3a

0.30

1.225a

1.168b

0.005

1.275a

1.181b

1.133a

0.006


1 Interaction (P < 0.01): Higher positive response of the use of HP-SBM at the low CP level of the diet

Processing conditionsTIA vs. Maillard reactions

Origin of the beansGrowing, storage conditions

Gastrointestinal conditions in the birdGizzard function and pHPepsine activity

Use of enzymesProteases

Improvement in digestibilitySoybean meal

OriginBacteria, fungi (fermentation)Animal and plants (extraction)

pH rangeAcid, neutral, alkaline

Bond specificityEndo- and exopeptidasesSpecific (AA) protease

Classification of proteases

Angel, 2012

ExamplesTrypsin and chymotrhypsinElastaseEnteroquinase

PROACTBacterialAlkalineSerine protease

Bond specifitySerine proteases

Protease and fullfat soybeansAID of amino acids, %

Control Protease200 mg/T P

ArgLysCysMetThr

81.982.548.880.763.2

86.084.059.585.069.8

**+******

Rostagno et al., 2009

General commentsUSA SBM vs. others

BrazilLess sucrose and P and more fiber and FeLess indispensable AA per unit of CPLower KOH, PDI, and HDI (lower AA digestibility)

ArgentinaLess CPLower KOH, PDI, and HDI (lower AA digestibility)

IndiaMore ash, crude fiber, and Fe (less energy)Less fat, sucrose, and indispensable AA per unit ofCPLower KOH (lower AA digestibility)

General conclusions

Current methods used by the industry to evaluate protein quality are not capable of detecting existing differences among SBM

The composition and the protein quality of SBM vary with the origin of the bean

Different matrixes should be used for SBM of different origins, NIR technology might help

Proteases might improve the uniformity and nutritive value of SBM batches

Thanks