a. A?
mm; .-
A.

;~.-:,'., 7,.
3151: (1»

it.

a W». .
{M'- N“

nﬂ 31:“

. .IIVW’I'I .

V um [mu-J.)

 

 

 

 

 

 

,‘.

d..-

.M. ..
4*..‘1‘.

u
9" u
. -'.
‘J& M
a
.4..."-

if?
5?».

I -!
. o.-
J \-

i!-

or us'
~53: his: '
., egg}:
J
F r

n w‘.’
:5.
”u,

,

 

cm
a
v.

a M.
........
g. ...,. -

 

’ Lao“)

This is to certify that the
dissertation entitled

REPLACEMENT OF FISH MEAL WITH SOYBEAN MEAL IN

DIETS FOR ATLANTIC SALMON, SALMO SALAR, EFFECTS

ON GROWTH, PROTEASE ACTIVITY, DIGESTIBILITY AND
INTESTINAL HISTOLOGY

 

presented by

Christopher Todd Weeks

LIBRARY
Michigan State
University

 

 

has been accepted towards fulﬁllment
of the requirements for the

 

Ph.D. degree in Fisheries and Wildlife

 

   

 

 

Major rofessor’s Signature //

&////o7

 

Date

MSU is an afﬁrmative-action, equal-opportunity employer

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p:/CIRClDateDue.indd-p.1

REPLACEMENT OF FISH MEAL WITH SOYBEAN MEAL IN DIETS FOR
ATLANTIC SALMON, SALMO SALAR, EFFECTS ON GROWTH, PROTEASE
ACTIVITY, DIGESTIBILITY AND INTESTINAL HISTOLOGY

By

Christopher Todd Weeks

A DISSERTATION

Submitted to
Michigan State University
In partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Fisheries and Wildlife

2007

ABSTRACT

REPLACEMENT OF FISH MEAL WITH SOYBEAN MEAL IN DIETS FOR
ATLANTIC SALMON, SALMO SALAR, EFFECTS ON GROWTH, PROTEASE
ACTIVITY, DIGESTIBILITY AND INTESTINAL HISTOLOGY
By

Christopher Todd Weeks

Wild harvested seafood resources are at or near maximum sustained yields
worldwide. This has, in part, caused a signiﬁcant increase in aquaculture production over
the last 50 years. Further expansion of farmed ﬁsh production is likely in order to meet
nutritional needs of a growing global human population. Since aquaculture depends
heavily on high quality ﬁsh meal (FM) as a feed ingredient, demand for ﬁsh meal is
expected to exceed global supplies in the near future. This could have major economic
impact on commercial aquaculture facilities raising carnivorous ﬁsh such as Atlantic
salmon. There has been increasing effort aimed at utilizing alternate protein sources to
help alleviate the growing demand for FM. Soybean meal (SBM), in particular, is
considered by many as a leading alternative to FM in formulated feeds for aquaculture.

Extensive research on SBM diets for carnivorous ﬁsh has identiﬁed various anti-
nutritional factors inherent to soya and other plant-based products. Among SBM anti-
nutritional factors, trypsin inhibitors can be a serious problem for salmonids because they
bind with protein digestive enzymes and interrupt nutrient absorption.

A series of experiments were conducted to examine effects of soybean trypsin
inhibitors (SBTI) and high nutrient dense (HND) practical SBM diets on Atlantic salmon

ﬁngerlings and smolts. In the trypsin inhibitor studies, stock SBTI were added to

standardized semi-puriﬁed diets containing 50% crude protein and 19% crude fat at
graded levels from 0 - 60% SBM equivalencies. In one trial, small Atlantic salmon (17.5
:t 1.4 grams) were fed in triplicate either a commercial control or test diets containing 0,
15, 30, 45, and 60% SBM equivalency TI for a period of 8 weeks. No signiﬁcant
differences were observed between growth rates of Atlantic salmon fed SBTI diets,
although the commercial control diet resulted in signiﬁcantly lower growth rates than
those fed the test diets. There were no differences in feed conversion rate (FCR, 0.78—
083), protein efﬁciency ratio (PER, 0.24—0.27), or apparent protein retention (APR,
32.9—35.7%) between diets. Slight differences were observed in proximate body
composition data, but they did not appear to be related to SBTI levels. In another trial,
similar experimental diets containing an inert marker were fed to Atlantic salmon smolts
(89.2 :t 4.1grams) for a period of 21 days. Slight differences were observed between
dietary SBTI levels in digesta dry matter and intestinal trypsin activity, and small
intestine trypsin activities varied over time. Neither apparent protein digestibility nor
body lipid compositions showed any affects from SBTI over the 21-day trial.

For the ﬁnal experiment HND diets containing 30% FM and 0, 20, 25, 30%
SBM, or 24% PM with 20 and 30% SBM were fed to Atlantic salmon for 12 weeks. No
differences were observed in grth (SGR, 1.88—1.94), FCR (0.78-0.82), PER (2.20—
2.32), trypsin activity, or intestinal histology. A negative linear response was observed
between SBM content, body lipid composition, and fecal dry matter. Fish whole body
lipid composition decreased from 22.0% to 12.9% for 0% SBM and 30% SBM diet
groups respectively. Study ﬁndings indicate use of HND SBM diets may contribute to

protein sharing ﬁInctions of Atlantic salmon to SBM carbohydrates.

This work is dedicated to all individuals contributing to the development for
sustainable aquaculture - past, present and future, especially my Mom - simple things like
the unrelenting support for a son, and aquaculture news article clippings from places out

west...

iv

ACKNOWLEDGEMENTS

First and foremost I thank my wonderful wife, Ianthe. In truth no words could
adequately tell you how lucky and grateﬁrl I am. Thank you for everything!

I wish to extend deep gratitude to Dr. Don Garling, my true friend and mentor.
Thank you, Don, for the opportunity to embark on this path, for your camaraderie and
guidance along the way. I also give sincere appreciation to my esteemed committee
members Dr. Ted Batterson, Dr. Mohamed Faisal and Dr. Nathalie Trottier for their
expertise, advice, and wealth of knowledge.

I give special thanks to Michigan State University Department of Fisheries and
Wildlife, Julie Moore of the Department of Animal Science, Dr’s. Steven Hart and Paul
Brown of Purdue University, and Dr. Rick Barrows, USDA Agricultural Research
Service. I also thank the United Soybean Board Soy-in-Aquaculture Program for
contributing ﬁnancial support for my research.

Lastly I give a very special thanks to my mentor and friend Harry Westers. I

would not be where I am today if it were not for him.

TABLE OF CONTENTS

Chapter 1 — Introduction ..................................................................................................... 1
Study Hypothesis ........................................................................................................ 2
Study Description ........................................................................................................ 3

Chapter 2 - Literature Review: Effects of Soybean Meal as an Alternative Protein

Source in Formulated diets for Atlantic Salmon (Salmo salar) .......................................... 4

Introduction ..................................................................................................................... 4
Atlantic Salmon .............................................................................................................. 5
The Fish Meal Dilemma ................................................................................................. 8
Feed Utilization in Intensive Fish Culture ...................................................................... 9
Protein Sparing Effects of Aquaculture Feeds .......................................................... 10
Alternative Protein Sources .......................................................................................... 12
SBM Manufacturing Process .................................................................................... l4
SBM Nutritional Characteristics ............................................................................... 15
SBM Anti-nutritional Factors ................................................................................... 16
Protease Inhibitors ................................................................................................ 17
Non-starch Polysaccharides .................................................................................. 18
Oligosaccharides ................................................................................................... l9
Saponins ................................................................................................................ 19
Lectins ................................................................................................................... 20
Antigens ................................................................................................................ 20
Isoﬂavones ............................................................................................................ 21
Phytic Acid ............................................................................................................ 21
Soybean meal-induced enteritis ................................................................................ 22
Continued Development of SBM Diets for Atlantic salmon ........................................ 23
Chapter 3 — Trypsin Inhibitor Effects on Atlantic Salmon (Salmo salar) Fingerlings ..... 25
Introduction ................................................................................................................... 25
Materials and Methods .................................................................................................. 26
Experimental Diets .................................................................................................... 26
Experimental System and Animals ........................................................................... 26
Sample Analysis and Calculations ............................................................................ 29
Feeding levels ....................................................................................................... 29
Growth .................................................................................................................. 30
Feed conversion .................................................................................................... 30
Whole body composition ...................................................................................... 31
Protein efﬁciency and retention ............................................................................ 31
Statistical Analysis .................................................................................................... 32
Results ........................................................................................................................... 33
Growth Characteristics .............................................................................................. 33
Feed Conversion and Protein Retention ................................................................... 33
Proximate Body Composition ................................................................................... 33

vi

Mortalities ................................................................................................................. 37

Discussion ..................................................................................................................... 38
Experimental Diets .................................................................................................... 38
Growth Characteristics .............................................................................................. 39
Feed Conversion and Protein Retention ................................................................... 40
Proximate Body Composition ................................................................................... 4O
Mortalities ................................................................................................................. 41

Chapter Summary ......................................................................................................... 42

Chapter 4 — Trypsin Inhibitor Effects on Atlantic Salmon (Salmo salar) Smolts ............ 44

Introduction ................................................................................................................... 44

Materials and Methods .............................................. ' .................................................... 44
Experimental Diets .................................................................................................... 44
Experimental System and Animals ........................................................................... 45
Sample Analysis and Calculations ............................................................................ 47

Digesta dry matter ................................................................................................. 47
Trypsin activity ..................................................................................................... 48
Apparent protein digestibility ............................................................................... 49
Whole body composition ...................................................................................... 49
Statistical Analysis .................................................................................................... 50

Results ........................................................................................................................... 50
Digesta Dry Matter ................................................................................................... 50
Trypsin Activity ........................................................................................................ 51
Apparent Protein Di gestibility .................................................................................. 52
Proximate Body Composition ................................................................................... 56

Discussion ..................................................................................................................... 58
Experimental Animals .............................................................................................. 58
Digesta Dry Matter ................................................................................................... 58
Trypsin Activity ........................................................................................................ 59
Apparent Protein Di gestibility .................................................................................. 60
Proximate Body Composition ................................................................................... 61
Fish Health and Compensatory Mechanisms ............................................................ 61

Chapter Summary ......................................................................................................... 62

Chapter 5. Fixed Formulation Model Development of Practical High Energy SBM Diets
for Atlantic Salmon (Salmo salar) .................................................................................... 64

Introduction ................................................................................................................... 64

Methods ......................................................................................................................... 66
Baseline Diet Formulation ........................................................................................ 66
Essential Amino Acid Requirements for Atlantic Salmon ....................................... 66
Vitamin and Mineral Requirements for Atlantic Salmon ......................................... 68
Ingredient Selection and Nutritional Properties ........................................................ 68
Ingredient Digestibility Indices ................................................................................. 7O
SBM Content ............................................................................................................ 70
Diet Formulation Model ........................................................................................... 71

Results ........................................................................................................................... 72

vii

Diet Formulation Model ........................................................................................... 72
Practical SBM HND Diet Formulations for Atlantic Salmon .................................. 73
Discussion ..................................................................................................................... 76
Chapter Summary ......................................................................................................... 80
Chapter 6 — Replacement of PM with SBM in High Energy Practical Diets of Smolting
Atlantic Sahnon (Salmo salar) .......................................................................................... 82
Introduction ................................................................................................................... 82
Methods and Materials .................................................................................................. 82
Diet Formulations and Processing ............................................................................ 82
Experimental System and Animals ........................................................................... 84
Sample Analysis and Calculations ............................................................................ 86
Feeding levels ....................................................................................................... 86
Growth .................................................................................................................. 86
Feed conversion .................................................................................................... 87
Protein efﬁciency .................................................................................................. 87
Whole body composition ...................................................................................... 88
Digesta dry matter ................................................................................................. 89
Trypsin activity ..................................................................................................... 89
Plasma insulin ....................................................................................................... 89
Hepatic somatic indices ........................................................................................ 90
Intestinal histology ................................................. . .............................................. 90
Muscle Tissue Amino Acids ..................................................................................... 90
Statistical Analysis .................................................................................................... 91
Results ........................................................................................................................... 91
Feed Compositions .................................................................................................... 91
Growth Characteristics, Feed Conversion and Protein Retention ............................ 92
Proximate Body Composition ................................................................................... 92
Digesta Dry Matter ................................................................................................... 96
Trypsin Activity ........................................................................................................ 97
Hepatic Somatic Index .............................................................................................. 97
Plasma Insulin ........................................................................................................... 98
Intestinal Histology ................................................................................................... 98
Muscle Tissue Amino Acids ..................................................................................... 98
Mortalities ................................................................................................................. 99
Discussion ................................................................................................................... 100
Feed Compositions .................................................................................................. 100
Growth Characteristics, Feed Conversion and Protein Retention .......................... 101
Proximate Body Composition ................................................................................. 102
Di gesta Dry Matter ................................................................................................. 102
Trypsin Activity ...................................................................................................... 103
Plasma Insulin and HSI ........................................................................................... 103
Intestinal Histology ................................................................................................. 104
Muscle Tissue Amino Acids ................................................................................... 105
Mortalities ............................................................................................................... l 05
Chapter Summary ....................................................................................................... 106

viii

Chapter 7 — Project Summary, Conclusions and Recommendations .............................. 107
Project Summary ......................................................................................................... 107
Conclusions ................................................................................................................. 108
Recommendations ....................................................................................................... 1 09

APPENDIX ..................................................................................................................... 111

REFERENCES ............................................................................................................... 156

ix

LIST OF TABLES

Table 2.1 Estimated ﬁsh meal and oil supply and usage in 1999 compared to 15 and 30
year projections. Source: New and Wijkstrdm (2002) ............................................... 9

Table 2.2 Estimated 2003 and projected 2010 feed conversion efﬁciencies (FCE) for
various farmed ﬁsh species. ...................................................................................... 11

Table 2.3 Comparison of ingredient properties between anchovy meal and potential
alternative protein sources for use in commercial salmonid diets. Proximate
compositions include crude protein (CP), crude fat (CF), crude ﬁber (CFbr) and ash.
................................................................................................................................... 13

Table 2.4 Essential amino acid content of Peruvian anchovy meal and soybean meal... 17

Table 3.1 Formulations of semi-puriﬁed test diets containing graded levels of stock
soybean trypsin inhibitors in g/kg. Diets are identiﬁed based on percent soybean
meal equivalency. ............................................................................. 27

Table 3.2 Mean ﬁnal weight, speciﬁc grth rate (SGR), condition factor (k), feed
conversion rate (FCR), protein efﬁciency ratio (PER), and apparent protein
retention (APR) in ﬁngerling Atlantic salmon fed diets containing graded levels of
trypsin inhibitors (T1) at 0, 15, 30, 45, and 60% soybean meal equivalencies. Mean
standard errors are in parentheses. Values in each row awarded common
superscripts are not signiﬁcantly different (P > 0.05). ................................. 34

Table 3.3 Percent body compositions of ﬁngerling Atlantic salmon fed a commercial
control and experimental diets containing graded levels of trypsin inhibitors (T1) at
O, 15, 30, 45, and 60% soybean meal equivalencies. Mean standard errors are in
parentheses. Values in each row awarded common superscripts are not
signiﬁcantly different (P > 0.05). .......................................................... 36

Table 4.1 Formulations of semi-puriﬁed test diets containing graded levels of stock

soybean trypsin inhibitors (TI) in g/kg. Diets are identiﬁed based on percent
soybean meal equivalency. ................................................................. 45

Table 4.2 Dietary trypsin inhibitor (TI) affects on percent dry matter (lyophilized) of
intestinal digesta from sampled regions of small intestine (SI), proximal large
intestine (LI) and distal large intestine (F) on day-10 and day-21. Diet C is the
commercial control. TI# represents the percent soybean meal equivalency (0, 15,
30, 45 and 60%) of puriﬁed T1 in test diets. Values in each row awarded common
superscripts are not signiﬁcantly different (P > 0.05). ................................. 52

Table 4.3 Dietary trypsin inhibitor (TI) affects on fecal sample mean total apparent
protein digestibility (TAPD). MSE represents the mean square error value. TI#
represents the percent soybean meal equivalency (0, 15, 30, 45 and 60) of puriﬁed
T1 in test diets. ............................................................................... 56

Table 5.1 Dietary ingredients and base-line nutritional characteristics of the Ontario
Ministry of Natural Resources open formula diet, MNR—98HS, used as a basal
control diet to formulate soybean meal-based practical diets for Atlantic salmon.
.................................................................................................. 67

Table 5.2 Atlantic salmon essential amino acid requirements obtained from literature (%
protein) and values selected for use in the diet formulation model (% as fed) based
on 50% crude protein diets. ................................................................. 67

Table 5.3 Atlantic salmon diet formulation vitamin and mineral requirements obtained
ﬁ'om various sources. ........................................................................ 69

Table 5.4 Di gestible protein (DP) and digestible energy (DE) values of practical
ingredients used in diet formulations for Atlantic salmon. .............................. 71

Table 5.5 Atlantic salmon basal (MNR-98HS), and soybean meal-based diet
formulations. Soybean meal treatment diets were identiﬁed according to percent
soybean meal (SB) and ﬁsh meal (F) inclusion. Diets containing 24% ﬁsh meal
were formulated with 20% reduction in ﬁsh meal from the control. Diet MNR-
98HS is an open formula high nutrient dense diet developed by the Fish Nutrition
Research Laboratory, University of Guelph and Ontario Ministry of Natural
Resources. .................................................................................... 75

Table 6.1 Compositions of test diets (as fed basis). Measured crude protein (CP), crude
fat (CF), ash, carbohydrate (CH0) and energy content (E). MNR-98HS is the open
formula control diet. Test diets are identiﬁed according to percent soybean meal
(SB) and percent ﬁsh meal (F). Formulated values are in parentheses. ..............92

Table 6.2 Mean ﬁnal weight, speciﬁc grth rate (SGR), condition factor (k), feed
conversion rate (FCR), and protein efﬁciency ratio (PER) in smolting Atlantic
salmon fed diets containing varying levels of soybean meal and ﬁsh meal. Mean
standard errors are in parentheses. MNR-98HS is the open formula control diet.
Test diets are identiﬁed according to percent soybean meal (SB) and percent ﬁsh

meal (F). Values in each row awarded common superscripts are not signiﬁcantly
different (P > 0.05). .......................................................................... 93

xi

Table 6.3 Body composition analysis of smolting Atlantic salmon fed diets varying
levels of soybean meal and ﬁsh meal. Mean standard errors are in parentheses.
MNR-98HS is the open formula control diet. Test diets are identiﬁed according to
percent soybean meal (SB) and percent ﬁsh meal (F). Values in each row awarded
common superscripts are not signiﬁcantly different (P > 0.05). ...................... 94

Table 6.4 Hepatic somatic index (HSI), intestinal trypsin activity (TA) and whole body
energy composition of smolting Atlantic salmon fed diets varying levels of soybean
meal and ﬁsh meal. Mean standard errors are in parentheses. MNR-98HS is the
open formula control diet. Test diets are identiﬁed according to percent soybean
meal (SB) and percent ﬁsh meal (F). Values in each row awarded common
superscripts are not signiﬁcantly different (P > 0.05). .................................. 97

Table 6.5 Muscle tissue amino acid compositions (umol/g) of smolting Atlantic salmon

fed a control (MNR-98HS) and a 30% soybean meal (SB) 24% ﬁsh meal (F) diet
for 22 weeks. Values with asterisk (*) were signiﬁcantly different (P < 0.05). ...99

xii

LIST OF FIGURES

Figure 2.1 Global commercial Atlantic salmon production (bars) and value (line) from
aquaculture 1964-2004. Source: FAO FIGIS data base (FAO 2002—2007). ............. 7

Figure 3.1 Box plot of mean speciﬁc growth rate of ﬁngerling Atlantic salmon fed a
commercial control and experimental diets containing graded levels of stock trypsin
inhibitors (T1) at O, 15, 30, 45, and 60% soybean meal equivalencies. ............... 35

Figure 3.2 Box plot of mean body fat composition of ﬁngerling Atlantic salmon fed a
commercial control and experimental diets containing graded levels of stock
trypsin inhibitors (TI) at 0, 15, 30, 45, and 60% soybean meal equivalencies. .. . ...36

Figure 3.3 Number of mortalities by treatment (diet) and replicate (tank) of ﬁngerling
Atlantic salmon over the 8-week feed trial. Diet C is a commercial control. TI#
represents percent soybean meal equivalency (O, 15, 30, 45 and 60%) of dietary
trypsin inhibitor (TI). ........................................................................ 38

Figure 4.1 Day-10 trypsin enzyme activity by dietary treatment groups from samples of
A) small intestine (SI), B) proximal large intestine (LI) and C) distal large intestine
(fecal; F). Trend lines based on percent soybean meal equivalent (0, 15, 30, 45 and
60) trypsin inhibitor (TI) level in treatment diets (commercial control, C, excluded).
P(linear) signiﬁes P-value for linear regression analysis. Plotted values in graphs
awarded common superscripts are not signiﬁcantly different (P > 0.05). ............53

Figure 4.2 Day-21 trypsin enzyme activity by dietary treatment groups from samples of
A) small intestine (SI), B) proximal large intestine (LI) and C) distal large intestine
(fecal; F). Trend lines based on percent soybean meal equivalent (0, 15, 30, 45 and
60) trypsin inhibitor (TI) level in treatment diets (commercial control, C, excluded).
Plotted values in graphs awarded common superscripts are not signiﬁcantly
different (P > 0.05). .......................................................................... 54

Figure 4.3 Apparent protein digestibility values (APD) of Atlantic salmon smolts based
on dietary trypsin inhibitor (TI) treatments from samples taken in regions of small
intestine (SI), proximal large intestine (LI), and distal large intestine (fecal; F). TI#

represents the percent soybean meal equivalency (O, 15, 30, 45 and 60) of puriﬁed
T1 in test diets. ................................................................................ 55

Figure 4.4 Cumulative apparent protein digestibility in intestinal segments of Atlantic
salmon smolts fed experimental diets containing graded levels of trypsin inhibitor
(T1) at O, 15, 30, 45 and 60% soybean meal equivalencies. ........................... 55

xiii

Figure 4.5 Box plot of mean body fat composition of Atlantic salmon smolts fed a
commercial control (C) and experimental diets containing graded levels of trypsin
inhibitor (TI) at 0, 15, 30, 45 and 60% soybean meal equivalencies. ................. 57

Figure 5.1 Digestible protein (DP), relative ﬁber and digestible energy (DE) contents of
principle ingredients used to formulate high nutrient dense soybean meal diets for
Atlantic salmon. Relative ﬁber is equal to % ﬁber x 10.0 as a scalar to the y-axis.
Optimal ingredients would be high in DP and DE and low in ﬁber. .................. 78

Figure 6.1 Percent whole body lipid compositions of smolting Atlantic salmon as a
function of dietary soybean meal. P(linear) signiﬁes P-value for linear regression

analysis. Values in each row awarded common superscripts are not signiﬁcantly
different (P > 0.05). .......................................................................... 95

Figure 6.2 Large intestine digesta dry matter (lyophilized) of smolting Atlantic salmon as
a function of dietary soybean meal. P(linear) signiﬁes P-value for linear regression

analysis. Values in each row awarded common superscripts are not signiﬁcantly
different (P > 0.05). .......................................................................... 96

Figure 6.3 Mortalities by treatment (Diet) of smolting Atlantic salmon over the 12-week
feed trial. MNR-98HS is the open formula control. Test diets are identiﬁed
according to percent soybean meal (SB) and percent ﬁsh meal (F). ................ 100

xiv

LIST OF ABBREVIATIONS

#FIShf - number of ﬁsh alive at end of study
#TU - number of thermal units in oC above 00°C
% PFeed - percent protein in diet as fed basis
%BW - percent body weight

0C - temperature in Celsius

ALA - alanine

ANF - anti-nutritional factor

ANOVA — analysis of variance

APD - apparent protein digestibility coefﬁcient
APR - apparent protein retention

Arg - arginine

ASN - asparagine

BAPA - Benzoyl-DL—arginine-p-nitranilide

Ca - calcium

CF - crude fat

CFbr - crude ﬁber

CHO - carbohydrate

CP - crude protein

CID - percent marker (CrzO3) in diet

Crp - percent marker (Cr203) in faeces

Cu - copper

XV

Cum Feed - cumulative weight of feed

Cys - cystine

d - time

DE - digestible energy

DMFeed - percent dry matter of feed

DNR - Department of Natural Resources

DP - digestible protein

E - energy

EVol - extract volume

F - fecal

FAO - Food and Agriculture Organization of the United Nations
F CE - feed conversion equivalencies

F CR - feed conversion rate

FDf - ﬁnal percent freeze-dried matter of ﬁsh samples
FD, - initial percent freeze-dried matter of ﬁsh samples
FE - feed efﬁciency

Fe - iron

FM - ﬁsh meal

FPG - ﬁsh protein gain

FRCLO - theoretical feed conversion rate of 1.0

FWS - Fish and Wildlife Service

GLU - glutamic acid

GLY - glycine

xvi

His - HIS - histidine

HND - high nutrient dense

HSI — hepatic somatic index

HYP - hydroxyproline

LAA - indispensible amino acid
116 - ILE - isoleucine

k - condition factor

kg - condition factor at the start of the trial
Leu - LEU - leucine

LF - ﬁnal length

L,- - initial length

LI - large intestine

me - liters per minute

LW - liver weight

Lys - LYS - lysine

Magn - Maganese

Met - MET - methionine

Mg - magnesium

mmt - million metric tonne
M8222 - tricaine methanesulfonate
MSE - mean square error

MSU - Michigan State University

N - nitrogen

xvii

Na - sodium

NMons - number of mortalities removed

NSP - non-starch polysaccharide

OMNR - Ontario Ministry of Natural Resources
ORN - ornithine

Panta—Acid - pantothenic acid

PD - percent protein of diet

PER - protein efﬁciency ratio

Pf - ﬁnal percent protein in body compositions of freeze-dried sample
Pp - percent protein of feces

Phe - PHE - phenylalanine

Phos - phosphorous

P, - initial percent protein in body compositions of freeze-dried sample
PIFeed - protein intake from feed

Pot - potassium

P(reg) - P-value for regression analysis

PRO - proline

Pyrdox - pryidoxine

Ribo - riboﬂavin

SBM - soybean meal

SBTI - soybean trypsin inhibitor

Se - selenium

SER - serine

xviii

SGR - speciﬁc grth rate

SI - small intestine

TA - trypsin activity

TAPD - total apparent protein digestion
TAU - Taurine

Thia - thiamine

Thr - THR - threonine

TI - trypsin inhibitor

TPFeed - total protein in feed

TRIS - Tris (hydroxyrnethyl) aminomethane
Trp - TRP - tryptophan

Try - trypsin value

TUG - temperature unit growth rate

Tyr - TYR - tyrosine

USB - United Soybean Board

Val -VAL - valine

W - wet weight of ﬁsh at time of sampling
Wf - average ﬁnal wet weight A
WGM - wheat gluten meal

W, - initial wet weight

WMon, - wet weight of mortalities removed

Zn - zinc

xix

Chapter 1 — Introduction

Soybean meal (SBM) has been widely considered as a viable alternative protein
source for ﬁsh meal in formulated ﬁsh feeds for aquaculture (Hardy 1996, 2003,
Storebakken et al. 1998). Studies have shown, however, that SBM diets can lead to
signiﬁcantly reduced growth rates and cause severe health problems for certain, mainly
carnivorous, ﬁsh species such as salmonids (Arnesen et al. 1989, Refstie et al. 2000,
Krogdahl et al. 2003).

Atlantic salmon are an important commercial aquaculture species in the European
Union, and North and South America with total production in 2005 estimated at over 1.2
million metric tonnes (mmt) and an value of $4.7 billion US (FAQ 2002—2007). Due to
increasing demands in salmon production and ﬁsh meal, a relatively large number of
studies have been conducted in effort to determine the maximum amount of dietary
ﬁshmeal that can be replaced by SBM in Atlantic salmon and rainbow trout feeds (Hardy
2003). After over a decade of research the maximum safe level of dietary SBM in
formulated diets for these salmonid species remains uncertain.

In 2002, a collaborative project was funded by the United Soybean Board (USB),
Soy—in-Aquaculture Program, to examine the extent of SBM anti-nutritional properties in
formulated feeds for salmonids. The project was designed to develop a commercially
acceptable SBM based formulated feed for Atlantic salmon and other species. Michigan
State University (MSU), Department of Fisheries and Wildlife, was awarded a 2-year

project funded through the USB Soy-in-Aquaculture Program to examine effects of

trypsin inhibitors and SBM on Atlantic salmon. This dissertation describes the
experimental design, methods and results of studies conducted at MSU.

The research conducted at MSU was originally intended as a 2-phase
experimental design to examine affects of trypsin inhibitors in SBM-based diets on
Atlantic salmon. Phase I of the design focused on the effect of growth, feed
consumption, digestibility, and pancreatic proteolytic enzyme activity of juvenile and
smolting Atlantic salmon (Salmo salar) fed puriﬁed diets with graded levels of trypsin
inhibitors. The second phase of this research project was initially intended to examine
effects of trypsin inhibitors in SBM-based diets containing practical feed ingredients
under different processing conditions.

Based on results obtained by other researchers in the Soy-in—Aquaculture Program
in 2003 and an extensive literary review, MSU expanded the focus of the research
scheduled in phase II to include a detailed assessment of commercial SBM diets for
Atlantic salmon. The overall goal of this study was to develop a potentially
commercially viable, practical diet for Atlantic salmon containing the highest level of

SBM possible based on best available knowledge.

Study Hypothesis

Hypothesis:

 

Based on literature review and recent unpublished study data the maximum level
of SBM incorporation into formulated diets for Atlantic salmon is expected to be
in the range of 20-3 0% wet ingredient weight without adverse affects on growth,

feed efﬁciency, and/or other observable ﬁsh health characteristics.

 

 

 

Study Description

This dissertation is presented in 7 chapters. This ﬁrst chapter provides a brief
introduction of the research undertaken at MSU and the hypothesis developed by the
researcher in regards to SBM diets for Atlantic salmon. Chapter 2 is a literature review
providing relevant background information of the status of SBM as an alternative protein
source in formulated diets for Atlantic salmon.

Chapters 3 and 4 describe two phase I trypsin inhibitor feed trials completed in
2003. Chapter 5 - Fixed Formulation Model Development of Practical High Energy
SBM Diets for Atlantic Salmon (Salmo salar), draws upon widely dispersed information
available speciﬁcally on Atlantic salmon diet formulations to develop a practical feed
formulation model. Diet formulation model parameters are provided in the Appendix.
Chapter 5 was added for potential use as a “blueprint” for anyone initiating a similar
exercise. Chapter 6 - Replacement of FM with SBM in High Energy Practical Diets of
Smolting Atlantic Salmon (Salmo salar), provides details and results of the 2004 practical
SBM diet study on Atlantic salmon. This study was designed to determine maximum
levels of SBM that can be safely added to commercially viable diets for Atlantic salmon
based on best available knowledge.

Chapter 7 summarizes the research results and hypothesis testing. The chapter

also provides conclusions and recommendations for further research.

Chapter 2 - Literature Review: Effects of Soybean Meal as an Alternative Protein
Source in Formulated diets for Atlantic Salmon (Salmo salar)

Introduction

Global production from capture ﬁsheries and aquaculture supplied about 140
million metric tonnes (mmt) of food ﬁsh in 2004 (FAO 2006). According to Wijkstrdm
(2003), human consumption of seafood products is expected to increase to approximately
121 mmt by the year 2010, and 271 mmt by 2050. Most experts agree that wild capture
ﬁsheries are at maximum yield at about 100 mmt per year. Global aquaculture
production must increase then in order to meet projected demands in seafood
consumption.

According to the Food and Agriculture Organization of United Nations (FAO),
current expansion rates observed in world aquaculture production combined with
anticipated needs for human consumption of seafood products are expected to cause a
global shortage of ﬁsh oil by the year 2010 and ﬁsh meal by 2015 (New and Wijkstrdm,
2002). In a recent survey of 600 ﬁsh species, 77% of the world’s marine ﬁsh stocks are
estimated to be either fully exploited, over exploited or depleted (FAO 2004). Adding to
this dilemma are reports and highly publicized news events linking elevated contaminant
levels found in farmed Atlantic salmon destined for human consumption to ﬁsh meals
and oils used to manufacture aquaculture feeds (Hites et al. 2004, Foran et al. 2005).
Based on current trends, future demands for high quality ﬁsh meals and oils are likely to
have severe economic impacts on global aquaculture production. Clearly, incentives
exist for the development of alternate protein and lipid sources for use in formulated

feeds for aquaculture.

Atlantic Salmon

Indigenous to the North Atlantic, the native range of Atlantic salmon (Salmo
salar) extends from the Arctic Ocean and Baltic Sea to Portugal in the east, and ﬁ'om
Iceland, to Southern Greenland, Canada, and Northern US in the western Atlantic
(Netboy 1974 in Danie et a1. 1984). Original landlocked populations existed in Maine,
New Brunswick and Nova Scotia, and transplanting has occurred to literally hundreds of
inland lakes (Danie et a1. 1984). Distinct segments of Atlantic salmon ﬁ'om Maine have
been listed as either endangered or threatened under the Endangered Species Act of 1973.

Atlantic salmon are cold water carnivorous species with a preferred water
temperature range of 8-180C (Sedgewick 1988, Jensen et a1 1991). Adults can survive in
both fresh and salt water. They are an anadromous - wild adults return from the ocean or
landlocked lakes to spawn in gravel areas of freshwater streams, and iteroparous - spawn
more than once. Various populations migrate into rivers anytime from the spring to late
fall, and peak spawning season is typically from mid-October to November (Bigelow and
Schroeder 1953). Their egg incubation period is temperature dependent (Ojanguren et al.
(1999), and may range from 2-3 months in a hatchery to several months in the wild under
normal winter conditions. Wild newly hatched fry (alevins), remain in the gravel
approximately 6 weeks until their yolk sacks are depleted of nutrients (Bigelow
and Schroeder 1953). At this time they must emerge and begin foraging for food.

At approximately 4.0 cm in length, normally achieved in the ﬁrst summer, young
Atlantic salmon are classiﬁed as parr or ﬁngerlings (Danie et al. 1984). Most parr remain
in the stream for 2-3 years (125-150 mm length), although Schaffer and Elson( 1975)

have reported populations in Ungava Bay region of Canada may remain in fresh water for

4-8 years (180 mm long). Prior to seaward migration Atlantic salmon undergo a
physiological transformation called smoltiﬁcation. These changes allow for a variety of
salmonids to survive in a salinity environment (Sedgewick 198 8), and occurs whether
smolt migration is from river to either sea or fresh water.

Wild ﬁsh typically grow to maturity in 2-3 years in the sea or landlocked lakes
before returning to their home stream to spawn. Returning Atlantic salmon usually range
between 3-9 kg in weight, although much larger ﬁsh have been observed and recorded
(Scott and Crossman 1973).

Atlantic salmon culture began in the 19th century in the United Kingdom in
freshwater as a means of stocking waters with parr in order to enhance wild returns for
anglers (F AO 2000—2007). North Atlantic commercial salmon catch data indicated a
high abundance cycle from the mid-19608 to the 1970s with a maximum harvest of
12,000 tonnes in 1967 (Mills 1989). Numbers of returning ﬁsh declined greatly over this
time frame, and presently nearly all commercial ﬁsheries for wild Atlantic salmon are
closed (Parrish et al. 1998). According to these authors, over ﬁshing, dam construction,
pollution, and dewatering of streams caused the decline and extirpations of Atlantic
salmon.

Atlantic salmon are a widely popular commercially farmed aquaculture species in
Europe, North and South America and Australia. Commercial aquaculture of Atlantic
salmon ﬁrst began in the early 19605, and production has increased dramatically over the
past few decades (Figure 2.1). In 2005, total annual production from commercial

Atlantic salmon farms exceeded an estimated 1.2 mmt worth for $4.7 billion US (FAQ

2002—2007).

1400 5000

 

 

 

 

 

 

 

A 1200

8 4000

E 1000 E

E 3000 E

g 800 - . E

3 3
w

g 600 2000 3'

a 2

E 400 g

S 1000

0 200 III

0 V 1 1 ‘__V Illll 0

 

1965 1969 1973 1977 1981 1985 1989 1993 1997 2001 2005

Year

Figure 2.1 Global commercial Atlantic salmon production (bars) and value (line) from aquaculture
1964-2004. Source: FAO FIGIS data base (FAO 2002—2007).

Current intensive rearing programs utilize culture techniques that provide a great
deal of control over production cycles (AAFC 2004). Temperature manipulation, for
example, can signiﬁcantly speed up the time required to grow from egg to ﬁngerling
stages (Ojanguren et a1. 1999). Sahnon eggs are collected from either captive or wild
female broodstock during fall spawning cycles. Eggs are mixed with milt from males
and incubated at around 6-10°C optimal water temperatures (Massachusetts Ofﬁce of
Coastal Zone Management 1995, FAO 2000—2007). Newly hatched fry (sac fry)
typically emerge in 2-4 months depending on incubation water temperatures (AAFC

2004). Sac fry use up oil sack energy reserves in approximately 3-6 weeks, at which

point they become swim-up fry and readily begin to feed on formulated starter diets.
Cultured Atlantic salmon are kept on formulated diets throughout the entire production
cycle. Fish are usually graded several times to maintain ﬁsh of the same size together in
individual rearing units (AAF C 2004). This practice reduces cannibalism and facilitates
product uniformity. Photoperiod manipulation can shorten the fresh water production
phase to about 6 months (IFREMER 2005). Smolts are transferred to ocean grow out
facilities (cages or net enclosures) in the spring, where they are raised to a market size of
between 8 and 10 pounds. The overall production cycle of farmed Atlantic salmon takes

approximately 20 months from hatch to harvest (AAF C 2004).

The Fish Meal Dilemma

Signiﬁcant expansion of the commercial aquaculture industry faces a major
dilemma. Most processed feeds for aquaculture contain varying amounts of ﬁsh meal
and oils. Carnivorous species, such as trout and salmon require up to 45—75% ﬁsh meal
of very high quality. High quality ﬁsh meals used to make commercial feeds for
aquaculture are obtained from wild harvested small pelagic ﬁsh species including
Peruvian anchoveta, Icelandic herring, menhaden (Gulf of Mexico), and Norwegian
capelin. These species are currently being harvested at or close to maximum yields.

New and Wijkstrdm (2002) examined global projections of aquaculture
production rates and projected 15 and 30 year ﬁsh meal usage requirements for
aquaculture (Table 2.1). Their conclusions indicate that a supply and demand crisis for
ﬁsh meal is highly likely over the projected timeline. This crisis is further compounded

by the point that aquaculture uses a disproportionate amount of ﬁsh meal as compared to

Table 2.1 Estimated fish meal and oil supply and usage in 1999 compared to 15 and 30 year
projections. Source: New and Wij kstriim (2002).

 

 

 

Fish Meal Fish Oil
Year Global Supply Usage by Aquaculture Global Supply Usage by Aquaculture
(thous. mmt) (°/o) (thous. mmt) (°/o)
1999 6548 32 1360 49
2015 6526 70 1283 145
2030 6526 ' 1 59 1283 460

 

 

 

 

 

 

 

other forms of agriculture. While aquaculture makes up only approximately 3% of global
animal feed production, 45% of ﬁsh meal usage goes into aquaculture feeds (Gill 2005,
Pike 2005). In order for aquaculture to successfully meet projected needs for food ﬁsh
for humans, ways must be found to reduce the amount of wild ﬁsh required to feed ﬁsh,
livestock and poultry. This realization has stimulated a rather intensive international
effort aimed at reducing global reliance on ﬁsh meal and oils in animal feeds. Most of
the focus of research to date has been placed on improving feed utilization of cultured

species, and replacement of ﬁsh meal by alternative protein sources.

Feed Utilization in Intensive Fish Culture

Feed utilization in intensive aquaculture is most often measured by feed
conversion ratio (FCR) or by feed efﬁciency (FE). FCR is equal to feed fed divided by
weight gain. FE is equal to the reciprocal of FCR expressed in percent. For example, a
FCR of 1.4 provides a reciprocal of 0.71 and FE equal to 71%. The lower the FCR, the
higher the FE and hence feed utilization efﬁciency. One important aspect of FCR is that

it does not account for differences between feed and ﬁsh dry matter content. For

example, an FCR of 1.0 (FE = 100%) is considered very good by industry standards
(Harry Westers, Aquaculture Bioengineering Corporation, personal communication). For
dry matter contents of 95% and 28% for feed and ﬁsh respectively, the true feed
conversion would equate to 3.4, or 30% efficiency.

Fish feed utilization rates in the aquaculture industry have substantially improved
over the past few decades due to advances made in feed formulation methods, feed
manufacturing technology, and feed management practices. According to Tacon (2005),
the average F CR for commercial Atlantic salmon farming operations has been reduced
from >2.0 before 1985, to 1.3 in 2003. This author also points out that FCRS for farmed
salmon and large rainbow trout are the lowest of all the major aquaculture species.

Another way to assess feed utilization is in terms of ﬁsh conversion
equivalencies. Approximately 4 to 5 tons of whole ﬁsh are required to produce 1 ton of
dry ﬁsh meal (Miles and Chapman, 2006). Feed conversion equivalencies (FCE) then is
the apparent conversion efﬁciency of pelagic ﬁshes (wet weight basis; calculated by
summing total ﬁshmeal and ﬁsh oil consumption ﬁgures and then multiplying by 4 or 5)
to farmed ﬁsh (Tacon 2004, 2005). According to Tacon, F CE’s should continue to

decrease over the next several years (Table 2.2)

Protein Sparing Effects of Aquaculture Feeds
Carnivorous ﬁsh are notably very efﬁcient at using protein as an energy source.
Gaitlin III (1995), attributes this due to the ability of ﬁsh to efﬁciently deaminate

ammonia from protein and excrete it through the gills with limited energy expenditure.

10

Table 2.2 Estimated 2003 and projected 2010 feed conversion efﬁciencies (FCE) for various farmed
fish species.

 

 

FCE (Fish inputzoutput) 20031 20102
Marine eels 3.1 — 3.9 1.8-2.3
Salmon 3.1 - 3.9 1.2-1.5
Marine ﬁsh 2.5 - 3.2 1.5-1.9
Trout 2.5 - 3.1 0.8-1.0
Marine shrimp 1.6 - 2.0 1.0-1.2
Freshwater crustaceans 0.9 - 1.1 0.5-0.6
Milkﬁsh 0.30 - 0.37 0.11-0.14
Tilapia 0.23 - 0.28 0.11-0.14
Catﬁsh 0.22 - 0.28 0.16-0.20
Feeding carp 0.19 - 0.24 0.02

1 FAQ Fisheries Department, Fishery Information, Data and Statistics Unit. Fishstat Plus (2005) in Tacon

2005)
Tacon (2004)

The concept of protein sparing typically refers to the protein utilization efﬁciencies of
monogastric animal to feeds containing carbohydrates. With a greater fraction of energy
supplied by carbohydrates, more protein is utilized for protein anabolism. Studies have
shown that inclusion of carbohydrates improves protein sparing effects for nearly all
species (Hemre et al. 1993, Sanchez-Muros et al. 1996). The later authors suggested
protein Sparing and growth promotion is aided by depression of gluconeogenic activity by
dietary carbohydrates, which divert amino acids away from oxidative pathways.
Carnivorous ﬁsh species, however, such as salmonids are limited to approximately 10%
starch before growth is depressed (Hernre et al. 1993, 1995). Increased levels of

indigestible carbohydrates have Shown to increase hepatic lipogenesis (Brauge ct a1.

11

1994, 1995). Signs of negative impacts on protein sparing include decreased protein

deposition, increased liver size and increased liver lipid content.

Alternative Protein Sources

In terms of commercial aquaculture production, high valued carnivorous species
(salmon, trout, eels) require much greater amounts of high quality protein from ﬁsh meal
(FM) than omnivorous species (catﬁsh, tilapia). Salmonids, for instance, require 40-50%
protein (Hardy 1996) and approximately 35% FM (Tacon 2005), compared to channel
catﬁsh requirements of 32-36% protein (Garling and Wilson 1976) and only 3% FM
(Hardy 2000). Further reduction of ﬁsh meal in carnivorous ﬁsh diets requires
development of nutritionally and environmentally acceptable, low cost alternative protein
sources.

Major feed ingredients for salmonids must be high in protein, have high
digestibility value, provide necessary essential amino acids, provide essential fatty acids,
and contain relatively low levels ash, carbohydrates and ﬁber. From an environmental
standpoint, fecal output, and ammonia and phosphorus content in efﬂuent water must
satisfy facility discharge requirements. Obviously, the ingredient or ingredients must
also be available at a lower cost than ﬁsh meal. According to Hardy (1996), the selective
properties required of protein sources for salmonid diets limit potential choices to a small
list of ingredients.

Alternative practical ingredient protein sources best suited for salmonid diets are
listed in Table 2.3. Note that each of the ingredients listed has one or more negative

attributes when compared to a standard anchovy meal. Plant-derived meals are usually

12

Table 2.3 Comparison of ingredient properties between anchovy meal and potential alternative
protein sources for use in commercial salmonid diets. Proximate compositions include crude protein
(CP), crude fat (CF), crude fiber (CFbr) and ash.

 

 

Ingredient Proximate composition Negative qualities1
(°/o)

CP CF CFbr Ash
Anchovy meal 70 5.3 1.0 16.9 -- - -- -- -- --
Soybean meal 50 0.9 3.4 5.8 antinutritional factors
Canola meal 38 3.8 11.1 6.8 high ﬁber, phytic acid
Corn gluten meal 60 1.8 1.5 2.1 ﬁber, colors ﬁsh ﬂesh yellow
Wheat gluten meal 80 1.5 0.5 0.7 cost
Soy protein concentrate 70 0.75 4.2 7 low protein solubility, cost
Rapeseed protein concentrate 77 0.8 -- 14.2 cost, availability, phytic acid
Pea protein concentrate 76 3 1 4 cost, antinutritional factors?
Feather meal 83 5.4 1.2 2.9 variable digestibility, 2
Poultry by-product meal 60 13.6 2.1 14.5 low essential amino acids, 2
Meat meal 55 8.7 2.3 27 high ash, phosphorus, 3
Blood meal 89 0.7 1.0 2.3 cost, 3

 

‘ Modiﬁed from Hardy (1996)
2 Avian ﬂu concerns
3 Mad cow and TB concerns

the least expensive, but most are lower in protein than ﬁsh meal. In addition, all of the
plant meals listed in Table 2.3 have been shown to contain various anti-nutritional
properties in association with use in salmonid diets (Carter and Hauler 2000, Buttle et al.
2001, Francis et al. 2001).

Of all the alternative ingredients examined for use in aquaculture feeds, SBM has
probably received the most attention. Soybeans constitute about 50% of the total oilseed
crops worldwide and have become the most important source of plant proteins in the diets

of monogastric animals (Alexis and Nengas 2001). SBM is a cost effective alternative to

13

ﬁsh meal and is an established agriculture product. in the US. SBM was ﬁrst tested in
diets for trout in the early 19403 (Hardy 2003), and today is a common ingredient in
many formulated feeds for aquatic and terrestrial animal species. Presently SBM may
comprise 50% or greater of all dietary ingredients for omnivorous aquaculture species
such as catﬁsh, tilapia and carp (ElSaidy and Gaber 2002, Peres et a1. 2002, J ahan et al.
2003). Its use in diets of carnivorous species like Atlantic salmon, however, has been
severely limited due to a number of associated anti-nutritional properties. These will be
discussed in more detail in following sections of this review.

Animal by—product meals have been commonly used in conjunction with ﬁsh
meal in salmon diets. However, these ingredients are also expensive and can vary
substantially in digestibility properties from batch to batch. Fairly recently, global threats
of highly contagious emerging diseases (e. g. mad cow, avian ﬂu), have resulted in
international quarantines of beef and poultry products (EC Regulation 999/2001 , CDC

2007)

SBM Manufacturing Process

“Dehulled solvent extracted” SBM is the most common form used in aquaculture
feeds. The processing method for this product is described in an American Soybean
Association Technical Bulletin by Behnke entitled U.S. Soybean Meal Extraction,
Processing and Specifications (K. Behnke, Kansas State University, personal
communication) and is described as follows:

Soybeans are cracked and dehulled on special rollers, then heat conditioned at a
temperature of 60°C for approximately 10 minutes. The conditioned beans are ground

into a “ﬂake” through another series of rollers, cooled, and solvent extracted. In this

14

process, the solvent, normally hexane, passes through the ﬂake in a counter current
exchange, removing most of the soybean oil and other soluble materials. Extracted ﬂakes
are dried under heat to volatilize and remove the solvent, and dried ﬂakes are cooled and
ground into meal. The oil rich material, miscella, is most often heated (distilled) to
recover the solvent. The remaining miscella can be reﬁned into oils for cooking and

other purposes.

SBM Nutritional Characteristics

Dehulled solvent extracted SBM typically contains 47-50% protein and 3-4%
crude ﬁber. In comparison, high quality ﬁsh meal contains 64-72% protein, 0.5-1.0%
ﬁber (NRC 1993, Hertrampf and Piedad-pascual 2000). Differences in fat content
between SBM and ﬁsh meal has, until recently, been of low concern in diet formulations
because of the high energy supplied by ﬁsh oil. It is fairly clear by the projections of
New and Wijkstrdm (2002, Table 2.1), that future demands for ﬁsh oil will exceed that of
ﬁsh meal. Researchers have begun to look for alternative energy sources for aquaculture
diets. For example, Torstensen et a1. (2005), reported that 100% of ﬁsh oil can be
replaced with a vegetable oil blend without compromising growth or ﬂesh quality of
Atlantic salmon. The current challenge in this area is ﬁnding an energy source rich in
polyunsaturated omega-3 fatty acids, EPA (eicosapentaenoic acid, C20:5n-3), and DHA
(docosahexaenoic acid, C22:6n-3). Presently, high levels of EPA and DHA can only be
obtained commercially through marine ﬁsh oils.

SBM has one of the best amino acid proﬁles of all protein-rich plant-based

ingredients for meeting essential amino acid requirements of ﬁsh (Mohsen 1989 in NRC

15

1993). According to Wilson (2002), there are 10 indispensable amino acids (1AA) for all
ﬁsh species known to date. Anchovy meal and SBM compositions of these amino acids
are compared in Table 2.4. SBM contains much lower levels of Met, which is often
considered the ﬁrst limiting 1AA in ﬁsh diets. SBM diets formulated for salmonids may
be deﬁcient in Met and thus require supplementation. While SBM appears to have a
fairly good 1AA composition, questions remain as to whether the essential amino acids
are sufﬁciently balanced to replace large amounts of ﬁsh meal. Nutritional studies have
shown that an imbalance of amino acids may affect feed utilization and growth of
Atlantic salmon. Berge et al. (1999) examined effects of LyszArg ratios and found that
lysine had both a stimulatory and inhibitory effect on the uptake of arginine on Atlantic
salmon, depending on the relative concentration of the two amino acids. In another study
on rainbow trout, Davies et al. (1997) concluded that the optimal LyszArg ratio is 1:1 for

SBM diets.

SBM Anti-nutritional Factors

Researchers have identiﬁed several anti-nutritional factors (ANFs) in SBM and
other plant meals which have shown to severely impair the health of carnivorous ﬁsh.
These researchers have observed negative impacts including poor growth, inﬂammation
of the cellular lining of the distal intestine (enteritis), mortality and disease (Dong et al.
2000, Storebakken et al. 2000, Krogdahl et al. 2003). ANFs of concern for Atlantic
salmon include protease inhibitors, non-starch polysaccharides, Oligosaccharides,

saponins, lectins, antigenic proteins, isoﬂavones and phytic acid.

16

Table 2.4 Essential amino acid content of Peruvian anchovy meal and soybean

meal.

 

 

Amino Acid Peruvian anchovy Soybean meal
meal (%) (%)
Arg 3.85 3.67
His 1.61 1.22
Ile 3.17 2.14
Leu 5.05 3.63
Lys 5.04 3.08
Met 1.99 0.68
Cys1 0.60 0.75
Phe 2.78 2.44
Tyr1 2.24 1.76
Thr 2.82 1.89
Trp 0.75 0.69
Val 3.50 2.55

1' Considered as semi-essential amino acids

Protease inhibitors

Protease inhibitors are proteins that inhibit proteolytic enzymes, or protease
activity, in the digestive track of animals (Krogdahl and Holm in Krogdahl et al. 1994).
Inhibition occurs when inhibitors bind with the enzymes forming compounds unavailable
for hydrolysis. The primary protease inhibitor of concern in SBM diets are trypsin
inhibitors (TIS). Trypsin inhibitors have been isolated in two forms: Kunitz inhibitors,

which inhibit mainly trypsin and is heat labile, and Bowman-Birk inhibitors, which

17

inhibit both trypsin and chymotrypsin (Liener 1980). In raw soybeans, TIs account for
about 6% of the protein content (Alexis and Nengas 2001), and 2-6 mg TI/g in
commercial soybean products (Snyder and Kwon 1987). Heat treatments in SBM
processing results in values of about 3.0-3.5 mg TI/g meal (Tacon et a1. 1983 in Hardy
2003). Further TI deactivation occurs in cooking-extrusion processes that occur during
diet manufacturing. Atlantic salmon have been Shown to tolerate up to 5 mg TI/g meal

(Krogdahl et al. 1994).

Non-starcleolvsacchaﬁdes

Non-starch polysaccharides (N SPs), also referred to as dietary ﬁber, form about
14-18% of the total carbohydrate content of defatted SBM (Alexis and Nengas 2001), and
up to 200 g/kg meal (Snyder and Kwon 1987). Insoluble NSPs such as cellulose and
hemicelluloses are structural polysaccharides that do not dissolve in water. Soluble NSPs
do not dissolve in water completely, but swell to form a gel in the presence of water
(NRC 2003). While ﬁber is important for nutrient passage of various omnivorous ﬁsh
species (e. g. catﬁsh), ﬁnﬁsh have no capacity to digest most ﬁbrous material (De Silva
and Anderson 1995, Roberts 2002).

Studies conducted on Atlantic salmon have identiﬁed NSPs as probable causes for
reduced lipid absorption and increased fecal water content (Reftsie et al. 2000, 2001).
According to Storebakken et al. (2000) NSPs likely impair difﬁision, convective transport
and/or micelle formation within the gastrointestinal tract of Atlantic salmon. These
structural carbohydrates are not readily destroyed by normal heat treatments or solvent

extraction. Fairly recently, scientists have been exploring the development of low NSP

18

soybeans for use in aquatic animal feeds through rapid detection methods such as infrared
spectroscopy and proteomics (Hollung et a1. 2005), and through genetic modiﬁcation

(Sanden et al. 2006).

Qﬂgsaccharides

SBM contains approximately 10-15% Oligosaccharides (sucrose, rafﬁnose and
stachyose, Dersjant-Li 2002, Alexis and Nengas 2001). Like NSPS, Oligosaccharides are
a carbohydrate fraction of SBM not digested by endogenous ﬁsh enzymes and thus
expected to be an energy loss to the ﬁsh (Alexis and Nengas 2001). Oligosaccharides
impair nutrient absorption, increase water content in feces, and may cause enteritis in
Atlantic salmon (Reﬁsie et al. 2000, Storebakken et al. 2000). Since Oligosaccharides are
alcohol soluble, they can be removed from SBM via alcohol solvent extraction. This
does not appear, however, to be a common industry practice at this time, perhaps because

of the added cost.

Saponins

Saponins are steroid or triterpene glycosides found in many plant-derived feed
ingredients and present in commercial SBM from 0.43-0.67% (Ireland et al. 1986). They
are bitter in taste and are highly toxic to ﬁsh in sufﬁcient doses. Bureau et a1. (1998)
identiﬁed soya saponins as a causative agent for reduced feed intake and intestinal
damage to Chinook salmon. Krogdahl et al. (1995 in Francis et al. 2001), however, did

not ﬁnd any negative effects of dietary saponins on Atlantic salmon up to 30-40% SBM

l9

equivalencies. Saponins are highly water soluble and can be removed rather easily by

aqueous extraction.

Lelia-S

Plant lectins are glycoproteins found in many legume seeds (Chrispeels and
Raikhel 1991). These compounds interfere with absorption and transport of nutrients by
binding with membrane receptors of carbohydrates (Tacon 1995 in Alexis and Nengas
2001). Lectins, also known as phytohaemagglutinins, or agglutinins, have been shown to
bind in vivo to the intestinal epithelium of Atlantic salmon and rainbow trout,
contributing to the pathological events in the distal intestine associated with ﬁsh feeds
containing high levels of SBM (Buttle et al. 2001). Since lectins, like trypsin inhibitors
are proteinic, they can be destabilized by heat treatment and are destroyed to large extent
in typical heating processes used to manufacture practical SBM-based aquaculture diets

(Alexis and Nengas 2001).

Antigenic proteins

Antigenic proteins in SBM include active immunoglobular compounds, glycinin
and IS-conglycinin (Alexis and Nengas 2001). Inclusion of soya products containing high
levels of these compounds reduced grth and caused enteritis in the distal intestine of
rainbow trout (Rumsey et al. 1994). These same compounds also stimulate non-speciﬁc
defense mechanisms in trout, but it is unknown if antigens in SBM result in higher
disease resistance (Rumsey et. al. 1995). Soybean meal antigens are heat stable and

alcohol soluble, and thus could be removed via alcohol solvent extraction.

20

Isoﬂavones

Soybean isoﬂavones are estrogenic compounds (phytoestrogens) both with and
without a sugar molecule attached (Eldridge and Kwolek 1983). According to You et al.
(2002), isoﬂavone concentration in raw soybeans ranges from 0.3 to 6.0 mg/ g. The two
primary isoﬂavones in soybeans are daidzein and genistein and their respective
glucosides, genistin and daidzin. Little is known about the effect of isoﬂavones on
sahnonids, although Mambrini et al. (1999), postulates they may be responsible for
reduced growth of rainbow trout fed a soy protein concentrate diet. Soy isoﬂavones are

heat stable and alcohol soluble, and thus could be removed via alcohol solvent extraction.

Phyaic acid

Phytic acid, or phytate, is a hexophosphate of myo-inositol and is common in soy
beans and other legumes (Francis et al. 2001). Phytic acid is problematic for most ﬁsh
species because they do not produce phytase, the enzyme required to hydrolyze phytate
(Alexis and Nengas 2001). Approximately 75% of the phosphorus in SBM is in the form
of phytic acid (Hardy 2003) and thus unavailable for digestive processes. Phytate can
chelate with mineral ions (Ca2+’ Mg”, Zn“, Cu3+ and Fe“) making these ions also
unavailable for use by consumers (NRC 1993). Sajjadi and Carter (2004) showed
signiﬁcantly reduced protein digestibility for Atlantic salmon fed ﬁsh meal diets with
addition of synthetic phytic acid. These same authors, and others, have postulated that

this effect was negated with addition of dietary phytase.

21

Dephytinized plant protein concentrates have resulted in good growth for Atlantic
salmon and rainbow trout in feed trials (Thiessen et al. 2004, Carter and Hauler 2000,
Brown et a1. 1997), but currently protein concentrates are too costly for large-scale
commercial use. Heat treatment and fermentation can also reduce phytate content in
meals and grains (Francis et al. 2001). Presently the level of SBM inclusion in diets for
carnivorous ﬁsh does not necessitate supplemental phytase, and phytic acid does not

appear to be a major factor limiting SBM usage.

Soybean meal-induced enteritis

Subacute enteritis of the distal intestine has been reported by several researchers
as a common side effect on Atlantic salmon and rainbow trout fed SBM diets (Bakke-
McKellep et al. 2000, Refstie et al. 2000, Krogdah12003). Observed pathological
changes include hypertrophic and hyperaemic mucosa, shortened secondary mucosal
folds, and loss of supranuclear vacuolization in epithelial cells (Beaverfjord and Krogdahl
1996). Those researchers also describe a widening of the lamina propria with inﬁltration
of a mixed layer of leucocyte populations. While the causative agent in SBM is
unknown, one or more of the alcohol soluble components is suspected since alcohol-
extracted soy protein concentrate did not induce enteritis in salmonids at concentrations
equal to SBM diets (011i and Krogdahl 1994). Similar symptoms have also been
observed with enteric disease of poultry and piglets (Morin et a1 1983, Dekich 1998),
which has been attributed to coccidiosis, feedbome toxins, bacteria and viruses. Since
problems observed with salmonids are believed to be of non-infectious origin

(Beaverfjord and Krogdahl 1996), feedbome toxins may be a common factor across

22

species. Examples of feedbome toxins include mycotoxins, which are molds and ﬁmgi
found in cereal grains and SBM, biogenic amines, and ingredient impurities (Dekich

1998).

Continued Development of SBM Diets for Atlantic salmon

Soybean meal is widely accepted as a standard ingredient in most commercial
aquaculture feeds. However, its use in commercial Atlantic salmon feeds likely remains
below 8-10% (Hardy 2003). Diets containing as little as 10% SBM resulted in moderate
intestinal histopathological changes in Atlantic salmon (Krogdahl et al. 2003). These
authors express that caution should be taken with use of even low levels of extracted
SBM in salmon feeds.

Extensive research has been designed to identify the causative agents in SBM that
negatively affect salmonids. Feed trials on salmon and trout have identiﬁed various
ANFS in SBM related to speciﬁc nutritional deﬁciencies. We now understand that heat
labile factors, such as trypsin inhibitors, lectins, and perhaps phytic acid, may be of less
concern because they can be destroyed or partially inactivated through meal and diet
extrusion fabrication processes (Alexis and Nengas 2001, Hardy 2003). According to
several authors, additional problems appear to be linked to dietary carbohydrate factions
(NSP and Oligosaccharides, Refstie et a1. 2000, 2001), and alcohol soluble components.
The later is supported by ﬁndings that alcohol extracted soya concentrate is of high
nutrition value to salmonids and can replace up to 50% of dietary ﬁsh meal (Olli and

Krogdahl 1994 in Krogdahl et al. 2003).

23

Improvements in feed processing technology have likely contributed to increased
usage of SBM in aquaculture feeds. Diet formulations and pellet extrusion methods have
improved substantially over the past 10-20 years resulting in very water stable, low
polluting commercial ﬁsh feeds (Todd Prowless, Ziegler Feed, personal communication).
Heat treatments and solvent extraction methods are physical processes which, through
advancements in research and technology, may be affordable means to overcome SBM
anti-nutritional properties in carnivorous ﬁsh. In the late 19903, high energy diets were
developed (Einen and Roem 1997), and made commercially available for salmon and
trout. These diets improved feed efﬁciencies of intensively grown Atlantic salmon
(Einen and Roem 1997, Azevedo et a1. 2002), and are becoming widely incorporated in
production cycles across the industry. Further research with high energy diets may
provide valuable information between SBM carbohydrate factions in ﬁsh diets and

protein sparing effects.

24

Chapter 3 -— Trypsin Inhibitor Effects on Atlantic Salmon (Salmo salar) Fingerlings

Introduction

Protease inhibitors are a class of proteins that react with speciﬁc proteolytic
enzymes in the digestive process of animals (Krogdahl and Holm in Krogdahl et al.
1994). Under controlled settings, feeds containing anti-nutritional factors such as
proteinase inhibitors have shown to effect grth rate and feed utilization of animals and
livestock. TIs are a class of proteinase inhibitors found in SBM, which is currently a
leading alternate protein source candidate for ﬁsh meal in formulated feeds used in
aquaculture (Hardy 2003, Alexis and Nengas 2001). Omnivorous ﬁsh species such as
carp, tilapia and catﬁsh have shown to accept rather high amounts of SBM in formulated
diets (ElSaidy and Gaber 2002, Jahan et a1. 2003). This has led to the development of
commercial SBM based diets for a number of cultured ﬁsh species and contributed to
reducing the global demand for expensive high quality ﬁsh meal. Ongoing research in
the area of ﬁsh nutrition continues to focus on SBM anti-nutritional properties for other
aquaculture species, particularly carnivorous species such as salmonids (Ollie et al. 1994,
Sveier et al. 2001, Krogdahl et al. 2003).

This study was part of an integrated research project funded by the United
Soybean Board, Soy-in-Aquaculture Program. The project was designed to develop a
commercially acceptable SBM based formulated feed for Atlantic salmon and other
species. The main objective of this study was to examine effects of soybean trypsin

inhibitors (SBTIs) on ﬁngerlings (young-of-year) Atlantic Salmon.

25

Materials and Methods
Experimental Diets

Five experimental diets containing graded levels of Us and one standard
commercial salmonid diet control were used to study effects of Us on Atlantic salmon
ﬁngerlings. The commercial control (Zeigler Brothers Finﬁsh Starter, Slow sinking)
contained a minimum of 50% crude protein (CP) and 19% crude fat (CF). Test diets
were formulated and manufactured by research collaborators at Purdue University under
the supervision of Dr. Steve Hart. Diets were formulated to contain the same CP and CF
levels as the control feed (Table 3.1). Crude protein content of commercial and
experimental diets were conﬁrmed using a Leco nitrogen/protein analyzer. Trypsin
inhibitor (Soybean Trypsin Inhibitor CAS #9035-81-8, USB Corporation) inclusion rates
were 0, 0.975, 1.950, 2.925, and 3.900 gTI/kg feed representing estimated SBM
equivalencies of 0, 15 , 30, 45, and 60% respectively. TI inclusion rate SBM
equivalencies were based on the average value of 6.5 mgTI/g SBM from the range of 5.0

—— 8.0 mgTI/g SBM (Russet 1998).

Experimental System and Animals

Young-of-the-year Atlantic salmon were obtained from the Michigan Department
of Natural Resources Lake Superior State University rearing facility in Sault Saint Marie,
Michigan in August, 2003. The ﬁsh were transported to Michigan State University’s
Aquaculture Research Laboratory and acclimated to water conditions in a 710-L ﬂow-

through tank culture system over a 30-day period. Fish were fed the commercial control

diet over the acclimation period.

26

Table 3.1 Formulations of semi-purified test diets containing graded levels of stock soybean trypsin
inhibitors in Miets are identiﬁed based on percent soybean meal equivalency.

 

 

 

 

Diet TIO T115 T|30 T145 TI6O

SBM Equivalency 0% 15% 30% 45% 60%
Egredient ' '
Esein 469 469 469 469 469
Gelatin 110 110 110 110 110
L-Methionine 5 5 5 5 5
Dextrin 130 1 30 1 30 1 30 1 30
a-Cellulose 10.5 9.525 8.55 7.575 6.6
Carboxymethylcellulose 20 20 20 20 20
Salmon Mineral Premix 33 33 33 33 33
Salmon Vitamin Premix 20 20 20 20 20
Ascorbic Acid (Stay C 35%) 3.5 3.5 3.5 3.5 3.5
Choline Chloride (74%) 4 4 4 4 4
Trypsin Inhibitor (TI) 0 0.975 1.95 2.925 3.9
Lecithin 5 5 5 5 5
Menhaden Oil (500 ppm ethoxy) 190 190 190 190 190

 

1000.0 1000.0 1000.0 1000.0 1000.0

Note: Commercial control was Zeigler Brothers Finﬁsh Starter, slow sinking.

A total of 450 ﬁsh were randomly distributed in eighteen, 110-L tanks, at 25 ﬁsh
per tank. Fish were acclimated to feed trial rearing units and fed the commercial control
diet for an additional 30—day period. Flow rates were maintained between 2.8—3.6 me
fresh well water based on target exchange rates of 1.5—2.0 water exchanges per hour.
Water temperature for all tanks remained between 11.7—12.2 0C for the duration of the
study with one potential exception on day-2 when the well went down for approximately
2 hours due to power failure. Dissolved oxygen varied from 8.9 to 9.1 mg/l; total
ammonia nitrogen concentrations remained below 1.0 mg/L (0.006 mg/L unionized
ammonia). All other water quality parameters fell within acceptable limits for salmonids

(Piper et al. 1982).

27

During the acclimation period several ﬁsh showed signs of bacterial infection.
Fish samples were sent to MSU’S Aquatic Animal Health Laboratory for diagnosis.
Isolates indicated presence of F exibacter columnaris suggesting moderate chronic
bacterial infection. All ﬁsh received a 10ppt salt bath for approximately 30 minutes to
treat the infection. This treatment was continued on a prophylactic basis once per week
through the end of the feed trial.

At the end of the rearing unit acclimation period, 3—5 ﬁsh were randomly selected
and removed from each tank. Of those removed, 15 random ﬁsh were pooled, euthanized
in tricaine methanesulfonate (MS-222) at a concentration of 500 mg/L (AVMA 2000),
frozen, and held at -20°C for subsequent whole body composition analysis prior to
starting the feed trial. Weight and length data were recorded for the sample of 15. The
feed trial was initiated with 20 ﬁsh in each of the 18 tanks, 360 total. Total weights of all
ﬁsh from each tank were recorded at the start of the trial. The average weight of ﬁsh per
tank was 17.5 :t 1.4 grams.

Triplicate tanks of ﬁsh were fed, either the commercial control diet, or one of ﬁve
treatment diets, three times daily (8:00—9:00 am, 12:00—1 :00 pm, 4:30—5:30 pm) for eight
weeks. Total weights of ﬁsh in each tank were recorded every 2 weeks. Feed levels
were maintained at a constant percent body weight (%BW), and were adjusted (by
weight) based on recorded weight samples. For the trial, %BW was calculated as 90% of
the theoretical optimal feed level for salmonids (Westers 1987). Feed levels fell both
above and below satiation levels of the ﬁsh across feeding times based on observations of

excess feed in tank bottoms at various times through the feed trial.

28

Mortalities were removed on a daily basis and weights of dead ﬁsh were recorded.
Eighteen mortalities (5%) were observed over the course of the feed trial. At the end of
the trial, total weight samples were recorded, and 10 ﬁsh per tank were selected at
random for length and weight measurements. Five ﬁsh pooled from each tank were
randomly selected, euthanized in MS-222 at a concentration of 500 mg/L (AVMA 2000),

and frozen as a group sample at -20°C for future analysis.

Sample Analysis and Calculations
Feeding levels
Trial feed levels were calculated as 90% of the theoretical optimal percent body
weight (%BW) for sahnonids as developed by Westers (1987):
%BW = (300 x TUGs x FCR1,o)x #TU/(W/1<,-)“3 x 0.90
TUG = (Li — h)/(°C x d)
where:
TUG = Temperature Unit Growth Rate,
TUGS = Theoretical TUG for Salrnonids = 0.006 (cm/OC/d),
FRCLO = Theoretical feed conversion rate of 1.0,
#TU = Number of thermal units in 0C above 00°C (0C),
W = Wet weight of ﬁsh at time of sampling (g),
k, = condition factor at the start of the trial,
LF = average ﬁnal length (cm),
L, = average initial length,

0C = temperature in Celsius,

29

d = time (days).

mm

Weight and length data were used to determine condition factor (k) and speciﬁc
growth rate (SGR) over the course of the study:

k = W/L3

SGR = (1n Wf— ani)/d x 100
where:

W = wet weight (g) of ﬁsh at time of sampling,

L = length (cm),

Wfori = average ﬁnal or initial wet weight (g), and d = time (days).

Feed conversion

Feed conversion rates were calculated as the standard apparent FCR with
adjustment for mortalities:

FCR = Cum F eed/(Net W Gained + (WMons - (NMorts X Wt»)
where:

Cum Feed = cumulative weight of feed (g),

WMom = Wet weight of mortalities removed (g),

NMons = Number of mortalities removed.

30

Whole body composition

Whole body composition analysis included lipid, protein and ash. Pooled, frozen
whole body samples were thawed quickly under cool water and homogenized in a
commercial-grade food processor. The samples were then weighed and stored at -20°C
until lyophilization. All freeze-dried samples were ﬁnely ground and homogenized again
and stored at -20°C. Whole body lipid content was determined on duplicate 1.0—3.0 g
samples by lipid extraction with diethyl ether (Soxtec System HT/ 1043 Extraction Unit,
Tecator, Sweden). Nitrogen was determined on 0.5 g samples according to combustion
method AOAC (2000) using a Leco nitrogen/protein analyzer (model FP-2000, Leco
Corp., St. Joseph, MI). Crude protein was calculated as N x 6.25. Dry matter was
obtained after oven drying 2.0 g samples at 105°C for 18-24 hours. Ash content was
determined after placing dry matter samples in muffle fumace at 500°C for 18 hours. All

body composition samples were performed in duplicate.

Protein efﬁciency and retention

Protein efﬁciency ratios were based on the formula provided by De Silva and
Anderson (1995) slightly modiﬁed to account for mortality:

PER = (Net W Gained + (WMcms — (N Mons x Wi))/ TPFeed

TPpecd = Cum Feed x % PFecd / 100

APR(%) = FPG/ PIFecd x 100

FPG = (fo Pg’lOO x FDp’lOO) - (W, x Pi/10O x FDi/IOO)

PIFeed = (Cum Feed x % PFeed / 100) / #FIShf

31

where:
PER = protein efﬁciency ratio,
TPpeed = total protein in feed (%),
APR(%) = apparent protein retention (%),
FPG = ﬁsh protein gain (gP),
PIFccd = Protein intake from feed (gP/ﬁsh),
% Freed = % protein in diet as fed basis (%P/gfeed),
Pro” = Final or initial % protein in body compositions of freeze-dried sample
(%P/gFD).
FDfori = Final or initial % ﬁ'eeze-dried matter of ﬁsh samples (%),

#FIShf = Number of ﬁsh alive at end of study.

Statistical Analysis

Data were analyzed by one-way analysis of variance (ANOVA) using SPSS
(SPSS© Release 12.0, 2003) and SAS (SAS© Release 9.1, 2003) statistical software.
Homogeneity of variance was conﬁrmed by Levene's statistic analysis. Signiﬁcant
differences between means were compared by Duncan’s multiple range test. Trend
analyses were conducted using regression analysis and orthogonal contrasts. Treatment

effects were considered signiﬁcant at P S 0.05 unless otherwise noted.

32

Results
Growth Characteristics

Slight differences were observed in growth characteristics among dietary
treatment groups (Table 3.2). Mean ﬁnal weight for ﬁsh fed the control diet was lower
than all treatment groups except TIO (P5005). The SGR value of 1.44 for the
commercial control group was lower than those obtained from ﬁsh fed experimental diets
which ranged between 1.58 (T10) and 1.69 (T145). There were no signiﬁcant differences
found in growth rates among treatment diets (Figure 3.1) Only the T115 group had a

higher condition factor, and k for all groups ranged between 0.0097 and 0.0105.

Feed Conversion and Protein Retention
No differences were detected in mean FCRS, PERs or APRs (Table 3.2). F CR
values ranged between 0.74 and 0.83 (T145 and T160 respectively), PER from 2.40

(Control) to 2.67 (T145), and APR from 32.94 (T130) to 35.98 (T115).

Proximate Body Composition

Slight differences were observed in proximate body compositions across diets.
Overall, ﬁsh fed the commercial diet had a greater percentage of protein and less fat
(P5005) than ﬁsh fed semi-puriﬁed diets (Table 3.3). Whole body crude protein content
was greatest for Atlantic sahnon fed the control (52.94%) and least for ﬁsh fed TIO

(48.29%). Crude protein compositions were statistically insigniﬁcant across all TI levels

tested.

33

Table 3.2 Mean ﬁnal weight, speciﬁc growth rate (SGR), condition factor (k), feed conversion rate
(FCR), protein efﬁciency ratio (PER), and apparent protein retention (APR) in ﬁngerling Atlantic
salmon fed diets containing graded levels of trypsin inhibitors (TI) at 0, 15, 30, 45, and 60% soybean
meal equivalencies. Mean standard errors are in parentheses. Values in each row awarded common
superscripts are not signiﬁcantly different (P > 0.05).

 

 

Control no r115 1130 1145 1160

Final weight (g) 39.55a 41.28 a” 44.49 b 44.16 b 43.59 b 44.35 b
(1.25) (1.64) (0.41) (1.17) (0.81) (1.21)

sea (%Id) 1.44 a 1.58 b 1.64 b 1.59 b 1.69 b 1.63 b
(0.04) (0.02) (0.01) (0.06) (0.03) (0.01)

k 0.0101 3" 0.0100ab 0.0105 8 0.0098 ” 0.0099b 0.0097 "
(0.0001) (0.0002) (0.0001) (0.0002) (0.0001) (0.0002)

FCR 0.82 0.78 0.75 0.78 0.74 0.83
(0.01) (0.01) (0.01) (0.05) (0.04) (0.09)

PER 2.40 2.58 2.64 2.57 2.67 2.49
(0.04) (0.04) (0.04) (0.16) (0.15) (0.25)

APR (%) 35.71 34.71 35.98 32.94 34.62 33.36
(0.44) (0.93) (0.59) (2.09) (1.91) (2.86)

34

 

1.75

1.70%
1.65- g E

1.55-

SGR (%Iday)
'8
1
HI]

8
l

1.45-

1.40-

1.35-

 

 

 

l I I I F I
Control TIO TI15 TI3O TI45 TI60

Diet

Figure 3.] Box plot of mean speciﬁc growth rate of ﬁngerling Atlantic salmon fed a commercial
control and experimental diets containing graded levels of stock trypsin inhibitors (T1) at 0, 15, 30,
45, and 60% soybean meal equivalencies.

35

Table 3.3 Percent body compositions of ﬁngerling Atlantic salmon fed a commercial control and
experimental diets containing graded levels of trypsin inhibitors (TI) at 0, 15, 30, 45, and 60%
soybean meal equivalencies. Mean standard errors are in parentheses. Values in each row awarded
common superscripts are not signiﬁcantly different (P > 0.05).

 

 

 

 

 

 

Control TIO T115 7130 1145 TI60
Dry matter 28.57 a” 29.67 a 29.58 a 27.54 b 28.11 b 28.74 a”
(0.18) (0.49) (0.29) (0.4) (0.47) (0.42)
Protein 52.94 a 48.29 b 48.20 b 50.37 b 49.20 b 48.58 b
(0.44) (1.07) (0.48) (1.21) (0.64) (0.6)
Fat 15.12 a 19.30 b 27.78 ° 23.74 “ 26.10 °“ 18.62 a”
(0.75) (0.75) (1.49) (1.53) (1.35) (0.88)
Ash 8.28 a 8.58 a 8.62 a 9.61 b 9.35 b 9.54 b
(0.15) (0.13) (0.05) (0.3) (0.2) (0.3)
30‘
25-i
E
a
u.
.5- E
l I I l l i
Control TIO T115 Tl30 Tl45 Tl60
Diet

Figure 3.2 Box plot of mean body fat composition of ﬁngerling Atlantic salmon fed a commercial
control and experimental diets containing graded levels of stock trypsin inhibitors (TI) at 0, 15, 30,
45, and 60% soybean meal equivalencies.

36

Differences in body fat content between dietary treatments were more pronounced
(Table 3.3 and Figure 3.2). Whole body crude fat content for the T115 group (27.78%)
was 12.66% higher than the cormnercial control group (15.12%). Atlantic salmon fed
T115, T130, T145 diets had 5 to 8% higher fat content than those fed TIO or T1 60.
Orthogonal contrasts showed a statistically signiﬁcant (P = 0.0003) quadratic trend for
SBTIS on body CF compositions.

Fish fed 30% SBM equivalency SBTI diets or greater had higher body ash
compositions (P S 0.05) than those fed diets containing 0 or 15% SBM equivalency.
Slight differences were observed in percent whole body dry matter, which ranged from
27.54 (T130) to 29.67 (T10). Differences detected in dry matter composition do not

appear to be attributed to treatment effects.

Mortalities

Mortalities occurring over the course of the experiment were highest for ﬁsh fed
TI60 (Figure 3.3). Mean mortality rates, however, were not signiﬁcantly different across
treatment levels, but variation within treatments was relatively high ranging from zero to
ﬁve ﬁsh per tank. Five of the six mortalities occurring in the T160 group came from a
single tank. Each treatment had at least one tank that experienced no deaths. Mortalities
increased as the trial progressed across all treatments: two mortalities were observed. in

weeks 1-3 of the trial, four observed in weeks 4-6, and thirteen in weeks 7-9.

37

 

 

: I

2 2 3 1 2 3
C 3Tl15 3T130 T145 T160
Tank and Diet

Figure 3.3 Number of mortalities by treatment (diet) and replicate (tank) of ﬁngerling Atlantic
salmon over the 8-week feed trial. Diet C is a commercial control. TI# represents percent soybean
meal equivalency (0, 15, 30, 45 and 60%) of dietary trypsin inhibitor (TI).

Discussion
Experimental Diets

The experimental diets manufactured by Purdue University were similar in
formulations to those used in a number of ﬁsh nutritional studies conducted at Purdue
University (Dr. Steve Hart, Purdue University, personal communication). All
ingredients, with exception of menhaden oil, were purchased in puriﬁed forms, and all
ingredients were obtained from reputable manufacturers. The main protein source of
experimental diets was ultra pure, vitamin free casein purchased from the USB chemical.
Study vitamin and mineral premixes are typically > 99% pure. Based on study results, no

evidence indicated that any of the ingredients, other than puriﬁed TIS, contained ANFS or

38

other impurities that potentially affected the outcome of the feed trial. This is in part
conﬁrmed by the inclusion of the commercial control and the results described within the

following section.

Growth Characteristics

Atlantic salmon fed the semi-puriﬁed test diets containing SBTIS grew faster (P S
0.05) than ﬁsh fed the commercial control feed. The commercial control diet was
obtained from a well known commercial feed supplier and considered by many to be of
good to high quality. Reduced growth for ﬁsh fed the commercial control diet is likely
due to differences in ingredient selection and processing methods.

Based on SGR alone, puriﬁed HS in diets of young-of-year Atlantic salmon had
no affect on growth to 60% SBM equivalency (3.9 gTI/kg feed). This ﬁnding differs
slightly with results by Olli et al. (1994) who reported that Atlantic salmon were able to
compensate for T1 levels up to 30% SBM equivalency (2.08 gTI/kg feed), but had
reduced growth at 48% SBM equivalent (3.15 gTI/kg feed). According to those
researchers TI compensation by Atlantic salmon appeared to be accomplished through
increased trypsin secretion, which eventually is exhausted at elevated SBTI activities.
Examination of Figure 3.1 may suggest an asymptotic treatment effect of TI on Atlantic
salmon growth; however, no statistical differences were detected in SGR across test diets.
Increased variability in the T130 group growth data could not be explained by study
results, but could be due to treatment effects, random effects and/or sample error.

No effects of SBTIS were observed on ﬁsh condition factor. All k values fell

within 0.005 of the value obtained from ﬁsh fed the control diet (k = 0.01). A k value of

39

0.01 is considered average for most salmonid species (Harry Westers, Aquaculture

Bioengineering Corporation, personal communication).

Feed Conversion and Protein Retention

No differences were detected in FCR across diets. In addition there were no
observable effects of Us on feed intake, protein efﬁciency or protein retention. Based
on these results, it appears that protein utilization by Atlantic salmon was unaffected by

T1s up to 60% SBM equivalency.

Proximate Body Composition

Whole body compositions showed differences across all categories (Table 3.3).
Most notably, Atlantic salmon fed the commercial control diet had more crude protein
and less crude fat compared to ﬁsh fed semi-puriﬁed test diets. As previously stated, this
difference is most likely due to differences in diet processing methods and ingredient
digestibility.

Fish protein levels remained relatively unchanged across dietary treatments.
This is in slight contrast to two related studies on salmonids. Krogdahl et a1. (1994)
observed minor effects of puriﬁed SBTIS on protein utilization by rainbow trout fed feed
containing 57% SBM equivalence (3.7 gTI/kg feed). Olli et al. (1994) reported observing
similar effects at 48% SBM equivalence (3.15 gTI/kg feed) with Atlantic salmon.

Crude fat content of T11 5 and T145 groups were approximately 8% and 12%
higher than the T160 and control group respectively. While crude protein content

remained relatively constant over dietary treatments, TI levels appeared to affect lipid

40

digestibility and/or accretion (Table 3.3; Figure 3.2). This observation agrees with
ﬁndings by several researchers examining the effects of puriﬁed trypsin inhibitors (Olli et
al. 1994, Krogdahl et al. 1994) and of SBM practical diets on salmonids (Ref’tsie et al.
2001, Storebakken et a1. 1998). Results from this study indicated a statistically
signiﬁcant (P = 0.0003) quadratic trend for SBTI level on CF. Body lipid content
increased from 19% CF at 0%SBTI to a peak of about 27% CF between 15 and 45%
SBTI, and then decreased to 18% for the highest SBTI (TI60) group. The peak observed
between 15 and 45% SBM equivalency, then, is likely a compensatory response by
Atlantic salmon to SBTIS. In the presence of the mid level amounts of SBTIS, metabolic
pathways appeared to favor body lipid deposition. This suggests that internal
mechanisms favored storage of additional energy reserves at mid-level SBTI’s. The
sharp decline at 60% SBM equivalency further suggests that more energy in the form of
body lipids was required at high levels of SBTI in order to maintain similar grth and
protein deposition.

Whole body ash content of ﬁsh from T130, T145 and TI60 groups were greater
than the control, T10 and T115 groups (P S 0.05). Thus, while grth remained constant
across treatments, internal physiological characteristics including whole body ash and

lipid compositions were affected by dietary levels of Tls.

Mortalities
All of the 18 mortalities occurring over the course of the feed trial showed similar

external signs of bacterial infections on the body and ﬁns. One-way ANOVA indicated

41

no signiﬁcant differences in mortality frequency across diets, but high variation within
treatments.

Typically, wild North American populations of Atlantic salmon remain in fresh
water as parr until an age of approximately 2-3 years when they undergo the
physiological changes associated with smolting (Danie et al. 1984). Thus, potential
stressors that may have contributed to mortalities would likely be more associated with a
par stage (pre-smolt) rather than stressors associated with the smolting process.
Examples of possible contributing factors include genetics, conﬁnement, handling stress

and/or presence of opportunistic bacteria.

Chapter Summary

Based on results from this feed trial, Atlantic salmon appeared to undergo
physiological changes in response to increasing dietary TIs. While speciﬁc deﬁciencies
could not be attributed to statistical signiﬁcance over measured parameters, slight
changes were observed in response of body fat and ash compositions to SBTIS. Speciﬁc
mechanisms for these changes are currently unknown. In this study Atlantic salmon were
able to consume up to 60% SBM equivalency SBTI without affecting grth rates, while
a comparable study by Ollie et al. (1994) showed depressed growth at a level of 48%
SBM equivalency. Both studies agree that Atlantic salmon appear to have a certain
capacity to compensate for SBTIS in their diets. Results from this study support the
hypothesis by Ollie et al. (1994) that TI compensation by Atlantic salmon is likely due to

increase trypsin production and secretion in the pancreas. Further research is required to

42

assess the pathways leading to these responses, and whether the changes are detrimental

to the health of the ﬁsh.

43

Chapter 4 — Trypsin Inhibitor Effects on Atlantic Salmon (Salmo salar) Smolts

Introduction

This study was part of an integrated research project funded by the United
Soybean Board, Soy-in-Aquaculture Program. The project was designed to develop a
commercially acceptable SBM based formulated feed for Atlantic salmon and other
species. Speciﬁcally, this study focused on short term physiological effects of soybean
trypsin inhibitors (SBTIS) on Atlantic salmon smolts (age-1+). Puriﬁed SBTIS were
added to standardized semi-puriﬁed diets at graded levels from 0—60% SBM dietary
inclusion equivalencies. Atlantic salmon intestinal digesta dry matter, trypsin enzyme
activity, apparent protein digestion indices and body lipids were evaluated over a 3-week

feed trial.

Materials and Methods
Experimental Diets

Five experimental diets containing graded levels of TI and one standard salmonid
diet control were used to study effects of trypsin inhibitors on Atlantic salmon smolts.
The commercial control (Silver Cup Extruded Salmon Feed, slow sinking) was reported
to contain a minimum of 45% crude protein and 19% crude fat. Protein content was
conﬁrmed and measured to be 47.3%. Test diets were formulated to contain 50% crude
protein 19% crude fat (Table 4.1), and manufactured by research collaborators at Purdue
University under the supervision of Dr. Steve Hart. Measured values for crude protein

were 49:1: 1%. Trypsin inhibitor (Soybean Trypsin Inhibitor CAS #9035-81-8, USB

44

Table 4.1 Formulations of semi-puriﬁed test diets containing graded levels of stock soybean trypsin
inhibitors (TD in glkg. Diets are identiﬁed based on percent soybean meal equivalengy.

 

 

 

 

Diet TIO T|15 Tl30 Tl45 Tl60
SBM Equivalency 0% 15% 30% 45% 60%

ﬂredient
Casein 425 425 425 425 425
Gelatin 80 80 80 80 80
L-Methionine 5 5 5 5 5
L-Arginine-HCL 5 5 5 5 5
Dextrin 1 30 130 1 30 130 1 30
ct -Cellulose 74.5 73.525 72.55 71.575 70.6
Carboxymethylcellulose 20 20 20 20 20
Salmon Mineral Premix 33 33 33 33 33
Salmon Vitamin Premix 20 20 20 20 20
Ascorbic Acid (Stay C 35%) 3.5 3.5 3.5 3.5 3.5
Choline Chloride (74%) 4 4 4 4 4
Trypsin Inhibitor (TI) 0 0.975 1.95 2.925 3.9
Lecithin 5 5 5 5 5
Menhaden Oil (500 ppm ethoxy) 190 190 190 190 190
Chromic Oxide 5 5 5 5 5

 

1000.0 1000.0 1000.0 1000.0 1000.0

Note: Commercial control was Silver Cup Extruded Salmon Feed, slow sinking.

Corporation) inclusion rates were 0, 0.975, 1.95, 2.925, and 3.9 gTI/kg feed representing
estimated SBM equivalencies of 0, 15, 30, 45, and 60% SBM, respectively. TI inclusion
rate SBM equivalencies were based on the average value of 6.5 mgTI/g SBM from the
range of 5.0—8.0 mgTI/g SBM (Russett 1998). Test diets also contained 0.5% Chromic

oxide as an inert marker for apparent digestibility analysis.

Experimental System and Animals

One-year old Atlantic salmon were obtained from the Michigan Department of

Natural Resources Lake Superior State University rearing facility in Sault Saint Marie,

45

Michigan in August, 2003. The ﬁsh were transported to Michigan State University’s
Aquaculture Research Laboratory and acclimated to water conditions in a 1,890-L ﬂow-
through tank culture system over a 30-day period. Fish were fed the commercial control
diet over the acclimation period. A total of 576 ﬁsh were randomly distributed in
eighteen, 110-L tanks, at 32 ﬁsh per tank, and acclimated to feed trial rearing units for an
additional 30-day period. Fish were fed the commercial control diet over the acclimation
period. Flow rates were maintained between 2.8—3.6 me well water based on target
exchange rates of 1.5—2.0 water exchanges per hour. Water temperature for all tanks
remained between 11.8—11.9 0C for the duration of the study with one potential exception
on day-3 when the supply well was out-of-service for approximately 2 hours due to a
power failure. Dissolved oxygen varied from 8.6 to 9.4 mg/L; total ammonia nitrogen
concentrations remained below 1.3 mg/L (0.009 mg/L unionized ammonia). All other
water quality parameters were within acceptable limits for salmonids (Piper et al. 1982).

During the acclimation period, a few ﬁsh showed signs of bacterial infection.
Fish samples were sent to MSU’s Aquatic Animal Health Laboratory for diagnosis.
Isolates indicated presence of F exibacter columnaris suggesting a moderate chronic
infection. All ﬁsh received a 10 ppt salt bath for approximately 30 minutes to treat the
infection. This treatment was continued on a prophylactic basis once per week through
the end of the feed trial. Two mortalities occurred over the course of the 21-day feed
trial.

At the end of the rearing unit acclimation period, 1-2 ﬁsh were randomly removed
from each tank in order to start the feed trial with 30 ﬁsh per tank, 540 ﬁsh total. Whole

body samples from 15 ﬁsh (randomly selected from those removed) were euthanized in

46

MS-222 at a concentration of 500 mg/L (AVMA 2000), frozen and held at -200C for
subsequent analysis. Initial total weights were obtained for each tank on day-one of the
feed trial. Average weight per ﬁsh in each tank was 89.2 5: 4.1g. Fish were fed three
times daily (8:00—9:00 am, 12:00—1:00 pm, 4:30—5:30 pm), at a feed rate of 1.1% initial
body weight per day for a period of 21 days.

On days 10 and 21 of the feed trial, 10 ﬁsh were randomly selected from each
tank and euthanized in MS-222 at a concentration of 500 mg/L (AVMA 2000). Digesta
were collected separately from portions of the small intestine, proximal large intestine,
and fecal samples were collected from the distal portion of the large intestine. Samples
were pooled within tank replicates, freeze-dried (Tri-Philizer MP, FTS systems, for 48
hours minimum), and stored frozen until analyzed. At the termination of the feed trial,
three ﬁsh from each tank were randomly selected and pooled into one sample for
evaluation of proximate body composition. Euthanized ﬁsh were frozen and stored at

-200C for further analysis.

Sample Analysis and Calculations
Di gastja dry matter
Digesta and fecal lyophilized dry matter content were determined from samples

obtained from the small intestine and proximal and distal (fecal) regions of the large

intestine.

47

Trypsin activity

Intestinal and fecal trypsin activities were determined colorimetrically using
BAPA (benzoyl-DL-arginine-p-nitranilide) as a substrate as described by Kakade et al.
(1969) with slight modiﬁcation. Lyophilized 50.0 mg digesta samples of small intestine
and proximal and distal (fecal) large intestine were extracted with 5.0 ml 0.01 N NaOH
for 3 hours with periodic mixing. Supernatant 1.0 ml samples were centriﬁrged at 4000
rpm for 10 minutes. Tris buffer solution contained 0.60 5 g Tris, 0.295 g CaClz, 100 ml
distilled water (DI), pH adjusted to 8.2 using 1.0N HCL. The BAPA solution was
prepared by dissolving 0.04 g BAPA completely in 1.0 ml dimethyl sulfoxide, and 100
ml Tris buffer (solution stable for 4 hours). Trypsin solution was prepared by mixing 4.0
mg trypsin (Bovine 2x Lyo, TRL3 702, Worthington Biochemical Corp) with 100 m1
0.001M HCL (solution stable for 30 days at 40C). Absorbance readings were analyzed
bichromatically by pipetting 5.0 pl extract in triplicate to a microreader plate, adding 35
pl D1, 20 p1 acetic acid to blank (ﬁrst cell), and adding 140 pl BAPA to all cells. The
trypsin standard curve was determined similarly by placing 10, 20, 30 and 40 pl trypsin
in triplicate to a microreader plate, adjusting each cell to 40 pl with DI, adding 20 pl
acetic acid to blank (ﬁrst cell), and by adding 140 pl BAPA to all cells. The plate was
inserted into the microplate spectrophotometer (Molecular Devices Spectramax 340,
Sunnyvale, CA 94089), and warmed to 370C for 10 minutes. The reaction was
terminated after 10 minutes with 20 pl acetic acid. Absorbance was read at 410 nm.
Trypsin activity in mg/ g fecal sample was obtained from the trypsin standard curve.

Trypsin activity (TA) in mg trypsin per gram intestinal fecal material was calculated as

follows:

48

TA (mg/g saInple) = Try (pg/v01) / wt (8)
Try (pg/vol) = Try (pg/ml from standard curve) x EVol (ml)

where:

Try = trypsin value; EVol = extract volume.

Apparent protein digestibim

Apparent protein digestibility was measured indirectly using Chromic oxide as an
inert marker. Chromic oxide levels in feces were obtained by digestion and atomic
absorption on 10 mg samples (Williams et al. 1962). Diet and fecal nitrogen was
determined on 50 mg samples according to combustion method AOAC (2000) using a
Leco nitrogen/protein deterrninator (model FP-2000, Leco Corp., St. Joseph, MI).
Protein was calculated as N x 6.25. Apparent protein digestibility coefﬁcient (APD) was
calculated as follows (Cho and Slinger 1979, NRC 1993):

APD = IOOX[I—(PF/PD)X(CTD/CI'F)]
where:

PF = % protein of feces,

PD = % protein of diet,

CrD = % marker (Cr203) in diet,

Crp = % marker (0203) in feces.

Whole body composition
Due to the short duration of the feed trial, whole body composition was not

expected to differ greatly between treatments. Based on subsequent related research,

49

whole body lipid appears to be a sensitive parameter to SBM anti-nutritional properties.
For this reason lipid analysis was used as a benchmark to assess whether additional body
composition data might be warranted. Pooled, frozen whole body samples were thawed
quickly under cool water and homogenized in a commercial-grade food processor.
Samples were weighed, frozen and freeze-dried. All freeze-dried samples were ﬁnely
ground and homogenized again and stored at -20°C. Whole body crude lipid content was
determined on duplicate 1.0 g samples by lipid extraction with diethyl ether (Soxtec

System HT/ 1043 Extraction Unit, Tecator, Sweden).

Statistical Analysis

Data were analyzed by one-way analysis of variance (ANOVA) using SPSS
(SPSS© Release 12.0, 2003) and SAS (SAS© Release 9.1, 2003) statistical software.
Homogeneity of variance was conﬁrmed by Levene's statistic analysis. Signiﬁcant
differences between means were compared by Duncan’s multiple range test. Trend
analyses were conducted using regression analysis and orthogonal contrasts. Lyophilized
dry matter (DM) day-10 small intestinal and fecal samples, and day-21 fecal samples
tested statistically signiﬁcant in homogeneity of variance and were compared using a
Games-Howell test. All treatment effects were considered signiﬁcant at P S 0.05 unless

otherwise noted.

Results
Digesta Dry Matter

Dry matter of digesta material sampled from the small intestine (SI), proximal

large intestine (LI), and distal large intestine (fecal; F) regions are provided in Table 4.2.

50

Generally, the control diet resulted in slightly higher DM contents than the semi-puriﬁed
diets. This difference was signiﬁcant in three of six cases (P S 0.05). Regarding
treatment diets, the T145 group had less DM than all other test diets including TI60. The
only other differences noted were from day-21 SI samples where T10 and T115 were

higher in DM than T130, T145 and TI60.

T rypsin Activity

Digesta and fecal TAs varied between intestinal sections, day of sampling, and
between diets to some degree (Figures 4.1, 4.2). Overall, mean TA was higher in the SI
than the L1 and F regions of the large intestine.

Trypsin activity in the SI was higher for the control group than semi-puriﬁed diets
(P _<_ 0.05), and higher for ﬁsh fed TIO than those receiving dietary SBTIS (Figure 4.1 A).
No differences in TA were observed from S1 samples between test diet groups T115
through T160 groups; however, there was a negative linear response of TA to SBTI on
day-10 in the SI (linear, P = 0.002, R=0.75).

T rypsin activity in the L1 on day-10 showed slight ﬂuctuations (Figure 4.1B).
Trend analysis indicated a statistically signiﬁcant negative linear trend of SBTI on TA
(linear, P = 0.02), although a linear regression showed low correlation (R = 0.56) of this
ﬁt. No other orthogonal contrasts were signiﬁcant. Fish fed the control diet, T10 and
T115 had higher TA levels than those fed T145. In addition, TA from ﬁsh fed T115 was
higher than those fed 30-60% SBM TI equivalency (P S 0.05). No differences were

observed in day-10 fecal TA.

51

Table 4.2 Dietary trypsin inhibitor (TI) affects on percent dry matter (lyophilized) of intestinal

digesta from sampled regions of small intestine (SI), proximal large intestine (LI) and distal large
intestine (F) on day-10 and day-21. Diet C is the commercial control. TI# represents the percent
soybean meal equivalency (0, 15, 30, 45 and 60%) of puriﬁed T1 in test diets. Values in each row
awarded common superscripts are not signiﬁcantly different (P > 0.05).

 

 

Day Sample c TIO 1115 r130 7145 T160

10 SI 29.3 ' 24.4 ' 20.6 ' 20.5 ' 17.6 b 20.0 '
Ll 26.0 ' 17.0 ° 13.7 ° 17.9 b 15.5 ° 16.3 "
F 26.2 16.5 12.8 15.1 14.6 13.2

21 st 25.5 ' 24.3 ' 24.1 ' 16.3 b 13.4 b 16.9 °
Ll 23.7 ' 17.7 " 16.5 b 14.0 b 13.5 " 14.6 b
F 20.1 ' 13.3 b 12.6 ° 13.3 ° 12.8 b 11.4 "

Day-21 TA values (Figure 4.2) differed slightly to those observed for day-10. SI
samples on day 21 (Figure 4.2A) again had higher TA than all test diet groups, but the
negative response observed in Figure 4.1A is no longer present. Also no differences were
detected in day-21 LI TA samples. Fecal TA for the T10 group was higher than that

observed for T130 (P S 0.05), but no other differences were detected.

Apparent Protein Digestibility

Apparent protein digestibility indices and cumulative percent protein digestion
occurring through the SI, LI, and F regions are shown in Figures 4.3 and 4.4. No
differences were detected in total apparent protein digestibility coefﬁcients (Table 4.3)
across treatments. No differences were detected in cumulative APD across diets for

individual segments of the digestive tract from which samples were taken (SI, LI, F).

52

Trypsin Activity (mglg)

Trypsin Activity (mg/g)

Trypsin Actlvlty (ma/9)

 

 

 

 

3 SI
“-0 § b y = -0.038x + 4.13
l I R2 = 0.56
‘00 ’l _ ..._° P(linear) = 0.002
0.0 L» - —- 4* —ﬂ————— . . ——
c 110 1115 1130 1145 TI60
Diet
8.0 1 B
I Ll
6.0
i a y = -0.038x + 4.05
J R2 = 0.32
4.0 L? iii-‘9... ,,,,,,, . P(linear) = 0.027
1 '
bc
2.0 l .....
l
o o “T _—_T—T T —— T— I— ——’—_"'—T"_ —- —_'T—T __—'_-—TT/ * l
c TIO Tl15 n30 Tl45 T160
Diet
8.0 .
~ C
6.0 , F
4.0
2.0 4
0.0 l L_____ —.————————— T- — ~ . e _. ..
c TIO n15 1130 1145 TI60
Diet

Figure 4.1 Day-10 trypsin enzyme activity by dietary treatnrent groups from samples of A) small
intestine (SI), B) proximal large intestine (LI) and C) distal large intestine (fecal; F). Trend lines
based on percent soybean meal equivalent (0, 15, 30, 45 and 60) trypsin inhibitor (TI) level in
treatment diets (commercial control, C, excluded). P(linear) signiﬁes P-value for linear regression

analysis. Plotted values in graphs awarded common superscripts are not signiﬁcantly different (P >
0.05).

53

Trypsin ACtMty (mg/9)

Trypsin ActiVItv (ma/9)

Trypsin Activity (mglg)

4.0

2.0

8.0 1

6.0 -

4.0

2.0

0.0

a” f” b {b {b

. "_T ...- __. L__T _

TI30 Tl45 TI60

TIO 1115
Diet
8
LI
TIO Tl15 TI30 T145 T160
Diet
C
F
a
ab
i ab
" l
TIO T115 T130 T145 TI60
Diet

Figure 4.2 Day-21 trypsin enzyme activity by dietary treatment groups from samples of A) small
intestine (SI), B) proximal large intestine (LI) and C) distal large intestine (fecal; F). Trend lines
based on percent soybean meal equivalent (0, 15, 30, 45 and 60) trypsin inhibitor (TI) level in
treatment diets (commercial control, C, excluded). Plotted values in graphs awarded common
superscripts are not signiﬁcantly different (P > 0.05).

54

ISI

       

l N
ZS
. S
N
: :\\\ .311“.

   

1130 1145
Diet

Figure 4.3 Apparent protein digestibility values (APD) of Atlantic salmon smolts based on dietary
trypsin inhibitor (TI) treatments from samples taken in regions of small intestine (SI), proximal large
intestine (LI), and distal large intestine (fecal; F). TI# represents the percent soybean meal
equivalency (0, 15, 30, 45 and 60) of puriﬁed T1 in test diets.

A 100

3 l

1: l /+\+—__+

o l

“g 75 . +/"/+

.9 ‘

D l

.3 50 1

E —A— Small intestine

&

g 25 l —o—Proxlmallarge

'ﬁ ‘ Intestine

E l —+— Distal large intestine

3 o . - . 7 , ~ "

° TIO Tl15 1130 Tl45 Tl60
Diet

Figure 4.4 Cumulative apparent protein digestibility in intestinal segments of Atlantic salmon smolts
fed experimental diets containing graded levels of trypsin inhibitor (TI) at 0, 15, 30, 45 and 60%
soybean meal equivalencies.

55

Table 4.3 Dietary trypsin inhibitor (TI) affects on fecal sample mean total apparent protein
digestibility (TAPD). MSE represents the mean square error value. TI# represents the
percent soybean meal equivalency (0, 15, 30, 45 and 60) of puriﬁed T1 in test diets.

 

 

Diet TAPD MSE
TIO 74.79 3.13
TI15 81 .17 5.33
T|30 86.86 1 .87
Tl45 82.75 3.41
TI60 82.03 3.56

Proximate Body Composition

Body lipid compositions showed no differences. between control or dietary
treatment groups; however, data was highly variable within treatments (Figure 4.5).
Affects of TI on ﬁsh body fat was not apparent over the 3-week feed trial. As a result ash
and protein body composition assays were not undertaken under the assumption that the

feed trial period was too short to produce meaningful body composition assessment

results.

56

 

29.0- ...—

28.0-

 

27.0-

26.0-1

 

 

Fat (%)

25.0-

24.0 -

23.0-

 

22.0-

 

 

 

I l l l
C TIO TI15 T130 Tl45 TI60

Diet

Figure 4.5 Box plot of mean body fat composition of Atlantic salmon smolts fed a commercial
control (C) and experimental diets containing graded levels of trypsin inhibitor (T1) at 0, 15, 30, 45
and 60% soybean meal equivalencies.

57

Discussion
Experimental Animals

The one-plus year old Atlantic salmon (89.2 g) used in the feed trial were
scheduled for a spring release as smolts by the Michigan DNR. The trial was conducted
in early October (prior to expected spring release), which should have been relatively
close to the time interval these speciﬁc ﬁsh typically undergo physiological
transformations associated with smolting. Over the course of the study, parr marks were
visibly receding from the sides of the ﬁsh, and ﬁsh scales were becoming more silver in
color presenting a silvery sheen appearance. Based on these combined observations the

ﬁsh were considered to be classiﬁed as smolts at the time of the feed trial.

Digesta Dry Matter

Differences were noted in digesta and fecal dry matter; however, it is unclear
whether these differences were a result of SBTIS. At the time of sampling, digesta
samples taken from the digestive tract of the commercial control group were dark brown
and rather dry in appearance. In comparison, samples taken from ﬁsh fed test diets were
bright green due to the marker and appeared much more liqueﬁed. Results of weight
analysis after freeze drying indicated that the commercial control diet generally resulted
in higher digesta and fecal DM content than the semi-puriﬁed diets. Most likely these
results are again due to differences between practical diet and semi-puriﬁed dietary
ingredients and processing techniques.

Studies examining SBM replacement of ﬁsh meal in salmonid diets have reported

SBM increased fecal water content and gastric emptying rates in Atlantic salmon

58

(Storebakken et al. 2000, Reftsie et al. 2001). In this study, SI material from ﬁsh fed T10
and T115 had signiﬁcantly higher DM content than T130, T145 and TI60. This could be
an expected result if SBTIS were responsible for increasing fecal wet weight in SBM
diets. However, the previously mentioned studies were based on fecal material DM
content, not SI digesta. In the current study neither day-10 nor day-21 fecal DM samples
showed effects from SBTI. Based on data obtained from all intestinal regions, it appears
that SBTIS had minimal affect on the water absorption processes in the digestive tract of

Atlantic salmon.

T rypsin Activity

Trypsin activities in ﬁsh typically decrease along the digestive tract, from the
pyloric ceaca and small intestine to fecal excretion (Krogdahl et al. 1994, Rust 2002).
According to the last authors, this is likely due to continuous autodigestion of the
enzymes taking place throughout the digestive tract. In the current study, TA from the
commercial control group clearly followed this pattern. For the semi-puriﬁed diets,
however, TA on day-21 was higher in the L1 and possibly fecal region than those
recorded for the S1. The cause for this deviation is unclear. Conceivably, nutrient
characteristics of the semi-puriﬁed diets increased absorption and gastric emptying rates.

Figure 4.1A shows strong evidence of a negative dose dependent response of TI
on TA in the S1 of Atlantic sahnon. While this was an expected result, due to the nature
of trypsin enzyme/inhibitor interactions, the effect observed from this study was less
severe than those previously reported (Krogdahl et al. 1994, Ollie et al. 1994, Sveier et al.

2001). Day-21 TA results (Figure 4.2) essentially show no differences in TA across

59

dietary SBTI. The main difference between this study and previous studies was that the
primary source of protein for the earlier studies was ﬁsh meal while in this current study
puriﬁed casein and gelatin were used. In this case, lack of TA differentiation across
treatment diets support the previous authors’ assertions that Atlantic salmon can
compensate for T1 activity to a certain degree.

Based on results from this study, it appears that Atlantic salmon compensated for
up to 60% SBM equivalency T1 (3.9 gTI/kg feed). This is somewhat higher than that
reported by Ollie et al. (1994). Those authors reported a reduction in TA in digestive
tract of Atlantic salmon at approximately 48% SBM equivalency and a TA of zero at

64% SBM equivalency.

Apparent Protein Digestibility

Apparent protein digestibility values showed no signiﬁcant dietary effects of
SBTI on protein digestion over the range 0 to 60% SBM equivalencies (Figures 4.3-4.4
and Table 4.3). In contrast, Ollie et al. (1994) reported a sharp decline in protein
digestion of Atlantic salmon at a feed intake level of 48% SBM equivalency TI (3.15
gTI/kg feed). Test diets used by Ollie et al., however, consisted of 70% FM with up to
11% wheat bran, while diets from this current study were made of primarily puriﬁed
protein source ingredients. Combined with the observation that test diets in the previous
study out performed a commercial control in terms of growth, it would appear that the
selection of protein sources of test diets has direct impact on affects of puriﬁed SBTIS in

feed trials for Atlantic salmon.

60

Proximate Body Composition

Body lipid composition data (Figure 4.5) at the end of the 3-week trial provided
no additional information regarding short term effects of SBTIS on Atlantic salmon
smolts. In the 8-week ﬁngerling feed trial (Chapter 3), the commercial diet group
showed the least amount of variation in body fat. Differences observed between this and
the previous study could be due to random effects, smolting and/or the short duration of
this feed trial. In a short term feed trial, one would typically expect to obtain more
information from measurable physiological traits such as enzymatic concentrations or
protein digestion, than from grth characteristic parameters and body composition
analysis.

It is important to note that commercial trout and salmon diets typically contain
some level of SBM. According to Hardy (2003), this could be as high as 8% for
salmonids. Commercial feed manufacturers operate under closed formula policy, and the
amount of SBM in the commercial control diet was unknown. Since the control was a
standard salmonid diet obtained from a highly reputable feed supplier, the likelihood that
variations in body fat levels between the control and test diets observed in this study were
a result of dietary effects was assumed to be low. More likely, these variations were due
to the condition of the smolts when received from the DNR and/or random variations

associated with tank stocking.

Fish Health and Compensatory Mechanisms

Results of this study suggest that SBTIS can cause short term affects on Atlantic

salmon smolts. In the long term ﬁngerling study (Chapter 3), SBTIS appeared to affect

61

body composition, to a degree, but did not show any effects on growth, feed conversion,
or health conditions. Neither the short term nor long term study provided speciﬁc
evidence of immediate detrimental health effects on Atlantic salmon. These results
appear to support a hypothesis that Atlantic salmon are able to compensate, through
physiological compensatory pathways, for up to 60% SBM equivalency dietary T15 and
maintain good health and growth. Ollie et al. (1994), reported a physiological
compensatory factor to SBTI in trypsin enzyme activity occurring in pyloric ceaca of
Atlantic salmon. Krogdahl et al. (1994) reported a similar response in the same species
occurring in the intestine. Other researchers have reported compensatory responses to
SBM based diets in coho salmon (Haard et al. 1996).

Present observations may support a hormesis affect as proposed by Calabrese and
Baldwin (2003), where a toxic substance acts like a stimulant in small doses, but as an
inhibitor in larger doses. This point was also made by Krogdahl et al. (2003).
Conversely, if the observed physiological compensations are indicators of a pending
impairment, then, based on results herein, additional impairments would be required to

reach a point of nutritional deﬁciency (reduced growth, disease or death).

Chapter Summary

SBTIS had no effect on digesta dry matter content or apparent protein digestion,
but had a slight inhibitory effect on TA in the small intestine. There was no indication of
detrimental impacts on ﬁngerling and smolting Atlantic salmon by SBTI concentrations
up to 60% SBM equivalency. Moreover, results provide some evidence that Atlantic

salmon are able to compensate for SBTI over the range of trypsin inhibitor tested.

62

In regards to SBM-based diets, research has shown that Atlantic salmon
experience serious nutritional deﬁciencies when fed diets containing 10—30% SBM
depending on various factors (Storebakken et al. 1998, Ref’tsie et al. 2000, Krogdahl et al.
2003). Data from this research strongly suggests that SBTIS alone may not be the main
cause of these problems, but that other anti-nutritional components (6. g. soluble and non-
soluble carbohydrate factions) must be contributing factors to effects observed by other

researchers.

63

Chapter 5. Fixed Formulation Model Development of Practical High Energy SBM
Diets for Atlantic Salmon (Salmo salar)
Introduction

SBM has received worldwide attention as a potential alternative for ﬁsh meal in
formulated feeds for aquaculture. To date its use in formulated feeds for carnivorous ﬁsh
has been limited. Extensive studies on trout and salmon have identiﬁed anti-nutritional
factors associated with SBM and other plant-based protein sources (Refstie et al. 2000,
Storebakken et al. 2000, Krogdahl et al. 2003). While research conducted in this area has
shown varying results, the consensus among most researchers is that rainbow trout and
Altantic salmon begin to show signs of nutritional deﬁciencies when fed diets containing
5% to 30% SBM (Hardy 2003, Krogdahl et al. 2003).

A current trend in commercial Atlantic salmon'production appears to be a
transition from the use of conventional salmonid diets to high nutrient dense (HND) diets
(Einen and Roem 1997, Azevedo et al. 2002). Data obtained from the Fish Nutrition
Research Laboratory in Canada indicated that HND salmonid starter diets require
approximately 49% CP, 20% CF, DE 20 MJ/kg and DP:DE 22g/MJ. These values were
in close agreement with Azevedo et a1. (2002), who reported that an HND diet (46% CP,
25.6% CF, digestible energy (DE) 22 MJ/kg, DP:DE 20 g/MJ) fed to Atlantic salmon
improved feed efﬁciency and lowered solid waste output over conventional diets. Einen
and Roem (1997) recommended slightly lower DP:DE ratios of 19 g/MJ for 1 kg Atlantic
salmon, and reported that carcass quality decreased with decreasing DP/DE ratios. Based
on results from these researchers, optimal HND diets for young Atlantic salmon should

contain between 46-52% CP, 19-25% CF, and 19-22 g/MJ DP:DE. Few studies have

64

analyzed elevated SBM levels in practical HND diets for Atlantic salmon.

The term “practical diet” refers to diets formulated with commercial ingredients,
which are normally purchased in bulk from commodity markets. Formulated feeds used
in commercial, State, Federal, and Provincial aquaculture facilities are typically
purchased from commercial feed suppliers. In the US, commercial feed companies
operate under a “closed formula” policy. Information on feed labels and data sheets
usually include proximate analysis and feed ingredients, while ingredient quantities are
considered proprietary information and not required to be listed (NRC 2003).
Commercial feed manufactures also tend to use least-cost diet formulations, where
ingredient quantities are routinely changed between feed lots based on the quality, price
and availability of feed ingredients (Guillaume et a1. 2001).

Non-proprietary “open formula” diets are those that ingredient quantities are
available for public inspection. As apposed to least-cost formulations, ﬁxed diets are
those that ingredient quantities are ﬁxed, or constant. Open formula, ﬁxed diets are often
selected for nutritional research purposes.

This study was part of an integrated research project funded by the United
Soybean Board, Soy-in-Aquaculture Program. The project was designed to develop
commercially acceptable SBM-based formulated feeds for Atlantic salmon and other
species.

The goal of this portion of the study was to formulate potentially commercially
viable, open formula practical diets for Atlantic salmon for use in subsequent feed trials
designed to identify maximum SBM replacement levels for FM.

Study objectives included:

65

1) literature review in order to utilize to the extent possible the best available
knowledge pertaining to
o optimal commercial diets presently used in the Atlantic salmon industry,
. practical ingredient nutrient utilization speciﬁc to Atlantic salmon,
0 related SBM nutritional studies conducted on salmonids,

2) maintain commercial viability through ingredient selection and cost

minimization, and
3) minimize phosphorus output into the environment by maintaining feed

phosphorous levels below 1.2 %.

Methods
Baseline Diet Formulation

The basal diet for this study was the open formula diet MNR-98HS (Table 5.1).
MNR-98HS is an HND diet formulated by the Fish Nutrition Research Laboratory,
University of Guelph and Ontario Ministry of Natural Resources, speciﬁcally for
commercial Atlantic salmon ﬁngerling production. The baseline properties of the basal

diet are as follows: 49% CP, 20% CF, and a DP:DE ratio of 22 g/MJ.

Essential Amino Acid Requirements for Atlantic Salmon

Essential amino acid requirements for Atlantic salmon are provided in Table 5.2.
The two most common limiting amino acids in formulated ﬁsh diets are Met and Lys
(Craig and Helﬁich 2002). Since FM contains more of both these amino acids than

SBM, close attention was given to Met and Lys in the diet formulation process. Diets

66

 

Table 5.1 Dietary ingredients and base-line nutritional characteristics of the Ontario Ministry of
Natural Resources open formula diet, MNR-98HS, used as a basal control diet to formulate soybean
meal-based practical diets for Atlantic salmon.

 

 

 

Ingredient Amount
(9E2

Fish meal, anchovy 300
Blood meal 70
Poultry by-product meal 60
Whey 90
Brewers yeast 50
Corn gluten meal 250
Lysine.HCL 5
Vitamin premix 10
Mineral premix 5
Fish oil, menhaden 160
1000

Crude protein (%) 49
Crude fat (%) 20
Digestible energy (MJ/kg) 20
DP:DE (g/MJ) 22
Ash (%) < 8

A
.5

Total phosphorus (%)

Table 5.2 Atlantic salmon essential amino acid requirements obtained from literature (°/o protein)

and values selected for use in the diet formulation model (% as fed) based on 50% crude protein
diets.

Amino Acid Requirement Requirement Reference
(% protein) (°/o as fed)

 

Arg 4.1 - 5.1 2.3 Lall et al. (1994), Berge et al. (1997), Rollin (1999)

His 1.8 0.9 Rollin (1999)

He 3.2 1.6 Rollin (1999)

Leu 5.2 2.6 Rollin (1999)

Lys 3.2 - 6.1 2.1 Anderson et al. (1993), Berge et al. (1998), Rollin (1999)
Met + Cys 3.1 1.55 Rollin (1999)

Phe + Tyr 5.8 2.9 Rollin (1999)

Thr 3.2 1.6 Rollin (1999)

Trp 0.17 NRC (1993, for Paciﬁc salmon)

Val 3.9 1.95 Rollin (1999)

67

were also formulated to achieve a LyszArg ratio of 1:1, considered optimal for Atlantic

salmon fed SBM diets (Davies et al. 1997).

Vitamin and Mineral Requirements for Atlantic Salmon
Vitamin and mineral requirements for Atlantic salmon and similar species are
provided in Table 5.3. Values obtained for closely related species were used in the diet

formulation model for cases where Atlantic salmon data were missing.

Ingredient Selection and Nutritional Properties

Basal diet ingredients (Table 5.1) were evaluated along with other practical ﬁsh
diet ingredients for addition, reduction, or elimination to/from SBM treatment diet
formulations. Wheat gluten meal (WGM) was added to the list of ingredients based on
research by Storebakken et a1. (2000), and Davies and Baker (1997), which showed high
protein digestibility of WGM by Atlantic salmon and rainbow trout.

The NRC (1993) publication The Nutrient Requirements of Fish provided the
majority of ingredient nutritional properties (Appendix 1) with a few exceptions. The
ﬁsh meal used by feed processors to make test diets for the practical feed trial was
Peruvian anchovy meal. The NRC (1993) publication reported a CP level of 65.4% for
mechanically extracted anchovy meal. This value appeared to be somewhat lower than
expected for high quality anchovy meal. Diet formulations instead were based on the
more current Hertrampf and Piedad-Pascual (2000) values of CF 70.7%, CF 5.3%, and
ash 16.9% for anchovy meal (true). Formulations, however, did include NRC (1993)

amino acid proﬁles for anchovy meal because they were more conservative. Properties

68

Table 5.3 Atlantic salmon diet formulation vitamin and mineral requirements obtained from various

sources.

 

 

Vitamin/ Dietary Requirement Species Reference

mineral unit

Ca (%) 1 Rainbow trout NRC (1993)

Phos. (%) 0.6 Atlantic salmon Lall (2002)

Pot (%) 0.8 Paciﬁc salmon NRC (1993)

Chlorine (%) 0.9 Rainbow trout NRC (1993)

Mg (%) 0.04 Atlantic salmon Lall (2002)

Na (%) 0.6 Rainbow trout NRC (1993)

Cu (mg/kg) 5 Atlantic salmon Lall (2002)

F6 (mg/kg) 60 Atlantic salmon Lall (2002)

Magn. (mg/kg) 10 Atlantic salmon Lall (2002)

Se (mg/kg) 0.3 Rainbow trout NRC (1993)

Zn (mg/kg) 67 Atlantic salmon Lall (2002)

Biotin (mg/kg) 1.5 salmon Halver (2002)

Choline (mg/kg) 1000 Rainbow trout NRC (1993)

Folacin (mg/kg) 10 Atlantic salmon Halver (1972), NRC (1973)
Niacin (mg/kg) 200 salmon Halver (2002)

Panta-Acid (mg/kg) 50 Atlantic salmon Halver (1972), NRC (1973)
Pyrdox (mg/kg) 15 Atlantic salmon Halver (1972), NRC (1973)
Ribo (mg/kg) 25 salmon NRC (1993)

Thia (mg/kg) 15 Atlantic salmon Halver (1972), NRC (1973)
B12 (mg/kg) 0.01 Rainbow trout NRC (1993)

E (IU/kg) 50 Paciﬁc salmon NRC (1993)

K (mg/_kg) 10 salmon NRC (1993)

 

 

and compositions of WGM were obtained from Hertrampf and Piedad-Pascual (2000),

and from the intemet (www.mutritiondatacom, © 2007 CondéNet Inc), since this

ingredient was not listed in the NRC publication. Menhaden ﬁsh oil was assumed to be

100% digestible with an energy value of 9019 kcal/kg (www.food-stats.com).

Vitamins and mineral supplements used in this study included US Fish and

Wildlife Service (U SFWS) vitamin premix #30 at 0.3% and USFWS mineral premix #3

69

at 0.2%, based on recommendations made by project collaborators (Steve Halt, Purdue
University, personal communication; Rick Barrows, USDA ARS, personal
communication). Vitamin and mineral premixes were consistent for all diet formulations
unless otherwise noted. Di-calcium phosphate was selected as a phosphorus supplement
and added to diet formulations as required in order to meet target phosphorous levels of

10.0—11.5 g P/kg feed.

Ingredient Digestibility Indices

Dietary ingredient DP and DE values obtained from literature review are provided
in Table 5.4. Amino acid digestibility data has been published for various feed
ingredients, but for relatively few individual species (NRC 1993, Hertrampf and Piedad-
Pascual 2000). Data obtained for similar species (e.g. rainbow trout, Chinook or other
Paciﬁc salmon) were used where data for Atlantic salmon were lacking. The most
conservative value was chosen for formulations when data from two or more similar

species were available and Atlantic salmon data were absent.

SBM Content

SBM content in commercial salmonid diets most likely ranges between 1-10%
(Hardy 2003). A median level of 5% was selected as a reasonable estimate of SBM
typically included in standard commercial Atlantic salmon diets. The ﬁrst treatment diet
therefore was formulated to contain 5% SBM, which was considered equivalent to a
typical commercial diet. Additional treatment diet formulations started at 20% SBM

inclusion and increased at 5% intervals. FM was reduced by 20% in treatment diets

70

Table 5.4 Digestible protein (DP) and digestible energy (DE) values of practical ingredients used in
diet formulations for Atlantic salmon.

 

 

Ingredient DP DE
(%) (MJ/kg)
Wheat gluten meal 100 1 20.3 ‘
SBM 86.8 2 13.6 5
Fish meal, anchovy 88.5 4 20.2 5'“
Blood Meal 69.0 3 14.3 7
Poultry by-product meal 68.0 3 16.6 7
Corn gluten meal 89.7 " 17.8 3
Brewers yeast 78.8 5 14.7 3
Whey 87.8 5 11.6 8

I Glencross and Hawkins (2003) rainbow trout

2 NRC (1993) Atlantic salmon

3 NRC (1993) rainbow trout

‘ Hertrampf and Piedad-Pascual (2000) Atlantic salmon

5 Hertrampf and Piedad-Pascual (2000) salmonids

6 NRC (1993) Chinook salmon

7 Hertrampf and Piedad-Pascual (2000) rainbow trout

3 Hertrampf and Piedad-Pascual (2000) unspeciﬁed ﬁsh species

containing 20% and 30% SBM in order to assess further reductions in FM. Finally, SBM

inclusion rates above 30% were evaluated based on nutritional requirements of Atlantic

salmon and selected ingredients.

Diet Formulation Model

Three diet formulation methods were evaluated for potential use in this study: 1) a
commercial diet formulation software program, 2) a livestock feed ration balancer
program available through Department of Animal Science, Michigan State University,

and 3) a new model construction using standard computer spreadsheet software such as

Excel.

71

Ingredient nutritional property values were ﬁxed inside the model (see “Ingredient
Nutritional Properties” matrix in the Appendix. Ingredient inclusion levels (model input
parameters), were adjusted to evaluate formulations containing baseline diet
characteristics, graded levels of SBM, reductions in FM, and meet essential amino acid
requirements of Atlantic salmon (model output variables). Amino acids Lys and Met,
were synthetically added to test diet formulations as needed. Cost estimates were based
on 2004 economic commodity values provided by project collaborators at Iowa State
University and USDA ARS (Robert Stunmerfelt, Richard Barrows, personal
communication, respectively). Model output values for the control MNR-98HS diet were
compared with baseline characteristics reported by the University of Guelph Fish
Nutrition Research Laboratory (Table 5.1), to assess relative accuracy of the diet

formulation model.

Results
Diet Formulation Model

Of the three model methods evaluated, the two diet formulation software
programs included least cost diet formulation options; however, neither contained
nutritional characteristics for Atlantic sahnon. No single method appeared to be more
user friendly, and each method appeared near equally challenging to learn. This was
compounded by the point that both diet software packages required modiﬁcations to
ingredient selections, speciﬁc nutritional requirements, and digestibility characteristics
for Atlantic salmon. Since the objective was to develop a simpliﬁed ﬁxed cost diet

formulation model, it was decided that building a model in Excel would provide

72

information on diet formulation model development not readily apparent with packaged
software programs. Excel spreadsheet diet formulation model results are provided in the

Appendix.

Practical SBM HND Diet Formulations for Atlantic Salmon

Atlantic salmon practical SBM diet formulations selected for commercial
application feed trials are provided in Table 5.5. SBM treatment diets were identiﬁed
according to the percent SBM and FM inclusion accordingly: SBS/F 30, SB20/F 30,
SBZO/F24, SBZS/F 30, SB30/F 30, and SB30/F24. Diets containing 24% FM were
formulated with 20% reduction in F M from the control. Differences between model
output values and reported values (Table 5.1) for the control diet were within 1.5% for
DP:DE ratios, 1% for CF, 0.5% for CP, DE and ash, and 0.1% for phosphorus.

All diet formulations met baseline characteristic criteria established for HND diets
with a few exceptions. Diet formulation crude fat levels ranged between 18.1% and
19.27%. Crude fat levels of 4 out of 6 diets were slightly below the HND criteria of
19%.

All diet formulations met minimum amino acid requirements of Atlantic salmon
(Table 5.2 and Appendix). Amino acids Arg and Thr typically fell within 20% of
minimum requirements while all remaining essential amino acids ranged between 20—
250% above minimum levels). The control diet was estimated to contain the lowest
amount of Arg (2.31%). This value is equivalent to the minimum requirement reported
for Atlantic salmon. All diet formulations were deﬁcient in sodium by 0.1—0.25% and

the control diet formulation was also potentially deﬁcient in potassium. Phosphorus

73

content varied between 108—1.13%. Increasing levels of phosphorous supplementation
were required in the form of di-calcium phosphate as dietary SBM levels were increased
with concomitant reductions in F M (Table 5.5).

Formulations containing 35% SBM were achieved but had 55% greater crude
ﬁber content than the control (Appendix). Fiber content for 40 and 45% SBM
formulations were 1.71 and 1.79% respectively as compared to 1.05% for the control
diet. Formulations containing 40% or more SBM required substantial increases in ﬁsh oil
as well as methionine and phosphorus supplementations. In addition, diets over 35%
SBM failed to meet one or more HND diet requirements and fell below a 20% safety
margin for essential amino acids Iso, Thr and Trp.

Diet costs estimates ranged from $0.22/1b for SBS/F 30, to $0.18/1b for SBZO/F 24,
SB30/F 30 and SB30/F 24. The 30% SBM diets showed a cost reduction of $0.03/1b from

that of the control.

74

83? 560888 28880 voom :0 women 803 35:58 .80 e

5o: sea; as. 8.88283 9:3 a Base mmwamzz 5. .9 5:68 Ease maam: .
so: as? as. 8.853%. we»... a 858 mmwaizz 8: do. 882.. 5855 maam: N
:83 £8.52 .05 80m :5. whom—oz Soc 335:5 _

 

 

75

8.. ...... m... 8.. m... m... o... gasococoea
5.3.. 3.8 «EN 8...... Fun 3% 3.8 3.. oaocoscoeno
8.. 3.. 8.. 8.. B. 5.. B. 390 + .62
.3 m: .3 use BC. 8.. a: 3.. 5.... 820
mm... 8.. a... o... 8.. 8.. 8.. 3°C case 820
Rm. 8.... we... m3. 3». ...m. 5.2 as. E 820
.34 8.9. 2.8 3.8 8.8 ... .m 9...... EC 529.. 8.5
.... N... N... .... m... e... o... 6283
5. .N 8. .u 8. .N 8. .N am. .N B. .N 8.8 :25. women
mam. 8m. 8m. 8.... 8m. .NBN 8.8. 635.5: mg
2.0 2.0 85 2.0 8.0 «No .~.o .35. 2 5a .80
odoo. odoo. odoo. .38. 0.80. odoo. odoo. 33:2 5.6
m m m m m m 0.66
o e e o e o 8:25 65.96
3 Mn mm ...o ma 8.. 9658.5 5.5.8.5
m... m... a. no ... 65828.2
2. me. we. a. .8. mm. ....m. 8. coocoaeoe .__o 5:
...N 3 m ...N N m m .8an .8652
m m m m m m o. ~5an SEES
m N . o. 3 m .616on
8 on 2 mm. mm... 2.. omm .85 85.6 58
m. 8. mm man on om cm .89. possum
.N 8.8 .... ... mm 8 cm >22.
no... 8 ...8 me 8 8 8 .85 885.3 ana
man 8 3.. B no.6. 2 S .85 82m
as 8... 8... SN 8... 8m 8m 265:6 ..ooE 5m
8m 8.” com com com cm .88 58.8
8. on E E o. on .85 .555 .855
emuammm omuammw 8%wa emuammw 9&5wa omuamm mzwadzs. seasons
.65

 

68.583— .35.. Z .3 .C..c._:_2 223:0 :5 £330 be 3383:: 58.3.53— :oaaomom 5.5.5: Z

gm...— 2: .3 cone—96: 3:. 8:3. Esta—E :3: 255.5.— ..er :a 3 mineral—Z .85 495:3 2: E9... :38 an: E 55260.. e\ec~
55$ Baa—=55. 9.03 .3:— nmc .33 9:53:69 365 .:e_m=_u:_ 0.: .35 no: e5. 59 30:. :aoazem E33.— 8 ”59.32. 3532:
one». 30:. 23:53.: :3:— :aon>em 6:622:53. 3:. Conant—«2: 5.218 :5. .Amlwath—zv .83 .3813. 93:23. Wm «Bah.

Discussion

The decision to develop the diet formulation model in Excel proved to be a
worthwhile undertaking in that each step of the diet formulation modeling process was
completed by the modeler. As the modeler builds the formulation model, he or she
undertakes a series of exercises that ultimately require a thorough understanding of
ingredient properties, ingredient bioavailability and ﬁsh nutritional requirements.

Finding suitable basal diets for diet formulation and development can be difﬁcult
due to the closed formula policies of commercial feed companies. In this study, we
found a suitable HND open formula diet formulated speciﬁcally for Atlantic salmon. Not
only did MNR-98HS provide an established base-line diet formulation, but published
data on the basal diet also provided a means to compare results between our and previous
experiments. The ﬁxed formulation model from this study was within 1.5% of all
proximate compositions for MNR-98HS reported by the Fish Nutrition Research
Laboratory, University of Guelph.

Ingredient nutritional properties directly inﬂuence physiological and chemical
interactions affecting ﬁsh health and growth. Ingredient selection also affects diet
stability and cost of production. Study dietary ingredients were selected based on criteria
of meeting HND speciﬁcations, Atlantic salmon nutritional requirements, and
commercial viability (availability, costs, etc.). One commercial feed supplier (Todd
Prowless, Zeigler Brothers, personal communication) strongly suggested avoiding animal
by-product ingredients (e. g. blood meal and meat meal). While these have shown to have
characteristically high nutritional value for ﬁsh, infectious diseases associated with

dietary animal by-products have been reported over the past several years (CDC 2007,

76

Latouche et al. 1998). Diets made with animal by-products currently risk importation
bans to various countries. For example, the European Union has banned the use of
processed animal protein (notably meat and bone meal) in feeds for all farm animals
since 2001 due to the threat of bovine spongiform encephalopathy (BSE or mad cow
disease, EC Regulation 999/2001 of the European Parliament and of the Council). Meat
meal and meat meal with bone also have relatively high phosphorus levels (4.1 and 4.6 %
respectfully, NRC 1993).

Due to differences in protein content, replacement of FM with the addition of
signiﬁcant quantities of SBM in standard commercial salmonid diets required reductions
of other main ingredients such as poultry and beef by-products. Atlantic salmon HND
diets became increasingly more difﬁcult to formulate when SBM levels exceeded 30%.
Digestible energy, protein, and ﬁber content of primary dietary ingredients are provided
in Figure 5.1. As the level of SBM increased, moderate increase in ﬁsh oil were required
to maintain Atlantic salmon energy requirements. Few options were available to increase
protein. The added ingredient, wheat gluten meal, has a relatively high nutritional value.
WGM, however, was limited between 7-10% based on recommendations by USB project
collaborators. Diets exceeding 10% WGM produced very hard diet pellets with poor
palatability for rainbow trout (Richard Barrows, USDA ARS, personal communication).
According to Figure 5.1 then, of the remaining ingredients, corn gluten appears to be the
best available replacement for ﬁsh and blood meals.

Diet formulations showed increased dietary ﬁber levels with increased SBM.
Certain types of ﬁber have been reported to result in poor digestion, faster gastric

emptying rates and depressed grth in salmonids (Davies 1988, Hilton et a1. 1983).

77

 

 

100.0 , 25.0
90.0 4‘
800 ~ — 200
a? 70.0 . —
v —
a 60.0 .... > 15.0 3
g — ‘
o 50.0 1 05
z ‘ l v
g 40.0 1-03
of 30.0
D l
20.0 7
0'0 7 '7 ,. ' ‘7 c l P I
Wheat ‘ Blood FlSh orn outryb y- Brewers ‘
‘ gluten ‘ . meal. gluten SBM product ‘
Meal 1 ast
meal . anchovy meal ‘ l mea al ‘
13 DP (%) 80.1 1 61.5 69.0 54.2 . 43.4 40.6 ‘ 33.6 3 11.7 1
;I Fiber(% x10) 5 ‘ 10 . 10 k 15 l 34 i 21 T 32 l 2
- . . 1 - , . . . - .. . . t . . . - . .
-DE 20.3 14.3 ‘ 20.2 17.8 . 13.6 L 16.6 l 14.7 ‘ 7117.6

 

Figure 5.1 Digestible protein (DP), relative ﬁber and digestible energy (DE) contents of principle
ingredients used to formulate high nutrient dense soybean meal diets for Atlantic salmon. Relative
ﬁber is equal to % ﬁber x 10.0 as a scalar to the y-axis. Optimal ingredients would be high in DP and
DE and low in ﬁber.

According to the National Research Council (NRC1993) most ﬁsh can tolerate up to 8%
ﬁber in their diets; however, most types of ﬁber are indigestible for salmonids. Dietary
ﬁber (non-starch polysaccharide), was identiﬁed as an additional parameter of concern
over the course of this study. Attempts were therefore made to limit dietary ﬁber content
to no more than 50% excess of the basal (control) level of 1.05% (Table 5.5). Replacing
FM and blood meal with SBM and corn gluten could result in a substantial increase in
dietary ﬁber (Figure 5.1).

Diet formulations met Atlantic salmon vitamin and mineral requirements with
minor exceptions. The control diet resulted in a potassium content of 0.65%, which was

slightly below the required 0.8% for Atlantic salmon. In addition, all diets showed slight

78

deﬁciencies in sodium content. According to De Silva and Anderson (1995), however,
sodium, potassium and chlorine are extremely common in the environment, and among
the most common elements found in ﬁsh. As such, supplementation of these elements is
probably not required.

Phosphorus (P) supplementation was not required for the control diet, but was
required for nearly all SBM diets. According to the NRC (1993) F M is high in
phosphorus (2.43%) while SBM is comparatively low (0.64%). Low FM high SBM diets
increased P supplementation requirements considerably. Supplemental P also helps
ensure the presence of an available form of P since most of the phosphorus in SBM is
tied up in the form of phytic acid (Hardy 2003).

Diet formulations generally met minimum essential amino acid requirements
across all diets up to 35% SBM inclusion. However, higher level SBM diets required
increased amounts of methionine supplementation. Diets containing 40 and 45% SBM
required 2.5 and 3.5% supplementation respectively, while the control formulation did
not require any supplemental methionine (see diets SB35/F 20, SB40/F 20, and SB45/F 10
in Appendix).

Diet formulations above 30%SBM were difﬁcult to maintain both HND diet
protein and energy requirements. They also required substantial decreases in other
(including more concentrated) protein sources, increased ﬁsh oil for energy and increased
supplementation of methionine and phosphorus. Loss of protein sources would likely
reduce the overall nutritional balance expected of a HND diet containing multiple protein

sources. Due to these factors and elevated ﬁber content in high SBM diets, formulations

79

selected as having the best opportunity at being commercially viable were limited to 30%
SBM or less (Table 5 .5).

Cost comparisons showed a reduction of $0.03 US per pound ($0.014/kg) feed
between the control diet and the high SBM diets. Based on 2004 economic commodity
values, a 3% cost reduction in feed is conceivable provided Atlantic salmon can tolerate
up to 30% SBM. At a feed conversion rate (F CR) of 1.0, this would equate to a 3% cost
reduction per kg ﬁsh across a 1.5 million metric tonne (2004 value) Atlantic salmon

industry.

Chapter Summary

The Atlantic salmon diet formulation modeling process undertaken in this study
provided valuable information regarding diet formulation model development and
Atlantic salmon nutritional requirements. Corrunercial and/or academic diet formulation
software offer additional beneﬁts including least cost diet formulations. A logical
progression in future diet formulations for Atlantic salmon would be to incorporate the
information summarized within the conﬁnes of this study into least cost, nutritional
optimization diet formulation model(s).

High nutrient dense (HND) diets containing 5 - 30% SBM, 30%F M were
formulated after the basal diet MNR-98HS (Ontario Ministry of Resources). All diets
were based on speciﬁc nutrient requirements of Atlantic salmon, and formulated for 50%
crude protein, 19% crude fat and DP:DE of 21 W]. SBM treatment diets were
identiﬁed according to percent SBM and FM inclusion accordingly: SB5/F30, SB20/F 30,

SB20/F24, SB25/F30, SB30/F 30, and SB30/F 24. Diets containing 24% FM were

80

formulated with 20% reduction in PM from the control. The resulting diet formulations
will be used in a feed trial designed to identify maximum SBM replacement levels for

FM in commercial diets for Atlantic salmon smolts.

81

Chapter 6 — Replacement of FM with SBM in High Energy Practical Diets of
Smolting Atlantic Salmon (Salmo salar)
Introduction

In 2002, a collaborative project was ﬁinded by the United Soybean Board (U SB),
of the United States, to examine the extent of SBM anti-nutritional properties on
salmonids. The goal of this collaboration was to determine how to overcome SBM
associated nutritional problems in order to increase the potential use of SBM in
formulated ﬁsh feeds. This particular study was the ﬁnal feed trial of an integrated
research project at Michigan State University, and was designed to develop a
commercially viable practical diet SBM diet for Atlantic salmon. The main goal of this
study was to determine maximum SBM replacement levels for FM in practical diets for

Atlantic salmon.

Methods and Materials
Diet Formulations and Processing

The control diet (MNR-98HS) for the feed trial was an open formula high nutrient
dense (HND) Atlantic salmon diet formulated by the Fish Nutrition Research Laboratory,
University of Guelph and Ontario Ministry of Natural Resources. All test diets were
formulated to contain 50% CP, 19%CF, with 20 MJ/kg digestible energy and DP:DE
equal to 21 WI The control and test diet formulations are provided in Table 5.5 (see
Chapter 5). Additionally, all diets were formulated to contain a minimum of 1.87% Met
+ Cys, LyszArg ratio of 1.1, and phosphorus concentration between 1.0 and 1.15% (10.0

and 11.5 gP/kg feed). The control diet contained 0% SBM and 30% FM, and was

82

compared to six test diets containing 5 to 30% SBM. In two of the test diets, the amount
of ﬁsh meal was reduced by 20% from that of the control diet. Test diets were identiﬁed
according to percent SBM and FM inclusion accordingly: SBS/F 30, SB20/F 30,

SBZO/FZ4, SB25/F30, SB30/F 30, and SB30/FZ4.

All diets were processed under the same conditions by project collaborators at
USDA ARS and USFWS Bozeman Fish Technology Center (Richard Barrows, USDA
ARS, personal communication). Diet processing methods were based on optimal
processing parameters determined by Barrows et al. (2004) for practical SBM diets for
rainbow trout. Ingredients were ground using an air-swept pulversizer (J acobsen, J ASP
18-H) and then mixed into a complete feed. The diets were manufactured using a twin-
screw cooking extruder (DNDL-44, Buhler AG, Uzwil, Switzerland). The mix was
introduced to the steam conditioning chamber with a volumetric feeder (K-tron, Inc.,
Pitrnan, NJ). Feed rate (1200 rpm) was set to provide 18 seconds of heat exposure in the
extruder. The extruder had 6 barrel sections and temperature was controlled by heating
section 2, 3, 4, and 5, with a maximum extruder barrel temperature of 262°F, and an
average die head pressure of 230 PSI. Sections 1 and 6 were cooled to insure smooth
feeding and discharge. Upon discharge the diets were dried in a pulse bed drier (Buhler,
Uzwil, Switzerland) for 25 minutes at 250°F with a 10 minute cooling period which
resulted in moisture levels less than 10%. After the diets had dried they were top-coated

with the remaining oil (6%) at ambient pressures.

83

Experimental System and Animals

One-year old Atlantic salmon were obtained from the Michigan Department of
Natural Resources, Thompson State Fish Hatchery, Manistique, Michigan, in November
of 2004. The ﬁsh were transported to the Michigan State University Aquaculture
Research Laboratory and were acclimated to water conditions in a 1,890-L ﬂow-through
tank culture system over a 30-day period. Fish were fed a commercial trout diet over the
acclimation period. A total of 420 ﬁsh were randomly distributed in 21, 225-L tanks, at
20 ﬁsh per tank, and acclimated on the MNR-98HS control diet to feed trial rearing units
for an additional 15-day period. Flow rates were maintained between 5.6—7.5 me fresh
well water based on target exchange rates of 1.5—2.0 water exchanges per hour. Water
temperature for all tanks remained between 12.3 i 10°C for the duration of the study.
Dissolved oxygen, total ammonia nitrogen, and all other water quality parameters fell
within acceptable limits for salmonids (Piper et a1. 1982).

At the start of the trial, parr marks were clearly visible on all ﬁsh. During the
acclimation period several ﬁsh showed signs of bacterial infection. Prophylactic
treatments of 10 ppt salt were administered once per week through the end of the feed
trial. Thirty six mortalities occurred over the course of the 85-day feed trial. All
mortalities showed external signs consistent with those observed in 2003 Atlantic salmon
ﬁngerling and smolt feed trials at the MSU Aquaculture Laboratory. Upon completion of
the trial, nearly all ﬁsh had developed a silver sheen appearance and showed less
noticeable parr marks.

Total weights of ﬁsh per replicate tank were recorded at the start of the feed trial.

Average initial weight was 27.6 :1: 1.0 grams/ﬁsh/tank. Triplicate groups of Atlantic

84

sahnon were fed either the open formula MNR-98HS control diet, or one of six SBM
treatment diets, 2 times daily (8:00—9:00 am, 4:30—5:30pm). Total weight samples were
obtained every 2 weeks throughout the study. Daily feed levels were determined on a
constant percent body weight basis (%BW) adjusted bi-weekly based on initial condition
factor (k), average water temperature and total weight samples. For the trial, %BW was
calculated as 90% of the theoretical optimal feed level for salmonids (Westers, 1987)
Feed levels fell both above and below satiation levels of the ﬁsh across feeding times
based on observations of excess feed in tank bottoms at various times through the feed
trial.

The feed trial was conducted for a period of 12 weeks. At the end of the trial,
total weights were recorded for each tank. Ten ﬁsh ﬁ'om each tank were randomly
selected for length and weight measurements, collection of blood, whole liver, and
intestinal digesta samples. Fish were euthanized in MS-222 at a concentration of
500mg/L (AVMA 2000). Blood samples were frozen at -20°C until centrifuging at 4000
rpm, 4°C for 10 minutes. Clear natant plasma was pipetted from the centrifuged blood
samples and frozen at ~80°C for future analysis. Whole livers were removed and
weighed. Intestinal digesta was scraped from the entire length of ﬁsh large intestines,
pooled by tank, and frozen at -20°C. Small intestines were excised from the ﬁrst three
ﬁsh sampled ﬁ'om each tank and ﬁxed in 10% neutral buffered formalin for subsequent
histological examination. Three ﬁsh from each replicate tank were pooled by tank and
frozen at -20°C for subsequent proximate body composition analysis.

At the end of the 12-week feed trial, 13 ﬁsh from the control group and 16 ﬁsh

from the SB30/F 24 group were available for further study after completion of the

85

sampling procedures described above. These ﬁsh were stocked into a two tanks, by
dietary treatment group, and were fed their original treatment diet for an additional 10-
week period. At the end of the second feeding period, all ﬁsh were sacriﬁced as
described above and muscle tissues were collected from the mid-dorsal muscular regions.
Tissue samples were then frozen at -80°C for subsequent amino acid composition

analysis.

Sample Analysis and Calculations

Feeding levels

 

Trial feed levels were calculated as 90% of the theoretical optimal %BW for
salmonids as developed by Westers (1987) for standard commercial diets and a feed
conversion rate (FCR) of 1.0:

%BW = (2 x 0C) /(W/ki)“3 x 0.90
where:

0C = temperature in Celsius

W = Wet weight of ﬁsh at time of sampling (g),

ki = condition factor at the start of the trial.

912m

Weight and length data were used to determine condition factor (k) and speciﬁc
growth rate (SGR) over the course of the study:

k = W/L3

SGR = (1n Wf — ani)/d x 100

86

where:
L = length (cm),
Wf ori = average ﬁnal or initial wet weight (g),

d = time (days).

Feed conversion

Feed conversion rates were calculated as the standard apparent FCR with
adjustment for mortalities:

FCR = Cum Feed/(Net W Gained + (WM...s — (NMons x Wi)))
where:

Net W Gained = net weight of ﬁsh gained over the feed trial (g),

Cum Feed = cumulative weight of feed (g),

WM,“s = Wet weight of mortalities removed (g),

NMons = Number of mortalities removed.

Protein efﬁciency

Protein efﬁciency ratios (PER) were based on the formula provided by De Silva
and Anderson(1995) slightly modiﬁed to account for mortality:

PER = (Net W Gained + (WMom —- (N Mons x Wi))/ TPFecd

TPpeed = Cum Feed x % PFeed / 100
where:

TPFeed = total protein level in feed fed,

% PFeed = % protein in diet as fed basis (%P/gfeed),

87

Whole body composition

Whole body composition analysis included lipid, protein, ash, dry matter and
energy. Pooled, frozen whole body samples were thawed quickly under cool water and
homogenized in a commercial-grade food processor. Samples were weighed, frozen and
freeze-dried. All freeze-dried samples were ﬁnely ground and stored at -200C. Whole
body lipid content was determined on duplicate 1.0—3.0 g samples by lipid extraction
with diethyl ether (Soxtec System HT/ 1043 Extraction Unit, Tecator, Sweden). Nitrogen
was determined on 0.5 g samples according to combustion method AOAC (2000) using a
Leco nitrogen/protein analyzer (model FP-2000, Leco Corp., St. Joseph, MI). Protein
was calculated as N x 6.25. Dry matter was obtained after oven drying 2.0 g samples at
105°C for 18-24 hours. Ash content was determined aﬁer placing dry matter samples in
mufﬂe ﬁrmace at 500°C for 18 hours. Gross energy was determined by bomb
calorimetery on whole body samples from control and SB30/F 24 replicates for
comparison of the two most extreme treatment levels. All body composition samples
were performed in duplicate, and repeated if differences between individual samples
exceeded 3.0%.

Feed lipid, protein, ash and gross energy were determined for all diets using the
methods described above for whole body samples. Feed percent total carbohydrate levels
were determined using the difference method, by subtracting the sum of percent protein,

lipid and ash from 100%.

88

Digesta dry mm

Large intestine digesta was lyophilized using a Tri-Philizer MP, F TS systems, for
48 hours minimum by laboratory personnel at MSU’s Department of Animal Science
(Dave Main, MSU, personal communication). Dry matter content was determined by

measuring weight of digesta samples before and after freeze drying.

Trypsin activity

Lyophilized large intestine digesta samples were held at -20°C until assayed for
trypsin activity. Trypsin activities were determined colorimetrically using BAPA
(Benzoyl-DL-arginine-p-nitranilide) as a substrate as described by Kakade et al. (1969)
with slight modiﬁcation. This procedure is described in detail in Chapter 4 of this
dissertation. Trypsin activity (TA) in mg trypsin per gram intestinal fecal material was
calculated as follows:

TA (mg/g 831111316) = Try (Pg/vol) / Wt (g)

Try (pg/vol) = Try (pg/ml from standard curve) x EVol (ml)
where:

Try = trypsin value; EVol = extract volume.

Plam insulin

Plasma from three ﬁsh per tank were assayed for insulin concentrations at MSU’s
Diagnostic Center for Population and Animal Health Laboratory, using protocols (DSL
Insulin RIA product insert, July 27, 1999) used for a 2004 Atlantic salmon feed trial

conducted by fellow USB project collaborators, Purdue University.

89

Mic somatic indices

Excised whole liver weights were used to calculate and contrast hepatic somatic
indices (HSI):

HSI = LW/Wf
where:

LW = Liver weight (g).

IntestinaLl histology

Fixed small intestine tissue samples were cut into multiple cross-sections,
parafﬁn-embedded, sectioned at Sum, and routinely processed for staining with
hematoxylin and eosin (Prophet et al. 1992). All slides were evaluated microscopically
by Dr. Scott Fitzgerald, an American College of Veterinary Pathologist, at MSU’s
Diagnostic Center for Population and Animal Health Laboratory. Sections of intestine
were evaluated for villous to crypt ratio, whether or not the mucosal epithelium and
microvillous brush border were intact, presence of inﬂammatory inﬁltrates, and for any

other evidence of degeneration.

Muscle Tissue Amino Acids

Muscle amino acid composition analysis was performed on tissue samples from
ﬁsh fed the two extreme diets of the feed trial (control and SB30/F 24 diets), 5 samples
per group. Amino acids were identiﬁed following HPLC separation (Waters 486
absorbance detector, Waters Corporation) at 254 nm by Julie Moore of MSU Department

of Animal Science, according to the method described in Guay and Trottier (2006).

90

Statistical Analysis

Data were analyzed by one-way analysis of variance (ANOVA) using SPSS
(SPSS© Release 12.0, 2003) and SAS (SASG Release 9.1, 2003) statistical software.
Homogeneity of variance was conﬁrmed by Levene's statistic analysis. Signiﬁcant
differences between means were compared by Duncan’s multiple range test. Trend
analyses were conducted using regression analysis. Treatment effects were considered
signiﬁcant at P S 0.05 unless otherwise noted.

Results from the amino acid assay were subjected to statistical analysis even
though each sample group came from a single tank. Since these ﬁsh were fed their
respective diets for a total of 22-weeks, it was assumed that tank effects would have little
to no inﬂuence on muscle tissue amino acid compositions. Individual ﬁsh were therefore

treated as replicates, for the amino acid analyses only.

Results
Feed Compositions

Diet formulation results (target levels) were compared to measured feed protein,
lipid, ash and carbohydrate compositions (Table 6.1). Actual feed CP levels were within
0.8% from each other, but consistently 5-6% higher than the expected formulated values.
Actual feed crude fat contents were about 1—2% higher than formulated values except for
diet SBS/F 30. Feed ash and carbohydrate compositions were 2—4% lower than
formulated values. Both the OMNR control and the low SBM (5%) diets appeared to
contain the greatest amounts of carbohydrates (20.4% and 21.7% respectively), while the

30% SBM diet contained the least (18.1—18.3%).

91

Table 6.1 Compositions of test diets (as fed basis). Measured crude protein (CP), crude fat (CF),
ash, carbohydrate (CHO) and energy content (E). MNR-98HS is the open formula control diet.
Test diets are identiﬁed according to percent soybean meal (SB) and percent ﬁsh meal (F).

Formulated values are in parentheses.

 

Diet %cp %CF %Ash %CHO E (IN/Kg)
nun-90115 55.3 (49.5) 19.7 (19.0) 4.7 (7.0) 20.4 (23.7) 22.1
sesn=30 50.0 (51.1) 17.7 (10.1) 4.0 (7.9) 21.7 (22.9) 23.2
SB20IF3O 55.0 (50.7) 20.4 (10.0) 5.2 (0.0) 10.0 (22.0) 23.2
SBZOIF24 55.0 (50.3) 20.3 (10.7) 5.0 (0.9) 19.7 (24.1) 23.0
31325le 55.3 (50.2) 20.7 (10.0) 5.1 (7.9) 10.9 (23.1) 23.5
SB30IF30 55.3 (49.9) 21.3 (19.0) 5.3 (7.0) 10.1 (23.3) 23.3
SB30IF24 55.2 (49.7) 21.2 (19.3) 5.3 (0.0) 10.3 (24.2) 23.3

 

Growth Characteristics, Feed Conversion and Protein Retention

No differences were observed in ﬁnal weight, growth (SGR), feed conversion
(F CR), or protein retention (PER) of ﬁsh fed control and experimental diets (Table 6.2).
Feed conversion rates ranged between 0.78-0.82. Only slight differences were observed
in the condition factor (k). Condition factors for all diet groups ranged from 0.010—
0.0107, and are well within optimal values for salmonids (Harry Westers, Aquaculture

Bioengineering Corporation, personal communication).

Proximate Body Composition

Whole body composition analysis showed minor differences in DM, CP, and ash,
but no trends were observed in these parameters (Table 6.3). Increased SBM in the diet,
however, was inversely related to whole body fat content (Figure 6.1). A plot of body
lipid composition with the amount of SBM in the diet produced a highly correlated
negative linear ﬁmctional response (linear, P = 0.0001) of:

Body Lipid Composition = -0.32 x %SBM +2259 (R2 = 0.88).

92

 

....3 .33 .....3 .83 .83 .83 .....3

.3 8.. 8.. .3 8.. .... 8.. «ma
.33 83.3 .893 ..83 .893 38.3 ..83
«.8... 8.... 8.... 8.... 8.... 8.... «.8... mo...
.883 $88.3 $883 .383 .883 .883 .883

a 8.2. a. 8.3. a. 8.3. . 8...... a 8...... a. 8...... . .23 x
.83 .83 .83 .83 .83 .....3 .83
8.. 3.. 8.. 8.. 8.. 3.. 3.. .23 «3
8.... .83 .83 .83 .83 .83 .83
8.8 8.8 8.8 3.8 8.... .... ... 8.8 .3 £225 .2...

 

ﬁnkcmmm own—...mmm can—.mmmm ﬁNuBumm emu—Bumw onuﬁmm 9.5?sz

 

Ame... A a. 393...... m..:3:.:u.w 3: 9... 33.39333 525:3 3933..

B9. .33 :. 8......» A... .3... ...... 33.3.. :3 $5.13.: 53:3... .393: 3 ”5.9.3“... 35.3... 9... «3:. .8... 3.1—95:3 «38...... :2... a... a.
$534.22 33:39:... :. 9... 9.9.3 9.3.538 :32 ..3:. ..m... ..:.. .3... 32.38 .... 8.23. «5%....» 3.53:3 «3:. .3. 5.5.8 8.3.3. «:22...
:. $5... 3.3.. 333:3 :33... ..:.. 5.0.5 3.... :e.u.3>:3 .33.. A... .83.... :e::.:3 AMUmV 3.... 5.39.» 953% 4.3.9: 3:... :32 N... 3.3...

93

 

.83 ......3 .....3 .83 .83 .83 ....3

3.. a... . ...... .... 8.. 2.. 3... 3.. ...... .. .... s ....o :2
.83 ....3 .83 .8... .....3 .83 .....3

3 ...... u 8... 8.. 8.... s 8..... ._ 8.... a 8... a 8.8 ...".
....3 .83 ....3 .8... .33 .83 .83

.... 8.8 a .....m a 8..... .... 8.8 a 8.8 n 2.8 a 9...... £32..
.83 .....3 .....3 .83 .83 .....3 .....3

a 8.... a 8.8 .... 8.8 .... 8.... a .8. a 8... ... 8.8 3...... ....

 

QuuBmmm ammBmumw enuBNmm «9:0wa Gnu—Bumw emu—Bum 2.5?sz

A3... A .5 593...... 35:35:»... ...: ...... 33.38358 5553 3.33:. 3.... 33 :. 3......» A...

.35 ..o... 533.. ..5. Em. .35 :39»... .33.... ... u:.....33 3.....3... ...... v3... .3... .3... .353 2:53.. 3.... o... a. minaénzz 803.333.. :.
...... 3...... .3353 :32 ..35 ..m... :5. .35 :39»... .... 8.93. «5%....» 3o... ...... 5.5.8 u..:....< «5:958 .... 8.9....5. 5.3.3353 ......m n6 2......

94

 

24
g 22 ia‘ fa y=-0.32x+22.59
° R2=0.88
E. 20 l P(linear)= 0.0001
I
18 --
g ! b
g 16 4 bu. . Ibod
.- C _
g14—1 "id
E11 d
10 I T g Y i- -2. _
0 5 10 15 20 25 30

SBM In diet (%)

Figure 6.1 Percent whole body lipid compositions of smolting Atlantic salmon as a function of
dietary soybean meal. P(linear) signifies P-value for linear regression analysis. Values in each row
awarded common superscripts are not signiﬁcantly different (P > 0.05).

Mean body fat content of ﬁsh fed the control (0% SBM) was approximately 9% higher
than those of ﬁsh fed diets containing 30% SBM. This result was consistent for both
30% SBM diets regardless of the level of FM (30% and 24%).

Whole body gross energy values obtained by bomb calorimetery showed no
signiﬁcant differences between ﬁsh fed the control and ﬁsh fed the highest SBM /lowest

FM diet (SB30/F24; Table 6.4).

95

PHI»

24 a b
.2: ;

§

3'?

r 20 i " “ it” ”Cd
y=-0.20x+23.74 bcd Cd

g 13 ‘ R2=0.72 ._

a l P(linear) = 0.0001 id

5 16 t

g 14 1

a |

gum. in- --u [he .

- 0 5 10 15 20 25 30

SBM in diet (%)

Figure 6.2 Large intestine digesta dry matter (lyophilized) of smolting Atlantic salmon as a function
of dietary soybean meal. P(linear) signifies P-value for linear regression analysis. Values in each
row awarded common superscripts are not signiﬁcantly different (P > 0.05).

Digesta Dry Matter

Large intestine digesta dry matter content (lyophilized) showed differences across
dietary treatments and provided evidence of a negative relationship between dry matter
content and increased levels of dietary SBM (Figure 6.2). Digesta sampled from the
0%SBM control had a signiﬁcantly higher dry matter content than dietary groups fed
20% SBM diets or greater. The observed negative linear response (linear, P = 0.0001)
between intestinal digesta dry matter and SBM was as follows:

Large intestine digesta dry matter = -0.20 x %SBM + 23.74 (R2 = 0.72).

96

Table 6.4 Hepatic somatic index (HSI), intestinal trypsin activity (TA) and whole body energy
composition of smolting Atlantic salmon fed diets varying levels of soybean meal and fish meal.
Mean standard errors are in parentheses. MNR-98HS is the open formula control diet. Test diets
are identified according to percent soybean meal (SB) and percent fish meal (F). Values in each row
awarded common superscripts are not significantly different (P > 0.05).

 

 

 

Diet HSI TA E
(111919) (MJIKQL
MNR-98HS 1.11 a 2.03 23.77
(0.08) (0.30) (0.19)
SB5IF30 1.09 a” 1.49
(0-02) (0.08)
$320II=30 1.12 a 1.87 ..--
(0.05) (0.21)
$3201F24 1.03 a” 1.64
(0.04) (0.43)
$825lF30 1.09 a” 1.73
(0.02) (0.13)
SB30IF30 1.02 .1. 1.18
(0.01) (0.20)
SB30IF24 0.96 b 1.37 23.95
(0.04) (0.23) . (0.09)
T rypsin Activity

Trypsin enzyme activity was slightly greater for the control group than all other
dietary treatments (Table 6.4). TA was signiﬁcantly greater for the control than all other
treatment groups at P _<_ 0.10 signiﬁcance level, but no signiﬁcant differences were

detected at P 5 0.5.

Hepatic Somatic Index

Hepatic somatic index was signiﬁcantly greater for the control than the groups fed

SB30/F24 at P 5 0.05, and SBZO/FZ4 and SB30/F 30 groups at P _<_ 0.10 (Table 6.4). No

97

signiﬁcant differences in HSI were observed between the control, SBS/F 30, SB20/F 30, or

SB25/F30 groups.

Plasma Insulin

Plasma insulin concentrations for all treatment groups fell below the minimum
detection limit (30pmol/l) required by assay protocols (MSU’s Diagnostic Center for
Population and Animal Health Laboratory). As a result, no differences in plasma insulin

could be attributed to dietary treatment effect.

Intestinal Histology

Histological sections of small intestine from representative ﬁsh in each study
group were examined microscopically. The villouszcrypt ratio was uniformly in the 2:1
to 3:1 range for all ﬁsh in all study groups. Neither the mucosal epithelium nor the
microvillous brush border exhibited attenuation or discontinuity in any ﬁsh. Only one
ﬁsh, from the SB30/F 24 group, exhibited mild lymphocytic-plasmacytic inﬁltrates within
the mucosal lamina propria. According to the histological examination report this type of
defect was likely to be of an infectious cause rather than an effect of the diet. No
signiﬁcant microscopic alterations in the gastrointestinal tract were observed from any

treatment group.

Muscle Tissue Amino Acids

Muscle tissue amino acid proﬁles (Table 6.5) were signiﬁcantly different (P S

0.05) between the control (0%SBM) and SB30/F 24 dietary groups for 14 of the 19 amino

98

Table 6.5 Muscle tissue amino acid compositions (umol/g) of smolting Atlantic salmon fed a control
(MNR-98HS) and a 30% soybean meal (SB) 24% fish meal (F) diet for 22 weeks. Values with
asterisk (*) were signiﬁcantly different (P < 0.05).

MNR-98HS SB30IF24

GLU 2.62 2.80 *
HYP 5.10 5.75 *
SER 3.73 3.89

ASN 3.35 3.70 *
GLY 12.56 12.04

TAU 3.41 3.71

HIS 7.58 5.10 *
THR 3.95 3.98

ALA 5.19 5.38

PRO 3.05 3.28 *
TYR 2.86 3.07 *
VAL 2.91 3.14 *
MET 2.98 3.22 *
ILE 2.91 3.13 *
LEU 3.66 3.99 *
PHE 3.07 3.31 *
TRP 2.97 3.19 *
ORN 2.96 3.20 *
LYS 3.51 4.04 *

acids analyzed. Only one amino acid, HIS, was higher in ﬁsh fed the 0% SBM control.
Fish fed the SB30/F 24 diet had higher levels of 12 of the remaining 17 amino acids than

ﬁsh fed the 0% SBM diet.

Mortalities

No correlations were found between average mortality per tank and level of SBM
in the diet. Survival in each tank varied from 60% to 100%, with a 92% overall survival
rate. The control diet had the highest number of mortalities with 8 (Figure 6.3).
Mortalities did, however, decline considerably over time. Of the 36 total mortalities, 23

occurred over the ﬁrst one-third of the trial (28 days), 10 occurred over the middle one-

99

Mortalities

0. Q
(8' «“9 «‘5
0 we go)

6&0
q. '1. ’b
65‘” 63’ 98 98’

 

Diet

Figure 6.3 Mortalities by treatment (Diet) of smolting Atlantic salmon over the lZ—week feed trial.
MNR-98HS is the open formula control. Test diets are identiﬁed according to percent soybean meal
(SB) and percent ﬁsh meal (F).

third of the trial (day 29 - 56), and only 3 mortalities were observed over the last one-
third period of the trial (day 57 - 85). No mortalities were observed over the extended

lO-week feeding trial.

Discussion
Feed Compositions

Differences in formulated versus measured diet compositions (Table 6.1) could be
attributed to intrinsic and extrinsic factors as identiﬁed by Gizzi and Givens (2001).

Intrinsic variability is that caused by real differences among feedstuffs (genetics,

100

processing techniques, etc). Extrinsic variability is caused by differences in sampling and
analysis procedures. The diet formulation model (Chapter 5), slightly underestimated
actual, or measured CP and CF levels. The model also slightly overestimated ash and
carbohydrate compositions in the feed. Future applications of the diet formulation model
used in this study should account for intrinsic variability to the extent possible and
minimize variability arising from extrinsic factors. Potential improvements include

testing individual diet ingredients and additional review of model input parameters.

Growth Characteristics, Feed Conversion and Protein Retention

Atlantic salmon smolts showed excellent growth characteristics across all dietary
treatments. No differences were observed in ﬁnal weight, SGR, FCR or PER. These
results were encouraging since nearly all previous studies had shown depressed growth of
Atlantic salmon fed SBM based diets with approximately equivalent FM replacement
rates. For example, Refstie et al. (2000) reported a gain of 44% grth of Atlantic
salmon fed 7 0%FM over those fed 32%FM/30%SBM. Those researchers also reported
much lower SGR values than those observed in this study. Krogdahl et al. (2003) found a
negative dose-dependent effect of SBM on growth parameters of Atlantic salmon fed
extruded diets containing up to 27% SBM (3 5% of total protein). Differences in diet
formulations, dietary ingredients and processing methods were likely causes for the
differences observed between studies. Mean k values were slightly different, but this

difference does not appear to be an effect of feed SBM content (Table 6.2)

101

Proximate Body Composition

Body lipid composition data provided strong evidence that SBM diets can have
long term impacts on Atlantic salmon. Results from this trial showed a similar negative
does-dependent response of lipid retention to that observed by other researchers
(Storebakken et al. 1998, Krogdahl et al. 2000, Reﬁsie et al. 2000). Those authors
theorized that salmon were unable to digest SBM Oligosaccharides, which are alcohol
soluble, and also non-starch polysaccharides (non soluble). A similar trend, however,
was not observed by Carter and Hauler (2000) in a comparison of high energy diets
containing 0%SBM 60%F M and 22%SBM 45%FM. Lipid utilization appears to be an
important function affected by consumption of SBM diets by carnivorous ﬁsh.

No signiﬁcant differences were detected in body dry matter, protein, ash content
(Table 6.3) or energy (Table 6.4). These results agreed in part to those of Krogdahl et al.
(2003) who found no differences in Atlantic salmon protein and ash body compositions
fed up to 27% SBM (35% of total protein). The previous authors did, however, show a
negative response of body dry matter compositions (P = 0.007) and body energy (P =
0.004) to increasing SBM levels. Neither body dry matter nor body energy was affected

by SBM content in the study reported here.

Digesta Dry Matter

Figure 6.2 indicates a highly correlated functional response of decreasing digesta
dry matter content to SBM level in diets for Atlantic salmon. In general, large intestine
digesta increased in water content with increased levels of SBM. This ﬁnding agrees

with results reported by others (Storebakken et al. 1998, Reﬁsie et al. 1999, 2000, 2001).

102

Reﬁsie et al. (2000) attributed reduced fecal dry matter and reduced lipid retention in
Atlantic salmon to non-starch polysaccharides in SBM, while Amesen et al. (1989)

attributed a similar condition to alcohol soluble carbohydrates.

T rypsin Activity

Lack of differences in TA across diets suggests that soybean TIs were either not
present in sufﬁcient quantities to impair trypsin enzyme activity, or were destroyed by
heat processes during diet manufacturing. This ﬁnding agrees with results from the
previous TI studies (Chapters 3 and 4) which failed to show direct response of TA to
SBTIS up to a level of 60%SBM equivalency. Our data also supports others who have
reported that TIs are heat liable and can be destroyed or reduced by diet processing

methods (Wilson 1992, Anderson and Wolf 1995).

Plasma Insulin and HSI

Increased levels of indigestible carbohydrates in diets of carnivorous ﬁsh have
resulted in high glucose and insulin levels (Hemre et al. 1995, Cowley and Sheridan
1993), increased liver size and weight (Hilton and Atkinson 1982) and increased hepatic
lipogenesis (Brauge et al. 1995, 1994). Conversely, increased levels of digestible
carbohydrates in feeds for rainbow trout increased plasma glucose concentrations (Bergot
1979), and increased gelatinized starch intake has shown protein sparing effects for the
same species (Kaushik and Oliva—Teles 1985).

Effects of the SBM practical diets on plasma insulin in this study were

inconclusive. Plasma insulin concentrations for all treatment groups fell below the

103

minimum detection limit (30pmol/l) required by assay protocols (MSU’s Diagnostic
Center for Population and Animal Health Laboratory). Reserve plasma samples were
destroyed when the cold storage unit holding the samples malfunctioned and the samples
thawed for an extended period of time. Thus, plasma insulin tests could not be repeated
after the initial assay.

Fish fed the control and SBZOF30 diets had higher HSI values (P S 0.05) the ﬁsh
fed the high SBM low FM diet (SB30/F 24). No other differences were observed with
HSI. Increased HSI (i.e. liver to body weight ratio) might be an indicator of lipid build
up in the liver. Thus it does not appear that ﬁsh fed high SBM experienced high
lipogenic activity. Rather, lower HSI values combined with decreased liver weight and
body lipid suggested more of a protein sparing condition. According to Buhler and
Halver (1961) the balance between lipid and starch affect protein sparing in Chinook

salmon. Conceivably, use of HND diets aided carbohydrate utilization in the SBM diets.

Intestinal Histology

No diet effects were observed relating SBM to problems in the small intestine.
Distal intestine (DI) sections, however, were not examined by histological methods.
According to Baeverﬁord and Krogdahl (1996), the D1 is the primary region where SBM
induced enteritis is likely to occur. In mammals, the vast majority of absorption of feed
materials takes place in the anterior small intestine, while the hindgut is more for
bacterial fermentation and ﬂuid absorption (Dr. Scott Fitzgerald, Michigan State
University, personal communication). Samples sent to MSU’s Pathobiology and

Diagnostics Center were analyzed using similar protocols to those previously done by

104

USB project collaborators at Purdue University. These protocols concentrated on the
anterior intestine. Focusing on small intestinal histology instead of D1 was an

unfortunate oversight of this study.

Muscle Tissue Amino Acids

SBM induced changes were observed in muscle tissue amino acid proﬁles (Table
6.5), but not on growth, protein utilization efﬁciency (PER in Table 6.2), or protein
composition (Table 6.3). Thus, it would appear that the SBM diets in this study had a
direct impact on various metabolic pathways without impairing growth rates. Krogdahl
et al. (2003), stated that lack of diet effects on nutrient and energy deposition suggested
that the main effect of SBM on nutrient utilization can be attributed to digestive rather
than metabolic processes. In that study, however, amino acid deposition was not
assessed. Based on our ﬁndings, dietary induced changes in energy utilization as
observed with changes to body lipid compositions could in turn cause alterations to

physiological processes involved in tissue amino acid deposition.

Mortalities

Based on our results there is no evidence that mortality rates were correlated with
increasing SBM content in feed. Rather, the declining mortality rates over time would
suggest that factors such as initial transport stress, handling stress and/or smolting were

more likely causes for the 36 mortalities observed over the course of the study.

105

Chapter Summary

This study was the ﬁnal feed trial of an integrated research project at MSU
designed to develop a commercially viable practical SBM diet for Atlantic salmon. The
main goal of this study was to determine maximum SBM replacement levels for FM in
practical diets for Atlantic sahnon. High nutrient dense practical diets (55% CP, 20%
CF), containing graded levels of SBM were fed in triplicate to one-year old Atlantic
salmon for a period of 12 weeks. A control diet of 0% SBM and 30% FM was compared
to test diets containing 5, 20, 25, 30% SBM and 30% FM. In two additional diets (20 and
30% SBM), the level of ﬁsh meal was reduced by 20% to that of the control. At the end
of the trial, Atlantic salmon receiving the control and the 30% SBM/reduced FM diets
were continued on their respective diets for 10 additional weeks to assess SBM effects on
long term muscle tissue amino acid compositions.

Study results indicated no differences in growth, protein or energy retention, feed
conversion or large intestinal TA for Atlantic salmon fed HND diets. Negative linear
functional responses were observed between SBM content, body lipid composition and
fecal dry matter. Changes were also observed in muscle amino acid composition but no
detrimental affects of dietary SBM on Atlantic salmon were found. Results suggest that
HND diets containing up to 30% SBM provided adequate energy to facilitate protein
sparing mechanisms in Atlantic salmon; however, SBM induced changes occurred along

various metabolic pathways and into muscle tissues.

106

Chapter 7 — Project Summary, Conclusions and Recommendations

Project Summary
This study was part of an integrated research project designed to develop a
commercially acceptable SBM based formulated feed for Atlantic salmon. This research

was conducted to test the following hypothesis:

 

Based on literature review and recent unpublished study data the maximum level
of SBM incorporation into formulated diets for Atlantic salmon is expected to be
in the range of 20-30% wet ingredient weight without adverse affects on growth,

feed efficiency, and/or other observable ﬁsh health characteristics.

 

 

 

Results from trypsin inhibitor feed trials indicated that SBTIS of 60% SBM
equivalency had no signiﬁcant affect on growth or feed efﬁciency. Slight changes were
observed in short term trypsin concentrations and possibly long term body compositions.
Data provided evidence of ability by Atlantic salmon to compensate for up to 60% SBM
equivalency TI levels in semi-puriﬁed diets; however, questions remain as to whether
such mechanisms would be indicative of signs of nutritional deﬁciencies.

Nutrient requirements of Atlantic salmon were reviewed and incorporated into a
diet formulation model developed speciﬁcally for that species. Using this model
optimally balanced HN D diets containing SBM were formulated for Atlantic salmon
using commercial-based, practical ingredients. Formulations above 30% SBM became
increasingly more difﬁcult to maintain proper balances between selected protein, energy,

and amino acid requirements. Practical diets containing 0 to 30%SBM were processed

107

using methods determined optimal for SBM diets for a closely related salmonid species
(Barrows et a1. 2004). Based on results of the practical feed trial, no single parameter
including mortality, growth, protein or energy retention, feed conversion nor TA
provided evidence of nutritional induced stressors. Moreover, SGR was considered very
good and slightly improved over the SBTI grth trial. Mortalities were highest for the
0%SBM control and regressed as the smolts appeared to complete physiological
transformation. No mortalities were observed over the ﬁnal 10 weeks of a 22 week
combined study. Changes in fecal dry matter and body lipid retention were observed in
direct response to the level of SBM in the diets. Modiﬁcations in long term amino acid
compositions were also observed. Effects of HND SBM diets appeared to be a function
of physiological protein sparing mechanisms, and results provided little evidence that the

diets were nutritionally deﬁcient.

Conclusions

Based on the results of this research, the following conclusions were made:

1. Puriﬁed soybean Us in experimental diets for Atlantic salmon had no effect on
growth to a TI level of 60% SBM equivalency.

2. Slight effects of dietary puriﬁed TIs were observed in body composition and
trypsin enzyme activity. These effects may be due to physiological compensation

factors by Atlantic salmon.

‘ 108

3. Trypsin inhibitors in practical Atlantic salmon diets containing 30% SBM are
sufﬁciently reduced or destroyed by heating processes in SBM and diet
manufacturing methods.

4. Atlantic salmon fed HND practical diets containing 30% SBM 24% FM had
similar growth and feed efﬁciency rates as those fed the open formula control diet
MNR-98HS.

5. High nutrient dense diets containing up to 30% SBM provided adequate energy to
facilitate protein sparing mechanisms in Atlantic salmon, however, SBM induced
changes occurred along various metabolic pathways and into muscle tissues.

6. Observations from this research suggest that, at least on a short term basis
(example: last several weeks of grow-out), HND diets containing 30% SBM 24%
PM may be commercially viable for Atlantic salmon production providing a

slightly leaner ﬁsh is a desirable product by consumers.

Recommendations

Further testing of the practical diets developed within this study is recommended
on Atlantic salmon and other salmonids prior to commercial use. Although no signs of
detrimental health were observed at inclusions of 60% SBM equivalency T1 (3.9 gTI/kg)
in experimental diets, or 30% SBM in practical diets, additional feed trials should include
distal intestine histology, plasma glucose and glycogen assays, and examine SBM effects
on endocrine hormonal controls. Use of HND diets may be a key factor in overcoming

SBM anti-nutritional factors observed on salmonids. Relationships between dietary

109

energy, lipid and carbohydrate utilization, and protein sparing mechanisms by Atlantic

salmon should be further explored.

110

APPENDIX

lll

o ...N
cod
cud
0mg:
and
OWE.
owd
05.0
3.;
2mm 0

omé
cud
0N6
o...N
on:
8...
o¢.m
om.o
33
..2E 0

cow

of
Se
cod
8.?
Ed
one
cad
of
3°C

83 .282 8:238» 85 - CREPE...

ovdw
ova
enamp
Sham
omdw
ends
oodm
o P dm
3%

Q0

omom

coma
Nmmm
vwnm
vmmm
movm
mev
comm
mmwv

ages;

m0

5mm
was
mﬂw
no
mo
ream
mdw
02.
ex;
an.

5
8
8
8
8
8
8
5
3e
.20

0.35

828.20 959.0
2282.... 52285
05:250.).

:0 cmE

3&9 seen .985
came 5an 5595
40:859..

.moE 556 :50
38> mamasocm

>oc>>

.88 6335.3 bzaon
_mos. coca

_moE >>ococ<

2mm

.88 c235 5055
«2.2029...

28:00 233:2 E20935

112

00.0
00.0

00.N
00.5
0.0
00.0
0N0
300
_m>

00.0
«0.0
0 F .o
00.0
00.5
05.0
00.0
3.0
$00

9.—

SN
30
ad

3.0
8..»
New
8..
8..
3.0

.E

00.0
0.5

00.0
50...
00.0
#00
05.5

3.0
5

00.0
N0...
00.0
5.0

00.0
05.N
30
00.0
300

9.5

N;
8.0
«.0

3.0
«we
00

Ed
80
3.0

90

98 382 802258 05 - 502mm,?

005

8.0
8.0
2.0
E
8..
8.0
8.0
mm;
3.0
.02

005

:#
50.N
0.0
00.0
00.5
v0.0
00.0
0N...
300

N05
00.0
05...
.....v
00.05
00.0
00.0
00.0
3.00
:0.—

00.0
00.5
05.0
0.0
50.0
55.0
30
00.N
300
m:

50.5
00...

00.5
3.0
50.?
NN.«
00.?
0x;

0:...

00.0
00.0
v0.0
00.0
05.0
00.0
50.0
3.0
3.0

02

0.50

0000.20 05.020
0.0200050 820.005
0500.522

=0cnu

§~.e 5890 .9052
$.08 5520 2695
0018.03

.005 0220 E00
.0005 90390

>mc>>

_moE «00090.50 9.000
.022 000.0

.09: 56:22

.200

.006 005.0 000:;
300.080...

22:00 055:: £285...

113

05

0.0
5.0
5.0
3
0.0
0.0
0.50

6.050
000.2

000
00 _.
000
00.0
0050
000
..0 _.

6.05.
on.

5.00
0.00
0
05.3
0.0
00.0
0.00

.305.
000

50.0
00.0

00.0
00.0
00.0
3.0

$00
:0

00.0
50.0
00.0
00.0
00.0
_..F

_.0.0

3..
02

50.0
00.0
05.0
0...0
0 5.0
00.0
0.0

300

50.0
50.0
50.0
00.0
00.0
_.

3.0

$00
2.5.5

05.0
00._.
00.0
00.0
0 ...0
0.0

0 ...0
00.5
300

“an.

00

3.0
03
8.0
8.?
no
9.“
8.0
:o
.00
donn—

93 .0002 000200000 .05 - 59200.2

0.50

8.8.5 2.05
220850 52200.0
8.8.50.2

..o 5.“.

£00. 5.80 .98....
3.08 550... 550.5

61.0%....
50.0 .000. 005.0 500
3.0 000.” 0.03000
8.0 >22,
50.0 .000. “0:00.043 .3500
3.0 .002 000.0
m; .85 0652
00.0 5.00
«Na .85 C220 .822,
...... 35.85:.

00
«03:00 «00502 «00.00.00.

114

0.0..

3.00..
v.

0.00
...0

0.0

0
0.0

30.0....
m

00

0.0.0
0.50
0..
000

.205.
«E

0.00

0.0
0.00

0.00

0.00

0.0
.200

3.0
00.0
00...
0.0

80.05.
.62....

0.00?

0.0
0.0:
00.00
«.5
0.0
0..
0.:

30.05.

02.980 0.8.2

000 N.m_.

00 0.0

00¢ 0.0

NV Fwd

NN 0.0

00.. N6

NN No
A0¥0EV 80:00...

0.00.0".

000

0000
000?
0000
000

003
0000

6.0.0.
05.2.0

2.0
00.?
00.0
00.0
00.0
00.0
00.0

80:00..

56.0

05 .252 8.0.088... 0.0 - 0002000....

00..

F0
00
0.0
0v
000
00..
00

6.05.
5

00.0.0

0000.00 05.000

0.0000000 0.5.0.005

00.02505.

..0 00.“.

.0000. 5.00.0 .0055.

03.0. 0an 0.50.5

400.20.:

00.0 .000. 00.0.0 E00

.00 .002. 9030.0

>0c>>

00.0 .00.: 60090.3 0.30..

.005. 000.0

00.. .000. >>0000<

..o 500

.000. 00.0.0 .0055

.0é0E. 800.0800.
00

20.000 .0005... 50.00.30.

115

0.... 00.. .00.
0 0 0.

0 0 0

0 0 0

0 0 0
00.0 00.0 00.0
00.0 0.0 0.00
00.0 ~00 00.0
8.0 2.0 «0.0
000 .00 00.0
~00 00.0 00..
0 0 0

0 0 0
3°. ...... 00.

0000 .0000 00

3.0000

0.03;
0

0

0
00.000.
000:
00.000
00.50
00.000
00.03.
0

0

008..
00

00.;

00.00

0000

00.0—
00.0
00.?
00.0
00.0
0. 00

......
00

0.00 .0005. 00020000... .05 - 0052000...

00...
00.0
00.0
.00..
0.0.0.0
00.0
00.00
00.00

00.0000

0..

IDOOOOO

2888§02

000

o

o

0&0
.0000.<

AQovgo ... “02
.0... 000 000.0
...... .00... 000.0
3... .0. 000.0

00. 0.0.0.0 000.0
0.400....

00.00

.0022. 00

00m.
0.50

0090

0000.00 05.000
0.0000000 0.00.005.
0500.505.

..0 00.0

50.0.0 .9052
50.0.0 50.0.5
610%....

.000. 00.0.0 0.00
.000.» 0.02.0.0

>055

.000. 0000.005 0.500
.002 000.0

.005 00.0

5.00

.000. 00.0.0 .0002.
300.020...

0:00.122

000.00.05.00 .05

116

m0...
00d

0omd
00.0

0.0
50..

GOOD

N06
0:.
mod
mod
0N6
mad

.0...
....

 

 

 

 

 

 

0.0 00..
00.. 00.0 .00 00..
0
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0 0
00.0 00.0 00.0 ....0
00.0 00.0 ~00 00.0
00.0 00.0 00.0 00.0
...0 0.0 00.0 .00
0.0 ....0 00.0 00.0
.00 00.0 0.0 00.0
0 0 0 0
0 0 0 0
3... ...... .00. ......
.0. 000 0.0 .05.

.000 .0002 00.00.0000”. .20 - 000200.?

..N 0.0.
00.0 00.0
o o

o o

o o
0.0 o
00.0 00.0
m ..o 0.0
00.0 ....o
0 ..o 00.0
«0.0 00.0
.0. 00..
o o

o o
.00. .0.
0.... 00..
02.0.20...“— .05

o...
00...

0000

000

506
0rd
nod
m0...

......
0_.

0.0
mm...

COCO

mod
cod
nod
and
00.0

3...
0.:

...... + 000 u 000. .0. 00.00.3000
0>0 + .02 u .02 .0. 0.00.9503.

0N
..mN

0000

5.0
:6
mod
0N6
mm...
0.....

......
0.<

0.E 30.00 .20 .00. - x

03.
Saw

0090

0000.00 00..000
0.208.... 0.00.005
00.02505.

..0 00.“.

00.0.0 0.00.5.
550.0 0.0.0.5
001.00.»...

.008 00.0.0 E00
.002. 0.0.50.0

>0c>>

_00E 80090.3 .....0on.
.005. 000.0

.000. 00.“.

Saw

.000. 00.0.0 .000>>
0.00.0200.

0100.122

117

or
3.05

on

mmé
3.0
vmd
mod
mvd
mwN

o
63.5
dams.

om
3».me

COMO

mNKm
min
ode
Nm.oN
mwdmr
00.00

c
axaé
o...

m
no.2

OONO

Na...
mvd
mod
nmd
{MN

o
axaé
goo

o
and

0000

55
ed
and

0000

3.0
cod
no.0
mod
mod
and
o

o
3.;
m2

X
vod
Ed

COCO

Nod
55
Nod
56
8.0
So
o

o
3
m2

X

0
0nd

0000

No.0
cod
and
mod
mod
and

o

0

30V
QCtOF—O

sown—35.5“. «20

SE
wd
mod

00°C

85
86
«No
85
:3
Rd
0

o
g
an.

ad
of?

.mocm

SS 382 SEEEE 35 - 592mm,?

31‘

E: 26.3 2m .08 - x
com
Eam

0.55

023.50 950:0
2239:. 52285
mchEEE

:0 5E

x755 .9952
AEBQEESS
6:659:

.35. 92:3 :50
33> ﬂmamﬁ

>oc>>

.99: 8335.3 Essen.
Ems. voo_m

.88 2mm

2mm

.92: 556 ammc>>
3:23.55

9.5?sz

118

N?
00.0..

(D
‘—

9.350

X

0
00.x.

o
3360
m

«No.0
00.3?

o

0

on

o

8.0
mod
8o
8.?
So
some
o

o
A9350
NE

0..
ONKm

0N0

00.0
0N6
00.0
F00
No.0
00.0

3&9»:
m2...

VN
50.3

Ndh

00.0
PEP
00.0
00.0
0N0
0 rd

0
3:95
09m.

0w
mném

Nov

mné
00;.
00.0
0N0
5.0
00,—.
0

o
axaev
x8?

x

00
006:

0

0
v.03
0
00.0
3.0
00.0
no.0
NN.0
006
0

o
6350
u_o<-scma

x

x

00*
00.2.5

0

0

0.0

0
00.00
00wa
00.00..
vs. 50
00.Nv
0v.NNm_.
0

0 0
Bios: 8x380

our NF

5. Se 8.?
o o

o 0

8m «.2

o o
8.? mod
3% m3
cod 8d
«3.. mod
3; 8.0
8.8 mod

o o

o

63.5

5092 58.0“.
5.5350“. 36

05.020

Né
NNé

01-00

00.0
00.0
No.0
90.0
No.0
no.0
0

o
6350
565

83 $on songssuom 35 - 592me

no
50.09

00..

mnK
mm...
00.0
wNK
NV. «N
00.00

3&9»:
:N

0.0
N50

.00

5.0
00.0
00.0
00.0
00.0
3.0

3x380
ww

8E-x
0mm
:50

5:. 25.8 2m

0.5m

3:020 959.0
92395 52285
0:60.502

.6 car.

5an .8252
529a SEES
40:97.3

_moE :220 :80
“mag 90390

>23

_mmE 6:00.93 >238
.85. 820

_mmE 2mm

.200

.85 :920 “was;
8.3.0205

w 100.12!

119

00.x

0000

and
8.0
3.0
Ed
:2
mm...
mod
......
58

00...

COCO

..N.0
9.0
No.0
20
no.0
00.0
to
v0.0
......
.0000

3.3

Eumw

and
mod
mod
and
mod
8..
mod
:5
......

“.0

VN.0N0¢

3.89
o

o

o
8.25
2.0:
$63
32.8
«2.8
8.3...
3%.
8.2%
:8...
mo

5.0V

0
0
o
0
05.x.
00...
00.?
SN
..mé
the
EM
5.0
3.0
10

£233.50". «05
95 .252 8.33.59. .05 - 502me

5.5

50000
and

2d
8..
won
em...
a. a
8......
am
.00
no

no...
00.x.
00.?
3.0..
5.5
3.4.
50. ..N
rN.0N

00000..

F

0001:0000
-
IO
r

COCO”
NCOODIDV'
F

com

on

on

9:0
EsoE<

0096 + 62
...... %m 820
...... .8... 820
...... a. 255

0x0 5905 820
02.9..

memo

6&3. mo

3m
E30

0466

m6.5.50 0::20
2288.". 62285
3.5.56.2

:0 ..mE

x.E9a .8055.
AESQEESS
401.0593

.3... c920 :50
.80.. 99590

>m5>>

_mmE 63099.3 Each.
.85. 08.0

_mmE 5.“.

20m

.93.. c920 .855
3.3.020...

own—3mm

120

00...
8N

30000
0'

N...0
00.0
:20
N00
004
2.0
NN.0
3.;

_m>

VON.0
N00

0000

mod
Nod
8d
35
mud
8.0
8.0
300

eh

oé
00.?

COCO

3.0
00.0
00.0
0N0
00.0
00.0
00.0
300

E...

 

 

 

 

_..N
mod

 

 

mm mm;
No; «in «3 m:
o
o o o o
o o o o
o o o o
o o o o
:8 5o :6 «No
3o 85 mod mod
mod 8.0 85 mod
:d 2.0 mod 5o
86 :3 8.0 8.0
$5 86 $6 85
8.0 who Ed mod
so mad 2.0 8.0
E: 3.0 g 3.;
5 of. go as.

33 382 cousspuom Em - 592me

0000

who
26
8.0
who
and
:5
96
mod
30

2:

gangs—Eon— «0.0

0.N
05v

0000

o:
Ed
:a
mud
Sue
N?
Ed
2.0
g

:3

0...
00.N

00°C

2.0
N00
30
50.0
00.0
:0
..N.0
300
m:

0.0
Nm._.

COCO

mod
cod
.35
8.0
23
8.0
:d
3...

.2:

SF + of u man. .8 Eoonacmm
£6 + 52 u E: .8 EwEQEcom

0N
mvN

COCO

2.0
mod
ad
85
o:
”to
Ed
3:

9<

58 25.8 5.0 «m5 - x

com
.26

0.36

$2.630 950:0
2238.". 522.85
058285.

:0 2mm

5:85 .8952
x_Eo._Q E895
81.853

.88 55.0 :80
0mm?» £9520

>23

_moE 8:09.93 $30....
.82 coca

.35 5E

20m

_mmE 55.0. «355
3.3520:-

own—Bum

121

0..
00.55

05

00.0
V0.0
v0.0
00.0
3.0
00.0
00. r
00.0
.986.
.592

00
00.500

0000

2.8
88
8.8
8.8
8.8.
8.8
8d
8d
.985.
on.

0
0N.0.‘

OONO

«we
mvd
00.0
50.0
F50
No;
8.0
.985.
goo

0
V0.0

0000

00.0
00.0
00.0
00.0
0.20
00.0
00.0
.8.

.20
0.0
0.00

0000

5d
8d
8d
8d
8d
8d
8d
8d
3°.
mz

x
#00
0 rd

0000

Ed
3d
«ed
8d
8d
Ed
8d
8d
30.

as.

x x x
d dd 8d
Ed 8d 8..
88d
d d d
d d d
d d d
d d d
3d 8d 8d
8d 8d Ed
8d 8d 8d
8d 8d 2 rd
8d Ed 8d
8d Ed 2d
8d 3d 8d
8d 8d 3d
8. $0. 8.
8:25 an. .85.

cosmicton. «20

35 .082 883.58 .05 - x5208...

0v. ..

COCO

5.0
3.0
00.0
..N.0
00.0
0...?
..0.0
No.0
.8.
00

5:. 30.2. .20 5E - x
cam
:50

0.88

00.3.20 05.2.0
9288.". 8.285
3.5.522

=0 5.“.

8an .2055.
AEEQEEES
81.8.93

.mmE c920 £00
.305 29590

>mc>>

.mmE 8:85.50 Egan.
.85. 08.0

.99: :2“.

.200

.8... :920 .853
3.3.3.0...

00E0mm

122

N..
00.0..

(O
‘-

000000000000

A

985.

X

0
00.0

0000

3.0
00.0
050
00.0
00...
550
00.0
.905.

0.0.0.0
00.00..

d
d

8

d

8d
8d
8d
8d.
.0d
8.8.
8d
8d
.985.
«.0

0..
00.50

000

8d
00...
8d
5d
8d
8d
9d
8d
.985.
a...»

00.00

d

d
«d.
d
8d
E...
8d
8d
8d
d ..0
3d
8d
.985.
8E

0F
00. 0

0.00

00.0
00..
00.0
00.0
5.0
00..
00.0
00.0
.98....
.805...

x

00
50.05?

8d
.985.

0.o<.a.:0n.

8. V m.
0.08 8d.

d d

d d

d8 «.2

d d

8d :3
0.00 8d
8d 8d
8.0 8d
8.. 8d
8.8 8d
0... 3d
8d 8d
.985. .985.
2.8.2 58.0.
5.5250. .20

x

000
00.0500.

d

d

0.0

d

3.8
8.8.
8.8.
.... .8
8.0..
8.08.
8...?
8d
8.8....

05.020

Né
..Né

OPOO

00.0
00.0
00.0
5.0
00.0
50.0
00.0
00.0
.0035.
5.0.0

80 082 88.88. .20 - 00020.02

00.00..

d
8.
d

d
0.....
8..
8d
8..
N... .N
8.8
8...
8d
.985.
..N

0.0
00.0

3.0

......d
8d
8d
8d
8d
...d
:3
8d
.985.

.02. - x
80
5:0

...... 26.00 .20

0.50.0

00.20.20 05.020
0.08820 80.80
05:52.05.

..0 20.“.

52.0.0 .8055.
52.0.0 52.0.5
001.052...

.00E 20.0.0 E00
.0005 0.02.0.0

>022.

.005 80090.52 5...:00
.002 000.0

.005 20.“.

200

.005 20.0.0 .0022.
38.028.

00...:000

123

50.5

0000

8d
8d
2d
3d
8.0
0...
8d
......
280

00.5

0000

mcd
09.0
rod
—.—..0
00.0
00.0
$0.0
#06
A00.
200:0

50.05

0.0..

3d
8d
8d
8d
8d
8..
Ed
..d
3...

.0

00.550

8.8...
d

d

d
2.8...
8.8.
«.8.
8.8.
8.8m
8.8.:
8.80
8.98
89..
00

05.00

0000

8d
5..
8d
8.0
8d
Ed.
8d
.0...
......
.0

2030.353“. «0.0

Sun .0002 sausage“. .0_Q - 592mm?

00.00

0000

8d
8..
ﬂd
80
Ed
.0. .0
8d.
8.0
......
.0

50...
50.5
00.?
50.3
00.00
0..?
00. ..N
00.0w

00.000.

0.00

00

00

00

0.00
000
000

05

00:0
.5022

Ao\ovm>0 + .02
.... 28 08.0
...... .82 08.0
...... .m. 08.0

...... ...99. 08.0
020.3

00.0

.9222. 00

00..
2.00

0.8.0

0020.20 05.020
0.088... 2.22.0.0
05252.05.

..0 20....

52.0.. .8055.
55205595
201.050....

.002. 20.0.0 2.00
.0005 0.0.50.0

>022.

.002. 8:00.93 52.00.
.005. 000.0

.002. 20..

.200

.002. 20.0.0 .0055
820.0202.

000.0000

124

x x
00.. .000
00.0 000
0 0

0 0

0 0

0 0
0.0 000
.0 0 000
00.0 .00
...0 000
0.. 0 00.0
00 . 000
.0 0 ...0
«N 0 000
...... ......
_0> 0..

0..
00...

COCO

«0.0

00.0
00.0
NN.0
00.0
00.0
00.0
...0.

.2...

 

 

 

 

 

 

x x x x
0.0 00..
00.. 00.0 00.0 0...
...0
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0 0
000 00.0 .00 0.0
000 00.0 .00 000
.00 ~00 ~00 .00
000 ...0 .00 00.0
0.0 000 ..00 000
.00 00.0 0.0 000
000 00.0 0.0 ...0
00.0 00.0 0.0 00.0
...... ...... ...... ......
... 02a. 0.20 .05.

0.00 .0002 0000.08.00 .20 - 00020.0...

x x
..N 0.0
00.0 00.0
0 o

o 0

0 0
50.0 0
50.0 00.0
00.0 00.0
00.0 00.0
0 ...0 5.0
3.0 00.0
...... 00..
00.0 05.0
00.0 00.0
...... ......
0.3 00..

2030.053“. .05

0..
SN

0.0
..0...

...... + 02... u 02... .0. .2060..000m.
9.0 + .02 n .05. .0. .20E0..000m

0N
05.N

2.2. 30.02 20 .02. - x

002
2.00

0.00.0

00.20.20 05.020
0.9.0820 2.0.0.005
02.20.50...

..0 20.“.

52.0.0 .0.02.s.
x.E0.0 2.2.0..>
002.020....

.00.: 20.0.0 2.00
.009 0.0.50.0

>02>>

.002. 0000.93 0200,...
.00.). 000.0

.002. 20.”.

5.00

.002. 20.0.0 .002>>
0.20.0202.

emu—Bum...”

125

0.
00.0

05

50.0
00.0

m ..0
00.0
00.0
00.0
3.5
00.0
.0005.
.2005.

00
0.0.0

00000

00.0.
.00
00....
0..-
0.00.
0000
0~0~
00.0
.000...
0.

0
00d.

OONO

0 ...
00.0
.50
00.0
.50
00...
00.0
.0005.
000

0
00.0

0000

.0.0
00.0
00.0
«0.0
0 . .0
00.0
00.0
30.

.20
0.0
00.0

0000

00.0
00.0
00.0
3.0
No.0
00.0
00.0
00.0
0.0.
02

x
00.0
5.0

0000

00.0
.00
.00
.00
.00
..00
000
00.0
30.

05.

X

0
00.0

0000

00.0
00.0
0N0
00.0
.0.0
00.0
.0.0
00.0
.05..
050020

20:03:23“— .20

0.0
.0...

0000

.0.0
00.0
0 ..0
No.0
.0.0
5N0
00.0
00.0
3...

.00.

0.0
0...

0.00 .0005. 2000.08.00 .05 - 00920000.

.
00. .

0000

00.0
00.0
0.0
~00
~...
000
~00
.05.

2.2. 30.02 5.0 .02. - x
000
2.00

0.00

0000.20 02..020
0.0200020 2.00.00...
02.20.2005.

..0 20....

02.0.0 .0.02.5.
ﬁEEQEEES
00:020....

.002. 20.0.0 E00
.0005 0.0.50.0

50255

.002. 8000.952 5.2000
.005. 000.0

.00E 20.”.

5.00

.002. 20.0.0 002?
0.20.0202.

00...:0Nmm

126

N.
00.0.

(D
‘—

0002..

X

0
N50

0000

00..
00.0
00.0
...0
00.0
00..
00.0
00.0
.0020E.
w

vmod
0.... .0 .

00

00.0
00.0

00.0.
...0
00.00.
00.0
00.0
.0005.
~.0

0.
00.00

0N0

~00
00.~
00.0
.00
~00
000
~00
00.0
.0008.
.0....

cm
05.00

0

0
~.0.
0
~.0
~0..
00.0
000
...0
0..~
000
00.0
.000...
00.2

0.
00.00

0

0

«.0..

0

....0
....
00...
N0...
00...
00..
00.0
00.0
.0020...
x00.5n.

x

00
00.N5.

0

0

000.

0

.~0
~00
~0.~
000
0.0
00...
00.~
00.0
.0002.
0.2-0.00...

00. N.
00.000 05.0.
o 0

o o

000 0.0.

0 0

00.0 No.0
00.0. 00.0
00.0 00.0
00.0 00.0
.0.. 00.0
00.00 00.0
00... 3.0
00.0 00.0
3.202.. 32.02..
2.00.2 2.00.0".
23.0.0220“. .05

x

000
00.000N

0

0

00

0

..00~
.00..
0~. .0.
0... .00
0.00
0...-0.
00000
00.0
.0005.

05.020

N.
.N..

0
0
.
0

.00
00.0
.00
00.0
~00
B0
000
00.0
.0005.
2.0.0

000 .0005. 00.0.0220... .20 - 0002000...

«0.00.

0
00.
0

0
.0..
~....
000
000
.~0.
0000
0... ..
00.0
.0005.
2N

0.0
00.0

2.2. 30.02 5.0 .02. - x
000

2.00

050.0

0020.20 02..020
0.0200020 2.0.0.005
02.20.505.

..0 20.”.

x520... .0.02.5.
AEQQEESS
001.0203

.002. 20.0.0 E00
.0005 0.0.50.0

502.5

.002. 8000.052 5.200.“.
.005. 000.0

.002. 20.".

5.00

.002. 20.0.0 .002>>
920.0202.

cngumw

127

mad 0.....
o o

o o

o o

o o
mud owd
m..o so...
and .0...
...o o..o
...0 co...
co... vmd
m... we...
mod vod
.0... ..0...

.08 .0000

mud.

...0.
o

o
0N6
No.0
Nod
bod
v0.0
5N.
o..o
...o
.0...
”.0

vméohv

00.000.
0

0

0
0.0...
00.00
0......
00.00.
......0.
0.00..
00.000
00.000
.00...
mo

.09.

0000

.m.»
...0
wed
mm.
.m.m
Nod.
mod
.06
..0...
...0

..0—«0.35.3.- «0.0

©N.om

0000

0....
00.0
00.0
00.0
00.0
.00.
00.0.
.00
.0...
00

0.0a .0005. 0000.08.00 .05 - .90.me0...

mm.
NWO
ow.
mud.
omdm
....
ow. .N
wad.

00.2....

m

0

v.0
0d

0. .0.
«am

0

m...
mm.
Qua
.v
av
..m
ova
com
on
9.5
.c:oE<

......08 + .05
.0... .00 000.0
...... .00... 000.0
...... .0. 0.5.0

.0... 0.0.0.0 000.0
0205..

may...

.0022. we

00m.
.55

0.50

02.020 05.9.0
0.05000... 50.0.00...
05.5.50...

..0 20.”.

050... .0055.
x.E0.n c.E0..>
00:00.0...

.00... c053 E00
.0005 0.0305

>0c>>

.00E 8:00.93 5.0.30...
.005. 0005

.00... ...0...

2mm

.00... :05... .0052.
0.02.8.0...

vuntoumm

128

00..
.0.N

VON...
V0.0

0000

.00
.00
00.0
00.0
0.0
...0
00.0
.0...

0..

0..
5.0...

0000

00.0
.00
00.0
.00
00.0
00.0
00.0
...0.

....

 

 

 

 

 

 

0.0 00..
00.. 00.0 ...0 v...
00.0
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0 0
00.0 00.0 0.0 00.0
00.0 .00 .00 00.0
.00 .00 .00 .00
00.0 0.0 .00 00.0
0.0 00.0 ..00 00.0
00.0 .00 ....0 00.0
00.0 00.0 0.0 0.0
00.0 00.0 0.0 00.0
...... .0... ...0. ......
.5. 0.0 050 .05

0.00 .0005. 00.0.0000”. 00.0 - 0002000..

..0 0.0
00.0 00...
0 0

0 0

0 0
0.0 0

m ..0 00..
50.0 50.0
3.0 00.0
m ..0 00.0
0.0.0 00.0
.0.. .N.
00.0 05.0
00.0 00.0
...... ..0.
05.. 00..
0030.05.00. «0.0

0..
00..

0000

V0.0

000
...0
00.0
0.0
00.0
.00
.0...
0__

0.0
mm.

0000

00.0
00.0
00.0
00.0
00.0
00.0
...0
..0.

0.:

.5 . + 0.... u 000 .0. 2.0.5.500”.
050 + .05. u .05. .0. 2.080.500”.

0N
NON

OOOO

00.0
.00
00.0
.00
~00
0.0
.00
......

0.<

5... 30.00 5.0 .0.: - x

000
600

0.50.0

05.0.50 05.9.0
0.0000000 50.0.00...
05:03.05.

..0 ..0...

00.0.0 .0.055.
560.0 560.5
40:00.05.

.00... 00.0.0 500
.0005 0.0.50.0

500>>

.00E 8:00.050 5...:00
.005. 000.0

.00... 50....

5.00

.00... 00.3.0 .0055
0.00.0205

QNKBwa

129

0..
3.00

.00

00.0

m ..0

. ..0
00.0
00.0
00.0
3....
00.0
.9005.
.0005.

00
0 .000

0.0

00.8
8.0
00.0.
00. .N
00.0.
8.00
8.00
8.0
80.0.5
0“.

m
00.0..

Wm

«0.0
00.0
.00
00.0
.....0
n ..N
00...
00.0
.9005.
000

0
00.0

0000

5.0
8.0
8...
No...
0.0
8.0
8.0
30.

5.0
0.0
50.0

0000

3.0
00.0
00.0
00.0
No.0
00.0
00.0
00.0
.0...
02

x
00.0
0.0

COCO

5.0
5.0
.0...
5.0
5.0
8.0
00.0
8.0
.0..

0.2

x

0
00.0

0000

.0...
8.0
0.0
8.0
6.0
.00
5.0
8.0
......
05.5.50

00:05:50.... «0.0

0.0
N00

0000

00.0
#00
0 ...0
No.0
3.0
mud
00.0
00.0
$0.

.on.

0.0
00...

02:0

0000

00.0
00.0
00.0
No.0
00.0
0 ...0
..0.0
.0...
.mOSQ

0.00 .0002 0000.08.50. .05 - 0002000...

P
ON...

0000

5.0
8.0
8.0
...0
8.0
00.0
00.0
No.0
.0...

0.0.. 30.00 .20 .0E - x
000.
E00

0590

02.0.00 00..000
0.2.02.0. 52000.0
00.00....05.

..0 00.0.

50.0.0 .0.055.
5E90£EES
00:00.03

.00E 00.0.0 E00
.002. 0.0.50.0

.0c>>

.00.: 6000.93 .....00n.
.00.). 000.0

.008 00.”.

5.00

.000. 00.0.0 .000>>
0.00.0200.

QNuBNmm

130

N..
00.0..

(D
‘.

0500..

X

0
5.0.0

...0
00.0
00..
00.0
00.0
.0590.
m

vmod
00.000

0

0

00

0
00.0
~00
00.0
0.....
...0
0......
00.0
00.0
.0505.
0.0

0..
0.0.00

0N0

..00
00..
00.0
.0.0
00.0
00.0
00.0
00.0
.0505.
0.0.

x

«.N
5.00

0

0
0.0.
0
.00
0.0
00.0
.00
...0
0...
00.0
00.0
.0590.
00.0

0..
v0.00

0

0

0.00

0

00.0
00.0
00.0
~00
00.0
....
00.0
00.0
.0505.
.6250

00
00.05..

00.0

00.0

00.0
.0505.
0.2.0.000

X

00. 0.
00.000 0.0.

0 0

0 0

000 0.0.

0 0

0.0 3.0
....0. ~00
00.0 00.0
00.0 00.0
00.. 00.0
00.0 00.0
0..... «..0
00.0 00.0
.0505. .0505.
500.2 0.00—on.
02.0.0000“. .20

X

000
omNm ..N

o

0

0.0

o

No.50

0 0.00
30.05
Ntmmm
0N0».
«0.500..
00.000
00.0
.0500...

05.000

.20
NF
omé

0
0
..
0

00.0
00.0
.0.0
00.0
00.0
00.0
00.0
00.0
.0505.

5.0.0

0.00. .0002 0030.38.50. .20. - ED750500.

50
00.00..

0
0m.

0

0
0....
00.0
.00
00.0
3....
«...0
00. ..
00.0
.0505.
0N

0.0
00.0

00.0

.. ...0
N00
00.0

00.0
00.0
00.0
00.0
.0505.
00

08-x
000
E00

...0. 30.00 5.0

0.50.0

05.0.00 05.000
0.0000000. 0000.005
05050.05.

..0 00.0.

50.0.0 .0.055.
5EQ05ESS
00:00.03

.00... 00.0.0 0.00
.0005 0.02.0.0

500>>

.00E 8000.050 55000.
.005. 000.0

.000. 00.0.

5.00

.00E 00.0.0 .0002.
300.080....

34.0—50wa

131

.05 50. F
0 0

0 0

0 0

0 0
00.0 00.0
50.0 00.0
00.0 00.0
50.0 00.0
3.0 00.0
50.0 00.0
00... 00.0
00.0 00.0
.05.. .05..

0000 .0000

N0.0..

N0.

00.0
00.0
00.0
00.0
00.0
00...
00.0
3.0
.05..

“.0

00.0550

00.0...
0

0

0
0.00.
00.00
00.00..
0.00.
05.0.
00.0%.
00.0...
00.000
000..
00

No.00
0
0
0
0

50.0
.05..
...0

scan—3E3“. «0.0

00.00

0000

000
50.0
00.0
m0.N
00.0
..N. wN
cod.
..0...“
.05..
«..0

0.00 .0005. 0000.08.00. .0... - 005me0...

00...
00.5
50..
N00.
0 ...0...
N50
00. Fm
0000

00.000 P

0

0
0d
0. ..
N0.
N

0

F

00
mm
00
0.00
0.50
000
com
05

05.0
.0000.<

05.050 105.
.05.. 000 000.0
.05.. .00.. 000.0
.05.. .0. 000.0

.05.. 0.0.0.0 000.0
0205..

00.00

.0505. 00

000
0.00

0.50.0

05.0.00 05.000
0.0000000. 0.0.0.00...
0500.0.05.

..0 00.“.

5.00.0 .0.055.
5E905ES5
00:00.05.

.00... 00.0.0 0.00
.0005 0.02.0.0

500>>

.00E 8000.050 50.000
.005. 000.0

.000. 00.0.

5.00

.00... 00.0.0 .0002.
0.00.0200.

000?.«00

132

00..
50.N

0000

00.0
00.0
...0
00.0
00..
00.0
00.0
.05..

_0>

00N.0
00.0

COCO

.0...
.0...
«0.0
no...
mad
5....
00...
.05..

a...

0..
00..

0000

00.0
3.0
.00
0.0
00.0
...0
00.0
.05..

....

 

 

 

 

..N
0.0

 

 

0.0 00..
00.. 00.0 00.0 0...
0.0
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0 0
...0 ...0 00.0 50.0
3.0 3.0 .00 00.0
.00 ~00 .00 .00
50.0 00.0 00.0 .00
0.0 00.0 00.0 00.0
50.0 00.0 0.0 00.0
3.0 .00 0.0 ...0
00.0 00.0 0.0 00.0
.05... A00. .05.. A0}.
.5. 00.0 050 .02

0.00 .0005. 0000.08.00. .0... - 50200.02

0
0

0
..0
00.0
50.0
.00
«..0
00.0
.0..
55.0
00.0
.05..
05.

co=a_:E._ou_ .20

0N
.00

0000

00.0
50.0
00.0
0.0
.00
N0...
.00
00.0
.00.

30..

0..
NON

0000

00.0
00.0
00.0
00.0
00.0
00.0
.N.0
.05..
0..

0.0
5N.

COCO

00.0
00.0
00.0
00.0
00.0
.00
...0
..5..

0.:

.5... + 0.5 u 0...“. .0. 0.050.500”.
050 + .05. u .05. .0. 2080.500”.

0N
05.N

0000

00.0
.0...
0....
0....
0...
N0...
.N...
.05..

94.

EE 30.00 .20 .0.: - x

00m
:50

0.50.0

00.00.50 05.9.0
0.0500050 0.5.0.005
05:03.05.

..0 20.“.

5.003 .0.055.
...EQQ 5.095
610%...

.008 00.3.0 500
.0005 0.02.0.0

505$

.00.: .039950 5...:0n.
.005. 000.0

.00.: 20....

5.00

.00.: 00.0... .0055
0.00.020...

emu—Bumw

133

0.
00.00

05

50.0

5 ...0
«..0
00.0
on...
00.0
00.0
00.0
.9005.
.0002

00
05.05N

CCOOC

05.0
.0... ..
....5.
00. .0.
00.00
05.00
00.0
.0005.
00.

m
00.0.

CCNO

00.0
00.0
00.0
00.0
...0
00.0
00.0
.0....02.
000

0
00.0

CCOC

00.0
.00
00.0
00.0
00.0
0.0
...0
00.0
05..
00

.20
0.0
«0.0

COCO

00.0
00.0
00.0
No.0
00.0
00.0
00.0
.05..
02

x
00.0
0.0

COCO

00.0
.00
.00
.00
.00
50.0
00.0
00.0
.05..

0.2

.00

COCO

00.0
00.0

0 ..0
No.0
.0.0
00.0
.0.0
00.0
.00.
Otto—£0

00.3.3500... .05

0.0
50..

CCCC

.00
..00
. ..0
00.0
.00
00.0
00.0
00.0
.05..

.00

0.0
0...

0.00.0

CCCC

00.0
00.0
00.0
50.0
.00
05.0
0.0
.00
0.5..
.mocn.

0.0m. .0002 02.0.0550... .20. - 099750500.

.
00..

COCO

00.0
00.0
....0
00.0
0...
50.0
00.0
0.5..

0..: 550.00 20 .0E - x
00”.
E00

0.50.0

0000.00 0....000
0.0000000. 220.005
00.02502

..0 00....

.0an 0.00.2
AEQQEESS
00:00.05.

.00.: 00.0.0 E00
.0005 0.0.50.0

500>>

.00E 6:00.050 5...:0n.
.002 000.0

.00.: 00.“.

200

.00.: 00.0.0 .000>>
0.00.020...

emu—Bumm

134

NF
00.0..

(D
‘—

A

0005.

x

0
5Y0

COCO

.0..
00.0
00.0
00.0
00.0
00..
00.0
00.0
.0005.
0

¢No.o
3.00;

0

0

00

0

00.0
00.0
00.0
50. :
00.0
00.00.
00.0
00.0
.0008.
«.0

05
55.00

0N0

00.0
000
00.0
00.0
00.0
00.0
05.0
00.0
.0005.
0::

0N
00.00

0
0
0.05
0
00.0
00.0
00.0
3.0
3.0
2.0
05.0
00.0
.0005.
00.0

0v
mvdm

o

0

N00

o

and
8.0
00.0
50.0
Fwd
mmé
mm;
00.0
.0005.
x0250

x

00
50.55..

0

0
0.00..
0
00.0
55d
00.?
3.0
00.0
00%
05.0
00.0
.9090.

0.0/$8ch

x

on? N?
modom 05.2
o o

o 0

can «.9

o o

and 3.0
00. 3 vwd
00.0 cod
00.0 «0.0
no.0 No.0
0000 00.0
00.0 0.....
00.0 00.0
.0005. .0005.
0.00.2 0.00.0”.
00.30.50“. «0.0

x

00¢
0N.N5¢N

0

0

0.0

0

3.3

0 v.00
00.00
3.5mm
00.0w
ovNNMr
0N.000
00.0
.9005

05.000

Né
0N9

0
0
w
0

00.0
00.0
.00
00.0
00.0
50.0
00.0
00.0
.0005.
000.0

300 .0002 800.0000... .20 - 00020000.

50
00.50..

0

00..

0

0
00%
00.0
0N0
55.0.
v0.3.
00.00
0~.:
00.0
.9090.
0N

0.0
00.0

5.0

30
~00
00.0
00.0
00.0
3.0
00.0
00.0
.0006.

0.0. 30.00 .20 .00. - x
00m

.000

0.530

0000.00 05.000
0.20020 0.00.8.0
00.00.50:

..0 00.“.

000.000 .9003
00.0000 0.0.0.5
000.0520

.006 00.0.0 E00
.0005 0003000

500>>

.0000 000090.50 5.0.000
.00.). 000.0

.00E 00.“.

.200

.008 00.0.0 .000>>
800.0200.

catﬁumm

135

05.5

OCCO

:0
00.0
00.0
00.0
50.0
05..
00.0
..0...
0000

00...

COCO

00.0
00.0
..0.0
00.0
00.0
00.0
N0. F
3.0
.05...
000:0

00.00

0.00

50.0
N00
N00
00.0
No.0
00.?
5N0
3.0
.05...

00

0N.055v

00.000.
0

0

0
0000.
00.00
00.05
00.0:
00.0:
00.000.
00.050
00.000
0080
00

05.0.0 00.0...

0 0
0 0

0 0

0 0
:0 00.0
00 0 00.0
.0 0 00.0
0. . 05..
00 0 00.0
55.0. .0. a
00.0. 00.0.
50 0 .00
..0... ..0...
00 00

0.030.353“. «0.0

0.00 .0002 0000.08.00 .20 - 00020000.

50.?
05.5
00.0
00.0..
00.000
N00
00. E
00.00

00000..

0

0
5.0
0. 0
00.
~......
0

N
00
0.
0.00
00
00
000
000
05
000

E0002.

30.050 + .05.
..0... 000 0020
..0... .000 0020
..0... .0. 0020

..0... 0.0.0.0 0020
03.05..

00.00

.0022. 00

000
E00

0.50.0

0000.00 00..000
0.0000000 62200.0
00.00.50:

..0 00.0

0.0000 .0.055.
06905E95
00:00.03

.00E 00.0.0 000
.0005 0.03000

500>>

.00E 800000.50 50.000
.00.). 000.0

.00E 00.0

.200

.00.: 00.0.0 .000>>
800.0200.

000.0000

136

00.:
mad

vomd
50.0

0000

5.0
00.0
5.0
00.0
mud
3.0
00.0
A00.

9.

0...
05. F

 

 

 

 

 

 

0.0 00..
00.. 00.0 000 0...
0.0
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0 0
0.0 0.0 000 000
«00 80 .00 .00
.00 .00 .00 .00
000 000 ~00 80
000 000 30 30
...00 000 0.0 000
000 20 000 000
000 000 0.0 000
.0. ...0. ...0. ...0.
E 0:0 0.0 .02

0.00. .0002 02.0—38.5”. .29 - 592mm?

..0 ed
5 v.0 :00
o 0

0 0

o 0
«.0 0
v0.0 3.0
00.0 00.0
00.0 00.0
00.0 «...0
3.0 00.0
.0.: ~0..
N00 00..
00.0 00.0
...0. ...0.
0.... 30..
5.3.355“. .05

0.:
mod

000°

00.0
No.0
50.0
00.0
00.0
V0.0
0N0
...0.
0..

0.0
mm...

0000

~00
000
000
:0
000
B0
:0
...0.

0i

.3... + 0.... u 020 .0. 30800503.
0.5 + .22 u .05. .0. 60600500.”.

0N
mmN

EE 26.00 .20 .0E - x

00m
:50

0.50

00.0020 05620
0.050.005 820.005
05:25.05.

..0 50.“.

580.0 _00052
5805 52.0.5
001.050.:

.00E 00.20 E00
.000.» 2030.0

>0c>>

.00E 5:00.043 2:500
.005. 000.0

.008 20.“.

.200

.00E c220 .0055
«...0—080:.

own—30mm

137

or
mm. 5

kn

mud
Ed
no.0
de
Fwd
mad
or. _. F
cod
8.3.5
.522

cm
mmdmw

m.m

8.8
E. v
N3
and,
8. a
8.8
8.8
23
38,5
on.

m
3...?

NM

8..
5.0
9.0
:3
Rd
EN
88
23
8.8.5
goo

0
mad

0000

So
cod
mod
5o
26
«to
cod
5

2m
06
and

COCO

cod
No.0
No.0
3.0
mmd
cod
cod
Gav
m2

3.0
Ed

0000

cod
85
85
5.0
8.0
So
mod
23
g

as.

x x x
o ad 3
33 F: m:
386
o o o
o o o
o o o
o o 0
8d :3 86
8.0 85 86
8.0 86 mod
86 :3 mod
:3 8o :3
8.o Rd 83
So 85 2.0
23 8.0 :3
30V 3.; g
2:25 5a .85“.

:0.~u.::tou «05

«an 332 883.88 85 - 594me

mm:

0000

cod
No.0
2.6
5.0
N..._.
mod
Nod
$9
$0

ES 323 Em EE - x
com
Eam

0.38

«2520 05.050
macamoﬁ 52285
0522522

:0 5E

5:55 .9952
559a£EES
401.883

.92: 556 :50
ammo.» «.5305

>23

.85 633543 bzaon.
_moE coca

_mmE 5mm

2mm

_moE 556 “was;
35:35:.

own—\cmmm

138

Nw
009

(O
‘—

A

555555

X

o
omd

0000

5.5
55.5
55.5
55.5
55.5
8.5
55.5
55.5
855555

vmod
wnévp

5
5

8

5

55.5
55.5
55.5
2.5
85.5
8.58
55.5
55.5
655.5
«E

or
mfmm

wNm

5.5
8..
55.5
5.5
5.5
8.5
8.5
55.5
35555
5::

NN.mw

Ndh

55.5
8.5
55.5
8.5
8.5
8.5
55
55.5
8595
SE

2.
omdm

o

o

Nov

o

Rd
and
cod
m;
2.5
mm;
3;
86
65:55:
589?.

x

ow
end:

5

5
5.53
5
3.5
N:
N:
8.5
:5
8.5
33
55.5
855555

22.585

x

555 NF
9.5.5.5 8.2
5 5

5 5

58 «.9

5 5

8.5 5.5
55.» 8.5
55.5 55.5
8. F 5.5
2.5 5.5
55.8 55.5
8.5 5.5
55.5 55.5
655555 65:95
EOQZ 5090“.
5.55:5". in

x

8v
hv.on¢N

5

5

m5

5

2.2
8. 5
55.55
55.5:
8.2
55.5va
8.85
55.5
655555
8:55

NF
rm...

OPOO

55.5
3.5
55.5
5.5
3.5
55.5
55.5
6555::
5555

98 8552 55525585 85 - 592%?

5d:

5

5:

5

5
a;
55.5
:5
5.5
8.8
55.8
8.:
55.5
85:55
5N

md
de

5.0

55.5
5.5
55.5
55.5
55.5
3.5
55.5
55.5
35:95
3

EE 3202 SE “9: u x
55m

65

0.565

8.320 05.050
9258:.“ 620.5505
05525555.

.6 5E

55:55 .8055.
5ESQEESS
501.283

_mmE 556 E00
68> 2m32m

>mc>>

.92: 825993 >239".
.8: 82m

.99: 5?.

2mm

.38 556 623
852.529..

own-Baum

139

5.0

0000

55.5
2.5
55.5
8.5
8.5
E. F
85
$00
88

mmé

0000

00.0
«0.0
00.0
00.0
v0.0
de
N0. r
no.0
$00
.008

NW?

C

:5
8.5
85
55.5
8.5
R;
R5
95
5x5

“6

undonv

8.89
5

5

5
8.58
8.2.
8.8
850
2.3:
«32 P
8.53
8.85
895
m5

Nwdw Cumv

0 0

0 0

0 0

0 0
mwd no.0
3.0 mm 0
0N0 mm 0
mNé v0;
hQN mvd
No.9. 50 0..
mud? 00.0..
.00 P00
350 080
no no

5.3.3.50“. 5.0

33 382 8522855 55 - 525.me55.

00.9
5.0
mmé
hudw
{.05.
2....
50. FN
00.2

00.000 w

0.5
mé
0:.
vN

8
5 v

a

5.8
5.8
555
58
8,
55:5
anE<

A8090 + 85.
£5 85 55.5
$00 .85 5520
g 5 5555

3.5 5995 820
52”»:

mode

A5530 mo

0mm
an

0.5m

8:020 05.20
22885 525.85
m:_co_£o_2

:0 :mE

5:505 .9055.
56955895
401.263

.85 c920 :50
88> 2526.0

>555)

_moE 83025.3 Egon.
Ems. 002m

.38 cmE

2mm

.85.: 55.0 89:5
85520:.

tun—Emma

140

mmé
NmN

0000

8.5
55.5
8.5
555
85
E5
85
3.5

_m>

3N0
00.0

0000

85
55.5
85
85
85
85
85
3.5

e»

0.?
N5...

0000

85
«55
8.5
3.5
8.5
B5
25
3.5

5,:

 

 

 

 

_..N

 

 

55 85
m: 8.5 35 m:
2.5
5 5 5 5
5 5 5 5
5 5 5 5
5 5 5 5
85 85 85 55.5
~55 .85 8.5 8.5
55 85 8.5 8.5
8.5 85 85 85
25 85 55.5 35
85 85 35 555
85 2.5 8.5 8.5
55.5 85 R5 25
3.5 3... 3.5 3.5
E 5E 96 .52

5550 3552 cousspcom 555 - 555me32

00.0

5
5

5
55
B5
55.5
85
85
85
2!
555
$5
$00
9:

5.0323550“. «20

0N
Nod

0000

5.0
V00
N00
0.20
Nvd
_.N._.
00.5
00.0
0.80

:3

0.5
00.N

0000

95
85
555
B5
85
85
55.5
85
3.5
5..

0.0
”N...

0000

8.5
55.5
85
8.5
85
85
$5
3.5

55..

S» + of. u man. .8 EmEQScmm
90 + 8.2 u 6.2 .2 55:525ch

0N
0~..N

COCO

no.0
5.0
9.5
3.0
N00
05.?
00.0
3.5

9<

SE 328 2m .9: - x

com
:50

0.5%

28530 955:0
958855 625.85
95:02.22

:0 55E

5an .8952
x_Em.a EEES
50:555.:

:59: 55.0 :80
88» 82590

>55>>

.82. 83595.3 9.58
Ems. coca

_mmE 55E

20m.

:59: 5920 $ch
355.020:—

VNn—Bnmm

141

or
mmdm

#0

00.0
00.0
00.0
v0.0
30
0N.N

0 v. P r
00.0
.9855
.0092

00
3.0mN

0.0

in?
5.1
55.5
5.2
5.8.
8.55
8.8
55.5
.585.
on.

m
00.8

vN

55
00.0
3.0
010
«0.0
5 FN
00.0
00.0
.5585
500

0
mmd

COCO

50.0
00.0
No.0
..0.0
m...0
0. F0
00.0
350

SE
0.0
00.0

x
#00
5.0

x

0
00.0

mESEU

cozﬁacton. «0.0

0.0
00...

0000

55
55.5
85
8.5
55
~55
55.5
5.5
$00

.55

0.0
00...

0..t.0

0000

85
555
85
8.5
5.5
85
5.5
85
3°.
.mocl

30D _oﬁoz Gown—3509A 8mm - EQZWAEAN

F
Ear

SE 26.3 Em .mE - x
00¢
:50

0.5m

25.50.20 00:20
9588i 52285
0500.522

..0 25....

5an .92.:
5890 0.895
50:55.93

.38 c920 E00
.89» 22590

>mc>>

.09: 6252043 Egan.
.022 000.0

.09: 5.“.

2mm

.85. 02.5.0 .09.;
3.3.320...

VNmEmmm

142

N?
00.05

(D
‘—

5.555.

X

0
00.0

0000

55..
8.5
55.5
555
55.5
55..
8.5
55.5
.5555.
w

0N0.0
NN.VN..

5
5

8

5
55.5
8.5
55.5
55.5
8.5
53.5
55.5
55.5
.9555.
5.0

0..
00.00

mNm

55.5
2..
55.5
.55
85
55.5
8.5
55.5
.5555.
5::

0N
tth

N05

N ...0
3.0
00.0
N00
3.0
05...
50.0
00.0
.936.
00.”.

av
00.00

0

0

5.0..

0

3.0
0.00
00.0
3.0
2.0
35
~...;
00.0
.9555.
x005...

00
3.00..

5

5

5.8.

5

.55
3..
85
85
5.5
8.5
35
55.5
.585.
22.585

x

8. 5.
55.555 8.8

5 5

5 5

58 5.8

5 5

8.5 55.5
55.5 85
55.5 55.5
55.. 55.5
8.5 55.5
8.55 8.5
85 55
55.5 55.5
.5555. .5555.
500.2 0.050“.
55.55.65“. 8.0

x

000
00.00NN

5

5

5.5

5

5.. .5
8.8
50.8
8.8.
5.55
55.58.
8.555
55.5
.5555.

05.9.0

55
N...
05...

0
0
..
0

85
.55
85
55.5
85
85
85
55.5
.985.
5520

0.00— .0002 coma—08.8.... .0.Q - EQZm—mmxx

50
no.0:

0N..

8..
8.5
$5
8.5
55. 3
55.5
2.:
55.5
.985.
:5

0.0
00.0

..0.0

8.5
8.5
55.5
55.5
55.5
8.5
85
55.5
.5555.

.06 - x
00.".
6:0

5.: 30.00 2m

0.50

0000.50 05.9.0
0.85850 55000.0
0560.502

..0 50.".

5606 .005.).
56805695
50:05.»:

.006 50.5.0 600
.0005. 9030.0

>022.

.006 83090.3 .5300
.005. 000.0

.006 ...0...

.200

.006 605.0 .0055
8520855

3.63an

143

00.0

0000

55.5
05.5
0.5
5.5
55.5
55.5
.55
3...
850

N05

COCO

no.0
No.0
00.0
No.0
50.0
0N0
05.5
00.0
3...
.0000

50.0N

N05

55.5
.55
55.5
....5
05.5
05..
8.5
5.5
......

“.0

VNKvnv

8. ..0.
5

5

5
0355
.0...
05.5.
8....
05.80
8.85
8.8..
05.8.0
5595
05

00.00

0000

05.5
C5
05.5
55.5
.5...
8.0.
5.8.
.55
3...
00

town—aston— «0.0

0.00 .0032 0030.0.Eom .2Q . 09%me

N500

5

5

5

5
55.5
.55
55.5
8.5
.00
......
8.:
.55
3...
no

00...
50.0
~0..
.0.0N
N500
are
..0. FN
00.0..

00.0005

0
0
0..

.8005 + .0.).
...... :8 0520
3... .55... 0520
...... .5. 0520

...... 5.0.95 0520
520.3

0500

.9522. 05

00w.
600

0090

00.8.20 05.9.0
0.2820 55.5.8.0
05:550.).

..0 50.“.

560.0 .0055.
56905685
50105.0...

.006 505.0 600
.000» 0.0305

>05>>

.006 6206043 09.300
.00.). 000.0

.006 20.“.

5.00

.006 60.0.0 .00..>>
8.8.00.3...

ouuﬁnmm

144

00...
00.N

0000

.55
55.5
05.5
~05
5.5
85
85
3...

_0>

vomd
00.0

0000

05.5
55.5
55.5
55.5
.55
0.5
.05
05.5
......

55

0...
00.5

0000

..0.0
00.0
..0.0
0N0
00.0
00.0
0.20
3...

E...

 

 

 

 

...N

 

 

55 00..
50.. .05. 0.5 ....
0.5
5 5 5 5
5 5 5 5
5 5 5 5
5 5 5 5
0.5 5.5 05.5 05.5
.55 .55 55.5 55.5
55.5 55.5 55.5 55.5
55 N55 .55 .55
0.5 .55 55.5 05.5
05.5 00.5 5.5 55.5
50.5 00.5 00.5 .05
55.5 00.5 .05 0.5
...... ...... ...... ......
5 9.0 0.0 .02

050 .0552 5550.58.50 05 - 0002000....

0.20

5
5

5
0.5
05.5
.5.5
55.5
55.5
~05
.5..
05..
0.5
.05.
0.:

eggs-Eon. ~20

0N
0.20

0

0

0

0
00.0
..0.0
50.0
3.0
05.0
..0...
sm...
00.0
3...
:05

0..
50.5

000°

.55
55.5
55.5
.55
8.5
0.5
8.5
......
2.

0.0
00.5

COCO

.55
55.5
.55
05.5
05.5
05.5
0.5
......

2:

.5 + 0:0 u 050 .0. 606053000
0>0 + .05. u .0.... .0. 606050000

0N
NAN

0000

.00
00.0
3.0
8.0
the
0~..
00.0
......

0.3

56 30.00 5.0 .06 - x

000
650

0590

00.8.50 05.9.0
0.058550 52200.0
05:550.).

..0 20....

560.0 .0.055.
56905695
501.0503

.006 605.0 600
.000.» 0.02.0.0

>55>>

.006 60060.3 >530
.005. 000.0

.006 50.0

.200

.006 505.0 .0052.
85.025...

ouuﬁnmm

145

0..
00.00

0~.

00.0
00.0
.00

N ..0
00.0
00..
«0.0.
00.0
.5555.
600.2

00
0N.vom

00000

05.. .
00.5
8. .
..0...
00.05.
55.3
00.0..
55.5
.5555.
0".

m
5.0..

0
N00

.20
0.0
5N0

00.0
:20

x

0
0N0

05.0.20

co_.0_:E._ou_ 00.0

0.0
00...

0000

.55
.55
.55
55.5
.55
0.5
0.5
0.5
.5.

.50

0.0
.0...

000 .0552 558.5005... 0.0 - 0002000...

.20

00.0

56 30.00 .20 .06 - x
.000
650

0.5.00

00.8.20 05.020
0.028520 55.5008
0520.505.

..0 20.0

5690 .0.055.
56905695
50105.0...

.006 20.5.0 600
.000.» 0.0390

>02>>

.006 80090.8 8.300
.00.2 000.0

.006 20.0

.200

.006 20.0.0 .002>>
0.20.0902.

0Nn=mnmm

146

N.
00.0.

(D
r

0000000000000

555....

x

0
«0.0

0000

5...
.5.5
55.5
55.5
55.5
55..
0...
55.5
.5555.
w

.000
00.00.

5
5

8

5
55.5
.55
55.5
0.5
.55
5.50
55.5
55.5
.5555.
5.5

0.
0.0.00

0N0

.00
00.0
00.0
00.0
No.0
No.0
00..
00.0
.0506.
0.2...

NNNO

5

5
0.5..
5
5.5
....5
55.5
...5
555
~....
55..
55.5
.5555.
55.0

0.
00.00

0

o

~0..

o

00.0
0....
00.0
00.0
.0...
00.0
N».
00.0
.0506.
.6050

X

00
00.00.

5
5

5.00.

5

....5
00.5
.55
5.5
55.5
555
0.0
55.5
.5556.
5.2.9550

00.
00.000

5

5

58

5
55.5
55.0
55.5
55.5
.0..
55.55
5:
55.5
.5555.
500.2

N.
00.0.

5

5
N5.
5
.55
05.5
55.5
.55
55.5
..55
05.5
55.5
.5555.

500.00

20.8.0680 .05

950 .5552 550955550 .90 - 00020500

X

000
0 . .000.

5
5

0.5

5

55...
55.5.
8.5
8.8
55.55
8. .00
00.505
55.5
.5552.
55520

20
N...
0...

01-00

.55
.55
55.5
55.5
55.5
05.5
...5
55.5
.5555.
505.5

00.00.

5

55.

5

5

8. .
55.5
55.5
0~..
51.5
8.50
05.5.
55.5
.5555.
5N

0.0
50.0

.00

..55
55.5
55.5
.55
55.5
R5
..55
55.5
.5555.
50

56 2.0.05 20 .06 - x
000

600

0090

00.8.20 05.020
0.0200020 620.005
058.502

..0 20.0

5690 .0852
56905695
50:05.50.

.006 20.0.0 600
.009. 0.0.590

002.5

.006 85008.02 ......500
.002 000.0

.006 20.0

200

.006 20.0.0 .002>>
355.080...

own—00000

147

00.0

000°

.55
05.5
6.5
0.5
00.0
~00
B5
3...
0000

.0..

COCO

~55
.5.5
55.5
.5.5
05.5
50.5
00..
05.5
......
.0000

.N.0N

0.0.
0

0

0
No.0
00.0
00.0
00.0
00.0
00..
00.0
0.0
...x...
“.0

00.0.5.

.0000.
5

5

5

50.0..
3..
00.0
~00.
05. .00
50.000
55.550.
00.00..
:8...
00

00.00

0000

.05
0.5.5
..55
50.5
00.0
.00.
00...
.50
......
00

...0—«0.35200— .05
«...0 .0052 8.0.650 .20 - 0002000...

00.0.0

0000

50.5
05.5
V5.5
505
2.0
......
55.50
.50
......
00

00..
00.0
.0..
.N.0N
0.0.00
N...
0~.. .N
00.0.

00.000.

0

0

0.
0.0
00.
0.0
0

0

0.

N

0

0
0.00
com
000
00.
0.00
6006.0.

Ao\o.w>0 + .05
3... 200 0020
...... .00.. 0520
...... .0. 85.0

3... 2.990 0020
0.2.0....

00000

.0002. 00

000
600

0.50

00.00.20 02.020
0.20820 62200.0
02.20.2005.

..0 20....

x690 .0055.
06906695
00:90.:

.006 205.0 600
.009. 0.0390

>02>>

.006 .0:00.0.>0 9.000
.00.). 000.0

.006 20....

.200

.006 20.0.0 .002>>
35.080...

000.30

148

00..
00.0

0000

00.0

55.5
.55
5.5
5.5
05..
~05
......

_m>

x x
.505 0. .
00.5 N...
5 5

5 5

5 5

5 5
55.5 05 5
55.5 55 5
55.5 55 5
55.5 55 5
.55 ..0 5
0.5 00.5
00.5 9.5
05.5 0. 5
...... 22..
e. .6.

 

 

 

 

 

 

x x x x
0.0 00..
00.. 000 9.5 0...
00.5
5 5 5 5
5 5 5 5
5 5 5 5
5 5 5 5
05.5 .55 .55 05.5
55.5 55.5 55.5 55.5
55.5 55.5 55.5 55.5
.55 .55 55.5 .55
....5 00.5 055 B5
0.5 00.5 0.5 5.5
5.5 00.5 50.5 R5
55.5 00.5 .05 0.5
...... ...... ...... ......
.... 2.0 0.0 .22

.....0 .25: 02.25650 .20 - 0002000....

x x
..N 0.0
0.0 00.0
o 0

0 0

0 0
0.0 o
.00 0.0
.00 .00
00.0 00.0
N00 000
00.0 00.0
.0.. .0..
0~.. 0....
0.0 00.0
...... ......
0.3 00..

€233.50". .05

0..
00..

0000

55.5
55.5
.55
05.5
00.5
00.5
50.5
......
m:

0.0
N0.

0000

.55
55.5
55.5
.55
.05
~05
0.5
0.5
3...

0.:

.3 + 020 n 020 ..0. 606060000
0>0 + .0.2 n .0.2 .0. 60695000

0N
00.N

66 30.02 .20 .06 - x

00m.
600

02.0.0

0020.20 06.020
20250520 250.000
06262.05.

..0 20.“.

.2690 .9065.
.2690 660..>
00:050....

.006 20.0.0 600
.000» 90390

.02>>

.006 83090.3 3.300
.00.... 000.0

.006 20....

.200

.006 20.0.0 .0022.
060.0902.

oumaﬁmm

149

or
00.00

t.

00.0
5.0
5.0
00.0
«V...
00..
00.3
cod
.9055.
.2005.

00
5.00m

0.0

00.0
00.5
00.5
.00
.00:
55.3.
50.00
55.5
.9005.
a“.

m
00.0..

Nd

00.0
00.0
«0.0
no.0
00.0
.0. _.
N F .0
cod
80.0.5
000

0
N00

0000

.55
55.5
55.5
55.5
05.5
:5
0.5
55.5
...0.

$5
0.0
0N0

COCO

00.0
00.0
00.0
No.0
NNd
00.0
00.0
3...
02

3.0
00.0

x

o
0N0

02.20.20

22.03.23“. .05

0.0
0.....

0.0
50

NNd

COCO

55.5
55.5
55.5
.55
05.5
00.5
00.5
.55
......
.00:&

0.00 .0052 800.5025... .05 - 5920000.

SE

«0.0

OOOO

55.5
55.5
05.5
05.5
00.5
5.5
05.5
......

52. 26.02 5.0 .06 - x
00m
250

0.00.0

0020.20 02.620
2025820 52285
02.5.2.0...

__o 20.“.

x2202. .0.055.
50.2.03 2.2.95
401.050....

.008 20.3.0 .200
.000.» 0203020

>02>>

.002. .03238 Egan.
.00.). 000.0

_00E 20.“.

5.00

.00E 20.20 .002>>
0.20.0202.

cumBQmm

150

NP
00.0..

(D
...

A

0005.

X

0
nmN

COCO

0N0
00.0
00.0
00.0
00.0
00...
~0..
00.0
.9202:
m

¢N0.0
mn.N0—.

0

0

00

0
00.0
00.0
00.0
5...
v0.0
010x.
00.0
00.0
820.2.
N 00

0..
mmém

0N0

55.5
0.5
55.5
55.5
05.5
05.5
00..
55.5
.9095
02.

VN
.....N0

0

0
N00.
0
No.0
no.0
00.0
00.0
0 ...0
N0;
0 ....
00.0
62.02..
00.0

0.
Ede

0

o

«.00

0

05.5
0.0.0
00.0
05.5
00.5
00.0
00. F
00.0
.9205.
.6050

00
0.0.50?

0

0
v.00...
0
v0.0
NN.0
0 ...0
00.0
5.0
00.0
No.0
00.0
65.0....

22.050

x

50. 0.
00. .00 00.0.

5 5

5 5

500 0.0.

5 5

50.5 55.5
00.5 05.5
55.5 55.5
00.5 55.5
0.... 05.5
55.50 3.5
50.0 00.5
55.5 55.5
.0206. .0005.
EOEZ Sow—0....
20.0.5220“. .20

X

00v
N0. EON

0

0

0.0

0

«0.0
00.x.
«0.0
0.200
3.00
00. 000
ON. F0:
00.0
8.202..
05.020

.20
NF
0N...

801-00
0'

00.0
00.0
00.0
N00
000
0 ...0
00.0
.9292.

5.0.0

900 .0052 800.0020”. .20 - 00520002

x

50
nméh _.

5

5. .

5

5
.05
05.5
05.5
.05
00.0.
50.50
50.00
55.5
.0205.
50

0.0
#00

00.0

.55
55.5
55.5
55.5
55.5
00.5
..55
55.5
.9005.

2.22 30.02 .20 .02. - x
000

E00

0.50

0020.20 02..020
0.0200020 52285
05222.02

..0 20.“.

5.200 .0.055.
5E90£ESS
001.0503

.002.. 20.0.0 E00
.000.» 0030.0

>02>>

_00E 8:00.043 2.30.”.
.002 000.0

002. 20.".

.200

002. 20.0.0 .002>>
0.20.0202.

ouqumw

151

mad 0~.?
0 0

0 0

0 0

0 0
00.0 00.0
00.0 00.0
00.0 00.0
00.0 00.0
m ..0 50.0
NWN 0 ..0
..0.N mm...
50.0 00.0
...... .0...

5.8 .85

an. .N

on

00.0
00.0
00.0
00.0
00.0
05.0
3.0
m ..0
......

“.0

Edam.»

00.000.
0

0

0

00.0
00.0
00.0
00.0

mm. FNN
00.0 _L.
8&0:
omdwv
..mox.
mo

..N...v

000°

00.0
00.0
00.0
00.0
004‘
and
00.0.
.00
......
a0

cog—ache“. no.0

mmdv

0000

55.5
55.5
55.5
55.5
55.5
.35.
55.00
3.5
3..
05

35 .0552 5550.555”. 85 - 500.2me

mm...
00.0
mm...
an. ..N
mmév
0w...
hadw
00.0?

00.000?

o
9.5.;
one
55.
9:0
anE<

30.96 + 85.
3.. cam 820
...... .8... 820
...... a. 820

...... 5225 325
mi“...

mode

.953. 05

gm.
Ezw

5.55

55.55.20 55.9.0
22585... 5228.5
55.52555.

..5 25.“.

5695 .8555
éEQQEESS
81.85....

.09: 505.5 550
.509 95265

>oc>>

.05.: 85585.3 .959.
.052 552m

.068 55.“.

2mm

.00E 523.5 .855
35.525...

0 £330

152

mm...
055..

x x
550.5 5..
.55 .5.
5 5

5 5

5 5

5 5
55.5 55.5
55 5 55.5
55.5 55.5
55.5 55.5
.55 50.5
...5 05.5
.55 55.5
55.5 5.5
...... ......
e. 5..

 

 

 

 

 

 

x x x x
5.0 55..
50.. 50.0 00.5 5...
55.5
5 5 5 5
5 5 5 5
5 5 5 5
5 5 5 5
55.5 55.5 55.5 55.5
55.5 55.5 55.5 55.5
55.5 55.5 55.5 55.5
55.5 55.5 55 5 55.5
....5 55.5 55.5 85
55.5 ....5 55.5 55.5
5.5 5... .55 .55
55.5 55.5 .05 5.5
...... ...... ...... ......
5 55a. 5.5 .52

..0 5.0
3.5 55.5
o 5

5 5

5 5
5.5 o
55.5 55.5
55.5 55.5
55.5 55.5
55.5 55.5
55.5 50.5
50.5 mud
55... 55..
0.5 55.5
...... .95.
5.... 55..
5.5.250". «5.0

0...
00.5

80000
0'

55.5
55.5
55.5
55.5
.55
55.5
55.5
......
5__

5.55 .5552 5555.55.55 .25 - 555255.?

0.0
mm...

0000

00.0
00.0
00.0
and
«N0
00.0
05.0
.55.

w...

...... + 5...”. u 555 .5. E5E5..:o5~.
5.0 + .53. u .55. .5. 555855.555”.

0N
\kN

EE 26.55 2m .5... - x

55m
an

5.555

5555.50 55.9.0
5.25855 25.2555
5555.553.

..5 25....

5.55.5 .5555...
5.55.5 5.595
551.555...

.55.: 555.5 550
5555.. 5.52.55

>55>>

.55.: 5:55.543 >5.on
.55.). 555.5

.55E 55.“.

2mm

.55.: 55.5.5 .55..>>
355.529..

crutmvmm

153

0.
ENC.

50

55.5
55.5
05.5
05.5
05.5
N... .
50.5.
55.5
.5555.
555.2

00
vaNN

Qm

55.5
55.5
55.5
55.5
55.5..
55.05
55.55
55.5
.5552.
on.

m
.50.

5N

55.5
55.5
55.5
55.5
55.5
55..
3.5
55.5
.5555.
550

o
00.0

0000

55.5
55.5
55.5
05.5
55.5
50.5
55.5
......

SE
0.0
0.0

0000

55.5
55.5
55.5
05.5
5.5
55.5
55.5
......
52

3.0
0.0

0000

55.5
55.5
55.5
55.5
.55
.55
....5
55.5
......

5s.

0
0.0

0000

55.5
55.5
55.5
55.5
05.5
5.5
05.5
55.5
.05.
mate—...0

cows-35.3”. .53

0.0
mm.

0000

55.5
55.5
55.5
55.5
.55
5.5
55.5
5.5
3...

.55

0.0
.0..

and

COCO

55.5
55.5
55.5
55.5
05.5
55.5
50.5
.55
.5.
.5021

5.5Q .5502 5.5.5.5353”. .53. - wand/ma:

2m

N50

0000

55.5
55.5
55.5
55.5
55.5
0.5
05.5
.5...

5:. 2.5.55 2m .58 - x
55m.
E50

5.5.55

55.5.5 55.9.0
55555555 5.5.5.505
5555.555.

..5 55.“.

5585.5 5.55.).
56955595
551555....

.55E 55.5.5 550
.559. 5.5355

.653

.55E 8555.543 3:555
.55.... 555m

.55.: 25.“.

2mm

.55E 55.5.5 .55..>>
355.555....

0 ...—35mm

154

N5
095..

(O
‘.

55:52..

X

o
mNN

vmod
55.55

c

c

on

o
cod
00.0
05.0
cod
mad
55.3
cod
8.0
35:58.
N rm

8
mmém

wNm

55.5
55.5
55.5
55.5
55.5
5.5
53
55.5
.5555.
5::

55.5

5
5
5.55
5
55.5
55.5
55.5
55.5
5.5
55..
5..
55.5
.5555.
558

m?
55.5...

o

o

«.55.

o

8.5
cod
55.5
55.5
55.5
55.5
55.5
55.5
.5555.
.8535

x

00
05.50..

5
5

5.55.

5

55.5
55.5
55.5
55.5
.55
55.5
55.5
55.5
.5555.

x

oar
55309.

5

5
555
5
55.5
55.5
55.5
55.5
5....
.55.
55.5
55.5
.5555.

5.o<.m.c5n. 565.2

N..
5.5.?

5

5
5.5.
5
55.5
55.5
55.5
55.5
55.5
55.5
55.5
55.5
.9555.

5.55.5“.

5.0.5.3550“— 50.0

5.55 5552 55.55.5555“. 55 - 55525555.

x

005
55555..

5

5

5.5

5

55.5
55.5
55.5
55.5
55.55
5355
55.555.
55.5
.5555.

55.9.0

Em
N6
om...

OFOO

55.5
55.5
55.5
55.5
«5.5
55.5
2.5
55.5
.5555.
5525

mm
8.05..

, o

55.
5

5

55.5
55.5
55.5
55.5
55.5.
55.5.
55.55
55.5
.5555.
5N

in
5.0
0N6

..od

00.0
00.0

55.5
55.5
55.5
55.5
55.5
.5555.

SE $0.09 2m .58 - x
55m
E5m

0.5.5.5

55.55.50 55.9.0
5.5555555 522555
55:52.55.

..5 55.“.

.5855 .5555.
5&95EES5
501.555....

.55E 55.5.5 E50
.552. 5553555

>55>>

.55.... 8555.543 9.555
.552 555.5

.55.: 25.“.

2mm

.55E 55.5.5 .55..>>
5.55.5555...

awn—Ram

155

REFERENCES

AAF C, 2004. Farmed salmon. Agriculture and Agri-Food Canada Online Fact Sheet,
http://www.seafoodcanadagc.ca/farmed_salmon-e.htm.

Alexis, MN. and I. Nengas, 2001. Current state of knowledge concerning the use of soy
products in diets for feeding sea bass and sea bream needs for ﬁiture research. Am.
Soybean Assoc., 5, 32.

Anderson, J. 8., SP. Lall, D.M. Anderson and MA. McNiven, 1993. Quantitative dietary
lysine requirement of Atlantic sahnon (Salmo salar) ﬁngerlings. Can . J. Fish Aquat Sci,
50, 316-322.

Anderson R. L. and W.F. Wolf, 1995. Compositional changes in trypsin inhibitors, phytic
acid, saponins, and isoﬂavones related to soybean processing. J. Nutr. 1995;125:581.

AOAC, 2000. Ofﬁcial Methods of Analysis. 17th ed. Association of Ofﬁcial Analytical
Chemists, Arlington, VA.

Amesen P., L.E. Brattas, J. Ollie, and A. Krogdahl, 1989. Soybean carbohydrate appear
to restrict the utilization of nutrients by Atlantic salmon (Salmo salar). Pro. Third. Int.
Symp. On Feeding and Nutr. In Fish. Taba Aug. 281— Septl. Japan. 273-280.

AVMA. 2000. The 2000 Report of the AVMA Panel on Euthanasia. Journal of the
American Veterinary Medical Association 218(5):669-696.

Azevedo, P.A., D.P. Bureau, S. Leeson and CY. Cho, 2002. Growth and efﬁciency of
feed usage by Atlantic salmon (salmo salar) fed diets with different dietary protein:
energy ratios at two feeding levels. Fisheries Science 68: 878-888.

Baeverﬁord G., and A. Krogdahl, 1996. Development and regression of soybean meal
induced enteritis in Atlantic salmon, Salmo salar L., distal intestine: a comparison with
the intestines of fasted ﬁsh. J. Fish Dis. 19, 375-387.

Barrows, R., R. Hardy and D. Stone, 2004. Effect of processing and pre-processing
cooking on nutritional value of soybean meal for rainbow trout (Oncorhynchus mykiss).
Aquaculture 2004. Honolulu, Hawaii.

Bakke-McKellep, A.M., C. McL Press, G. Baevertjord, A. Krogdahl and T. Landsverk,
2000. Changes in immune and enzyme histochemical phenotypes of cells in the intestinal

mucosa of Atlantic salmon, Salmo salar L., with soybean meal-induced enteritis. Journal
of Fish Diseases 2000, 23, 115—127.

156

Berge, G.E., A.M. Bakke-McKellepb and E. Lied, 1999. In vitro uptake and interaction
between arginine and lysine in the intestine of Atlantic salmon (Salmo salar),
Aquaculture 179, 1-4, pp.81-193.

Berge, G. E., H. Sveier and E. Lied, 1998. Nutrition of Atlantic salmon (Salmo salar).
The requirement and metabolic effect of lysine. Comp. Biochem. Physiol. 120A, 477.

Berge, G. E., E. Lied, and H. Sveier, 1997. Nutrition of Atlantic salmon (Salmo salar).
The requirement and metabolism of arginine. Comp. Biochem. Physiol. 117A, 501.

Bergot, F., 1979. Effects of dietary carbohydrate and of their mode of distribution on

glycemia in rainbow trout (Salmo gairdneri Richardson). Comp. Biochem. Physiol. 64,
534-547.

Bigelow, H.B., and WC. Schroeder. 1953. Fishes of the Gulf of Maine. US. Fish Wildl.
Serv. Fish. Bull. 53(74):1-577.

Brauge, C., G. Corraze, and F. Medale 1995. Effects of dietary levels of carbohydrate
and lipid on glucose oxidation and lipogenesis from glucose in rainbow trout,

Oncorhynchus mykiss, reared in freshwater or seawater. Comp. Biochem. Physiol. 111A,
1 1 7-124.

Brauge, C., F. Medale and G. Corraze, 1994. Effect of dietary carbohydrate-levels on
growth, body-composition and glycemia in rainbow-trout, Oncorhynchus-mykiss, reared
in seawater. Aquaculutre 123 (l-2):109-120.

Brown, P.B., K.A. Wilson, Y. Hodgin, and J. Stanley, 1997. Use of soy protein
concentrates and lecithin products in diets fed to coho and Atlantic sahnon. Journal of
the American Oil Chemists Society 74:187-194.

Buhler, DR. and J .E. Halver, 1961. Nutrition of salmonid ﬁshes. IX. Carbohydrate
requirements of Chinook salmon. J. Nutr,. 33, 219-231.

Bureau, D.P., A. M. Harris, and CY. Cho, 1998. The effects of puriﬁed alcohol extracts
from soy products on feed intake and growth of Chinook salmon (Oncorhynchus
tshawytscha) and rainbow trout (Oncorhynchus mykiss). Aquaculture, 161, 27—43.

Buttle, L.G., A.C. Burrells, J .E. Good, P.D. Williams, P]. Southgate, and C. Burrells,
2001. The binding of soybean agglutinin (SBA) to the intestinal epithelium of Atlantic
salmon, Salmo salar and rainbow trout, Oncorhynchus mykiss, fed high levels of soybean
meal. Veterinary Immunology and Immunopathology 80: 237-244.

Calabrese, E.J. and LA. Baldwin, 2003. Toxicology rethinks its central belief. Nature
421, 691-692.

157

Carter, CG. and RC. Hauler, 2000. Fish meal replacement by plant meals in extruded
feeds for Atlantic salmon (Salmo salar). Aquaculture 185, 299-311.

CDC, 2007. Avian Inﬂuenza. http://www.cdc.gov/ﬂu/avian/. Centers for Disease
Control and Prevention.

Cho, CY. and SJ. Slinger. 1979. Apparent digestibility measurement in feedstuffs for
rainbow trout. Finﬁsh Nutrition and F ishfeed Technology (Halver, J .E. and Tiews, K.,
eds.), Vol. H. Hennemann GmbH & Co.. Berlin, Germany. Pp. 239-247.

Chrispeels M.J. and NV. Raikhel, 1991. Lectins, lectin genes, and their role in plant
defense. The Plant Cell 3, 1-9.

Cowley, DJ. and Sheridan, M.A., 1993. Insulin stimulates hepatic lipogenesis in
rainbow trout, Oncorhynchus mykiss. Fish Physiol. Biochem., 11,421—428.

Craig S. and L. A. Helfrich, 2002. Understanding ﬁsh nutrition, feeds, and feeding.
Virginia Cooperative Extension Publication Number 420-256. Blacksburg, VA.

Danie, D.S., J .G. Trial, and J .G. Stanley, 1984. Species Proﬁles: Life histories and
environmental requirements of coastal ﬁsh and invertebrates (North Atlantic) Atlantic
salmon. National Wetlands Research Center. U. S. Fish and Wildlife Service,
FWS/OBS- 82/11. 22. July. 1984.

Davies, S. J. 1988. The role of dietary ﬁbre in ﬁsh nutrition. Pp. 219-249 in Recent
Advances in Aquaculture, J. F. Muir and R. R. Roberts, eds. London.

Davies, S.J., P.C. Morris, and R.T. Baker, 1997. Partial substitution of ﬁsh meal and full-
fat soya bean meal with wheat gluten and inﬂuence of lysine supplementation in diets for

rainbow trout, Oncorhynchus mykiss (Walbaum). Aquaculture Research 28, 317-328,
1997.

Dekich M.A., 1998. Broiler industry strategies for control of respiratory enteric disease.
Poultry Science 77:1 176-1 180.

Dersj ant-Li, Y., 2002. The use of soy protein in aquafeeds. In: Cruz-Suarez, L.E., D.
Ricque-Marie, M. Tapia-Salazar, M.G. Gaxiola—Cortes and N. Simoes (Eds). Avances en
Nutricion Acuicola VI. Memorias del VI Simposium Internacional de Nutricion
Acuicola. 3a] 6 de Septiembre del 2002. Cancun, Quintana Roo. Mexico.

De Silva SS. and TA. Anderson, 1995. Fish nutrition in aquaculture. Chapman & Hall.
London.

Dong, F.M., R.W. Hardy, and DA. Diggs, 2000. Antinutritional Factors. In R.R.
Stickney (Ed), Encyclopedia of Aquaculture (pp. 4551). New York: John Wiley and
Sons, Inc.

158

EC Regulation 999/2001 of the European Parliament and of the Council laying down
rules for the prevention, control and eradication of certain transmissible spongiform
encephalopathies. Ofﬁcial Journal of the European Communities, 31 May 2001, L 147,
pp 1—40.

Einen, O. and A. J. Roem, 1997. Dietary protein/energy ratios for Atlantic salmon in
relation to ﬁsh size: growth, feed, utilization and slaughter quality. Aquaculture
Nutrition, 3, 115—116.

Eldridge AC. and W.F. Kwolek, 1983. Soybean isoﬂavones: effect of environment and
variety on composition. J. Agric. Food Chem. 1983, 31, 394-396.

EISaidy, D.M.S.D. and M.M.A. Gaber, 2002. Complete replacement of ﬁsh meal by
soybean meal with dietary L-lysine supplementation for Nile tilapia Oreochromis
niloticus (L. ) ﬁngerings. J. World Aquaculture Soc. 33(3): 297-306.

FAO, 2004. The State of World Fisheries and Aquaculture — 2004. Food and
Agriculture Organization of the United Nations Fisheries Department, Rome.

FAO, 2006. The State of World Fisheries and Aquaculture — 2006. Food and
Agriculture Organization of the United Nations Fisheries Department, Rome.

Food and Agriculture Organization of the United Nations, 2002—2007. Fisheries Global
Information System (FIGIS). Food and Agriculture Organization of the United Nations
Fisheries Department, Rome.

Foran, J .A., D.O. Carpenter, M.C. Hamilton, B.A. Knuth, and S. J. Schwager, 2005.
Risk-based consumption advice for farmed Atlantic and wild Paciﬁc salmon

contaminated with dioxins and dioxinlike compounds. Environm. Health Perspectives,
l 1 3#5.

Francis, G., H.P.S. Makkar, and K. Becker, 2001. Antinutritional factors present in
plant-derived alternate ﬁsh feed ingredients and their effects in ﬁsh. Aquaculture 199;
197-227.

Gatlin III, D.M., 1995. Review of red drum nutrition. In Nutrition and Utilization
Technology in Aquaculture (eds. Limm CE. and DJ. Sessa), AOCS Press, Champaign,
Illinois, USA.

Garling D.L. Jr. and RP. Wilson, 1976. Optimum dietary protein to energy ratio for
channel catﬁsh ﬁngerling, Ictalurus punctatus. J. Nutr., 106 : 1368 - 1375.

Gill, C., 2005. World feed panorama: disease takes toll, but feed output bounces back.
Feed International, 26(1): 4-9.

159

Gizzi G. and D.I. Givens, 2001. Variability in feed composition and its impact on animal
production. 27/7/2001. AGRIPPA FAO electronic journal: http://www.fao.org/agrippa/.

Glencross, B.D. and WE. Hawkins, 2003. A comparison of the digestibility of several
lupin (Lupinus spp.) kernel meal varieties when fed to either rainbow trout

(Oncorhynchus mykiss) or red seabream (Pagrus auratus). Aquaculture Nutrition.
Volume 10 (2) 65-73.

Guay F. and N. L. Trottier, 2006. Muscle growth and plasma concentrations of amino
acids, insulin-like grth factor-I, and insulin in growing pigs fed reduced-protein diets.
J. Anim Sci. 2006. 84:3010-3019.

Guillaume J ., S. Kaushik, P. Bergot and R. Metailler, 2001. Nutrition and feeding of ﬁsh
and crustaceans. Praxis Publishing Ltd, Chichester, UK.

Haard, N.F., L.E. Dimes, R.E. Amdt, and RM. Dong, 1996. Estimation of protein
digestibility — IV. Digestive proteinases from the pyloric caeca of coho salmon

(Oncorhynchus kisutch) fed diets containing soybean meal. Comp. Biochem. Physiol.
115B, 533-540.

Halver , J .E. 2002. The vitamins. In Fish Nutrition 3rd ed. (eds J .E. Halver and R.W.
Hardy), Academic Press, California, USA.

Halver, J .E., ed. Fish nutrition. Academic Press, New York NY, 1972.

Hardy, R., 2003. Use of soybean meals in diets of salmon and trout. United Soybean
Board and American Soybean Association publication AQ 141 - 2003 — EN.

Hardy, R.W., 2000. New developments in aquatic feed ingredients, and potential of
enzyme supplements. In: Cruz -Suarez, L.E., Ricque-Marie, D., Tapia—Salazar, M.,
Olvera-Novoa, M.A. y Civera-Cerecedo, R., (Eds). Avances en Nutricion Acuicola
V.Memorias del V Simposium Intemacional de Nutricion Acuicola. 19-22 Noviembre,
2000. Mérida, Yucatan, Mexico.

Hardy R., 1996. Alternate protein sources for salmon and trout diets. Animal Feed
Science Technology 59, 71-80.

Hemre G.I., K. Sandnes, O. Lie, 0. Torrissen, and R. Waagbo, 1995. Carbohydrate
nutrition in Atlantic salmon, Salmo salar L.: growth and feed utilization. Aquaculture
Research 26: 149—154.

Hemre G.I., O. Lie, and A. Sundby, 1993. Dietary carbohydrate utilization in cod (Gadus
morhua): metabolic responses to feeding and fasting. Fish Physiology and Biochemistry
10:45 5—463.

160

Hertrampf, J.W. and F. Piedad-Pascual, 2000. Handbook on ingredients for aquaculture
feeds. Kluwer Academic Publishers, Netherlands.

Hilton J .W. and J .L. Atkinson JL, 1982. Responses of rainbow trout to increased levels of
available carbohydrate in practical trout diets. Br J Nutr 47, 597—607.

Hilton, J. W., J. L. Atkinson and S. J. Slinger, 1983. Effect of dietary ﬁber on the growth
of rainbow trout (Salmo gairdneri). Can. J. Fish. Aquat. Sci. Vol. 40, no. 1, pp. 81-85.

Hites, R.A, J .A. Foran, D.O. Carpenter, M. Coreen Hamilton, B.A. Knuth, and SJ.
Schwager, 2004. Global assessment of organic contaminants in farmed salmon. Science,
303:226-229.

Hollung K, M. Overland, M. Hrustic, P. Sekulic, J. Miladinovic, H. Martens, B. Narum,
S. Sahlstrom, M. Sorensen, T. Storebakken, and A. Skrede. 2005. Evaluation of
nonstarch polysaccharides and oligosaccharide content of different soybean varieties

(Glycine max) by near-infrared spectroscopy and proteomics. J. Agric. Food Chem. 53,
9112-9121.

IFREMER, 2005. Atlantic salmon. French Research Institute for Exploitation of the Sea
Online Technical Bulletin, http://www.ifremer.fr/aquaculture/en/ﬁsh/salmonidae.htm.

Ireland PA, S.Z. Dziedzic and M.W. Kearsley, 1986. ‘Saponin content of soya and some
commercial soya products by means of hi gh-performance liquid chromatography of the
sapogenins. J Sci Food Agric 37:694—698, 1986.

J ahan, P., T. Watanabe, V. Kiron and S. Satoh, 2003. Improved carp diets based on plant
protein sources reduce environmental phosphorus loading. F siheries Science 69; 219-
225.

Jensen A.J. , B.O. Johnsen and T. G. Heggberget, 1991. Initial feeding time of Atlantic
salmon, Salmo salar, alevins compared to river ﬂow and water temperature in Norwegian
streams. Environmental Biology of Fishes, 30 (4), 379-385.

Kakade, M.L., N. Simons and LE. Liener, 1969. An evaluation of natural vs. synthetic
substrates for measuring the antitryptic activity of soybean samples. Cereal. Chem. 40,
5 1 8-5 26.

Kaushik, S.J., Oliva-Teles, A., 1985. Effect of digestible energy on nitrogen and energy
balance in rainbow trout. Aquaculture, 50: 89-101.

Krogdahl A., A.M. Bakke-Mckellep, and G. Baeverﬁord, 2003. Effects of graded levels
of standard soybean meal on intestinal structure, mucosal enzyme activities, and

pancreatic response in Atlantic salmon (Salmo salar, L.). Aquaculture Nutrition 9:361-
371.

161

Krogdahl, A., A.M. Bakke-McKellep, K.H. Roed, and G. Baeverfjord, G., 2000. Feeding
Atlantic salmon (Salmo salar) woth soybean products: Effects on disease resistance
(ﬁrrunculosis), and lysozyme and IgM levels in the intestinal mucosa. Aquaculture
Nutrition 6, 77—84.

Krogdahl, A., T.B. Lea, and 1.]. Olli, 1994. Soybean proteinase inhibitors affect
intestinal trypsin activities and amino acid digestibilities in rainbow trout (Oncorhynchus
mykiss). Comp. Biochem. Physiol. 107A, 215-219.

Lall, SP. 2002. The minerals. In Fish Nutrition 3” ed. (eds J.E. Halver and R.W.
Hardy), Academic Press, California, USA.

Lall, S.P., S.J. Kaushik, P.Y. Le Bail, R. Keith, J .S. Anderson and E. Plisetskaya, 1994.
Aquaculture 124, 13.

Latouche, K., P. Rainelli, and D. Verrnersch, 1998. Food Safety Issues and the BSE
Scare: Some Lessons from the French Case. Food Policy 23-5: 347-56.

Liener, I.E. (Editor), 1980. Toxic constituents of plant foodstuffs. 2nd edition, Academic
Press, N.Y., 502pp.

Mambrini, M., A.J. Roem, J .P. Cravedi, J .P. Lalles, and SJ. Kaushik, 1999. Effects of
replacing ﬁsh meal with soy protein concentrate and of DL-methionine supplementation
in high-energy diets on the grth and nutrient utilization of rainbow trout,
Oncorhynchus mykiss. Journal of Animal Science, 77, 2990-2999.

Massachusetts Ofﬁce of Coastal Zone Management, 1995. Commercially important
ﬁnﬁsh. Massachusetts Ofﬁce of Coastal Zone Management White Paper. Boston, MA.

Miles RD. and F. A. Chapman, 2006. The beneﬁts of ﬁsh meal in aquaculture diets.
Department of Fisheries and Aquatic Sciences, Florida Cooperative Extension Service,
Institute of Food andAgricultural Sciences, University of Florida. FA122.

Mills, DH. 1989. Ecology and management of Atlantic salmon. Chapman and Hall,
London.

Morin M., D. Turgeon, J. J olette, Y. Robinson, J .B. Phaneuf, R. Sauvageau, M.
Beauregard, E. Teuscher, R. Higgins and S. Lariviere, 1983. Neonatal Diarrhea of Pigs
in Quebec: Infectious Causes of Signiﬁcant Outbreaks. Can J Comp Med. 47(1): 11—17.

New, MB. and UN. Wijkstrom, 2002. Use of ﬁshmeal and ﬁsh oil in aquafeeds: further
thoughts on the ﬁshmeal trap. FAO Fisheries Circular No. 975. Rome. 61 pp.

NRC (National Research Council), 2003. Nutrient Requirements of Nonhuman Primates:
Second Revised Edition. National Academy Press, Washington, DC.

162

NRC (National Research Council), 1993. Nutrient requirements of ﬁsh. National
Academy Press, Washington, DC.

NRC (National Research Council). 1973 Nutrient requirements of 20 domestic animals:
Nutrient requirements of trout, salmon and catﬁsh. National Academy of Sciences,
Washington DC.

Ojanguren A.F., F.G. Reyes-Gavilan and RR. Munoz, 1999. Effects of temperature on
growth and efﬁciency of yolk utilization in eggs and pre-feeding stages of Atlantic
salmon. Aquacult Int. 7:81—87.

Ollie J .J ., K. Hjelmeland, and A. Krogdahl, 1994. Soybean trypsin inhibitors in diets for
Atlantic salmon (salmo salar, L): effects on nutrient digestibilities and trypsin in pyloric
caeca homogenate and intestican content. Comp. Biochem. Physiol 109A (2) 923-928.

Parrish, D. L., R.J. Behnke, S.R. Gephard, S.D. McCormick and G.H. Reeves, 1998.
Why aren’t there more Atlantic salmon (Salmo salar)? Canadian Journal of Fisheries
and Aquatic Sciences 55 (Suppl. 1), 281—287.

Peres, H., C.E. Lim and PH. Klesius, 2002. Nutritional value of heat-treated soybean
meal for channel catﬁsh (Ictalurus punctatus). Proceeding of Fish Nutrition and Feeding
Symposium Proceedings. p. 190.

Pike, 1., 2005. Eco-efﬁciency in aquaculture: global catch of wild ﬁsh used in
aquaculture. International Aquafeed, 8(1):38—40.

Piper, R. G., I. B. McElwain, L. E. Orme, J. P. McCraren, L. G. Fowler, and J. R
Leonard. 1982. Fish hatchery management. US. Fish and Wildlife Service,Washington,
D.C.

Prophet, E.B., B. Mills and J .B. Arrington, 1992. In Sobin L.S. (ed) Laboratory Methods
in Histology. Washington DC: Armed forces Institute of pathology.

Refstie S., T.Storebakken, G. Baevertjord and A.J. Roem, 2001. Long-term protein and
lipid growth of Atlantic salmon (Salmo salar) fed diets with partial replacement of ﬁsh
meal by soy protein products at medium or high lipid level. Aquaculture 193 (2001) 91 -
106.

Refstie, S., O.J. Korsoeen, T. Storebakken, G. Baeverﬁord, I. Lein, and A.J. Roem, 2000.
Differing nutritional responses to dietary soybean meal in rainbow trout (Oncorhynchus
mykiss) and Atlantic salmon (Salmo salar). Aquaculture, 193(1-2):91-106.

Refstie, S., B. Svihus, K.D. Shearer, T. Storebakken, 1999. Nutrient digestibility in
Atlantic salmon and broiler chickens related to viscosity and non-starch polysaccharide

content in different soyabean products. Animal Feed Science and Technology 79 (1999)
33 1-345.

163

Roberts J .R., 2002. Nutritional Pathology. In Fish Nutrition 3rd ed. (eds J .E. Halver and
R.W. Hardy), Academic Press, California, USA. 453-504.

Rollin, X. 1999. Critical Study of Indispensable Amino Acids Requirements of Atlantic
Salmon (Salmo salar L.) Fry, Ph.D. thesis. Universite catholique de Louvain, Louvain,
Belgium.

Rumsey, G.L., J .G. Endres, P.R. Bowser, K.A. Eamest-Koons and AK. Siwicki, 1995.
Soy proteins in diets of rainbow trout: effects on growth, protein absorption,
gastrointestinal histology and non-speciﬁc serologic and immune response. In C. Lim
and DJ. Sessa (ed), Nutrition and Technology in Aquaculture (166-188). Champaign,
IL: AOCS Press.

Rumsey, G.L., A.K. Siwicki, D.P. Anderson and RR. Bowser, 1994. Effect of soybean
protein on serological response, non-speciﬁc defense mechanisms, growth and protein
utilization in rainbw trout. Veterinary Immunology and Immunopathology. 41, 323-339.

Russett, J. C. 1998. Soy protein concentrate for animal feeds. Chemurgy Research Notes,
SPC-T-47, Central Soya Company, Fort Wayne, IN.

Rust, MB, 2002. Nutritional physiology. In Fish Nutrition 3” ed. (eds J .E. Halver and
R.W. Hardy), Academic Press, California, USA.

Sajj adi, M. and CG. Carter, 2004. Dietary phytase supplementation and the utilization of
phosphorus by Atlantic salmon (Salmo salar L.) fed a canola-meal-based diet.
Aquaculture, 240:417-43 1 .

Sanchez-Muros M.J., L. Garcia-Rejon, J.A. Lupiafiez and M. De la Higuera, 1996. Long-
tenn nutritional effects on the primary liver kidney metabolism in rainbow trout
(Oncorhynchus mykiss). 2. Adaptative response of glucose 6-phosphate dehydrogenase
activity to high-carbohydrate/low-protein and high-fat/non-carbohydrate diets.
Aquaculture Nutrition 2: 193-200.

Sanden, M., A. Krogdahl, A.M. Bakke-McKellep, R.K. Buddington, and GI. Hemre,
2006. Growth performance and organ development in Atlantic salmon, Salmo salar L.

parr fed genetically modiﬁed (GM) soybean and maize. Aquaculture Nutrition 2006 12;
1—14.

Schaffer, W.M., and PF. Elson, 1975. The adaptive signiﬁcance of variations in life
history among local populations of Atlantic sahnon in North America. Ecology
56(3):577-590.

Sedgewick, S. 1988. Salmon farming handbook. Fishing News Books, Oxford.

164

Scott, W.B., and E.J. Crossman. 1973. Freshwater ﬁshes of Canada. Bull. Fish. Res.
Board Can. 184. 996 pp.

Shepherd OJ. and Bromage NR. 1988. Intensive Fish Farming. BSP professional books,
Oxford, pp. 404.

Snyder, HE. and T. W. Kwon, 1987. Soybean Utilization. Van Nostrand Reinhold,
New York.

Storebakken, T., S. Refstie and B.Ruyter, 2000. Soy products as fat and protein sources
in ﬁsh feeds for intensive aquaculture. Drackley, J. K. (Ed), Soy in Animal Nutrition.
Federation of Animal Science Societies, 127-170.

Storebakken, T., I.S. Kvien, K.D. Shearer, B. Grisdale-Helland, S.J. Helland and GM.
Berge, 1998. The apparent digestibility of diets containing ﬁsh meal, soybean meal or
bacterial meal fed to Atlantic salmon (Salmo salar): evaluation of different faecal
collection methods. Aquaculture 169, 195-210.

Sveier H., B.O. Kvamme, and A.J. Raae, 2001. Growth and protein utilization in

Atlantic salmon (Salmo salar L.) given a protease inhibitor in the diet. Aquaculture
Nutrition 2001 7; 255-264.

Tacon, A.G.J., 2004. Use of ﬁsh meal and ﬁsh oil in aquaculture: a global perspective.
Aquatic Resources, Culture & Development, l(l):3-14 .

Tacon, A.G.J., 2005. State of information on sahnon aquaculture feed and the
environment. Prepared for the World Wildlife Fund/Salmon Aquaculture Dialogue.
http ://www.worldwildlife.org/cci/pubs/Feed_ﬁnal_resaved2 .pdf.

Thiessen D.L., D.D. Maenz, R.W. Newkirk, H.L. Classen and MD. Drew, 2004.
Replacement of ﬁshmeal by canola protein concentrate in diets fed to rainbow trout
(Oncorhynchus mykiss). Aquaculture Nutrition 10; 379-388.

Torstensen BE, J .G. Bell, G. Rosenlund, R.J. Henderson, I.E. Graff, D.R. Tocher, O. Lie,
and J .R. Sargent, 2005. Tailoring of Atlantic salmon (Salmo salar L.) ﬂesh lipid

composition and sensory quality by replacing ﬁsh oil with a vegetable oil blend. J Agric
Food Chem. 28;53(26):10166-78.

Westers, H., 1987. Feeding levels of ﬁsh fed formulated diets. Progressive F ish-
Culturist 49 (2): 87-92.

Wijkstrém, UN, 2003. Short and long - term prospects for consumption of ﬁsh.
Veterinary Research Communications, 27 Suppl. 1: 461-468.

Williams, C. H., D. J. David and O.Iismaa, 1962. The determination of chromic oxide in
faeces samples by atomic absorption spectrophotometry. J. Agric. Sci. 59:381-385.

165

Wilson RP, 2002. Amino acids and proteins. In Fish Nutrition 3rd ed. (eds J .E. Halver
and R.W. Hardy), Academic Press, California, USA. 144-175.

Wilson, T.R., 1992. Full-fat soybean meal-an acceptable, economical ingredient in
Chinook salmon grower feeds. 1992. Seattle, WA, University of Washington.

You, T., J. Guo, I. Baianu, and R. Nelson. Determination of isoﬂavone contents for

selected soybean lines by Fourier transform near infrared reﬂectance spectroscopy.
Cellular and Molecular Biology of Soybean Biennial Conference. 2002. p.505.

166

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

illWWI111111111!llll11111111111llillll1|