INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may
be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the
copy subm itted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UM I a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize m aterials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand corner and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in
reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6" x 9" black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly
to order.

University Microfilms International
A Bell & Howell Information C om pany
300 North Z eeb Road. Ann Arbor. Ml 48106-1346 USA
313/761-4700 800/521-0600

O rder N u m b er 9406510

I m p a c t o f Renibacterium salmoninarum o n n o n -lip id en erg y , lip id
a n d w a te r c o n te n ts in liv e r o f in fe c te d c h in o o k sa lm o n d u rin g
fa ll a n d sp r in g in L ake M ic h ig a n , 1 9 9 0 -1 9 9 2
Kabre, Jean Andre Tinkoudgou, Ph.D .
Michigan State University, 1993

UMI

300 N. Zeeb Rd.
Ann Arbor, MI 48106

IMPACT OP RENIBACTERIUM SALMONINARUM ON NON-LIPID ENERGY,
LIPID AND WATER CONTENTS IN LIVER OP INFECTED CHINOOK SALMON
DURING PALL AND SPRING IN LAKE MICHIGAN, 1990-1992

By

Jean Andre Tinkoudgou Kabre

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
1993

ABSTRACT

IMPACT OP RENIBACTERIUM SALMONINARUM ON NON-LIPID ENERGY,
LIPID AND WATER CONTENTS IN LIVER OP INFECTED CHINOOK SALMON
DURING FALL AND SPRING IN LAKE MICHIGAN, 1990-1992

By

Jean Andre Tinkoudgou Kabre

(Body of Abstract)

Since 1987, Bacterial Kidney Disease

(BKD) is

considered to have significantly contributed to increased
mortality and decreased growth of Chinook salmon in Lake
Michigan.

To assess the impact of this disease during the

fall and spring on the general condition of the fish, the
non-lipid energy, lipid and water contents were evaluated
from the liver of chinook salmon.

A total of 346 fish were

collected at 2 sites (Pentwater and Ludington)
Michigan during 1990-1992.
between 1+ to 3+ years.

in Lake

The age of the fish ranged

All fish were examined for BKD

infection and then grouped into non-BKD and diseased (BKD)
fish based on their clinical signs.

Following Renibacterium

salmoninarum infection, a 3.15 % decrease in non-lipid
energy and 48.55 % in liver lipid contents were observed.
This loss in lipid reserves indicates the critical role of

lipid as an energy source during BKD stress conditions.

The

losses in energy reserve were followed by a significant
accumulation of water (6.74 % increase)

in the liver.

Evaluation of the relationship of the severity of the
disease and loss of energy showed that fish experience a
sharp decrease in energy reserves as the virulence of R.
salmoninarum increases during the spring.
The influence of the season was explained by a
significant use of non-lipid energy in the fall while
important losses in lipid were observed during the Spring.
Concurrently, the liver water increased (2.04 %)
significantly as fall fish enter the winter and spring.
It was concluded that the BKD and season-linked changes
in energy reserve and water content in liver reflected the
general condition of fish and that lipid reserves constitute
a critical source of energy during disease stress condition.
Increase in mortality and decreased growth are likely
functions of BKD infection and loss of lipid during fall and
spring.

Key w o r d s : Chinook salmon; Oncorhvnchus tshawvtscha;
Bacterial Kidney Disease; Renibacterium salmoninarum; liver
contents; proximate analysis; Lake Michigan.

To my father Tinga Kabre and
my mother Manegre Tinoaga Zongo
who respectively passed away in 1969 and 1983.

ACKNOWLEDGMENTS
I would like to thank the members of my dissertation
guidance committee, Dr. William W. Taylor, Dr. Scott
Winterstein, Dr. Patrick M. Muzzall, Dr Richard W. Hill and
Dr. Carl Latta for their frequent advice aimed at improving
the quality of this research.
I have worked closely with several fisheries
researchers and technicians of the Michigan Department of
Natural Resources during these 4 years of study, to all of
them I am very thankful. I am particularly grateful to Mr.
Myrl Keller and Mr. Asa Wright for the technical support and
the particular interest they had for this work. My infinite
thanks go to Mr. John G. Hnath and Mrs. Julia E. Zischke
that have made their Fish Health Laboratory and their great
experience available to me.
Without the help of the crew of the Stealhead Boat
(Charles Cutler, Robert Russell and Jerry Meggison), John
Gulvas

(Consumers Power Company) and Robert Elliot and his

interns at the Ludington Pump Storage, I never could have
collected all the fish needed for this study.

For the work

you did for me I thank you a l l .
I am also thankful to Dr. John P. Giesy who allowed me
to use his Laboratory for my early attempt to isolate the
bacteria through culture media.
I also thank my fellows graduate students, the interns
and work-study students working in the Fisheries Lab.
v

Finally,

I remain grateful to my late parents, Tinga

Kabre and Manegre Tinoaga Zongo for instilling in me the
desire to further my education.

TABLE OF CONTENTS

Page
List of

Tables.............................................. ix

List of

Plates...................................... ...... xiii

List of

Figures.

List of

Appendices.

CHAPTER I.
BACTERIAL

................................... xiv
...................

xxviii

INTRODUCTION AND LITERATURE REVIEW ON
KIDNEY DISEASE. . .... ........ ........ ........ 1

1.1.

Statement of the Problem

1.2.

Need for the study

1.3.

Choice of the liver............................. 3

1.4.

Goal of Study ....................................3

1.5.

Statistical analysis........ .................. 4

................ 1
......... 2

1.6. Literature review on Bacterial Kidney
Disease.......................................... 5
CHAPTER II.

MATERIALS AND METHODS

2.1. Collection and diagnosis of
2.2. Evaluation of water,

fish...............10

lipid and non-lipid

energy contents.......
2.3.

................... 10

14

Statistical analysis............................ 15

CHAPTER III.

RESULTS..............

19

3.1.

Separation of fish............................ ..19

3.2.

Factorial desi gn

......................... .34

CHAPTER IV.

DISCUSSIONS AND CONCLUSION................. 83

4.1. Occurrence and impact of Bacterial
Kidney Disease.................................. 83
4.2. Quantification of Renibacterium salmoninarum..84
4.3. Pattern of fish separation.......... .......... 85
4.4. Impact of Renibacterium salmoninarum
infection........................................85
4.5. Control of Bacterial Kidney Disease...........86
APPENDICES......... ........................................ 89
Appendix A - l ............. . ........ ................. 89
Appendix A-2

.............................. ....... 95

Appendix B .......... ............................... . .101
LIST OF REFERENCES.........................................102

viii

LIST OF TABLES

Page
Table

1: Number of Renibacterium salmoninarum
counted under fluorescent light and body
fluid content related to three different
stages of Bacterial Kidney Diseased (BKD)

fish:

stage 1 (BKD1), stage 2 (BKD2) and stage 3

Table

2: Correlation matrices for n'on-lipid
energy (LE), lipid (LF) and water (LW) contents
in liver of Chinook salmon caught in Fall and
Spring in Lake Michigan (1990-1992).................. 26

Table

3: Principal Component Analysis

of non-lipid

energy, lipid and water contents in liver of
non-Bacterial Kidney Diseased (non-BKD) and
Bacterial Kidney Diseased (BKD) Chinook salmon
caught in Fall in Lake Michigan
Table

(1990-1992)......... 27

4: Principal Component Analysis

of non-lipid

energy, lipid and water contents in liver
of non-Bacterial Kidney Diseased (non-BKD)
and Bacterial Kidney Diseased (BKD) Chinook
salmon caught in Spring in Lake Michigan
1992)...............................
ix

(199028

Table

5: Eigenvectors of the measured non-lipid
energy,

lipid and water contents in liver of

non-Bacterial Kidney Diseased (non-BKD) and
Bacterial Kidney Diseased (BKD) Chinook salmon
caught in Fall and Spring in Lake Michigan (19901992)...........................
Table

31

6: ANOVA of 23 factorial design of non­
lipid energy (LE), lipid (LF) and water (LW)
contents in liver of non-Bacterial Kidney
Diseased (non-BKD) and Bacterial Kidney Diseased
(BKD) chinook salmon caught in Fall and Spring
in Lake Michigan

Table

(1990-1992)............ ............35

7: One factor analyses and
risons of means of non-lipid

Pairwise compa­
energy,

lipid and

water contents in liver of chinook salmon using
Fish er 's Least Significant Difference (LSD).
The fish were caught in Fall and Spring in Lake
Michigan (1990-1992).............
Table

38

8: Means and Standard Errors (SE) of non­
lipid energy (LE), lipid (LF) and water (LW)
contents in liver of chinook salmon caught in
Fall and Spring in Lake Michigan (1990-1992).
Two factor combinations of treatment were
considered. See Table 9 for Pairwise comparisons
of m e a n s ..................

43

Table

9: Two factors analyses and Pairwise compa­
risons of means of non-lipid energy (LE), lipid
(LF) and water (LW) contents in liver of chinook
salmon using Fisher's Least Significant Diffe­
rence (LSD).

The fish were caught in FA11 and

Spring in Lake Michigan (1990-1992)................. 44
Table

10: Means and Standard Errors (SE) of non­
lipid energy (LE), lipid (LF) and water (LW)
contents in liver of chinook salmon caught in
Fall and Spring in Lake Michigan (1990-1992).
Three factor treatment combinations were
considered.

See Table 11 for Pairwise comparisons

of m e a n s . ........
Table

54

11: Three factor analyses and Pairwise
comparisons of means of non-lipid energy (LE) ,
lipid (LF) and water (LW) contents in liver
of chinook salmon using Fisher's Least Signi­
ficant Difference (LSD).

The fish were caught

in FA11 and Spring in Lake Michigan (19901992).................
Table

55

12: ANOVA of 4x2 factorial design of
non-lipid energy (LE), lipid (LF) and water
(LW) contents in liver of non-Bacterial
Kidney Diseased (non-BKD) and Bacterial Kidney
Diseased (BKD) chinook salmon caught in Spring
in Lake Michigan (1990-1992)......................... 63
xi

Table

13: Means and Standard Errors

(SE) of non­

lipid energy (LE), lipid (LF) and water (LW)
contents in liver of chinook salmon caught
in Spring in Lake Michigan (1990-1992).

One

factor treatment combinations were considered.
See Table 14 for Pairwise comparisons of
means
Table

......... ..........................

14 : One factors analyses and Pairwise
comparisons of means of non-lipid energy (LE),
lipid (LF) and water (LW) contents in liver
of chinook salmon using Fisher's Least Signi­
ficant Difference (LSD).

The fish were caught

in Spring in Lake Michigan (1990-1992)....
Table

15: Means and Standard Errors

(SE) of non­

lipid energy (LE), lipid (LF) and water (LW)
contents in liver of chinook salmon caught
in Spring in Lake Michigan (1990-1992).

Two

factor treatment combinations were considered.
See Table 16 for Pairwise comparisons of
m e a n s ............................................
Table

16: Two factor analyses and Pairwise com­
parison of means of non-lipid energy (LE),
lipid (LF) and water (LW) contents in liver
of chinook salmon using Fisher's Least Signi­
ficant Difference (LSD).

The fish were caught

in Spring in Lake Michigan (1990-1992)

.... .

LIST OF PLATES

Page
Plate 1: Gross lesions produced by Renibacterium
salmoninarum in chinook salmon. Three stages
of the disease

with corresponding severity of

necrotic areas or nodules (indicated by the
arrows) are displayed: Bacterial Kidney Disease
stage 1 (A) , Bacterial Kidney Disease stage 2 (B)
and Bacterial Kidney Disease stage 3 (C). The
small arrow in stage 3 (C) indicates the
spleen covered with cheesy materials................ 20
Plate 2: Cells of Renibacterium salmoninarum stained
with crystal violet. Three stages of the disease
with their relative density of cells are displayed:
Bacterial Kidney Disease stage 1 (A), Bacterial
Kidney Disease stage 2 (B) and Bacterial Kidney
Disease stage 3 (C). The arrows indicate the cells
(diplobacilli)

of

R. salmoninarum.................. 22

xiii

LIST OF FIGURES

Page
Figure

1: Map of Lake Michigan with sampling

sites (1, 2 ) .......................................... 11
Figure

2: Direct optic viision count of cells of

Renibacterium salmoninarum using the Fluorescent
Antibody Technique (FAT). Three disease stages
ranging from low to high virulence are iden­
tified: Bacterial Kidney Disease stage 1
(BKD1), stage 2 (BKD2) and stage 3 (BKD3). The
vertical bars represent +/- one Standard Error
(SE)
Figure

...............

24

3: Principal Components plots for non­

lipid energy (LE), lipid (LF) and water (LW)
contents in liver of non-Bacterial Kidney
Diseased (0) and Bacterial Kidney Diseased (2)
chinook salmon caught in Fall in Lake Michigan
(1990-1992). Plot B gives a complete separation
of the fish after (0) fish, overlapping with (2)
fish in plot A, were discarded (see text for more
detail)................................................ 29

xiv

Figure

4: Principal Components plots for non­

lipid energy (LE), lipid (LF) and water (LW)
contents in liver of non-Bacterial Kidney
Diseased (0) and Bacterial Kidney Diseased
stagel

(1), stage2

(2) and stage3

(3) chinook

salmon caught in Spring in Lake Michigan
(1990-1992). Plot B gives a complete separation
of the fish after

(0) fish, overlapping with (1),

(2) and (3) fish in plot A, were discarded (see
text for more detail)
Figure

......

30

5: Eigenvectors plot for non-lipid energy

(LE), lipid (LF) and water (LW) contents in
liver of non-Bacterial Kidney Diseased and
Bacterial Kidney Diseased chinook salmon caught
in Fall in Lake Michigan (1990-1992). The lines
indicate the direction of change of the variables
LE, LF and L W .
Figure

........

32

6: Eigenvectors plot for non-lipid energy

(LE), lipid (LF) and water (LW) contents in
liver of non-Bacterial Kidney Diseased and
Bacterial Kidney Diseased chinook salmon caught
in Spring in Lake Michigan (1990-1992). The lines
indicate the direction of change of the variables
L E , LF and L W

.....

33

xv

Figure

7: Pairwise comparisons of means of non­

lipid energy content in liver of non-Bacterial
Kidney Diseased (Non-BKD) and Bacterial Kidney
Diseased

(BKD) male and female chinook salmon

caught in Fall and Spring in Lake Michigan
(1990-1992). The fish were grouped into
different treatment combinations correspon­
ding to each level of the factors health,
sex and season. The vertical bars represent
+/- one Standard Error............
Figure

.....40

8: Pairwise comparisons of means of lipid

content in liver of non-Bacterial Kidney
Diseased (Non-BKD) and Bacterial Kidney
Diseased (BKD) male and female chinook salmon
caught in Fall and Spring in Lake Michigan
(1990-1992). The fish were grouped into diffe­
rent treatment combinations corresponding
to each level of the factors health, sex and
season. The vertical bars represent +/- one
Standard Erro r.................

xv i

41

Figure

9: Pairwise comparisons of means of water

content in liver of non-Bacterial Kidney
Diseased (Non-BKD) and Bacterial Kidney
Diseased (BKD) male and female chinook salmon
caught in Fall and Spring in Lake Michigan
(1990-1992). The fish were grouped into diffe­
rent treatment combinations corresponding to
each level of the factors health, sex and
season. The vertical bars represent +/- one
Standard Error.
Figure

.....

42

10: Pairwise comparisons of means of non­

lipid energy content in liver of non-Bacterial
Kidney Diseased (Non-BKD) and Bacterial Kidney
Diseased (BKD) chinook salmon caught in Fall
and Spring in Lake Michigan (1990-1992). The
fish were grouped into treatment conbinations:
sex*health (A) and sex*season (B)......
Figure

46

11: Pairwise comparisons of means of lipid

content in liver of non-Bacterial Kidney
Diseased (Non-BKD) and Bacterial Kidney Diseased
(BKD) chinook salmon caught in Fall and Spring
in Lake Michigan (1990-1992). The fish were
grouped into treatment combinations: sex*health
(A) and sex*season (B).................................48

xvii

Figure

12: Pairwise comparisons of means of water

content in liver of non-Bacterial Kidney
Diseased

(Non-BKD) and

Bacterial Kidney

Diseased

(BKD) chinook

salmon caught in Fall

and Spring in Lake Michigan (1990-1992). The
fish were grouped into treatment combinations:
sex*health (A) and sex*season (B)................... 49
Figure

13: Pairwise comparisons of means of non­

lipid energy content in liver of non-Bacterial
Kidney Diseased (Non-BKD) and
Diseased

(BKD) chinook

Bacterial Kidney

salmon caught in Fall

and Spring in Lake Michigan (1990-1992). The
fish were grouped into treatment combinations:
heath*season (A) and season*health (B). The
vertical bars represent +/- one Standard Er r o r ..... 50
Figure

14: Pairwise comparisons of means of lipid

content in liver of non-Bacterial Kidney
Diseased

(Non-BKD) and Bacterial Kidney

Diseased (BKD) chinook salmon caught in
Fall and Spring in Lake Michigan (1990-1992).
The fish were grouped into treatment combinations:
heath*season (A) and season*health (B)..............51

xviii

Figure

15: Pairwise comparisons of means of water

content in liver of non-Bacterial Kidney
Diseased (Non-BKD) and Bacterial Kidney Diseased
(BKD) chinook salmon caught in Fall and Spring
in Lake Michigan (1990-1992). The fish were
grouped into treatment combinations: heath*
season (A) and season*health (B).................... 52
Figure

16: Pairwise comparisons of means of non­

lipid energy content in liver of non-Bacterial
Kidney Diseased (Non-BKD) and Bacterial Kidney
Diseased (BKD) male and female chinook salmon
caught in Fall and Spring in Lake Michigan
(1990-1992). The fish were grouped into Treatment
combinations: Non-BKD*sex*season (A) and BKD*
sex*season (B) . ...........................
Figure

56

17: Pairwise comparisons of means of lipid

content in liver of non-Bacterial Kidney
Diseased (Non-BKD) and Bacterial Kidney Diseased
(BKD) male and female chinook salmon caught in
Fall and Spring in Lake Michigan (1990-1992).
The fish were grouped into treatment combinations:
treatment: Non-BKD*sex*season (A) and BKD*sex*
season (B). The vertical bars represent +/- one
Standard Error. ....................................

xix

57

Figure

18: Pairwise comparisons of means of water

content in liver of non-Bacterial Kidney
Diseased (Non-BKD) and Bacterial Kidney Diseased
(BKD) male and female chinook salmon caught in
Fall and Spring in Lake Michigan (1990-1992).
The fish were grouped into treatment combinations:
Non-BKD*sex*season (A) and BKD*sex*season (B).
The vertical bars represent +/- one Standard
E r r o r ...............................
Figure

58

19: Pairwise comparisons of means of non­

lipid energy content in liver of non-Bacterial
Kidney Diseased

(Non-BKD) and Bacterial Kidney

Diseased (BKD) male and female chinook salmon
caught in Fall and Spring in Lake Michigan
(1990-1992). The fish were grouped into treatment
combinations: Male*season*health (A) and Female*
season*health (B). The vertical bares represent
+/- one Standard Error............................

xx

. .59

Figure

20: Pairwise comparisons of means of lipid

in liver of non-Bacterial Kidney Diseased
(Non-BKD) and Bacterial Kidney Diseased (BKD)
male and female chinook salmon caught in Fall
and Spring in Lake Michigan (1990-1992). The
fish were grouped into treatment combinations:
Male*season*health (A) and Female*season*health
(B). The vertical bars represent +/- one Standard
Err o r ........................... .....................
Figure

21: Pairwise comparisons of means of water

in liver of non-Bacterial Kidney Diseased
(Non-BKD) and Bacterial Kidney Diseased (BKD)
male and female chinook salmon caught in Fall
and Spring in Lake Michigan (1990-1992). The
fish were grouped into treatment combinations:
Male*season*health (A) and Female*season*health (B)
The vertical bars represent +/- one Standard
Erro r................................ .................

Figure

22: Pairwise comparisons of means of lipid

content in liver of non-Bacterial Kidney
Diseased (Non-BKD) and Bacterial Kidney Diseased
(BKD) stagel (BKD 1), stage2

(BKD 2) and stage3

(BKD3) male and female Chinook salmon caught
in Spring in Lake Michigan (1990-1992). The
fish were clustered in pairs (clusterl-cluster7)
of treatment combinations: Non-BKD*BKD(1, 2 or 3)
and Male*Female. The vertical bars represent +/one Standard Err o r ......................... . „„.....
Figure

23: Pairwise comparisons of means of non­

lipid energy content in liver of non-Bacterial
Kidney Diseased

(Non-BKD) and Bacterial Kidney

Diseased (BKD) stagel
and stage3

(BKD 1), stage2

(BKD 2)

(BKD3) male and female chinook

salmon caught in Spring in Lake Michigan
(1990-1992). The fish were clustered in pairs
(clusterl-cluster7) of treatment combinations:
Non-BKD*BKD(1, 2 or 3) and Male*Female. The
vertical bars represent +/- one Standard
Err or...............................................

xxii

Figure

24: Pairwise comparisons of means of water

content in liver of non-Bacterial Kidney
Diseased (Non-BKD) and Bacterial Kidney Diseased
(BKD) stagel (BKD 1), stage2

(BKD 2) and stage3

(BKD3) male and female Chinook salmon caught
in Spring in Lake Michigan (1990-1992). The
fish were clustered in pairs

(clusterl-cluster7)

of treatment combinations: Non-BKD*BKD(1, 2 or 3)
and Male*Female. The vertical bars represent +/one Standard Error.
Figure

......

69

25: Pairwise comparisons of means of non­

lipid energy content in liver of non-Bacterial
Kidney Diseased (Non-BKD) and Bacterial Kidney
Diseased (BKD) stagel
and stage3

(BKD 1), stage2

(BKD 2)

(BKD3) male Chinook salmon caught

in Spring in Lake Michigan (1990-1992). The
fish were clustered in pairs

(cluster1-cluster6)

of treatment combinations: NBKD male*BKD(l,
2 or 3) male. The vertical bars represent
+/- one Standard Err or............................... 73

xxiii

Figure

26: Pairwise comparisons of means of non­

lipid energy content in liver of non-Bacterial
Kidney Diseased (Non-BKD) and Bacterial Kidney
Diseased (BKD) stagel (BKD 1), stage2
and stage3

(BKD3)

(BKD 2)

female Chinook salmon caught

in Spring in Lake Michigan (1990-1992). The
fish were clustered in pairs (clusterl-cluster6)
of treatment combinations: NBKD female*BKD(l, 2
or 3) female. The vertical bars represent
+/- one Standard Error..... .....................
Figure

27: Means of non-lipid energy content in

liver of male and female non-Bacterial Kidney
Diseased (NBKD) and Bacterial Kidney Diseased
stage 1 (BKD1), stage 2 (BKD2) and stage 3
(BKD3) Chinook salmon caught in Spring in Lake
Michigan (1990-1992)..............................

xx iv

Figure

28: Pairwise comparisons of means of lipid

content in liver of non-Bacterial Kidney
Diseased (Non-BKD) and Bacterial Kidney
Diseased (BKD) stagel
and stage3

(BKD 1), stage2

(BKD 2)

(BKD3) male Chinook salmon caught

in Spring in Lake Michigan (1990-1992). The
fish were clustered in pairs

(clusterl-cluster6)

treatment combinations: Non-BKD femal.e*BKD(l,
2 or 3) male. The vertical bars represent
+/- one Standard Error.
Figure

.......

76

29: Pairwise comparisons of means of lipid

content in liver of non-Bacterial Kidney
Diseased (Non-BKD) and Bacterial Kidney
Diseased (BKD) stagel
and stage3

(BKD 1), stage2

(BKD 2)

(BKD3) female Chinook salmon caught

in Spring in Lake Michigan (1990-1992). The
fish were clustered in pairs

(clusterl-cluster6)

of treatment combinations: Non-BKD female*
B K D (1, 2 or 3) female. The vertical bars
represent +/- one Standard Err o r .................... 77

xxv

Figure

30: Means of lipid content in liver of

male and female non-Bacterial Kidney Diseased
(NBKD) and Bacterial Kidney Diseased stage 1
(BKD1), stage 2 (BKD2) and stage 3 (BKD3)
chinook salmon caught in Spring in Lake Michigan
(1990-1992).....
Figure

.78

31: Pairwise comparisons of means of water

content in liver of non-Bacterial Kidney
Diseased (Non-BKD) and Bacterial Kidney
Diseased (BKD) stagel
and stage3

(BKD 1), stage2

(BKD 2)

(BKD3) male chinook salmon caught

in Spring in Lake Michigan (1990-1992). The
fish were clustered in pairs (clusterl-cluster6)
oftreatment combinations: Non-BKD male*BKD(l,
2 or 3) male. The vertical bars represent
+/- one Standard Error.
Figure

.....

79

32: Pairwise comparisons of means of water

content in liver of non-Bacterial Kidney
Diseased (Non-BKD) and Bacterial Kidney
Diseased (BKD) stagel
and stage3

(BKD3)

(BKD 1), stage2

(BKD 2)

female chinook salmon caught

in Spring in Lake Michigan (1990-1992). The
fish were clustered in pairs (clusterl-cluster6)
of treatment combinations: Non-BKD female*BKD(l,
2 or 3) female. The vertical bars represent
+/- one Standard Error.

............................. 80
xxv i

Figure

33: Means of water content in liver of

male and female non-Bacterial Kidney Diseased
(NBKD) and Bacterial Kidney Diseased stage 1
(BKD1), stage 2 (BKD2) and stage 3 (BKD3)
chinook salmon caught in Spring in Lake
Michigan (1990-1992)..........................

xxvii

LIST OF APPENDICES

Page
Appendix A-l: Means and Principal Components (PCI,
PC2 and PC3) scores of non-lipid energy (LE),
lipid (LF) and water (LW) contents in liver
of non-Bacterial Kidney Diseased (0) and Bacterial
Kidney Diseased (2) male (1) and female

(2)

chinook salmon caught in Fall in Lake Michigan
(1990-1992).......................................
Appendix A - 2 : Means and Principal Components

(PCI,

PC2 and PC3) scores of non-lipid energy (LE),
lipid (LF) and water (LW) contents in liver
of non-Bacterial Kidney Diseased (0) and Bacterial
Kidney Diseased stage 1 (1), stage 2 (2) and
stage 3 (3) male (1) and female (2) chinook
salmon caught in Spring in Lake Michigan (19901992).............................

xxviii

.89

Appendix B: Graphical demonstration of the apparent
interaction between the factors health and season
(A), for lipid content, and health and sex (B),
for water content in liver of non-Bacterial Kidney
Diseased (Non-BKD) and Bacterial Kidney Diseased
(BKD) chinook salmon caught in Fall and Spring in
Lake Michigan (1990-1992)

.... ..................... 101

xxix

CHAPTER I.

INTRODUCTION AND LITERATURE REVIEW ON BACTERIAL

KIDNEY DISEASE

1.1.

Statement of the problem

Chinook salmon (Oncorhvnchus tshawvtcha) were
introduced in Lake Michigan in 1967
1973, Mercer 1980).

(Scott and Crossman

The objectives of this introduction by

the Michigan Department of Natural Resources

(MDNR) was to

control the overabundant alewife (Alosa pseudoharenqus)
population and to create a sport fishery (Keller et al
1989).

The success of this stocking was so spectacular that

the other states bordering Michigan soon joined the program
(Mercer 1980).

The introduction of the salmonids into the

Great Lakes created a boom in sport fishing activity with
the Chinook salmon being the most important sport-fish
salmonid caught in Lake Michigan (Rakoczy 1988).

Recently,

this species has experienced a dramatic decline in growth
and abundance.

This decline has been concurrent with a

decline in abundance of alewives and an increase in the
severity of Bacterial Kidney Diseased (BKD).

The decline in

chinook salmon abundance in Lake Michigan has resulted in a
decline in the sport fishing industry.
Renibacterium salmoninarum. the causative agent of BKD,

existed as a mild pathogen in Lake Michigan until 1987 when
dead and dying fish plagued with the disease were observed
by MDNR fisheries biologists during the Spring (Myrl Keller,
MDNR Charlevoix, Michigan, personal communication).

The

MDNR estimated the impact of the BKD infection to be a 50 %
decrease in the chinook salmon population of Lake Michigan
since 1980.
In an attempt to contribute to a better understanding
of the impact of BKD on Lake Mich i g a n 's chinook salmon
population it was the purpose of this study to address
whether there were significant changes in energy stores
(non-lipid energy,

lipid)

and water content in Age 1+ and

older chinook salmon infected with BKD as compared to fish
without BKD during the Fall and Spring.

1.2.

Need for the study

The occurrence of disease vectors and their impact on
the population are factors which should be accounted for in
an integrated salmon fisheries management program.

Stress

from such vectors is often apparent in changes in liver non­
lipid (proteins and carbohydrates), lipid and water
contents.

Lipids primarily store energy that will be used

during stress conditions

(Black 1958, Jezierska et al 1982)

while proteins are used for growth (somatic and gonadal),
feeding and spawning activities

(Love 1970, lies 1984,

3
Dygert 1990).
It is surprising that the energetic cost of
Renibacterium salmoninaruxn infection has not been well
investigated in chinook salmon, especially since this
disease is specific to salmonids

(Fryer and Sanders 1981,

Post 1983, Pascho et al 1987) and causes most of the disease
problems in the salmon fisheries.

In Lake Michigan, Rakoczy

(1988), Hay (1989) and Stewart and Ibarra

(1991) noted that

the chinook salmon harvest declined 32 % in number; the size
of individual sport caught fish also decreased.

The

increase in mortality and decrease in growth are presumed to
be related to Bacterial Kidney Disease (John Hnath, Fish
Health Laboratory, Wolf Lake, Michigan, personal
communication).

1.3.

Choice of the liver

Preliminary analysis of the energy and water contents
of the liver, the filet (trunk muscle) and the pyloric
caecum of chinook salmon indicated that the condition of the
liver was representative of the health of the fish.

The

choice of the 1iver as an adequate organ to study the impact
of BKD infection is further supported by other researchers
(Idler and Bitners 1960, Tarver 1951, 1963, Pesch 1963,
Jungermann and Katz 1986) who reported that the liver was
the best indicator of proteins, carbohydrates and lipids

4
changes during stress condition in animals.

1.4.

Goal of Study

The goal of this research is to describe the changes in
energy reserves (non-lipid and lipid) and water content in
the liver of Chinook salmon as a result of the incidence of
Renibacterium salmoninarum.

1.5.

Statistical analysis.

Data were analyzed using Principal Component Analysis
(PCA), Factorial design and Fisher's Least Significant
Difference

(LSD). The use of PCA was to demonstrate the

multidimensionality of the general condition of the fish and
to separate fish into two groups, non-Bacterial Kidney
Diseased (non- BKD)
Disease (BKD)
1991).

fish and fish with Bacterial Kidney

(Burkhead and Williams 1991, Valcarce et al.

The PCA transforms the observed variables into a set

of new, uncorrelated variables called Principal Components
(Morrison 1990).

Each Principal Component

linear combination of the
such as Yj=a1j-X1 + .....

(Yj) is the

observed variables

+ apj-Xp) .

(X1 ...... Xp)

In PCA the investigator

has different planes on which to view the data and can
define the planes on the directions of variance within the

data itself (Neff and Smith 1979).
The LSD was used to make all pairwise comparisons
(Petersen 1985) of the mean values of total non-lipid
energy, lipid and water contents from different treatment
combinations of season, sex and health.

The mean values

were computed using the General Linear Model

(GLM) in SAS

(SAS Institute Inc. 1988) prior to using the LSD analyze the
effect of the 3 main factors and their interactions on the
measured variables.

1.6.

Literature review on Bacterial Kidney Disease

Geographical distribution
Historically Bacterial Kidney Disease (BKD, under the
name of Dee Disease) was first described in 1930 from
Atlantic salmon (Salmon solar) in Aberdeenshire Dee and the
river Sprey in Scotland (Smith 1964).

The disease was

reported in the United States of America (prior to 1955),
Japan (in 1973), and France (in 1974).

By 1956, in the USA,

the disease was only noted in limited areas of the states of
Michigan, Minnesota, Wisconsin, Iowa, South Dakota, Nebraska
and Wyoming (Post 1983).

Susceptible species
Bacterial Kidney Disease, caused by Renibacterium

salmoninarum. is prevalent among salmonid fishes
al 1974, Sakai et al 1989).

(Bullock et

It is a chronic systemic

infection with a protracted cause and an insidious nature
(Warren 1983, Pascho et al 1987).

The disease is considered

to be among the most important disease problems in salmon
population today (Plumb and Bowser 1983, Turaga et al 1987).

Etiological agent
Taxonomically, the etiological agent of BKD, R.
salmoninarum. belongs to the order Actinomycete, the family
Corynebacteriaceae, the genus Renibacterium and the species
salmoninarum (Post 1983) . The bacteria was formerly named
Corvnebacterium salmoninarum in 1978. It is a small (0.4 n
by 0.8 /u) bacteria, strongly Gram-positive, non-acid-fast,
non-motile, non sporulating and rod-shaped occurring usually
in pairs (diplobacilli)

(Fryer and Sanders 1981). It reaches

a maximum growth at 15°c and will grow at 5°c but not above
25°c (Post 1983).

Epizootiologv
The epizootiology of R. salmoninarum is not well known.
The principal routes of transmission noted by Fryer and
Sanders (1981) are through infected or carrier fish
(horizontal transmission)

and the eggs

(vertical

transmission).

Water temperature has a direct impact on the

epizootiology of the pathogen.
and Warren

Belding and Merrill

(1935)

(1983) indicated the seasonal occurrence of BKD

and the relationship of changing water temperature to
increased mortality: most epizootics occur during Winter and
early Spring time under conditions of declining and cold
water temperatures.

In addition to the influence of the

water temperature, Post (1983) reported that BKD is more
prevalent in soft water conditions than in hard water.

Pathology
Externally the clinical signs of BKD are exophthalmia,
abdominal distension, hemorrhagic areas, deep abscesses on
various
parts of the body surface and darkening of the skin (Fryer
and Sanders 1981, Warren 1983).

Internally R. salmoninarum

produces gray-white necrotic nodules primarily on the kidney
and sometimes on the liver and massive cheesy material (dead
cells entangled in haemopoietic tissue)
liver (Kinkelin 1974).

in the spleen or

The kidney usually appears swollen,

convex and has a lumpy surface.

A bloody, or turbid fluid

often accumulates in the abdominal cavity. Fryer and Sanders
(1981) and Hsu et al

(1991) reported that the bacteria can

exist as an intracellular and extracellular pathogen.

8
Detection and identification of R. salmoninarum
The diagnosis of Bacterial Kidney Disease must be
confirmed by the detection and identification of the
etiological agent because others bacterial diseases such as
Aeromonas Furunculosis, and Visceral Granuloma may occur
simultaneously with somewhat similar symptoms
Axelrod 1971).

(Snieszko and

Several known techniques for the detection

and identification of R. salmoninarum involve the
Immunodiffusion procedure (Chen et al 1974), Fluorescent
Antibody Technique (FAT)

(Bullock and Stuckey 1975, Laidler

1980), Coagglutination test (Kimura and Yoshimizu 1981),
Counterimmunoelectrophoresis (CIE)

(Cipriano et al 1985),

Enzyme-Linked Immunosorbent Assay (ELISA)

(Pascho and

Mulcahy 1987, Hsu et al 1991) and Dot Blot Assay
(Sakai et al 1987).

(DBA)

The Fluorescent Antibody Technique is

known to be the most rapid and most sensitive technique.
There have been many attempt to diagnose R. salmoninaram
through culture media but, because of the fastidious nature
of the bacteria to grow in culture, such media must be
highly fortified with cystine or cysteine.

Meuller-Hinton

medium with 0.1 % cysteine-Hcl is one of the known culture
media for the bacteria.

The isolation in culture media will

take 5 to 18 days before round, very convex and white
(usually)

colonies with complete margins can be observed

(Post 1983).

9

Disease control
Bacterial Kidney Disease is the most difficult of the
bacterial fish pathogens to control.

At present no drugs

are known that will cure the disease; there is no effective
vaccination treatment and the control through chemotherapy
gives temporary results (Eliott et al 1989).

The fact that

the bacteria is mainly intracellular and not sensitive to
the usual antibiotics or chemotherapeutic drugs make the
control very difficult.

On a temporary bas i s , Erythromycin

seems to be the most effective chemotherapy when used in
preventive or curative treatment.

Other drugs as

Sulfaguanidine, Sulfamethazine, Sulfisoxazole and
Sulfamerazine have also been used but generate inconsistent
results and risks of the development of drugs resistant
bacteria.

Approaches to control the BKD have focused on the

effect of diet on resistance of fish to the lethality of BKD
and

the creation of a strain of fish more resistant to the

disease

(Withler and Evelyn 1990).

10
CHAPTER II.

2.1.

MATERIALS AND METHODS

Collection and diagnosis of fish

Collection of fish
This study was conducted in the Michigan waters of Lake
Michigan near the cities of Ludington and Pentwater (Figure
1).

I collected a total of 346 immature salmon ranging

from age

1+ to age 3+ in Fall and Spring at these two

locations.

These fish ranged in size from 300 mm to 910 mm

in total length.
nets.

Chinook salmon

were caught using gill

These nets were 3000 to 4000 feet in length and 24

feet in depth.

The sampling period was chosen to be

representative of fish condition during periods of low and
high level of virulence of Bacterial Kidney Disease.
Bacterial Kidney Disease incidence is low in the Fall and
high in the Spring (Snieszko and Azelrod 1971, Warren 1983).

The fish caught were immediately put in a container
with ice

and taken to the laboratory for diagnosis of

which occurred within the next 48 hours.

BKD

Ludingi on

4^ 00’

Pentwater

43 30’

Illinois

Indiana
4130’

scale:

crr=i

50

Z3

Kilometers

Figure 1: Map of Lake Michigan with sampling sites (1, 2).

12
Diagnosis of fish. Preliminary separation of fish.
The identification of the diseased fish was based on
the presence of gray-white necrotic abscesses in the kidney
and cheesy material on the spleen (Kinkelin 1974, Bullock et
al 1975, Fryer and Sanders 1981).

The fish could be easily

grouped into BKD and non-BKD fish based on the gross
lesions.

The BKD fish in the Spring were sorted into 3

subgroups

(BKD1, B K D 2 , BKD3) based on an increasing presence

of necrotic area in the kidney. BKD1 (first stage)
less than 3 nodules and BKD2

(second stage)

more nodules and a substantial
in their body cavity.

The BKD3

fish had

fish had 3 or

(23 to 30 ml) amount of fluid
(last stage of the disease)

fish had large necrotic area on their swollen kidney, cheesy
material on their spleen and a high (101 to 101.2 ml) amount
of body fluid.
macroparisites
discarded.

During this diagnosis 84 fish infected by
(mainly Echnorhvnchus salmonis) were

The macroparasitic infection by the helminth

Echinorhynchus salmonis and other species was also noted by
Muzzal

(1989) on Lake Michigan chinook salmon.

Identification and quantification of R. salmoninarum. Second
separation of fish.
A portion of the posterior kidney was homogenized in a
mortar and then used for the Gram stain and the Fluorescent
Antibody Technique

(FAT) procedures for identification of

Renibacterium (Bullock et al 1975; John Hnath, Fish Health
Laboratory, Wolf Lake, Michigan, personal communication).
smear of 10
Stain.

11

A

of the homogenized kidney was used in the Gram

The observation of stained slides under oil

immersion revealed short blue diplobacillus in sick fish
while these bacteria were rare or non-existent on slides
from the non-BKD fish.

After the Gram Stain the rest of the

homogenized kidney was used to run the FAT procedures.
A volume of 1/10 ml of the homogenized kidney was diluted to
10"1 in 9/10 ml of 0.01 M PBS

(Phosphate-Buffered saline.) .

One microliter of the dilution was put in each of the 12
wells of a microslide supercured.

The wells were allowed to

dry then fixed in absolute methanol for 10 mn, covered with
a prepared Goat antiserum-Rhodamine working solution and
incubated for 1 hour in dark chamber.

The Goat anti-R.

salmoninarum immunoglobulin conjugated with fluorescein
isothiocyanate was diluted 1:40 (v/v) in 0.01 M PBS and a
Rhodamine counterstain was added.

The prepared wells were

covered with mounting solution before observation under
fluorescent blue light with oil immersion. Renibacterium
salmoninarum appeared as bright green short diplobacilli.

I

counted the number of R. salmoninarum in each optic field.
A microgrid subdivided the field into 4 subsampling areas.
I sampled 10 fields in each w e l l : 5 fields across
horizontally and 5 vertically. The total number of bacteria
per field was converted to number of cells per milliliter of

14
the original sample as per the methods of Ecker and Lockart
(1959) :

/ convezsion\J dilution\^ .
.
^
^
I* (number of cells counted)
cel Is/ml- \ factor
/ I factor ) _________________________
number of fields counted

The diameter of the well was 5 mm and the optic field
was 0.175 mm in diameter measured at a 100X magnification.
The conversion factor was calculated to be 815.06

(ie 19.643

mm2/ 0 .0241 mm2).

2.2.

Evaluation of water,

lipid, and non-lipid energy

contents.

Evaluation of liver water content
A sample of fresh liver of each fish was taken within
42 hours after catch, weighed and dried in a forced hot air
oven at 64°c for 72 hours.

The dry weight was then measured

and the percentage of water in the wet weight calculated:

txratcr-

wei-9 ht - dry weight)* 100
wet weight

15
Evaluation of liver lipid content
The dried liver was homogenized in an electric grinder
and 3 replicates (wl) of 0.5 to 0.75 g were taken for lipid
extraction using a Tecator Soxtec System HT 1043 unit.

The

method uses diethyl ether as solvent; the lipid is extracted
into cups weighed with boiling chips (w2) and filled with 25
ml of the solvent.

After extraction the cups are dried in a

forced hot air oven at 100°c

for 30 mn and weighed

(w3) .

The percentage of lipid was calculated as:

%

i i p i d -

~ w2± * 100

wl

Evaluation of liver non-lipid energy content
I used the remains of the subsamples after lipid
extraction to estimate the non-lipid energy content of the
liver.

The caloric content of this material was analyzed

using an adiabatic oxygen bomb calorimeter 1241 obtained
from Parr Instrument company, Moline, Illinois. Three
replicates of 0.20 to 0.75 g

were used to estimate the

energy content for each fish liver.

2.3 Statistical analysis

16
Use of multivariate methods (Principal Component Analysis)
in the separation of fish. Final separation.
The Principal Component Analysis (PCA) was performed on
the data of 262 fish to further separate the two groups of
fish based on the

distribution of the scores of each

individual on the

first two Principal Components (PCI and

PC2).

The approximate test was done on the correlation

matrix for no equality with the identity matrix.
analysis used the

correlation

The

matrix because of the

heterogeneity of the variables (Gnanadesikan,

1977).

The

data matrix was generated in SAS system.
The scores of 262 fish (128 non-BKD and 21 BKD infected
in Fall and 92 non-BKD and 21 BKD infected in Spring) were
used to make the plots of the first two principal
components.

The plots permitted a visual assessment of the

position of individual data point and also the variation
within and between groups.

The dispersion of the groups

scores in an Euclidian plane is a function of both variance
of the individual characters

(PCI and PC 2 ) and the degree of

correlation between them (Neff, 1979).

The biplots of the

values of the first two eigenvectors linked the variation
between groups to the spatial dynamic changes of the
observed variables.
The final separation was obtained on the scatter
diagrams by considering the sign of the PCI scores.

The

fish (94 individuals) that were diagnosed "non-BKD" but had

17
a negative PCI were reconsidered as sick fish without R.
salmoninarum infection and discarded for the current
analysis: these fish were experiencing stress of diverse
nature (Aeromonas and Vibrio infection, physical
abnormalities, et c ) .

The diagnosis of the Aeromonas

hvdrophila and Vibrio sp infection was confirmed by John
Hnath (Fish health laboratory, Wolf lake, Michigan)

and

Robert Walker (Animal health diagnostic laboratory, Michigan
State University).
necropsy,

When all separation procedures

(general

identification and quantification of the bacteria,

principal component) were performed, a total of 168 fish
(Appendix A) were highly selected for the factorial analysis
and the multiple comparisons of the m e a n s .

In all the

separation procedures a total of 178 fish (84 before PCA and
94 after PCA) out of the initial sample (346 fish) were
discarded.

Factorial analysis of changes in non-lipid energy,

lipid and

water contents.
The General Linear Model

(GLM)

was applied to test the

main effect and interaction of the 3 factors: sex, health
and season on

the non-lipid energy, lipid and water

contents in the liver.

The F test on the main effect and

interaction was executed in 3 different situations: one
factor, two factors and three factors analysis.

The

18
comparisons between means of non-lipid energy,

lipid and

water content were made possible by a pairwise comparison
using Fisher's Least Significant Difference (FPLSD).
Differences in mean values in the Spring were graphically
displayed in a 3 dimensional plot to visualize the direction
of changes in non-lipid energy, lipid and water content as a
consequence of the effect of BKD.

19
CHAPTER III. RESULTS

3.1. Separation of fish

Diagnosis and preliminary separation of fish
I diagnosed BKD in 23 (13.3%) of the 173 Chinook salmon
captured in the Fall and 29

(16.8%) of 173 fish caught

during the Spring sampling season.

Plate 1 depicts the

necrotic areas related to the severity of the disease in
three identifiable stages

(from low to advanced stage).

These 3 stages are related to the chronic nature of the
disease in these salmon.

A fish infected by R^_ salmoninarum

will undergoes first a stage of infection (BKD1), then a
stage of invasion of the kidney where optimum growth of the
bacteria occurs

(BKD2) and finally a mortem stage (BKD3).

Identification and quantification of Renibacterium
salmoninarum.

Although the gross lesions were typical

of Bacterial Kidney Disease,

I used a second method to

identify and quantify the bacteria.
identification of

The first

salmoninarum employed the Gram Stain

which determined the presence or absence of blue short
diplobacilli readily sorted the fish into non-BKD and BKD

20a

Plate

1: Gross lesions produced by Renibacterium

salmoninarum in Chinook salmon. Three stages of the disease
with corresponding severity of necrotic areas or nodules
(indicated by the arrows) are displayed: Bacterial Kidney
Disease stage 1 (A), Bacterial Kidney Disease stage 2 (B)
and Bacterial Kidney Disease stage 3 (C). The small arrow in
stage 3 (C) indicates the spleen covered with cheesy
materials.

20

21
fish (Plate 2).

The results from the Gram Stain were

confirmed by a second and more precise identification of R.
salmoninarum using the Fluorescent Antibody Techniques
(FAT).

This bacteria appeared as bright green short rods of

0.735 ii in length.

Table 1 depicts the number of bacteria

for each disease stage identified.

The three bacterial

concentrations differed significantly (a= 5 %) with BKD2
fish having the highest concentration of cells/ml
2).

(Figure

The different concentrations reflected the level of

development of the disease and matched with the
classification based on the gross lesions.

Only one BKD2

fish was reclassified as BKD3 after the FAT procedures.
Thus an experienced technician can readily and reliably sort
fish exposed to BKD into 4 categories

(non-BKD, BKD1, BKD2

and BKD3) based on their gross lesions and the Gram Stain
results.

Analysis of liver non-lipid energy,

lipid and water

contents. Final separation of fish.
I analyzed the differences in liver contents of BKD and
non-BKD fish during the fall and spring sampling periods
using Principal Component Analysis

(Appendix A ) .

The

Principal Component Analysis (PCA) was run using the
correlation matrices for the variables of non-lipid energy,
lipid and water contents in the liver during Fall and Spring

22a

Plate

2: Cells of Renibacterium salmoninarum stained with

crystal violet. Three stages of the disease with their
relative density of cells are displayed: Bacterial Kidney
Disease stage 1 (A), Bacterial Kidney Disease stage 2 (B)
and Bacterial Kidney Disease stage 3 (C). The arrows
indicate the cells (diplobacilli) of R. salmoninarum.

23

Table

1: Number of Renibacterium salmoninarum counted under

fluorescent light and body fluid content for three different
stages of Bacterial Kidney Diseased (BKD) fish: stage 1
(BKD1), stage 2 (BKD2) and stage 3 (BKD3).

Mean number

Number of

Standard

Body

of cells/ml

fish examined

error

fluid

(ml)

BKD1

2387

5

699

13.5-19.4

BKD2

31277

11

15975

23-30

BKD 3

15530

5

1132

101-101.2

24

Number of cells/ml
35,000

30,000

25,000

20,000

15,000

10,000

5,000

BKD 1

BKD 2

BKD 3

BKD stage
Figure 2: Direct optic vision count of cells of Renibacterlum salmoninarum
using the Fluorescent Antibody Technique (FAT). Three d isease stag es ranging
from low to high virulence are identified: Bacterial Kidney D isease stage 1
(BKD1), stag e 2 (BKD2) and stag e 3 (BKD3). The vertical bars represent +/- one
Standard Error (SE).

(Table 2).

Table 3 and Table 4 show that the first two

principal components

(PCI and PC2) explained

over 80 % of the variability of the data in both Fall and
Spring.

As such, I was able to accurately separate non-BKD

and BKD chinook salmon based on liver contents.

Such

differences were identified in a multivariate space by the
discrete grouping of non-BKD and BKD fish for the Fall
(Figure 3-A,B) and the Spring (Fig 4-A,B).

Our initial

separation of fish based on PCA indicated that a small
percentage of non-BKD fish partially overlapped with BKD
fish for both Fall (Figure 3-A) and Spring (Figure 4-A).
This

overlap indicated that these fish which I had

previously diagnosed as non-BKD individuals were in fact
experiencing other stresses of various nature.

When I

reconsidered the diagnosis of these fish, I determined that
they suffered from a variety of bacterial infections
including Aeromonas s p . and Vibrio sp. or macroparasites
infection and physical abnormality.

Given this information

I removed these fish from the analysis and re-analyzed my
data.

The final separation is plotted in Figure 3-B for

Fall and Figure 4-B for Spring and shows that non-BKD and
BKD fish are completely separated.

BKD salmon displayed

negative PCI values while non-BKD salmon PCI values were
positive.

The biplot of the first two eigenvectors scores

(Table 5) for the Fall data (n= 84)
Spring data

(n=84)

(Figure 5) and the

(Figure 6) demonstrated that non-BKD fish

26

Table

2: Correlation matrices for non-lipid energy (LE),

lipid (LF) and water

(LW) contents in liver of Chinook

salmon caught in Fall and Spring in Lake Michigan

(1990-

1992).

Fall

Spring
1
1

LE

LF

LW

|

LE

LF

LW

1
1
LE
LF
LW

1

0.063
1

-0.106

|

-0.455

|

1

1

1

-0.065
1

-0.36
-0.28
1

27

Table

3: Principal Component Analysis

of non-lipid energy,

lipid and water contents in liver of non-Bacterial Kidney
Diseased (non-BKD) and Bacterial Kidney Diseased
chinook salmon caught in Fall in Lake Michigan

(BKD)

(1990-1992).

Principal Components

1

Eigenvalues

2

3

1.485

0.982

0.542

0.503

0.440

-

0.495

0.324

0.181

Difference between
consecutive eigenvalues

Proportion

of the total

variation indicated by
each Principal Component

Cumulative proportion
indicated by the
Principal Components

0.495

0.819

1

28

Table

4: Principal Component Analysis

of non-lipid energy,

lipid and water contents in liver of non-Bacterial Kidney
Diseased (non-BKD) and Bacterial Kidney Diseased

(BKD)

chinook salmon caught in Spring in Lake Michigan (19901992) .

Principal Components

Eigenvalues

1

2

1.430

1.062

0.368

0.555

0.477

0.354

0.477

0.831

3

0.507

Difference between
consecutive eigenvalues

Proportion of the total
variation indicated by
each Principal Component

0.169

Cumulative proportion
indicated by the
Principal Components

1

29
6

s

A

4
3
2

2 0®

1

0

0
2Z

-1

£

•2

i

■3
-4

&
a

8

I

2 27

J
0) 0

#»°

.

0
0

0

-5
-6
t

-7

i 2 l

l

.... I

I

1

1

6

1

5
4

Vi

3

B

2

1
0

( 2^2
2

■1 -

Z
ly

$2

-2

flw

!A X f o ,
W ° »
0

■3
-4

®

0 8
0

• °n°
4 **

,
«
o

0

-5
-6

■7

..

. i

2

.i_______

i

-2

i

i

0

i

i

i

2

First Principal component (PCI)

Figure 3: Principal Component plots for non-lipid energy (LE), lipid (LF) and water (LW)
contents in liver of non-Bacterial Kidney Diseased (0) and Bacterial Kidney
Diseased
(2) chinook salmon caught in Fall in Lake Michigan (1990-1992). Plot B gives a
com­
plete separation of the fish after (0) fish, overlapping with (2) fish in plot A
, were
discarded (see text for more detail).

30

A

CM

if
0

£

c
CD
c
o
Q.
E
o
O
(0
Q.
a

c

-tvr
2o
0
0

-1

0 0
f)

0

000 OoU
00
0------

)Q
00
0

00

%
co

4

T3
C

o

3

8
a>
2

1

tw

0
•1
-2

-4

■2

0

2

4

First Principal C o m p o n en t (PC1)

Figure 4: Principal Components plots for non-lipid energy (LE), lipid (LF) and
water (LW) contents in liver of non-Bacterlal Kidney Diseased (0) and
Bacterial
Kidney Diseased stagel (1), stage2 (2) and stages (3) Chinook salmon caught in
Spring In Lake Michigan (1990-1992). Plot B gives a complete separation of the
fish after (0) fish, overlapping with (1), (2) and (3) fish in plot A, were
discarded
(see text for more detail).

31

Table

5: Eigenvectors of the measured non-lipid energy,

lipid and water contents in liver of non-Bacterial Kidney
Diseased (non-BKD) and Bacterial Kidney Diseased (BKD)
chinook salmon caught in Fall and Spring in Lake Michigan
(1990-1992) .

Eigenvectors

Variables

Fall

Spring

1

2

3

non-lipid energy

0.239

0.968

0.069

lipid

0.681

-0.218

0.699

water

-0.692

0.121

0.712

non-lipid energy

0.561

-0.603

0.567

lipid

0. 390

0.796

0.462

water

-0.730

-0.038

0.682

32

0.9
0.8
0.7
0.6
0.5
0.4
03

-

-

02
0.1 - LW
-

0.1
0.2
03

LF

0.5
0.6
0.7
0.8
0.9
-0.7

-0.5

-03

-

0.1

0.1

03

0.7

first eigenvector

Figure 5: Eigenvectors plot for non-lipid energy (LE), lipid (LF) and water (LW) contents
in liver of non-Bacterial Kidney Diseased and Bacterial Kidney Diseased

chinook

salmon caught in Fall in Lake Michigan (1990-1992). The lines indicate the

direction

of change of the variables LE, LF and LW.

33

0.9

0.8
0.7

0.6
0.5
0.4
0.3

0.2
0.1
LW

0.1
0.2
-0.3
-0.4
-

-

-0.5
-

0.6

-0.7
-

0.8

-

0.6

-0.4

-0.2

C

0.2

0.4

0.6

First eigenvector

Figure 6: Eigenvectors plot for non-iipid energy (LE), lipid (LF) and water (LW)
contents in liver of non-Bacterial Kidney Diseased and Bacterial Kidney Diseased
chinook salmon caught in Spring in Lake Michigan (1990-1992). The lines
indicate the direction of change of the variables LE, LF and LW.

34
possessed more non-lipid energy and lipid content than BKD
fish while livers of BKD fish had greater water contents.
Further comparisons of the Fall and Spring biplots of these
eigenvectors and their scores (Table 5) demonstrate that
salmon use lipid content during the winter and as a
consequence increase their water

content.

3.2. Factorial design

2- factorial design
I evaluated the relationships between fish health (BKD
versus non-BKD fish) and sex within and between season (fall
and spring)

using the General Linear Model

(Table 6).

The

linear model produced was highly significant for the three
variables.

Health and season were significantly related to

changes in liver contents while sex was not.

The apparent

significant interaction between health and season for lipid
and health and sex for water content does not impair the
interpretation of the main effect based on subsequent
graphical analysis

(Appendix B ) .

Thus the interpretation of

all main effects was possible regardless of the interaction
between the factors.
The results of the multiple comparisons of means using
the Fisher's Least Significant Difference were congruent

Table

6: ANOVA of 23 factorial design of non-lipid energy (LE), lipid (LF) and water (LW)

contents in liver of non-Bacterial Kidney Diseased (non-BKD) and Bacterial Kidney Diseased
(BKD) chinook salmon caught in Fall and Spring in Lake Michigan (1990-1992).

Liver non-lipid

Liver lipid

energy(LE)

Source

df

MS

Liver water

(LF)

F

MS

(LW)

F

MS

F

Model

7

414983.47

13.66 *

1689.13

33.93 *

135.99

24.64 *

Health

1

827933.06

27.25 *

2298.33

46.16 *

785.17

142.29 *

Sex

1

19968.98

0. 66

9.46

0.19

10.33

1.87

Season

1

1934755.25

63.69 *

8321.47

167.14 *

90.88

16.47 *

Health*Sex

1

31947.56

1.05

62.82

1.26

41.86

7.59 **

Sex*Season

1

1386.53

0.05

76.81

1.54

2.12

0.38

Health*Season 1

4347.60

0.14

973.76

19.56 **

16.64

3.02

Health*Sex*

Table 6 (continued)

Season
Error

1

84545.27

160

30377.80

2.78

81.23
49.78

1.63

4.95

0.90

5.52

*= significant at a= 5 %
**= Graphically demonstrated as biologically non significant. See text and Appendix-B for
further detail.
U)
On

37
with the factorial analysis in that season and health
independently had significant effect on changes of non-lipid
energy, lipid and water contents in the liver.

In contrast,

the factor sex did not yield significant effect on changes
of the liver contents.

In order to fully assess the impact

of health and season on BKD and non-BKD fish I evaluated my
data using a one, two and three factor analysis.

One factor analysis.
When one factor treatment combination was considered
(Table 7) LSD indicated the liver of non-BKD fish contains
significantly more (3.26%) non-lipid energy (Figure 7), more
(94.36%)

lipid (Figure 8) and less (6.31%) water (Figure 9)

than BKD fish.
energy,

No significant differences in non-lipid

lipid and water content between male and female fish

were noted.

When the effect of the season was taken in

account, chinook salmon liver in the Fall contained
significantly less non-lipid energy (4.14 %), more lipid
(164.19 %) and less water (2 %) than fish in the Spring.

Two factors analysis.
In the two factors combination of treatments (Table 8
and Table 9) the LSD revealed that non-BKD males and females
contained more non-lipid energy (Figure 10-A), more lipid

Table

7: One factor analyses and Pairwise comparisons of means of non-lipid energy, lipid

and water contents in liver of chinook salmon using Fisher's Least Significant Difference
(LSD).

The fish were caught in Fall and Spring in Lake Michigan (1990-1992).

Treatment combinations

Non-BKD

Means of non-lipid

energy

(LSD(0.05)= 52.71)

BKD

Male

Female

5142.64

4980.52

5095.55

5110.65

(16.84)

(37.88)

(24.64)

(21.06)

Fall

Spring

4994.24 5209.97
(26.47)

(11.53)

Means of lipid content

17.59

9.05

15.41

15.52

22.43

8.49

(LSD(0.05)= 2.13)

(1.01)

(0.93)

(0.96)

(1.48)

(1.21)

(0.44)

Table

7 (continued)

Means of water content

74.08

79.07

75.02

75.73

74.57

76.09

(LSD(0.05)= 0.71)

(0.21)

(0.45)

(0.37)

(0.33)

(0.43)

(0.25)

Mean values within the same pair of treatment combinations and joined by a dotted line do
not differ significantly.

Elements of the pair of treatment combinations to be compared

are joined with a solid horizontal line.
Values inside the parenthesis below each mean are Standard Errors.

Non-lipid energy (cal/g)
6,000
Non-BKD

5,000

Spring

Fall

BKD

4,000

3,000

2,000

1,000

Health

Sex

Season

Figure 7: Pairwise comparisons of means of non-lipid energy content in liver of non-Bacterial Kidney
Diseased (Non-BKD) and Bacterial Kidney Diseased (BKD) male and female chinook salmon caught
in Fall and Spring in Lake Michigan (1990-1992). The fish were grouped into different treatment combi­
nations corresponding to each level of the factors health, sex and season. The vertical bars represent
+/- one Standard Error.

Lipid content (%)
100

80

60

Fail
Female

Male
BKjp

Health

Sex

Season

Figure 8: Pairwise comparisons of means of lipid content in liver of non-Bacterial Kidney Diseased
(Non-BKD) and Bacterial Kidney Diseased (BKD) male and female Chinook salmon caught in Fall
and Spring in Lake Michigan (1990-1992). The fish were grouped Into different treatment combi­
nations corresponding to each level of the factors health, sex and season.
represent +/- one Standard Error.

The vertical bars

Water content (%)
100

BKD

80
Non-BKD

Female

Male

Fall

Spring

60

40

20

0

Sex

Season

Figure 9: Pairwise com parisons of m eans of water content in liver of non-Bacterial Kidney Diseased
(Non-BKD) and Bacterial Kidney Diseased (BKD) male and female Chinook salmon caught in Fall and
Spring in Lake Michigan (1990-1992). The fish were grouped into different treatment combinations
corresponding to each level of the factors health, sex and season.
The vertical bars represent
+/- one Standard Error.

43

Table

8: Means and Standard Errors (SE) of non-lipid energy

(LE), lipid (LF) and water (LW) contents in liver of Chinook
salmon caught in Fall and Spring in Lake Michigan (19901992). Two factor treatment combinations were considered.
See Table 9 for Pairwise comparisons of m e a n s .

liver non-lipid
energy (LE)

liver lipid

liver water

in cal/g

(LF) in %

(LW) in %

Treatment

mean

SE

mean

SE

mean

SE

Non-BKD male

5136.82

22 .81

17.31

1.07

73.60

0.28

Non-BKD female

5150.65

25.03

17.98

1.90

74.75

0.29

BKD male

4958.60

68.27

9.09

1.51

79.73

0.62

BKD female

5004.62

27.87

9.01

1.09

78.35

0.64

Male in Fall

5003.15

38.93

21.46

1.17

74.35

0.59

Male in Spring

5207.29

15.33

8.08

0.46

75.84

0.38

Female in Fall

4979.77

29.34

24.00

2.56

74.94

0.62

Female in Spring 5212.79

17.48

8.91

0.77

76.35

0.31

Non-BKD Fall

5032.26

25.86

25.88

1. 26

73 .13

0. 35

Non-BKD Spring

5253.02

9.03

9.31

0.53

75.03

0.16

BKD Fall

4880.20

67.38

12.07

1.54

78.90

0.86

BKD Spring

5080.83

18.30

6.03

0.53

79.25

0.32

44

Table

9: Two factor analyses and Pairwise comparisons of

means of non-lipid energy (LE), lipid (LF) and water (LW)
contents in liver of Chinook salmon using Fisher's Least
Significant Difference (LSD).

The fish were caught in FA11

and Spring in Lake Michigan (1990-1992).

Significant
difference
ixsauXuenu coituDina uions
tested

LE

LF

Non-BKD male— ---BKD male

*

*

*

Non-BKD female---BKD female

*

*

*

3.53

1.17

LSD (0.05)

87.21

LW

Male in Fall----- Male in Spring

*

*

*

Female in Fall— -Female in Spring

*

*

*

3.53

1.17

*

LSD (0.05)

87.21

Non-BKD Fall— -- — Non-BKD Spring

*

*

BKD Fall------- ---- BKD Spring

*

*

Non-BKD Fall------ BKD Fall

*

*

Non-BKD Spring----BKD Spring

*

LSD (0.05)

96.23

*
*

3.90

1.29

Table

9 (continued)

*= Significant difference (ie. difference > LSD(0.05))

46
Non-lipid e n erg y (cal/g)
6,000
Non-BKD

BKD

Non-BKD

BKD

5,000

4,000

3,000

2,000

1,000

Male

F em ale

Non-lipid energy (cal/g)
6,000
S pring

Fall

S pring

Fall

5,000

4,000

3,000

2,000

1,000

-

Male

Fem ale

Figure 10: Pairwise comparisons of means of non-lipid energy content in liver of
non-Bacterial Kidney Diseased (Non-BKD) and Bacterial Kidney Diseased (BKD)
Chinook salmon caught in Fall and Spring in Lake Michigan (1990-1992). The fish
were grouped into treatment combinations: sex*health (A) and sex*season (B).

47

(Figure 11-A) and less water (Figure 12-A) than BKD males
and females counterparts.
When evaluating the impact of sex and season on liver
contents,

I found that male and female Chinook salmon had

significantly less non-lipid energy in the Fall
4.47 % respectively)

(3.92 % and

than in the Spring (Figure 10-B).

Males and females possessed more (165.59 % and 169.36 %
respectively)

lipid (Figure 11-B) and less (1.96 % and 1.85

%) water (Figure 12-B) in the Fall than in the Spring.
When the two factors health and season were considered
non-BKD and BKD chinook salmon in the Fall contained less
(4.20 % and 3.95 % respectively) non-lipid energy (Figure
13-A) and more (177.98 % and 100.17 % respectively)

lipid

(Figure 14-A) than their non-BKD and BKD salmon counterparts
in the Spring (Table 8 and 9).

The within season comparison

showed that non-BKD fish had significantly more non-lipid
energy (Figure 13-B)
Spring

(Table 8).

than BKD fish in both the Fall and the

Additionally non-BKD fish had

significantly more lipid (Figure 14-B) than BKD fish in the
Fall while in contrast both groups of fish did not exibit
significant differences in lipid content in the liver during
the Spring.

In terms of changes in the liver water, I

observed a significant difference between non-BKD fish in
the Fall and the Spring with fall salmon containing less
water than those captured in the spring (Figure 15-A).
However, BKD infected fish did not exhibit a significant

48

Lipid content (%)
100

40

20

Non-BKD

Non-BKD
BKD

BKD

Spring

Spring

Lipid content (%)

100

Male

Female

Figure 11: Pairwise comparisons of means of lipid content in liver of non-Bacterial
Kidney Diseased (Non-BKD) and Bacterial Kidney Diseased (BKD) Chinook salmon
caught in Fall and Spring in Lake Michigan (1990-1992). The fish were grouped
info treatment combinations: sex*health (A) and sex*8eason (B).

•3

49

Water content (%)
100

80

BKD

BKD
Non-BKD

Non-BKD

^r^rr

20

Male

Female

Water content (%)
100

80

Spring

Spring

40

20

Male

Female

Figure 12: Pairwise comparisons of means of wafer content in liver of non-Bacterial
Kidney Diseased (Non-BKD) and Bacterial Kidney Diseased (BKD) Chinook salmon
caught in Fall and Spring in Lake Michigan (1990-1992). The fish were grouped
into treatment combinations: sex*health (A) and @ex*@eason (B).

50
Non-lipid energy (cal/g)
6,000
Spring

Spring

5,000

4,000

3,000

2,000

1,000

0
Non-BKD

BKD

Non-lipid energy (cal/g)
6,000
Non-BKD

Non-BKD

BKD

BKD

5,000

4,000

3,000

2,000

1,000

Fall

Spring

Figure 13: Pairwise comparisons of means of non-lipid energy content in liver of
non-Bacteria! Kidney Diseased (Non-BKD) and Bacterial Kidney Diseased (BKD)
Chinook salmon caught in Fall and Spring in Lake Michigan (1990-1992). The fish
were grouped into treatment combinations: health*season (A) and 8eason*health
(B).
The vertical bars represent +/- one Standard Error.

51

Lipid content (%)
100

Spring

Fall
Spring

BKD

Non-BKD

Lipid content (%)
100

Non-BKD

BKD

Non-BKD

BKD

Spring

Figure 14: Pairwise comparisons of means of lipid content in liver of non-Bacterlal
Kidney Diseased (Non-BKD) and Bacterial Kidney Diseased (BKD) Chinook salmon
caught in Fall and Spring in Lake Michigan (1990-1992). The fish were grouped into
treatment combinations: health*season (A) and season*health (B).

52

Water content (%)
100

Fall

80

Spring

Spring

Fall

60

40

20

0
Non-BKD

BKD

Water content (%)
100

BKD

80

BKD
Non-BKD

Non-BKD

60

40

20

0
Fall

Spring

Figure 15: Pairwise comparisons of means of water content in liver of non-Bacterlal
Kidney Diseased (Non-BKD) and Bacterial Kidney Diseased (BKD) chinook salmon
caught in Fall and Spring in Lake Michigan (1990-1992). The fish were grouped
into treatment combinations: health*season (A) and season*health (B).

53
difference in liver water content between the two seasons
(Figure 15-A).

The within season comparisons demonstrated

that BKD fish contained significantly more water than nonBKD fish in either season (Figure 15-B).

Three factors analysis
When the comparisons between season, sex and health
were performed (Table 10 and 11), non-BKD males and females
in Fall were found to have significantly less non-lipid
energy

(Figure 16-A) and more lipid (Figure 17-A) than their

non-BKD counterparts in Spring.
males in Fall contained

Additionally BKD infected

significantly less non-lipid energy

(Figure 16-B) and more lipid (Figure 17-B) than BKD infected
males in the Spring.

In contrast, BKD infected females did

not show signifant differences in non-lipid energy and
lipid contents in liver between Fall and Spring.

No

significant change in water content for non-BKD (Figure 18A) and BKD infected (Figure 18-B) males and females fish was
observed between fall and spring.
The within season comparison (Table 11) indicated that
non-BKD males in the Fall contained more (4.70 %) non-lipid
energy than BKD males in Fall (Figure 19-A) while no
significant difference was noted between non-BKD and BKD
females in Fall

(Figure 19-B).

and females had significantly

In the Spring, non-BKD males
greater non-lipid energy than

54
Table

10: Means and Standard Errors

(SE) of non-lipid

energy (LE), lipid (LF) and water (LW) contents in liver of
Chinook salmon caught in Fall and Spring in Lake Michigan
(1990-1992). Three factor treatment combinations were
considered.

See Table 11 for Pairwise comparisons of means.

liver non lipid
energy (LE)
in cal/g

SE

liver lipid
(LF)

in %

mean

SE

liver water
(LW) in %

Treatment

mean

NonBKD male Fall

5051.08 33.83

23.87

1.01 72.84

0.45

NonBKD male Spring

5246.68 13.08

8.90

0.54 74.57

0.20

NonBKD female Fall

4997.18 38.76

29.62

2.97 73.67

0.53

9.72

0.92 75.51

0.24

NonBKD female Spring 5259.57 12.53

mean

SE

BKD male Fall

4824.50 124.21 12.48

2.66 79.97

1.16

BKD male Spring

5092.71 23.43

5.70

0.40 79.50

0.52

BKD female Fall

4941.48 39.21

11.63

1.57 77.73

1.22

BKD female Spring

5067.77 29.27

6.38

1.05 78.97

0.37

55
Table

11: Three factor analyses and Pairwise comparisons of

means of non-lipid energy (LE), lipid (LF) and water (LW)
contents in liver of Chinook salmon using Fisher's Least
Significant Difference

(LSD).

The fish were caught in FA11

and Spring in Lake Michigan (1990-1992).

Significant
difference
Treatment combinations

-----------------------

tested

LE

LF

NonBKD male Fall----- -NonBKD male Spring

*

*

-NonBKD female Spring *

*

*

*

NonBKD male Fall--- - -BKD male Fall

*

*

NonBKD male Spring--- -BKD male Spring

*

NonBKD female Fall—

BKD male Fall-------- -BKD male Spring

LW

BKD female Fall---- — -BKD female Spring

NonBKD female Spring- -BKD female Spring
LSD (0 .05)

*
*

NonBKD female Fall--- -BKD female Fall
*
143.93

*

*
*

5.S3 1.94

*= Significant difference (ie. difference > LSD(0.05))

56

Non-lipid energy (cal/g)

e.ooo

Spring

Fall

Fall

5,000

4,000 -

3,000

2,000

1,000

Non-BKD mala

Non-BKD female

Non-lipid energy (cal/g)
6,000

5,000

Fall

Spring

Fall

Sprjng

4,000

3,000

2,000

1,000

BKD male

BKD female

Figure 16: Pairwise c o m p a ris o n s of means of non-lipid energy content In liver of
non-Bacterial Kidney Diseased (Non-BKD) and Bacterial Kidney Diseased (BKD) male
and female chinook salmon ca u g h t in Fall and Spring In Lake Michigan (1990-1992) •
The fish were grouped into treatment combinations: Non-BKD*sex*season (A) and
BKD*8ex*sea8on (B).

57

Lipid content (%)
100

80

60

40
Fall
Fall

20
Spring

Spring

0
Non-BKD male

Non-BKD female

Lipid content (%)
100

80

60

40

20

Fall

Fall
Spring

0

BKD male

Spring

BKD female

Figure 17: Pairwise comparisons of means of lipid content in liver of non-Bacterial
Kidney Diseased (Non-BKD) and Bacterial Kidney Diseased (BKD) male and female
Chinook salmon caught in Fall and Spring in Lake Michigan (1990-1992). The fish
were grouped into treatment combinations: Non-BKD*sex*season (A) and BKD*
8ex*season (B).
The vertical bars represent +/- one Standard Error.

58

Water content (%)
100

80

S p rin g

Fall

S p rin g

Fall

60

40

20

0

Non-BKD mala

Non-BKD female

Water content (%)
100

Fall

S p rin g

Fall

S p rin g

80

60

40

20

0

BKD male

BKD female

Figure 18: P airw ise comparisons of means of water content in liver of non-Bacterial
Kidney Diseased (Non-BKD) and Bacterial Kidney Diseased (BKD) m ale and female
Chinook salmon caught in Fall and Spring in Lake Michigan (1990-1992). The fish
were grouped into treatment combinations: Non-BKD*sex*8eason (A) and BKD*
8ex*season (B). The vertical bars represent +/- one Standard Error.

59

Non-lipid energy (cal/g)
0,000
Non-BKD

Non-BKD

BKD

5,000

BKD

4,000

3,000

2,000

-

1,000

-

Male Spring

Non-lipid energy (cal/g)
6,000
Non-BKD
5,000

Non-BKD

BKD

BKD

4,000

3,000

2,000

1,000

Female Fall

Female Spring

Figure 19: Pairwise comparisons of means of non-lipid energy content in liver of
non-Bacterial Kidney Diseased (Non-BKD) and Bacterial Kidney Diseased (BKD) male
and female Chinook salmon caught in Fall and Spring in Lake Michigan (1990-1992) The fish were grouped into treatment combinations: Male*season*health (A) and
Female*8eason*health (B). The vertical bars represent +/- one Standard Error.

60
BKD infected males and females.

Non-BKD males and females

also had significantly more lipid than BKD males (Figure 20A) and females

(Figure 20-B)

in the Fall.

No significant

difference were found in lipid contents of the liver when
non-BKD males and females are compared to BKD males and
females during the Spring.
When comparing BKD infected fish to non-BKD fish I
found that non-BKD males

(Figure 21-A) and females

(Figure

21-B) contained signifciantly less water than their BKD
counterparts in both Fall and Spring.

2x4 factorial design
The pattern of the impact of the BKD, as related to its
severity, on chinook salmon was analyzed in the Spring using
a 2x4 factorial analysis.

The main effect of health was

categorized in 4 levels: non-BKD, BKD stage 1, BKD stage 2,
and BKD stage 3.

The analsysis demonstrated that the

severity of the disease significantly impacted the non-lipid
energy,

lipid and water contents in the liver of infected

fish (Table 12).

One factor analysis
Both non-BKD and BKD chinook salmon use their lipid
reserve during the winter with larger depletions being

61

Lipid content (%)
100

80

60

40
Non-BKD

20

BKD
Non-BKD

Male Fall

BKD

Male Spring

Lipid content (%)
100

80

60

40
Non-BKD

20
BKD

Female Fall

Non-BKD

BKD

Female Spring

Figure 20: Pairwise comparisons of m eans of lipid content in liver of non-Bacterial
Kidney Diseased (Non-BKD) and Bacterial Kidney Diseased (BKD) male and female
chinook salm on caught in Fall and Spring in Lake Michigan (1990-1992). The fish
were grouped into treatment combinations: Male*season*health (A) and Female*
season*health (B). The vertical bars represent +/- one Standard Error.

62

Water content (%)
100

80

BKD

BKD
Non-BKD

Non-BKD

60

40

20

Male Spring

Water content (%)
100

80

Non-BKD

BKD

Non-BKD

BKD

60

40

20

Female Fall

Female Spring

Figure 21: Pairwise comparisons of means of water content in liver of non-Bacterial
Kidney Diseased (Non-BKD) and Bacterial Kidney Diseased (BKD) male and female
chinook salmon caught in Fall and Spring in Lake Michigan (1990-1992). The fish
were grouped into treatment combinations: Ma!e*8eason*healfh (A) and Female*
sea8on*heaith (B). The vertical bars represent +/- one Standard Error.

Table

12: ANOVA of 4x2 factorial design of non-lipid energy (LE), lipid (LF) and water

(LW) contents in liver of non-Bacterial Kidney Diseased (non-BKD) and Bacterial Kidney
Diseased (BKD) chinook salmon caught in Spring in Lake Michigan (1990-1992).

Liver non-lipid

Liver lipid

energy(LE)

Liver water

(LF)

F

(LW)

df

MS

Model

7

69388.88

11.97*

27.06

1.74

43.34

25.86*

Health

3

159336.02

27.48*

56.82

3.64*

95.74

57.11*

Sex

1

129.25

0.02

12.65

0.81

6.68

3.98*

Health*Sex 3

2528.28

0.44

2.10

0.13

1.89

0.14

Error

5798.22

76

*= significant at a= 5 %

MS

15.59

F

F

Source

MS

1.67

64

directly related to the severity of the disease (Table 13,
14, Figure 22). Further, diseased fish unlike non-BKD fish
significantly reduced the non-lipid energy reserve of the
liver and magnitude of this loss was related to the degree
of infection (Table 13, 14, Figure 23).

Additionally, the

water content in the liver also increased significantly as
the severity of the disease increased (Table 13 and 14,
Figure 24).

Two factors analysis
The LSD on the two factors treatment combinations
(Table 15, Table 16) indicated that males

(Figure 25) and

females (Figure 26) chinook salmon significantly deplete
their non-lipid energy resources during the course of the
BKD infection.

The degree of non-lipid energy loss is

graphically summarized in Figure 27.

Lipid content, unlike

the non-lipid energy, dropped sharply with the onset of BKD
infection for both male
salmon.

(Figure 28) and female (Figure 29)

The liver lipid reserve was intensively used by

both non-BKD and
BKD fish.

Figure 30 depicts the losses in lipid during R.

salmoninarum infection.

Infected male (Figure 31) and

female (Figure 32) chinook salmon increased significantly
their liver water content when compared to non-BKD
counterparts.

In contrast but inversely related to lipid

65
Table

13: Means and Standard Errors (SE) of non-lipid

energy (LE), lipid (LF) and water (LW) contents in liver of
chinook salmon caught in Spring in Lake Michigan (19901992).

One factor treatment combinations were considered.

See Table 14 for Pairwise comparisons of means.

liver non-lipid
energy (LE)
in cal/g

liver lipid

liver water

(LF) in %

(LW) in %

Treatment

mean

SE

mean

SE

mean

SE

Non-BKD fish

5253.02

9.03

9.31

0.53

75.03

0.16

BKD1 fish

5093.52

17.81

6.15

0.96

79.7

0.85

BKD2 fish

5093.70

32.38

6.13

3.02

78.69

0.32

BKD3 fish

5039.84

19.99

5.67

0.65

79.99

0.69

Male

5207.29

15.33

8.08

0.46

75.83

0.38

Female

5212.79

17.48

8.91

0.77

76.35

0.31

66

Table

14 : One factor analyses and Pairwise comparisons of

means of non-lipid energy (LE), lipid (LF) and water (LW)
contents in liver of Chinook salmon using Fisher's Least
Significant Difference (LSD).

The fish were caught in

Spring in Lake Michigan (1990-1992).

Significant
difference
Treatment combinations
tested

---------------------LE

LF

LW

-------- BKD1

*

*

*

Non-BKD----------- ----------- BKD2

*

*

*

------- ---- -------- BKD3

*

*

*

Non-BKD------- — — —

Non-BKD-—

*

BKD1-------------- ---- ------- BKD2
BKD1-------------- — --------- BKD3

*

BKD2-------------- ----------- BKD3

*

Male— --------- —

*

LSD (0.05)

*=significant at a= 5 %

39.89

2.07

0.67

Lipid content (%)
100

80

40

CM

cluster 1

cluster 2

cluster 3

cluster 4

CM

clusters

cluster 6

cluster 7

Figure 22: Pairwise comparisons of means of lipid content in liver of Bacterial Kidney
Diseased (Non-BKD) and Bacterial Kidney Diseased (BKD) stagel (BKD 1), stage2 (BKD 2)
and stage3 (BKD 3) male and female Chinook salmon caught in Spring In Lake Michigan
(1990-1992). The fish were clustered in pairs (clusterl-cluster?) of treatment combinations:
Non-BKD*BKD(1,2 or 3) and Male*Female. The vertical bars represent +/- one Standard
Error.

Non-lipid energy content (cal/g)
6,000
CM

to

CM

U.

m
5,000

4,000

3,000

2,000

1,000

cluster 1

cluster 2

cluster 3

cluster 4

cluster 5

cluster 6

cluster 7

Figure 23: Pairwise com parisons of m eans of non-lipid energy content innon-Bacterlal Kidney
Diseased (Non-BKD) and Bacterial Kidney Diseased (BKD) stage 1 (BKD 1), stage2 (BKD 2)
and stageS (BKD 3) male and female Chinook salmon caught In Spring in Lake Michigan
(1990-1992). The fish were clustered In pairs (cluster1-cluster7) of treatmentcomblnations:
Non-BKD*BKD(1,2 or 3) and Male*Female. The vertical bars represent +/- one Standard

Water content (%)
100

CM

o

CO

CM

CD

80

u.

60

cluster 1

cluster 2

cluster 3

cluster 4

cluster 5

clusters

cluster 7

Figure 24: Pairwise com parisons of m eans of water content in liver oftion-Bacterial Kidney
Diseased (Non-BKD) and Bacterial Kidney Diseased (BKD) stagel (BKD 1), stage2 (BKD 2)
and stag®3 (BKD 3) male and female Chinook salmon caught in Spring in Lake Michigan
(1990-1992). The fish were clustered In pairs (clusterl-cluster7) of tfeatmentcombinations:
Non-BKD*BKD(1,2 or 3) and Male*Female. The vertical bars represent +/- one Standard

70

Table

15: Means and Standard Errors (SE) of non-lipid

energy (LE), lipid (LF) and water (LW) contents in liver of
Chinook salmon caught in Spring in Lake Michigan (19901992) .

Two factor treatment combinations were considered.

See Table 16 for Pairwise comparisons of means.

liver non-lipid
energy (LE)
in cal/g

liver lipid

liver water

(LF) in %

(LW) in %

Treatment

mean

SE

mean

SE

mean

SE

Non-BKD male

5246.68

13.09

8.90

0.54

74.57

0.20

Non-BKD female 5259.57

12.53

9.72

0.92

75.51

0.24

BKD1 male

5117.29

45.75

5.86

0.79

80.48

2.08

BKD1 female

5077.67

6.83

6.35

1.68

79.23

0.83

BKD2 male

5102.32

38.69

5.43

0.60

78.88

0.49

BKD2 female

5083.35

58.94

6.97

1.92

78.46

0.43

BKD3 male

5057.08

30.87

6.14

0.87

80.09

1.16

BKD3 female

5013.96

4 .68

4.96

1.05

79.86

0.89

71
Table

16: Two factor analyses and Pairwise comparisons of

means of non-lipid energy (LE), lipid (LF) and water (LW)
contents in liver of Chinook salmon using Fisher's Least
Significant Difference (LSD).

The fish were caught in

Spring in Lake Michigan (1990-1992).

Significant
difference
--------------------

Treatment combinations
tested

LE

Non-BKD male--------- — — BKD1 male

*

Non-BKD male

BKD2 male

*

Non-BKD male—

-BKD3 male

*

LF

LW

*
*

*
*

BKD1 male-------- — ------ BKD2 male

*

BKD1 male---------------- BKD3 male
BKD2 male----— ---------- BKD3 male

*

Non-BKD female----------- BKD1 female

*

Non-BKD female—

BKD2 female

*

Non-BKD female— ---------- BKD3 female

*

*

*
*

*

*

BKD1 female--------- — --- BKD2 female
BKD1 female

BKD3 female

BKD2 female— —

BKD3 female
LSD (0.05)

*
63.64

3.30

1.08

72

Table

16 (continued)

*=significant at a= 5 %

Non-lipid energy content (cal/g)
6,000

/

#

/

.o'

5,000

4,000

3,000

2,000

1,000

0

cluster 1

cluster 2

cluster 3

cluster 4

cluster 5

Figure 25: Pairwise comparisons of means of non-lipid energy content in liver nonBacterlal Kidney Diseased (NBKD) and Bacterial Kidney Diseased stagei (BKD1), stage2
(BKD2) and stages (BKD3) male chlnook salmon caught In Lake Michigan (1990-1992). The
fish were clustered In pairs (clusterl -cluster6) of treatment combinations: NBKD male*
BKD(1,2 or 3) male. The vertical bars represent +/- one Standard Error.

Non-lipid energy content (cal/g)
----- 7 5 ^ — 35--- ^ ^ --

6.000

/

/

/

/

/

5,000

4,000

3,000

2,000

1,000

cluster 1

cluster 2

cluster 3

cluster 4

cluster 5

cluster 6

Figure 26: Pairwise comparisons of means of non-llpid energy content in liver of nonBacterial Kidney Diseased (NBKD) and Bacterial Kidney Diseased stagel (BKD1), stage2
(BKD2) and stages (BKD3) female Chinook salmon caught in Spring in Lake Michigan
(1990-1992). The fish were clustered In pairs (cluster1-cluster@) of treatment combinations:
NBKD femaie*BKD(1,2 or 3) female. The vertical bars represent +/- one Standard Error.

c

Non-lipid energy (cal/g)
H Male
H Female

5,300
5,250
5,200
01

5,150
5,100 5,050 5,000
NBKD

BKD1

BKD2

BKD3

Figure 27: Means of non-lipid energy content in liver of male and female
non-Bacterial Kidney Diseased (NBKD) and Bacterial Kidney Diseased stage 1
(BKD1), stage 2 (BKD2) and stage 3 (BKD3) Chinook salmon caught In Spring In
Lake Michigan (1990-1992).

Lipid content (%)
100

60

40

cluster 1

cluster 2

cluster 3

cluster 4

cluster 5

cluster 6

Figure 28: Pairwise comparisons o? means of lipid content in liver of non-Baderial Kidney
Diseased (Non-BKD) and Bacterial Kidney Diseased stagel (BKD1), stage2 (BKD2) and stages
(BKD3) male Chinook salmon caught in Spring in Lake Michigan (1990-1992). The fish were
clustered in pairs (clusterl-clusters) of treatment combinations: Non-BKD male* BKD(1,2 or 3)
male. The vertical bars represent +/- one Standard Error.

Lipid content (%)
100

40 -

>l

-J

CM

CM

CM
CO

cluster 1

cluster 2

cluster 3

cluster 4

cluster 5

cluster 6

Figure 29: Pairwise comparisons of means of lipid content in liver of non-Bacteria! Kidney
Diseased (Non-BKD) and Bacterial Kidney Diseased stagel (BKD1), stage2 (BKD2) and stage3
(BKD3) female Chinook salmon caught In Spring In Lake Michigan (1990-1992). The fish were
clustered in pairs (clusterl-clusters) of treatment combinations: Non-BKD femaie*BKD(1,2
or 3) female. The vertical bars represent +/- one Standard Error.

Lipid content (%)
□ Male
H Female

10

9
8

7
6
5
4

NBKD

BKD1

BKD2

BKD3

Figure 30: Means of lipid content In liver of male and female non-Bacterial
Kidney Diseased (NBKD) and Bacterial Kidney Diseased stage 1 (BKD1), stage 2
(BKD2) and stage 3 (BKD3) chlnook salmon caught in Spring in Lake Michigan (1990
-1992).

Water content (%)
100

CM

CM

80

60

40

20

0

cluster 1

cluster 2

cluster 3

cluster 4

cluster 5

cluster 6

Figure 31: Pairwise comparisons of means of water content in liver of non-Bacteriai Kidney
Diseased (Non-BKD) and Bacterial Kidney Diseased stagel (BKD1), @tage2 (BKD) and stages
(BKD3) male chinook salmon caught in Spring in Lake Michigan (1 ©90-1992). The fish were
clustered in pairs (clusierl-clusters) of treatment combinations: Non-BKD male* BKD(1,2 or 3)
male. The vertical bars represent -f/- one Standard Error.

Water content (%)
100

80

60

40

20

0
cluster 1

cluster 2

cluster 3

cluster 4

cluster 5

cluster 6

Figure 32: Pairwise com parisons of means of wafer content In liver of non-Bacl@rial Kidney Diseased
(Non-BKD) and Bacterial Kidney Diseased sfagel (BKD1), @t@g®2 (BKDd2) and stegsS (BKD3) female
chlnook salmon caught In Spring in Lake Michigan (1S9©-1®92). Th® fish were clustered In pairs
(elusfer1-c!usterf») of treatment combinations: Non-BKD female*BKD(1,2 or 3) female. The vertical
bars represent +/- one Standard Error.

o

81

content, the water content in theliver increased
dramatically with the onset of BKD
(Figure 33).

Water content (%)

■

NBKD

BKD1

BKD2

BKD3

Figure 33: Means of water content in liver of male and female non-Bacterial
Kidney Diseased (NBKD) and Bacterial Kidney Diseased stage 1 (BKD1), stage 2
(BKD2) and stage 3 (BKD3) chinook salmon caught in Lake Michigan (1990-1992).

83

CHAPTER IV.

4.1.

DISCUSSIONS AND CONCLUSION

Occurrence and impact of Bacterial Kidney Disease.

The occurrence of Bacterial Kidney Disease as a
significant mortality factor for Chinook salmon in Lake
Michigan was noted by biologist of the Michigan Department
of Natural Resources since 1987 (Keller et a l . 1989).

The

dramatic decline in alewives in the early to mid 1980' s
possibly created the adverse (poor food resources)
environment leading to the significant outbreak of
Bacterial Kidney Disease.

Such interactions between the

host, the parasite and the environment has been described by
Snieszko (1973). The introduction of Renibacterium
salmoninarum in Chinook salmon in the Great Lakes probably
occurred during their initial planting. This bacterium was
reported to be present in these lakes prior to 1955
1983) .

(Post

However, Renibacterium salmoninarum was considered

to be a mild pathogen in the Great Lakes prior to 1987 when
dead and dying Chinook salmon with the disease were observed
washing up on the beach during the spring.

The impact of

BKD on these salmon has been dramatic as Keller
Charlevoix, Michigan, Personal communication)

(1992, MDNR,

estimates that

it has been responsible for the observed 50% decrease in the
Lake Michigan Chinook salmon abundance since the early

84

1980's.

Concurrent with this decrease in abundance, Hay

(1989) noted a decline in the

individual growth and catch

of this species in Lake Michigan.

The increase in

susceptibility of the fish to the infectious agent appears
to be related to the amount of energy reserves of the
individual fish during the fall and spring.
contained significantly less lipid
spring compared to non-BKD fish.

BKD fish

during the fall and
Lipids, although

comprising only 2-12 % of the total wet weight of fish
(Weatherly and Gill 1987) constitute the main source of
energy for these animals.

Thus, small changes in this

energy reserve can significantly affect the health and
longevity of the fish (Salz 1989).

Additionally,

it is this

energy resource which is used during starvation (Hadley
1985).

4.2.

Quantification of Renibacterium salmoninarum

The bacteria reached its optimum growth at the second
stage (BKD2) of the disease during the Spring.

Spring

temperature conditions are ideal for the growth of
Renibacterium salmoninarum.

During the spring, I noted that

salmon infected with this bacterium displayed severe
necrotic regions in the kidney and sometimes in the 1iver.
It appears that in these fish, the stress to overcome the
bacterial invasion by stimulation of the immune response

85

failed,

leading to the destruction of the fish cells with

immediate consequence being the production of massive deep
necrotic areas in the kidney (Kinkelin 1974, Fryer and
Sanders 1981). Simultaneously the number of bacteria cells
declined in the mortem stage (BKD3) because of the drop in
number of living fish cells for the bacteria to feed on.

4.3.

Pattern of fish separation.

This study has shown the existence of 3 distinct stages
of BKD infection.

The pathological condition of the fish

could be easily detected by visual observation of the gross
lesions.

Further, I found that I could use Principal

Components scores to completely separate non-BKD from BKD
fish.

Principal Component Analysis has great potential to

be useful in separating other microbially infected fishes
based on information on their general health condition.

4.4.

Impact of Renibacterium salmoninarum infection

The factorial analysis and the multiple comparison of
non- lipid energy,

lipid and water contents demonstrated

that non-BKD Chinook salmon use primarily non-lipid energy
during the Summer for growth and maintenance while they
store lipid for use during the Winter and early Spring.

86
Jezierska et al.

(1982) reported that rainbow trout

(Oncorhvnchus mvkissl relied heavily on lipid reserves
during the overwintering period.
It was apparent that BKD infected fish store
significantly less lipid during the summer when compared to
non-BKD fish.

Consequently these salmon deplete their lipid

reserve during the winter period and intensively use their
non-lipid energy stores to survive through to Spring.

My

analysis showed that non-BKD fish had significantly more
non-lipid energy and lipid than BKD fish in both Fall and
Spring.

The severity of the disease during the Spring,

coupled with the stress due to seasonal transition weakens
the efficiency of the immune system and favors the virulence
of the pathogen; such a state of stress necessarily
generates an intensive use of energy resources for
biological needs.

The relationship between disease

condition and liver content in the fall and spring clearly
demonstrated that storage in non-lipid energy and lipid are
linked to fish health. As fish health deteriorated energy
reserves were depleted while water content increased.

4.5.

Control of Bacterial Kidney Disease.

Once established Renibacterium salmoninarum is an
extremely difficult pathogen to eradicate

(Warren 1983).

To

87

date there is no effective vaccination treatment and control
through chemotherapy gives only temporary relief (Eliott et
al. 1989).
Given our inability to control

salmoninarum a

possible approach to solve the BKD problem is the creation
of strains of chinook salmon that are resistant to the
infectious agent (Suzumoto et al. 1977, Withler and Evelyn
1990, and Beacham and Evelyn 1992). Although the evaluation
of the lipid reserve in this study is not directly related
to the mechanism of the immune defense of the
fish, this work indicated the critical role played by lipid
reserve during the stages of infection.

Another possibility

is to use a strain of fish which has the ability to more
efficiently convert the current food resources in the Great
Lakes to lipid, thereby increasing their chance of surviving
the stress related to BKD infection.

As energy storage is

related to fish health a sufficient energy reserve to meet
the biological need during stress condition can extend the
period of resistance of the fish and thus increase the
chance of the build up of more efficient immune response.
Time has been reported to be important for successful immune
responses

(Roberts 1978).

Although it was not demonstrated

in this study, it is possible that there exists chinook
salmon in Lake Michigan which may be more efficient energy
converters than other or that possess a natural or non­
specific antibodies resistant to R^. salmoninarum.

If this

88
is the case, it may explain why some fish exhibit advanced
stages of BKD while others do not.

In conclusion,

I found that Renibacterium salmoninarum

infection significantly effected the energy and water
contents of the liver of chinook salmon from Lake Michigan.
The severe depletion of lipid and non-lipid energy resources
in BKD infected salmon during the fall and winter results in
fish deaths during the Spring.

This disease has likely

contributed to the major decline in the abundance noted for
this species in the late 19 8 0 's in Lake
Michigan and efforts to eliminate the disease or remediate
its effects are essential for a future healthy and
productive chinook salmon fishery.

APPENDICES

Appendix

A-l: Means and Principal Components (PCI, PC2 and PC3) scores of non-lipid energy

(LE), lipid (LF) and water (LW) contents in liver of non-Bacterial Kidney Diseased (0) and
Bacterial Kidney Diseased (2) male (1) and female (2) chinook salmon caught in Fall in Lake
Michigan (1990-1992).

Observation

Fish

LE

LF

Tag

(cal/g)

(%)

LW
(%)

Health Sex

PCI

PC2

PC3

1

268

4943.37

26.8387

74.0554

0

1

0.82919 -0.51128

0.21975

2

113439

5002.8

20.1354

75.1973

0

1

0.20508 -0.03461

0.00168

3

113441

5155.95

23.0357

76.3855

0

2

0.34872

0.68442

0.50794

4

258

4978.29

44.1324

74.3991

0

1

1.99862 -0.71365

1.53228

5

254

5005.82

23.8789

73.3422

0

1

0.84354 -0.16856 -0.11769

6

205

5099.7

22.4882

75.8422

0

1

0.35346

0.40507

0.33652

7

113455

5067.5

33.3132

71.321

0

2

1.97989 -0.15065

0.15287

8

113460

4993 .9

17.6376

70.799

0

1

0.91223 -0.17773 -1.09511

Appendix A-l continued
9

223

5035.19

18.4301

73.2138

0

1

0.52754

0.0899 -0.52157

10

113433

5178.06

29.6361

70.935

0

2

1.93598

0.45242 -0.15087

11

220

5155.49

17.5384

72.7229

0

1

0.70935

0.67466 -0.64586

12

270

5253.14

30.4904

68.6576

0

1

2.54628

0.71643 -0.53877

13

113478

5037.93

32.3669

69.4091

0

1

2.26616 -0.34037 -0.32282

14

110450

4976.83

23.5158

72.7796

0

2

0.89762

15

113481

5034.69

36.9376

72.1252

0

2

2.02862 -0.36146

16

110448

5059.7

27.4618

73.6661

0

1

1.09053

0.02422

17

113462

5211.44

16.6843

71.9876

0

1

0.86619

0.93844 -0.84052

18

113487

5121.59

15.9941

74.4148

0

1

0.21919

0.60462

19

113431

5215.01

15.2601

72.9386

0

1

0.57927

1.02092 -0.74239

20

113434

5062.3

26.9463

72.1548

0

1

1.36403 -0.00517 -0.12767

21

110449

5219.23

30.5328

75.5514

0

1

1.11247

0.79496

22

110414

5142.49

24.6824

72.3119

0

2

1.27157

0.43876 -0.22827

23

110409

5179.49

38.9556

66.1487

0

2

3.55217

0.08351 -0.48516

24

110406

6032.59

23.8111

73.9684

0

1

1.94221

4.82529

25

113490

5043.06

15.9642

74.9581

0

2

0.01299

0.24433 -0.33072

-0.3207 -0.27066
0.56669
0.2229

-0.4148

0.88868

0.36073

Appendix A-l continued
0.50058 -0.19774 -0.13564

26

113493

4983.47

21.5207

74.0974

27

113500

5055.83

26.779

75.9347

0.5792

0.10083

0.64564

28

113499

4944.79

24.2785

74.2196

0.6204 -0.44186

0.0725

29

113466

4912.6

22.8379

72.6979

30

113492

5117.22

22.6694

73.2849

0.9049

0.39547 -0.17731

31

113494

5061.18

16.6075

73.3685

0.40117

0.2616 -0.60993

32

113483

4848.45

25.1067

71.507

1.11167

-1.0224 -0.46682

33

113479

5269.39

29.712

76.4398

0.93581

1.0874

1.03267

34

110415

5020.51

20.9436

76.0636

0.1068

0.06383

0.24567

35

110430

4903.31

18.3982

73.7607

36

113401

5152.52

24.5022

74.4485

37

110412

4493.89

23.3937

73.6432

0.13564 —2.62526 -0.26537

38

113409

4661.02

25.8855

71,3525

0.97231 -1.95243 -0.50807

39

113464

4983.46

39.2874

73.4021

1.87132

-0.6164

0.98207

40

110407

4815.23

32.8284

74.1386

1.07346 -1.26148

0.61869

41

113472

4831.92

27.0347

73.5047

0.82076 -1.07458

0.08067

42

113407

5016.13

30.0428

67.2862

2.50907 -0.46939

-0.9377

0.79029 -0.61951 -0.35794

0.25655 -0.52845 -0.45521
0.8384

0.56682

0.20747

o

Appendix A-l continued
43

211

4936.97

15.9816

71.9612

0

1

0.49404 ■
-0.37549 ■-0.99026

44

113473

5086.2

64.1752

73.4505

0

2

-0.66937
3.70751 •

45

110410

4929

26.5058

76.8694

0

2

0.21902 -0.47405

0.77736

46

110403

4824.34

28.3794

76.1481

0

2

0.36942 -1.04771

0.72428

47

110424

5000.66

24.4382

73.9154

0

1

0.76001 -0.18568

0.03968

48

113756

4423.42

25.8804

72.4344

0

2

0.46818 -3.06421 -0.36469

49

224

5034.87

25.1854

75.6719

0

1

0.49698

0.02543

0.47044

50

225

5217.13

26.6093

78.9004

0

1

0.16009

0.99017

1.30681

51

113425

5265.46

15.2937

74.6644

0

1

0.29253

1.32538 -0.36315

52

113422

4930.65

22.302

74.8254

0

1

0.34395 -0.44504

53

113424

5125.99

19.5107

63.5118

0

1

2.67599

54

113402

4947.32

19.0729

74.0708

0

2

0.29319 -0.31941 -0.32755

55

113432

4946.93

19.5124

68.518

0

1

1.44772 -0.52725

56

113444

5044.55

17.0427

75.3229

0

1

0.01557

57

113482

4843.5

19.2436

73.0352

0

1

0.39037 -0.86232 -0.56684

58

113436

5109.37

15.0939

69.2383

0

1

1.19058

0.3825 -1.56132

59

113447

5252.75

29.3018

66.7676

0

1

2.8463

0.67413 -1.01709

2.79605

0.05338

0.16264 -2.43467

-1.4532

0.24051 -0.17758

Appendix A-l continued
60

110432

4977.2

73.0571

76.0371

3.66796 -1.30255

3.92862

61

113486

4925.13

23.1435

75.3437

0.29063 -0.47209

0.21926

62

110421

4958.1

29.3589

74.888

0.85269 -0.46649

0.57733

63

113485

4976.32

22.8598

73.2481

64

275

4798.38

11.6888

67.0731

65

292

4900.28

21.3074

79.671

66

216

4909.91

3.9204

67

252

4988.51

68

288

69

0.7567 -0.29208 -0.21985
0.74264 -0.39877

0.98171

3.06219

0.2302

0.90915

85.238

1.04518

0.19661

0.0649

11.6986

78.4023

1.53018

0.42101

0.33367

5012.42

9.8163

80.2952

2.71129

0.59278

0.45568

212

4990.74

2.3746

83.4552

2.10354 -0.91772

0.04942

70

113467

4721.66

6.8939

80.4072

1.43458

0.55884

0.70689

71

296

5048.84

12.7273

81.0337

1.20227

1.13421 -0.13842

72

273

5159.83

7.2359

78.6661

0.89469 -0.75842

0.88861

73

300

4822.03

20.4205

79.6557

0.90793

0.31168

0.04099

74

289

5017.06

12.2066

78.0672

2.72161 -6.50083 -0.78616

75

282

3598.86

8.9898

77.5331

1.25544

0.28433

0.49575

76

299

5001.35

13.0646

79.9834

1.05488

0.2668

0.03165

Appendix A-l continued
77

215

5001.19

11.2543

78.3734

2

2

-0.688

-0.1412

0.52298

78

113446

4946.49

17.9662

78.5325

2

2

-2.4439

0.20016

1.09126

79

294

4936.24

9.3789

84.2068

2

1 -1.00567

0.32569

0.70929

80

207

4975.04

34.77

70.9862

2

1 -1.10984 -0.36065

0.13403

81

276

4878.4

12.9368

78.4934

2

1 -2.07999 -0.12878 -0.01099

82

113437

4874.8

4.943

80.5301

2

2 -1.41879 -0.17874 -0.09243

83

113440

4895.71

8.9195

78.7482

2

1 -0.89489

84

298

5006.57

11.129

77.5724

2

2

-1.5627

0.26732 -0.14228
0.1087 -0.25352
VO

Appendix

A-2:

Means and Principal Components (PCI, PC2 and PC3) scores of non-lipid energy

(LE), lipid (LF) and water (LW) contents in liver of non-Bacterial Kidney Diseased (0) and
Bacterial Kidney Diseased stage 1 (1), stage 2 (2) and stage 3 (3) male (1) and female (2)
Chinook salmon caught in Spring in Lake Michigan (1990-1992).

Observation

Fish

LE

LF

LW

Tag

(cal/g)

(%)

(%)

Health

Sex

PCI

PC2

PC 3

1.69574

0.06989

0.14631

1

110467

5304.72

11.4651

73.8879

0

2

2

111398

4885.6

14.227

79.1357

2

2 -2.20786

3

111399

5136.36

5.4479

74.2451

0

1

4

111388

5077.14

6.7653

80.7821

1

2 -2.50234

0.3166

0.66677

5

111400

5051.13

4.5854

80.8294

2

1 -2.87845

0.02435

0.28245

6

111377

4984.52

6.3439

79.251

2

1 -2.49623

0.80218 -0.40942

7

111384

5038.85

4.8304

77.6419

2

2 -1.76996

0.20614

8

111379

5067.03

4.9527

78.7554

2

1 -2.00628

0.04501 -0.28683

9

111386

5062.49

7.852

78.0004

3

1 -1.46942

0.67596 -0.22462

2.98841 -0.06269

0.04952 -0.17544 -1.36943

-0.8322

Ln

Appendix A-2 continued
1.53977

10

111391

5163.04

5.0662

82.5618

1

1 -2.84623 -0.56591

11

111345

5009.28

6.0139

78.973

3

2

-2.2938

0.59521 -0.40595

12

111343

5085.28

7.7326

77.5361

2

1 -1.18946

0.52689 -0.27005

13

111341

5071.54

6.6477

78.4G34

1

1 -1.68565

0.37023 -0.18087

14

111376

5066.1

3.2507

78.9679

1

2 -2.25773 -0.29998 -0.42101

15

111342

5100.17

7.5865

77.3737

2

2 -1.06424

16

5992

5266.7

4.6014

75.9494

0

2

17

110436

5001.12

5.0513

82.0032

3

1 -3.52839

0.39004

0.45846

18

111305

5178.54

3.4696

79.2755

2

1 -1.73426 -0.91987

0.32737

19

111352

5335.55

10.8287

76.8709

0

2

0.72317 -0.29638

1.24782

20

111367

5403.77

9.2493

78.5229

0

2

0.34098 -1.04859

1.99457

21

111364

5018.64

3.9165

80.7515

3

2 -3.09404

0.07998 -0.00155

22

111303

5089.76

9.0271

77.9502

1

2 -1.18556

0.75636

0.04713

23

72

5287.13

27.5166

77.6905

0

2

1.82678

3.36867

3.22676

24

110438

5304.74

9.0353

76.3571

0

1

0.56204 -0.47128

0.69328

25

110441

5295.76

3.8677

75.8382

0

2

0.18538 -1.46078 -0.14093

26

67

5241.57

8.6679

77.6592

0

2

0.41299 -0.26017

0.06 -1.14331 -0.17676

0.5114 -1.26153

0.55556

Appendix A-2 continued
27

111372

5175.48

9.2156

74.2583

0

1

0.63349

0.36208 -0.70522

28

59

5308.23

12.9249

75.2507

0

2

1.36825

0.32081

0.79789

29

111312

5228.32

13.2326

74.4011

0

2

1.27025

0.86753

0.10768

30

110428

5305.27

10.1318

76.8486

0

2

0.49674 -0.26045

0.99147

31

65

5356.87

6.2314

74.8055

0

1

1,12687 -1.31819

0.12561

32

73

5286.99

5.6108

75.4042

0

2

0.46803 -1.04644 -0.13004

33

68

5133.56

12.0176

74.9614

0

2

0.43045

1.16475 -0.36798

34

111371

5288.45

14.4115

74.3816

0

1

1.72249

0.75564

35

111304

5202.82

5.1775

74.1791

0

1

0.40856 -0.61854 -1.05789

36

110462

5238.65

6.3005

75.3805

0

2

0.28194 -0.62244

-0.3227

37

110466

5321.49

8.437

74.9

0

1

1.11983 -0.66376

0.22301

38

110478

5323.55

9.4376

75.0762

0

2

1.16715 -0.47549

0.41186

39

110426

5235.16

9.5145

74.8369

0

2

0.77957

0.06245 -0.14623

40

110468

5171.8

12.8708

73.226

0

1

1.35053

1.14709 -0.64271

41

110465

5267.02

8.2841

74.517

0

2

0.94606 -0.36861 -0.22406

42

110487

5190.36

10.1639

72.5309

0

1

1.43282

0.50045 -1.09454

43

110463

5363.37

16.4079

73.6532

0

2

2.59276

0.73686

0.57104

0.97295

Appendix A-2 continued
44

110471

5324.95

5.4904

74.3776

1.03361 -1.27399 -0.28192

45

110454

5278.63

8.0698

75.1384

0.76356 -0.49194

0.02432

46

110445

5320.25

6.1969

75.3819

0.71577 -1.12156

0.11457

47

111314

5179.57

8.914

73.4036

0.93433

48

111313

5299.34

8.3115

77.315

49

111317

5186.43

9.4249

72.8308

1.22948

50

111322

5159.29

4.2942

78.4796

1.46958 -0.62427

51

110439

5167.4

9.0557

73.6738

0.78456

52

111315

5318.18

10.6225

75.7809

1.00152 -0.21611

53

110489

5334.5

11.0589

72.7782

2.21824 -0.16636 -0.11234

54

111361

5267.1

10.5517

76

0.63701

0.06458

55

110456

5163.31

18.7562

72.2147

2.25605

2.41381 -0.33648

56

111359

5314.26

6.1596

73.9835

1.18436 -1.06775 -0.39484

57

111353

5216.13

4.977

74.6331

0.29715 -0.74586 -0.85503

58

110452

5232.85

3.9335

79.6784

1.53757 -1.15116

59

111318

5256.59

6.4394

74.4681

0.72299

-0.6821 -0.51558

60

110460

5172.94

10.8936

75.4569

0.35408

0.69595 -0.11662

0.29281 -1.00679

0.11457 -0.60502

0.90152

0.36741 -1.10213
0.05007

0.38785 -0.96584
0.75999

0.54451

0.81697

Appendix A-2 continued
-0.1941

0.27104

74.9385

0.39669 -0.69429

-0.5243

5.8478

74.8572

0.61381

5255.97

8.8504

76.5631

0.20346 -0.22712

110470

5244.65

6.1329

74.9024

0.47055 -0.68272 -0.47083

66

110455

5261

6.6008

73.8531

0.98519

-0.6635 -0.67987

67

111370

5140.22

13.5238

74.7245

0.70244

1.4368 -0.23361

68

110474

5323.04

8.4525

75.7215

0.83318 -0.68515

69

110482

5198.78

3.6456

75.8422

0.32609 -0.50732 -0.54483

70

110486

5293.33

8.0796

77.0884

0.14052 -0.61276

0.76465

71

69

5184.06

10.6338

75.9837

0.19853

0.56805

0.09173

72

61

5148.44

15.3062

74.3766

1.05051

73

60

5294.01

4.9356

75.3784

0.44833 -1.22453 -0.17977

74

58

5197

14.2682

75.0148

0.98123

1.25022

0.26466

75

111308

5107.65

5.5206

80.262

-2.27232

-0.1057

0.51227

76

57

5268.94

6.7493

74.5967

0.77477 -0.69379 -0.36764

77

110485

5318.74

7.1847

74.4603

1.13881

61

111360

5265.32

9.2011

75.6909

62

110497

5237.98

5.8874

63

110451

5273.16

64

111366

65

0.6043

-0.9069 -0.36283
0.47263

0.51067

1.758 -0.09554

-0.8943 -0.08827

Appendix A-2 continued
78

64

5099.32

10.8506

74.1885

0

1

79

110459

5253.54

2.4049

74.0149

0

2

0.46785 -1.47693 -1.16129

80

111309

5407.4

5.4566

76.9147

0

1

0.56336 -1.81157

1.02422

81

110496

5089.09

17.5442

76.0751

0

2

0.33681

0.41524

82

110457

5252.48

6.148

76.1369

0

2

0.06892 -0.74872 -0.00925

83

616912

5247.44

5.4868

77.6592

2

1

-0.5741 -0.88244

84

616915

5160.62

12.7398

72.8689

0

2

1.40547

0.4066

1.14235 -0.95498

2.52926

0.39889

1.19261 -0.84022

100

101
Lipid content (%)
100

80

60

40
NorvBKO

20
BKD
NorvBKD
BKD

0

Spring

Season
Water content (%)
100

BKD

BKD
NorvBKD

40

20

Male

Female

Sex

Appendix B: Graphical demonstration of the apparent interaction between the
factors health and season (A), for lipid content, and health and sex (B), for
water content in liver of non-Bacterial Kidney Diseased (Non-BKD) and Bacterial
Kidney Diseased (BKD) Chinook salmon caught in Fall and Spring In Lake Michigan
(1990-1992).

LIST OF REFERENCES

102
List of References
Black, E. C. 1958. Energy stores and metabolism in relation
to muscular activity in fishes. Page 51-67 in Larkin,
P. A . , editor. The investigation of fish-power
problems. H. R.

MacMillan lectures in Fisheries.

University of British Columbia. Ill pp.
Beacham, T. D. and T. P. T. Evelyn. 1992. Population and
Genetic Variation in Resistance of Chinook Salmon to
Vibriosis, Furunculosis, and Bacterial Kidney Disease.
Journal of Aquatic Animal Health 4:153-167.
Belding, D. L. and B. Merrill. 1935. A Preliminary report
upon a Hatchery Disease of the salmonidae.
T r a n s . Am. Fish. Soc. 65:76-84.
Bullock, G. L . , H. M.

Stuckey , and P. K. Chen. 1974 .

Corynebacterial Kidney Disease of Salmonids: Growth and
Serological studies on the Causative Bacterium.
Applied Microbiology 28(5):811-814.
Bullock, G. L . , and H. M. Stuckey.

1975. Fluorescent

antibody identification and detection of the
corynebacterium causing kidney disease of salmonids. J.
Fish. Res. Board Can. 32:2224-2227.
Burkhead, N. M. and J. D. Williams.

1991. An Intergeneric

Hybrid of a Native Minnow, the Golden Shiner, and an
Exotic Minnow, the Rudd. Trans. Am. Fish. Soc.
795.

120:781-

103
Chen, P. K . , G. L. Bullock, H. M. Stuckey, and A. C.
Bullock.

1974. Serological diagnosis of Corynebacterial

Kidney Disease of salmonids. J. Fish. Res. Board Canada
31:1939-1940.
Cipriano, R. C . , C. E . , Starliper, and J. H. Schachte.

1985.

Comparative sensitivities of diagnostic procedures used
to detect bacterial kidney disease in salmonid fishes.
Journal of Wildlife Diseases 21:211-218.
Dygert, P. H. 1990. Seasonal Changes in Energy Content and
Proximate Composition associated with Somatic Growth
and Reproduction in a Representative Age-Class of
Female English Sole. Transactions of the American
Fisheries Society.119:791-801.
Elliott, D. G . , R. J. Pascho, and G. L. Bullock.

1989.

Developments in the control of bacterial kidney disease
of salmonid fishes. Diseases of Aquatic Organisms
6:201-215.
Ecker, R. E . , and W. R. Lockhart.

1959. A rapid membrane

filter method for direct counts of microorganisms from
small samples. Journal of Bacteriology 77:173-176.
Fryer, J. L. and, J. E. Sanders.

1981. Bacterial Kidney

Disease of salmonid fish. Ann. Rev. Microbiol. 35:273298.
Gnanadesikan, R. 1977. Methods for statistical data analysis
of multivariate observations. John Wiley and Sons, New
York. 311 pp.

104

Hadley,

F. N. 1985. The adaptive role of lipids in

biological systems. John Wiley and Sons. New York,
Toronto.

Pages 149-193. 319 pp.

Hsu, H. M . , P. R. Bowser, and J. H . , Schachte, Jr. 1991.
Development and Evaluation of a Monoclonal-AntibodyBased Enzyme-Linked Immunosorbent Assay for the
Diagnosis of Renibacterium salmoninarum infection.
Journal of Aquatic Animal Health 3:168-175,

1991

Hay, R. L. 1989. Little Manistee River harvest weir and
Chinook salmon egg-take report. MDNR, Fisheries
division, Technical Report N° 89-5, July 1989.
Idler, D. R. and,

I . Bitners.

1960. Biochemical Studies on

Sockeye Salmon During Spawning Migration.

IX. Fat,

Protein and Water in the Major Internal Organs and
Cholesterol in the Liver and Gonads of the Standard
Fish. J. Fish. Res. Bd. Canada 17:113-122.
lies, T. D. 1984. Allocation of resources to gonad and soma
in Atlantic herring Clupea harengus L. Pages 331-347 in
G. W. Potts and R.J. Wootten, Editors. Fish
reproduction : Strategies and Tactics. Academic Press,
London. 410 pp.
Jezierska,

B . , J. R. Hazel, and S. D. Gerking.

1982. Lipid

mobilization during starvation in the rainbow tro u t ,
Salmon gairdneri Richardson, with attention to fatty
acids. J. Fish. Biol. 21:681-692.

105
Jungermann, k. , and N. Katz.

1986. Metabolism of

carbohydrates. Pages 211-235 in R. G. Thurman et al.
Regulation of hepatic metabolism.

Intra and

intercellular compartmentation. Plenum Press. 489 pp.
Keller, M. K . , K. D. Smith, and R. W. Rybicki.

1989. Summary

of salmon and Trout Management in Lake Michigan. MDNR,
Fisheries Division, Technical Report N° 89-1, March 17,
1989.
Kimura, T . , and M. Yoshimizu.

1981. A Coagglutination test

with antibody-sensitized staphylococci for rapid and
simple diagnosis of bacterial kidney disease

(BKD).

Development in Biological Standardization 49:219-224
Kinkelin, P. 1974. Corynebacteriose des salmonides: premiere
observation en France. Bulletin Frangais de
Pisciculture 254:3-7.
Laidler, L. A. 1980. Detection and identification of the
bacterial kidney disease (BKD) organism by the indirect
Fluorescent Antibody Technique.
Journal of Fish Diseases 3:67-69,

1980.

Love, A. M. 1970. The chemical biology of fishes. Academic
Press, London. 547 pp.
Mercer, H. P. 1980. Movement and harvest of Coho salmon in
Lake Michigan 1978-1979. MDNR, Fisheries Research
Report N° 1889, Dec. 15, 1980.
Morrison,

D. F. 1990. Multivariate statistical m e t h o d s .

Third Edition. McGraw-Hill Publishing Comp a n y . 495 pp.

106
Muzzal, P. M. 1988. Endohelxninths of salmonids from two
localities in eastern Lake Michigan, with emphasis on
Echinorhvnchus salmonis. Can. J. zool. 67:1604-1607.
Neff, N. A . , and G. R. Smith.

1979. Multivariate analysis of

hybrid fishes. Systematic zoology 28:176-196.
Pascho, R. J . , D. G. Elliott, R. W. Mallett, and S. Mulcahy.
1987. Comparison of Five Techniques for the Detection
of Renibacterium

salnoninarum in Adult Coho Salomon.

Transactions of the American Fisheries Society 116:
882-890.
Pascho, R . , and D. Mulcahy,

1987. Enzyme-1inked immuno­

sorbent assay for a soluble antigen of salmonid
bacterial kidney disease. Can. J. Fish. Aqa. Sees.
44:183-191.
Pesch, L. A. 1963. The liver and carbohydrate metabolism.
Pages 605-633 in CH. Rouiller. The liver Morphology,
Biochemistry, Physiology. Volume I . Academic Press, New
York and London.

683 PP.

Petersen, R. G. 1985. Design and Analysis of Experiments.
Marcel Dekker, Inc, New York . 429 PP.
Plumb, J. A. and P. R. Bowser.

1983. Microbial fish disease

laboratory m a n u a l . Brown Printing company, Montgomery,
Alabama.

95 PP.

Post, G. 1983. Textbook of fish health.
Inc. Ltd. 256 PP.

TFH Publications,

107
Rakoczy G. P. 1988. Sportfishing catch and effort from the
Michigan waters of Lakes Michigan, Huron, Superior and
Erie, and their important tributary streams. MDNR,
Fisheries Division, Technical Report N° 88-9a.
Roberts, R. J. 1978. Fish pathology. First edition. Pages
92-104. Baillere Tindall. 318 pp.
Sakai, M. N. A., S. Atsuta, and M. Kobayashi. 1987.
Comparative sensitivities of several dot blot methods
used to detect bacterial kidney disease.
Journal of Fish Diseases 10:229-231,
Sakai, M . , S. Atsuta, and M. Kobayashi.

1987.

1989. Comparison of

methods used to detect Renibacterium salmoninarum. the
causative agent of Bacterial Kiney Disease.
J. Aqua. Ani. Health 1:21-24,

1989.

Salz, R. J. 1989. Factors Influencing Growth and Survival of
Yellow perch from Saginaw Bay, Lake H u r o n . MDNR,
Fisheries Research Report N° 1964.
SAS Institute Inc. 1988. SAS Language Guide for Personal
Computers, Release 6.03 Edition. Cary, NC: SAS
Institute Inc. 558 PP.
Scott, W. B. and E. J. Crossman.

1973. Freshwater Fishes of

Canada. Fisheries Research Board of Canada, Bulletin
184. Pages 138-229. 966 pp.
Smith,

I . W. 1964. Freshwater salmon.

Fish. Res. 34:3-12.

108
Snieszko, S. F. and H. R. Axelrod.

1971. Diseases of Fishes.

Pages 94-104. T. F. H. Publications, Jersey City, New
Jersey. 41 PP.
Snieszko, S. E. 1973. Recent advances in scientific
knowledge and developments pertaining to diseases of
fishes. Adv. Vet. Sci. Comp. Med. 17: 291-314.
Stewart, D. J. and M. Ibarra.

1991. Predation and Production

by Salmonine Fishes in Lake Michigan,

1978-88. Can. J.

Fish. Aqu a t . Sci. 48:909-922.
Suzumoto, B. K . , C. B. Schreck and J. D. McIntyre.

1977.

Relative Resistances of three Transferrin Genotypes of
Coho salmon (Onchorhvnchus kisutch) and their
Hematological Responses to Bacterial Kidney Disease.
Journal of the Fisheries Research Board of Canada 34:18.
Tarver, H. 1951. The metabolism of amino acids and proteins.
Pages 769-908 in D. M. Greenberg, Editor. Amino Acids
and Proteins. C. C. Thomas, Springfield, Illinois. 950
pp.
__________. 1963. Metabolism of amino acids and proteins.
Pages 449-548 in CH. Rouiller. The liver Morphology,
Biochemistry, Physiology Volume I. Academic Press, New
York and London. 683 PP.

109
Turaga, P. S. D . , G. D. Wiens, and S. L. Kaattari.

1987.

Analysis of Renibacterium salmoninarum Antigen
Production in situ. Fish Pathology 22(4): 209-214, 1987
Valcarce, R . , D. Stevenson, G. G. Smith, and J. W. Sigler.
1991. Discriminant Separation of Fishes: A Preliminary
Study of Fish Eggs by Curie-Point Pyrolysis-Mass
Spectrometry-Chemometric Analysis. T r a n s . Am. Fish.
SOC.

120:796-802

Warren, J. W. 1983. Bacterial Kidney Disease. Pages 185-197.
In Meyer, F. P. et a l ., editors. A Guide to Integrated
Fish Health Management in The Great Lakes Basin. Great
Lakes Fishery Commission- Special Publication 83-2. 262
PPWeatherly, A. H. and H. S. G i l l . 1987. The biology of fish
growth. Academic Press, London. 443 pp.
Withler, R. E. and T. P. T. Evelyn.

1990. Genetic Variation

in Resistance to Bacterial Kidney Disease within and
between Two Strains of Coho Salmon from British
Columbia. Transaction of the American Fisheries Society
119:1003-1009