- .
.
. :-I
.
I- *7 I ‘r‘ AV II: I
“ I ‘1‘.“ . I. O
J! .5 .. . ii. . I... s.
I II I I . .IIA QIIII an. ,I 1.0. "I
. IIIInuJflI I.
‘ '4 ' 'I O. C
303;”...“1. II . :4 u
I III
.. .. . I.
.. . .
:0 a
. (.5...
'1! Y. O).

4". I '1 V.
.l . 1 TM a y
a. In: ¢ II I
I A "' ~ II;
I - . ufl \ . - , I ,
) - ‘ " . ‘ ‘n
.. c . . I I _ .IIIIV ‘
.- II - {III-luau, . v
I I . u v
\ I 6‘ (I ‘1'. u-
| “II. I r I , Util-
.

1 I IV . II: .. n .
, A I t ‘1. 0
I I: I “Kin“... . .rI! . :3:
I .l '0 i I. .
- - k I 4% {l 3:..-
- A I! I. I 4-I nu n Q - ‘4.
3. III! .1lIvIKII.‘ II.1a.I...quI.I .
3 14%» “fin“! I11. I WVIIW. Ilsrur...
.PNHUOI- 2m.

It!
.1
In

‘1' I‘ll; a
II.

 

o
.I
4 .IA
WM?
.0)
“huh" . u
. .? I I
I'lvr'l. I!“ .
I fluflgmhv) I III. . . . c I- 3
I IIIIIIIIIaulrIIIH u LII III.‘IIII|II I.....u.t.:..IIl II...“ I
.I.‘ v I! 1'50“ - C nulll.|ll . u ll'l‘lullvlol-iiv‘vt‘. A ' II I
319.0%? II‘II‘IIdI It‘ll... {III-III!!!“ .1. lIIIIIIIoII I. v\ o 7 I
|| '3‘!“ III!!! I il‘lll‘ltil! Dfillili‘)’ I I
I . [RE I II I I}-.. 1‘1...‘II4I§1’OIII]|I .. I1;
II‘lIIluVl I 1.4“”! II II I . .IHRHI‘IlliftlI (II-IIlII‘IIlulIII‘IIII...‘ in. I.”
.III. .hMHI.\ IIuI ISXII . 3 IIIBNDIIIIIIIIHIIII II IIII...‘||II .ITIHIInIKI .I III I g I. I
. ‘ . . IILI‘“ , “.CI at! Oll‘ *“Y.vlu"‘..llhlvili\lw...l..vll’lcvllo|lbulthl'tl . .10..“ III. I I. |.l.‘ .I I I {it 1 IAI I‘ll. ‘Mi
.I- l- ..IIIIII...-..IIIIH.“..VIIIIIII¥IIIIIIIIIIIIIII\I.IIHIIII1I IIIIIIIIIII'IIIIIIEIIIIIIHIIIIIIII.-. -IIIIIluIIIIIVIIIIfIIIII .. «.1 II IIIIIP I.”
1 no I II'- «I ‘i‘filIIlglII-LII II‘III‘.‘ III ”$1.6! I. ' u I
LII»! fI-l 7‘ I... III...» I‘Icvll‘lti'lr)‘ I Illvb.\~ 'IEIII“I I
L|.IIO.II|\I I E‘III .I‘II‘L.‘I.I\I101AI Io’-."i ’01.!‘I
. I. III .lclIIIIIZ . IIlI‘Illp‘I‘flhu‘lJILI.VIIIIIIIIIIIIIIII‘II‘1II\I}IIIIIII
IIIIIIIIIIII III tlumI‘..ILu’\IIIII1I\IIIIIflII )JII‘III'UII' '. III.‘.I.I\IIIII. IL
IN! - I‘I. II.“ l‘rv.‘ cl. III .IINIH‘IIUF\‘IIIIIIIIQ“|IIIW‘Ii {than
D ' - I‘D ‘Il‘l I‘ ! '1
. O llI km)! 0"}.%
1 . xl‘d'

                 

   

 

III
:IlIIIIIIIIuIIII
leIlIIIIII . (“9J1
III III , -
- I fl‘nl (Ill i“!
'IIIIIlII I “IIIIIIII [IIIIIIII fl.‘ 1
IIIIlIII‘lI (-01!“ Ill! .l'IIIIoIrIIIIIIIrI .IIIVII
I: III. III. .I HI. L?III|III..Ifl.IILII«IIIIII. H I...IIIIII. “III III
"‘1 lIIIHuD‘lIII gin I! IIILIIII. IIII”III “an,
I ‘I-I. 'tl! I !|
I leIlqun’IIlII - 3‘31»... ‘lchIIIPIIhII IIIIIIIIIIIIIII‘IIIIIII. ‘l ‘l‘ III";
IIIIUMILILMHI III! III III-3! I. -rII..|InlI IIIIIIHIWNNVFII‘LI [IRIII‘IIII III-III»!
I..|\ 1"! I‘il . l ‘1 I 1" l
.IIIIIII III .I.1 {IIIIIIIIIIIIIIIIIII IIIIIIIIIIIIII IIIIIIIIIIIIII. IIIIIIII II! I'lllllII II
.I . IIII. . IVIIIII lI.II..I.IIl.I3I .IIHIHIIIIIIIIIIIKIAIJanNII' II.IIII IIIIII‘IIIIII II “II .1
IL ‘IIIKIII I..III.QI.I‘|III.}I.I|.\III|!IVIIIIBIL{ I. {I If) III'I‘.
. I .I. ‘ III 2 I I. .IiIIIlIobil UPI II.I|I.7
‘II I illgu'gllio (I III! .
t .. . I . I {IIIIII III. .IIIHIIIIIIIIII.
. . llIIIoI-III'IIIII I .IIIIIIIIIIIIII
r.» I ‘iEII . III 'I‘It"
‘0 I f t.tl .
I- I I ,
a II. I II
I.

  
   

EH“
IIIIIII‘II‘I‘I-lll'll‘l
“ .‘ll‘lllltlil ‘ .l I.
IJI'II‘IL IIIIIII‘" .I
I‘IIIIIIUUIII..V . I , .V
I l‘!1‘ll . .- .‘IV I
. VA‘III‘olIII LI Ilv'tll I I»9I.
{I 3|- II I I'll...‘ I
. I \IIIIOI4II‘IIIILJII. Io . I
«4 ‘l‘lfll‘l‘- I‘ll I
.l'Io‘

 

 

I IIIIII III . IIIIIIII I. I
At. LOI’ I‘ll-'II'I HI‘I Il
9:3: III‘I'I‘TIIIII‘IIIIIII‘IIIIII‘III IIIIIII‘II.IIIIIIvk(.hNu”L IIIIIIIU
_ I .‘lIIIIIIIII‘... I'I I‘I'lnnkl‘l 'VI 1% ‘7‘!
II catfihfi'u‘ith‘rJ"; (4“ \llIuill ‘1! ('1’. \‘ ‘Illll't’v‘h. I.‘ l I II I
Hrcow .. quIIIIIIIIIIIIlllIII'i‘Ilo’IIlIIIIIIIIIIIIIIIIlIIIIII. IIIIIIIIIIIIunWUIDI InNUIIILIIIIIIIlIIIIII.I IIIIII. \IutI bur]. . I III IIIII'II .
.anI III. IIIAIIIIIIIIIIIII ‘IIIIIIIIHIIIIIIIIIII IIIIIII“ III. laIIIII IIIIIIIIIIIIIIIIII. IIII' po...II.III¢|( IUIIIIIIII: IIII..I«..I..II..I1II...uur.IIIII. .. 1.01? I.
no .i I II‘ 11' . II| "I‘IU‘ II--II.II‘P.I..\I‘I lItogI |.||I\.I\lll\l.9§1.$"~l‘llIl in. It“ .il‘ ‘1." 14“.

I \I ‘a‘Hl'I' | | I‘IIfIll' IH‘IIIIIIQ‘I.*I‘ED".I‘["’ ‘I‘I‘I-lnl .I‘fi“‘|a \I...~I 01‘.4k\|’|ll|b '11-'IJVII {*2
I '. ‘1 ‘l'ltr'l'l‘ll‘l '\ I III‘I . -IIIIIII‘IIIFUIA‘I‘QICIIIIIIQ.“ ”W‘IMIHNW 1.!!‘I‘IJUWI'LI.0..III .ndlII'HI.I.II1A.'0» I IE] I .
r J IIIIIII «fiIUHIIINI‘III-IIIIIIIIIIIIIIIIIII‘IQII IJIIIIII IIIIIIIHEIIIIHN H ,I. ,HJIHIII‘IIIIIIIIIIIII” III mlI: IIIIMHF: IIIIIIIII . ”III- IIIIIDnIIII I1]IIII|. IIIII . |
$9.11.. IIIIIIIIIIIHIIIIIII is. I {III I..\.II.IIJII IInHIIIII'II .m..\.. .I- III .HIIIIIIIIIIIIIIIFIIIILII .HIIIIIIIIII LIIIEI.

IIIJIKIIIIISIIIII‘ II nLMIolclINlIlvOtIIIINIMWIlIIIV‘IILHHMII" ' II Ii‘.| -I I
1 . [fli‘l Q!" . I '.I I I‘ll

51911111.? ,I .IIIuIIlI‘IIIIhHIII I. (1| Illlll
OI I‘lbvi‘lool -T‘I‘I’!“|I.II.III.UII"‘)IV 5"' I’ll
. I111!" ‘11, IO ‘III‘I'.
I‘ll! I \il‘fl’l' II I .
IIIIIIIIIIII'I

 

l

 

2IIIIIIIIIIIIII I,
U : _:_¥_:¥ ‘

a) 145:3
IVUI‘IIU",}¢-- Q-v—wz.’¢-

I. .2 o...
L: “I snaffa:£_ “I

II. w -l‘,

 

 

 

This is to certify that the

thesis entitled

I_N. VITRO MITOGENESIS OF PERIPHERAL BLOOD

LYMPHOCYTES FROM RAINBOW TROUT (Salmo gairdneri)

 

presented by

Donald Edward Tillitt

has been accepted towards fulfillment
of the requirements for

M.s-_..__degree in 4.1M 'ld.

 

Major professor

Date 7/25/86

 

0-7539 MS U is an Aflirrnative Action/Equal Opportunity Institution

 

 

 

RETURNING MATERIALS:
‘IV1ESI_] Place in book drop to
[ABRARJES remove this checkout from

_g=5!C!!-._ your record. FINES wiII
be charged if book is

 

 

returned after the date
stamped beIow;

 

 

 

 

 

 

 

lg yrrgg MITOGENESIS or PERIPHERAL BLOOD
LYMPHOCYTES FROM RAINBOW TROUT (§alm9.zainoneri)

By

Donald Edward Tillitt

A THESIS

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

MASTER OF SCIENCE

Department of Fisheries and Wildlife

1986

ABSTRACT
13 yIIRO MITOGENESIS or PERIPHERAL BLOOD
LYMPHOCYTES FROM RAINBOW TROUT (§alm9 zaizdnerJ
By
Donald Edward Tillitt

_I_n m mitogenesis of rainbow trout peripheral blood
lymphocytes (RBT PBL) was investigated to define the Optimal
culture conditions and repeatability of the assay for routine
laboratory use. The assay variables of media, mitogen, serum
supplementation, lymphocyte isolation procedure, and incuba-
tion period were assessed. Optimal proliferative response
was obtained when 138'! P81. were cultured in RPMI 161m supple-
mented with 10% fetal bovine serum and stimulated with 10 ug
Concanavalin A/ml for between four and five days. I observed
statistically significant variation among fish.‘ Power analysis
with variance estimates from this study reveal that sample size
requirements of further studies under the given conditions oould
severely limit the applicability of this procedure for RBT
health assessment. IFurther work in this area should center
around standardization of culture conditions pertaining to the

source of protein supplementation.

 

To my parents, Bart and Marvelle

ii

ACKNOWLEDGEMENTS

I would like to express my gratitude to my major profes-
sor, Dr. John Giesy, and the other members of my guidance com-
mittee, Dr. P.(L Fromm and Dr. Monte Mayes, for their support
during the course of this study.

There are also a number of people to whom I would 1 ike
to give special thanks for technical assistance. Dr. Robert
Bull performed screening tests on ConA; Dr. Louis King for
the assistance and use of the flowcytometer; Ken Weber, owner
of Green River Trout Farm, for his kind donation of rainbow
trout serum; Harry Westors, of the Michigan Department of
Natural Resources, for permission to collect salmon sera and
steelhead from the Little Manistee River; the Michigan State
University Clinical Center for donation of human sera; and
Iha Robert Ringer for use of his automated cell harvester.
Lastly; I would like to thank Christine Flaga and Elizabeth
Bartels for preparation of this thesis.

iii

 

 

TABLE

INTRODUCTION. . . . . .
MATERIALS AND METHODS . .
Fish . . . . . . . . .
Media. . . . . . . . .
Serum Supplements. . .
Mitogens . . . . . . .
RESULTS . . . . . . . . . .
Media. . . . . . . . .
Mitogens . . . . . . .
Serum Supplement . . .
Procedural Variability
Fish Source. . . . . .
DISCUSSION. . . . . . . . .
Culture Conditions . .
Variation and Sample 8
SUMMARY AND CONCLUSIONS . .
APPENDIX A. . . . . . . . .
APPENDIX B. . . . . . . . .
LIST OF REFERENCES. . . . .

OF CONTENTS

ize. I O O 0

iv

Page

«JO‘GO

19
19
3a
37
1:2
“3
“3
“£3
52
56
57’

100

LIST OF TABLES

Table Page
1 RPMI 16u0 Tissue Culture Media and Supplements. 1n
2 Medium 199 Tissue Culture Media and Supplements 15
3 Mean Uptake of 3H-Thymidine by RBT PBL Cul-

tured in RPMI 1640 With 10% FBS . . . . . . . . 20

u Mean, Standard Error, Standard Deviation,
Variance, Range, and Coefficient of Variation
of the Response of RBT PBL to ConA Across All
Experiments . . . . . . . . . . . . . . . . . . 22

5 Mean, Standard Deviation, Variance, Range, and
Coefficient of Variation of the Response of
Individual RBT PBL to 10 ug ConA/ml . . . . . . 23

6 Mean and Range of Responses of Four Subsamples
of PBL from Individual RBT to ConA. . . . . . . H1

7 Sample Sizes Required to Demonstrate Reductions
in Immunocompetence of RBT PBL Respnses to 10 ug
ConA/ml With an Incubation Period of u Days . . 51

A1 ConA Screening with Human Lymphocytes from Pro-
spective Organ Donors . . . . . . . . . . . . . 56

BI Model, Formula, and Coefficients of Variation of
Expected Mean Squares for the 3-Way ANOVA of 3H-
Thymidine Incorporation into Cultured RBT PBL in
Experiment 18 . . . . . . . . . . . . . . . . . 57

82 Estimate of Variance, Partition of Variance, and
3-Way ANOVA of DPM Response from RBT PBL Cultured
in Experiment 18. . . . . . . . . . . . . . . . 58

B3 Estimate of Variance, Partition of Variance, and
3-Way ANOVA of SI Response from RBT PBL Cultured
in Experiment 18. O O O O O O O O 0 I O O O O O 59

En Model, Formula, and Coefficients of Variation of
Expected Mean Squares for the 3-Way ANOVA of 3H-
Thymidine Incorporation into Cultured RBT PBL in
Experiment 20 . . . . . . . . . . . . . . . . . 60

BS Estimate of Variance, Partition of Variance, and
3-Way ANOVA of DPM Response from RBT PBL Cultured
in Experiment 20. . . . . . . . . . . . . . . . 61

LIST OF TABLES (cont.)
Table Page

B6 Estimate of Variance, Partition of Variance, and
3-Way ANOVA of SI Response from RBT PBL Cultured
in Experiment 20. O O I C I O O I O O I O O O O 62

B7 Model, Formula, and Coefficients of Variation of
Expected Mean Squares for the 2-Way ANOVA of 3H-
Thymidine Incorporation into Cultured RBT PBL in
Experiment 21 . . . . . . . . . . . . . . . . . 63

B8 Estimate of Variance, Partition of Variance, and
2-Way ANOVA of DPM Response from RBT PBL Cultured
in Experiment 21 O O O O O O O I O O O O O O O 0 6n

B9 Estimate of Variance, Partition of Variance, and
2-Way ANOVA of SI Response from RBT PBL Cultured
in Experiment 21. O O O O O O O O I O I O I O 0 65

B10 Model, Formula, and Coefficients of Variation of
Expected Mean Squares for the 2-Way ANOVA of 3H-
Thymidine Incorporation into Cultured RBT PBL in
Experiment 22 0 O O O O O O D O I O O 0 O O I O 66

811 Estimate of Variance, Partition of Variance, and
2-Way ANOVA of DPM Response from RBT PBL Cultured
in Experiment 22. . . . . . . . . . . . . . . . 67

B12 Estimate of Variance, Partition of Variance, and
2-Way ANOVA of SI Response from RBT PBL Cultured
in Experiment 22. O O I O O I O O O I I O O O I 68

B13 Model, Formula, and Coefficients of Variation of
Expected Mean Squares for the 3-Way ANOVA of 3H-
Thymidine Incorporation into Cultured RBT PBL in
Experiment 23 . . . . . . . . . . . . . . . . . 69

B1u Estimate of Variance, Partition of Variance, and
3-Way ANOVA of DMP Response from RBT PBL Cultured
in Experiment 23. . . . . . . . . . . . . . . . 70

B15 Estimate of Variance, Partition of Variance, and
3-Way ANOVA of SI Response from RBT PBL Cultured
in Experiment 23. . . . . . . . . . . . . . . . 71

B16 Model, Formula, and Coefficients of Variation of
Expected Mean Squares for the 3-Way ANOVA of 3H-
Thymidine Incorporation into Cultured RBT PBL in
Experiment 26 . . . . . . . . . . . . . . . . . 72

LIST OF TABLES (cont.)

Table
B17

B18

819

820

821

822

823

82“

825

826

827

828

Page

Estimate of Variance, Partition of Variance, and
3-Way ANOVA of DPM Response from RBT PBL Cultured
in Experiment 26. I O O O O l O O I O O O O O I 73

Estimate of Variance, Partition of Variance, and
3-Way ANOVA of SI Response from RBT PBL Cultured
in Experiment 26. . . . . . . . . . . . . . . . 7“

Model, Formula, and Coefficients of Variation of
Expected Mean Squares for the 3-Way ANOVA of 3H-
Thymidine Incorporation into RBT PBL Cultured with
Human Sera in Experiment 28 . . . . . . . . . . 75

Estimate of Variance, Partition of Variance, and
3-Way ANOVA of DPM Response from RBT PBL Cultured
with Human Sera in Experiment 28. . . . . . . . 76

Estimate of Variance, Partition of Variance, and
3-Way ANOVA of SI Response from RBT PBL Cultured
with Human Sera in Experiment 28. . . . . . . . 77

Model, Formula, and Coefficients of Variation of
Expected Mean Squares for the 2-Way ANOVA of 3H-
Thymidine Incorporation into Cultured RBT PBL Ex-
posed to ConA or LPS in Experiment 29 . . . . . 78

Estimate of Variance, Partition of Variance, and
2-Way ANOVA of DPM Response from RBT PBL Cultured
with ConA in Experiment 29. . . . . . . . . . . 79

Estimate of Variance, Partition of Variance, and
2-Way ANOVA of SI Response from RBT PBL Cultured
with ConA in Experiment 29. . . . . . . . . . . 80

Estimate of Variance, Partition of Variance, and
2-Way ANOVA of DPM Response from RBT PBL Cultured
with LPS in Experiment 29 . . . . . . . . . . . 81

Estimate of Variance, Partition of Variance, and
2-Way ANOVA of SI Response from RBT PBL Cultured
with LPS in Experiment 29 . . . . . . . . . . . 82

Model, Formula, and Coefficients of Variation of
Expected Mean Squares for the 2-Way ANOVA of 3H-
Thymidine Incorporation into Cultured RBT PBL Ex-
posed to PWM or PHA in Experiments 29 and 30. . 83

Estimate of Variance, Partition of Variance, and

2-Way ANOVA of DPM Response from RBT PBL Cultured
with FHA in Experiment 29 . . . . . . . . . . . 8n

vii

LIST OF TABLES (cont.)

Table

829

830

831

B32

333

83”

835

B36

B37

B38

839

8H0

Estimate of
2-Way ANOVA
with FHA in

Estimate of
2-Way ANOVA
with PWM in

Estimate of
2-Way ANOVA
with PWM in

, Estimate of

2-Way ANOVA
with PHA in

Estimate of
2-Way ANOVA
with FHA in

Estimate of
2-Way ANOVA
with PWM in

Estimate of
2-Way ANOVA
with PWM in

Page

Variance, Partition of Variance, and
of SI Response from RBT PBL Cultured
Experiment 29 . . . . . . . . . . . 85

Variance, Partition of Variance, and
of DPM Response from RBT PBL Cultured
Experiment 29 . . . . . . . . . . . 86

Variance, Partition of Variance, and
of SI Response from RBT PBL Cultured
Experiment 29 . . . . . . . . . . . 87

Variance, Partition of Variance, and
of DPM Response from RBT PBL Cultured
Experiment 30 . . . . . . . . . . . 88

Variance, Partition of Variance, and
of SI Response from RBT PBL Cultured
Experiment 30 O O O O I O O I O O O 89

Variance, Partition of Variance, and
of DPM Response from RBT PBL Cultured
Experiment 30 . . . . . . . . . . . 90

Variance, Partition of Variance, and
of SI Response from RBT PBL Cultured
Experiment 30 . . . . . . . . . . . 91

Model, Formula, and Coefficients of Variation of
Expected Mean Squares for the 2-Way ANOVA of 3H-
Thymidine Incorporation into Cultured RBT PBL Ex-
posed to ConA or LPS in Experiment 30 . . . . . 92

Estimate of
2-Way ANOVA

Variance, Partition of Variance, and
of DPM Response from RBT PBL Cultured

with ConA in Experiment 30. . . . . . . . . . . 93

Estimate of
2-Way ANOVA

Variance, Partition of Variance, and
of SI Response from RBT PBL Cultured

with ConA in Experiment 30. . . . . . . . . . . 9n

Estimate of
2-Way ANOVA
with LPS in

Estimate of
2-Way ANOVA
with LPS in

Variance, Partition of Variance, and
of DPM Response from RBT PBL Cultured
Experiment 30 . . . . . . . . . . . 95

Variance, Partition of Variance, and

of SI Response from RBT PBL Cultured
Experiment 30 O O I O O I O O O O O 96

\dii

LIST OF TABLES (cont.)

Table
Bu1

Bu2

13:13

Page

Model, Formula, and Coefficients of Variation of
Expected Mean Squares for the 3—Way ANOVA of 3H-
Thymidine Incorporation into RBT PBL Cultured with
Human Sera in Experiment 31 . . . . . . . . . . 97

Estimate of Variance, Partition of Variance, and
3-Way ANOVA of DPM Response from RBT PBL Cultured
with Human Sera in Experiment 31. . . . . . . . 98

Estimate of Variance, Partition of Variance, and
3-Way ANOVA of SI Response from RBT PBL Cultured
with Human Sera in Experiment 31. . . . . . . . 99

Figure

10

11

12

13

LIST OF FIGURES

Cytogram of RBT PBL Separated on Ficol-Paque
Gradient Material . . . . . . ... . . . . . . .

Cytogram of RBT PBL Separated on Percoll
Gradient Material I I I I I I I I I I I I I I I

Kinetics of Response of Fish 1 PBL to ConA in
Experiment 18 . . . . . . . . . . . . . . . . .

Kinetics of Response of Fish 2 P81 to ConA in
Experiment 18 I I I I I I I I I I I I I I I I I

Kinetics of Response of Fish 21 PBL to ConA in
Experiment 20 I I I I I I I I I I I I I I I I I

Kinetics of Response of Fish 24 PBL to ConA in
Experiment 20 I I I I I I I I I I I I I I I I I

Mean Response of RBT PBL to ConA Across Fish in
Experiment 23 I I I I I I I I I I I I I I I I I

Coefficients of Variation of the SI Response

Variable for Means Across Fish in Experiment 23.

Coefficients of Variation of the SI Response

Variable of Across Fish and Experiment Means for

RBT PBL Stimulated with ConA. . . . . . . . . .

Mean Response (DPM) of RBT PBL to ConA with 10%
Human Sera Supplementation in Experiment 28 ..

Mean Response (DPM) of RBT PBL to ConA with 10%
Human Sera Supplementation in Experiment 31 . .

Mean Response (DPM) of RBT PBL to ConA with 10%
Human Sera Supplementation Across Experiment 31
and 28I I I I I I I I I I I I I I I I I I I I I

Coefficients of Variation for DPM Response of

Page

12

13

25

26

27

28

3O

31

33

35

36

38

Across Fish and Experiment Means of RBT PBL Sti-

mulated with ConA and Media Supplemented with
10’ Human SeraI I I I I I I I I I I I I I I I I

39

INTRODUCTION

The importance of hematology and immunology in health
assessment of higher vertebrates has lead workers to adOpt a
variety of hematological and immunological techniques for fish
health assessment (Hesser, 1960; Blaxhall, 1972; Hickey, 1976;
Hedemeyer and Yasutake, 1977). Techniques adopted are classi-
cally descriptive in nature, which stress numbers and sizes of
fish blood cells. Examples of descriptive techniques include:
use of hematocrit (Soivio and Oikari, 1976; Munkittrick and
Leatherland, 1983), hemoglobin (Sniesko, 1960), leukocyte counts
or leucocrit (McLeay and Gordon, 1977; Wedemeyer et al., 1983),
chromosomal aberrations (Al-Sabti, 1985), macrophage aggregates
(Holke et al., 1985) or combinations of these parameters.

Reports of the utilization of functional immunological
tests for the evaluation of fish health are notably absent from
the literature in the area of aquatic toxicology. In their
review of reports dealing with the effects of toxic agents on
the immune systems of fish Zeeman and Brindley (1981) concluded
that such reports comprised merely sidelights and footnotes in
the literature. This is not the situation in mammalian toxi-
cology where a variety of functional studies have been recommen-
ded for routine screening for suppression of the immune system
(Vos, 1977; Koller, 1979; Vos, 1981; Sharma, 1981; Bleavins and
Aulrich, 1983). V03 (1977 and 1981) suggested tests for the
immunological functions of cell-mediated immunity, humoral immu-

nity, and phagocytoses by macrophages.

Cellular immune function is an important component of
both cell-mediated and humoral immunity and is crucial in
assessing immunocompetence of an organism. Lymphocyte proli-
feration or stimulation assay, commonly referred to as lym-
phocyte activation (LA), is an extremely useful test of
cellular immune function. .In7xjgrg LA techniques are used in
human medicine to assess cellular immunity in cases of immuno-
deficiency, autoimmunity, infectious diseases, and cancer
(Stites et al., 1982). LA techniques have also been utilized to
monitor the immunosuppressive activity of toxic compounds in
mammals (Vos and Moore, 197a; Bleavins et al., 1983; Greenlee et
al., 1985).

LA is the morphological and functional alteration process
which occurs when immunocompetent lymphocytes are stimulated by
antigens or nonspecific mitogens. This transformation occurs in
.1119_when lymphocytes are presented with properly processed
antigens. The site of antigen presentation may be in the sys-
temic circulatory system but more commonly occurs in mammalian
lymph nodes (Spent, 1977) or the pronephros of teleosts (Chiller
et al., 1969; Smith et al., 1970). These centers have concen-
trated numbers of lymphocytes in close proximity to one another
so as to increase the likelihood of an antigen-presenting macro-
phage associating with a lymphocyte that has surface membrane
immunoglobin receptors for that particular antigen. This trans-
formation process occurs in 11519 when lymphocytes are cultured
with either specific antigens or, more commonly, with nonspeci-

fic mitogens (mitogenesis). Mitogens are: 1) lectins derived

from various plants or 2) polysaccarides from bacterial cell
coats. They act to nonspecifically stimulate lymphocytes,
without the requirement of a sensitized host.

A myriad of biochemical events occur in the lymphocyte
upon activation by the mitogen. These events include changes in
lipid components of the plasma membrane, increased permeability
to divalent cations, adenylate cyclase and guanylate cyclase
activation and resultant elevation of intracellular cAMP and
cGMP in early and late phases, respectively. Synthesis of
protein, RNA and shortly thereafter DNA occurs in activated
cells (Hadden, 1981). The DNA synthesis is the basis for
measuring cell proliferation. Morphologically, proliferating
cells enlarge, form large pyroninophilic vesicles, fill with
endoplasmic reticulum, polysomes, free ribosomes, and have
marked increases in microtubule development. These changes give
the cells the appearance of primordial blastlike cells from
which the term blastogenesis is derived. The biochemical and
morphological changes are a prelude to the production of anti-
bodies in the case of B-cells or the synthesis soluble factors
(iueL, lymphokines, prostaglandins) and cell-mediated activities
with T-cells (Hume and Weideman, 1980).

A number of fish species have the ability to produce
lymphocytes that respond to LA techniques (Ethlinger et al.,
1976a; Cuchens and Clem, 1977; Sigel et al., 1978; Al-Sabti,
1983; Blaxhall, 1983b; Faulman et al., 1983; Narr and Simon,
1983; Caspi et al., 198“). Cells separated from lymphoid tis-

sues (spleen, thymus, and anterior kidney) and peripheral blood
of fish have produced various levels of response to the same
mitogens used in higher vertebrates. Authors of the above
mentioned studies were attempting to establish lymphocyte
heterogeneity in fish or develop methods for karyotyping fish.
Lymphocytes from teleosts, although not as clearly defin-
able as those from higher vertebrates, appear to have discern-
ible subpopulations resembling T and B-cells (Clem et al., 1981;
Warr and Simon, 1983). Partial evidence for this fact comes
from differential responsiveness to "classicalJ'T-cell mitogens,
phytohemagglutinin (PHA) and concanavalin A (ConA), as compared
to B-cell mitogens lipopolysaccharide (LPS) and purified protein
derivative of tuberculin (PPD) (Etlinger et al., 1976; Cuchens
and Clem, 1977; Harr and Simon, 1983). Although in disagreement
about the exact nature of tissue-specific mitogenic responsive-
ness, all of their studies demonstrated some degree of response
to "T-cell" and "B-cell" type mitogens by cultures of lympho-
cytes which were prepared from peripheral blood samples. This
is not unexpected based on the role of the circulatory system in
transporting lymphoid cells during the ontogeny of cellular
immunity in young fish and in adults (Tatner and Manning, 1983).
Adapting lymphocyte mitogenesis assay for routine use with
fish in the laboratory or in the field, could provide a useful
tool to workers interested in fish health. Such a tool could be
used to assess the status of the immune system during immuno-
toxicological screening tests, in fish culturing systems, and in

wild populations. To this end, my research was aimed at evalu-

ating the applicability of mitogenic responses of lymphocytes as
a routine tool for fish health assessment. To address this
problem, I conducted lymphocyte mitogensis assays on cells
separated from peripheral blood of rainbow trout. The para-
meters of incubation time, culture media, serum supplement,
mitogen type and concentration were evaluated to maximize the
proliferative response.

The specific objectives of my stidues were to:

1. Optimize in yjtrg culture parameters of culture
media, serum supplement, mitogen type, mitogen
concentration and incubation time for peripheral
blood lymphocyte mitogenesisassay in rainbow
trout.

2. Test and define the repeatability of the assay

as a tool for fish health assessment-

MATERIALS AND METHODS

Ejsh

The rainbow trout (Salmg gaicgnezi), RBT, was chosen as
the experimental organism because it is a common freshwater fish
about which an abundance of physiological, hematological and
husbandry information is available. Previous studies have also
demonstrated that lymphocytes from rainbow trout are capable of
mitogenic transformation (Etlinger et al., 1976a; Harr and
Simon, 1983). Rainbow trout of both sexes, 150-250g, were
purchased from Balders Fish Farming Enterprise, Big Rapids,
Michigan. Fish were held in 500 l fiberglass tanks, at 10 11
C, with a continuous flow (5 turnovers/day) of activated char-
coal filtered, aerated tap water. Twice a week the tanks were
cleaned and the fish were fed (Purina Trout Chow) to satiation.

The photoperiod was 16 h light, 8 h dark.

flsflla

Media were prepared fresh for each assay, generally 28
hours prior to use. RPMI 16u0 (Gibco Laboratoreis, Grand Is-
land, New York, Catalog No. 330-2511) with L-glutamine was
prepared from 10x liquid concentrate and supplemented according
to the method of Warr and Simon (1983, see Table 1). Medium 199
(Gibco Laboratories, Catalog No. 330-1181) with Hank's balanced
salts and L-glutamine was prepared from 10x liquid concentrate
and supplemented according to Blaxhall (1983a) with modifica-
tions as noted in Table 2. Media and supplements were diluted

to volume with double distilled water. pH was adjusted to 7.3

with 25% NaOH and monitored by an Electro-Mark pH meter (Markson
Science Inc., Del Mar, California). Osmolality of the media was
checked by a 5100 B vapor pressure osmometer (Wescor Inc.,
Logan, Utah) and adjusted to 312 m0s/Kg with NaCl. Media were
then sterilized by filtration with Type TC filter units (Nalgene
Co., Rochester, New York) which were equipped with 0.2 micron

membrane filters and stored at 2-8 C.

MW
Fetal bovine serum (FBS) and heat inactivated FBS

(Gibco Laboratories, Control Nos. 29P883u and 29K9051, re-
spectively) and human serum (HS) obtained from the Michigan
State University Clinical Center were tested for their abili-
ty to support growth and mitogenic stimulation of RBT peri-
pheral blood lymphocytes. Serum that was not already heat
inactivated was treated at 56 C for 30 minutes to destroy the
C3 component of the complement system. All serum was stored
at -20 C. Serum was added to media before filter steriliza-

tion.

Eiigzsns

Concanavalin A, (ConA; Pharmacia Fine Chemicals, Uppsala,
Sweden, Lot KA 35107), was obtained freeze dried, prepared by
chromatography on Sephadex with less than 0.T$ carbohydrate.
Lipopolysaccharide, LPS (Escherichia £211 0111:3u-w, Control No.
721935 and E; 9911,055:BS-W, Control No. 725525), pokeweed
mitogen, PWM, (Control No. 13N7932), and phytohemagglutinin-P,
PHA-P (Control No. 715136), were obtained in lyopholized form

from Difco Laboratories, Detroit, Michigan. All mitogens were
rehydrated with sterile double distilled water and stored at -20
C. Dilutions of each of the mitogens were made with the appro-
priate media just prior to use.

ConA was screened for its ability to stimulate human
lymphocytes bthn Robert Bull, College of Veterinary Medi-
cine, Michigan State University. Stimulation was noted from
all four individuals tested (Appendix A1).

Iégléilgn 21.Lxm2h92x&£§

Lymphocytes were separated from peripheral blood by a
modification of Boyum's (1968) method. Numerous variations of
dilution volume, centrifuge time and cell suspension to gradient
material ratios were tested. The purest separations resulted
from the following procedure. Whole blood samples were diluted
1:3 with cold, complete medium to reduce viscosity. Two milli-
liters of whole blood suspension was carefully'layered over u ml
Ficoll-Paque (Pharmacia Fine Chemicals, Piscataway, New Jersey)
in a 17 x 100 mm polystyrene tube (Falcon, Oxnard, California)
so as not to mix the two phases. The gradients were centrifuged
at 2000 RPM (500 x gravity) in an International Centrifuge Model
SBV (International Equipment, Boston, Massachusetts) with a
swing-bucket rotor at 10 C for 30 minutes. ‘The overlying medium
was removed by aspiration and the lymphocytes at the interface
collected with a polyethylene transfer pipette. A typical blood
sample was 3 ml whole blood, with a resultant 12 ml dilution

volume; Therefore, six gradients were required for a single

blood sample (2 ml/gradientL. The lymphocytes from these gra-
dients were collected and placed in another 17 x 100 mm tube.
Cells were washed twice with cold, complete medium at 100 x
gravity (700 RPM in the International Model SBV), 10 C, for 10
minutes. Cells were enumerated using an Improved Neubauer hema-
cytometer (Arthur H. Thomas Co., Philadelphia, Pennsylvania) as
described by Absher (1973L. Determination of cell viability by
trypan blue exclusion “Lu! trypan blue in 1% NaCl) was per-
formed concurrently (Phillips, 1973).

Alternatively, lymphocyte isolation techniques described
by Harr and Simon (1983) were assessed. This approach employs a
continuous gradient consisting of Percoll (Pharmacia Fine Chemi-
cals, Piscataway, New Jersey),£L5 M sucrose, and distilled
water in a 3:1:6 ratio. The sterile solutions were mixed and 8
ml placed in a 16 x 116 mm polycarbonate tube. .The gradient was
established by centrifuging for 35 minutes in a Sorvall RC-2
with a ss-2u angular head rotor at 20,000 RPM (u9,000 x gravity)
and 20 C. A 1 ml aliquot of 1:1 dilution of whole blood and
complete media was carefully layered over the Percoll gradient
and centrifuged for 8 minutes at 10 C in an International Cen-
trifuge Model SBV with a swing-bucket rotor at 2000 RPM (500 x
gravityL. The band of lymphocytes was collected, transfered to
a 17 x 100 mm polystyrene tube and washed twice with cold,
complete medium as in the Ficoll-Paque procedure.

Flow cytometric and histological staining examination of

the separated cells were done to characterize the population of

10

cells obtained from these procedures. Cells were stained with a
Ralphlsawright-Giesma procedure (Luna, 1968) and viewed under a
light microscope. The cirterion for cell classification was as
described by Ellis (1977) and Yasutake and Walles (1983L. Cell
collected from either of these procedures were composed of over
90% lymphocytes by morphologic criterion; the remaining cells
were an approximately even distribution of thrombocytes, mono-
cytes, and granular leukocytes. Red blood cells (RBC) were
generally less than 51 of the total cells counted. Separations
containing more than 5% RBC were discarded.

Flow cytometry was done on a Cytofluorograf equipped with
a 2150 Ortho Diagnostic Systems (Nestwood, Massachusetts) compu-
ter. Cells separated by both methods (Ficol-Paque and Percoll
gradients) were analyzed for their forward red light scatter (a
measure of cell size) and 90-degree blue light scatter (a
measure of cell granularity). Data is expressed as relative
light intensity (for each form of light scatter) and relative
frequency of cells. Cytograms from both methods (Figures 1 and
2) show very similar patterns of 90 degree blue light scatter
for both light intensity range and frequency of cells. The
pattern of forward red light scatter for Ficol-Paque separated
cells had a distinct peak narrower than that of Percoll separ-
ated cells.

The cytogram of forward red light scatter of Percoll
separated cells has more of a bell shape, but the boundaries
are similar to the Ficol-Paque separated cells. The cyto-

grams of red-light versus blue-light-scatter further shows

11

the similarity in cell populations attained by these two
gradient separation methods as well as the homogeneity of

size and internal granulation within each of the populations.

Assay Procedure

Fish were quickly netted from holding tanks and immobi-
lized by a blow to the head. This method is preferred over
anaesthesia for giving samples representative of resting
state (Oikari and Soivio, 1975). The ventral side of the
peduncle, just posterior to the anal fin, was topically
disinfected with 95% ethanol. Peripheral blood was taken
into a heparinized syringe (approximately 100 units/ml final
concentrationL. Blood samples were diluted with cold, com-
plete medium and lymphocytes isolated as previously discussed.
After enumeration and viability testing, cells were resuspen-
ded in complete medium to the desired concentration. All
lymphocyte suspensions contained 10,000 viable cells/ml un-
less otherwise noted.

Cells were cultured in 96-well microculture plates
(Falcon, Becton Dickinson and Co., Oxnard, California). Cells
were delivered to culture wells in a 50 ul volume of complete
medium that resulted in a cell concentration of 5 x105
cells/well. In certain cases the initial seeding density of
cells was less than 5 x 105 cells/well due to low recovery rates
from separation. Mitogens at the desired dilution were also

delivered to culture wells in a 50 ul volume of complete medium,

 

12

 

  

2II 4II III III IIII 2II III III III lIII
N2 PHI RED SCITIER NI 9I' ILUE SCIYIER

RIINIIH TROUT FlCIL-PIIUE
SEPIRI'EI

 

 

 

2'. (I'GYIYI'I'III
II raw-sc/v vs 90'-sczx

Figure 1. Cytogram of RBT PBL Separated on Ficol-Paque Gra-

dient Material.

l3

 

 

     

III IIII 2II III III III [III

 

 

 

200 400 see
uz run RED scattsa MI 90' ILUE scattsa
use. .
so. RBINIOH tnour
PERCOLL SRRDIENI sernaatea
sea ‘
4 ,gt$*l?
40+ '
20-
' ' f ' V V v’ ' V V ‘
I 20 40 so so no.

II FIN-SCIY VS II'-SCIX

Figure 2. Cytogram of RBT PBL Separated on Percoll Gradient
Material.

14

Table 1. RPMI 16u0 Tissue Culture Medium and Supplements.

 

W W
RPMI 16u0 (10x) 10% (V/V)
Penicillin-G 105 units/l
Streptomycin sulfate 50 mg/l
2-Mecaptoethanol 50 uM

Sodium bicarbonate 25 uM

15

Table 2. Medium 199 Tissue Culture Medium and Supplements.

 

Sunalgmgnt QQDSEBLLQLJQD
Medium 199 (10x) 10% (V/V)
Penicillin-G 2 x 105 units/l
Streptomycin sulfate 105 units/l
Nystatin 2.5 x 10" units/1
HEPES buffer (1) 5 uM
2-Mercaptoethanol 50 uM

 

(1) N-2-Hydroxyethylpiperczine-N'-2-ethanesulfonic acid, so-
dium salt.

16

resulting in a final culture volume of 100 ul. Cultures of each
mitogen concentration were conducted in triplicate.

Microculture plates containing cells were incubated in
an atmosphere of 95% air and 5% C02 at 20 C. Growth kinetics
were studied by varying the period of incubation from 1 to 7
days. Proliferation of the lymphocytes was monitored by the
uptake and incorporation of [methyl-3HJ-thymidine (ICN, Ir-
vine, California, specific activity'6.7 Ci/mmole, Lots
2235129 and 2618119, >991 purity) into DNA. Each culture
well was dosed with 1 uCi [methyl-3H1-thymidine in 25 ul of
complete media 2n h before harvesting. Etlinger et al.

(1976a) demonstrated that radioactivity incorporated in this
method was a valid measure of cellular proliferation.

Cell cultures were harvested using a multiple automated
sample harvester (MASH) that was constructed specifically for
Dr. R. Ringer, Animal Science Department, Michigan State
University, and is not available commercially. MASH units
were first introduced by Hartzman and coworkers (1971, 1972)
to simplify harvesting in conjunction with microculture sys-
tems. Together, the microculture plates and MASH units af-
forded a simple technique that required fewer cells and allowed
more replicate cultures. The MASH unit used in these assays
harvested 2n culture wells simultaneously. Culture cells were
rinsed with distilled water. Cells that were not disrupted by
the suction action of the MASH were lysed by the distilled
water. Cellular fragments, including DNA with incorporated
[methyl-3HJ-thymidine, were collected on glass microfiber filter

17

paper (Hhatman 93u-AH, Cambridge Technology, Cambridge, Massa-
chusetts).

Filter disks were placed in liquid scintillation counting
(LSC) vials and scintillation cocktail was added to the vial.
The scintillation cocktail was prepared with toluene and 42 ml
Liquifluor (New England Nuclear, Boston, Massachusetts) per
liter toluene. Liquifluor is a PPO-POPOP concentrate that re-
sults in u g PPO/liter and 50 mg POPOP/liter when diluted with
toluene. Radioactivity was quantitated by LSC on a Beta Tracor
by ESR. The Beta Tracor contained an internalized quench curve
for automatic converstion from counts per minute (CPM) to disin-

tegrations per minute (DPM).

.§La£i§&i£al Analxsis
All data was analyzed with the computer program Statis-

tical Analysis Systems (SAS) (SAS Institute Inc., Cary, North
Carolina). Means, standard deviation, range, variance, cor-
rected sums of squares, uncorrected sums of squares, standard
error of the mean, and the coefficient of variation for repli-
cate samples and/or across fish means were calculated with the
MEANS procedure for the dependent variables DPM, stimulation
index (SI = experimental DPM/control DPM) and standardized DPM
(SDPM = experimental DPM-control DPM).

Analysis of variance (ANOVA) of the variables DPM and
SI was accomplished with the General Linear Models (GLM)

procedure because of its flexibility to accept mixed models.

18

The models of response contained fixed main effects of mito-
gen concentration and incubation time, and random effects of
fish. The GLM procedure gave results in the form of sums of
squares for main effects, interactions,.and error, F-values
and associated probabilities. Models, formula and coeffi-
cients of expected mean squares for each of the experiments
are presented in Appendix B.

Expected mean squares were calculated according to the
procedures of Gill (1978). Comparison of mean response (DPM,
SI, and SDPM) from controls and experimentals was by t-test
(Gill, 1978) and least significant difference (LSD) multiple
range tests was calculated according to Sokal and Rohlf (1969a)
with t-values from Rohlf and Sokal (1969L. Comparison of mean
response was also performed by Tukey‘s Studentized Range (HSD)

with the MEANS/GLM procedures of SAS.

RESULTS
Media
Cell viability was assessed with lymphocytes cultured
in RPMI 16u0 (Table 1) or TC 199 (Table 2) culture media.
Either tissue culture medium supported lymphocytes when via-
bility was measured by Trypan blue dye exclusion. Viability
in all cases was greater than 90% when monitored daily

through seven days of incubation.

MILQEEDB

Stimulation of rainbow trout peripheral blood lymphocytes
(RBT PBL) was greatest when ConA was in the culture medium
(Table 3). The data presented in Table 3 represents the mean
responses of six or more fish with three or four replicate
cultures per fish. All cultures were incubated four days of
which the final 28 h was with radiolabelled thymidine. The
proliferative response of RBT PBL to the classical mammalian
mitogens PHA, LPS, or PWM was small, relative to that of mammals
at all concentrations. The lymphocytes' lack of response to
these three mitogens is noted when comparing both raw incorpora-
tion of labelled thymidine (DPM) and stimulation indices (SI).
The maximal mean SI of RBT PBL to PHA, LPS, or PWM was 1.83,
1.78, and 1.u0, respectively.

In contrast, the maximal mean SI of RBT PBL was n.87
when cultured with 10 ug ConA/ml. However, during this
series of experiments it became apparent that there was a

high degree of variation between the response of individual

19

20

.Aammv emcee

 

 

ommgcousum mkoxsh om mcfictooom $0.023 maogucoo so: 20.8.33 maucmoncmHm m 3
.Hoaaaoo zen . eoaaaaaaaa zen u :aam Am
maaoo Hogucoo :ma
mwmmm eased: an mum xeeaH eoaamflasaam u Am
.nmwm Lon mcoHamoHHaoL Lsou ou mots» ucm o N a saw: mcwufisansnz nu“: mm:
23:: mo meson :N puma on» .mamu Lao.“ mo coated couumnsocw cm mucomogam mumo HH< C
App.ov mo.o Am..ov mm.e “mo.ov ma.o Hs\ma ooa
Apo.ov o_.o Amp.ov o=.. Amo.ov m>.o Haxma om
Aeo.ov o Am_.ov .P.F Aop.ov mo.o Ha\ma m
22a
Amo.ov o “mo.ov m~.o Amo.ov ea.o Hs\ma omm
.Amp.ov aa.o afimm.ov m~.. aAmP.ov mF.P Hs\ma ooa
AF..ov am.o Amm.ov mm.P Aop.ov mm.c H2\ma om
was
Aop.ov o Am..ov .P.P Amo.ov oe.o Hex»: co,
aAm..ov _=.o eAam.oV mm.~ A=—.ov mo.F Ha\ma mm
AoF.ov am.o Amm.ov Fe._ Am_.ov oa.o stma m
«ma
Ame.ov o A=_.ov FF." arm.ov mo.m Ha\ma mp
eflmo.=v mo.aa eama._v pm.= aeflpo.av m~.~_ Hs\ma op
A>=.ov -.o Ama.ov mm.a Apm.ov .m.= Ha\ma m
«ado
Ammw muqfl m mmmm mammq mm Aumq mnqd a mum. mwmqum
fie N av mmqqqmmm_qmmm
Pam: no. 5a: 9.3 Hz: 5 3.3.25 and .22 3 2322.7: .8 0x32. and: .m 03.2.

21

fish (Table A). The range of stimulation indices was 2u.37 when
RBT PBL were cultured four days in RPMI 16110 with 10% FBS and 10
ug ConA/ml. The response of some individual fish are presented
in Table 5 as an example of this variation between individual
fish. The variation within individual fish was considerably
less than among-fish variation. Coefficients of variation from
triplicate samples were usually less than 20%. Analysis of
variance was performed on all individual experiments. The
models, estimate of mean squares, variance, percent of asso-
ciated variance, F-values, and probability values are presented
in Appendix B. The effect of fish was significant (p : 0.05) in
all experiments conducted with ConA as the mitogen when assessed
with the dependent variable DPM. The variation from fish to
fish was significantly greater than that of replication sample
from a fish. The variation in response to ConA by RBT PBL was
directly related to the degree of stimulation (Table AL. The
highest coefficients of variation are seen at what appears to be

the Optimal dose of ConA (10 ug/ml).

$1035.12: .91 Beams:

RBT PBL were cultured for periods of one to seven days
at doses of 1, 3, 10, 15, or 25 ug ConA/ml to determine the
optimal culture time and dose-response kinetics of the proli-
ferative response. Studies to optimize incubation period and
mitogen dose were limited to the use of ConA because the re-

sults of preliminary experiments indicated that ConA possessed

.cofiumfigm> mo acoaofimmooo u >0 .oocmHLm>
u m<>..:ofiumfi>oo nemccmum u new .cmos one no Lotto unmocmum u mm .moumoaaqog u z
.53... u x .mma 2: a3: neaeasoaaaaa sides 33 2:2 5 Page .38 etaaaadaa 3:5 2

22

 

om.me om.m mm.o m~.o =P.o .P.P am Hm
ea.mma ma.aa aroma mo.m Pm.o pm.m am Am1o. xv zen
Hs\ms mp
so.map >m.=~ o.mm mm.» m... em.a mm Hm
ma.aap m>.m> ammoma oo.mm 50.: «5.». mm Am-o_ xv zen
Ha\ma o,
aa.=e o..m me.o m>.o mp.o mm.P pm Hm
.m.ae Fo.FP mmpmma ~=.m em.o Pm.= am Amnop xv sac
as}: m
elm we; 35 mmd mod 8; S. 5
mm.>e mo.» ompa oe.~ P=.o =m.m oa Amuop xv :aa
HoLucoo
Anv >0 moz<m m<> new mm m z <eoo

 

.mucosfigoaxm HH< mmogo< <=oo on man am: ho uncommoz on» no :ofiamHLm> mo
ucwfiofimmmoo ucm .omcmm .oocmfigm> .cofiama>oa cemncmum .Loggm utmccmum .cmo:

.3 manmh

.cofiumagm>
no acofioamhmoo u >0 .oocmfigm> u m<> .cofiumfl>ou ugmocmum u new .moumoHHQoL u z
.cmms u x .mmu now new: noncoSmHQasm Enzyme oaop Him: :a mhmc Lao.“ cmumnsocam mHHmo 2.

23

 

mm.aa .~.o .o.o ma.o Pa.o m Hm
mm.m. a=.. mm» ao.o m=.a m Am1oa xv zam
a
em.m ma.o Fo.o o..o =N.P m Hm
em.m a... oaa ee.o ao.m m Am1o_ xv 2mm
eo.e ma.o Fo.ov ao.o ao._ m Hm
eo.e m_.o Pm oa.o am.. m Am-oa av zen
mm
ma.m m~.o aa.o mm.o em.> a Hm
m..m P..m mm.m~ Fm._ em.am a Amuo. av zen
Pm
ma.>_ om.m am.m mm.. as.» a Hm
ma.>_ ma.m. mammm em.> a=.ma a Am-op av zen
m
mm.m em.a ae.o om.o aw.mm a Hm
mm.m mo.e cape oe.~ ma.>~ a Amuoa xv zen
P
Auv >0 muzam ma> new m 2 dead

 

.Haxaeoo ma o. as and emu Haaea>aeaH do aaeoaaem has
mo :ofiumatm> ho acowo ammou cam .omcmm .oocmflgm> .coHumw>oa ugmvcmum .cmo: .

m magma

28

the greatest ability to stimulate RBT PBL. The culture media in
this set of experiments were supplemented with 10% FBS.

Those fish lymphocytes that responded well to ConA in
113:9 had distinctive kinetic patterns of both dose and incuba-
tion period. An example of RBT PBL kinetics of response to ConA
over a five day period is given in Figures 3 and 11. These are
the two fish from an experiment that tested ConA concentrations
of 1, 3, 10, and 25 ug/ml. The greatest response in both cases
was at 10 ug ConA/ml with an incubation period of five days. A
concentration of 25 ug ConA/m1 caused a greater initial proli-
feration of RBT PBL. In contrast, the activation of RBT PBL
with 3 ug ConA/ml required a longer incubation to attain maximum
stimulation. The response to 1 ug ConA/ml was not different
from control cell cultures. No plateau or decline of incorpora-
tion of radiolabelled thymidine was observed at the optimal dose
of ConA, additional experiments were done determine whether
proliferation of the cells increased past day five.

These results indicate an extreme variation in LA response
to ConA stimulation between the two fish tested even though the
control cells of these fish responded similarly to the culture
conditions. PBL from fish 21 (Figure 5) responded to ConA at 10
ug/ml under culture conditions, while PBL from fish 2a (Figure
6) failed to respond under the same culture conditions.

The second point taken from these results is that after
five days of incubation the proliferative response of RBT PBL to

ConA plateaus. At all concentrations of ConA, except the

25

8a a Control, 1 ug/ml

3 ug/ml

100
1
0

10 ug/ml

90
L
e

25 ug/ ml

0
1
x

DPMIIIO’
6

 

 

 

 

2 3 4 s
INCUBATION (days

Figure 3. Kinetics of Response of Fish 1 PBL to ConA in
Experiment 18.

26

8-

'2: In Control, 1 ug/ml
E?‘ o 3'1Lg/finfl

8“ o 10 ug/ml

8d x 25 ug/ml

DPM “0’
s

 

 

0 1 2 3I 4' 5
INCUBATION (days)

Figure u. Kinetics of Response of Fish 2 PBL to ConA in
Experiment 18.

27

   
    

 

 

 

D
D.—
V
In Control, 1 ug/ml
S-
m o 3 ug/ml
g- o 10 ug/ml
x 25 ug/ml
Si
N
E:
x 8..
a N
a.
C3
0
LD—A
O
2—
o-
[D
l I I I
Di 2 3 4 5 fl'7
INCUBATION (days)
Figure 5. Kinetics of Response of Fish 21 PBL to ConA in

Experiment 20.

28

 

 

 

8..
V‘
a: Control, 1 ug/ml
D
u)-
m o 3 ug/ml
g- o 10 ug/ml
x 25 ug/ml
8‘
N
"o
‘2 El
5
C3
8-
§J
04
L0
°1 2r ar 4' 5l 8'

INCUBATION (days)

Figure 6. Kinetics of Response of Fish 24 PBL to ConA in
Experiment 20.

29

smallest, the uptake and incorporation of radiolabelled thymi-
dine either reached a plateau or began to decline by day six
(Figure 5).

This experiment was repeated to further elucidate the
effect of the incubation time on proliverative response and the
variability in responsiveness among individual fish. Cells from
eight fish were incubated for periods of two to seven days at
ConA concentrations of 3, 10, and 15 ug/ml. The mean response
(DPM) across fish is presented in Figure 7.

Kinetics of response of RBT PBL to ConA was similar to
that of previous experiments, with the maximum amount of radio-
labelled thymidine incorporated by day.five at all concentra-
tions of ConA. However, the rate of proliferation of the con-
trol cells was the same as that of ConA-stimulated cells. The
optimal SI for individual fish ranged from 1.15 to ”.25 and
varied with both incubation period and mitogen concentration.
Analysis of variance, using the dependent variable DPM (Appendix
B) resulted in significance of all main effects of fish (p <
0.0001), ConA concentrations (p < 0.0001), and incubation time
(p < 0.0001), as well as all interaction terms (p < 0.0001).
The variation in proliferation response is dose-dependent, with
larger relative standard deviations seen at doses of ConA
leading to greater lymphocyte activation (Figure 8). It is
important to note that even though control cell cultures proli-
ferated at the same rate as the optimal doses of ConA, they had

much smaller relative standard deviations. Coefficients of

30

   

 

 

8-
O
x Control
.3: 03 ug/ml
0 10 ug/ml
o H 15 tug/tnl
8.4
D
g—
is
I §‘
5'.
n
D
8—
D
a—
8..
9‘ 2| 1 I 1 8I ”’j?

3 4 5
INCUBATION (days)

Figure 7. Mean Response of RBT PBL to ConA Across Fish in
Experiment 23.

31

J

a Control
0 3 ug/ml

9 10 ug/ml
x 15 ug/ml

l

100 110 120

90
L

Coefficient of Variation

 

 

r l I r I
°1 2 a 4 s a 7

INCUBATION (days)

Figure 8. Coefficients of Variation of the SI Response
Variable for Means Across Fish in Experiment 23.

32

variation for control cultures did not increase with the rate of
cell proliferation. This same phenomena is true when CV of SI
response are determined for across fish and experimental means
for the optimal stimulatory dose of ConA (Figure 9).—

Four points may be taken from this experiment: 1) The
optimal incubation period for RBT PBL stimulated with ConA is
five days. 2) The Optimal dose of mitogen for proliferative
response under the test conditions is in the range of 10-15
ug ConA/ml, however this dose range results in the greatest
variation in proliferative response. 3).Activation of RBT
PBL by ConA was slight as compared to control cultures with
optimal SI for individual fish ranges from 1.15 to 11.25 and a
mean optimal SI of 2.13. a) The positive correlation of
relative standard variation to rate of proliferation seen in
ConA-stimulated cells is absent in control cultures.

The question raised by the last two points is whether the
small amount of stimulation of lymphocytes treated with ConA
relative to controls is due to an inhibition of mitogen-mediated
proliferation or an artifact caused by the stimulation of con-
trols by something in the culture medium. ‘The next set of
experiments was designed to determine how much of the variation
was due to culture conditions, and how much was due to variation

in experimental procedures.

33

180
I

1

Control
8 10 ug/ml

L

l

105 120 185 150 165

l

90
i

Coefficient of Variation

 

30
‘

15

 

1 2 3' 4I 5' e 7
mcmnon (days

Figure 9. Coefficients of Variation of the SI Response
Variable of Across Fish and Experiment Means for
RBT PBL Stimulated with ConA.

34

Serum W

Experiments were performed in which the culture medium
was supplemented with human sera instead of FBS, to assess the
effect of an alternate source of protein on RBT PBL prolifera-
tive response. All other parameters of the culture media and
assay procedure were the same as in previous experiments with
FBS supplementation. Heat inactivated human serum (HS) was
added to the culture medium at 10% (v/v), lymphocytes were
stimulated with ConA at 3, 10, or 15 ug/ml, and incubations were
either 2, 3, 4, or 5 days.

The initial experimental design (Appendix B19) used lym-
phocytes from seven fish. Lymphocytes were exposed to three
mitogen concentrations harvested after 3, 4, and 5 days of
incubation. Maximum SI for ConA-stimulated lymphocytes ranged
from 1.19 to 3.12. The mean response (DPM) across all fish in
this experiment is presented in Figure 10. This experiment was
repeated because it appeared that the optimal day for harvest
may have been prior to day 3. The experimental design for this
assay included a 2-day incubation period. In the second experi-
ment maximal SI for individual fish ranged from 1.55 to 6.27.
The mean response (DPM) at each dose across fish for this assay
is presented in Figure 11. The mean maximal SI across both of
these experiments was 2.95.

The kinetics of response of RBT PBL in culture medium
supplemented with HS are slightly different from RBT PBL cul-
tured with FBS. The greatest amount of incorporation of 3H-
thymidine into DNA was on day 3 (Figures 10 and 11) in both

35

 

 

 

O
N-a
0")
In Control
0
g); 0 3 ug/ml
9 10 ug/ml
34 x 15 ug/ml
N .
O
Q—t
N
6
FIG
n all
a CI.
0..
C3
0
9:“
Cd
D
0‘
V’
r r T l I
“’1 2 3 4 s e 7

INCUBATION (days)

Figure 10. Mean Response (DPM) of RBT PBL to ConA with 10%
Human Sera Supplementation in Experiment 28.

 

 

 

 

O
81
m Control
.3“ ° 3 Ila/ml
9 10 ug/ml
O X 15 ug/ml /
g-d
O
8‘
5'3
x Eh.
a i.
a
O
8-
O
8‘
EA
°1 2I 3I 4’ 5I 3T 7

INCUBATION (days)

Figure 11. Mean Response (DPM) of RBT PBL to ConA With 10%
Human Sera Supplementation in Experiment 31.

37

experiments. The mean response (DPM) across all fish PBL cul-
tures, from both of these experiments, is presented in Figure
12. A dose-response is seen with increasing ConA concen-
tration. The maximum average stimulation of RBT PBL was ob-
served when lymphocytes were incubated with 15 ug ConA/m1 for
three days. Only those cultures harvested on days 11 or 5

had a mean response significantly greater than that of the
control cultures. The pattern of greater variation in response
associated with the treatment combinations that produce the
greatest proliferative response (Figure 13) is similar to the
results seen when FBS is used as a protein source.

General linear models were used to analyze and partition
variance among the main effects (individual fish, ConA concen-
tration, and duration of incubation) and their interactions for
these experiments (Appendices 820-821, Bu2-Bu3). .All main ef-
fectscontributed significantly to the total variance (p < OAK”)
relative to unexplained residual variance when the dependent
response variable DPM was analyzed. The interactions of fish *
ConA concentration and fish ' incubation period were also signi-
ficant (p < 0.005) in both experiments with culture medium

supplemented with human serum.

W W
An experiment was performed in which a single blood sample
from each of four fish was split into four portions and each of
these portions was then assayed for lymphocyte proliferative

response to determine how much of the variability in proliferative

 

 

D
o—
ID
0 Control
Ed 0 3 ug/ml
’ 10 ug/ml
and X 15 ug/ml
S
N
v-o-+
(’0
(3
x E}.
2
a.
C3
h
2.:
ID
2d
cu4
(D
l I f l I I
°1 2 3 4 5 s 7

INCUBATION (days)

Figure 12. Mean Response (DPM) of RBT PBL to ConA With 101
Human Sera Supplimentation Across Experiments 28
and 31.

120
1 J

100 110
L

Coefficient of Variation

 

39

m Control
a 3 ug/ml

’ 10 ug/rnl
.. x 15 ug/ml

 

 

Figure 13.

l I 1* 1* l l
2 3 - 4 5 8 7

INCUB TION (days)

Coefficients of Variation for the DPM Response of
Across Fish and Experiment Means of RBT PBL
Stimualted With 101 Human Sera.

“0

response was due to the separation of the cells and assay
procedure. The results were analyzed with the GLM procedure of
SAS. ‘The model and formulation of expected mean squares are
presented in Appendix B16. The estimates of variance, parti-
tioning of variance, and F-values from a 3-way ANOVA are given
for the dependent variables DPM (Appendix B17) and SI (Appendix
B18). When the dependent variable DPM is used as the response,
the main effects of individual fish, ConA concentration, and
separation were all significant (p < 0.05) while none of the
interaction terms explained significant portions of the var-
iance. When the data was normalized to the proliferation ob-
served in medium without mitogens (SI), the only main effect
that was significant is the concentration of ConA, however, the
interaction term between individual fish and separation was also
significant (p < 0.05).

The percentage of the variance attributed to the separ-
ation procedure was 21.01% when using raw DPM response data.
That is, 21.01% of the variance in DPM response was to lympho-
cyte isolation, cell counting, delivery volumes, and other pro-
cedural steps. When SI is used as the dependent variable, the
variance due to these procedures is reduced to zero. This
indicates that the procedure is not affecting the degree to
which RBT PBL will be stimulated by ConA. Cells from an indivi-
dual fish responded similarly to the mitogen. The variation in
DPM response between samples from the same fish was likely due

to error associated with cell counts.

41

 

Table 6. Mean and Range of Responses of Four Subsamples of
PBL From Individual RBT to ConA.
been 2.! items 2211 Mean SI Range .1.
Elan 61
Control 1292 1124-1457 ---- ---

3 ug/ml 1745 1020-2409 1.37 0.78-2.10
10 ug/ml 1532 1166-1967 1.20 0.89-1.75
15 ug/ml 1250 813-1426 0.99 0.62-1 60
Elan 62
Control 652 289- 996 ---- ---

3 ug/ml 1093 537-2179 1.68 0.86-2.21
10 ug/ml 723 355-1115 1.12 0.66-1.59
15 ug/ml 603 168-1091 0.99 0.27-1.65
Elsi: 6.3
Control 1058 552-1507 ---- ---

10 ug/ml 1398 1177-1896 1.49 0.78-2.21
15 ug/ml 952 603-1206 0.97 0.68-1.29
m 6.11

Control 893 434-1642 ---- ---

3 ug/ml 1417 1273-1675 2.08 0.87-2.97
10 ug/ml 926 509-1609 1.22 0.52-1.67
15 ug/ml 726 402- 969 0.95 0.57-1.38

 

1) RB;,;BL cultured in RPMI 1640 supglemented with 10% FBS,
10 ° cells/well, incubated with H-thymidine (1 uCi/well)
for 24 hours prior to harvest, and harvested on day 5 of
incubation. DPM = disintegrations per minute. SI = sti-
mulation index (DPM treatment/DPM controls).

1:2

Elfin Source

To determine whether the variability in response of RBT
PBL to mitogens was due to laboratory fish health problems or
holding conditions, PBL from fish taken at Balders Fish Farm
were tested for their ability to respond to ConA in vitro.
Three fish from Balders were bled and whole blood samples were
diluted with complete medium (RPMI 1640/10} FBSL. The samples
were transported to the laboratory and lymphocytes were isolated
from the diluted whole blood. Cells were assayed as in other
experiments and stimulated with 1, 3, 10, 25 or 50 ug ConA/m1.
Following a four day incubation, lymphocyte culture which had
not been treated with ConA, had activities of 1580, 1585 and
1074 DPM and optimal stimulation indices in experimental cul-
tures were 0.67, 1.19, and 0.83, respectively. The low degree
of stimulation was similar to that seen in previous experiments
with fish held in our facilities. Therefore our fish holding
facilities did not appear to affect the in yjtrg mitogenesis of
RBT PBL adversely.

DISCUSSION
MC Quinlan:

RPMI 1640, a culture medium specifically designed for the
in 11339 culture of human lymphocytes (Moore et al., 1967), is
well suited for the culture of RBT PBL. Viability tests re-
sulted in >901 viability over a seven day period. The pH.(7;3)
and osmolality (312 mOs/kg) of the medium suggested by others
for culture of salmonid cells (Wolfe and Quinby, 1969; Sigel et
al., 1973; Warr and Simon, 1983) also appear to be suitable for
growth of RBT PBL.

Hitcssns

ConA in the culture media at 10 ug/ml provided optimal
stimulation of RBT PBL with an incubation period of 4 to 5 days.
The mean SI under these culture conditions was 4.85. The mito-
gens LPS, PHA, and PWM had only a slight stimulatory effect on
RBT PBL under the same culture conditions. These results are
similar to that of Etlinger et al. (1976a) who reported that
maximal RBT PBL activation was obtained with 10 ug ConA/ml and a
4-5 day incubation period. When FBS was used as a protein
source they reported SI between 2a2 and 3.1 that of controls.
The results of Warr and Simon (1983) were somewhat different
with ConA producing maximal stimulation at 3 ug/ml on day 4 only
with RBT PBL. The major difference of these results and those
reported here is the degree of stimulation. Warr and Simon
reported SI of about 7.3 for RBT PBL activated with ConA, about
twice that of the mean optimal SI in these studiest This

43

an

difference may be due to the fact that Warr and Simon did not
report a mean across all fish tested. In the materials and
methods section of their paper they state, "For reasons that are
not understood lymphocytes from an occasional fish failed to
respond or responded very poorly;" ‘Additionallyg it was unclear
how many fish were used in their study. From the results it
appears that only a single fish with quadruplicate cultures was
reported. If this were true, then their results would coincide
with my own, for I had individual fish with $1 equal to or
greater than 7.3. Their objective was to define heterogeneity
in RBT lymphocytes populations through mitogenic response, it
was not to assess the precise degree of response.

ConA has often proven to be stimulatory to PBL from other
fish species (Clem et al., 1977; Faulmann et al., 1983), however
the source of protein supplementation for the media appears to
be astrong covariate in these studies. For this reason a set
of experiments were performed with an alternate source of serum

supplementation.

Serum

Heat-inactivated human serum, used as an alternate source
of protein for RBT PBL cultures in lymphocyte activation, re-
sulted in a similar dose-response pattern and kinetic pattern as
those studies with FBS. The optimal dose was slightly greater
(15 ug ConA/ml) and the kinetic response was slightly acceler-
ated (3 day optimal incubation) but the mean maximal SI across

fish (SI = 2.95) was slightly less than that of experiments

45

conducted with FBS. Human serum was chosen because it success-
fully enhanced stimulation of PBL by ConA in other fish species
(Faulmann et al., 1983), it had not been previously tested in
this species, and is available commercially.

Finding the optimal serum supplement for lymphocyte acti-
vation studies in fish has been a major undertaking in previous-
ly reported studies and has only been determined through empiri-
cal testing of a wide variety of homologous and heterologous
sera. Clem and his coworkers (Cuchens and Clem, 1977; Clem et
al., 1981) tested human, calf, fetal calf, rabbit, alligator,
bass, grouper, and homologous sera in an effort to find optimal
conditions for bluegill (BG) and catfish (CF) PBL activation.
Results obtained from these sera individually or in various
combinations with BC PBL were "unrewarding" and pooled BG sera
proved to be cytotoxic to BC PBL in some cases. They were able
to solve this problem through dialysis of the BG sera overnight
against 0.15 M NaCl. The reason for the loss of cytotoxic
effects of the homologous serum after dialysis were not known or
explained by Clem. A similar trial and error experimentation
with CF PBL resulted in the unlikely combination of 10% human
and 5% channel catfish sera as the preferred supplement. Again,
the rationale for this choice is unknown and according to the
authors "may seem absurd," however they report the synergistic
effects of the mixture to be quite impressive (Faulmann et al.,
1983).

Avtalion and his coworkers (Rosenberg-Wiser and Avtalion,

1982; Caspi et al., 1984) report similar problems when carp PBL

46

are cultured. There was only a low percentage of proliferation
when 12% homologous serum was employed. Later studies with rat,
rabbit, horse, calf, human, dog, chicken, Illaplg, and carp sera
showed that manipulation of the sera was required for optimal
results. Charcoal-absorbed pooled carp sera (2-45) was optimal,
although only with low efficiency and high individual variabili-
ty. Further, they tested charcoal-absorbed carp sera from
various individuals and found that certain fish served well as
donors while others had sera that failed to support lymphocyte
activation.

The only other sera previously tested (besides FBS) with
RBT PBL is homologous rainbow trout sera (RBTS). Etlinger et
al. (1976a) reported that SI of RBT PBL cultured with ConA, LPS,
or purified protein derivative (PPD) were enhanced when 20% RBTS
was employed as opposed to 10% FBS. .A possible reason for the
increased SI of cultures grown with RBTS comes from another
report by these same authors. Etlinger et al. (1976b) cultured
RBT PBL in media with 10% FBS, 30% FBS, 20$ RBTS, or no serum at
all. These experiments were conducted without the addition of
mitogens and cultures were incubated either 4 or 5 days. There
was little growth in cultures without serum or those supple-
mented with 201 RBTS, while those cultures supplemented with
either 10% or 30% FBS had marked growth in the absence of any
mitogens. ‘Therefore, the smaller SI observed with FBS sup-
plementation compared with RBTS supplementation (Etlinger et
al., 1976a) was most likely due to the mitogenic effect of FBS

1:7

on control cultures. This stimulatory effect of FBS was noted
in PBL cultures from some fish in this study (control cultures
having counts up to 10,000 DPM), although not to the same extent
reported by Etlinger who reported counts upwards of 45,000 CPM.
A stimulatory effect of FBS on control cultures may have been a
factor in the relatively small SI seen in this study, with the
mean response of 3840 DPM on day 4. This is approximately twice
that reported by Warr and Simon after the same incubation
period. However, it seems more likely that inhibitory factors,
such as hormones, present in the FBS were responsible for the
low SI observed in these studies. Evidence for this comes from
the fact that fish cells with the greatest SI often had greater
incorporation of 3H-thymidine in control cultures (Figures 3 and
4).

All of these studies reported above point out that serum
supplements play a critical role in LA assays, yet they remain
an unresolved problem. The concentrations of hormones and pro-
teins have been observed to vary widely in commercially acquired
lots of FBS (Horn et al., 1975), particularly those hormones
which may have an effect on immunocompetence. In particular,
fish cellular immune response has been shown to be suppressed by
cortisone (Weinreb, 1959). Rasquin (1951) also demonstrated
fishes immune system sensitivity to hormones with ACTH. There-
fore, the development of assay procedures that utilize chemical-
ly defined media could help standardize 1n 113:9 LA techniques
with fish PBL and enhance the possibilities of their use as a

diagnostic tool.

48

13:33.12!) and mm i112

Variance of SI among fish in a given experiment when
the main effects of incubation period (day) and ConA concen-
tration were considered ranged from 3.31 to 19.83% of the
total variance (Appendix B). The variance components of
incubation period (day) and ConA concentration in these same
experiments had ranges of O to 6.77% and 6.00 to 29.48% of
the total variance, respectively; These main effects and all
the interaction terms (except the Day'ConA interaction in
experiment 20, Appendix B6) were significant at the p = CL05
level. The error variance was between 0.23 and 29.64% of the
total variance in these experimentst These fractions were
all increased in experiments that had only a single incuba-
tion period. The variance attributable to harvest day ef-
fects and interactions of this variable were folded into the
other main effects and error terms.

Orthogonal contrasts incorporating experiment to exper-
iment variation were impossible due to loss of samples in the
lymphocyte isolation process and poor recovery efficiency on
others. These problems led to varied numbers of fish or
reduced harvest days in some of the experiments. A two-stage
nested contrast of control cultures of fish across experi-
ments was performed, however, and both experiment and fish
were significant effects (p <OJKH) on any given harvest day.
A possible explanation of these sources of variance is the

differences in serum lots used in the various experiments.

49

Variation within a fish sample due to procedural hand-
ling of the cells (separation, counting, and platting) was
21.01% of the total variance when the dependent variable DPM
was analyzed and 01 of the total when SI was used. The
difference seen here was most likely attributable to the fact
that error due to separation and cell counts is normalized
when SI are the dependent variable. Any variation in DPM
response of mitogen stimulated cultures that is caused by
altered cell counts or cell separation on the gradient mater-
ial is reduced when divided by the control cultures response
from that same separation. The control cultures having been
handled the same as the stimulated cells in a particular
separation.

A useful piece of information that can be obtained from
the estimates of variance is the sample sizes required in
future experiments. Previous investigations of RBT PBL LA
have either failed to report the number of fish used (Warr
and Simon, 1983) or standard errors were not reported
(Etlinger et al., 1976a). From the estimates of variance
reported in this study, power analyses were conducted accor-
ding to Gill (1977). The design of the studies to be conduc-
ted dictates the number of samples (fish) that are required
to detect a specified difference or change in response. In
an experimental design that included only a single mitogen
treatment at the optimal dose (10 ug ConA/ml) and a single
incubation period (4 days) the sample sizes required to

detect various degrees of immunosuppression are presented in

50

Table 7. Variance estimated for this power analysis were
taken from Table 4, Type I error (A) = 0.05, and Type II
error (B) = 0.20. The number of fish that would be required,
with three replications per fish, is prOhibitively large even
for the detection of a 50% reduction in response. An experi-
mental design similar to a typical experiment from this study
(4 ConA concentrations and 6 harvest days) reduces this
requirement. Power analyses were performed with A = 0.05, B
= 0.20, replication per fish = 3, and variance estimates from
one of my experiments (Appendix B14). Detection of a 1000
DPM reduction (approximately 101) in response would require a
sample size of 41 fish, while a 0.2 SI reduction (approxi-
mately 10%) would require a 54 fish sample size. The re-
quirements of holding facilities in a toxicological study for
such a large number of fish in a given exposure would be
prohibitive due to the size of fish necessary.) Each fish
blood sample is split into at least four tubes in the lympho-
cyte isolation procedure, therefore centrifuging would also
be a limiting step in the process. The personnel needed to
process the samples would also be inordinantly large.
Therefore, the use of this assay as a diagnostic tool
in RBT is limited under conditions specified in the litera-

ture and this work.

51

Table 7. Sample Sizes Required to Demonstrate Reductions in
Immunocompetence of RBT PBL Responses to 10 ug
ConA/ml With An Incubation Period of 4 Days.

 

Denenneni Reenenee lenieble
Bereeni Benneiien

in Beenenee DE! §l
so: 50 55
+110: 77 86
301 136 152
201 310 342
10: 1240 1370

 

1) Type I error (A) : 0.05, Type II error (B) = 0.20, 3
replications/fish, and variance estimates from Table 4.

SUMMARY AND CONCLUSIONS

In yjtrg mitogenesis of RBT PBL, which had been demon-
strated in the literature previously, was further investi-
gated to define the optimal conditions for the procedure and
the respeatability of the assay for routine laboratory use.
The assay variables of media, mitogen, serum supplementation,
lymphocyte isolation procedure, and incubation period were
assessed.

RPMI 1640 culture media adjusted to pH 7.3 and 312
mOs/kg provided ample nutrients and tonicity for growth of
RBT PBL. Isolation of lymphocytes from whole blood was by
density gradient centrifugation. Both of the commercially'
available gradient materials tested (Percoll and Ficol-Paque)
resulted in cell populations consisting of >901 lymphocytes
and the remainder a distribution of moncytes, thrombocytes,
and granular leukocytes. Cell classification was based on
staining characteristics and flowcytometrics.

Of the mitogens tested (ConA, LPS, PHA, and PWM), the
greatest degree of lymphocyte activation of RBT PBL was
observed with ConA. The Optimal dose of ConA in the culture
wells was 10 ug/ml, however this treatment variable showed a
significant interaction with both fish and incubation period
as well as a three-way interaction (p (0.0001) in most
experiments. Kinetic studies of ConA stimulated cells re-
sulted in optimal stimulation at 4 to 5 days, but as with
mitogen concentration, incubation period had significant

interactions (p <0u0001) with the other main treatment

52

53

effects (fish and ConA concentration). Therefore, comments
as to the optimal conditions of ConA concentration and incu-
bation period must be kept to a qualitative or general na-
ture. The optimal ConA concentration is 10 ug/ml with an
incubation period of 4 to 5 days when across fish averages
are considered, but individual fish may respond optimally at
a different mitogen concentration in a different time frame.
Mean SI across fish and experiments at 10 ug/ml incubated 4
days was 4.87 (i 1.18 SE).

Human serum was tested as an alternative source of
protein. Dose-response and kinetic pattern of ConA stimu-
lated cells was similar to that of cultures supplemented with
FBS, but the degree of stimulation was lower.

Variability in response was shown to be directly re-
lated to the degree of proliferation in mitogen stimulated
cultures. Coefficients of variation for across fish means of
SI response at optimal dose of ConA were in the range of
100%, and those for across fish and experiment means were up
to 150%. Control cultures generally had coefficients of
variation for across fish and experiment means of less than
301. Procedural steps of cell isolation and fish holding
facilities did not appear to account for the large amount of
variation. It is suggested that serum supplement may account
for a large portion of the variation. However, the optimal
serum supplement or combination of sera was not established in
this set of experiments. Development of assay procedures in

completely defined media may provide a solution to this problem.

54

Power analysis with variance estimates from these
studies reveal that with an experimental design consisting of
a single ConA concentration (10 ug/ml) and incubation period
(4 days), the sample sizes required to detect a 50% reduction
in mitogen responsiveness of RBT PBL is approximately 50 if
the DPM response variable is used or 55 if the SI response
variable is evaluated. Requirements of sample size are re-
duced if a multi-stage nested design is used, however, the
overall work load would not be reduced significantly because
of the increase in harvest days, mitogen concentrations, and
number of cells needed in such studies.

My specific conclusions are:

1) Cell separation with either Percoll or Ficol-Paque
gradient materials provided relatively pure prepara-
tions of RBT PBL.

2) Optimal conditions of RBT PBL in £1559 mitogenesis
were determined to be 10 ug/ml ConA, 4-5 day incuba-
tion, and 10% FBS supplementation in this study.

3) Medium supplemented with 10% human serum did not
enhance stimulation indicies of RBT PBL cultured with
ConA or reduce the variability in response.

4) Variability in proliferative response of RBT PBL
under optimal conditions of this study would pre-
clude the use of this assay for laboratory screening

of RBT immunocompetence.

55

5) Responsiveness of individual fish PBL suggests that
in 115:9 mitogenesis of RBT PBL may be a useful
technique if standardized culture conditions (i.e.

‘serum supplementation) are established.

APPENDICES

56

APPENDIX A1

ConA screening with human lymphocytes from prospective organ

onors. Assa refor d in Dr. Robert Bull's laborator ith
gH-thymidine gspa monTgor. y w

L

* SI=Stimulation Index

Mean £35 £11

Donar C

ControT 1571
ConA 1 mg/ml 39308
ConA 0.1 mg/ml 125789
ConA 0.01 mg/ml 107431
ConA 0.001 mg/ml 17760
Donar D

ControT 2481
ConA 1 mg/ml 53482
ConA 0.1 mg/ml 81678
ConA 0.01 mg/ml 101962
ConA 0.001 mg/ml 30682
Control )5

Control 1056
ConA 1 mg/ml 5788
ConA 0.1 mg/ml 190166
ConA 0.01 mg/ml 57786
ConA 0.001 mg/ml 3340
Control 1

Control 1732
ConA 1 mg/ml 5486
ConA 0.1 mg/ml 148285
ConA 0.01 mg/ml 62214
ConA 0.001 mg/ml 5318

CPH experimentals/ CPM controls.

57

APPENDIX 81

Model, Formula and Coefficients of Expected Mean Squares for the 3-Way
ANOVA of 3H-Thymidine Incorporation into Cultured RBT PBL in
Experiment 181.

 

Model
Y = u + Aa + 3b * cc * (AB)ab * (Ac)ac * (3c)bc

+ (ABC)abc * E(abc)r

 

Sources of Variation .gfz .ELEEIZ
1.Fish (e) 1 s2 + mas?c '
2.ConA (A) 4 s2 + zosZac + 102A 2,
3.Day (a) 4 s2 + 20s2bc + 1028 2o
4.Fish*ConA 4 $2 + ZOSZac
5.Fish*0ay 4 s2 + ZOSZbC
6.ConA*0ay 16 s2 + 4s?“c

+ 0.522(Aa) Zab
7.Fish*ConA*Day 16 52 + 452abc
8.Error 143 $2
9.Total 192

 

1)Model for experiment is mixed with fixed treatment effects of ConA
and Day, and random effect of Fish. The levels for each treatment
are: Fish, c = 2; ConA, a = 5; Day, b = 5. The experimental design
was completely cross-classified with r=4.

2)df. = degrees of freedom, E[MS] = expected mean squares

58

APPENDIX 82

Estimate of Variance, Partition of Variance, and 3-Way ANOVA of DPM
Response from RBT PBL Cultured in Experiment 181.

 

5222222 2:
1.Fish 1
2.ConA 4
3.Day 4
4.Fish*C0nA 4
5.Fish*0ay 4
6.ConA*Day 16

7.Fish*ConA*Day 16
8.Err0r 143
9.Total 192

32
0.4437 x
9.7607 x
8.9145 x
0.8728 x
0.2567 x
4.2021 x
0.6107 x
0.0782 x

6.2870 x

108
108
108
108
108
109
108
108
109

zlot. $2

0.71
15.53
14.13
1.25
0.41
66.84
0.97
.gélg
1002

.5

567.97
1448.20
1206.14
201.05
67.10
300.67
32.21

‘23::
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001

<0.0001

 

l)df - degrees of freedom, 52 = estimated variance, F-value from

General Linear Models procdure from SAS.

59

APPENDIX 83

Estimate of Variance, Partition of Variance, and 3-Way ANOVA of 51

Response

from RBT PBL Cultured in Experiment 181.

 

59.42222 2: :2

1.Fish 1 7.58
2.ConA 4 67.50
3.0ay 4 15.50
4.Fish*C0nA 4 12.07
5.Fish*0ay 4 3.42
6.ConA*Day 16 116.00
7.Fish*C0nA*0ay 16 6.37
8 .E rror 113 L53
9.T0tal 192 228.98

xrot. 52

3.31
29.48
6.77
5.27
1.50
50.66
2.78
0,2;
100%

.E
1430.97
1727.69

422.99
457.43
130.32
159.67

49.34

.2515
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001

 

l)df 8 degrees of freedom, s2 = estimated variance, F-value from

General Linear Models procdure from SAS.

60

APPENDIX 84

M0del,F0rmula and Coefficients of Expected Mean Squares for the 3-Way
ANOVA of 3H-Thymidine Incorporation into Cultured RBT PBL in
Experiment 203.

 

Model
Y = U I A8 I Bb I cc I (A3)ab I (ACIac I (BCIbc I

(ABC)abc I E(abc)r

 

Sources of Variation 'gfb .EL!§lb

1.Fish (C) 2 s2 + 100s2

2.ConA (A) 4 s2 + 20s2ac + 1524 2a

3.0ay (B) 4 52 + 20stc + 1528 2b
4.Fish*ConA 8 s2 + 20s2 a,

5.Fish*0ay 8 s2 + 20stc

6.C0nA*Day 16 52 + 4sZabc + 0.7522(AB) Zab
7.Fish*ConA*Day 32 52 + 452abc

8.Err0r .208 52

9.T0tal 282

 

a) Model for experiment is mixed with fixed treatment effects of ConA
and Day, and random effect of Fish. The levels for each treatment
are: Fish, 0 = 3; ConA, a = 5; Day, b = 5. The experimental design
was completely cross-classified with r=4.

0) df. = degrees of freedom, E[MS] = expected mean squares

61

APPENDIX 85

Estimate of Variance, Partition of Variance, and 3-Way ANOVA of DPM
Response from RBT PBL Cultured in Experiment 201.

 

Sources

1.Fish
2.C0nA
3.0ay

4.Fish*ConA
5.Fish*0ay
6.ConA*Day
7.Fish*ConA*Day

8.Err0r

9.T0tal

a: a: a. a- A. 14.

16
32

.229.
282

22
8.6479 x
3.9192 x
2.2615 x
1.7542 x

8.5663 x

1.4319 x
3.0302 x
7.8640 x

106
105
107
107
106

107
105
107

zTot. s2

11.00
4.98
28.76
22.31
10.89
0.0
18.21
.2222
100%

.5
286.39
136.18
169.48
116.78

57.54
12.34
19.90

M
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001

 

l)df 8 degrees of freedom, s2 = estimated variance, F-value from

General Linear Models procdure from SAS.

62

APPENDIX 86

Estimate of Variance, Partition of Variance, and 3-Way ANOVA of 51

Response from RBT PBL Cultured in Experiment 201.

 

Sources 9;
1.Fish 2
2.C0nA 4
3.0ay 4
4.Fish*ConA 8
5.Fish*Day 8
6.ConA*Day 16
7.Fish*ConA*Day 32
8.Err0r .298
9.T0tal 282

E2
0.2447
0.5168
0.0

0.5490
0.2160
0.0

0.5060

0.8569

2.8911

zTot. 52

8.46
17.94
0.0
18.99
7.47
0.0
17.50
22222
100%

29.55
22.89
2.62
13.81
6.04
1.35
3.34

PR>F

 

<0.0001
<0.0001
<0.0359
<0.0001
<0.0001
<0.1781
<0.0001

 

l)df = degrees of freedom, 52

General Linear Models procdure from SAS.

= estimated variance, F-value from

63

APPENDIX 87

M0del,F0rmula and Coefficients of Expected Mean Squares for the 3—Way
ANOVA of 3H-Thymidine Incorporation into Cultured RBT PBL in
Experiment 213.

 

Model

Y = U I Aa I cc I (ACIac I E(abc)r

 

Sources of Variation 3E0 'ELMSJP
1.Fish (0) 7 s2 + 12s2

2.C0nA (A) 3 s2 + 3s?ac + 82A 2,
3.Fish*C0nA 21 s2 + 3523c

4.Err0r 64 s2

5.T0tal 95

 

a) Model for experiment is mixed with fixed treatment effects of ConA,
and random effect of Fish. The levels for each treatment are: Fish, c
= 8; ConA, a = 4. The experimental design was completely
cross-classified with r=3.

0) df. = degrees of freedom, E[MS] = expected mean squares

64

APPENDIX 88

Estimate of Variance, Partition of Variance, and 3-Way ANOVA of DPM
Response from RBT PBL Cultured in Experiment 211.

 

 

5212.222 21: .42 2121.22 .E. £82}:
1.Fish 7 0.6814 x 105 8.77 3.29 <0.0048
2.ConA 3 3.5242 x 105 45.36 8.79 <0.0001
3.Fish*ConA 21 0 0.0 0.88 <0.6155
4.Err0r 21 3.5631 x 105 .45,86 -- --
5.T0tal 95 7.7688 x 105 100%

 

l)df = degrees of freedom, s2 = estimated variance, F-value from
General Linear Models procdure from SAS.

65

APPENDIX 89

Estimate of Variance, Partition of Variance, and 3-Way ANOVA of 51
Response from RBT PBL Cultured in Experiment 211.

 

5.092222 .0: .22 2122.22 5 £826
1.Fish 7 0.0267 2.63 1.65 <0.1380
2.C0nA 3 0.4804 47.34 8.83 <0.0001
3.Fish*C0nA 21 0.0122 1.20 1.07 <0.3973
4 .Error 64 M955 48:83 -- --
5.T0tal 95 1.0148 1001

 

l)df = degrees of freedom, 52 = estimated variance, F-value from
General Linear Models procdure from SAS.

66

APPENDIX 810

M0del,F0rmula and Coefficients of Expected Mean Squares for the Z-Way
ANOVA of 3H-Thymidine Incorporation into Cultured RBT PBL in
Experiment 229.

 

Model

Y = U I Aa I cc I (Ac)ac I E(abc)r

 

Sources of Variation 'gfb .ELMSJD
1.Fish (0) 2 s2 + 12s2c

2.ConA (A) 3 52 + 3s2ac + 324 2,
3.Fish*C0nA 6 s2 + 3s2ac

4.Err0r .23 s2

5.T0tal 35

 

a) Model for experiment is mixed with fixed treatment effects of ConA,
and random effect of Fish. The levels for each treatment are: Fish, c
= 3; ConA, a = 4. The experimental design was completely
cross-classified with r=3.

0) df. = degrees of freedom, E[MS] = expected mean squares

67

APPENDIX 811

Estimate of Variance, Partition of Variance, an 3-Way ANOVA of DPM
Response from RBT PBL Cultured in Experiment 22 .

 

 

£925.22 £1.15 .22 21.22.22 E 282.:
1.Fish 2 2.9130 x 104 39.67 3.29 <0.0048
2.ConA 3 0 0.0 8.79 <0.0001
3.Fish*C0nA 6 3.0516 x 104 41.56 0.88 <0.6155
4.Err0r 25 1.3777 x 104 .1§;16 -- --
5.T0tal 35 7.3423 x 104 100%

 

l)df = degrees of freedom, 52 = estimated variance, F-value from
General Linear Models procdure from SAS.

68

APPENDIX 812

Estimate of Variance, Partition of Variance, and 3-Way ANOVA of SI
Response from RBT PBL Cultured in Experiment 22 .

2222222 .2: 22 21222.22 .5. 22:2
1.Fish 2 0.0508 22.08 10.34 <0.0006
2.C0nA 3 0.0 0.0 0.55 <0.6528
3.Fish*C0nA 6 0.1141 49.55 6.24 <0.0005
4 .Error 24 L065; M -- --
5.T0tal 35 0.2302 100%

 

l)df = degrees of freedom, 52 = estimated variance, F-value from
General Linear Models procdure from SAS.

69

APPENDIX 813

M0del,F0rmula and Coefficients of Expected Mean Squares for the 3-Way
ANOVA of 3H-Thymidine Incorporation into Cultured RBT PBL in
Experiment 23“.

 

Model
Y 8 u + Aa + 3b + Cc + (AB)ab I (ACIac I (BCIDC I

(ABC)abc I E(abc)r

 

Sources of Variation .gjb ELMélb

1.Fish (0) 7 s2 + 72s2c

2.C0nA (A) 3 s? + 1852ac + 362A 2,

3.08y (8) 5 s2 + 12stC + 1628 20
4.Fish*ConA 21 52 + 18szac

5.Fish*0ay 32 s2 + 1252bC

6.C0nA*Day 15 s2 + 3s2abc + 1.612(48) Zoo
7.Fish*ConA*Day 96 s2 + 3523bC

8.Err0r .EEQ 52

9.T0tal 537

 

a) Model for experiment is mixed with fixed treatment effects of ConA
and Day, and random effect of Fish. The levels for each treatment
are: Fish, c = 8; ConA, a = 4; Day, b = 6. The experimental design
was completely cross-classified with r=3.

b) df. = degrees of freedom, E[MS] = expected mean squares

70

APPENDIX 814

Estimate of Variance, Partition of Variance, and 3-Way ANOVA of DPM
Response from RBT PBL Cultured in Experiment 231.

 

 

29.9.2223 9.: .22
1.Fish 7 2.3126 x 106
2.C0nA 3 0
3.06y 5 1.2057 x 107
4.Fish*ConA 21 2.9221 x 106
5.Fish*0ay 32 1.3094 x 106
6.ConA*0ay 15 0
7.Fish*ConA*Day 96 2.0432 x 106
8.Err0r .léé 0.7995 x 106
9.T0tal 282 2.1444 x 107

%Tot. 52

10.78
0.0

56.23

13.63
6.11
0.0
9.53
.3;13
100%

.5

209.26
58.65
261.93
66.79
20.65
5.92
8.67

.2315
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001

 

l)df = degrees of freedom, 52 = estimated variance, F-value from

General Linear Models procdure from SAS.

71

APPENDIX 815

Estimate of Variance, Partition of Variance, and 3-Way ANOVA of SI
Response from RBT PBL Cultured in Experiment 231.

 

Sources

1.Fish
2.C0nA
3.0ay

4.Fish*C0nA
5.Fish*0ay
6.C0nA*Day
7.Fish*ConA*Day

8.Err0r

9.T0tal

21
32
15
96

.222
537

22
0.1098
0.0332
0.0
0.1295
0.0929
0.0
0.1113
2.2112

0.5537

%T0t. 52

19.83
6.00
0.0

23.38

16.78
0.0

20.11

13,29
1002

.5
103.72
46.83
2.98
31.28
15.49
4.75
5.34

M
<0.0001
<0.0001
<0.0120
<0.0001
<0.0001
<0.0001
<0.0001

 

l)df = degrees of freedom, 52 = estimated variance, F-value from

General Linear Models procdure from SAS.

72

APPENDIX 816

Model, Formula nd Coefficients of Expected Mean Squares for the
3-Way ANOVA of H-Thymidine Incorporation into Cultured RBT PBL
in Experiment 261.

 

Model
Y = u + Aa + 36 + Cc + (AB)an + (Aclac + (Bclbc

I (ABCIabc I E(abc)r

 

Sources of Variation .012 E|MS|2
1.Fish (0) 3 s2 + 4852c
2.ConA (A) 3 52 + 12s2ac + 162A 2,
3.Separati0n (8) 3 52 + 12stC + 1628 2b
4.Fish*C0nA 9 s2 + 1252ac '
5.Fish*Sep 9 s2 + 12stc
6.C0nA*Sep 9 52 + 352abc

+ .522(A8) Zeb
7.Fish*ConA*Sep 27 s2 + 3szabc
8.Err0r .128 52
9.T0tal 191

 

1)Model for experiment is mixed with fixed treatment effects of
ConA and Separation, and random effect of Fish. The levels for
each treatment are: Fish, c = 4; ConA, a = 4; Separation, b = 4.
The experimental design was completely cross-classified with r=3.

2)df. = degrees of freedom, E[MS] = expected mean squares

73

APPENDIX 817

Estimate of Variance, Partition of Variance, and 3-Way ANOVA of DPM
Response from RBT PBL Cultured in Experiment 26 .

 

 

2222222 .2 .22 2122.22 2 282.2
1.Fish 3 9.0173 x 104 5.74 14.06 <0.0001
2.C0nA 3 2.9066 x 105 18.50 13.47 <0.0001
3.Separati0n 3 3.3012 x 105 21.01 15.11 <0.0001
4.Fish*C0nA 9 0 0.0 0.87 <0.5547
5.Fish*Sep 9 0 0.0 0.79 <0.6254
6.C0nA*Sep 9 0.0170 x 104 0.01 0.71 <0.7035
7.Fish*ConA*Sep 27 0 0.0 0.71 <0.8533
8.Err0r .122 8.6030 x 105 .54,15 A -- --
9.T0tal 191 1.5714 x 106 100:

 

l)df 8 degrees of freedom, 52 = estimated variance, F-value from
General Linear Models procdure from SAS.

74

APPENDIX 818

Estimate of Variance, Partition of Variance, and 3-Way ANOVA of 51
Response from RBT PBL Cultured in Experiment 261.

 

5.222222 .2 22 21222.22 .5 22:2
1.Fish 3 0.86 x 10-2 0.87 1.55 <0.0890
2.C0nA 3 4.57 x 10-3 46.41 13.47 <0.0001
3.Separation 3 0 0.0 1.31 <0.0676
4.Fish*ConA 9 0.10 x 10-2 1.12 0.87 <0.2006
5.Fish*Sep 9 1.32 x 10-3 13.81 3.98 <0.0001
6.C0nA*Sep 9 0 0.0 0.71 <0.4103
7.Fish*ConA*Sep 27 2.30 x 10-2 3.04 .°-71 <0.1949
8.Err0r 128 3.43 x 10'3 .34,15 -- --
9.T0tal 191 9.86 x 10-3 100%

 

l)df = degrees of freedom, s2 = estimated variance, F-value, and
probability value from General Linear Models procdure from SAS.

75

APPENDIX 819

M0del,F0rmula and Coefficients of Expected Mean Squares for the 3-Way
ANOVA of 3H-Thymidine Incorporation into RBT PBL Cultured with Human
Sera in Experiment 28“.

 

Model
v=o+n+8b+m+(mno+umn+(mmc+

(ABCIaDC I E(abc)r

 

Sources of Variation gjb ELMSJP

1.Fish (0) 6 s2 + 3652c

2.C0nA (A) 3 s? + 952ac + 211A 2,

3.0ay (B) 2 s2 + 12stC + 4228 2b
4.Fish*C0nA 18 s2 + 9szac

5.Fish*0ay 10 s2 + 12stc

6.C0nA*Day 6 s2 + 352abc + 3.522(AB) Zab
7.Fish*ConA*Day 30 s2 + 3s?“c

8.Err0r 151 s2

9.T0tal 226

 

a) Model for experiment is mixed with fixed treatment effects of ConA
and Day, and random effect of Fish. The levels for each treatment
are: Fish, c = 7; ConA, a = 4; Day, 0 = 3. The experimental design
was completely cross-classified with r=3.

b) df. = degrees of freedom, E[MS] = expected mean squares

76

APPENDIX 820

Estimate of Variance, Partition of Variance, and 3-Way ANOVA of 0PM
Response from RBT PBL Cultured with Human Sera in Experiment 281.

 

222222 2:
1.Fish 6
2.C0nA 3
3.0ay 2
4.Fish*C0nA 18
5.Fish*Day 10
6.C0nA*Day 6

7.Fish*ConA*Day 30
8.Err0r 151
9.T0tal 226

32
1.9260 x
1.2464 x
4.0745 x
0.9670 x
2.3349 x

0.5041 x

7.2142 x
1.8267 x

105

105

105
105
105
105

105
106

zTot. sZ

10.54
6.82
22.31
5.29
12.78
2.76
0.0
39,49
100%

10.61
5.83
28.61
2.21
4.88
0.72

0.48

.2315
<0.0001
<0.0010
<0.0001
<0.0051
<0.0001
<0.6332
<0.9903

 

l)df = degrees of freedom, s2 = estimated variance, F-value, and

probability value from General Linear Models procdure from SAS.

77

APPENDIX 821

Estimate of Variance, Partition of Variance, and B-Hay ANOVA of SI
Response from RBT PBL Cultured with Human Sera in Experiment 281.

 

2222222 2:. 22 1122.22 .2 282.2
1.Fish 6 0.2270 12.64 8.46 <0.0001
2.C0nA 3 0.2596 14.45 7.21 <0.0002
3.06y 2 0.0 0.0 0.40 <0.6685
4.Fish*C0nA 18 0.1497 8.34 2.23 <0.0046
5.Fish*0ay 10 0.0 0.0 0.85 <0.5858
6.ConA*Day 6 0.0643 3.58 0.61 <0.7183
7.Fish*ConA*Day 30 0.0 0.0 0.41 <0.9973
8.Err0r 151 1111913- M I -- --
9.T0tal 226 1.7959 1001

 

l)df = degrees of freedom, 52 = estimated variance, F-value, and
probability value from General Linear Models procdure from SAS.

78

APPENDIX 822

Model,F0rmula and Coefficients of Expected Mean Squares for the 3-Way
ANOVA of 3H-Thymidine Incorporation into Cultured RBT PBL Exposed to
ConA or LPS in Experiment 293.

Model

Y = U I Aa I Cc I (ACIac I E(ac)r

 

Sources_gf Variation .919 ‘115114
1.Fish (C) 3 s2 + 12s2c

2.C0nA (or LPS) (A) 4 s2 + 3s2ac 4 42A 2,
3.Fish*C0nA (or LPS) 9 52 + 3s2ac

4.Err0r '11 s2

5.T0tal 47

 

a) Model for both ConA and LPS stimulated cells in this experiment.
Model is mixed with fixed treatment effects of ConA or LPS, and random
effects of Fish. The levels for each of the treatments are: Fish, c
= 4; ConA (or LPS), A = 4. The experimental design was completely
cross-classified with r a 3.

b) d.f. = degrees of freedom, E[MS] = expected mean squares

79

APPENDIX 823

Estimate of Variance, Partition of Variance, and Z-Way AN VA 0f DPM
Response from RBT PBL Cultured with ConA in Experiment 29 .

 

 

2222222 2: .22 31222.22 2 222.2
1.Fish 3 7.9251 x 104 19.88 4.03 <0.0154
2.C0nA 3 0.5490 x 104 1.38 0.49 <0.6895
3.Fish*C0nA 9 0 0.0 0.42 <0.9127
4.Err0r 32 3.1398 x 105 .1311; -- --
5.T0tal 47 3.9872 x 105 100:

 

l)df = degrees of freedom, s2 = estimated variance, F-value, and
probability value from General Linear Models procdure from SAS.

80

APPENDIX 824

Estimate of Variance, Partition of Variance, and Z-Hay AN VA of SI
Response from RBT PBL Cultured with ConA in Experiment 29 .

 

 

 

2222222 .22 .42 31222.22 2 PM
1.Fish 3 0.0998 16.34 3.68 <0.0220
2.ConA 3 0.0642 10.52 1.05 <0.3825
3.Fish*ConA 9 0 0 0.48 <0.8784
4.Err0r .11 0.4466 .11111 -- --
5.T0tal 47 0.6105 100%

 

l)df = degrees of freedom, 52 a estimated variance, F-value, and
probability value from General Linear Models procdure from SAS.

81

APPENDIX 825

Estimate of Variance, Partition of Variance, and Z-Way ANOVA of DPM
Response from RBT PBL Cultured with LPS in Experiment 291.

 

 

 

2222222 .22 .22 2122.22 2 PM
1.Fish 3 1.3965 x 104 2.55 1.60 <0.2088
2.LPS 3 2.5487 x 105 46.49 3.97 <0.0163
3.Fish*LPS 9 0 0.0 0.32 <0.9616
4.Err0r 32 2.7941 x 105 591g§ -- --
5.T0tal 47 5.4825 x 105 100:

 

l)df = degrees of freedom, s2 = estimated variance, F-value, and
probability value from General Linear Models procdure from SAS.

82

APPENDIX 826

Estimate of Variance, Partition of Variance, and Z-Way ANOVA of SI
Response from RBT PBL Cultured with LPS in Experiment 291.

2222222 2: 22 21222.22 2 282.6
1.Fish 3 0.0353 4.10 2.15 <0.1138
2.LPS 3 0.4550 52.92 5.36 <0.0042
3.Fish*LPS 9 0.0 0.0 0.43 <0.9080
4 .Error 33 9_._3_§_9_§ 3.53.9.9. -- --
5.T0tal 47 0.8598 100%

 

l)df = degrees of freedom, 52 = estimated variance, F-value, and
probability value from General Linear Models procdure from SAS.

83

APPENDIX 827

M0del,F0rmula and Coefficients of Expected Mean Squares for the Z-Way
ANOVA of 3H-Thymidine Incorporation into Cultured RBT PBL Exposed to
PWM 0r PHA in Experiments 29 and 303.

 

Model

Y = U I Aa I Cc I (ACIac I E(ac)r

 

Sources of Variation ‘gfb .EL!§lb
1.Fish (0) 3 s2 + 12s2c

2.PWM (or PHA) (A) 3 s2 + 3s2ac + 42A 2,
3.Fish*PWM (0r PHA) 9 s2 + 3s2ac

4.Err0r 11 s2

9.T0tal 47

 

a) Model for both PWM and PHA stimulated cells in these experiments.
Model is mixed with fixed treatment effects of PWM 0r PHA, and random
effects of Fish. The levels for each of the treatments are: Fish, c
= 4; PWM (0r PHA); a = 4. The experimental design was completely
cross-classified with r = 3.

b) df. = degrees of freedom, E[MS] = expected mean squares

84

APPENDIX 828

Estimate of Variance, Partition of Variance, and Z-Way ANOVA of DPM
Response from RBT PBL Cultured with PHA in Experiment 291.

 

 

 

2922222 .2: .22 11222.22 2 PM
1.Fish 3 1.8564 x 105 20.38 10.70 <0.0001
2.PHA 3 4.7827 x 105 52.52 9.56 <0.0001
3.Fish*PHA 9 0.1714 x 105 1.88 1.22 <0.3153
4.Err0r .32 2.2965 x 105 .3512; -- --
5.T0tal 47 9.1068 x 105 1001

 

l)df . degrees of freedom, s2 = estimated variance, F-value, and
probability value from General Linear Models procdure from SAS.

85

APPENDIX 829

Estimate of Variance, Partition of Variance, and Z-Way ANOVA of SI
Response from RBT PBL Cultured with PHA in Experiment 291.

 

 

2222222 21. .22 2122.22 f. PM
1.Fish 3 0.4241 13.01 6.40 <0.0016
2.PHA 3 1.4383 44.13 7.47 <0.0006
3.Fish*PHA 9 0.4545 13.95 1.36 <0.2452
4 .E rror 11 112110 18191 -- ~-
5.T0tal 47 3.2589 100%

 

l)df 8 degrees of freedom, 52 8 estimated variance, F-value, and
probability value from General Linear Models procdure from SAS.

86

APPENDIX 830

Estimate of Variance, Partition of Variance, and Z-Way ANOVA of 0PM
Response from RBT PBL Cultured with PWM in Experiment 291.

 

 

2222222 .2: 22 21222.22 2 23:2
1.Fish 3 2.6780 x 105 48.50 12.32 <0.0001
2.PWM 3 0.0058 x 105 0.11 0.46 <0.7134
3.Fish*PWM 9 0 0.0 0.45 <0.8967
4.Err0r .32 2.8377 x 105 .51132 -- --
5.T0tal 47 5.5214 x 105 100:

 

l)df 8 degrees of freedom, s2 8 estimated variance, F-value, and
probability value from General Linear Models procdure from SAS.

87

APPENDIX 831

Estimate of Variance, Partition of Variance, and Z-Way ANOVA of SI
Response from RBT PBL Cultured with PWM in Experiment 291.

 

2228.222 22 22 21222.22 .5 23:2
1.Fish 3 0.0937 17.83 4.37 <0.0109
2.PWM 3 1.0985 18.74 1.98 <0.1368
3.Fish*PHM 9 0 0.0 0.80 <0.6206
4 .Error 12 211114. 11111 -- --
5.T0tal 47 0.5257 100:

 

l)df 8 degrees of freedom, 52 = estimated variance, F-value, and
probability value from General Linear Models procdure from SAS.

88

APPENDIX 832

Estimate of Variance, Partition of Variance, and Z-Hay ANOVA of 0PM
Response from RBT PBL Cultured with PHA in Experiment 301.

 

 

2222222 .22 .22 31222.22 2 .2822
1.Fish 3 2.1324 x 104 10.10 2.39 <0.0873
2.PHA 3 0.5464 x 104 2.59 0.46 <0.7096
3.Fish*PHA 9 0 0.0 0.35 <0.9522
4.Err0r '32 1.8440 x 105 Igzggg -- --
5.T0tal 47 2.1119 x 105 1001

 

l)df 8 degrees of freedom, 52 8 estimated variance, F-value, and
probability value from General Linear Models procdure from SAS.

89

APPENDIX 833

Estimate of Variance, Partition of Variance, and Z-Way ANOVA of SI
Response from RBT PBL Cultured with PHA in Experiment 301.

 

2222222 22 22 21222.22 .E 28.25
1.Fish 3 0.0483 7.96 2.08 <0.1227
2.PHA 3 0.0209 3.44 0.57 <0.6361
3.Fish*PHA 9 0.0 0.0 0.42 <0.9153
4 .Error 12 0111111 M -- --
5.T0tal 47 0.6075 100%

 

l)df 8 degrees of freedom, s2 8 estimated variance, F-value, and
probability value from General Linear Models procdure from SAS.

90

APPENDIX 834

Estimate of Variance, Partition of Variance, and Z-Way ANOVA of DPM
Response from RBT PBL Cultured with PWM in Experiment 301.

 

 

2222222 .2: 22 21222.22 .6 23.22
1.Fish 3 0 0.0 0.14 <0.9331
2.PWM 3 3.1254 x 104 15.51 1.34 <0.2795
3.Fish*PWM 9 0 0.0 0.60 <0.7847
4.Err0r .3; 1.7024 x 105 .81159 -- --
5.T0tal 47 2.0150 x 105 100%

 

l)df 8 degrees of freedom, s2 8 estimated variance, F-value, and
probability value from General Linear Models procdure from SAS.

91

APPENDIX 835

Estimate of Variance, Partition of Variance, and Z-Way ANOVA of SI
Response from RBT PBL Cultured with PWM in Experiment 301.

 

 

2222222 2:. .22 21222.2?- 2 PM
1.Fish 3 0.1198 15.92 3.75 <0.0204
2.PWM 3 0.1104 14.68 1.68 <0.1904
3.Fish*PWM 9 0 0.0 0.84 <0.5886
4 .E rror 12- W M -- --
5.T0tal 47 0.7521 100%

 

l)df 8 degrees of freedom, 52 8 estimated variance, F-value, and
probability value from General Linear Models procdure from SAS.

92

APPENDIX 836

M0del,F0rmula and Coefficients of Expected Mean Squares for the 2-Way
ANOVA of 3H-Thymidine Incorporation into Cultured RBT PBL Exposed to
ConA or LPS in Experiment 303.

 

Model

 

Y = U + Aa + CC + (AC)ac + E(ac)r

 

Sources of Variation gjb .ELߤlb
1.Fish (C) 1 s2 + 1stc

2.C0nA (or LPS) (A) 3 s2 + 3s2ac + 22A 2,
3.Fish*C0nA (or LPS) 3 s2 + 3523c

4.Err0r ‘19 s2

5.T0tal 23

 

a) Model for both ConA and LPS stimulated cells in this experiment.
Model is mixed with fixed treatment affects of ConA or LPS, and random
effects of Fish. The levels for each of the treatments are: Fish, c
8 2; ConA (or LPS), A 8 4. The experimental design was completely
cross-classified with r 8 3.

0) df. 8 degrees of freedom, E[MS] 8 expected mean squares

93

APPENDIX 837

Estimate of Variance, Partition of Variance, and Z-Way AN VA of DPM
Response from RBT PBL Cultured with ConA in Experiment 30 .

 

 

2222222 .22 22 21222.22 2 22:2
1.Fish 1 1.2256 x 105 26.51 16.07 <0.0010
2.C0nA 3 1.3992 x 105 30.27 7.01 <0.0032
3.Fish*C0nA 3 1.0225 x 105 22.12 4.14 <0.0237
4.Err0r _6 0.9758 x 105 .1111; -- --
5.T0tal 23 4.6231 x 105 1001

 

l)df 8 degrees of freedom, s2 8 estimated variance, F-value, and
probability value from General Linear Models procdure from SAS.

94

APPENDIX 838

Estimate of Variance, Partition of Variance, and Z-Way ANOVA of SI
Response from RBT PBL Cultured with ConA in Experiment 301.

 

228222 2: 22 5122.22 .6 662.6
1.Fish 1 0.0 0.0 0.04 <0.8477
2.C0nA 3 0.3878 59.25 6.76 <0.0037
3.Fish*C0nA 3 0.0869 13.28 2.45 <0.1010
4.Err0r 13 [211128 .11111 -- --

5.T0tal 23 0.6545 100%

 

l)df 8 degrees of freedom, 52 8 estimated variance, F-value, and
probability value from General Linear Models procdure from SAS.

95

APPENDIX 839

Estimate of Variance, Partition of Variance, and 2-Way ANOVA of DPM
Response from RBT PBL Cultured with LPS in Experiment 301.

 

 

 

29.9.6223. 9.1 .132 11262.22 E. PR>F
1.Fish 1 0 0.0 0.23 <0.6410
2.LPS 3 3.0406 x 105 69.97 5.00 <0.0123
3.Fish*LPS 3 0 0.0 0.34 <0.7965
4.Err0r 12 1.3048 x 105 .3019; -- --
5.T0tal 23 4.3451 x 105 100%

 

l)df 8 degrees of freedom, s2 8 estimated variance, F-value, and
probability value from General Linear Models procdure from SAS.

96

APPENDIX 840

Estimate of Variance, Partition of Variance, and Z-Way ANOVA of SI
Response from RBT PBL Cultured with LPS in Experiment 301.

 

2222222 .26 .22 2122.22 .6 .6626
1.Fish 1 0.1387 5.81 3.11 <0.0969
2.LPS 3 1.4600 61.51 4.33 <0.0205
3.Fish*LPS 3 0 0.0 0.63 <0.6081
4.Err0r .11 Q11§§2 .11101 -- --
5.T0tal 23 2.3876 100%

 

l)df 8 degrees of freedom, s2 8 estimated variance, F-value, and
probability value from General Linear Models procdure from SAS.

97

APPENDIX 841

M0del,F0rmula and Coefficients of Expected Mean Squares for the 3-Way
ANOVA of 3H-Thymidine Incorporation into Cultured RBT PBL Cultured
with Human Sera in Experiment 313.

Model

 

Y 8 u + A3 + 30 + Cc I (AB)ab I (ACIac I (Bc)bc I

(ABC)abc I E(abc)r

 

Sources of Variation .gfb .EL!§lb

1.Fish (C) 3 s2 + 48s?c

2.ConA (A) 3 s2 + 12sZac+ 1624 2,

3.0ay (B) 3 s2 + 12s2bc+ 1628 20
4.Fish*ConA 9 s2 + 12$Zac

5.Fish*0ay 9 s2 + 12stC

6.C0nA*Day 9 s2 + 352abcI 1.3322(A8) Zab
7.Fish*ConA*Day 27 s2 + 3s23bc

8.Err0r .118 s2

9.T0tal 19l

 

a) Model for experiment is mixed with fixed treatment effects of ConA
and Day, and random effect of Fish. The levels for each treatment
are: Fish, c 8 4; ConA, a 8 4; Day, 0 8 4. The experimental design
was completely cross-classified with r83.

0) df. 8 degrees of freedom, E[MS] 8 expected mean squares

98

APPENDIX 842

Estimate of Variance, Partition of Variance, and 3-Hay ANOVA of 0PM
Response from RBT PBL Cultured with Human Sera in Experiment 311.

 

2222222
1.Fish
2.C0nA
3.0ay
4.Fish*ConA
5.Fish*0ay
6.C0nA*0ay

7.Fish*ConA*Day

8.Err0r

9.T0tal

\D to 00 DJ 00 5'00

9
27

1.2.2
191

 

.22
4.2308 x 106
2.5453 x 106
1.1018 x 106
2.2909 x 105
1.5748 x 105
3.6213 x 106
0.7318 x 105
0.8621 x 106
1.6959 x 107

%T0t. 52

24.95
15.01
6.50
13.51
9.29
21.35
4.32
.5,08
1001

.E

236.56
80.13
43.37
32.89
22.92
9.13
3.55

.3315
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001

 

l)df 8 degrees of freedom, s2 8 estimated variance, F-value, and

probability value from General Linear Models procdure from SAS.

99

APPENDIX 843

Estimate of Variance, Partition of Variance, and 3-Way ANOVA of SI
Response from RBT PBL Cultured with Human Sera in Experiment 311.

 

2222222
1.Fish
2.C0nA
3.0ay
4.Fish*C0nA
5.Fish*0ay

£0 40 to DJ 0) 00 I"!

6.ConA*0ay
7.Fish*ConA*Day 27
8.Err0r 128
9.T0tal 191

32

0.1450
0.7142
0.5776
0.2938
0.1132
1.8225
0.2409
2.2212
4.1314

%T0t. s2

3.51
17.29
13.98

7.11

2.74
44.11

5.83

5.43

100%

.E
32.02
67.67
48.26
16.72

7.06
15.03
4.22

'23::
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001

 

l)df 8 degrees of freedom, s2 8 estimated variance, F-value, and

probability value from General Linear Models procdure from SAS.

REFERENCES

LIST OF REFERENCES

Absher, M. 1973. Hemocytometer counting. In: IRK. Kuse

and M.J. Patterson (edsJ. Iiggue Cgltgce; Methggg gag
Applieatigns. Academic Press, NY, NY.

Al-Sabti, K. 1983. Karyotypical studies on three salmonidae
in Slovenia using leukocyte culture technique. Ichthyo-
logis 15:92-99.

Al-Sabti, K. 1985. Frequency of chromosomal aberrations in

the rainbow trout, §§lmm ggjrgngnl Rick., exposed to
five pollutants. J. Fish Biol. 26:13-19.

Blaxhall, PAL 1972. The haematological assessment of the
health of freshwater fish. A review of selected litera-

Blaxhall, PJL 1983. Electron microscope studies of fish
lymphocytes and thrombocytes. J. Fish Biol. 22:223-229.

Blaxhall, FhC. 1983a. Factors affecting lymphocyte culture
for chromosome studies. J. Fish Biol. 22:61-76.

Blaxhall, PAL 1983b. Lymphocyte culture for chromosome
preparation. J. Fish Biol. 22:279-282.

Blaxhall, P.C., and K.W. Daisley. 1973. Routine haematolo-
gical methods for use with fish blood. J. Fish Biol.
5:771-782. '

Bleavins, M.R., R. J. Aulerich,.and R.K. Ringer. 1983.
Hexachlorobenzene-induced effects on lymphocyte blasto-
genic response to concanavalin A in the mik and eur0pean
ferret. Environ. Toxicol. Chem. 2:411-418.

Bleavins, FLR., and R. J. Aulerich. 1983. Immunotoxicologic
effects of polychlorinated biphenyls on the cell-
mediated and humoral immune systems. Res. Rev. 90:57-
67.

Boyum, A. 1968. Isolation of leucocytes from blood and bone
marrow. Scandinavian J. Clin. Lab. Invest. Suppl. 97:1-
12.

Caspi, R.R., R. Shahrabani, and R. R. Avtalion. 1980. The
cells involved in the immune response of fish. I. The
separation and study of lymphocyte sub-populations in
carp--a new approach. In: M. J. Manning (ed.). P134119-

891)! 9.1: 1.1111111021231291 new. Elsevier/North-Holland
Biomedical Press, p. 131-142.

100

101

Caspi, R.R., R. Shahrabani, T. Kehati-Dan, and R. R. Avtalion.
1984. Heterogeneity of mitogen-responsive lymphocytes in
carp (Cyprimgs canpio). Dev. Comp. Immunol. 8:61-70.

Chiller, J.M., H. O. Hodgins, V. C. Chambers, and R. S.
Weiser. 1969. Antibody response in rainbow trout (Sal-
mg ggiggnezl). I. Immunocompetent cells in the spleen
and anterior kidney. J. Immunol. 102:1193-1201.

Clem, L.W., C. J. Lobb, E. Faulman, and M. A. Cuchens. 1981.
Lymphocyte heterogeneity in fish: Differential environ-
mental effects on cellular functions. Develop. Biol.
Standard 49:279-284.

Cuchens, M.A. and L. W. Clem. 1977. Phylogeny of lymphocyte
hererogeneity. II. Differential effects of temperature
on fish T-like and B-like cells. Cellular Immunol.
34:219-230.

Ellis, A.E. 1977. The leucocyte of fish: a review. J.
Fish. Biol. 11:453-491.

Etlinger, H.M., H. O. Hodgins, and J. M. Chiller. 1976a.
Evolution of the lymphoid system. 1; Evidence for
lymphocyte heterogeneity in rainbow trout revealed by
the organ distribution of mitogenic responses. J. Immu-
nol. 116(6):1547-1553.

Etlinger,lLMg,lL O.Hodgin,.L M.Chiller. 1976b. Rainbow
trout leukocyte culture: a simplified method. In Vitro
.8:599-6010 ‘

Eskola, J., E. Soppi, M. Viljanen, and 0. Ruuskanen. 1975.
A new micromethod for lymphocyte stimulation using whole
blood. Immunol. Comm. 4(4):297-307.

Faulmann, E., M. A. Cuchens, C. J. Lobb, N. W. Miller, and L.
W. Clem. 1983. An effective culture system for
studying in vitro mitogenic responses of channel catfish
lymphocytes. Trans. Amer. Fish. Soc. 112:673-679.

Gill. J.L. 1978. Design and Anelxeie 21 Exnecimente in 211.2
Animal and 8.6.0.1221 §eieneee. loll .1. Iowa State Uni-

versity Press, Ames, Iowa. 409 p.

Greenlee, W.E., K. M. Dold, R. D. Irons, and R.- Osborne.
1985. Evidence for direct action of 2. 3. 7, 8-tetra-
chlorodibenzo-p-dioxin (TCDD) on thymic epithelium.
Toxicol. Appl. Pharmacol. 79:112-120.

Hadden, J. W. 1981. Cyclic nucleotides and related mechan-
ism in immune regulation: a mini review. In: N.
Fabins, E. Garaci, J. Hadden, and N. A. Mitchison

102

(eds.). Immunocggulatiog, Plenum Press, New York. p.
201-230.

Han, T., and J. Pauly. 1972. Simplified whole blood method
for evaluating 1n yitcg lymphocyte reactivity of labora-
tory animals. Clin. Exp. Immunol. 11:137-142.

Hartzman, R. J., M. Segall, M. L. Bach, and F. H. Bach.
1971. Histocompatibility matching. VI. Miniaturization
of the mixed leucocyte culture test: A preliminary report.
Transplantation 11:268-273.

Hartzman, R.J. M. L. Bach, F. H. Bach, G. B. Thurman, and K.
W. Sell. 1972. Precipitation of radioactively labelled
samples: A semi-automatic multiple-sample processor.
Cellular Immunology 4:182-186.

Hesser, ELF. 1960. Methods for routine fish hematology.
Prog. Fish-Cult. 22:164-170.

Hickey, CJL, Jr. 1976. Fish hematology, its uses and
significance. N.Y. Fish Game J. 23:170-175.

Horn, K.V., J. A. Singley, and W. Chavi’n. 1975. Fetal
bovine serum: a multivariate standard. Proc. Soc. Exp.
Biol. Med 149:344-349.

Hume, D.A., and M. J. Weideman. 1980. mm Lymphocyte
T h 820221222113 in lnnnneleex 1.0L
2. Elsevior/North-Holland Biomedical Press, Amsterdam.
251 P. '

Koller, LAD. 1979. Effects of environmental contaminants on
the immune system. Adv. Veterinary Sci. Compar. Med.
23:267-295.

Lee, LnF. 1978. Chicken lymphocyte stimulation by mitogens:
a microassay with whole-blood cultures. Avian Diseases
22(2):296-3070

Legendre, P. 1975. A field-trip technique for studying fish
leukocyte chromosomes. Can. J. 2001. 53:1443-1446.

Liewes, E.W., and R. H. Van Dam. 1982. Procedures and applica-
tion of the in yjtrg fish leukocyte stimulation assay. In:

Cooper (ed.). Innenelem end W ef £13.11. Dev.
Compar. Immunol. Supp. 1. Pergamon Press Ltd. p. 223-
232.

Luna, L.G. (ed.). 1968. Manual of Histological Staining
Methods of the Armed Forces Institute of Pathology.
McGraw-Hill Book Co., New York.

103

McLeay, D.J., and M. R. Gordon. 1977. Leucocrit: A simple
hematological technique for measuring acute stress in
salmonid fish, including stressful concentrations of
pulpmill effluent. J. Fish. Res. Board Can. 34:2164-
2175.

Moore, G.E., R. E. Garner, and H. A. Franklin. 1967. Culture
of normal human leukocytes. J. Amer. Med. Assn.

199:519-524.
Munkittrick, K.R., and J.F. Leatherland. 1983. Haematocrit
values in feral goldfish, garaaalaa aaxataa L., as

indicators of the health of the population. .L Fish
Biol. 23:153-161.

Oikari, A., and A. Soivio. 1975. Influence of sampling
methods and anaesthetization on various haematological
parameters of several teleosts. Aquaculture 6:171-180.

Paty, DAL, and D. Hughes. 1972. Lymphocyte transformation
using whole blood cultures: an analysis of responses.
J. Immunol. Methods 2:99-114.

Pauly, J.L., J. E. Sokal, and T. Han. 1973. Whole-blood
culture technique for functional studies of lymphocyte
reactivity to mitogens, antigens, and homologous lympho-
cytes. J. Lab. Clin. Med. 82(3):500-512.

Phillips, Huh 1973. Dye exclusion tests for cell viabili-
ty. In: P.K. Kruse and M.J. Patterson (eds.). 11mg

W Deepen: and Annlieetiens. Aoademio Press. NV.
NY. p 406-408.

Rasquin, P. 1951. Effects of carp pituatary and mammalian ACTH
on the endocrine and lymphoid systems of the teleost,

Aatlanax nexieane. J. Exp. 2001. 117:317-358.

Rohlf, F.J., and R. R. Sokal. 1969. Statiatical Tablas). W.
H. Freeman and Co., San Francisco, CA.

Rosenberg-Wiser, S., and R. R. Avtalion. 1982. The cells
involved in the immune response of fish. III. Culture
requirements of PHA-stimulated carp (gypzjnta gargig)
lymphocytes. Dev. Compar. Immunol. 6:693-702.

Sharma, RJL 1981. Splemic lympocyte transformation in
culture as a tool for immunotoxicological evaluation of

chemicals. In: R.P. Sharma (ed.), M Can;-
13 Taxicglggy Vol. II. CRC Press, Inc., Boca

Raton, Florida. p. 133-145.

Sigel, M.M., E.C. McKinney, and J.C. Lee. 1973. Fish
lymphocytes and blastogenesis. In: ELF. Kruse and M.K.

104

Patterson (eds.), 11mg Culture. Academic Press, N.Y.
pp. 135-1380

Sigel, M.M., J. C. Lee, E. C. McKinney, and D. M. Lopez.
1978. Cellular immunity in fish as measured by lympho-
cyte stimulation. Mar. Fish. Rev. 40(3):6-11.

Smith, A.M., N. A. Wivel, and M. Potter. 1970. Plasmacyto-
poiesis in the pronephros of the carp, (anzinua
9.3.2.0152). Anat. Rec. 167:351-370.

Sniesko, S.F. 1960. Microhematocrit as a tool in fishery
research. U.S. Dept. Interior Fish Wildl. Ser. Spec.
Sci. Rep.-Fisheries No. 341.

Soivio, A., and A. Oikari. 1976. Haematological effects of
stress on a teleost, Lag; 13313.: L. J. Fish Biol.
8:397-u11o

Sokal R.R., and F. J. Rohlf. 1969. My. W.H. Freeman
and Co., San Francisco, CA.

Sprent, J. 1977. Migration and lifespan of lymphocytes.
In: F. Loor and 0.5. Roelants (eds.), B and I 9311.: in,
Impede Beeennieien. Wilev. New York. pp. 59-82.

Stites, D.P., J. D. Stobo, H. H. Fudenberg, and J. V. Wells.
1982. 55.3.1.2 and .Qlinieel 11001111215121. Lance Medical
Publications. Los Altos, California. 775 p.

Tatner, M.F. and M. J. Manning. 1983. Growth of the lym-
phoid organs in rainbow trout, §a_1.m_q W, from one
to fifteen months of age. J. Zool., Lond. 199:503-520.

Viljanen, M.K. and J. Eskola. 1977. PPD-induced lymphocyte
transformation in 1.131.129 using whole blood. Clin. Immu-
nol. Immunopath 8:28-33.

Vos, JLG. and J. A. Moore. 1974. Suppression of cellular
immunity in rats and mice by maternal treatment with 2,
3, 7, 8-tetrachlorodibenzo-p-dioxin. Int. Arch. Allergy
Appl. Immunol. 47:777-794-

Vos, J.C. 1977. Immune suppression as related to toxicolo-
gy. CRC Crit. Rev. Toxicol. 5:67-101.

Vos, J.C. 1981. Screening and function tests to detect
immune suppression in toxicity studies. In: R.P.
Sharma (ed.). lnnnnelenie geneidecetiene in W o
Vol. II. CRC Press, Inc., Boca Raton, Florida. p. 109-
122.

105

Warr, G.W. and R. C. Simon. 1983. The mitogenic response
potential of lymphocytes from the rainbow trout (_$_a_l.mg
gairgnagl) re-examined. Dev. Comp. Immunol. 7:379-384.

Wedemeyer, G.A. and W. T. Yasatake. 1977. Clinical methods
for the assessment of the effects of environmental
stress on fish health. U.S. Dept. Interior Fish Wildl.
Ser. Tech. Pap. No. 89, 18 pp.

Wedemeyer, G.A., R. W. Gould, and W. T. Tasatake. 1983.
Some potentials and limits of the leucocrit as a fish
health assessment method. J. Fish Biol. 23:711-716.

Weinreb, E.L. 1959. Studies on the histology and histopath-
ology of rainbow trout (fialm W indius). II.
Effects of induced inflammation and cortisone treatment
on the digestive organs. Zoologica N.Y. 44:45-52.

Wolfe, K. and M.C. Quimby. 1969. Fish cell and tissue
culture. In: W.S. Hoar and D.J. Randall (eds.) £1311

Pnyaiglggx. Academic Press, N.Y., Vol. 3. p. 253-305.

Wolke, R.E., R. A. Murchelano, C. D. Dickstein, and C. J.
George. 1985. Preliminary evaluation of the use of
macrophase aggregates (MA) as fish health monitors.
Bull. Environ. Contam. Toxicol. 35:222-227.

Yasatake, W.T. and J. H. Wales. 1983. Microscopic Anatomy
of Salmonids: An Atlas. U.S. Fish and Wildlife Service
Res. Publ. No. 150, Wash., D.C.

Zeeman, M.C. and W. A. Brindley. 1981. Effects of toxic
agents upon fish immune systems: A review. In: R.P.
Shanna (ed.). Immnneleeie Qeneidenetiene in 19.121.221.221-
Vol. II. CRC Press Inc., Boca Raton, Florida. p. 1-60.