3 a: a... .. . {v.0 . o . To 83‘ d0
7. kg 1 .... . 5.15%... w... :5...
.{an 3.... . . .Qflmuzu.~m\km¢mm . 3.x... Atarfia»... ......
3...... . 91.3.. #5. in. .33....
. ink. .wuw..........b: P .1. . .... ‘ L .
.. n. in. 1...uvw........n.... v... v. . ..
fiflv *WS‘.............c................:_.... uh... .f.
.011.” doc \wuL-AN. .fl} c.3145;

.. , a. i. $3.5. . . .._. .
. whammy”... "Wtswzu .2 mwyrfi 5%“..an an...“ .a .
‘ .

 

 

 

 

   

  

‘. x

 

 
         

l‘lv: 4

.0...
u 7%.. nc.| H."
..

       

    
  

 

       

       

  

‘ —. u. . L
.

. ’.I}I~c .

. 0 pk: .OUov .
ho ..L.m..c.xu:..: . ‘ .r:
o .0 . 35!. 0A.. ‘ K rs.'-n_(5l

‘ 9....tv' or. u... . . v tn. I
n 0000-! Uviloulifl; VL»54;\Y§...I .wo.ro
% ~III. v.6.u < . ,4! 0‘4 nuvr
v.10..n7 ‘4. . \‘UONQ-k‘vd‘;
. . u.-

.. D" I.’ \‘I. h wl‘IOL\-O I
. . .
. 3:2: or 5.x. .
~ '-

     

       

 

 

... b, a.

             

‘ n

        

     

 

V
'D'l '
‘-

    

.|o i.

ci.\ cut 5..
-6 ' .f .c a >. .
.llvoaflfl Klrfl‘ooflvw." riolt .
o0... ‘IIV\|L.9 tpfiol.£'t.oi\ul.
$ AntOt 16‘.rtl|vb. A27
.b‘IQl-vt 1004.‘fl01... .0} 9‘9 5‘ .Q

vhtlngmI—ul‘u ‘9; ‘but‘.:
.tl . 7.3. II. . ‘ ,

.

   

 

n
_..
_-l

 

   

 

   

. . 3.
. . I. .otufilvvw .: o‘c.v1:.-L
A in I v v. :03].

.Ht‘.’ {IID'ILflHo'Ith'IOI‘ a. {with-an

   

   

  

. I .7.
3;... V 9 0,4,. ‘9‘. T

.Q: . I W. .4... 1.1%-.. IHJ...{.RK....JH~...).M....I ..

. Alt 1 I £3.,.lr§!..!.!.r:ufipi. ......-. .

      

      

(I... \Vbl O . ‘00" . u. 11
It [\N .| op .0!!! ‘9

9000... :-

  

  

‘0‘. .1:
a: 9‘

 

. .Jv.

 

            

no |
wt c o
I ...u t n ‘ ~ \ V ‘
:1“.in T.) q... .1131 ylxluuintklnwn
:E... .5355 £9-12. £,r.u..‘"v.|¢..
. v90.-. {igl‘u u .1. ..Ia.|.v .vyll‘
l .u.-.1 ufi or
.1 y|wt. A1. l
I ’96 ‘ v . .Il
:‘Mt‘a . til Oil-t!
. I {i .33.]...
; II .‘ 1:33

v’l AYv‘noAr .3... libflutlllcf
. A ‘ ‘1! .l

.. .o‘lvcy: .»ts’ LN.

‘.|-l ot 1st?

.u...‘.|«f.f. ”p.41

.9933]. r..Ll.v‘|.-I‘Ihrtdl.l

‘0 v I If:

. . II‘VA ’bo’lt‘.“

. ..b§0‘fiv,thfi“llxvu.il

. . . :voog‘solu‘ll. ‘0'::’! l

1.0 .i..-1a.o1.~lx.v.i uvu . 1 .lis'l‘! furl: 21!:

u 3.. 85.1. 3.!!! L. nulpol, 1.3;?! a
vlo~<nf.l¢:.\..v‘t3vl

‘ - 2.6...-

A". . 6.3%". 4.100.. v. on. \k . hon-avid ‘lfl
Z .11.‘ . > . .. . n. .. . . . . ... cl ( LIL-MD
3 {I . 5|): i.......}(:: ll: 3290!}-..35. azl... fizfldurfitprfib. .YJHH...J.. . . 1. . .213... .u...«Lt...fl...«1..|

Ira-.9. Ian 1.3.5... Ralf?! . :1 31:21? ivy...|.tll...u{.ll.2 o... :. . l . . .. - i In . DI i 1 .lo. 1
«IIHNH Yum... ennui”... . Filay: ‘l. )1»: . .n».w....=1.hv§n.l.hux. v .4321... .l...,. ob..n...lw milimvtthhA . .
. r . I .
. .

 

          

        

    

     

l

    

 

1‘.
‘l
0'»:le 5 111.12.]... .Ilunhlf. y: 00 {$9.31} fr.
4 t .u . ‘bulQ‘b ‘R A .T ‘V. 5(r-I. .1. . o‘VT rt nV-(,1I. _
. QIYAQLV‘V II. |.IJ..¢I.AY|I\.\:'|.O v.. .. va| 2- 3“. I‘ ‘ .. v. c ‘
tr... ! «1;dun...“.....12.....d.1¢.....:5w ‘1..‘...L$.‘d....c..¢h..:.. .1“. (:21. 5nd. 1 1:45!
0v '8: . cl!!!“ ll.fl9‘.0. .o‘f. Inflno‘pwdflflq'd n no... r>.~..(:dWIQI-o‘ho‘vK-A-N-‘LNWNJR 0v ‘N3‘ '9‘."
O. I! i-.‘ ‘ I . ‘10:!» 00‘ lo
0-! u .\o 0 é o\! s '1 .lufl.‘¢ - Q. ‘1 I La 0.50
Qua o u ‘
\nu I I

. 3::
a bk ‘ .4 - lq4onVo~oI ~ It! to ' l-n- Q ~\ 14”}..5 "V I
V E a“ €iv§9.l.. "ltv-u- pi ‘00“ o .‘v.4 I Nv‘ I ‘1‘LL 1“. C.
u r I 5! ‘t 1" ' Y] 1““ |
: .l: _1.25.5.~§ki¥nhr.r.vfl.§1j:.lb.5312?! 1....an pin-.13» .2. 3
.

  

I‘llu- Viiaz vol

I .. II; 1‘. . . in II
to. V . V . it: 41 1‘43123-hr‘a dwghphlt'mi

 

LIBRARY
Michigan State
University

 

 

 

This is to certify that the

dissertation entitled

Natural Killer Cell Activity
in Canine Lymphoid Malignancies

presented by

Rose Esther Raskin

has been accepted towards fulfillment
of the requirements for

PhoDo degree in PathOlogy

 

 

03%0/M

Ma?» professor

Date October 9; 1987

MS U is an Affirmative Action/Equal Opportunity Institution 0- 12771

 

 

 

 

MSU

LIBRARIES
.—_—.

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

NATURAL KILLER CELL ACTIVITY
IN CANINE LYMPHOID MALIGNANCIES

BY

Rose Esther Raskin

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Pathology

1987

ABSTRACT

NATURAL KILLER CELL ACTIVITY
IN CANINE LYMPHOID MALIGNANCIES

BY

Rose Esther Raskin

The research objective was to evaluate natural cell-
mediated cytotoxicity as an immunosurveillant mechanism in
spontaneously occurring canine lymphoma. Specific goals
involved morphologic, cytochemical and immunologic
characterization of an effector cell population containing
canine natural killer (NK) cells. Functional studies were
performed in healthy and lymphoma dogs.

Peripheral blood lymphocytes isolated by routine
density gradient centrifugation and nylon wool adherence
resulted in low numbers of large granular lymphocytes (LGL)
that morphologically resembled NK cells in people and
rodents. ‘Ultrastructurally, binding lymphocytes were LGL
with membrane bound granules in close association with
target cell membranes. These binding interactions were
similar to those of other species. Lymphocytes had villous
projections with minimal binding while extensive contact
involved smooth or flattened surfaces.

Cytochemically, effector cell populations were negative

for granulocytic markers but focally positive for acid

Rose Esther Raskin

phosphatase and non-specific esterases.

Direct immunofluorescent staining for surface
immunoglobulins was generally negative while cells bearing
Fc receptors for IgG (Peg) were present up to 26% of the
cells.

Overall NK cell activity in healthy dogs was measured
as percent cytotoxicity in 12 and 16 hour chromium release
assays using canine thyroid adenocarcinoma cells as targets.
A wide range of normal activity existed but levels from
individual dogs were generally constant. Percent
cytotoxicity correlated poorly with numbers of LGL but well
with FcG receptor positive cells. Morphologic features were
less reliable in predicting NK cell activity levels than
immunologic markers.

The relative frequency of active binding and cytotoxic
NK cells was determined using a single cell binding assay.
No correlation was found between percent dead binding cells
and percent cytotoxicity in healthy dogs.

Untreated lymphoma dogs had significantly lower percent
cytotoxicity but normal numbers of active binding and
cytotoxic effector cells compared to controls.
Glucocorticoid treated dogs had no change in NK cell
activity or in percent live or dead binding. It appeared
that canine NK cells were glucocorticoid resistant. In
contrast, combined chemotherapy protocols resulted in marked

inhibition of NK cell activity but effector cell numbers

Rose Esther Raskin

were unaffected. Thus, immune dysfunction in both untreated
and combined chemotherapy treated canine lymphoma patients

involved defective recycling of NK cells.

To my parents, brother and special friends whose loving
support and encouragement was essential in achieving this

goal.

To my cats, D.J. and Paws whose purring and affection
restored my sanity each evening and whose annoying morning

antics got me up on time each day.

ACKNOWLEDGMENTS

Special thanks to my major advisor, Dr. Harold Tvedten,
for his constant support, concern and endless patience plus
sense of humor during my residency and graduate work. His
tireless efforts in review of this dissertation were also
greatly appreciated.

Appreciation is extended to Dr. Robert Bull, my
research advisor, for use of his laboratory. I also wish to
thank the rest of the members of my graduate advisory
committee which included Drs. Steven Crow, Robert Dunstan
and Janver Krehbiel.

Crucial to the success of this research was the
technical assistance from Dr. Esther Roege, Ms. Irene Brett
and Ms. Peggy Coffman; statistical advice from Dr. Gregory
Fink and secretarial advice from Ms. Jamie Lyon. I would
also like to acknowledge the assistance of the clinic‘s
oncology staff for helping to provide patient materials.

Research and salary support were provided by the
Michigan State University Companion Animal Fund, Department
of Pathology and Animal Health Diagnostic Laboratory. Target
cells were kindly provided by Dr. Steven Krakowka of Ohio

State University.

vi

TABLE

LIST OF TABLES . . . . . .

LIST OF FIGURES . . . . .

LIST OF ABBREVIATIONS . .

INTRODUCTION . . . . . . .

CHAPTER 1

NATURAL CELLULAR CYTOTOXICITY AGAINST

A REVIEW OF THE LITE

Historical Perspective .

Assay Methods .
Target Cells . .

Cytolytic Effector Cells .
Cytolytic Mechanisms and Regu

CHAPTER 2

OF CONTENTS

RATURE . .

lati

on

MORPHOLOGIC, CYTOCHEMICAL AND IMMUNOLOGIC
CHARACTERIZATION OF CANINE NATURAL KILLER
CE LIS O O O O O O O O O O O O O O O O O 0

INTRODUCTION . . . . . . . . . . . . .
MATERIALS AND METHODS . . . . .
Animals and Lymphocyte Isolation
Light Microscopic Studies . . . .
Preparation of Cells for Electron
Microscopy . . . . . . . . . . . .
Immunofluorescent Staining of Effector
Cells . . . . . . . . . . . . . .
RESULTS . . . . . . . . . . . .
Isolation and Light Microscopic
Studies . . . . . . . . . . .
Ultrastructural Studies . . . . . .
Immunofluorescent Staining Studies
CONCLUSIONS . . . . . . . . . . . . . .
DISCUSSION . . . . . . . . . . . . . . .

vii

10
11
12

16

17
20
20
22

22

23
24

24
26
40
52
54

Page

CHAPTER 3
NATURAL KILLER CELL ACTIVITY IN NORMAL DOGS . . . 62
INTRODUCTION . . . . . . . . . . . . . . . . 63
MATERIALS AND METHODS . . . . . . . . . . . 65
Animals . . . . . . . . . . . . . . . 65
Effector and Target Cell Isolations . . 65
Light Microscopic Studies . . . . . . . 65
Chromium Release Microcytotoxicity
Assay . . . . . . . . . . . . . . 65
Single Cell Binding Assay . . . . . . . 66
Immunofluorescent Staining of
Effector Cells . . . . . . . . . . 67
Statistical Analysis . . . . . . . . . 68
RESULTS . . . . . . . . . . . . . . . . . . 68
CONCLUSIONS . . . . . . . . . . . . . . . . 78
DISCUSSION . . . . . . . . . . . . . . . . . 79
CHAPTER 4
NATURAL KILLER CELL ACTIVITY IN UNTREATED
AND TREATED CASES OF CANINE LYMPHOMA . . . . . 86
INTRODUCTION . . . . . . . . . . . . . . . . 87
MATERIALS AND METHODS . . . . . . . . . . . 91
Animals . . . . . . . 91
Effector and Target Cell Isolations . . 91
Chromium Release Microcytotoxicity
Assay . . . . . . . . . . . . 91
Single Cell Binding Assay . . . . . . . 92
Statistical Analysis . . . . . . . . . 92
RESULTS . . . . . . . . . . . . . . . . . . 93
CONCLUSIONS . . . . . . . . . . . . . . . . 98
DISCUSSION . . . . . . . . . . . . . . . . . 102
SUMMARY . . . . . . . . . . . . . . . . . . . . . . . 112
APPENDIX A

NATURAL KILLER CELL ACTIVITY IN NORMAL DOGS . . . 115
APPENDIX B
NATURAL KILLER CELL ACTIVITY IN UNTREATED AND
TREATED CASES OF CANINE LYMPHOMA . . . . . . 117

viii

Page
CHAPTER 3
NATURAL KILLER CELL ACTIVITY IN NORMAL DOGS . . . 62

INTRODUflION . C O O C O O O O O O O O I O O 6 3
MATERIALS AND METHODS . . . . . . . . . . . 65

Animals . . . . . . . . . . . . . . . 65
Effector and Target Cel Isolations . 65
Light Microscopic Studies . . . . . . . 65
Chromium Release Microcytotoxicity

Assay . . . . . . . . . . . . . . 65
Single Cell Binding Assay . . . . . . . 66

Immunofluorescent Staining of

Effector Cells . . . . . . . . . . 67
Statistical Analysis . . . . . . . . . 68
RESULTS . . . . . . . . . . . . . . . . . . 68
CONCLUSIONS . . . . . . . . . . . . . . . . 78
DISCUSSION . . . . . . . . . . . . . . . . . 79
CHAPTER 4
NATURAL KILLER CELL ACTIVITY IN UNTREATED

AND TREATED CASES OF CANINE LYMPHOMA . . . . . 86
INTRODUCTION . . . . . . . . . . . . . . . . 87
MATERIALS AND METHODS . . . . . . . . . . . 91
Animals . . . . . . . . . . . . . . . . 91
Effector and Target Cell Isolations . . 91

Chromium Release Microcytotoxicity
Assay . . . . . . . . . . . . . . 91
Single Cell Binding Assay . . . . . . . 92
Statistical Analysis . . . . . . . . . 92
RESULTS . . . . . . . . . . . . . . . . . . 93
CONCLUSIONS . . . . . . . . . . . . . . . . 98
DISCUSSION . . . . . . . . . . . . . . . . . 102

SMARY O O O O O O O O O O O O O O O O O O O O O O O 112

APPENDIX A
NATURAL KILLER CELL ACTIVITY IN NORMAL DOGS . . . 115

APPENDIX B
NATURAL KILLER CELL ACTIVITY IN UNTREATED AND
TREATED CASES OF CANINE LYMPHOMA . . . . . . 117

viii

Page

APPENDIX C
DETAILED TECHNIQUE PROCEDURES . . . . . . . . . . 125

LIST OF REFERENCES . . . . . . . . . . . . . . . . . . 135

VITA O O O O O O O O O O O O O O O I O O O O O O O O O 148

ix

LIST OF TABLES

Table Page

1 Frequency of lymphocytes and LGL in peripheral
blood smears and mononuclear isolations for 26
samples taken from healthy dogs . . . . . . . . . 25

2 Frequency of positive staining lymphocytes with
surface immunoglobulin (Ig) and Fe receptor for
IgG (FcGR) as related to the percentage of LGL
per isolated lymphocytes . . . . . . . . . . . . 50

3 NK cell activity in 26 samples from 18 healthy
dogs 0 O O O I O O O O O C O O O O O O O O O O O 6 9

4 Frequency of FcG receptor positive cells and
correlation to percent cytotoxicity, percent
live or dead cell binding and percent LGL . . . . 77

5 The group placement of 11 canine lymphoma
patients (I = untreated, II = glucocorticoid-
treated, III = combined chemotherapy-treated) . . 92

6 NK cell activity in canine lymphoma patients
versus paired and all controls . . . . . . . . . 94

7 Single cell binding activity in canine lymphoma
patients versus paired and all controls . . . . . 97

8 Degree of reduction in NK cell activity
following combined chemotherapy relative to
initial levels as compared to the initial NK
cell activity levels relative to normal
controls . . . . . . . . . . . . . . . . . . . . 101

9 NK cell activity of 26 samples from 18 healthy
dogs (raw data) . . . . . . . . . . . . . . . . . 115

10 Comparison between percent LGL, percent
cytotoxicity and percent live or dead binding
in 26 samples from 18 healthy dogs . . . . . . . 116

Table

11

12

13

14

15

16

17

Summary of clinical data for canine lymphoma
patients 0 O O O O O O O O O O O O O O O O O O 0

NK cell activity in untreated lymphoma patients
versus paired controls . . . . . . . . . . . .

NK cell activity in glucocorticoid-treated
canine lymphoma patients versus paired controls.

NK cell activity in combined chemotherapy-
treated canine lymphoma patients versus paired
contr°1 s O O O O O O O O O O O O O O O O O O O 0

Single cell binding activity in untreated canine
lymphoma patients versus paired controls . . . .

Single cell binding activity in glucocorticoid-
treated canine lymphoma patients versus paired
cont r0 1 s O O O O O O O O O O O O O O O O O O O 0

Single cell binding activity in combined

chemotherapy-treated canine lymphoma patients
versus paired controls . . . . . . . . . . . . .

xi

Page

117

119

120

121

122

123

124

LIST OF FIGURES

Figure Page

1 Granular lymphocyte present in the peripheral
blood from a clinically healthy dog. Wright's
stain O O O O O O O O O O O O O O O O O O O O O O 2 7

2 Cytocentrifuge preparation of isolated non-
adherent mononuclear cells with three granular
lymphocytes (arrows). Wright's stain. . . . . . . 27

3 Weak focal staining of isolated lymphocytes with
alpha naphthyl acetate esterase (arrows). . . . . 28

4 Strong focal paranuclear staining of isolated
lymphocytes with alpha naphthyl butyrate
esterase (arrows) . . . . . . . . . . . . . . . . 28

5 Focal (arrows) and diffuse (arrowheads) granular
staining of isolated lymphocytes with acid
phosphatase . . . . . . . . . . . . . . . . . . . 29

6 Granular lymphocyte with several short pseudopod
projections. Lead citrate-uranyl acetate stain;
x 18,360. 0 O O O O O O O O O O O O O O O O O O 0 3o

7 Deep clefted nucleus in a granular lymphocyte in
contact with target cells. Lead citrate-uranyl
acetate stain; X 18,360 . . . . . . . . . . . . . 31

8 Granular lymphocyte with cytoplasmic organelles
found within the indented portion of the nucleus.
Lead citrate-uranyl acetate stain; X 18,360 . . . 33

9 Gathered to one end of the cytoplasm are several
mitochondria (M), golgi apparatus (GA) and
variably dense granules (g). Lead citrate—uranyl
acetate stain; x 31,500 . . . . . . . . . . . . . 34

10 High magnification of granular lymphocyte with
mitochondria (M), centriole (arrow), membrane
bound granules (G) and myelin figures (MF). Lead
citrate-uranyl acetate stain: X 36,720 . . . . . 35

xii

Figure Page

11 Granular lymphocyte with irregularly shaped
nucleus, prominent nucleolus (N) and abundant
rough endoplasmic reticulum (arrows). Lead
citrate-uranyl acetate stain; X 18,360 . . . . . . 36

12 High magnification of granular lymphocyte with
variably dense granules and numerous vacuoles
(V). Lead citrate-uranyl acetate stain; X 36,720 . 37

13 Possible aggregation of ribosomes or a lysed
granule within the cytoplasm of a lymphocyte (L)
in contact with a target cell. Lead citrate-
uranyl acetate stain: X 36,720 . . . . . . . . . . 38

14 Lymphocyte bound to tumor target cell (T) with
increased focal density where cell membranes
touch (arrow). Incubation of 30 minutes. Lead
citrate-uranyl acetate stain; X 18,360 . . . . . . 39

15 Granular lymphocyte with smooth cell surface
bound extensively to a target cell. Cytoplasmic
granule located immediately adjacent to the
target cell membrane. Incubation of 10 hours.
Lead citrate-uranyl acetate stain; X 10,200 . . . 41

16 Isolated granular lymphocytes with numerous short
finger-like surface projections. Incubation of 30
minutes. Lead citrate-uranyl acetate stain:

X 10,200 . . . . . . . . . . . . . . . . . . . . . 42

17 Granular lymphocyte with long pseudopod surface
projections minimally touching a target cell.
Incubation of 12 hours. Lead citrate-uranyl
acetate stain; X 18,260 . . . . . . . . . . . . . 43

18 Granular lymphocyte with minimal binding to a
tumor target (T) via pseudopod projections of the
cell membrane. Incubation of 16 hours. Lead
citrate—uranyl acetate stain; X 16,110 . . . . . . 44

19 A granular and an agranular lymphocyte binding to
the same tumor target cell. Incubation of 16
hours. Lead citrate-uranyl acetate stain;
X 6,593 . . . . . . . . . . . . . . . . . . . . . 45

xiii

 

 

 

 

Figure

11

12

13

14

15

16

17

18

19

Granular lymphocyte with irregularly shaped
nucleus, prominent nucleolus (N) and abundant
rough endoplasmic reticulum (arrows). Lead
citrate-uranyl acetate stain; X 18,360 . . . . .

High magnification of granular lymphocyte with
variably dense granules and numerous vacuoles
(V). Lead citrate-uranyl acetate stain; X 36,720

Possible aggregation of ribosomes or a lysed
granule within the cytoplasm of a lymphocyte (L)
in contact with a target cell. Lead citrate-
uranyl acetate stain; X 36,720 . . . . . . . . .

Lymphocyte bound to tumor target cell (T) with
increased focal density where cell membranes
touch (arrow). Incubation of 30 minutes. Lead
citrate-uranyl acetate stain; X 18,360 . . . . .

Granular lymphocyte with smooth cell surface
bound extensively to a target cell. Cytoplasmic
granule located immediately adjacent to the
target cell membrane. Incubation of 10 hours.
Lead citrate-uranyl acetate stain; X 10,200 . .

Isolated granular lymphocytes with numerous short
finger-like surface projections. Incubation of 30
minutes. Lead citrate-uranyl acetate stain:

x 10,200 . . . . . . . . . . . . . . . . . . . .

Granular lymphocyte with long pseudopod surface
projections minimally touching a target cell.
Incubation of 12 hours. Lead citrate-uranyl
acetate stain; X 18,260 . . . . . . . . . . . .

Granular lymphocyte with minimal binding to a
tumor target (T) via pseudopod projections of the
cell membrane. Incubation of 16 hours. Lead
citrate-uranyl acetate stain; X 16,110 . . . . .

A granular and an agranular lymphocyte binding to
the same tumor target cell. Incubation of 16
hours. Lead citrate-uranyl acetate stain;

X 6,593 . . . . . . . . . . . . . . . . . . . .

xiii

Page

36

37

38

39

41

42

43

44

45

Figure

20

21

22

23

24

25

26

27

Cytoplasmic granules located to a side furthest
from target cell in a lymphocyte with minimal
binding. Incubation of 16 hours. Lead citrate-
uranyl acetate stain; X 8,950 . . . . . . . . .

Granules are present between the lymphocyte
nucleus and the target cell to which the
lymphocyte has minimal attachment. Incubation of
2 hours. Lead citrate-uranyl acetate stain;

X 27,450 . . . . . . . . . . . . . . . . . . . .

Granule placement directly adjacent to target
cell membrane (T) in a lymphocyte with extensive
binding. Incubation of 2 hours. Lead citrate-
uranyl acetate stain; X 18,360 . . . . . . . . .

Coalescence of several granules and their
location immediately adjacent to the target cell

to which the lymphocyte has extensively attached.

Incubation of 16 hours. Lead citrate~uranyl
acetate stain; X 25,704 . . . . . . . . . . . .

Positive staining of 2 lymphocytes (L) and 2
neutrophils. Direct immunofluorescence staining
for EC receptor of IgG; X 40 . . . . . . . . . .

Percent cytotoxicity in healthy adult dogs at

various effector to target cell ratios for 12 and

16 hours of incubation. Number of samples tested
per incubation time is given in parentheses . .

Comparison of percent cytotoxicity at various
incubation times and effector to target cell
ratios for males, intact and neutered females.
Number of samples per group is given in
parentheses . . . . . . . . . . . . . . . . . .

Samplings of 5 dogs taken twice over 1-3 months.
Percent cytotoxicity determined at 12 hours

incubation with various effector to target cell
ratios. Paired symbols identify individual dogs

xiv

Page

46

47

48

49

51

72

74

76

Figure Page

28

29

NK cell activity in percent cytotoxicity (mean

plus standard error) for canine lymphoma

patients and paired controls. Horizontal area

with vertical hatching represents the mean +/- 1
standard error for all controls . . . . . . . . . 96

Percent live target binding to lymphocytes (mean

plus standard error) for canine lymphoma patients

and paired controls. Horizontal area with

vertical hatching represents the mean +/- 1

standard error for all controls . . . . . . . . . 100

ACP

ADCC
ANAE
ANBE

CHOP

COP
CPM
CTAC
CTL
E:T
F CG
FCGR

LCFC
LGL

MEM
NCI
NK
NKCF
PAS
PBS
PLL
RI
SCBA
SEM
WHO

LIST OF ABBREVIATIONS

acid phosphatase

antibody dependent cellular cytotoxicity

alpha naphthyl acetate esterase

alpha naphthyl butyrate esterase

chloroacetate esterase

cyclophosphamide, doxorubicin, vincristine
prednisone

cyclophosphamide, vincristine, prednisone

counts per minute

canine thyroid adenocarcinoma

cytolytic T lymphocyte

effector to target cell ratio

Fc portion of immunoglobulin G

receptor for FcG

immunoglobulin

killer

lytic conjugate forming cells

large granular lymphocyte(s)

lytic units

minimal essential medium

National Cancer Institute

natural killer

natural killer cytotoxic factor

periodic acid-Schiff

phosphate buffered saline

poly-L-lysine

refractive index

single cell binding assay

standard error of the mean

World Health Orgnaization

xvi

INTRODUCTION

Several specific and nonspecific mechanisms are
involved in the immunosurveillance of tumor cells (Trainin &
Essex, 1982; Wheelock & Robinson, 1983). Specific reactions
include cytolytic T-lymphocytes or tumor specific antibodies
with or without complement activation. Macrophages and
antibody dependent cellular cytotoxicity may act in both a
specific and nonspecific manner. Natural killer (NK) cells
act rapidly within hours, nonspecifically and spontaneously
against a wide variety of targets. They do not require prior
sensitization, presence of antibody or complement and show
no restriction related to the major histocompatibility
complex (MHC) (Herberman &‘ Ortaldo, 1981; Trinchieri &
Perussia, 1984).

NK cells lack many of the characteristics of T
lymphocytes (e.g., thymic dependency, T cell surface
markers) or of B lymphocytes (e.g., surface immunoglobulin)
or of macrophages (e.g., adherence, phagocytosis) (Herberman
& Ortaldo, 1981; Trinchieri & Perussia, 1984). NK cell
activity has been closely related to a subpopulation of
_lymphocytes, morphologically identified as large granular

lymphocytes (LGL) (Timonen g; 11., 1979a,b). In vivo, NK

2
cell activity has been associated with early tumor cell
elimination prior to metastasis as well as participation in
cytolysis of virus infected cells, graft rejection,
hematopoietic stem cell regulation and other general
immunoregulatory functions (Herberman, 1983; Trinchieri &
Perussia, 1984).

NK cells have been well characterized in humans and
mice and to a lesser extent in other species.
Characterization of these effector cells in the dog has been
infrequent with reference only to their surface markers
(Ringler & Krakowka, 1985; Betton, 1980). One aim of the
present study was to further characterize the morphologic,
cytochemical, immunologic and functional aspects of these
cells in healthy dogs.

In beige mice, an animal homologue of the human
Chediak-Higashi syndrome, NK cell activity is known to be
profoundly reduced. Lymphoid malignancies grew faster and at
higher frequencies in this mouse strain than in normal
littermates (Karre gt a_l., 1980). Studies in people with
lymphoma or leukemia have demonstrated reduced NK cell
activity (Tursz gt 21., 1982; Nasrallah & Miale, 1983; Kay &
Sinkovics, 1984). This activity increased during remission
of disease (Yoda gt _a_l., 1983; Lotzova g3; gl., 1986) and
with administration of biological response modifiers such as
interferon or interleukin 2 (Hooper gt gl., 1986).

Evaluation of the significance of these changes in NK cell

3
activity during the course of disease or treatment must be
coupled with the effect of chemotherapeutic agents commonly
used to treat these conditions. Early studies indicate that
cytoreductive chemotherapy produces either increases or
decreases in NK cell activity depending upon the level of
immunocompetence of patients prior to treatment (Braun &
Harris, 1986). Due to the close relationship between NK cell
activity and lymphoproliferative malignancies in other
species, a second aim of this study involved the evaluation
of this activity in untreated and treated cases of canine

malignant lymphoma. If in vitro measurement correlates well

 

with 12 1129 activity, then NK cell assays may be
beneficial in determining response to treatment or prognosis
for survival.

It is hoped that future applications of this current
work would aid in the understanding and treatment of canine
and human lymphomas. Canine lymphoma is a frequent,
spontaneously occurring disease that clinically and
pathologically resembles its human counterpart (Crow, 1982;
Carter gt gl., 1986). Treatment protocols for lymphoma are
similar for the two species. Therefore it is possible that
the canine disease could assist future studies involving

biological response modifiers for the treatment of lymphoma.

CHAPTER 1

NATURAL CELLULAR CYTOTOXICITY AGAINST TUMORS:

A REVIEW OF THE LITERATURE

 

 

 

 

1)
tu
tu.
(W,
prc

tum

ind
Cytc
Thes

 

NATURAL CELLULAR CYTOTOXICITY AGAINST TUMORS:

A REVIEW OF THE LITERATURE

H' ! . J E !'v

During the 1960's many investigators pursued the
premise that cell mediated mechanisms were involved in tumor
defense. Researchers sought to demonstrate that specific
tumor associated antigens provided the recognition mechanism
for this process. Hellstrom gt g1. (1967, 1968, 1971) using
colony inhibition and visual cytotoxicity assays
demonstrated both humoral and cell mediated activity of
lymphoid cells against murine and human induced and cultured
tumor cells. Several years later a similar demonstration of
tumor associated antigens was made in dogs with lymphoma
(Warren gt a_l., 1975). These specific cell mediated
processes were considered at the time the major mechanism of
tumor defense.

During this time, experimental observations began to
indicate that besides specific lymphocyte mediated
cytolysis, non-specific cytolytic processes were occurring.
These had been originally considered to be technical errors

or artifacts. This involved reactivity by cells of normal

6

controls during trials investigating the cellular activity
of tumor patients. Rosenberg gt g1. (1972) was among the
first to report this reactivity by normal human cells
against fresh, noncultured leukemia cells. He used identical
twins in which one had leukemia for a genetically controlled
study. Lymphocyte mediated cytotoxicity by cells of the
nonleukemic twin was demonstrated against leukemia cells
from the diseased twin. This activity was attributed to
possible prior sensitization to leukemia associated
antigens. Soon after, other reports followed that supported
the occurrence of cytotoxicity by normal cells. Takasugi gt
g1. (1973) noted that lymphocytes from normal individuals
possessed higher reactivity against tumor cell lines than
those of tumor patients having the same histologic cell type
as the target cells. This suggested prior sensitization was
not a prerequisite for reactivity. Kay & Sinkovics (1974)
similarly found that healthy human subjects possessed
lymphocytes that could reduce the number of cultured tumor
cells measured in a test system. This study suggested the
existence of immunosurveillance in people.

Further characterization of this process continued
using a variety of tumor targets, most of which gave similar
results. These experiments suggested the presence of a
natural or spontaneous cytolytic process which neither
required prior immunologic sensitization or the presence of

antibody or complement. De Vries gt g1. (1974) demonstrated

7

that the non-T cell fractions from normal individuals and
melanoma patients were more cytotoxic than the T cell
fractions or total lymphoid population. Initially these
cells were classified as "null cells" i.e., lacking
characteristic markers of either T or B lymphocytes
(Herberman gt g_., 1979). Later studies would indicate that
they share some of the markers identifying T cells and
myelomonocytic cells (Trinchieri & Perussia, 1984).

Simultaneous to the human studies, similar observations
of natural cytotoxicity were reported in rats and mice. The
term ”natural killer cell" was generally accepted, when
Kiessling gt _1. (1975a, 1975b) and Herberman gt 31. (1975a,
1975b) independently characterized the first NK cells, those

of the mouse spleen.

Assay Methods

Early investigations of cell mediated cytotoxicity
utilized visual assays of colony inhibition and
microcytotoxicity to evaluate tumor systems. Cell numbers
were calculated following visual inspection. Detection of
colony reduction or lysis of tumor cells by these methods
was tedious and time consuming.

Brunner gt, 11. (1968) developed a sensitive
quantitative radiolabel technique which involved chromium-51

release in a microcytotoxicity assay to measure cell death.

8

The chromium release microcytotoxicity assay has since
become the standard testing procedure. This involves placing
a fixed number of radiolabelled tumor cells together with
variable numbers of effector cells into tubes or microtiter
plates. When killed or lyzed, the target cells release their
radiolabel into the supernatant which can be measured as
counts per minute (cpm) by a gamma counter. A direct
relationship exists between the amount of radiolabel
released and the number of effector cells present since
increases in effector to target cell ratios (E:T) produce an
increase in the amount of Cr51 measured.

A qualitative calculation of percent cytotoxicity can
be used to define natural killer activity within various
situations. Another commonly used calculation of NK cell
activity is that of lytic units. These are semi-quantitative
and require the use of multiple E:T ratios. The units are
mathematically derived and define the number of cells
necessary to produce a predetermined percentage of
cytolysis.

The radiolabel release method measures the overall
activity of the cells and considers their recycling ability
(Pross gt gl., 1986). Because cytotoxic lymphocytes are not
destroyed during cytolysis, they may cause death of
additional target cells through a stage in, the process
termed recycling.

A second commonly used assay for cellular cytotoxicity

9
studies is the single cell binding assay (SCBA). The
advantage of this method is the prevention of cell
recycling. This is accomplished by allowing effector cells
and targets to form conjugates which can then be immobilized
in a medium such as agarose (Grimm & Bonavida, 1979:
Timonen g _a_l_., 1982) or poly-L—lysine (Vargas-Cortes gt
g;., 1983). This simple method allows the direct microscopic
visualization of cell binding and cytolysis. Calculations
such as percent live binding or percent cytolysis are
determined by counting the number of cell conjugates having
target cells that respectively, have not or have stained
with trypan blue, a dye that labels dead cells. A high
correlation was found between lytic conjugate forming cells
(LCFC) frequency i.e., the percentage of lymphocytes bound
to dead targets and NK cell cytotoxicity defined by the
chromium release assay, suggesting that LCFC are NK cells
(Rubin gt gl., 1982). However, Rubin gt gt. (1982) noted a
low correlation between total conjugate forming cells and NK
cytotoxicity implying that all lymphocytes that bound to
target cells were not NK cells. The SCBA provides a semi-
quantitative estimate of the frequency of NK cells active
against the target cell tested (Pross gt g;., 1986).
Combination of the results from chromium release and single
cell binding assays provides information about the recycling
and target binding capacity of NK cells (Ullberg 8 Jondal,

1981).

10

Wells

The target cells used in natural killer cell assays
have been variable but all were identified by the ability of
NK cells to spontaneously lyze these tumor or virus-infected
cells. Although this activity is not entirely species
specific, optimal response is obtained with allogeneic
target cells. For human studies, K562 cells, an
erythroleukemic cell line derived from a chronic
myelogeneous leukemia case in blast crisis, has become the
standard target. Mouse studies involve YAC-l, a cultured
murine lymphoma induced by Maloney leukemia virus.

Various cell lines including heterologeous K562 and
Vera (African Green monkey kidney) or homologenous canine
kidney, melanoma, thyroid adenocarcinoma (CTAC) and canine
distemper virus infected cells have been used in cell
mediated cytotoxicity studies in dogs (Betton, 1980; Shek gt
1.1., 1980; Krakowka, 1983; Ringler & Krakowka, 1985). Of
all of these, the CTAC appears to be the most sensitive for
use in dog studies (Krakowka, 1983). It was cloned from a
single cell and is characterized by adherence and an
epithelioid syncytial growth pattern (Kazna, 1964). Use of
this cell line requires trypsinization which could affect
target antigens (Pross gt g;., 1986). However, this problem

has not been encountered.

11
WW

There are several types of naturally occurring
lymphocyte effector cells involved in the immune response to
tumor cells (Trainin 8 Essex, 1982). These may be subdivided
into those requiring major histocompatibility complex (MHC)
antigens for target cell recognition and those that are non-
MHC restricted. The MHC-restricted lymphocytes possess the T
cell antigen receptor with rearrangement and transcription
of the genes for it confirming the T cell origin of these
cells (Marrack 8 Kappler, 1986; Lanier gt g1“, 1986). The
cytolytic T lymphocyte (CTL) requires a class I MHC antigen
and prior sensitization to the target cell for recognition
to occur. The cytolytic process is slow requiring several
days such that it is thought to constitute the second line
of tumor defense (Herberman, 1983).

The second group of cytolytic lymphocytes are non-MHC
restricted and are frequently classified by function. This
involves the requirement of antibody for target cell
recognition. The killer (K) cell of antibody-dependent
cellular cytotoxicity (ADCC) has Fc receptors for the IgG
molecule which bind to the‘ antibody coated target cells
(Djeu, 1982). The NK cell also has Fc receptors but antibody
is not necessary for target cell binding to occur. Treatment
to remove antibody does not reduce cytolysis by NK cells of

sensitive targets. The K and NK cells share in common some

12

surface 'markers and. the formation. of pores that induce
target cell damage (Muchmore, 1986). Additional evidence to
suggest these are the same cell involve the demonstration of
a single effector cell binding to and killing both NK and
ADCC targets (Bradley 8 Bonavida, 1982). On the other hand,
there is a temporal difference in their appearance in
gnotobiotic pigs. ADCC was fully expressed by 2 weeks
whereas NK cell activity was not apparent until 6 weeks of
age. However, it is also possible that NR and ADCC activity
are mediated by the same cell at different stages of
maturation (Muchmore, 1986).

The non-MHC restricted lymphocytes may also be
classified by surface markers. Lanier gt gt. (1986) proposed
dividing these cells into two groups. The most abundant
effectors of non-MHC-restricted cytotoxicity would be
distinct from T cells and include NK cells and those with
NK-like activity. A smaller group would include those

possessing T cell receptor gene rearrangements.

C o c Mec an'sms nd e tio

The main stages of the lymphocyte mediated cytolytic
process include: (i) recognition and binding of target
cells; (ii) activation and release of cytotoxic factors; and
(iii) target cell lysis and effector cell recycling

(Henkart, 1985) . Recognition mechanisms were characterized

13

for the various effector cells in the previous section.
Target recognition receptors for K cells involves functional
Fc receptors which require antibody to be present. Unlike
CTLs, the T cell receptor is not involved. in INK. cell
recognition mechanisms as gene transcripts of the receptor
are lacking (Tutt gt _a_l., 1986). A possible recognition
mechanism. may involve laminin, a membrane glycoprotein,
since it has been identified on NK cells and receptors for
it. have been reported occurring’ on. NKI sensitive target
cells. Blockage of these receptor sites on target cells or
anti-laminin treatment of NK cells has resulted in reduced
binding and cytolytic activity (Hiserodt gt g1., 1985).
Other target cell antigens proposed to attract NK cells
include those involved in the differentiation process as
mature cells are insensitive to NK lysis and immature cells
are more susceptible (Herberman, 1985). Storkus gt .4;.
(1987) reported an inverse correlation between NK conjugate
or binding formation and the expression of MHC class I
antigens on the surface of target cells. This expression may
be increased by agents such as interferon (Hansson gt g;.,
1980). Much research is presently directed towards answering
this question.

Binding of NK cells to targets is necessary for
cytolysis to occur, and has been shown to require Mgz". A
Ca2+ dependent activation stage follows which is sensitive

to a number of pharmacologic agents (Trinchieri 8 Perussia,

Mm“.

 

 

 

 

14

1984) . The ion involvement lends support to a secretory
process triggered by membrane receptors as the mechanism of
cytolysis (Henkart, 1985). Targan and Newman (1983) reported
that binding alone does not trigger or activate the lytic
event. Using a monoclonal antibody against the T-200
glycoprotein, they were able to fully inhibit NK cell
activity following conjugate formation.

Several of the post-binding stages are similar for the
process of cytolysis by NK, K and CTL cells (Henkart, 1985;
Bonavida 8 Wright, 1986; Muchmore, 1986). Following
activation, granule exocytosis occurs. The extent to which
these granules perform in the cytolytic process is not fully
understood. Ultrastructurally, these granules orientate
towards the target cell, fuse with each other and release
their contents in the space between the two cells (Henkart,
1985).

Two types of soluble factors are associated with the
granules. One has been termed pore forming proteins and
consists of perforins or cytolysin (Marx, 1986). These
effect polymerization of proteins forming a tubular
structure that inserts into the target cell membrane. This
has been viewed ultrastructurally as thickened rings on the
cell surface akin to the complement attack complex (Henkart,
1985). A protein resembling perforin has recently been
isolated from the cytoplasmic granules that is

immunochemically related to the ninth component of human

 

 

15

complement (Zalman gt g1., 1986). The second type of factor
is a glycoprotein termed NK cytotoxic factor (NKCF) . The
similarities between NKCF mode of activity and that observed
in NK cell reactions suggest that the factor is closely
involved in NK cell activity (Bonavida 8 Wright, 1986). It
is suggested that NK cell resistant targets are such because
they lack receptor sites for NKCF (Kiyohara gt g1., 1986).

The last stage in the cytolytic process involves target
cell lysis and recycling of the effector cell. Target cell
death may be related to colloid osmotic lysis due to the
release of small ions or to the insertion of pore forming
proteins into cell membranes (Henkart, 1985). This final
step, termed lethal hit, is independent of the effector
cell. The NK cell is not destroyed in the cytolytic process
and may recycle to kill again following a refractory period
(Timonen gt g;., 1982).

Regulation of NK cell cytolytic activity is mostly
associated with interferon and interleukin-2 (Djeu gt g1.,
1979; Henney gt g;., 1981). Interferon is known to activate
pre-NK cells to recognize and bind targets, increase binding
rate and reduce the refractory period for recycling of NK
cells (Timonen gt g1., 1982). In addition, it increases NKCF
release (Bonavida 8 Wright, 1986). Interleukin-2 has also
been shown to booster NK cell activity, possibly due to

increased NKCF release (Bonavida gt g_., 1986).

CHAPTER 2

MORPHOLOGIC, CYTOCHEMICAL AND IMMUNOLOGIC

CHARACTERIZATION OF CANINE NATURAL KILLER CELLS

MORPHOLOGIC, CYTOCHEMICAL AND IMMUNOLOGIC

CHARACTERIZATION OF CANINE NATURAL KILLER CELIS

INTRODUCTION

Natural killer (NK) cells have been defined in terms of
their function of spontaneous cytotoxicity to a wide variety
of targets including virus-infected cells, tumor cells and
some poorly differentiated cells (Herberman, 1983). NK cells
along with cytotoxic T lymphocytes, specific antibodies,
macrophages and antibody dependent cellular cytotoxicity
killer lymphocytes represent the major mechanisms involved
in anti-tumor immunity (Trainin 8 Essex, 1982).

The morphology of cells associated with this functional
activity has been well described in humans and rodents
(Ferrarini gt g_., 1980; Reynolds gt at, 1981; Timonen gt
11., 1979a,b; Saksela g 91., 1982; Gastl gt Q1., 1983).
There is strong correlation between NK cell activity and
cells which are termed large granular lymphocytes (LGL)
(Timonen gt g1., 1979a; Savary 8 Lotzova, 1986). Up to 80%
of LGL can function as NK cells (Timonen g 11., 1982). In
Wright's or Giemsa stained preparations these cells are

medium to large sized lymphocytes having variably numbered

17

18
small azurophilic stained cytoplasmic granules.

The LGL has been identified to a variable extent in
peripheral blood, bone marrow, lung, intestine and major
lymphoid structures such as lymph nodes, spleen and thymus
(Savary and Lotzova, 1986). They may be found in peripheral
blood smears of normal human subjects between 2-6% of the
total leukocyte population (Timonen gt gl., 1981).

Ultrastructurally, the granules most resemble the
lysosomal structures seen in granulocytes (Grossi 8
Ferrarini, 1982) . However, others report parallel tubular
arrays in association with these granular structures (Huhn
gt g_., 1982; Zucker-Franklin gt a_l., 1983; Payne, 1984).
Despite disagreement regarding morphology, it is generally
regarded that the granules contain cytotoxic factors that
contribute significantly to the cytolytic process (Henkart,
1985; Bonavida 8 Wright, 1986).

Binding between NK cells and tumor targets is necessary
as the initial step in cytolysis. This activity usually
involves close physical contact between the two cells (Roder
gt g_l_., 1978; Hiserodt 8 Beals, 1985). It is thought that
orientation of the granules occurs towards the target
surface followed by release of their lytic factors (Henkart,
1985). Circular lesions have been observed ultrastructurally
on the surface of the target cell due to polymers similar to
those of the complement system (Podack 8 Dennert, 1983;

Zalman gt a_l., 1986). It is possible that these circular

19
lesions or pores provide a mechanism for target cell death
through loss of ions or insertion of cytotoxins.

Cytochemically, the granules stain similarly to
lysosomes being strongly positive for acid phosphatase.
Alpha naphthyl acetate esterase is weakly positive while
another nonspecific esterase, alpha naphthyl butyrate
esterase, is negative in humans and rodents (Ferrarini 8
Grossi, 1986). In addition, chloroacetate esterase, a
granulocytic marker stain has been reported positive in some
human LGLs (Heumann gt gl., 1983). However, other
granulocytic marker stains such as peroxidase, are negative
(Huhn _et a1” 1982) .

NK cells are heterogenous with respect to their surface
markers. They are not considered purely B— or T-lymphocytes
or myelomonocytic in origin although they share in common
some of the markers found on these cell types. This suggests
their origin is different but related to these cell lines.
Characteristic immunologic features associated with human or
rodent NK cells include non-adherence, non-phagocytosis,
absence of surface immunoglobulin (Ig), presence of
receptors for the Fc portion of IgG (FcG) and a tendency to
form low affinity rosettes with sheep erythrocytes
(Trinchieri 8 Perussia, 1984).

Most information known about canine NK cells has been
associated with functional or immunologic marker studies

(Betton 8 Gorman, 1978; Betton, 1980; Shek e_t gt” 1980;

20

Krakowka, 1983; Ringler 8 Krakowka, 1985a,b; Savary 8
Lotzova, 1986). There is one account of the light
microscopic appearance (Savary 8 Lotzova, 1986) and none
that involve cytochemical staining of the effector cells.

The purpose of this report will be to further
characterize the morphologic, cytochemical and surface
marker features of the effector cells associated with NK
cell activity in the dog. These peripheral blood lymphocytes
will be evaluated for their light and electron microscopic
features. In particular, the percentage of LGL in fresh
blood and in isolated lymphoid cell populations will be
determined. LGL will be evaluated ultrastructurally to
assess the composition of the granules. Electron microscopy
will also be used to visualize the cell membrane interaction
between effector and. target cells. In laddition, routine
cytochemical stains will be applied to the effector cell
population. The results of these studies will be compared to

those found in other species, notably rodents and people.

MATERIALS AND METHODS

A a s d ho solatio

A total of 18 healthy appearing adult pure and mixed
breed dogs of both sexes were used. These had been housed in
the Veterinary Clinical Center at Michigan State University.

.A standard lymphocyte isolation procedure to obtain canine

21
peripheral blood NK cells (Krakowka, 1983) was used with
slight modification. Twenty milliliters of heparinized whole
blood were collected from each subject, seven of which were
sampled more than once. Blood samples were incubated at 37 C
for 1 hour with carbonyl iron filings to assist in the
removal of phagocytic cells. This blood was then diluted 1:5
with saline and layered over Ficoll-Hypaque solution (RI
1.3550). This solution is a high molecular weight polymer of
sucrose (Ficoll) with sodium diatrizoate used to separate
mononuclear cells from granulocytes and erythrocytes.
Mononuclear cells were recovered by centrifugation of this
solution at 450 g for 18 minutes. Contaminating red cells
were removed by hypotonic shock lysis and remaining cells
were washed with saline. Cells were resuspended in minimal
essential medium (MEM) supplemented with 10% (v/v) fetal
calf serum, 1% (v/v) glutamine and 1% (v/v) penicillin-
streptomycin. Adherent cells were removed by incubation at
37 C for 1 hour in a 10 ml syringe packed with 0.75g nylon
wool fiber (Fenwal Lab., Deerfield, IL). Non-adherent cells
were eluted using 15 ml of warm MEM without additives. The
final cell suspension was adjusted to 5.0 X 106 cells/ml.
This population of isolated lymphocytes demonstrated
spontaneous cytotoxicity in routine NK cell assays (data not
shown). Therefore these effector cells contained a

functional NK cell population.

22
s o s u 3

To evaluate the percentage of LGL in the initial
peripheral blood sample, smears were made from each sample
prior to further processing. The morphologic appearance of
the effector cell population was also examined from slide
preparations made from 5.0 X 105 cells with a cytocentrifuge
(Shandon Southern Instruments Inc, Sewickley, PA). All
slides were air-dried. One blood smear and one cytospin
slide were stained. with. a- modified. Wright-Giemsa stain
(Hema—tek, Ames Co, Elkhart, IN). Cytochemical staining was
performed on additional cytospin preparations which included
the following stains: peroxidase, naphthol AS-D
chloroacetate esterase (CAE) (Sigma Diagnostics, St. Louis,
MO) , alpha napthyl acetate esterase (ANAE) , alpha napthyl
butyrate esterase (ANBE), alkaline phosphatase, acid
phosphatase (ACP), periodic acid-Schiff (PAS) and Sudan
black B. The techniques for these stains have been fully
described elsewhere (Barka 8 Anderson, 1962; Jain, 1986).

Cell differentials were performed on a minimum of 100 cells.

e at on o ce for e tron crosco
The ultrastructural morphology of the effector cells
alone and those involved in target cell interactions were
examined following processing. Canine thyroid adenocarcinoma
(CTAC), a cell line maintained as monolayer cultures (Kasza,

1964), was kindly provided by Dr. Steven Krakowka (The Ohio

23

State University). The target cells were trypsinized prior
to use and concentrations were adjusted to 1.0 X 106
cells/ml. Non-adherent effector cells isolated as previously
described were also adjusted to a concentration of 1.0 X 106
cells/ml. An equal amount (1 ml) of each cell suspension was
placed together, pelleted and allowed to incubate at 37 C
for variable times. These times included: 0.5, 2, 8, 10, 12
and 16 hours.

Samples containing isolated lymphocytes (5.0 X 106
cells) and 1:1 mixtures of lymphocytes and target cells were
processed similarly following pellet formation. Karnovsky's
fixative (pH 7.3) was added to the pellet and left for 2
hours. Samples were rinsed in Zitterquist wash and post-
fixed in 1% osmium tetroxide. Following routine dehydration,
preparations were embedded in DDSA + Araldite 502. Thin
sections were stained with uranyl acetate and lead citrate
and examined in a transmission electron microscope (Zeiss EM

952).

Immunofluorescent staining of effector cells

A direct immunofluorescent procedure was applied to
cell suspensions to detect surface staining. Fluorescein
conjugated F(ab')2 fragments were used that included goat
anti-dog-IgM and goat anti-dog-IgG (Cooper Biomedical Inc,
Malvern, PA) for the detection of surface IgM and IgG

respectively. In addition, to detect the presence of Fc

24
receptors for IgG, fluorescein conjugated Fc fragment for
dog IgG (Jackson Immunoresearch Laboratories Inc, Avondale,
PA) was used. Dilutions (1:10) were made of the antibodies
and 0.1ml of each was added to 1.5 X 106 cells of the non-
adherent lymphocyte population. Incubation for 1 hour at 4 C
was followed by 2 phosphate buffered saline (PBS) washes.
Cells were resuspended in PBS, then transferred to a glass
slide and examined under a fluorescent microscope.
Mononuclear cells were counted and positive cells were

expressed as a percentage of the total number.

RESULTS

Isolattog and light microscopic studies

Twenty-six samples from the 18 dogs were evaluated for
the proportion of lymphocytes and LGL present in peripheral
blood smears and cytocentrifuge preparations of isolated
non-adherent mononuclears. Data is presented in Table 1.
Four isolation samples with total lymphocyte populations
below 80% were contaminated to a moderate degree with
neutrophils and eosinophils. Monocytes accounted for less
than 1% in each of the lymphocyte isolated preparations.

Morphologically, the LGL was recognized by a larger
size (10-15 um) compared to other well-differentiated
lymphocytes. The cytoplasm was moderately abundant and pale

blue with Romanowsky-type stains. Present within the

25

Table 1 - Frequency of lymphocytes and LGL in peripheral
blood smears and mononuclear isolations for 26 samples taken
from healthy dogs.

 

WW
Lymphocytes (%)

Mean 33.6 90.1

Range 21-54 64.5-99
LGL (%)

Mean 1.8 6.4

Range 0-5 0.5-16.5

LGL/Lymphocytes (%)
Mean 5.7 6.9

Range 0-18.5 0.7-16.8

26

cytoplasm to one side of the nucleus were variably numbered
small red azurophilic staining granules (Figures 1 and 2).

Cytochemical staining of the effector cell population
was negative with peroxidase, Sudan black B and leukocyte
alkaline phosphatase. One preparation had rare focal
granular staining by chloroacetate esterase accounting for
0.5% of all lymphocytes in that sample. All other isolations
were negative for that stain. The mean and (range)
percentage of positive staining lymphocytes are listed for
the following stains: PAS 6.5% (4-9), ANAE 10.5% (5-19),
ANBE 23.0% (1-53) and ACP 23.7% (3-55). The pattern of
staining for the latter 3 stains was generally focal rather

than diffuse (Figures 3-5).

Ulttgstttgtutg; Studieg
The morphology and binding interaction of LGL in dogs

was compared to that in other species in terms of the
presence and appearance of organelles or cytolytic
structures in addition to the manner of cell membrane
involvement. The LGL was characterized by a round to ovoid
shape with a smooth surface or more frequently irregular by
the presence of several short thin pseudopod projections
(Figure 6). The nucleus was also round to ovoid with rare
deep clefts (Figure 7). Nucleoli were occasionally noted but
often the chromatin stained densely and peripheralized to

the nuclear margins. Frequently a shallow nuclear membrane

27

 

Figure 1. - Granular lymphocyte present in the peripheral
blood from a clinically healthy dog. Wright's stain.

‘ l
‘llll‘ 4‘
“} .

L .1 . 1 . .. 7 ,1. . ..-. ...- _ ,.,,,_ 7., 1. 1‘ ..1.,, AW,

   

Figure 2 - Cytocentrifuge preparation of isolated
non-adherent mononuclear cells with three granular
lymphocytes (arrows). Wright's stain.

28

 

Figure 3 - Weak focal staining of isolated lymphocytes with
alpha naphthyl acetate esterase (arrows).

 

Figure 4 - Strong focal paranuclear staining of isolated
lymphocytes with alpha naphthyl butyrate esterase (arrows).

29

 

Figure 5 - Focal (arrows) and diffuse (arrowheads) granular
staining of isolated lymphocytes with acid phosphatase.

3O

 

Figure 6 - Granular lymphocyte with several short pseudopod
projections. Lead citrate-uranyl acetate stain: X 18,360.

31

«A.

L!" ’l

' .1
u .§-,§’ .

4.1;. fl.’ 5 ":1:
'11 iv -

 

Figure 7 - Deep clefted nucleus in a granular lymphocyte in
contact with target cells. Lead citrate-uranyl acetate
stain: X 18,360.

32

indentation was present. At this site or to one end of the
cell were gathered most of the cytoplasmic organelles
(Figure 8). These included the golgi apparatus,
mitochondria, centrioles, vacuoles, rough endoplasmic
reticulum and dense granules (Figures 9-12) . Mitochondria
were often large, round and few in number (2-6). The
granules were found occasionally scattered throughout the
cytoplasm or more frequently located all to one side. They
varied in number (3-7), size and density. Most were membrane
bound and several were strongly osmiophilic and homogeneous
in density (Figures 6 and 10). However granules with
variable density and myelin figure formation were common
(Figures 9, 10 and 12). Rarely noted in long incubation
preparations (16 hours) were cytoplasmic structures which
were not membrane bound. These consisted of a cluster of
granular densities resembling ribosomes or glycogen granules
that were arranged in chains (Figure 13) . Occasionally a
disparity existed between the frequency of LGL observed
under light microscopy and that seen ultrastructurally with
the latter being more sensitive.

The morphology of the binding between lymphocytes and
target cells was investigated at various incubation times.
Few attachments were present at 30 minutes or 2 hours. As
early' as 30 :minutes, broad. attachments were noted. with
increased focal density where the cell membranes touched

(Figure 14).

33

 

Figure 8 - Granular lymphocyte with cytoplasmic organelles
found within the indented portion of the nucleus. Lead
citrate-uranyl acetate stain; X 18,360.

34

 

Figure 9 - Gathered to one end of the cytoplasm are several
mitochondria (M), golgi apparatus (GA) and variably dense
granules (g). Lead citrate-uranyl acetate stain; x 31,500.

35

 

Figure 10 - High magnification of granular lymphocyte with
mitochondria (M), centriole (arrow), membrane bound granules

(G) and myelin figure (MF). Lead citrate-uranyl acetate
stain; X 36,720.

36

 

Figure 11 - Granular lymphocyte with irregularly shaped
nucleus, prominent nucleolus (N) and abundant rough
endoplasmic reticulum (arrows). Lead citrate-uranyl acetate
stain; X 18,360.

37

 

Figure 12 - High magnification of granular lymphocyte with
variably dense granules and numerous vacuoles (V). Lead
citrate-uranyl acetate stain; X 36,720.

38

 

Figure 13 - Possible aggregation of ribosomes or a lysed
granule within the cytoplasm of a lymphocyte (L) in contact
with a target cell. Lead citrate-uranyl acetate stain; X

36,720.

39

 

Figure 14 - Lymphocyte bound to tumor target cell (T) with
increased focal density where cell membranes touch (arrow).
Incubation of 30 minutes. Lead citrate-uranyl acetate stain;
x 18,360.

 

 

40

The cell surface varied with the degree of binding
present between effector and target cells. Generally, cells
with extensive binding had smooth membrane surfaces with few
projections (Figure 15) . Cells with no or minimal contact
frequently had long finger-like pseudopodia (Figures 16-18).
This form of interaction was the most common observed. Both
granular and agranular lymphocytes were found attached to
targets but the former cell type predominated. Occasionally
more than one cell bound to the tumor cell (Figure 19) .
Granules were often in various locations within the
cytoplasm when binding involved minimal attachment (Figures
20 and 21). However with extensive adherence, granules
appeared to be positioned more frequently in the cytoplasm
facing the target cells (Figures 15, 22 and 23). This was
subjectively determined while viewing microscopic material.
Occasionally coalescence of several granules was noted
(Figure 23). Evidence of epicellular pore-like structures or
the intercellular tubular material that has been observed

in other species were not found during these studies.

Immunofluorescent staining studies

The results are presented in Table 2. Surface IgG and
IgM were infrequent in that positive cells generally
accounted for less than 5% of the mononuclear population. A
relatively larger proportion of mononuclear cells stained

positive for receptors of the Fc portion of IgG. Neutrophils

41

 

Figure 15 - Granular lymphocyte with smooth cell surface
bound extensively to a target cell. Cytoplasmic granule
located immediately adjacent to the target cell membrane.
Incubation of 10 hours. Lead citrate-uranyl acetate stain; X
10,200.

42

 

Figure 16 - Isolated granular lymphocytes with numerous
short finger-like surface projections. Incubation of 30
minutes. Lead citrate-uranyl acetate stain; X 10,200.

43

 

Figure 17 - Granular lymphocyte with long pseudopod surface
projections minimally touching a target cell. Incubation of
12 hours. Lead citrate-uranyl acetate stain; X 18,260.

44

 

Figure 18 - Granular lymphocyte with minimal binding to a
tumor target (T) via pseudopod projections of the cell
membrane. Incubation of 16 hours. Lead citrate-uranyl
acetate stain; X 16,110.

 

Figure 19 - A granular and an agranular lymphocyte binding
to the same tumor target cell. Incubation of 16 hours. Lead
citrate-uranyl acetate stain; X 6,593.

46

 

Figure 20 - Cytoplasmic granules located to a side furthest
from the target cell in a lymphocyte with minimal binding.
Incubation of 16 hours. Lead citrate-uranyl acetate stain; X
8,950.

47

 

 

Figure 21 - Granules are present between the lymphocyte
nucleus and the target cell to which the lymphocyte has
minimal attachment. Incubation of 2 hours. Lead citrate-
uranyl acetate stain; X 27,450.

48

 

Figure 22 - Granule placement directly adjacent to target
cell membrane (T) in a lymphocyte with extensive binding.

Incubation of 2 hours. Lead citrate-uranyl acetate stain; X
18,360.

49

 

Figure 23 - Coalescence of several granules and their
location immediately adjacent to the target cell to which
the lymphocyte has extensively attached. Incubation of 16
hours. Lead citrate-uranyl acetate stain; X 25,704.

50
Table 2 - Frequency of positive staining lymphocytes with
surface immunoglobulins (Ig) and PC receptor for IgG (FcGR)

as related to the percentage of LGL per isolated
lymphocytes.

 

Dag 1911131 13ers WM
Spot 5 10 19 8.4
Frankie 6 10 26 6.8
Spice 1 4 2 8.2
Annie 2 2 6 4.2
Mean 3.5 6.5 13.3 6.9

SEM 1.2 2.1 5.6 1.9

51
acted as positive controls for the latter stain (Figure 24).
The percentage of positive mononuclears however, represented

only a minor population of the total mononuclear cells.

 

Figure 24 - Positive staining of 2 lymphocytes (L) and 2
neutrophils. Direct immunofluorescence staining for
Fcreceptor of IgG; X 40.

52

CONCLUSIONS

These studies indicated that LGL were present normally
in canine peripheral blood and generally accounted for 6% of
all lymphocytes. This percentage could be increased up to
17% following routine NK cell isolation procedures.

Morphology of LGL with Wright's stain was similar to
that observed in other species. Numerous azurophilic
staining granules were eccentrically placed within the
cytoplasm of a large lymphocyte that measured 10-15 pm).

The isolated effector cell population was heterogeneous
with variable cytochemical and surface marker staining. This
population contained canine NK cells as verified by the
demonstration of spontaneous cytolysis of tumor target cells
(data not shown), but these cells likely represented only a
minor portion of the total population.

Cytochemically, positive staining of effector' cells
resembled that in other species. The effector cells stained
negative for those stains that identify granulocytic markers
(Sudan black B, peroxidase, alkaline phosphatase, PAS and
CAE). They stained positive for ANAE, ANBE and ACP in a
variable percentage of the population. Generally ANAE was
the weakest of the: three and. all demonstrated. a focal
pattern of stain distribution. The variability in staining
indicated a mixed population of T cells with up to 55%

having a focal staining pattern and a much smaller

53
population with diffuse (ACP) staining.

Immunologically, the percentage of effector cells with
surface immunoglobulins was similar to other species.
However, cells with FcG receptors did not predominate as has
been reported in rodents and people. Immunofluorescent
staining with surface antibodies indicated the majority of
cells were negative for IgG and IgM. Presence of receptors
for ECG was found only in a small number of effector cells
(up to 26%).

Ultrastructurally, LGL and their binding interaction
with target cells were similar to that observed in rodents
and humans with the exception that no pore-like or ring
structures were demonstrated by these studies. The granules
'were generally membrane bound, osmiophilic staining
structures resembling the lysosomes of myelomonocytic cells.
Binding interactions varied from villous interdigitations
with minimal contact to a smooth and extensive membrane
adherence. Approximately 75% of the cells binding to tumor
targets involved granular lymphocytes which is similar to
most species. The location of the granules during these
interactions was variable but generally found on the side of
the nucleus closest to the target cell. This orientation was
particularly apparent where binding was extensive suggesting
a direct role of the granules in cytolysis, a process which
has been shown to require intimate contact between effector

and target cells.

54

DISCUSSION

Since the LGL is reported to mediate NK cell activity
in many species, an attempt was made to determine their
frequency' in canine peripheral blood, isolate and
characterize these cells morphologically, cytochemically and
immunologically and finally to relate their appearance in
binding interactions with target cells. The frequency of LGL
in peripheral blood was similar to that reported in people
being generally less than 5% of leukocytes (Timonen gt g_.,
1981).

The type of isolation procedure utilized may affect the
yield of LGL. The use of nylon wool adherence following
Ficoll-Hypaque gradient centrifugation increased, the LGL
purity (data not shown) by removing neutrophils, monocytes
and some B cells. Reynolds gt g_l_. (1981) demonstrated a
similar increase in purity in rats using nylon wool
adherence. However there are suggestions that activated NK
cells have an enhanced adherence to nylon wool or plastic
(Herberman, 1985b). This could presumably remove some
functional NK cells.

Purification procedures used in this study were less
extensive than those used for other species. Discontinuous
Percoll gradient centrifugation or flow cytometry for
specific NK markers are additional methods commonly used to

isolate NK cells. Simpler and less sensitive techniques were

55

used in this study. This may account for the marked
difference in the yield of LGL, 15% in the dog compared to
60-90% for rodent and human cells (Reynolds gt g;., 1981;
Timonen g gt” 1981; Hiserodt 8 Beals, 1985). However, a
canine study using nylon wool adherence and discontinuous
Percoll gradients yielded 15% LGL as the greatest
concentration of any fraction (Savary 8 Lotzova, 1986). This
suggests the canine granular lymphocytes have more variable
cellular densities than human or rodent cells and that dog
LGL may not be readily purified by adherence or density
methods. Similarly, porcine NK studies (Yang gt g1,” 1987)
using plastic adherence and discontinuous Percoll gradients
produced fractions abundant in non-granular small to medium
sized lymphocytes (4.5 - 6.0 pm) . Therefore these procedures
will yield a variety of cell types responsible for NK cell
activity.

Morphologically, the canine LGL most closely resembled
those of non-human primates and people (Savary 8 Lotzova,
1986). Cell size was similar to that of people (12-15 pm)
and rats (lo-14 pm). The size and numbers of azurophilic
granules were similar to those in people and primates
(Savary 8 Lotzova, 1986). However, mice and rats appear to
have more abundant and less localized cytoplasmic granules
than other species (Reynolds gt a_l., 1981; Savary 8 Lotzova,
1986).

Cytochemical staining overall indicated only a small

56

proportion of the isolated lymphocytes stained positive.
Similar to human and rodent studies, stains such as
peroxidase, Sudan black B, leukocyte alkaline phosphatase
and PAS were negative. One human study found positive
staining in a majority of LGL with chloroacetate esterase, a
granulocytic marker (Heumann gt g1., 1983). This could not
be supported by this study as only rare positive staining
resulted. This finding is in agreement with the majority of
human and rodent reports.

Rat and human studies indicate ANBE is negative in NK
cells. The presence of focal ANBE staining in this study in
up to 53% of the lymphocytes suggests a mixed population of
T cells and non-T cells. A similar focal appearance was
noted in ACP and ANAE stained canine preparations. The focal
or globular pattern seen with ANBE, ANAE or ACP has been
associated with T cells in people and rodents (Manconi gt
g_l., 1979; Bozdech 8 Bainton, 1981; Yang gt gt” 1982).
Specifically, this focal staining pattern with ACP or ANAE
in people corresponds to T non-G lymphocytes or T cells with
Fc receptors for IgM (Manconi gt gl., 1979; Ferrarini gt
g1., 1980). These cells have. little or no NK cell activity
(Ferrarini 8 Grossi, 1986).

A diffuse fine granular staining with ACP was present
in a small number of isolated canine lymphocytes. Human T
cells with Fc receptors for IgG and third population cells

which lack B or T cell markers but possess receptors for

57
Fcc, stain in a diffuse or paranuclear granular pattern with
ACP and ANAE (Ferrarini gt gl., 1980; Huhn gt gl., 1982).
The third population cells are primarily responsible for NK
cell activity.

Information regarding staining pattern by nonspecific
esterases (ANAE, ANBE) and ACP has not been well worked out
in dogs. Extrapolation of the significance of these patterns
from other species to this study suggests that the isolated
lymphocytes consisted of a small population of T cells and
non-T cells having FcG receptors. However a larger
proportion of the cells were likely T cells with IgM
receptors.

Ultrastructurally, there is strong similarity between
the dog and other species in the types of organelles present
within the cytoplasm. The majority of granules were
lysosomal in nature being membrane bound with dense internal
staining. One cell had a granule-like structure which
resembled the parallel tubular arrays reported in people.
However, this cluster of granular chain-like densities
likely represented an accumulation of glycogen or ribosomes.
Their frequency was rare compared to a more common
occurrence of parallel tubular arrays in people.

Ultrastructurally, the interaction between target cells
and granular or agranular canine lymphocytes was analogous
to that seen in rats, pigs and humans, particularly in the

early binding stages. The pore-like structures found on the

58

surface of target cells in human and murine studies (Henkart
8 Henkart, 1982; Podaok 8 Dennert, 1983) were not observed
in this study. The intercellular cylindrical or ring-like
structures reported by Hiserodt and Beals (1985) were
occurring naturally or produced within 15 minutes after the
addition of calcium. Their absence in the dog may be due to
differences in isolation procedures, incubation times,
fixation or microscopic resolution.

The addition of calcium was also associated with a
change in the cell surface characteristics which occurred 5-
15 minutes following calcium pulsing (Hiserodt 8 Beals,
1985). Prior to its addition and after 30 minutes,
microvillous projections were present. This morphologic
change which was also observed with EC bearing lymphocytes
adhered to antigen-antibody complexes, may be necessary in
the cytolytic process (Alexander 8 Henkart, 1976). Similar
to these studies, cell surface appearance in the dog changed
from villous to smooth with increased attachment of
lymphocytes to target cells.

The location of granules on the side of the nucleus
closest to the target cell is in agreement with the majority
of reports which also suggest this orientation is important
for the cytolytic process to occur. The exocytosis of
granule contents releases potent cytotoxic substances
(Henkart, 1985).

The observations made at various incubation periods

59

indicated few time-related differences relative to the
pattern of attachment of lymphocytes to target cells. This
may be due to the recycling nature of NK cells. Once cell
death has been initiated by the NK cell, it can attach to
other target cells and begin again the cytolytic process.
The preparation of cells for electron microscopy can disrupt
the interaction between targets and effector cells,
particularly in the prebinding or postbinding stages of
cytolysis. Manipulation of the cytolytic rate with serial
examinations performed within 1 hour of incubation would be
more helpful in evaluating this process.

Since binding is necessary for cytolysis to occur, one
can conclude that the cells most frequently bound to targets
are responsible for NK cell activity. This study indicated
that approximately 75% of binding lymphocytes contained
cytoplasmic granules. Other evidence to support the LGL as
the canine NK cell comes from data which demonstrates
parallelism between the frequency of canine LGL and NK cell
activity. Percoll fractions with larger numbers of LGL
produced increased cytotoxicity in lytic units as measured
in a chromium release assay (Savary 8 Lotzova, 1986).

Immunologic marker studies in dogs indicate the NK cell
variably rosettes with human and guinea pig erythrocytes
(Betton, 1980; Ringler 8 Krakowka, 1985). To additionally
support the T cell origin, Ringler and Krakowka (1985)

demonstrated the expression of Thy-1, a T cell marker of

60
mice on a majority of the canine NK cells.

Furthermore, Ringler and Krakowka (1985) demonstrated
that 39.0% of binding lymphocytes rosetted with IgG coated
ovine erythrocytes indicating the presence of Fc receptors
for IgG. They also showed that depletion of these cells
significantly reduced NK cell activity. The present study
using a direct immunofluorescence method detected 13.3% IgG
receptor positive cells. A low proportion (6.5%) of positive
cells stained by direct immunofluorence for surface 196 was
also found in this study. This compares to (1%) by Ringler
and Krakowka (1985). The variation in positive cell
percentages may be explained by differences in the
techniques and the population of the lymphocytes sampled.
The addition of nylon wool adherence in the present study
may have removed some of the FCC receptor positive cells as
well as granulocytes and monocytes. Ringler and Krakowka's
evaluations were performed on lymphocytes bound to dead
targets while the present study assessed a total lymphocyte
population.

There is a direct relationship between human LGL and
IgG receptor positive cells (Gastl gt g;., 1983). Enrichment
procedures for cells having FcG receptors have produced 90%
purity of LGL (Timonen gt g1., 1981). The percentage of LGL
per lymphocytes in this canine study did not correlate well
with the percentage of positive FcG receptor cells (p >.05)

using correlation analysis. However, there was significant

61
correlation between the percentage of ECG receptor positive
cells and NK cell activity (data not shown).

It is concluded from this study that several
morphologic, cytochemical and immunologic features of canine
NK cells resemble those described for other species
including humans and mice. These involve the confirmation of
the presence of LGL in peripheral blood and in NK effector
cell populations. In addition, morphologic studies suggest
the LGL may mediate NK cell activity in the dog. Effector
cells used in the NK cell assays were weakly positive for
ANAE, ANBE and ACP. Furthermore, they involve a small

population of ECG receptor positive cells.

CHAPTER 3

NATURAL KILLER CELL ACTIVITY IN NORMAL DOGS

NATURAL KILLER CELL ACTIVITY IN NORMAL DOGS

INTRODUCTION

Natural killer (NK) cell activity is a spontaneous
nonspecific cell mediated cytotoxicity reaction which acts
independently of the humoral defense mechanisms. In vivo, NK
cell activity has been attributed to early tumor cell
elimination, cytolysis of virus infected cells, graft
rejection, hematopoietic cell regulation and general
immunoregulatory functions (Trinchieri 8 Perussia, 1984;
Wheelock 8 Robinson, 1983). NK cell activity has been
recognized in almost every species including: invertebrates,
fish, amphibians, sheep, cattle, pigs, chickens, dogs, cats,
rodents, non-human primates and humans (Savary 8 Lotzova,
1986; Yamamoto _e_t_ g_., 1985; Yang gt g;., 1987; Sharma 8
Coulson, 1979; Betton, 1980; Tompkins gt g;., 1983;
Kiessling gt gl., 1975b; Reynolds gt a_l., 1981; Altman 8
Rapp, 1978; Gengozian gt _t., 1986; Timonen gt g;., 1979b).

Features used to identify the cells responsible for NK
activity include: nonadherence, nonphagocytosis, receptors
for Fc portion of immunoglobulin (Ig) G, some T cell and

myelomonocytic surface markers, absence of surface

63

64

immunoglobulin, cytolysis of NK cell-sensitive cell lines,
and the presence of azurophilic cytoplasmic granules
(Trinchieri 8 Perussia, 1984; Gastl gt g;., 1984; Betton,
1980; Kiessling gt gt” 1975b; Ringler 8 Krakowka, 1985a:
Timonen gt g1., 1975a). The frequency of the large granular
lymphocytes (LGL) has been closely associated with NK cell
activity in rodents and humans (Reynolds gt g1., 1981; Gastl
gt g1., 1983) .

Reports evaluating NK cell activity in dogs have been
few in number. Immunologic studies involving erythrocyte
rosetting; presence of Thy-l, surface Ig, or Fc receptors
for IgG; adherent and phagocytic properties have been
reported in the dog (Betton, 1980; Shek gt g_., 1980;
Ringler 8 Krakowka, 1985a). Investigation of NK functional
activity has included conventional and gnotobiotic dogs
(Savary 8 Lotzova, 1986; Krakowka, 1983; Betton, 1980),
canine distemper virus infected dogs (Krakowka, 1983; Shek
gt gl., 1980; Ringler 8 Krakowka, 1985b) and dogs with
spontaneous neoplasia (Betton 8 Gorman, 1978; Warren gt g;.,
1975). Reports involving morphology of canine NK cells are
limited (Savary 8 Lotzova, 1986).

The aim of this work was to evaluate NK cell activity
from healthy adult dogs through concurrent functional and
light microscopic studies. The relationship between
frequency of LGL present in effector cell populations and NK

cell activity was explored using both chromium release and

65

single cell binding assays.

MATERIALS AND METHODS

Animals
Eighteen clinically healthy pure and mixed breed adult

dogs were used. There were 3 intact males, 6 intact and 9
ovariohysterectomized females. Seven dogs were evaluated 2-3

times providing a total of 26 samples.

Effggtgr and target cell isolations

The isolation of effector and target cells was

performed by the method described in Chapter 2.

os o ' t d'

Determination of the frequency of LGL was accomplished
by the procedure described in Chapter 2. LGL were
distinguished by the presence of numerous small red
cytoplasmic granules located focalLy at one end of a large
lymphocyte. Percent LGL refers to their frequency in the

total isolated mononuclear population.

0 a 'cro t o 't assa
Following isolation of target cells, 100 uCi of sodium
chromate (51Cr) in saline (New England Nuclear, Boston, MA)

was added to approximately 5 X 106 cells and allowed to

66
incubate at 37 C for 90 minutes. Final target cell
suspension was adjusted to 5 X 104 cells/ml in supplemented
MEM. Fifty or 100 pl of the effector cell suspension were
placed in a 96 round bottom well microtiter plate (Flow
Lab., Inc, McLean, VA) along with 100 pl of the target cell
suspension for effector to target cell ratios (E:T) of 50:1
or 100:1. Wells containing 100,pl of target cells alone were
used to measure spontaneous release of the radiolabel.
Maximum release was defined by lysis of 100 pl of target
cells with 10% Triton X-100, a cell detergent. Plates were
centrifuged 200 g for 2 minutes at 4 C. Samples were read
immediately and after 12 or 16 hours incubation in a 37 C
humidified 5% C02 /95% air environment. Supernatants were
collected from wells using a Skatron harvesting set
(Sterling, VA) and read in a gamma counter (TM Analytic,
Inc, Elk Grove Village, IL). All tests were performed
minimally in triplicate. Cytotoxicity was measured using
counts per' minute (CPM) of radioactive release. In the

following formula, percent cytotoxicity equals:

CPM (effectors + targets) - CPM (spontaneous release)
X 100

 

CPM (maximum release) - CPM (spontaneous release)

Singlg ce;l binging assay

Preparation of poly-L-lysine (PLL) coated coverslips

were modified from the technique described by Vargas-Cortes

67
gt gl. (1983) . Poly-L—lysine hydrobromide (Sigma Chemical
Co, St. Louis, MO) was diluted with buffered Hanks solution
to a concentration of 2 pg/ml and 1 ml was placed on each
glass coverslip (22 mm X 22 mm) used. Coverslips were
incubated for 45 minutes at room temperature, rinsed with
Hanks solution and allowed to dry before use.

One hundred microliters of cell suspensions
concentrated to 1 X 105/ul were used for both effector cells
and targets. Targets alone and an E:T of 1:1 were pelleted
and allowed to incubate 15 minutes at 37 C. The supernatant
was then replaced with 0.5 ml serum free MEM and the cells
gently resuspended. The cell mixture was plated upon the PLL
coated coverslips and allowed to incubate 20 minutes at room
temperature to insure adherence. Samples were read
immediately and after 12 and 16 hours incubation in a 37 C
humidified 5% C02 environment. To detect dead cells,
coverslips were stained with 0.1% trypan blue in Hanks
solution, then rinsed to remove any excess. Coverslips were
then inverted onto glass slides and the number of conjugates
per 200 live lymphocytes was counted. The percentage of live
or dead conjugates with effector cells bound to targets was

then calculated.

Igmunotluoregcent staining of effector cells

A direct immunofluorescent procedure was applied to

cell suspensions to detect surface staining. To detect the

68
presence of receptors for the Fc portion of IgG (FcG) ,
fluorescein conjugated Fc fragment for dog IgG (Jackson
Immunoresearch laboratories Inc, Avondale, PA) was used. A
dilution (1:10) was made of the antibody and 0.1ml was added
to 1.5 X 106 cells of the nonradherent lymphocyte
population. Incubation for 1 hour at 4 C was followed by 2
phosphate buffered saline washes. Cells were resuspended in
phosphate buffered saline, then transferred to a glass slide
and examined under a fluorescent microscope. Mononuclear
cells were counted and positive cells were expressed as a

percentage of the total number.

ta '3 cal ana sis

Arcsin transformation of percentage data was performed
prior to analysis. Paired and unpaired Student's t tests
(two tailed) were used to compare incubation times, cell
ratios and sex differences. Relationships between frequency
of LGL, percent cytotoxicity, percent live and dead cell
binding and percent FcG receptor positive cells were

determined by Pearson's correlation analysis.
RESULTS
Table 3 lists the NK cell activity in healthy dogs as

measured by the chromium release assay. There was no

significant difference (p >.05) between the 12 and 16 hour

Table 3 - NK cell
dogs.

activity in 26 samples from 18 healthy

69

 

Males: (3)
Mean
SEM
# of Samples

Intact Females: (6)
Mean
SEM
# of Samples

Neutered Females: (9)

Mean
SEM
# of Samples

TOTAL:
MEAN
SEM
# 0F SAMPLES

 

12_HQE£§
50: 100:
26.3 35.0
9.7 10.3
3 3
14.4 17.5
4.3 5.2
10 8
16.6 24.8
2.3 3.4
13 12
16.9 23.6
2.3 2.9
26 23

 

16 Houts
50:1 100:
23.5 40.2
13.1 11.7

3 3
19.3 18.6
4.9 4.7
10 8
18.1 24.5
1.9 4.4
13 12
19.2 24.5
2.4 3.3
26 23

 

Data expressed as percent cytotoxicity at 12 and 16 hours of
incubation with 50:1 and 100:1 (E:T).

70

incubation periods for effector to target cell ratios of
50:1 or 100:1. Sixteen hour incubation generally was higher
in NK cell activity than 12 hours (Figure 25) . Significant
differences were present between 50:1 and 100:1 (E:T) at 12
hours (p <.0005) and at 16 hours (p <.005) demonstrating a
positive dose response effect. Relative to sex differences,
there was a suggestion of highest activity in males,
followed by ovariohysterectomized females, with intact
females having the lowest activity (Figure 26). However this
was not statistically significant (p >.05). In 5 dogs with
repeated samplings over 1-3 months, 3 had relatively stable
activities (Figure 27).

The data from 26 samples indicate the following mean
(SEM) for percent LGL: 6.9 (0.9), percent cytotoxicity: 16.9
(2.3) at 12 hours with E:T of 50:1, percent live binding:
8.4 (1.2) at 12 hours and percent dead binding: 1.0 (0.2) at
12 Ihours. When compared by correlation analysis, no
significant difference (p >.05) could be found between these
variables. There was significant positive correlation (p
<.05) between the 4 dogs evaluated for percent FcG receptor
positive cells and percent cytotoxicity (Table 4). In these
same dogs, negative correlation (p <.02) was present between
percent FcG receptor positive cells and percent live
binding. No correlation could be found between percent FcG

receptor positive cells and percent LGL.

71

.mmnmnucouoa
ca co>fim we 93» scapegosw mom ooummu mmaoaam mo Monasz
.coflpansocfi no musoc ma one NH now moans.“ HHoO woman» 0» uouoouuo
msowuo> am noon wagon canvases A: auflofixouogo unwouom I mm 05.5.3

E. n'll‘ll In

F.

 

[23)

72

B 12 HOURS
l8 HOURS

S

 

V/////////////4

vOOO0.00000’O'OOOO’OOOOOOOOOO00.9.09a
O000090.000.00.00.0.00.00.00.00.0
DO00000090900000.9000.900000009004
00.0.00....OOOQOOOOOOOOOOOOOOOOOO
OOOOOOOOOOOOO00090000009000.09000
OOO00000000000000.000.000.00.0000
'000.00.900.00.000000000000000000q
O0009000000OOOOOOOOOOOOOOOOOOOOOO
oO00O...0000.090.000000000000000.G
’O...O...0.0.0.0.0.0.0.0.0.0.0.0.0...0.0.0.0...0.0...0.0.0.0.0.0.0‘
..9...0.9.0.9.0.9. 0.......0.0.9...03.931... 0.9.9.0..-..9.0.9.0

VVOVVVVVVVVVVVVVV'VVVVVV
ooooooooooooooooooooooo-
oooooooooooooooooooooooo
ooooooooooooooooooooooo:
oooooooooooooooooooooooo
ooooooooooooooooooooooo'
oooooooooooooooooooooooo
oooooooooooooooo.ooooor
0.9000000000009000009090
0.0900000000000990000o0
'I’¢V9V¢V¢VJ’¢I¢3¢WJ¢33¢

    
    
 

[23)

  
      
        
      
 
           
      

(28]

    
 
        
 

   

   

   

l DD:|
TARGET RATIO

5|]: l
EFFECTDH

 

30-

 

l I
C3 C3 C3
pg _—

MJUIXDLDMO lNBIJHBd

Figure 25

73

.mmmwzucmumm
a“ cm>am ma msoum “mm moamamm Ho Hmnfisz .mmamfimu Umumusm:
«Em Human.“ $922: you mowumu. .300 #098» on Houomuum can mmfiwu
cowuunaoca maoflum> um huflonououmo unmouma mo comfihumfioo I mm whamflh

74

 

 

mmzax m.
:8. tom
6:
a. a:
a:
mmiEmu [r
63:. n a.
unison.
_umgmfimz E
moi: a "ma

 

 

 

 

 

maze: N.
.52 :8
a:

.2
a. _
a: .m.
a.
Li
.9

 

 

on Gunman

0.

ON

am

av

am

am

AllfllXUlDlACl lNHIJUBcI

75

.mmov Hmacw>wwcw huflucmufi maonahm cmuwmm .mOflumu HHmo umwumu on
noncommm m90flhw> nufls cowunnauCH muzon NH um umcwaumumc meOflxououho
unmoumm .msucoa mIH Hm>o wows» :mxmp mvoo m mo mmCHamfimm I hm mhsoflh

76

 

I

IDUII

50:]
EFFECTUR: TARGET RATIO

 

l
I:
m

l
CI
In

I l l l
I: I: CI I:
v m N —-

AlICIIXDllJlACI lNBCIHEJd

 

Figure 27

77

unmoflaflm uo: " m2
.5338”: no WEB 8982 gum 5 .3852

 

 

 

mz Bad mé ad ad «.m 33 w
m2 ommd o.~ o m5 m5 ad 38
mo. v a Sad: oé o.m m4 ms 83 «>3
"3.52am :8 w
m2 «485 92 m Q 38 3.3 $3 flood
mo. v a Sad v.3 m o 043 You 83 flow A
no. v m ommd «.3 5.... 9mm «:3 33 ~82
m2 «36 ed flu 93 o; Am: Ham
”36336 w
.. I m m mm 2 $8882 cum »
gfimaflm £86338 w m m a
Sfimgoo
man

 

.ggggn Hamoummcuogflfifioumm
.Bfiofifiub‘tnu £50.39 0» SflmamHuoo EM 330 «2.5309 Hang 00m mo in I _v «.38.

78

CONCLUSIONS

This study indicated no significant difference in NK
cell activity in healthy’ dogs between 12 and. 16 hours
incubation. However, increased cell concentrations
demonstrated significantly higher NK cell activities. Too
few individuals were available per' group but. there ‘was
nonetheless a definite trend towards higher NK cell activity
in males compared to females.

Similar to other species, percent cytotoxicity varied
widely within the total group of 26 canine samples. However,
healthy individuals tended to maintain the same level of
activity over time.

An unexpected finding was that neither frequency of LGL
nor percent dead binding correlated ‘well with NK cell
activity. The percentage of Cells possessing FcG receptors
did significantly and positively correlate with percent
cytotoxicity but negatively with percent live binding. While
cell isolation procedures and quantitation of NK cell
activity may account for some of the discrepancy, it appears
that identification of the canine NK cell is more closely
related to the presence of immunologic markers than to mere

morphologic criteria.

79

DISCUSSION

Use of percent cytotoxicity to evaluate NK cell
activity has been shown to be a qualitative method (Pross g;
al., 1986). Studies in people suggest this method produces a
large degree of variability with a broad range of activity
found in repeatedly tested normal individuals (Pross &
Baines, 1982). However, Gengozian gt a1. (1986) found
individual rhesus monkeys could be separated into low and
high responder categories. Within these groups, twice
repeated samples from individuals over time produced a
relatively constant level of activity using percent
cytotoxicity.

Two canine studies indicate a similar variability in NH
levels. Ringler & Krakowka (1985a) reported levels of NK
activity in a small population of adult dogs that ranged
from 12.4% to 51.5% using CTAC targets with 30:1 (E:T) at 16
hours of incubation. Betton (1980) measured percent
cytotoxicity in 10 dogs over 2-3 weeks and found
considerable differences in activity. In comparison, the
present study utilizing percent cytotoxicity to measure NK
cell activity, also found a wide range within the entire
population. However, stable activity was evident in most of
the individuals that were sampled twice over time.

To reduce some of the variability in NK cell activity,

mathematical calculations from several E:T may be performed

80
to express this activity in terms of lytic units (LU) or
relative cytotoxicity. Lytic units refers to the activity
per effector cells necessary to produce lysis of a
predetermined percentage of target cells. Calculation of a
ratio between the LU of an individual as compared to that of
normals run concurrently gives a more accurate and
quantitative measurement of NK cell activity (Pross &
Baines, 1982). Since 3 or more E:T are necessary to
accurately derive LU, this calculation could not be
attempted in the present study.

The variability of NK cell activity in humans has been
attributed to age, sex, drugs, exercise and disease (Pross &
Baines, 1982; Brahmi e_t a_l., 1985; Huwyler gt 1L, 1985;
Lipinski _e; a_l., 1982). Age differences in people involve
lower values at birth which gradually increase towards
middle age and decrease mildly in the elderly (Pross &
Baines, 1982) . Since this study involved only adult dogs,
this comparison could not be made. The only reference to age
in dogs is by Ringler & Krakowka (1985b) who evaluated pups
from 1 to 6 weeks of age. They found activity present as
early as 1 week of age that minimally fluctuated over the 6
weeks.

Relative to sex differences, Pross & Baines (1982)
reported that the mean activity in each age group tested
including_neonates, was higher for males than females. Sex-

related differences were not found in 23 gnotobiotic pups

81

evaluated from 1-6 weeks of age (Ringler & Krakowka, 1985b).
The results from the present study indicated a trend similar
to people with males having the highest activity. However
this was based on a small number per group and hence was not
statistically significant. Between intact and
ovariohysterectomized females, differences were much less
pronounced. Since dates of the surgery were unknown and most
were sampled once, little significance can be attributed to
this observation. Estrogens have been shown to inhibit NK
cell activity in mice (Seaman gt g_., 1978). It is therefore
possible that this hormone is responsible for the
differences seen between males and females in this study.

Other dog studies indicate activity increases over
time, particularly when CTAC targets are used (Krakowka,
1983; Savary & Lotzova, 1986). Results from this study using
8 (data not shown), 12, and 16 hours of incubation also
showed a gradual increase in NK activity. However there was
no significant difference between the 12 and 16 hour
incubation periods for either E:T (50:1 or 100:1) tested.
Therefore evaluations of NK cell activity in the dog against
CTAC targets may use either 12 or 16 hour incubation times.

Several human and rodent studies suggest strong
association between the number or percentage of LGL present
or binding to targets and NK cell activity, measured in
percent cytotoxicity or LU (Timonen gt g;., 1979b; Reynolds

gt g_., 1981; Gastl _et gt” 1983). Gastl _e_t gt. (1984)

82
reported this finding was particularly evident in malignant
and nonmalignant disease states. Despite the association,
Reynolds gt _l. (1981) felt LGL alone would not predict NK
cell activity since not all LGL have cytolytic function
against certain targets. They found that only 10-30% of LGL
formed conjugates with target cells.

The present study could find no significant correlation
between the frequency of LGL seen by light microscopy and
percent cytotoxicity. Savary and Lotzova (1986) evaluated
several Percoll fractions in the dog relative to percent LGL
and NK cell activity that was measured in LU. They found
parallelism between the highest levels for each variable
with peaks occurring in fraction 3 which consisted of
approximately 45% Percoll. Comparison of this work to the
present study is impossible due to differences in LGL
isolation and measurement of activity. In addition, the
number of animals tested was not indicated in Savary and
Lotzova's report. It is also possible that Percoll fractions
which separate cells on the basis of density may select for
a different functionally active LGL or nongranular
lymphocyte. There is information to suggest that the cell's
activity may be linked to granule maturation which could
affect. cell density. Furthermore, recent. porcine. studies
indicate that the cell mediating its NK cell activity is
primarily nongranular (Yang gt gl., 1987). Since percentages

of LGL following isolation procedures are much lower in dogs

83
compared to rodents and people, it is necessary that
additional studies be conducted in the dog to confirm or
deny the role of LGL in NK cell activity.

While the frequency of LGL may correlate well with
activity levels in some studies, the number of binding cells
may not as well (Ullberg & Jondal, 1981; Reynolds gt a]..,
1981; Rubin gt gl., 1982; Gastl gt g1., 1984). There is some
disagreement as Roder gt gt. (1978) reported that the
frequency of target binding conjugates did correlate well to
percent lysis. It is well known that agranular lymphocytes
may also participate in effector to target cell binding.
Therefore it seems ‘unlikely that there ‘would. be strong
correlation. This conclusion was supported by the present
study.

On the other hand, the percentage of dead or lytic
effector-target cell conjugates has been highly correlated
to NK activity for certain targets (Ullberg & Jondal, 1981;
Rubin gt g;., 1982). This observation could not be supported
by this study which may reflect differences in technique and
longer incubation periods which were required in the canine
analysis.

Instead of morphologic features to establish the
identity of the NK cell, functional or immunologic markers
have been more widely used in humans and rodents. Discussion
of all these markers is out of the scope of this report.

However, the marker that identifies the receptor for the Fc

84

portion of IgG was evaluated in this study. Ringler and
Krakowka (1985a) demonstrated marked reduction in NK
activity following depletion of Fee receptor positive cells.
In addition, they found 39% of functional canine NK cells
possessed this marker. Using a different technique, this
study demonstrated that up to 26% of isolated lymphocytes
were FcG receptor positive.

The importance of this marker in evaluating NK cell
activity is additionally supported by human studies which
highly correlate FcG receptor positive cells with target
binding cells or cytotoxicity (Ullberg & Jondal, 1981;
Timonen gt g;., 1979b). While the present study demonstrated
significant correlation between FcG receptor positive cells
and percent cytotoxicity, it established a significant
negative correlation with live conjugate formation. Since
only 4 dogs were tested, it would. be premature ‘to .be
definitive about this data. Additional studies are clearly
necessary before the role of FCC receptor positive cells in
NK cell activity is established for the dog.

It can be concluded from this study that certain
similarities and differences exist between the NK cell
activity of various species and that in normal or healthy
adult dogs. Similarities involve the presence of spontaneous
or natural cell mediated cytotoxicity against tumor target
cells, a wide range of normal NK cell activity, minimal

variation in activity among individual dogs over time, sex

85
differences in activity and a significant correlation
between FcG receptor positive cells and percent
cytotoxicity. Unexpected results were the lack of
significant correlation between LGL frequency or percent

dead binding and NK cell activity.

CHAPTER 4

NATURAL KILLER CELL ACTIVITY IN UNTREATED AND TREATED CASES

OF CANINE LYMPHOMA

NATURAL KILLER CELL.ACTIVITY IN UNTREATED AND TREATED CASES

OF CANINE LXHPHOHA

INTRODUCTION

Lymphoma is the most frequent hematopoietic neoplasm in
dogs accounting for 15% of all malignancies (Priester &
McKay, 1980). Affected dogs are usually middle to older aged
and sex predilection is not evident (Rosenthal, 1982).
Several breeds appear to be at high risk including boxers,
basset hounds, airedale terriers, Scottish terriers,
bulldogs, Saint Bernards, labrador retrievers and bull
mastiffs (Rosenthal, 1982; Onions, 1984). Elevated reverse
transcriptase activity has been associated with lymphoma
cell cultures but definitive proof of a viral etiology is
lacking (Madewell, 1985).

Classification of canine lymphoma has involved their
anatomic distribution, immunologic features and cytologic or
histologic type (Couto, 1985). The :most common form. of
anatomic presentation is multicentric characterized by
generalized lymphadenapathy which progresses to organomegaly
and terminally into leukemia. Immunologic studies

demonstrated most canine lymphomas were of B cell origin

87

88
(Holmberg gt a_l., 1976; Onions, 1977). Classification of
histologic cell types have been adapted to those described
in people with the most recent being the NCI Working
Formulation (Carter gt g;., 1986).

Prognosis has been related to clinical staging,
presence of paraneoplastic syndromes, sex and histologic
cell type (Crow, 1982; Cotter, 1983; Gray gt 11., 1984;
Couto, 1985; MacEwen _e_t g;., 1987). Survival without
treatment is short being about 2 months. Glucocorticoids
alone provide a survival range between 2-6 months. Combined
chemotherapy protocols may extend this range to 1.5 years
with a median survival of 6—7 months (Crow, 1982; Couto,
1935).

The role of NK cells in tumor immunosurveillance has
been suggested indirectly by studies of jg yittg NK cell
activity correlated to growth of transplantable tumors in
experimental animals. For example, beige mice with deficient
NK activity, develop more rapid and frequent tumors compared
to normal mice (Karre gt g;., 1980). On the other hand, nude
athymic mice with higher than normal NK activity, have
increased resistance to tumor growth (Warner gt g)", 1977).

The in m role of NK cells in the prevention of
spontaneous tumors has been derived indirectly from
observations that individuals with genetic immune
deficiencies, such as Chediak-Higashi with low NK cell

activity, have a high risk of developing lymphoid

89
malignancies (Roder gt a_l., 1980; Abo gt g;., 1982). The
same is true for transplantation patients receiving large
doses of immunosuppressant drugs that reduce NK cell
activity (Lipinski gt g;., 1982). In contrast, normalization
of NK cell activity occurs during periods of disease
remission (Stupp gt g_., 1978; Yoda gt g;., 1983).

An inverse in yiyg relationship between NK cell
activity and progression of disease can be derived from
experiments in mice which deplete the NK cell population
through specific NK antibodies (asialo 6M1, NK1.1) or drugs
(estradiol, cyclophosphamide) allowing increased tumor
growth and hematogenous metastasis (Hanna & Burton, 1981:
Hanna & Schneider, 1983; Barlozzari gt a_l., 1983; Hanna,
1987; Seaman gt g;., 1987). Restoration of the depleted NK
cell population produced decreased lung metastasis and
evidence to confirm the NK cell's role in host defense
against early metastasis (Hanna & Fidler, 1980; Hanna,
1986).

The malignancies most commonly associated with
decreased NK cell activity in various species involve
hemolymphatic conditions. These include lymphoid or myeloid
leukemias, Hodgkin's and non-Hodgkin's lymphoma and bovine
leukosis (Behelak gt gl., 1976; Stupp gt gl., 1978; Ziegler
gt g1., 1981; Hawrylowicz gt g;., 1982; Tursz gt g;., 1982;
Laroche 8 Kaiserlian, 1983; Nasrallah & Miale, 1983: Kay &

Zarling, 1984; Yamamoto _e_t a_l., 1984; Jezewska gt gt” 1985;

90

1., 1986: Fujimiya gt g1., 1986; Hooper gt g1.,

Fontana gt
1986). It is therefore possible for a similar association to
exist between canine lymphoma and reduced NK cell activity.

If NK cell activity plays an important role in the
progression of canine lymphoma, then it is desirable for
drug protocols that treat the condition, to enhance that
activity. Several investigators have reported in contrast,
an inhibition of NK cell activity by glucocorticoids,
cyclophosphamide and combination protocols presently used
for the treatment of lymphoproliferative disorders (Stupp gt
gl., 1978; Oehler & Herberman, 1978: Parrillo & Fauci, 1978:
Hochman & Cudkowicz, 1979; Riccardi gt g;., 1981; HcGeorge
gt a_l., 1982; Bray gt gl., 1983; Ballas, 1986; Braun &
Harris, 1986).

The purpose of this study was to investigate the
presence of natural cytotoxic mechanisms in spontaneous
untreated cases of lymphoma in dogs. In addition, the
effects of glucocorticoids and combined chemotherapy
protocols used to treat these patients would also be
evaluated relative to NK cell activity. The use of
concurrent chromium release and single cell binding assays
would help determine whether the disease and treatment
effects are associated. with low' NK cell numbers and/or

altered NK cell function.

91
MATERIALS AND METHODS

Animalg

A 'total of’ 11 adult. canine lymphoma. patients ‘were
evaluated. These were placed into 3 groups depending upon
prior treatment. The groups were as follows: I - untreated
(n=7), II - glucocorticoid-treated (n=5), III - combined
chemotherapy-treated (n=4). The combined chemotherapy
protocols involved COP (cyclophosphamide, vincristine and
prednisone) or CHOP (cyclophosphamide, doxorubicin,
vincristine and prednisone). Five of these dogs were placed
in 2 groups (Table 5). A summary of the clinical data for
these cases including sex, breed, age, clinical staging (WHO
classification), histologic cell typing (Carter' gt_ g1.,
1986) and specific treatment may be found in Appendix B. A
total of 11 clinically healthy control dogs were used for
concurrent or paired evaluations of lymphoma patients. A
larger comparison population, namely ”all controls”,
consisted of the above 11 paired controls plus 15 additional

clinically healthy dogs.

Effector and target cell isolations

The isolation of effector and target cells has been

previously described in Chapter 2.

e s m'c totoxicit ssa

Measurement of recycling ability or overall function of

92

NK cells was performed by the method described in Chapter 3.

Table 5 - The group placement of 11 canine lymphoma patients
(I = untreated, II = glucocorticoid-treated, III = combined
chemotherapy-treated)

 

Qog Number grotp I Qtoup II Gtoup III

1 X * X

2 X * X
5 X X
6 X

8 X X
9 X X
10 X

11 X

12 X

14 X

17 X

 

* Chromium release data unavailable for 100:1 effector to
target cell ratios

Singlg gel; binding assay
Methods to evaluate the frequency of target binding and

cytotoxic cells in effector cell populations were discussed

previously in Chapter 3.

Statigtical analysis

Arcsin transformation of percentage data was performed
prior to analysis. A split-plot factorial analysis of
variance was used to determine the between and within group
interactions of the 3 lymphoma conditions relative to cell

ratios and incubation times. Additionally, a paired

93
Student's t ‘test (one-tailed) was used. to compare :mean
differences between lymphoma patients and their concurrently
measured controls. Significant differences were considered

to be p <.05.

RESULTS

Table 6 lists the NK cell activity of canine lymphoma
patients as measured. by the chromium. release assay for
various incubation times and effector to target cell ratios
(E:T). Significant differences (p <.05) were found between
paired controls and Group I patients at 16 hours and an E:T
of 100:1. No significant. differences. were found. between
paired controls and prior glucocorticoid-treated lymphoma
dogs. Significant differences were present between controls
and COP or CHOP treated lymphoma patients at various
incubation times and E:T. Figure 28 illustrates the mean and
standard error differences in percent cytotoxicity between
paired controls and lymphoma patients at 16 hours and 100:1
(E:T).

The live and. dead target cell binding activity by
lymphocytes in lymphoma dogs for 12 and 16 hours of
incubation is listed in Table 7. No significant differences
were found between paired controls and lymphoma dogs of any
group. The mean and standard error differences for controls

and lymphoma patients measured as percentage of live and

94

Table 6 - NK cell activity* in canine lymphoma patients
versus paired and all controls.

 

HOURS l§_HQHB§
50: 00: 50:1 199;;
QBQQE_; (untreated)
Controls:
Mean 10.8 15.5 11.9 23.0
SEM 2.3 5.7 2.2 8.9
Number 7 5 7 5
Patients:
Mean 7.7 10.0 12.1 14.6 a
SEM 2.4 2.8 2.9 5.2
Number 7 5 7 5
GROUP II (glucocorticoid-treated)
Controls:
Mean 17.0 25.7 22.1 25.3
SEM 7.3 7.3 6.2 8.5
Number 5 5 5 5
Patients:
Mean 15.3 24.1 21.5 26.8
SEM 5.9 7.5 9.6 9.9
Number 5 5 5 5
GROUP III (combined chemotherapy-treated)
Controls: ‘
Mean 13.5 13.4 10.8 12.3
SEM 6.0 3.7 3.4 2.5
Number 3 4 4 4
Patients:
Mean 3.9 a 5.1 b 2.2 3.5 0
SEM 2.3 1.8 1.3 1.2
Number 3 4 4 4
A NT 0
Mean 16.9 23.6 19.2 24.5
SEM 2.3 2.9 2.4 3.3
Number 26 23 26 23

 

* Data expressed as percent cytotoxicity at 12 and 16 hours
of incubation with 50:1 and 100:1 (E:T).

a = p <.05
b = p <.01
c = p <.025

95

.maouucoo Ham you Hound ouoocmpm
H |\+ 533 on» mgsmmmummu anaconda Hoowuumtr €53 comm anacofiuom
.maouucoo condom can musmfiumm 38:05: deacon you Auouum pumps—mum
913 some; muwofixououmo usmoumm ca auflfinuom Sumo xz I mm whamflm

96

 

_mNnE

mnempzou
8%.. >155: m

35.5 sags: N

am

".5
30.5

a n5
no.5

mm musmmm

a
E
n: ”mu
1
m
ow m..
IL
nu
wIA
D
IL
am A

0v

97

Table 7 - Single cell binding activity* in canine lymphoma
patients versus paired and all controls.

 

12 HOURS 6 HO
Live Qggg Livg Dead
Gtoup I (untreated)
Controls:
Mean 15.9 0.9 15.2 1.0
SEM 4.4 0.6 4.6 0.4
Number 7 7 6 6
Patients:
Mean 14.7 0.4 14.3 0.2
SEM 6.2 0.2 4.5 0.2
Number 7 7 6 6
Group II (glucocorticoid-treated)
Controls:
Mean 11.5 3.4 14.0 2.8
SEM 3.4 2.3 3.0 1.3
Number 4 4 4 4
Patients:
Mean 10.5 0.5 16.3 0.3
SEM 1.2 0.3 2.1 0.3
Number 4 4 4 4
Group III (combined chemotherapy-treated)
Controls:
Mean 26.0 3.3 24.0 2.3
SEM 5.7 1.6 5.2 1.3
Number 4 4 4 4
Patients:
Mean 29.5 0.5 31.0 1.8
SEM 3.6 0.3 9.5 1.4
Number 4 4 4 4
ALL CONTROLS
Mean 19.0 2.0 21.2 1.2
SEM 3.8 0.2 2.8 0.2
Number 23 23 24 24

 

* Data expressed as number of live or dead target cells
bound per 200 lymphocytes at 12 and 16 hours of incubation.

98
dead targets bound to lymphocytes at 16 hours incubation are
illustrated in Figure 29. Although patients generally had
reduced binding to dead targets compared to controls
suggesting fewer cytotoxic cells, the differences were not
statistically significant.

Table 8 indicates the effects of combined chemotherapy
in relation to the initial or pre-treatment levels of NK
cell activity. The data suggested an association between
relative initial levels and the relative degree of reduction
in NK cell activity seen with combined chemotherapy. For
example, high initial NK: cell activity compared. to all
controls is not protective to the decrease in NK activity.
The lowest initial activity, on the other hand had the least

reduction following combined chemotherapy.
CONCLUSIONS

The results of this study were similar to human studies
and indicated a significant reduction in NK cell activity
for untreated lymphoma patients when compared to their
paired controls and measured in percent cytotoxicity. No
significant differences were found between lymphoma patients
and their controls relative to the relative frequency of
live target binding cells or cytotoxic binding cells in
effector cell populations. Despite individual effectiveness,

these active NK cells were less cytotoxic overall than

 

99

 

.waouucoo Ham you Mound choocmum
a |\+ coma can mucmmmummu mcflnoumn HMOfivuwer :35 moms HausoNHuom
.maouucoo 60.309 0.8 mucmfluwm waonmaha mafia—m0 you Conrad cumosmum
moan coma wouaoonmga on 0:35.; ummumu 95H unwouom I ma ougflm

 

 

100

3Nu£

a .35

“was a n5
__ no.5 _ no.5

 

mnempzou
85% ESE: m

929:: signs: @

mm munmfih

SNIUNIB 3N1 .LNEOHEd

0N

mN

101

Table 8 - Degree of reduction in NK cell activity following
combined chemotherapy relative to initial levels as compared
to the initial NK cell activity levels relative to normal
controls.

 

Relative Degree of

 

Dgg_£ Therapy Befote Tegting Initigl a gggpgtigp b
5 None ~72% 25%
2 None -31% C 61%
8 Glucocorticoids +16% 85% d
9 Glucocorticoids +139% 98%
a = mean difference in % cytotoxicity compared to all

controls for 4 experiments unless otherwise stated

b = mean difference in % cytotoxicity for 4 experiments
unless otherwise stated

c = mean of 2 experiments

d = mean of 3 experiments

102
controls. Therefore it appears that a mechanism of immune
dysfunction in canine lymphoma involves an inability of NK
cells to recycle.

Different effects from treatment were seen depending
upon the type used. Dogs having received glucocorticoids
prior to testing had normal NK cell activity, suggesting
canine NK cells are steroid-resistant. COP or CHOP treated
patients, on the other hand, had marked inhibition of NK
cell activity, the mechanism of which appears similar to
untreated lymphoma dogs. Neither 'treatment significantly
depressed individual effector cell function suggesting these

drugs were not preferentially cytotoxic to NK cells.

DISCUSSION

Unique insight into the likely nature of the NK cell
dysfunction in canine lymphoma is permitted by the
concurrent use of chromium release and single cell binding
assays. These procedures when performed together have helped
to determine the mechanism of the defect for other disease
conditions such as, Chediak-Higashi, Hairy cell leukemia and
advanced solid epithelial malignant tumors (Abo gt g1. ,
1982: Fontana gt gl., 1986: Steinhauer gt 91., 1982). The
presence of decreased NK cell activity in untreated canine
lymphoma as measured by a chromium release assay agrees with

all known reports involving human lymphoma (Lipinski gt gI.,

103
1982: Tursz gt gI., 1982: Hawrylowicz gt gI., 1982: Gastl gt
gI., 1984) . However, in none of these reports were single
cell target binding and chromium release assays performed
together.

In the present canine study using single cell binding
assays, the relative frequency of effector cells bound to
live or dead target cells was the same in untreated lymphoma
patients as compared to their controls indicating active NK
cells were present to the same degree in the effector cell
population used for the assay. Therefore the decrease in
chromium release from dead target cells could not be
attributed to relatively low numbers of active NK cells.
Individual cells in the chromium release assay were able to
detach from dead target cells and bind to additional targets
thereby accomplishing multiple cytotoxic reactions. This
ability to "recycle" has been determined to be defective in
certain solid tissue malignancies when decreased NK cell
activity is unrelated to the frequency of effector cells
(Steinhauer gt gI., 1982). It could be concluded that canine
lymphoma also involves a recycling defect.

The exact mechanism for decreased NK cell activity in
lymphoma is unknown but possible explanations, in addition
to recycling defects, include depressed levels of produced
or available interferon (Hawrylowicz _et gI., 1982). They
reported only half of the cases responded to pretreatment of

IQ yittg interferon suggesting low levels of it alone could

104
not account for all the differences. _I_1_1 yiyg studies (Foon,
1986) show again only partial response to interferon.

Other cytokines or biological response modifiers, such
as interleukin 2 have demonstrated correction of deficient '
NK cell activity against fresh human lymphoid leukemias
(Lotzova gt gI., 1987). This would suggest a relationship
between low amounts of interleukin 2 available and NK cell
defects. Specifically this cytoki'ne is known to affect the
rate of cell lysis and release of cytotoxic factors from NK
cells (see Chapter 1).

The relationship of interleukin 2 and NK cell activity
in human leukemias relative to cytolytic rate may help to
explain percent cytotoxicity differences between controls
and untreated lymphoma dogs. The cytolytic rate relative to
recycling involves the following conditions: 1)rate and
ability of attachment by effector cells to target cells,
2)availability of cytotoxic factors with the cell, 3)rate of
release of these factors by effector cells and 4)rate of
detachment and movement towards other target cells. The
single cell binding assay results indicated conditions 1 and
3 were satisfied to a normal degree. The combined effect in
having many slower moving NK cells or those with lower
quantities of cytotoxic factors would help to explain why
the only statistically significant difference between the
above 2 groups occurred at the longest incubation period (16

hours). As time evolved, fewer numbers of competent

105
cytotoxic cells would reach their targets. The difference in
cytolytic rate in untreated lymphoma supports a possible
role of interleukin 2 in the treatment of canine lymphoma.
Additional studies, using In 1119 or adoptive transfer of
interleukin 2 cultured or activated effector cells, are
clearly necessary to better define the mechanism of immune
dysfunction in canine lymphoma.

Investigation into the absolute number of NK cells
present would contribute mmch to further understanding the
disease mechanism, however these were not determined due to
the lack of a practical marker for this canine cell type. An
earlier study (Chapter 3) showed that cells which stained
positive by direct immunofluorescence for the Fog receptor
correlated well with canine NK cell activity. This could be
used for future studies as one such. marker, but more
specific ones are necessary. Monoclonal antibodies that are
specific for murine and human NK cells include asialo GMl,
NK1.1, NKH-1 and Len-11. Of the studies evaluating NK cell
numbers in human lymphoma, one report (Baumann gt gI., 1986)
quantitated circulating NK cells by flow cytometry using the
monoclonal antibody NKH-l (Len-7). They found in previously
treated lymphoma patients, reduced NK cell numbers compared
to controls. In addition, Baumann gt _I. (1986) reported a
correlation between NKH-l positive cells and the clinical
course of these patients. No simultaneous assay for NK cell

activity was performed in this study to verify a low overall

106
activity existed. One report argued that the cell marker
Baumann gt g1. used only detects a portion of the active NK
cell population and may not truly reflect the entire
population (Foa gt g1., 1986).

To contrast the report of reduced NK cell numbers in
lymphoma, Gastl gt g1. (1984) found when compared to
controls, an increase in the absolute number of large
granular lymphocytes (LGL) present in non-Hodgkin's
lymphoma. These cells have been closely correlated with NK
cell activity in species other than dogs (Chapter 3). Gastl
gt g_. (1984) concluded that "defects other than the simple
reduction of effector cells are involved in this malignancy-
associated impairment of NK function".

Effects of glucocorticoids on NK cell function are
variable as indicated. in the literature. This 'may’ well
explain the lack of inhibition or augmentation in cytotoxic
activity or target binding ability when compared to controls
for canine patients treated with glucocorticosteroids. Many
early rodent and human studies (Oehler s. Herberman, 1978:
Hochman & CMdkowicz, 1979: Parrillo & Fauci, 1978: Bray gt
gI., 1983) noted transient It yiyg and In yittg reductions
in NK cell activity with hydrocortisone and dexamethasone.
Bray gt _1. (1983) concluded that the drug 1p yittg blocked
an early membrane signaling event necessary for activation
of the lytic mechanism.

A recent human study (Katz gt _I., 1984) using Ip viyg

107
and In yIttg corticosteroids on normal human subjects, found
on the other hand, increased NK activity which was
attributed to a selective depletion of noncytoxic cells from
circulation and a relative increase in the frequency of
target binding cells. Also, single cell binding and
cytolysis were not affected by in yiyg corticosteroids.

To further characterize the unresponsiveness of LGL to
corticosteroids, Katz gt gI. (1985) examined corticosteroid
receptor numbers and affinity but found no differences. In
addition, they concluded that corticosteroid receptor
characteristics bore more resemblance to those of
neutrophils and monocytes than T cells which had lower
receptors sites per cell and a lower dissociation constant.

An earlier study by Onsrud and Thorsby (1981)
demonstrated a change in NK cell activity over time
following IQ yiyg administration of corticosteroids to human
subjects. Activity was increased in samples collected at 4
hours while decreased for samples at 24 hours. Levels
returned to normal at 48 hours. Parallel variations were
found in the relative percentage of positive cells for the
FcG receptor, but not in their absolute numbers. This
suggests the effects of corticosteroids are due to a
reversible time dependent redistribution of cells other than
NK cells.

The variation in reported effects for glucocorticoids

may suggest the time factor is most critical. Incubation

108
periods used in the present canine study were relatively
long. A study involving shorter time periods might help to
fully explain the observed effects in dogs.

This study in dogs agreed with human and rodent studies
which reported reductions in NK cell activity due to
combined chemotherapy protocols or to their component drugs
alone. It has been shown with in y_1yp administration of
cyclophosphamide alone in normal rodents that a transient
dose dependent decrease in NK cell activity was produced
(Mantovani gt gI., 1979: Hanna & Fidler, 1980: Riccardi gt
gI., 1981: Ballas, 1986). Robinson & Heath (1968) reported
that cyclophosphamide depressed levels of interferon in the
intact animal but not In yittg. Since interferon inducers
will not boost the NK activity of cyclophosphamide-treated
mice (Hanna & Fidler, 1980), this suggests the relationship
to interferon is only a secondary response and not the
primary' mechanism. of cyclophosphamide-induced. inhibition.
The depressant effects on NK cell activity were short lived,
returning to normal levels by 7-12 days (Mantovani gt 21.,
1978: Riccardi gt gI., 1981: Ballas, 1986). The drug reduced
the number of positive FcG receptor cells (Hunninghake 8
Fauci, 1976) . Permanent effects on NK cell maturation and
activity were not seen with cyclophosphamide administration
as adoptive transfer i.e. , reconstitution with activated
spleen and bone marrow cells could remove the

cyclophosphamide induced effects of enhanced pulmonary

109
metastasis of tumors (Hanna 8 Fidler, 1980: Riccardi gt gI.,
1981).

Adriamycin (doxorubicin) given alone produced no change
in NK activity but analogs of it resulted in marked
decreases of activity (Mantovani gt g_., 1978).

The use of combined chemotherapy to treat acute
lymphoblastic leukemia in children, acute myelogenous
leukemia in adults and several solid tissue tumors in people
resulted in changes in NK cell activity (McGeorge gt gI.,
1982: Stupp gt gI., 1978: Braun 8 Harris, 1986). The drugs
used in common with the present canine study included
vincristine, prednisone, cyclophosphamide and doxorubicin.
Two of the above studies involved cyclophosphamide which has
already been shown to suppress NK cell activity. The third
study utilized vincristine and prednisone along with 6-
mercaptopurine and methotrexate (McGeorge _e_t gI. , 1982) .
These investigators demonstrated marked depression in NK
cell activity receiving this combined therapy but not for
patients in remission whose treatment had been discontinued
for at least 6 months. Therefore, combined chemotherapy
appears to produce no long term effects on NK cell function.

One report related the effects of chemotherapy to NK
cell activity levels prior to treatment in patients with
solid tissue tumors (Braun 8 Harris, 1986). They
demonstrated that patients with low pre-treatment NK cell

activity levels had augmentation of that activity with

 

 

 

 

 

g

110
chemotherapy. On the other hand, patients with normal pre-
treatment NK cell activity later developed depressed
activity following chemotherapy. The study concluded that
the effects of combined chemotherapy depended in part on
pre-treatment levels which in turn were dependent upon glass
adherent cell (monocyte) regulation of NK cell activity.

The present canine study had similar results regarding
pre-treatment levels and chemotherapy effects. The initial
pre-treatment NK cell activity related directly to the
degree their activity decreased as a result of treatment.
Initial activity which was much higher than controls, was
associated with a high degree of reduction in activity
following treatment, causing treatment values to be quite
low. An important aspect to further explore is whether the
levels of NK cell activity measured during treatment have
any bearing on the response of the patient to these drugs or
to overall survival.

The role of NR: cell numbers or activity’ has been
correlated to prognosis in human lymphoma and acute
myelogenous leukemia (Baumann gt gI., 1986: Stupp gt gI.,
1978). Additional studies are necessary to determine this
relationship in canine lymphoma.

In summary of this chapter, the study indicated
depression of NK cell activity in canine lymphoma patients
compared to normal controls. It was concluded for untreated

canine lymphoma that the mechanism of immunologic

111
dysfunction. relative to the NK cell defect involves
recycling activity rather than a deficiency of active
binding or cytotoxic effector cells. Treatment affected only
NK cell activity measured as percent cytotoxicity, not the
frequency of NK cells but this effect depended upon the type
of treatment. Short or long term glucocorticoid therapy did
not influence NK cell activity as measured in a 12 or 16
hour incubation chromium release assay. However, more
aggressive therapy involving cyclophosphamide, vincristine,
prednisone and frequently doxorubicin did significantly

depress NK cell activity.

SUMMARY

Three aspects of canine NK cell activity were covered
in this study. These involved: 1) morphologic, cytochemical
and immunologic characterization of the NK effector cell
population: 2) recycling and single cell target binding
capability of NK cells from healthy and lymphoma dogs: and
3) treatment effects on NK cell function in lymphoma dogs.

Morphologic features at both light and electron
microscopic levels were similar to those observed in other
species. The majority of binding cells were large
lymphocytes having several lysosomal granules within their
cytoplasm. In binding to targets, NK cell surface features
changed from irregular villous interdigitations to smooth
and extensive tight junctions. Cytoplasmic granules were
oriented most frequently between the NK cell nucleus and
target cell membrane, supporting their role in cytolysis.

The effector cell population assumed to be NK cells was
cytochemically heterogenous with variable numbers of cells
staining focally with acid phosphatase and alpha naphthyl
butyrate esterase.

Immunologic characterization of the effector cell

population confirmed earlier canine studies which found a

112

113
majority of cells staining negative for surface
immunoglobulins. Cells staining positive for the Fc receptor
of IgG were found more frequently, but still involved a
minor portion of the population.

NK cell function was measured in both chromium release
and single cell binding assays to evaluate overall activity
and relative frequency of active binding and cytotoxic
cells, respectively. The population of healthy dogs studied
had a wide range of NK cell activity, but individual dogs
generally maintained constant levels over time. There was no
correlation between the number of large granular lymphocytes
and NK cell activity or percent dead target cell binding.
However cells positive for FcG receptors correlated
positively to NK cell activity and negatively to percent
live target binding. It was concluded that immunologic
markers related better to levels of NK cell activity than
morphologic features.

The NK cell activity for untreated and combined
chemotherapy treated lymphoma dogs was significantly less
than that of healthy controls. The relative frequency of
active binding or cytotoxic NK cells was not affected in any
of the lymphoma groups. It was concluded that the immune
defect in untreated and combined chemotherapy treated
lymphoma dogs relative to NK cell function involved
decreased recycling ability to lyse multiple target cells.

Therapy with glucocorticoids had little effect on either

114
overall activity or cytotoxic binding cell numbers which

suggested the canine NK cell may be steroid-resistant.

APPENDICES

APPENDIX A

NATURAL KILLER CELL ACTIVITY IN NORMAL DOGS

115

Table 9 - NK cell activity* of 26 samples from 18 healthy
dogs (raw data).

 

 

I; flours I6 Houts
50:1 100:1 0: 100:;
Males:
19.6 35.1 25.5 43.4
45.4 52.7 45.1 58.7
13.9 17.1 0 18.6
Mean (SEM) 26.3 (9.7) 35.0 (10.3) 23.5 (13.1) 40.2 (11.7)
Intact Females:
34.5 51.3 30.2 50.6
9.0 11.8 10.6 13.0
2.9 6.1 5.0 12.6
37.0 NT 47.3 NT-
13.4 23.3 22.4 13.3
28.1 NT 41.3 NT
2.6 12.8 8.1 9.7
1.8 8.8 8.4 18.4
8.1 9.7 6.2 13.3
6.9 15.8 13.2 18.2
Mean (SEM) 14.4 (4.3) 17.5 (5.2) 19.3 (4.9) 18.6 (4.7)
Neutered Females:
5.0 5.5 4.5 6.2
10.5 10.6 6.0 7.0
10.5 25.0 19.0 23.3
16.6 NT 19.6 NT
25.4 22.5 19.1 11.5
15.4 37.8 21.8 57.5
11.5 14.1 13.3 19.6
33.7 42.9 24.4 33.7
14.3 25.1 18.2 23.5
22.9 37.8 26.7 41.0
7.6 19.3 20.6 15.3
15.3 21.5 14.8 17.2
27.5 35.6 27.4 38.0
Mean (SEM) 16.6 (2.3) 24.8 (3.4) 18.1 (1.9) 24.5 (4.4)
TOTAL MEAN 16.9 23.6 19.2 24.5
SEM 2.3 2.9 2.4 3.3

 

* Data expressed as percent cytotoxicity at 12 and 16 hour

incubations with 50:1 and 100:1 (E:T).
NT = not tested

116

Table 10 - Comparison between percent LGL, percent
cytotoxicity and percent live or dead binding in 26 samples
from 18 healthy dogs.

 

% LGL % Cytotoxicity1 g Live Binding2 3 Deag LindIng2

 

 

12.6 19.6 0.5 0.5
2.6 45.4 4.5 0.5
0.7 13.9 3.5 2.5
2.3 34.5 10.0 1.5
4.2 9.0 3.0 2.0
7.1 2.9 9.0 0.5

13.5 37.0 21.5 1.5
0.6 13.4 2.0 0
4.6 28.1 13.0 1.5
6.4 2.6 9.5 1.0

15.6 1.8 10.5 0
8.2 8.1 9.0 0
5.8 6.9 9.0 5.5
5.4 5.0 13.0 0
7.3 10.5 17.5 2.0
6.8 10.5 1.5 0.5
1.0 16.6 1.5 0
4.0 25.4 21.0 1.0
6.2 15.4 4.5 0.5

12.9 11.5 11.0 0
4.4 33.7 2.0 1.5
9.7 14.3 17.5 1.5

10.1 22.9 0 1.5
8.4 7.6 3.5 0.5
2.1 15.3 12.0 0

16.8 27.5 5.5 0.5

TOTAL MEAN (SEM)
6.9 (0.9) 16.9 (2.3) 8.4 (1.2) 1.0 (0.2)

 

5...:
ll

Percent cytotoxicity determined at 12 hours incubation
and 50:1 (E:T).
2 = Binding assay determined at 12 hours incubation.

APPENDIX B

NATURAL KILLER CELL ACTIVITY IN UNTREATED AND TREATED CASES

OF CANINE LYMPHOMA

117

i x005 090 .. HHH 8x0} 08000

0300.3 m o no?

2003 5 0802680 0 > H She}. .955 238}
E, 9008 090 .. HHH $\m~\m $300M :8

s is

Amfig 8 08096000 n E H $\m\m 2392.5 2883
50m 93

083 a w a:

082 0 HHH H B\3\~ 00308 £280

8* 9003 090 .. HHH $33. .300 38
03.003 x m 5%

082 0 B H $\~H\~ has 83QO

E 003 08 .. HHH $\e\~ 50m .60 006
owumman h m a:

082 n > H .00me9 g 000nmm\~

181003 8 080.9808 n > H She} mug 33
602003.52 h e “ma

8003 ~ you 082 0 > H Skis Hagen 358}

3 ~33 a 3 Hi 3
0838.0? 03 can
t 03888

 

.3038 EH 05:8 .80 58 209:8 06 g .. S 088..

118

Age .5033866

0:33.023 .3358 u0.60.2695 6:60ng, .06Hfiwfimoaoaoaonm8v

83808008 92 8. 8.3.2800 0030 u m
530§£§H03Huzou§ooo0§30oua

082 0 HHH H B\~\e 59025 08832

5.800

_ 2 0 rm

082 0 E H $\0~\m 2392.5 20233

208000 0 8 33000 00680
gag a H. “0%
20 ed 08005303 0 HHH HH SSH} 5.9032 80832
893

0300.3 2 w “Ox:

082 0 > H 33>. .295 0009”th

0.02 38

0300.3 2 3 a:

8 S 080.9680 n > HH B\e~\m .295 ~088on
jg 0.89 no 3034003 mug 3 Hi lg
082832 000 300

e 045888

803083 HH 0309

Table 12

119

patients versus paired controls.

- NK cell activity* in untreated canine lymphoma

 

Mean
SEM
Number

Mean
SEM
Number

0:

16.6
16.6
10.5
15.4
11.5

2.9

1.8

PM PH
enough-mo:-
0....

houmouu

\JNH
N0

ONTRO

00'

NT
NT
10.6
37.8
14.1
6.1
8.8

12_EQQB§
BAILEHIfi
52:1 19911
20.3 NT
10.0 NT
7.2 7.3
5.6 20.3
1.3 5.2
7.7 10.8
1.7 6.2
7.7 10.0
2.4 2.8
7 5
16 HOURS
25.1 NT
15.3 NT
2.8 6.2
17.3 34.8
6.7 11.6
10.5 12.9
7.1 7.5
12.1 14.6
2.9 5.2
7 5

 

* Data expressed as percent cytotoxicity at 12 and 16 hours

of incubation with 50:1 and 100:1 (E:T)
NT = Not Tested

120

Table 13 - NK cell activity* in glucocorticoid-treated
canine lymphoma patients versus paired controls.

 

_LLHQLJRi
CONTRO EATIENTs

59:1 190:; 5051 19211
9.0 11.8 5.2 20.7
13.4 23.3 20.6 28.5
45.4 52.7 33.5 47.0
10.5 25.0 16.6 23.0
6.9 15.8 10.6 1.3
Mean 17.0 25.7 15.3 24.1
SEM 7.3 7.3 5.9 7.5

Number 5 5 5 5

16 HOURS
QQNTROLfi A NT

50:; 100:; 50: 100:1
10.6 13.0 11.8 17.5
22.4 13.3 22.0 26.6
45.1 58.7 57.5 63.3
19.0 23.3 11.9 22.6
13.2 18.2 4.3 4.1
Mean 22.1 25.3 21.5 26.8
SEM 6.2 8.5 9.6 9.9

Number 5 5 5 5

 

* Data expressed as percent cytotoxicity at 12 and 16 hours
of incubation with 50:1 and 100:1 (E:T)

121

Table 14 - NK cell activity* in combined chemotherapy
treated canine lymphoma patients versus paired controls.

 

 

W
QONTROLS EATIENIS

59:1 100: 0: 190;;
25.4 22.5 7.9 9.1
6.9 15.8 3.6 6.7
5.0 5.5 NT 3.1
8.1 9.7 0.1 1.3
Mean 11.4 13.4 3.9 5.1
SEM 4.7 3.7 2.3 1.8

Number 4 4 3 4

16 HOURS
CONTROLS PATIENTS

50: 100:1 50: 0:
19.1 11.5 0 0.5
13.2 18.2 4.3 6.4
4.5 6.2 4.4 4.0
6.2 13.3 0 3.2
Mean 10.8 12.3 2.2 3.5
SEM 3.4 2.5 1.3 1.2

Number 4 4 4 4

 

* Data expressed as percent cytotoxicity at 12 and 16 hours
of incubation with 50:1 and 100:1 (E:T)
NT = Not Tested

122

Table 15 - Single cell binding activity* in untreated canine
lymphoma patients versus paired controls.

 

 

 

 

12_HQQB§
CONTROLS w

Lime Dead Lixe Dead

21 0 17 0

18 1 11 O

22 0 10 0

9 1 10 1

35 4 50 1

3 0 1 0

3 O 4 1
Mean 15.9 0.9 14.7 0.4
SEM 4.4 0.6 6.2 0.2

Number 7 7 7 7

16 HOUR
CONTROLS PATIENTS

Live Dead Live Dead

11 2 12 0

14 0 15 0

18 2 19 0

2 0 NT NT

36 0 33 0

6 1 4 0

6 1 3 1
Mean 13.3 0.9 14.3 0.2
SEM 403 0.3 405 002

Number 7 7 6 6

 

* Data expressed as number of live or dead target cells
bound per 200 lymphocytes at 12 and 16 hours of incubation.
NT = Not Tested

123

Table 16 - Single cell binding activity* in glucocorticoid-
treated canine lymphoma patients versus paired controls.

 

 

 

 

12 flODBS
CON OLS BAILENI§

Live Dead Live Deeg

18 11 7 O

3 1 12 0

9 1 11 1

4 0 NT NT

16 4 12 1
Mean 10 3.4 10.5 0.5
SEM 3.1 2.0 1.2 0.3

Number 5 5 4 4

16 HOURS
ONTROLS Eéilzfllfi

Live Dead Live Deeg

17 6 19 ‘0

20 0 19 0

6 3 10 1

16 0 NT NT

13 2 17 0
Mean 14.4 2.2 16.3 0.3
SEM 2.4 1.1 2.1 0.3

Number 5 5 4 4

 

* Data expressed as number of live or dead target cells
bound per 200 lymphocytes at 12 and 16 hours of incubation.
NT = Not Tested

124

Table 17 - Single cell binding activity* in cmbined

diamtherapy treated canine lynphana patients versus paired
controls.

 

w
m w
Lug Dead Lila Dead

18 O 24 0

26 O 33 1

18 11 23 1

42 2 38 0
Mean 26.0 3.3 29.5 0.5
SEM 5.7 1.6 3.6 0.3
Nunber 4 4 4 4

16 m
WIS m
Leve Dead lee E111

17 1 17 0

23 1 25 0

17 6 23 6

39 1 59 1
Mean 24.0 2.3 31.0 1.8
SEM 5.2 1.3 9.5 1.4
Nunber 4 4 4 4

 

*Dataeamressedasnmberofliveordeadtazgetoells
bandparZOOlymmocytesatRammrmrsofirmbation.

APPENDIX C

DETAILED TECHNIQUE PROCEDURES

125

CANINE LXHPHOCYTE ISOLATION

TER Sam les

4 - 10ml tubes of heparinized blood

Nalgene .20M filter 115ml (Nalge Co; #120-0020 Rochester,NY)

4 - sterile 50 m1 plastic centrifuge tubes with plug seal
cap (Corning #25330)

sterile saline

60 ml filtered Ficoll-Hypaque (R.I. = 1.3550)

28-32, (13ml) 17 X 100mm sterile plastic Falcon tubes with
caps (#2057 Oxnard, CA)

sterile deionized water

sterile 2xEPS

sterile MEME with 10% FCS, 1% glutamine, 1% pen/strep

sterile MEME without additives

sterile disposable 10ml syringe

0.75g sterile scrubbed nylon fiber (3 denier,3.81 cm, type
200 - #4C2906 Fenwal Laboratories, Deerfield, IL)

sterile 3 way stopcock

hemacytometer

37 C incubator (5% C02)

sterile Pasteur pipettes

10 ml sterile pipettes

parafilm

sterile Petri dish or beaker

sterile forceps

sterile carbonyl iron powder

timer

PROTOCOL

1. incubate each 10ml tube of blood following addition of
sterile carbonyl iron for 1 hr at 37 C

2. dilute each 10ml tube of blood with saline to 45ml of
50ml tube

3. place 3ml Ficoll-Hypaque into 10 Falcon tubes for each
patient and layer gently 9ml of diluted blood over it; 1800
RPM (450 G) for 18 min (BEGIN STEP #8)

4. remove lymph layer by Pasteur pipette and place into 4
Falcon tubes per each patient; fill tube with saline; mix;
1500 RPM (300 G) for 8 min

126

5. pour off supernatant; fill tube halfway with deionized
water: wait 6 sec: agitate 6 sec; add 2xEPS to top; 1000 RPM
(140 G) for 8 min

6. pour off supernatant; add saline and combine tubes into 2
per each patient; 1500 RPM (300 G) for 8 min

7. pour off supernatant; resuspend into 5 m1 MEME with
(additives4 fill hemacytometer and differentiate
granulocytes/lymphocytes

8. attach stopcock and hydrate sterile nylon wool in Petri
dish with MEME without additives; place into syringe; fill
with 10 ml warm MEME without additives keeping medium above
column and allowing a moderate drip rate (use Pasteur
pipette)

9. add lymphocyte suspension slowly while eluding the media
with the stopcock; once the suspension is completely added,
close the stopcock; cover top 'with Parafilm; allow to
incubate 1 hr at 37 C (BEGIN TARGET CELL ISOLATION)

10. elute the non-adherent cells by slowly adding 15 ml warm
MEME without additives while keeping the level of media
above the nylon column

11. fill hemacytometer; resuspend in complete media and
adjust to 5 x 106 cells/ml (100 cells/5 hemacytometer
squares) REMEMBER TO AVERAGE COUNT TO 5 SQUARES!

Save cell suspension for cytospin preps (minimum of 6) for
Wright's, chloroacetate esterase, naphthyl butyrate
esterase, and acid phosphatase staining (3-4 drops of lymph
suspension/cytospin container, 1000 RPM for 1-5 min)

127
TARGET CELL ISOLATION and LABELLING

STV (trypsin) (filter periodically)

Chromium-51

Lml tuberculin syringe with 25g needle

37 C water bath

sterile MEME with additives

Triton X-100 solution

sterile tissue culture 96 U shaped well plates with lids
(Linbro Titertek #76-013-05, Mc Lean, VA)

supernatant collection system with harvesting frames and
macrowell tube strips (Skatron, Inc Harvesting Set #7075,
Sterling, VA)

dispensing pipette (200 p1)

PROTOCO

1. add 3ml STV to culture flask with target cells; pour out:
add fresh 3ml STV and allow to incubate 20 min at 37 C

2. split cells passing some into new flasks with remainder
placed in 10ml complete MEME; count cells/5 squares

3. use 10 ml of cell suspension (concentration of
approximately 10 cells/5 squares); save some unlabelled
targets for single cell binding assay
4. pellet target cells at 1500 RPM (300 G) for 10 min
5. Cr labelling calculated as follows:

Activity X 2 mCi X 1000 = units pCi needed

units pCi needed 100 pCi

6. add Cr and separate cells through syringe and needle;
incubate cells for 90 min in 37 C water bath (PREPARE PLL
COVERSLIPS)

7. add 5 ml cold MEME to target cell mixture; 1500 RPM for
10 min; discard supernatant

8. add 2 ml cold MEME to target cell mixture; agitate cells
gently by pipette; incubate in ice bath for 30 min
(BEGIN VIABILITY TESTING AND SCBA)

9. make 10% triton solution by adding 0.5m1 Triton x-1oo to
5ml MEME

128

10. determine cell count in the Zml; pellet cells at 1500
RPM for 10 min; remove supernatant and adjust count to 1
cell/5 squares with additional complete MEME; pellet; add
fresh medium especially if frequent aspirations were
necessary to obtain proper count

11. fill 2 U shaped well plates with labelled targets,
triton, and effector cells; centrifuge at 4 C at 1000 RPM
(200 G) for 2 min with blank plate

12. plates read at 0, 12, and 16 hours using supernatant
collection system with a gamma counter

Gamma Counter:
TM Analytic, Inc Model 1185
Elk Grove Village, IL

Chromium-51 settings:
Base 5.6; Window 1.6
Attenuator: Coarse (inside) 4 (green)
Fine (outside) 25 (red)
Coarse 900; Fine 3.6
Wide Differential
Set time to minutes; 1 min
High voltage - on
Test - off
Master - on
Auto Stop
Count Twice

129
Setting Up Trays
lymphocytes or effector cells:
100 cells/5 squares n 5 X 106/ml

target cells:
1 cell/ 5 squares = 5 X 104/ml

100:1 50:1 spont max
IREEQ'ELZQ'SQIT"MISS-{xi """ 33';I"""3"""""3""
targets (min 10ml) 100 pl . 100 pl 100 pl 100 pl
triton (min 2ml) 0 O O 100 pl

N913: total amount per well = 200 pl or .2 ml

SAMPLE TRAY DIAGRAM FOR CONTROL AND PATIENT

Os OS Om Om 12S 12s 12m 12m 16S 16s 16m 16m

12C 12C 12C 12P 12P 12P 16C 16C 16C 16P 16P 16F
50 100 100 50 100 100 50 100 100 50 100 100

SAMPLE COLLECTION ORDER

0 hour - Max(6) + Spont(8)

+

12, 16 hours - Max(6) + Spont(8) + P 100(8) + C 100(8)
P 50(4) + c 50(4)

130

PREPARATION OF POL!-LPLKSINE COATED COVERSLIPS
(Modified from Vargas-Cortes J Immunol Meth, 1983)

35 mm plastic petri dishes with lids (9)

22 mm X 22 mm glass coverslips

poly-L—lysine hydrobromide (# P6282 5 mg; MW 90,000; degree
«of polymerization 430; Sigma Chemical Co, St. Louis, MO

sterile water

HEPES buffered Hank's pH 7.3

OCO

1. add 50 ml sterile distilled water to 5mg lyophilized PLL
jproducing 0.1 mg/ml (keep frozen)

2. prepare HEPES buffered Hank's (pH 7.3)
450 m1 sterile distilled water
50 ml 10X (conc) Hanks Balanced Salt Solution
1.5 ml 7.5% Na bicarbonate solution
26.3 ml sterile HEPES buffer (pH 7.2)

3. add 1 ml PLL mixture to 49 ml HEPES buffered Hank's

solution; place 10ml in 5 Falcon tubes (2 pg/ml) and freeze
extra tubes

PROTOCOL FOR COVERSLIP COATING
1. keep 1 tube refrigerated and place 1 m1 (about 15-20
drops) to coat coverslip placed in petri dish; rock to cover
completely
2. incubate at room temperature for 45 min

3. wash well with HEPES buffered Hanks; lift coverslip off
dish and allow to dry before using

131

VIABILITY TESTING: SINGLE CELL BINDING AND CYTOTOXIC ASSAY

filtered 0.1 % trypan blue in HEPES buffered Hank's
9 dried PLL coated coverslips

12 conical plastic tubes (1.25 ml)

Pasteur pipettes

Fflme without additives (serum free)

HEPES buffered Hank's solution

1% formaldehyde in PBS

glass slides

hemacytometer

PROTOCOL

l. pellet isolated lymphocyte population (1500 RPM X 8 min);
add serum free MEME to 5 X remaining volume to produce 20
cells/5 squares or 1 X 106 cells/ml

2. pellet (unlabelled) target cell suspension (1500 RPM X 10
min); add serum free MEME to produce concentration of 20
cells/5 squares or 1 X 106 cells/ml

3. place 100 pl of target cell suspension in 10 conical
tubes for: 1 viability, 3 target control, 6 E:T mixtures:
place 100 pl of control/patient effector cell suspensions in
8 tubes for: 2 X 1 viability, 2 X 3 E:T mixtures (1:1).
Total tubes = 12.

4. test viability by adding equal volume (100 pl) of cell
suspension and trypan blue; leave stain for 5-15 min; count
% dead on hematocytometer

5. pellet remaining 8 tubes in microfuge (Fisher) at low
speed 1500 RPM (200 G) for 5 min

6. incubate tubes at 37 C for 15 min; suction off
supernatant

7. resuspend pellet with one suction by Pasteur pipette in
serum free MEME (0.5 ml/tube)

8. plate mixture onto PLL coated coverslips; incubate 20 min
at room temperature to allow cell attachment (0 hr); remove
1 target control and 1 each of E:T coverslips for initial
testing

9. incubate remainder coverslips for additional 12 and 16
hours at 37 C in 5% C02

132

10. following incubation 0, 12, 16 hour samples are tested
for: % spontaneous dead (target controls)
% conjugate E:T binding/200 lymphocytes (live and
dead)

11. coverslips are prepared as follows:

wash dropwise with HEPES buffered Hanks or serum
free MEME

stain with 0.1% trypan blue (few drops) for 5 min

wash/fix slides (several drops) 1% formaldehyde for 5
min

invert coverslips onto glass slides; drain excess
fluid

PROTOCOL FOR ELECTRON MICROSCOPY SAMPLES
1. place equal volumes of effectors and targets (E:T = 1:1)
into small falcon tube. BEST TO USE >1 ML OF EACH FROM SCBA
WITH SERUM FREE MEME
2. centrifuge 1000 RPM (140 g) for 5 min

3. incubate at 37 C for 15 minutes minimally to allow cell
binding

4. suction off supernatant leaving 0.5 ml; resuspend gently

5. pipette mixture into plastic microfuge tubes and incubate
at 37 C for desired time e.g., 10, 12, 16 hours

6. replace medium with Karnovsky's fixative leaving for 2
hours minimally

7. pellet at high speed for 30 seconds

8. ready for EM processing

133

IMMUNOFLUORESCENCE STAINING - REAGENTS AND EQUIPMENT

sterile PBS and 0.02 % NaN :
filter 1 ml sodium azide solution (lOOmg/ml) and add to
500 ml sterile PBS

PBS +-0.02 % NaN3 + 5 % FCS (complete PBS):
add 20 m1 FCS to 380 ml of above solution

reconstitute lyophilized antisera - FITC conjugated using
sterile distilled water to volume suggested by manufacture;
aliqpot 50 pl into plastic microfuge vials and freeze

Fluorescein Conjugated F(ab')2 Fragment Goat Anti-Dog
IgM (Mu Chain Specific) (Cooper Biomedical Inc,
Cat #1305-0201, 2 m1, Malvern, PA)

Fluorescein Conjugated F(ab')2 Fragment Goat Anti-Dog
IgG (Heavy & Light chains specific) (Cooper
Biomedical Inc, Cat #1305-0081, 2 ml, Malvern,
PA)

Fluorescein Conjugated ChromPure Dog IgG, Fc Fragment
(Jackson Immunoresearch Laboratories, Inc;
Cat #004e1008, 1.1 ml, Avondale, PA)

dilutions made with PBS + sodium azide solution:
1:100 50 p1 to 5.0 ml PBS - use for smears
1:10 50 pl to 0.5 ml PBS - use for live cells

acetone for slide fixation

coplin jar

wooden applicator sticks

dark moist chambers for slide incubation
buffered glycerol to preserve slides
coverslips for cells and smears

glass pipettes

134
TECHNIQUE FOR IMMUNOFLUORESCENCE STAINING - CELLS (819)
1. place 0.3m1 of lymphocytes (100/5 squares) and 0.1ml
FITC-Antibody at 1:10 dilution into small falcon tubes
2. incubate 4 C for 1 hour (minimal time = 30 minutes)

3. wash twice with 1 ml PBS + FCS + NaN3; use refrigerated
centrifuge at 1500 RPM (500 G) for 5 min following washes

4. add 0.1 ml of complete PBS to pellet
5. transport to fluorescent scope unexposed to light
6. place 1 drop to glass slide, coverslip and read

7. place heat filter on top of FITC filter in field
diaphragm

8. place red suppressor filter in condenser holder

9. swing out auxillary lens

10. use #53 filter to detect green fluorescence; 0 filter
for total blue count of cells; do 100 cell differential

TECHNIQUE FOR IMMUNOFLUORESCENCE STAINING - SMEARS (019)

1. fix slides 30 seconds in acetone; air dry

2. wash 3 changes for 15 min total in "complete" PBS

3. place wooden stick applicator across slide, lay edge of
coverslip on stick and fill area below with antisera
(1:100): 300 UL NEEDED

4. incubate in moist chamber in refrigerator overnight (24
hours)

5. wash 3 changes for 15 min total in "complete" PBS

6. mount coverslip on glycerol coated area and place into
moist chamber

7. transport to fluorescent microscope and read

L-“

 

 

 

 

LIST OF REFERENCES

LIST OF REFERENCES

Abo T, Roder JC, Abo W, Cooper MD, Balch CM. Natural killer
(PINK-1"") cells in Chediak-Higashi patients are present in
normal numbers but are abnormal in function and morphology.
J Clin Invest 1982;70:193-197.

Alexander E, Henkart P. The adherence of human Fc receptor-
bearing lymphocytes to antigen-antibody complexes II.
Morphologic alterations induced by the substrate. J Exp Med
1976;143:329-346.

Altman A, Rapp HJ. Natural cell-mediated cytotoxicity in
guinea pigs: properties and specificity of natural killer
cells. J Immunol 1979;121:2244-2252.

Ballas ZK. Lymphokine-activated killer (LAK) cells I.
Differential recovery of LAK, natural killer cells and
cytotoxic T lymphocytes after a sublethal dose of
cyclophosphamide. J Immunol 1986;137:2380-2384.

Barka T, Anderson PJ. Histochemical methods for acid
phosphatase using hexazonium pararosanilin as coupler. J
Histochem Cytochem 1962;10:741-753.

Barlozzari T, Reynolds CW, Herberman RB. E vivo role of
natural killer cells: involvement of large granular
lymphocytes in the clearance of tumor cells in anti-asialo
GMl-treated rats. J Immunol 1983;131:1024-1027.

Baumann MA, Milson TJ, Patrick CW, Libnoch JA, Keller RH.
Correlation of circulating natural killer count with
prognosis in large cell lymphoma. Cancer 1986;57:2309-2312.

Behelak Y, Banerjee D, Richter M. Immunocompetent cells in
patients with malignant disease. I. The lack of naturally
occurring killer cell activity in the unfractionated
circulating lymphocytes from patients with chronic
lymphocytic leukemia (CLL). Cancer 1976;38:2274-2277.

135

136

Betton GR. Natural cell-mediated cytotoxicity in the canine.
In: Shifrine M, Wilson FD, eds. The eanine as a Diomegieel

zeseerch model: Immunelggieal. Demetelegleel. and
encolegieel eepects, Washington DC: US Department of Energy,
1980;99-126.

Betton GR, Gorman NT. Cell mediated responses in dogs with
spontaneous neoplasms I. Detection of cell-mediated
cytotoxicity by the chromium-51 release assay. J Natl Cancer
Inst 1978;61:1085-1093.

Bonavida B, Katz J, Gottlieb M. Mechanism of defective NK
cell activity in patients with acquired immunodeficiency
syndromes (AIDS) and. AIDS-related complex. I. Defective
trigger on NK cells for NKCF production by target cells and
partial restoration by IL 2. J Immunol 1986;137:1157-1163.

Bonavida B, Wright SC. Natural killer cytotoxic factors
(NKCF) role in cell-mediated cytotoxicity. In: Lotzova E,

Herberman RB, eds. lmmunobiolm ef natural killer eell .
Vol 1. Boca Raton, Florida: CRC Press, Inc, 1986;125-139.

Bozdech MJ, Bainton DF. Identification of alpha-naphthyl
butyrate esterase as a plasma membrane ectoenzyme of
monocytes and as a discrete intracellular membrane-bounded
organelle in lymphocytes. J Exp Med 1981;153:182-195.

Bradley TP, Bonavida B. Mechanism of cell-mediated
cytotoxicity at the single cell level. IV. Natural killing
and antibody-dependent cellular cytotoxicity can be mediated
by the same human effector cell as determined by the two-
target conjugate assay. J Immunol 1982;129:2260-2265.

Brahmi 2, Thomas JE, Park M, Park M, Dowdeswell IRG. The
effect of acute exercise on natural killer-cell activity of
trained and sedentary human subjects. J Clin Immunol
1985;5:321-328.

Braun DP, Harris JE. Effect of chemotherapy on NK function
in the peripheral blood of cancer patients. Cancer Immunol
Immunother 1986;21:240-245.

Bray R, Abrams S, Brahmi Z. Studies on the mechanism of
human natural killer cell-mediated cytolysis I. Modulation
by dexamethasone and arachidonic acid. Cellular Immunol
1983;78:100-113.

Brunner KT, Mauel J, Cerottini JC, Chapuis B. Quantitative
assay of the lytic action of immune lymphoid cells on 51Cr-
labelled allogeneic target cells ln vitre; Inhibition by
isoantibody and by drugs. Immunology 1968;14:181-196.

137

Carter RF, Valli VEO, Lumsden JH. The cytology, histology
and prevalence of cell types in canine lymphoma classified
according to the National Cancer Institute Working
Formulation. Can J Vet Res 1986;50:154-164.

Cotter SM. Treatment of lymphoma and leukemia with
cyclophosphamide, vincristine, and prednisone: I. Treatment
of dogs. J Am Animal Hosp Assoc 1983;19:159-165.

Couto CG. Canine lymphomas: something old, something new.
Compend Contin Educ Pract Vet 1985;7:291-302.

Crow SE. Lymphosarcoma (malignant lymphoma) in the dog:
Diagnosis and treatment. Compend Contin Educ Pract Vet
1982;4:283-292.

DeVries JE, Cornain S, Rumke P. Cytotoxicity of non-T versus
T-lymphocytes from melanoma patients and healthy donors on
short- and long-term cultured melanoma cells. Int J Cancer
1974;14:427-434.

Dj eu J Y . Ant ibody-dependent cel l -mediated cytotoxic ity and
natural killer-cell phenomenon. J Am Vet Med Assoc
1982;181:1043-1048.

Djeu JY, Heinbaugh JA, Holden HT, Herberman RB. Augmentation
of mouse natural killer cell activity by interferon and
interferon inducers. J Immunol 1979;122:175-181.

Ferrarini M, Cadoni A, Franzi AT, Ghigliotti C, Leprini A,
zicca A, Grossi CE. Ultrastructure and cytochemistry of
human peripheral blood lymphocytes. Similarities between the
cells of the third population and TG lymphocytes. Eur J
Immunol 1980;10:562-570.

Ferrarini M, Grossi CE. Ultrastructure and cytochemistry of
the human large granular lymphocytes. In: Lotzova E,

Herberman RB, eds. lmunobiolm ef naturel killer eell .
Vol 1. Boca Raton, Florida: CRC Press, Inc, 1986;33-43.

Foa R, Fierro MT, Lpsso P, Raspadori D, Ferrando ML, Matera
L, Malavasi F, Lauria F. Reduced natural killer T-cells in
B-cell chronic lymphocytic leukaemia indentified by three
monoclonal antibodies: Len-11, A10, ABB.28. Br J Haematol
1986;62:151-154.

Fontana L, DeRossi G, DeSanctis G, Ensoli F, Lopez M, Annino
L, Mandelli F. Decreased NK activity in Hairy cell leukemia
(HCL): An analysis at the cellular level. Blut 1986;53:107-
113.

138

Foon KA. Interferon therapy of lymphoproliferative
disorders. Seminars in Hematol 1986;23 Suppl 1:10-13.

Fujimiya Y, Bakke A, Chang WC, Linker-Israeli M, Udis B,
Horwitz D, Pattengale PK. Natural killer cell immuno-
deficiency in patients with chronic myelogenous leukemia. I.
Analysis of the defect using the monoclonal antibodies HNK-1
(Len-7) and B73.1. Int J Cancer 1986;37:639-649.

Gastl G, Niederwieser D, Marth C, Huber H, Egg D, Schuler G,
Margreiter R, Braunsteiner H, Huber C. Human large granular
lymphocytes and their relationship to natural killer cell
activity in various disease states. Blood 1984;64:288-295.

Gastl G, Schmalzl F, Huhn D, Gattringer C, Huber C. Large
granular lymphocytes: morphological and functionsl
properties I. Results in normals. Blut 1983;46:297-310.

Gengozian N, Longley RE, Filler J, Good RA. Natural killer
cells in the blood and bone marrow of the rhesus monkey.
Cellular Immunol 1986;101:24-38.

Gray KN, Raulston GL, Gleiser CA, Jardine JH. Histologic
classification as an indication of therapeutic response in
malignant lymphoma of dogs. J Am Vet Med Assoc 1984;184:814-
817.

Grimm E, Bonavida B. Mechanism of cell-mediated cytotoxicity
at ‘the single-cell level. I. Estimation of cytotoxic 1r
lymphocyte frequency and relative lytic efficiency. J
Immunol 1979;123:2861-2869.

Grossi CE, Ferrarini M. Morphology and cytochemistry of
human large granular lymphocytes. In: Herberman RB, ed. N3

eells end other natural effector cells. New York: Academic
Press, 1982;1-8.

Hanna N. In give activities of NK cells against primary and
metastatic tumors in experimental animals. In: Lotzova E,

Herberman RB, eds. lmmunebl'olmy ef naturel killer; celle.
Vol 2. Boca Raton, Florida: CRC Press, Inc, 1986;1-10.

Hanna N, Burton RC. Definitive evidence that natural killer
(NK) cells inhibit experimental tumor metastasis in yiye. J
Immunol 1981;127:1754-1758.

Hanna N, Fidler IJ. Role of natural killer cells in the
destruction of circulating tumor emboli. J Natl Cancer Inst
1980;65:801-809.

139

Hanna N, Schneider M. Enhancement of tumor metastasis and
suppression of natural killer cell activity by beta-
estradiol treatment. J Immunol 1983;130:974-980.

Hansson M, Kiessling R, Andersson B, Welsh RM. Effect of
interferon and interferon inducers on the NK sensitivity of
normal mouse thymocytes. J Immunol 1980;125:2225-2231.

Hawrylowicz CM, Rees RC, Hancock BW, Potter CW. Depressed
spontaneous natural killing and interferon augmentation in
patients with malignant lymphoma. Eur J Canc Clin Oncol
1982;18:1081-1088.

Hellstrom I. A. colony inhibition (CI) technique for
demonstration of tumor cell destruction by lymphoid cells in
vitro. Int J Cancer 1967;2:65-68.

Hellstrom I, Hellstrom KE, Pierce GE, Yang JPS. Cellular and
humoral immunity to different types of human neoplasms.
Nature 1968;220:1352-1354.

Hellstrom I, Hellstrom KE, Sjogren HO, Warner GA.
Demonstration of cell-mediated immunity to human neoplasms
of various histological types. Int J Cancer 1971;7:1-16.

Henkart MP, Henkart PA. Lymphocyte mediated cytolysis as a
secretory phenomenon. Adv Exp Med Biol 1982;146:227-242.

Henkart PA. Mechanism of lymphocyte-mediated cytotoxicity.
Ann Rev Immunol 1985;3:31-58.

Henney CS, Kuribayashi K, Kern DE, Gillis S. Interleukin-2
augments natural killer cell activity. Nature 1981;291:335-
338.

Herberman RB. Natural killer cells. In: Dixon FJ, Fisher DW,

eds. W of immunologic disease. Sunderland,

Massachusetts: Sinauer Associates, Inc, 1983;75-85.

 

Herberman RB. Mechanisms and pharmacology of NK cell
activity. Adv in Immunopharmacol 1985a:189-192.

Herberman RB. Introduction. In: Herberman RB, Callewaert DM,

eds. Mechanisms e; cytotoxicity by NK cells. Orlando:
Academic Press, Inc, 1985b;1-13.

 

 

Herberman RB, Djeu JY, Kay HD, Ortaldo JR, Riccardi C,
Bonnard GD, Holden HT, Fagnani R, Santoni A, Puccetti P.
Natural killer cells: Characteristics and regulation of
activity. Immunol Rev 1979;44:43-70.

1110)

Hum
CYc

SL1;
197

11qu
hum:

140

Herberman RB, Nunn ME, Holden HT, Lavrin DH. Natural
cytotoxic reactivity of mouse lymphoid cells against
syngeneic and allogeneic tumors. II. Characterization of
effector cells. Int J Cancer 1975b;l6:230-239.

Herberman RB, Nunn ME, Lavrin DH. Natural cytotoxic
reactivity of mouse lymphoid cells against syngeneic and
allogeneic tumors. I. Distribution of reactivity and
specificity. Int J Cancer 1975a;16:216-229.

Herberman RB, Ortaldo JR. Natural killer cells: Their role
in defenses against disease. Science 1981;214:24-30.

Heumann D, Colombatti M, Mach J-P. Human large granular
lymphocytes contain an esterase activity usually considered
as specific for the myeloid series. Eur J Immunol
1983;13:254-258.

Hiserodt JC , Beals TF . Ultrastructural analysis of human
natural killer cell-target cell interactions leading to
target cell lysis . In: Herberman RB, Callewaert DM , eds .

Meehanisms e: eygetoxicl'ty by N15 cells. Orlando: Academic
Press, Inc, 1985;195-204.

Hiserodt JC, Laybourn KA, Varani J. Expression of a laminin-
like substance on the surface of murine natural killer (NK)
lymphocytes and its role in NK recognition of tumor target
cells. J Immunol 1985;135:1484-1487.

Hochman PS, Cudkowicz G. Suppression of natural cytotoxicity
by spleen cells of hydrocortisone-treated mice. J Immunol
1979;123:968-976.

Holmberg CA, Manning JS, Osburn BI. Canine malignant
lymphomas: comparison of morphologic and immunologic
parameters. J Natl Cancer Inst 1976;56:125-135.

Hooper WC, Barth RF, Shah NT. Lack of natural killer cell
activity in hairy cell leukemia patients and partial
restoration with interleukin-2. Cancer 1986;57:988-993.

Huhn D, Huber C, Gastl G. Large granular lymphocytes:
morphological studies. Eur J Immunol 1982;12:985-988.

Hunninghake GW, Fauci AS. Divergent effects of
cyclophosphamide administration on mononuclear killer cells:
quantitative depletion of cell numbers versus qualitative
suppression of functional capabilities. J Immunol
1976;117:337-342.

Huwyler T, Hirt A, Morell A. Effect of ascorbic acid on
human natural killer cells. Immunol Letters 1985;10:173-176.

141

Jain NC. Cytochemistry of normal and leukemic leukocytes.
In: W- 4th ed Philadelphia:
Lea & Febiger, 1986;909-939.

Jezewska E, Bjorkholm M, Giscombe R, Holm G, Tullgren 0.
Surface markers and cytotoxic activity of blood natural
killer cells studied at the single cell level in Hodgkin's
disease. Clin Exp Immunol 1985;61:96-102.

Karre K, Klein GO, Kiessling R, Klein G, Roder JC. Low
natural in 2129 resistance to syngeneic leukaemias in
natural killer-deficient mice. Nature 1980;284:624-626.

Kasza L. Establishment and characterization of canine
thyroid adenocarcinoma and canine melanoma cell lines. Am J
Vet Res 1964;25:1178-1185.

Katz P, Zaytoun AM, Lee JH. The effects of in give
hydrocortisone on lymphocyte-mediated cytotoxicity.
Arthritis Rheumatism 1984;27:72-78.

Katz P, Zaytoun AM, Lee JH. Characterization of
corticosteroid receptors in natural killer cells: comparison
with circulating lymphoid and myeloid cells. Cellular
Immunol 1985;94:347-352.

Kay HD, Sinkovics JG. Cytotoxic lymphocytes from normal
donors 1974 v.2:296-297.

Kay NE, Zarling JM. Impaired natural killer activity in
patients with chronic lymphocytic leukemia is associated
with a deficiency of azurophilic cytoplasmic granules in
putative NK cells. Blood 1984;63:305-309.

Kiessling R, Klein E, Pross H, Wigzell H. Natural killer
cells in the mouse. II. Cytotoxic cells with specificity for
mouse Maloney leukemia cells. Characteristics of the killer
cell. Eur J Immunol 1975b;5:117-121.

Kiessling R, Klein E, Wigzell H. Natural killer cells in the
mouse. I. Cytotoxic cells with specificity for mouse Moloney
leukemia cells. Specificity and distribution according to
genotype. Eur J Immunol 1975a;5:112-117.

Kiyohara T, Lauzon R, Haliotis T, Roder JC. Target cell
structures, recognition sites, and the mechanism of NK
cytotoxicity. In: Lotzova E, Herberman RB, eds.

11s. Vol 1. Boca Raton,
Florida: CRC Press, Inc, 1986;107-123.

Krakowka 8. Natural killer cell activity in adult
gnotobiotic dogs. Am J Vet Res 1983;44:635-638.

142

Lanier LL, Phillips JH, Hackett J, Tutt M, Kumar V. Natural
killer cells: Definition of a cell type rather than a
function. J Immmunol 1986;137:2735-2739.

Laroche L, Kaiserlian D. Decreased natural killer-cell
activity in cutaneous T cell lymphomas. New Eng J Med
1983;308:101-102.

Lipinski M, Dokhelar M-C, Tursz T. NK cell activity in
patients with high risk for tumors and in patients with

cancer. In: Herberman RB, ed. e s n tu
effeetez cells. New York: Academic Press, 1982;1183-1187.

Lotzova E, Savary CA, Herberman RB. Impaired NK cell profile
in leukemia patients. In: Lotzova E, Herberman RB, eds.
mmunob 010 o a k'l e ce ls. Vol 2. Boca Raton,
Florida: CRC Press, Inc, 1986;29-53.

Lotzova E, Savary CA, Herberman RB. Induction of NK cell
activity against fresh human leukemia in culture with
interleukin 2. J Immunol 1987;138:2718-2727.

MacEwen EG, Hayes AA, Matus RE, Kurzman I. Evaluation of
some prognostic factors for advanced multicentric
lymphosarcoma in the dog: 147 cases (1978-1981). J Am Vet
Med Assoc 1987;190:564-568.

Madewell BR. Canine Lymphoma. Vet Clinics N Am: Sm An Pract
1985;15:709-722.

Manconi PE, Marrosu MG, Paghi L, Correale G, Zaccheo D.
Alpha-naphthyl acetate esterase activity in human
lymphocytes: distribution in lymphocyte subpopulations and
in mitogen-activated cells. Scand J Immunol 1979;9:99-104.

Mantovani A, Luini W, Peri G,‘ Vecchi A, Spreafico F. Effect
of chemotherapeutic agents on natural cell-mediated
cytotoxicity in mice. J Natl Cancer Inst 1978;61:1255-1261.

Marrack P, Kappler J. The T cell and its receptor. Sci Am
1986;254:36-45.

Marx JL. How killer cells kill their targets. Science
1986;231:1367-1369.

McGeorge MB, Russell EC, Mohanakumar T. Immunologic
evaluation of long term effects of childhood ALL
chemotherapy: analysis of la vltre NK- and K-cell activities
of peripheral blood lymphocytes. Am J Hematol 1982;12:19-27.

143

Muchmore AV. A comparison of antibody-dependent cellular
cytotoxicity and NK activity. In: Lotzova E, Herberman RB,

edS- WWW. Vol 2. Boca
Raton, Florida: CRC Press, Inc, 1986;201-213.

Nasrallah AG, Miale TD. Decreased natural killer cell
activity in children with untreated acute leukemia. Cancer
Res 1983;43:5580-5585.

Oehler JR, Herberman RB. Natural cell-mediated cytotoxicity
in rats. III. Effects of immunopharmacologic treatments on
natural reactivity and on reactivity augmented by
polyinosinic-polycytidylic acid. Int J Cancer 1978;21:221-
229.

Onions D. B- and T-cell markers on canine lymphosarcoma
cells. J Natl Cancer Inst 1977;59:1001-1006.

Onions DE. A prospective survey of familial canine
lymphosarcoma. J Natl Cancer Inst 1984;72:909-912.

Onsrud M, Thorsby E. Influence of in vivo hydrocortisone on
some human blood lymphocyte subpopulations I. Effect on
natural killer cell activity. Scand J Immunol 1981;13:573-
579. '

Parrillo JE, Fauci AS. Comparison of the effector cells in
human spontaneous cellular cytotoxicity and antibody-
dependent cellular cytotoxicity: differential sensitivity of
effector cells to Ln vivo and in vitro corticosteroids.
Scand J Immunol 1978;8:99-107.

Payne CM. Letter to the editor: parallel tubular arrays in
large granular lymphocytes. Lab Invest 1984;51:598.

Podack ER, Dennert G. Assembly of two types of tubules with
putative cytolytic function by cloned natural killer cells.
Nature 1983;302:442-445.

Priester WA, McKay FW. The occurrence of tumors in domestic
animals. Natl Cancer Inst Monograph No 54. Bethesda MD: US
Department of Health and Human Services, 1980;63,152.

Pross HF, Baines MG. Studies of human natural killer cells.
I. In vivo parameters affecting normal cytotoxic function.
Int J Cancer 1982;29:383-390.

Pross HF, Callewaert D, Rubin P. Assays for NK cell
cytotoxicity - Their values and pitfalls. In: Lotzova E,

Herberman RB, eds. Immunopiolm of negurel killer celle.
Vol 1. Boca Raton, Florida: CRC Press, Inc, 1986; 1-20.

144

Reynolds CW, Timonen T, Herberman RB. Natural killler (NK)
cell activity in the rat I. Isolation and characterization
of the effector cells. J Immunol 1981;127:282-287.

Riccardi C, Barlozzari T, Santoni A, Herberman RB, Cesarini
C. Transfer to cyclophosphamide-treated, mice of natural
killer (NK) cells and in vlye natural reactivity against
tumors. J Immunol 1981;126:1284-1289.

Ringler SS, Krakowka S. Cell surface markers of the canine
natural killer (NK) cell. Vet Immunol Immunopathol
1985a;9:1-11.

Ringler SS, Krakowka S. Effects of canine distemper virus on
natural killer cell activity in dogs. Am J Vet Res
1985b;46:1781-1786.

Robinson TWE, Heath RB. Effect of cyclophosphamide on the
production of interferon. Nature 1968;217:178-179.

Roder JC, Haliotis T, Klein M, Korec S, Jett JR, Ortaldo J,
Herberman RB, Katz P, Fauci AS. A new immunodeficiency
disorder in humans involving NK cells. Nature 1980;284:553-
555.

Roder JC, Kiessling R, Biberfeld P, Andersson B. Target-
effector interaction in the natural killer (NK) cell system
II. The isolation of NK cells and studies on the mechanism
Of killing. J Immunol 1978;121:2509-2517.

Rosenberg EB, Herberman RB, Levine PH, Halterman RH, McCoy
JL, Wunderlich JR. Lymphocyte cytotoxicity reactions to
leukemia-associated antigens in identical twins. Int J
Cancer 1972;9:648-658.

Rosenthal RC. Epidemiology of canine lymphosarcoma. Compend
Contin Educ Pract Vet 1982;4:855-861.

Rubin P, Pross HF, Roder JC. Studies of human natural killer
cells. II. Analysis at the single cell level. J Immunol
1982;128:2553-2558.

Sakela E, Tarkkanen J, Carpen O, Virtanen I. Morphological
and functional characteristics of the human NK system. In:
Serrou B, Rosenfeld C, Herberman RB, eds. N tu a l
gelle. Human cancer immunology. Vol 4. New York: Elsevier
Biomedical Press, 1982;81-93.

Savary CA, Lotzova E. Phylogeny and ontogeny of NK cells.
In: Lotzova E, Herberman RB, eds. lempeeplelegy_efi_ge§p;el
M. Vol 1. Boca Raton, Florida: CRC Press, Inc,
1986;45-61.

Se!
‘mi<

sea
nat
IIK]
lcn
198

Sha
spe
1971

She)
of <
and

Stei.
natu:
of e
patie
2259.

Stork
susce
antige

StUpp
patien
1978;1

Takasu

10pm
Cancer

Targan
NR Cytc

the g1 y,

Timanen
hUman 1.
killer

C811

fmquen ‘
the ef ,

1932;12

 

145

Seaman WE, Blackman MA, Gindhart TD, Roubinian JR, Loeb JM,
Talal N. Beta-estradiol reduces natural killer' cells in
mice. J Immunol 1978;121:2193-2198.

Seaman WE, Sleisenger M, Eriksson E, Koo GC. Depletion of
natural killer cells in mice by monoclonal antibody to
NK1.1. Reduction in host defense against malignancy without
loss of cellular or humoral immunity. J Immunol
1987;138:4539-4544.

Sharma JM, Coulson BD. Presence of natural killer cells in
specific-pathogen-free chickens. J Natl Cancer Inst
1979;63:527-531.

Shek WR, Schultz RD, Appel MJG. Natural and immune cytolysis
‘of canine distemper virus-infected target cells. Infection
and Immunity 1980;28:724-734.

Steinhauer EH, Doyle AT, Reed. J, Kadish, AS. Defective
natural cytotoxicity in patients with cancer: normal number
of effector cells but decreased recycling capacity in
patients with advanced disease. J Immunol 1982;129:2255-
2259.

Storkus WJ, Howell DN, Salter RD, Dawson JR, Cresswell P. NK
susceptibility varies inversely with target cell class I HLA
antigen expression. J Immunol 1987;138:1657-1659.

Stupp Y, Rosenkovitch E, Izak G. Natural killer activity in
patients with acute myelocytic leukemia. Israel J Med Sci
1978;14:1212-1215.

Takasugi M, Mickey MR, Terasaki PI. Reactivity of
lymphocytes from normal persons on cultured tumor cells.
Cancer Res 1973;33:2898-2902.

Targan SR, Newman W. Definition of a "trigger" stage in the
NK cytolytic reaction sequence by a monoclonal antibody to
the glycoprotein T-200. J Immunol 1983;131:1149-1153.

Timonen T, Ortaldo JR, Herberman RB. Characteristics of
human large granular lymphocytes and relationship to natural
killer and K cells. J Exp Med 1981;153:569-582.

Timonen T, Ortaldo JR, Herberman RB. Analysis by a single
cell cytotoxicity assay of natural killer (NK) cell
frequencies among human large granular lymphocytes and of
the effects of interferon on ‘their activity; J Immunol
1982;128:2514-2521.

146

Timonen T, Ranki A, Saksela E, Hayry P. Human natural cell-
mediated cytotoxicity against fetal fibroblasts III.
Morphological and functional characterization of the
effector cells. Cell Immunol 1979a;48:121-132.

Timonen T, Saksela E, Ranki A, Hayry P. Fractionation,
morphological and functional characterization of effector
cells responsible for human natural killer activity against
cell-line targets. Cell Immunol 1979b;48:133-148.

Tompkins MB, Huber K, Tompkins WAF. Natural cell-mediated
cytotoxicity in the domestic cat: Properties and specificity
of effector cells. Am J Vet Res 1983;44:1525-1529.

Trainin 2, Essex M. Immune response to tumor cells in
domestic animals. J Am Vet Med Assoc 1982;181:1125-1133.

'rrinchieri G, Perussia B. Human natural killer cells:
Biologic and pathologic aspects. Lab Invest 1984;50:489-513.

Tursz T, Dokhelar MC, Lipinski M, Amiel JL. Low natural
killer cell activity in patients with malignant lymphoma.
Cancer 1982;50:2333-2335.

Tutt MM, Kuziel WA, Hackett J, Bennett M, Tucker PW, Kumar
V. Murine natural killer cells do not express functional
transcripts of the alpha-, beta- or gamma-chain genes of the
T cell receptor. J Immunol 1986;137:2998-3001.

Ullberg M, JOndal M. Recycling and target binding capacity
of human natural killer cells. J Exp Med 1981;153:615-628.

‘Vargas-Cortes M, Hellstrom U, Perlmann P. Surface markers of
human natural killer cells as analyzed in a modified single
cell cytotoxicity assay on poly-L-lysine coated cover slips.
J Immunol Meth 1983;62:87-99.

Warner NL, Woodruff MFA, Burton RC. Inhibition of the growth
of lymphoid tumours in syngeneic athymic (nude) mice. Int J
Cancer 1977;20:146-155.

Warren RP, Tsoi MS, Henderson BH, Weiden PL, Storb R.
Demonstration of tumor-associated antigens in dogs.
Transplant Proc 1975;7:481-484.

Wheelock EF, Robinson MK. Endogenous control of the
neoplastic process. Lab Invest 1983;48:120-139.

Yamamoto S, Onuma M, Kodama H, Koyama H, Mikami T, Izawa H.
Existence of cytotoxic activity against BLV-transformed
cells in lymphocytes from :normal cattle and sheep. Vet
Immunol Immunopathol 1985;8:63-78.

147

Yamamoto S, Onuma M, Kodama H, Mikami T, Izawa H.
Suppression of natural cytotoxic activity of lymphocytes
from cattle and sheep during the progress of bovine
leukosis. Vet Microbiol 1984;9:105-111.

Yang K, Bearman RM, Pangalis GA, Zelman RJ, Rappaport' H.
.Acid phosphatase and alpha-naphthyl acetate esterase in
:neoplastic and non-neoplastic lymphocytes: a statistical
analysis. Am J Clin Pathol 1982;78:141-149.

Yang WC, 'Schultz RD, Spano JS. Isolation and
characterization of porcine natural killer (NK) cells. Vet
Immunol Immunopathol 1987;14:345-356.

Yoda Y, Abe T, Tashiro A, Hirosawa S, Kawada K, Onozawa Y,
Adachi Y, Shishido H, Nomura T. Normalized natural killer
(NK) cell activity in long-term remission of acute
leukaemia. Br J Haematol 1983;55:305-309.

Zalman LS, Brothers MA, Chiu FJ, Muller-Eberhard HJ.
Mechanism of cytotoxicity of human large granular
lymphocytes: Relationship of the cytotoxic lymphocyte
protein to the ninth component (C9) of human complement.
Proc Natl Acad Sci 1986;83:5262-5266.

Ziegler H-W, Kay N, Zarling J. Deficiency of natural killer
cell activity in patients with chronic lymphocytic leukemia.
Int J Cancer 1981;27:321-327.

Zucker-Franklin D, Grusky G, Yang J-S. Arylsulfatase in
natural killer cells: its possible role in cytotoxicity.
Proc Natl Acad Sci 1983;80:6977-6981.

VITA

Rose Esther Raskin was born March 4, 1950 in Albany,
New York. Her family moved to Newark, New Jersey where she
attended elementary and high school. She graduated in 1971
from Rutgers University (Newark, New Jersey) with a Bachelor
of Arts in zoology/pre-med. She received. her' Doctor’ of
Veterinary Medicine in 1976 from Purdue University (West
Lafayette, Indiana). Following graduation she worked as an
associate small animal practitioner in Montclair, New Jersey
at the Cameron Animal Hospital until 1982. Dr. Raskin
pursued her interests in hematology and oncology by
accepting a 3 year combined anatomic and clinical pathology
resident/instructor position at Michigan State University.
She remained in the Department of Pathology as a senior
clinical pathology resident/instructor during which time she
was awarded a grant (Companion Animal Fund) from the College
of Veterinary Medicine to pursue doctoral research. In 1986,
she became a Diplomate of the American College of Veterinary
Pathologists (cdinical pathology). In 1987, following
completion of her doctoral research, she accepted a position
as Assistant Professor in the Department of Physiological

Sciences at the University of Florida (Gainesville).

148

149
Dr. Raskin is the author of 5 journal publications. Her
research interests involve the clinical, immunologic and
pathologic aspects of hematopoietic neoplasia including bone
marrow hi‘stopathology and cytochemistry. Honors received

include Phi Beta Kappa, Rutgers Scholar and Phi Zeta. She is

listed in ILL—0's 022 In 1226mm Meme (1987).