 

         

a E. ¢ I . I7 ,I-l :1
,‘Is' o'l 'IZL IVVII‘O. 0

  
     

      
   
 

(nth-I": v
I III'IE,,1'4
If"; -I' 11,54,151"? I; "

 
  
  
   
  

‘ 71:7 1:, L, LETIII ﬂ,
'.. .f 1.‘ II- ” .I A v
..;,‘;=,I.15: 1%,, I L. ' d

.i".I.‘,’~'EII§ ‘1‘,

  
   
 
 
   
  
 
 
 

  

  
 
   
 
  
  

        
    
 

   
 

 

WI. ,,¢E,? ,3, 7' ‘. MI; .IIEIIE q, E,
I l I II IV R IIII i'IlIIIJ IEIII ’23! II=III 3 ‘I-‘I'E’IIIII "5!’ I: 'I 1 III “HIMI
I ,hlII 1,“, x E “III? IiEI' [213;] “a? ”5,3, in. ,,, ,;:E;V‘:, 2,. v .

   

   

‘ ;:3,I;‘ " ,‘ IIIIIIIJ 313,, ”$1,, ”h
.‘n.', ‘. :—:;‘,:,II’I,:I,§, 1,,1‘,‘ :1I'EII1, ’ I145“ In,
' ‘§:;IIIIE,.III'1IEI ! II IRW- L33,

IIIIIIF III} '

II,,",‘{II IR IAIIEIIIII 'IJIIIII. I. . :~
I" III-III

, ’,,lII

  

IL: ‘ 1,1},
IIIIIIII «H IM‘. ‘l' a?! ‘5“
,,, ”III: I], SIM?
III“ :5‘ ’LSIEII‘IIIYI‘ WW:
‘ ﬁf'I‘Qli‘I' '53”
“1*“ 35‘” QQI‘ v
:J'EI 9:1.
£11.? LIIIEHEI
'1,
tag?“ N'IO ’- '
,IfIl‘Il‘éI‘IIJt I
g- 'iitci‘I‘L $4431

 

   

       
   
 

 
      

  

m3}, IIII'I; F,

  
   

 

ii"!

  
 
  
  

   

.' :II _,
IIHIgﬁI III':,_'_-:
. 4,,

     

  

”A, 'IIIIII #3,.
,1, L

  
 
 

  
  

    

       
   

  
  
  

 

    
    
 
 
   
   
  

 
    

I‘LL ';" “I, by W ML“ , '
f"; 7 I Ea: . III III?
2, ;; MIL“ 3:4,. ~ . 23'
I a 13‘? ,5: ‘5', "A: IIIEE' - I I
'3 II' III!“ IISII” 7
“III I 1 ~ 57'. z 2 ”I“ . «t
L , M -
\h: n I
:1 3’3” LEI

   
   
 
  
  
  

  

. -....."

m
.1317”
f“
:u
...

. . .
-— L'- ’ ”—A
L-n l“ 4 ..
...-... ' I .-
J-.:. J.- ' .- u‘
...-v ' ‘
_ .. ,..
- I-A A I g V -|
5.“..nm- JJ' .- - '
“..., ....--“ . . 1
"" ' .a ‘ W ‘- u
. .-V - . .
TWA-Tu . ~ - u ' -
... n .
ﬂ .... m“ __ {4-.. . ...
_. .1 I ‘5.’ In
..M. . . -- .. .
. .. ._. . . ...V v ,,
7:.”‘l ‘14" -- v- ;:'2‘. -’ ‘ ¢ .
r . - -
"‘ 4'.‘ -. H
" ‘.o-- I 1". ,
4 __ ‘ own-7.“
‘. m .-
—o. - ~— . .
. -

  
  
  
 

|v1,- 12,31,
ii. IEIII I1
4,15,

  
        
  
    
 
 

   

  
  
  
    
  
 
 
 
  
 

    

  
 

   
   
  
 
   

 
      

. -
'_,_7:1: . c
_ .m ’ n .. .. ~
. .‘f—‘p 1" -.. «...—4 ~
,
...
.--... v

   

     

   

  
  
   
  
 

  
 

    

 
         
 
 

 
 
 
  

  
  
    

 

  
 
 
 

     

,hflI I.
IIIII 1| “(I < 3' I "
IIIII,IIII%I,J,IIII,I' EL ,1 \II!‘ ~I'Iv2- E I LII'IIhIi'
2.“,3'1‘1'2’1'“; 'i.‘ ,7 ; ,' , . '1‘ H" In 5
I, IIII‘IIII‘IgIIIII‘IIIV' - ' ' Io, . 3,4,. ,. ,I'I L, .f‘ . II”, ,3, 1 [,4
,;,2,‘.“.'::~:I,,sén13:55" ' '1' ' ,‘I'V' ,5; ‘LE «.13 ,‘LI’L 54.1,, 1,, “III E," L5,:
.7 I .119th . : ;' . I , ' - 1:! v 11:: \ I.’~ 4
I: I? ‘1‘ "l" 4‘, f 5,, 1‘ , ,» {1,3,51,51,95 p3 ‘ ‘
., ' .13, 'Iui I”, i: ‘,'ftl‘-';I:II", ,9, I,E::-' {H.Ivn’étgﬁ IIILIIK ,,a
' ‘IIIIZ‘W‘I I I‘ 'U‘I'! 7‘ ,z-I II
IIEEEEIE‘II'I " . fr). *‘IE‘JI ’L L23 ‘ »
Iirﬁz'réie, LIL: ,. I-‘IL' IL, "MW
.531“? a 7‘ AL; “I?“ =‘7" =2.
II'IIII. ,,,,k,:,; , . 3521» ‘IA, 4! 3,3,5“ .- “A ,
f ' !‘ . St; E - t " . I} ' l. 7", 1
. EESIIIfir‘f’q ...,1‘1,,‘ 7 \ iIIP'il‘éLtﬁ, :
J. I 57? '.'.‘I‘ ‘ 1::
~ 9, ._«:
x

\‘o

_.o—

     
   
  

 
    
 
   
    

     
 

      
   
  

 

         

         

 

    
      
 
  

     
  
  

     
     
 

 

     
    
 
 
 

 
 
   

 

    
 

      
 

_.=
--..a—vv
_"__ . ~—-—~

. :5 «I7
*‘IfIIIII, Ix 7'71;
I; '9: 13‘I‘Ts‘;L'LI‘=L 2.3-»
, : , no
IIIIE:Z;:‘I::‘1 ,,'.,IIII,I,‘I;i,III‘L' I IIIII‘IJ-ﬁIIIiJn'I'II' .III,"III,IIEI§I;.[ ,.' 1.0: I: , . I , I.
32,; EE, it] , 1,4 L Lﬂageﬂi' E ,1».,,I,;¢'L,,IE”., ,_ , g, fle'IIEt,” ‘3, Ly, 59,1, :
I'LL, Ii" , L3,.I. E15,, almELEﬁ «I; Egg: ,,~E;J,;,§,, 19.32:},5: 4.717;, ,IIIII, W, E III L1,,
7: ..IIIIIILII I II ‘IEIIuII III-II I'IIIEIIII}. I 11‘ .g. ICE. IIII‘III‘IIII'II'IIHI" . :‘v NH, {III}? ,zpﬂ‘JI III IEEIIII _.
I.,;;1§.!’=,,':Iu,;II 3,5535ELIE;E,,E,;=,Eng,l'.;,EI,:;"~,e§,;2;!gz}gizE§;,§,§zEt'I' 'Iil‘mhIII.;.* ’,, ,‘ ,i-r’fiIIt,Jt-r J’ﬁﬁ'u‘"
III‘HI ,EIIEI‘LLHLuLI,LEis'pE'L‘r {,r I":.LIIIIquEE'i‘zLLELELI;’ ' 4,, E23, ’ ‘ ..M ~ I
,E , Leg-,EEL,,ILu,,,w;,t-,E E1, I,,I,,,,,,, II,£~,,- 1,“: LEE; ,-- L Em *L ,5. L, "If: XL
,'~ E,‘.I{,'I5I,:II “III,“ ”EL, ,III .,I IIII'HIIEIIHIIII1',,,I’I',H: . :1, I .. I' . ' gm”. ":34 v .r
'IIL!.I~:’I,‘",., ILLNEII 'I 431-111 '- ‘ ' "' 'LIILIIII‘I. ‘, =EI‘I L“ E I: ~.
- -1 ,EII', w,"I'i‘-IIII ‘ ,:.. WWI-5" “ ,3?" ‘4, M. . L51
"IIII‘I.I'LII!II1:,IILIII” I I ‘ IILIEEI IIII“ 5,2,
I , , :

...

 

W
'ﬁ.

—‘_ _._’ ‘ ..V
_ ..3.

.lI
III, .Ig: MI]:

I I
I-:
v

v?! I
I,(I. IEIIII i'I $333,.” (7&5: Kt,

 

_-."—v__ .

 

THE3EB

IlllllllllllllllllllllllllllllJlllllllll’llll’lllllllllll

301718 8495

This is to certify that the

thesis entitled

ALTERNATIVE MURINE MODELS FOR DIETARY VOMITOXIN-
INDUCED IMMUNE DYSREGULATION AND IgA NEPHROPATHY

presented by

Craig A. Banotai

has been accepted towards fulfillment
of the requirements for

M. S. degree in Food $91 gnge

JW Patti/4T

Major professor

 

Date 5//Y/?5

 

0-7 639 MS U is an Afﬁrmative Action/Equal Opportunity Institution

 

 

' vg-v—r v v—vﬁﬁ—vv—

AA v”

 

LIBRARY

MIchlgan State
University

 

 

 

PLACE IN RETURN BOX
to remove this checkout from your record.
TO AVOID FINES return on or before date due.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1M com-.mu

ALTERNATIVE MURINE MODELS FOR DIETARY VOMITOXIN-INDUCED
IMMUNE DYSREGULATION AND IgA NEPHROPATHY

By

Craig Allen Banotai

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

MASTER OF SCIENCE

Department of Food Science and Human Nutrition

1998

ABSTRACT

Alternative Murine Models for Dietary Vomitoxin-Induced
Immune Dysregulation and IgA Nephropathy

By

Craig Allen Banotai

Dietary exposure to the nichothecene VT results in reduced body weight gain,
elevated production of serum IgA, terminal diﬂ‘erentiation of Peyer’s patch B cells to IgA
secreting plasma cells, and elevated mesangial IgA accumulation. These eﬁ‘ects closely
mirror human IgA nephropathy (IgAN). The purpose of this research was to l) assess
whether intermittent consumption of dietary VT, as might be seen with human exposures,
has similar effects as continuous consumption, and 2) assess whether consumption of diet
containing VT will similarly affect mice with aberrant immune systems. To determine
the eﬁ‘ect an intermittent VT exposure would have on the mouse model, female B6C3F1
, mice were fed either a diet containing 20 ppm VT continuously or intermittently (every
other week) for 13 wks. The eﬁ‘ect of VT exposure on mice with aberrant immune
systems was determined using autoimmrme—prone MRL/lpr, NZBW/F ,, and BXSB mice
fed a diet containing either 5 or 10 ppm VT for 9, 11, and 14 wks, respectively. The eﬂ'ect
these diets had on the immunopathologic indicators of IgAN were assessed and compared
with each other as well to mice fed a clean diet. The results suggest that intermittent
exposure to VT had reduced toxic effects in B6C3F1 mice as compared to the continuous
model and that mice with aberrant immune systems were diﬁ‘erentially aﬁ‘ected by

continuous VT exposure.

In Memory

of my Mother
Kathaleen Orgeck Banotai

1946-1978

iii

 

 

 

It

I would like to thank ﬁrst and foremost my major professor and mentor, Dr.
James Pestka, for his guidance and support during my graduate work at Michigan State
University. I would especially like to express my gratitude for his patience throughout
this endeavor without which would have made this project nearly impossible to complete.

I would also like to thank the other members of my committee, Drs. John Linz
and Kathy Brooks, for their time and eﬁ‘ort during the planning stages of this project and
for their continued patience and support throughout. I would like to further thank Dr.
Kathy Brooks for being my undergraduate mentor and for taking a chance on a
inexperienced, young sophomore.

Special thanks go out to Drs. Roscoe Warner, Juan Azcona-Olivera, and Dana
Greene for their extensive help and expertise which helped my project run smoother than
could have been imagined.

A special thanks also goes out to Chris Sypien and Amy Dwyer, two of the closest
ﬁiends a person could ever expect to have, who not only assisted me with my project, but
were there to make my life both in and out of the lab a memorable one.

A large debt of gratitude is owed to my grandparents, Louis and Serah Banotai,
without whom, through their unselﬁsh support and faith in me, none of this would have

been possible.

iv

Finally, thanks is not enough to express my overwhelming gratitude to my wife,
Charlene, who has been with me through the good times and the bad, even though our
time together was very scarce. It is her devotion and unending support that has made this

journey worth taking.

TABLE OF CONTENTS

List of Tables .......................................................................................................... viii
List of Figures .......................................................................................................... x
List of Abbreviations .............................................................................................. xiii
Chapter 1. Literature Review .............................................................................. l
Mycotoxins .................................................................................................. 1
Trichothecenes ............................................................................................. 2
Vomitoxin .................................................................................................... 4
Immunoglobulin A ....................................................................................... 1 1
IgA Nephropathy ......................................................................................... 14
Systemic Lupus Erythematosus ................................................................... 15
Thesis Rationale ........................................................................................... 18
Chapter 2. The Eﬂ'ect of Intermittent Vomitoxin Exposure on Body Weight
Gain and the Scrum Immunoglobulin Proﬁle of the B6C3F, Mouse ............... 19
Abstract ........................................................................................................ 19
Introduction .................................................................................................. 21
Materials and Methods ................................................................................. 23
Experimental Design. ........................................................................ 23
ELISA .............................................................................................. 24
Urinalysis ......................................................................................... 26
Quantitation of Mesangial IgA, IgG, and C3 ................................... 27
Safety ............................................................................................... 27

vi

Animal Care ..................................................................................... 28

Statistical Analysis ........................................................................... 28
Results .......................................................................................................... 29
Discussion .................................................................................................... 44

Chapter 3. Differential Effects of Subchronic Dietary Vomitoxin Exposure

on Murine Models for Systemic Lupus Erythematosus ..................................... 48
Abstract ........................................................................................................ 48
Introduction .................................................................................................. 50
Materials and Methods ................................................................................. 53

Experimental Design ........................................................................ 53

Lymphocyte Culture ........................................................................ 55

lg Quantitation ................................................................................. 56

Hematuria ......................................................................................... 58

Quantitation of Mesangial IgA, IgG, and C3 ................................... 58

Statistical Analysis ........................................................................... 58

Results .......................................................................................................... 60
Discussion .................................................................................................... 95
Chapter 4. Summary ........................................................................................... 103
List of References ................................................................................................... 107

vii

LIST OF TABLES

Table 2.1. Effect of Dietary Vomitoxin (VT) on the Production of Serum
Autoantibodies ......................................................................................................... 39

Table 2.2. Eﬁ‘ect of Immediate and Long Term Dietary Vomitoxin (VT)
Exposure on Mesangial Deposition of IgA, IgG, and C3 ......................................... 43

Table 3.1. Effect of Dietary Vomitoxin (VT) on the Production of Serum
Autoantibodies in Female NZBW/Fl Mice .............................................................. 74

Table 3.2. Effect of Dietary Vomitoxin (VT) on the Production of Serum
Autoantibodies in Female MRL/lpr Mice ................................................................ 75

Table 3.3. Effect of Dietary Vomitoxin (VT) on the Production of Sermn
Autoantibodies in Male BXSB Mice ....................................................................... 76

Table 3.4. Effect of Dietary Vomitoxin (VT) on lg Production in Peyer’s
Patch Cultures of Female NZBW/Fl Mice .............................................................. 78

Table 3.5. Effect of Dietary Vomitoxin (VT) on Ig Production in Splenic
Cultures of Female NZBW/Fl Mice ........................................................................ 79

Table 3.6. Effect of Dietary Vomitoxin (VT) on Ig Production in Peyer’s
Patch Cultures of Female MRL/lpr Mice ................................................................ 81

Table 3.7. Effect of Dietary Vomitoxin (VT) on Ig Production in Splenic
Cultures of Female MRL/lpr Mice .......................................................................... 82

Table 3.8. Eﬂ‘ect of Dietary Vomitoxin (VT) on Ig Production in Peyer’s
Patch Cultures of Male BXSB Mice ........................................................................ 83

Table 3.9. Effect of Dietary Vomitoxin (VT) on lg Production in Splenic
Cultures of Male BXSB Mice .................................................................................. 84

Table 3.10. Eﬂ‘ect of Dietary Vomitoxin (VT) on Hematuria of
Autoimmune-Prone Mice ........................................................................................ 86

viii

Table 3.11. Effect of Subchronic Dietary Vomitoxin (VT) on Mesangial
Deposition of Ig’s in Female NZBW/Fl Mice ......................................................... 88

Table 3.12. Eﬁ‘ect of Subchronic Dietary Vomitoxin (VT) on Mesangial
Deposition of Ig’s in Female NRL/lpr Mice ........................................................... 89

Table 3.13. Effect of Subchronic Dietary Vomitoxin (VT) on Mesangial
Deposition of Ig’s in Male BXSB Mice .................................................................. 90

ix

LIST OF FIGURES

Figure 1.1. Chemical structure of Vomitoxin ......................................................... 5
Figure 1.2. Structure of human monomeric IgA (from Roitt et al., 1989) .............. 12
Figure 1.3. Structure of human secretory IgA (sIgA) (from Roitt et al., 1989) ...... 12

Figure 2.1. Eﬂ'ect of intermittent and continuous dietary vomitoxin exposure

on body weights of B6C3F1 mice compared to control. Data are means i SE

(8-9 mice/group). The intermittent group was fed 20 ppm vomitoxin on odd

weeks and a clean diet on even weeks. ‘Control group signiﬁcantly diﬂ‘erent

than both intermittent and continuous groups at p s 0.05. l’lndicates group is
signiﬁcantly different than the continuous group at p s 0.05 .................................. 31

Figure 2.2. Effect of intermittent and continuous dietary vomitoxin exposure

on serum IgA levels of B6C3F1 mice compared to control. Data are

means :1: SE (8—9 mice/ group). 'Indicates signiﬁcantly different than control
group at p s 0.05. ”Indicates signiﬁcantly different than other treatment group

at p s 0.05 ................................................................................................................ 34

Figure 2.3. Effect of intermittent and continuous dietary vomitoxin exposure

on serum IgG levels of B6C3F1 mice compared to control. Data are

means :1: SE (8-9 mice/group). 'Indicates signiﬁcantly diﬁ‘erent than 00an

group at p s 0.05. l’Indicates signiﬁcantly diﬂ‘erent than other treatment group

at p s 0.05 ................................................................................................................ 36

Figure 2.4. Eﬁ‘ect of intermittent and continuous dietary vomitoxin exposm'e

on serum IgM levels of B6C3F1 mice compared to control. Data are

means :1: SE (8-9 mice/group). ‘Indicates signiﬁcantly different than control

group at p s 0. 05 ...................................................................................................... 38

Km 2.5. Effect of intermittent and continuous dietary vomitoxin exposure

on hematuria of B6C3F1 mice compared to control. Data are means :1: SE

(8-9 mice/group). ‘Indicates signiﬁcantly different than control group at

p s 0.05. l‘Indicates signiﬁcantly different than other treatment group at

p s 0.05 .................................................................................................................... 41

X

Figure 3.1. Eﬁ‘ect of subchronic vomitoxin exposure on weight gain by

NZBW/Fl female mice. Data are means t SE (3-7 mice/group). Data was

analyzed by Student-Newman-Keuls method following one-way analysis of
variance (AN OVA). ‘Control group signiﬁcantly different than both treatment
groups at p s 0.05. bIndicates group is signiﬁcantly different than 10 ppm

treatment group at p s 0.05 ...................................................................................... 62

Figure 3.2. Effect of subchronic vomitoxin exposure on weight gain by

MRL/lpr female mice. Data are means :t SE (6-7 mice/group). Data was

analyzed by Student-Newman-Keuls and Drmn’s methods following one-way
analysis of variance (AN OVA). ‘Control group signiﬁcantly diﬁ‘erent than both
treatment groups at p s 0.05. bIndicates group is signiﬁcantly different than the

10 ppm treatment group at p s 0.05 ......................................................................... 64

Figure 3.3. Eﬁ‘ect of subchronic vomitoxin exposure on weight gain by

BXSB male mice. Data are means :1: SE (3-7 mice/group). Data was analyzed

by Student-Newman-Keuls method following one-way analysis of variance

(AN OVA). ‘Indicates group is signiﬁcantly diﬁ‘erent than the 10 ppm treatment
group at p s 0.05. l’Indicates group is signiﬁcantly different than the other

treatment group at p s 0.05 ...................................................................................... 66

Figme 3.4. Effect of subchronic dietary vomitoxin exposm'e on serum IgG, IgA,

and IgM levels of NZBW/Fl female mice compared to control. Data are

means 3: SE (3-7 mice/group). Data was analyzed by Dunn’s method following
one-way analysis of variance (AN OVA). ‘Indicates signiﬁcantly different than
other treatment group at p s 0.05 ............................................................................. 69

Figure 3.5. Effect of subchronic dietary vomitoxin exposure on serum IgG, IgA,
and IgM levels of MRL/lpr female mice compared to control. Data are

means :1: SE (7 mice/ group). Data was analyzed following one-way analysis of
variance (AN OVA) .................................................................................................. 71

Figure 3.6. Eﬁ‘ect of subchronic dietary vomitoxin exposure on serrun IgG, IgA,

and IgM levels of BXSB male mice compared to control. Data are means :1: SE

(3-7 mice/group). Data was analyzed by Student-Newman—Keuls method

following one-way analysis of variance (AN OVA). ‘Indicates signiﬁcantly

diﬂemnt than control group at p s 0.05. bIndicates signiﬁcantly different than

other treatment group at p s 0.05 ............................................................................. 73

Figure 3.7. Mesangial IgA deposition after dietary VT exposure. Kidney

sections were prepared at wk 14 ﬁ'om male control and VT treated BXSB mice

and stained with FITC-labeled anti-mouse IgA. (a) control group (b) 10 ppm
treatment group ........................................................................................................ 92

Figure 3.8. Mesangial C3 deposition after dietary VT exposure. Kidney
sections were prepared at wk 14 from male control and VT treated BXSB mice
and stained with FITC-labeled anti-mouse C3. (a) control group (1)) treatment

group ........................................................................................... 94

Con A
ELISA
GALT
IC

13

IgA
IgAN
IgA-1C
IgG
18E
IgM
IL
LPS

PP

sIgA
SLB

SP

VT

LIST OF ABBREVIATIONS

concavalin A

enzyme-linked immunosorbent assay
gut-associated lymphoid tissue
immune complex

immunogloban

immunogloban A

IgA nephrOpathy

immunoglobulin A-immune complexes
immunoglobulin G

immunoglobulin E

immunoglobulin M

interleukin

lipopolysaccharide

Peyer’s patch

red blood cells

secretory IgA

systemic lupus erythematosus

spleen

vomitoxin

xiii

 

 

 

 

Chapter 1. Literature Review

 

Mariam

Mycotoxins are a structurally diverse group of chemicals that are produced as
secondary metabolites by a variety of fungal genera (Pestka and Bondy, 1994). These
compounds are capable of causing a wide range of biological and toxicologic effects
(Pestka and Bondy, 1990). These secondary metabolites are often strain speciﬁc and are
not required for the natural survival of the producing fungus, but may confer selective
advantages for organism growth (Pestka and Casale, 1990). Mycotoxins are produced
during conditions favoring fungal colonization. Environmental crop stress due to
excessive rainfall, drought, and cold weather as well as improper storage conditions can
lead to the presence of mycotoxins in grain and cereal crops (Abouzied et al., 1991;
Tanaka et al., 1988). Approximately 25% of the food crops worldwide are contaminated
by mycotoxins annually (Rotter et al., 1996). Mycotoxins can be formd in many.
agricultural products such as corn, wheat, and nuts and this contamination can be carried
over into animal feed (Pestka and Bondy, 1994). This can cause severe economic
hardship for farmers and livestock producers and can pose a serious health threat to
humans consuming contaminated food products (Pestka, 1988). Though mycotoxins are
produced by various fungal genera, the mycotoxins produced by the Aspergillus—

Penicillium and F usarium groups are the most thoroughly studied since they have the

 

 

 

 

 

 

 

2

greatest potential impact on the food supply (Pestka and Bondy, 1994). These fungi
produce the major mycotoxins found in foods and animal feeds in the United States and
worldwide (Pestka and Casale, 1990). The most notorious mycotoxins produced by the
Aspergillus-Penicillium group are the aﬂatoxins and the ochratoxins, which are both
carcinogenic (Pestka and Bondy, 1994). Because aﬂatoxin is such a potent
hepatocarcinogen, it is currently regulated at low parts per billion in most industrialized
nations (Pestka and Bondy, 1990). Three other major mycotoxins, the trichothecenes,
furnonisins, and zearalenones are produced by the Fusm'ium group. These are not as
highly regulated, but have been shown to impair growth and cause immune dysregulation

(Pestka and Bondy, 1994).

Trichothecene!

Second only to the highly carcinogenic aﬂatoxin, trichothecene mycotoxins have
been implicated as a major cause in both animal and human toxicoses (Ueno, 1983).
Trichothecene mycotoxins are a group of over 148 structurally related compounds that
have been researched thoroughly because of their toxicity and the frequency in which
they are found in food staples (Rotter et al., 1996). Of these trichothecenes, vomitoxin
(deoxynivalenol), nivalenol, T-2 toxin, and diacetoxyscirpenol are most commonly
detected during human and animal exposures (Pestka and Bondy, 1994). These
trichothecenes are elaborated by at least six strains of fungi in the genera Fusarium,
Myothecium, Trichothecium, Stachybotrys, Cephalosporiwn, and Verticimonosporium

(U eno, 1983). The trichothecene mycotoxins are esters of sesquiterpenoid alcohols

3

containing the trichothecane tricyclic ring system (Pestka and Casale, 1990). All
trichothecene mycotoxins contain a double bond between carbons 9 and 10 and an
epoxide at the carbon 12, 13 position of the molecule, giving them the designation l2, l3-
epoxytrichothecenes (T amm, 1977).

Trichothecenes are among the most potent low molecular weight inhibitors of
protein and DNA synthesis known (McLaughlin et al., 1977). Their toxic mode of action
is attributed to selective binding of trichothecenes to a single site on the 60S subunit of
eukaryotic ribosomes which restricts further protein synthesis by inhibiting initiation and
elongation and by blocking peptidyl transferase activity (Pestka and Casale, 1990). This
inhibition of protein synthesis ultimately leads to the restriction of DNA synthesis.

Acute exposure to trichothecenes causes severe damage to actively dividing cells
in such tissue as bone marrow, lymph nodes, spleen, thymus, and intestinal mucosa
(Pestka and Bondy, 1990). Acute trichothecene poisoning has been characterized as a
. multi system shocklike syndrome and is presented by dermal irritation, nausea, emesis,
diarrhea, hemorrhage, and hematologic lesions such as leukopenia and anemia (Pestka,
1997). There is also evidence that these compormds can cause profound effects on host
resistance, cellular function, and humoral immunity (Pestka and Bondy, 1990).

Mycotoxicoses in livestock caused by chronic trichothecene exposure are a
signiﬁcant problem from an agricultural standpoint as well as of particular importance to
human health. Chronically exposed livestock may exhibit feed refusal and subsequent
decreased body weight gain, vomiting, diarrhea, reduced milk production in dairy cattle,

and hemorrhaging in the liver, stomach, heart, lungs, bladder, kidney, and intestines

4

(Bamburg, 1983). Immunotoxicity is often observed in animals ingesting trichothecene
contaminated grains. Repeated exposure of animals to trichothecenes results in increased
susceptibility to bacterial, fungal, and viral infections (Pestka and Casale, 1990).
Increased infections in food animals could lead to major public health concerns with
increased animal-to-human transmission of pathogens such as Salmonella and Listeria
(Pestka and Bondy, 1990). Other immunotoxic effects seen by long term exposru'e to
trichothecenes include a decreased humoral response to T-dependent antigens, increased
response to T-independent antigens, impaired macrophage function, increased skin graft

rejection times, and depressed acute phase response (Pestka and Casale, 1990).

I! 'I .
Deoxynivalenol or vomitoxin (VT) as it is more commonly known because of its
emetic eﬁect in swine (V esonder et al., 1976), is a trichothecene mycotoxin produced
primarily by the species F usarium graminearum. It was ﬁrst isolated (Morooka et al.,
1972) ﬁom Fusarium contaminated barley and (V esonder et al., 1973) from Fusarium-
infected corn from northwestern Ohio. VT is a toxic, secondary metabolite that has a
molecular weight of 296 (Figure 1.1) and is frequently found in cereal grains and grain-
based food products (Abouzied et al., 1991; Tanaka et al., 1988). It is produced druing
conditions that favor fungal colonization and varies yearly based on environmental
conditions. Environmental crop stress due to cold and wet weather as well as improper
storage conditions can lead to VT’s presence in these cereal grains (Abouzied et al., 1991;

Tanaka et al., 1988). Bad weather can also force delays in the harvest of crops which can

l6

    

onll'..'l'|':

 

 

 

"2

OIIIIIIII
I

O

I

CISHZOOG
MW. 296.13
Species:

E grmninemm
F. cubnorum

FIGURE 1.1. Chemical structure of vomitoxin

6

perpetuate mold growth and toxin production (V esonder et al., 1978). There is much
concern associated with the presence of VT in crops used for food. Even though VT is
not as toxic as other trichothecenes, it is one of the most common contaminants of crops
worldwide (Jelinek et al., 1989; Scott, 1989). The problem of VT contamination in such
food staples as corn, wheat, barley, rice, and cats is that it is such a stable compound and
does not degrade at high temperatures (Scott, 1991). The problem is further exacerbated
by VT’s resistance to inactivation during milling and processing which allows it to
contaminate grain-based food products worldwide.

The toxic eﬁects exhibited by the trichothecenes including VT are due to several
structural features of the compound. Two of the features that all trichothecenes share are
a double bond at the C9-C10 position and a 12, l3-epoxy ring (T amm, 1977). As
mentioned previously, VT is considered to be less toxic than other trichothecenes
including T-2 toxin, fusarenon-X, and nivalenol (Rotter et al., 1993; Ueno, 1985). The
difference in the extent of toxicity presented by each trichothecene is based on
substitution of hydroxl or other groups around the trichothecene nucleus and by the
structure and position of side chains on the molecule (Betina, 1989). The diﬁ‘erence of
only a hydrogen in place of a hydroxyl group at the C-4 position found in nivalenol
decreases VT toxicity by 10 fold (Ueno, 1985).

The LD50 for VT in B6C3F, female mice is 78 mg/kg body weight per as (p0)
and 49 mg/kg body weight administered intraperitoneally (F orsell et al., 1987). VT
toxicity is attributed to it’s binding to a speciﬁc site on the 608 subunit of eukaryotic

ribosomes and interfering with peptidyl transferase activity. This process inhibits protein

7
synthesis by preventing the normal initiation, elongation, and termination of protein
formation (Pestka and Casale, 1990). VT has also been implicated in suppressing DNA
synthesis (Ueno, 1983). This effect does not appear to be a direct cause of VT exposure
to the cells and is probably caused by protein synthesis inhibition which leads to DNA
damage. Evidence suggests that DNA damage in cells exposed to VT is not caused by
unscheduled DNA synthesis and VT is still toxic in cells that are defective in DNA repair
(Bradlaw et al., 1985; Robbana-Bamat et al., 1988).

VT has been implicated in both animal and human toxicoses. Chronic dietary VT
has been shown to cause a variety of toxic eﬂ'ects in experimental animals including
vomiting, feed refusal, reduced feed efﬁciency, and reduced body weight gain (F orsell et
al., 1986; Rotter et al., 1996). Subchronic dietary VT ingestion has also been shown
experimentally to impair host resistance, humoral immunity, and cell-mediated immune
function (Pestka and Bondy, 1990; Pestka et al., 1987). Immunosuppressive effects in
host resistance have been reported by Tryphonas et al. (1986) in which oral exposure to
VT caused a dose-related, reduced, time-to-death interval following exposure to Listeria.
Analogous experiments by Pestka et al. (1987) resulted in signiﬁcantly increased splenic
Listeria counts, suggesting that both macrophage and cell-mediated immune ﬁrnction
might have been altered. This diminished resistance to Listeria may be a result of VT-
induced feed refusal rather than a direct aﬁ‘ect of VT exposure (Pestka and Bondy, 1990).
Immunosuppression of cell-mediated immune function caused by VT exposure has also
been demonstrated by diminished lymphocyte stimulation by mitogens (T ryphonas et al.,

1986; Robbana-Bamata et al., 1988).

8

Besides the immunosuppressive eﬁ‘ects, dietary VT ingestion has been shown to
be simultaneously immunostimulatory in experimental animal models (Pestka and
Bondy, 1994). The most notable stimulatory effect of extended VT exposrue is elevated
production of immunoglobulin A (IgA) (F orsell et al., 1986), which is symptomatic of an
aberrant mucosal immune response. The net result of this elevated production is that IgA
becomes the major Ig isotype in the serum. Concmrent with this is an elevation of IgA
immune complexes (IgA-1C), increased polymeric to monomeric IgA ratio in serum,
decreased total serum IgG levels, an increase in the size and ﬁ'equency of Peyer’s patches,

an increased spontaneous and mitogen-stimulated IgA production by isolated

' lymphocytes, mesangial I gA accumulation and persistent glomerulonephritis

accompanied by hematuria in experimental mice (Dong and Pestka, 1993; Dong et al.,
1991). These VT-induced eﬁ‘ects closely parallel the human disease lmown as IgA
nephropathy, which is the most common form of human glomerulonephritis worldwide
(D’Amico, 1987). The observation that dietary VT induces murine IgA nephropathy is
important because foodbome VT might be an etiological factor in the human form of the
disease.

Humoral immunity can also be both stimulated or repressed by VT exposure.
Exposure to VT at a 10 ppm concentration inthe diet or at a 0.75 mg/kg gavaging dose
can impair the murine splenic plaque forming cell (PFC) response to sheep red blood
cells (SRBC) (Pestka et al., 1987; Robbana-Barnat et al., 1988; Tryphonas et al., 1984).
The isotype-speciﬁc effects of mice fed VT that were evaluated as mentioned before by

Forsell et al. (1986) not only detected a dramatic increase in total serum IgA, but also a

9

concurrent decrease in total serum IgM and IgG. The lowest concentration of VT to
induce this effect was 2 ppm, but the maximal effect was seen at concentrations in the 10-
25 ppm range (Dong et al., 1991; Pestka et al., 1989; 1990b). There also appears to be a
progressive increase in serum IgE aﬁer withdrawal ﬁom dietary VT exposrue (Pestka and
Dong, 1994).

Mice fed VT exhibited both an increase in the number and size of Peyer’s patches
removed ﬁ'om these mice. In addition, Peyer’s patch lymphocyte and to a lesser extent
splenic lymphocyte cultures prepared from VT fed mice produced signiﬁcantly larger
amounts of IgA than control cultures, both with and without mitogen stimulation and
showed an actual increase in the number of IgA-secreting cells in both the Peyer’s
patches and spleen (Pestka et al., 1989; 1990a; 1990c; Bondy and Pestka, 1991). These
results indicate that premature diﬁerentiation of IgA-secreting cells occru's at the Peyer’s
patch level and that this differentiation is later reﬂected in the systemic compartment
(Pestka and Bondy, 1994). This premature differentiation is possibly caused by T-cell
dysregulation which is suggested by an increase in T helper cells (CD4+) and the
CD4+:CD8“ T-cell ratio in the Peyer’s patches and spleens of VT-fed mice (Pestka et al.,
1 990a).

Terminal differentiation of B cells to IgA-secreting cells in the Peyer’s patches
and spleen along with increased CD4+ T helper cell population and CD4+:CD8” cell ratio
after VT exposure suggests that VT alters B cell diﬂerentiation by aﬁ‘ecting T cells. This
idea was further demonstrated when Warner et al. (1994) could not get VT to enhance Ig

secretion in puriﬁed B-cell cultures. It was also demonstrated that addition of VT-treated

10

mice T cells to a culture of B cells from untreated mice increased IgA production over
that of control B cells co-cultmed with control T-cells (Bondy and Pestka, 1991). It was
also shown that there is increased IgA secretion by untreated B-cells when they are co-
cultured with CD4“ cells pulsed in vitro with VT (Warner et al., 1994). Lastly, increased
messenger RNAs (mRN As) and secreted proteins for interleukin (IL) 2, IL-4, IL-5 and/or
IL-6 have been observed upon exposure to VT (50-250 ng/ml) or cycloheximide (CI-IX)
in Con A stimulated CD4+ T-cells (Warner et al., 1994; Azeona-Olivera et al., 1995;
Ouyang et al., 1996a). Any of these cytokines could be involved either directly or
indirectly in increased terminal diﬂ‘erentiation of IgA-secreting B-cells (Rotter et al.,

1 996).

The ability of VT to simultaneously stimulate and suppress immune function
presents a seemingly tmresolvable contradiction. Most of the immunotoxic events
demonstrated by VT exposure can be explained by T cell function aberrations. Even
. though other toxic mechanisms such as impaired membrane ftmction (Bunner and Morris,
1988), increased apoptosis (Pestka et al., 1994), or altered intercellular communication
(Jone et al., 1987) can not be discounted, VT-induced immunosuppression is most likely
explained by VT’s ability to bind to ribosomes and inhibit protein translation (Bamburg,
1983). ‘The immunostimulatory eﬁ‘ects of VT involve impairment of normal regulatory
mechanisms (Rotter et al., 1996). Besides stimulating T helper cell cytokines, VT
superinduces IL-1 production in macrophages (Miller and Atkinson, 1986; Mizel and
Mizel, 1981). A likely mechanism for superinduction of cytokines is interference with

synthesis of high turnover proteins that limit transcription (e.g., IxBa) or half-life (e.g.,

1 1
ribonuclease) of interleukin mRNA (Ouyang et al., 1996b).

The possible explanation for VT-induced IgA hyper production is via
superinduction of TH2 cytokines. The following sequence of events after VT exposure in
a lymph node is predicted to go like this: (1) protein synthesis inhibition, (2) enhanced TH
cytokine mRNA expression, (3) elevated TH cytokine production after VT metabolism or
removal and cessation of protein synthesis inhibition, and (4) terminal diﬁ‘erentiation of
Ig-secreting cells. The Peyer’s patches in particular may be prone to this type of
dysregulation since it is exposed to concentrated levels of VT in food and after

enterohepatic recirculation (Pestka and Bondy, 1994).

Immmlahnlm

The basic structure of IgA has been known for some time and has been described
by Roitt et al. (1989). In humans, IgA has a molecular weight of 160,000 and a half-life
of 6 days. In man, 80% of IgA occurs as a basic four-chain monomer, but in other
mammals serum IgA is often a dimer. The a chain consists of 472 amino acid residues
which are arranged in four domains, VH, Cal, Ca2, and Ca3 (Figure 1.2). Secretory IgA
(sIgA), which is the predominant 1g in seromucous secretions, exists mainly as a dimer
and has a molecular weight of 380,000. This molecule consists of two fora-chain units of
IgA, one secretory component (MW 70,000), and one I chain (MW 15,000) (Figme 1.3).
The J chain is produced by the plasma cell, but the secretory component is produced by
epithelial cells. IgA dimers secreted by submucosal plasma cells are held together by a J

chain when it actively binds secretory component as it travels through epithelial cell

 

Figure 1.2. Structure of human monomeric IgA (from Roitt et al., 1989)

    

l
secretory component

Figure 1.3. Structure of human secretory IgA (sIgA) (from Roitt et al., 1989)

13

layers. Binding the secretory component facilitates the transport of sIgA into secretions
as well as protecting it ﬁom proteolytic attack.

At ﬁrst glance, IgA may appear to be an insigniﬁcant class of Ig because it only
accounts for approximately 15% of the human serum Ig pool (Roitt et al., 1989). In
actuality, though it is just a small part of the serum Ig population, the total amount of IgA
produced both in man and experimental animals exceeds the amount of the rest of the Ig
isotypes produced (Mestecky, 1987). Most of this IgA is involved with mucosal
immunity and is the principle host defense against foodbome pathogens and their
products (Kagnoff, 1981; Shah et al., 1982). IgA is derived ﬁom plasma cells residing in
gut-associated lymphoid tissue (GALT). The GALT contains focal accumulations of
these lymphocytes in the lamina propria and Peyer’s patches. Gut epithelium overlies the

Peyer’s patch and is specialized to allow transport of antigens into the lymphoid tissue.
Only IgA, which contains a secretory component, can traverse mucosal membranes,
preventing entry of infectious microorganisms (Roitt et al., 1989). B cells residing in
Peyer’s patches are sensitized by antigens entering through the gut epithelium and
become committed to antigen-speciﬁc IgA production. These cells then enter the
systemic immune system, undergo further diﬂ‘erentiation, and return to the intestinal
lamina propria, where further antigenic stimulation causes them to become fully

differentiated IgA-secreting plasma cells (Bienenstock and Befus, 1980).

l4

Warmth:
VT-induced dysregulation of IgA production closely parallels human IgA

nephropathy (Berger’s disease) which is considered the most common form of
glomerulonephritis in the world (D’Amico, 1987). The major immunological eﬂ‘ects seen
with extended VT exposure include elevated production of IgA, increased IgA immune
complexes, mesangial IgA accumulation and persistent glomerulonephritis accompanied
by hematuria, which are all hallmarks of this autoimmune disease (Dong and Pestka,
1993; Dong et al., 1991). This elevation of serum IgA, mesangial IgA deposition, and
hematuria is so profound that these symptoms persist for up to 3 months after VT
removal from mouse diet (Dong and Pestka, 1993). I gA nephropathy occurs 2 to 6 fold
more often in males than in females, worldwide (D’Amico, 1987). Studies performed in
this lab have also shown a male predilection in the B6C3F, mouse in terms of threshold
toxin dose, onset, and magnitude of response (Greene et al., 1994a). Both the human
disease and the mmine model result in production of polyvalent “natural” IgA that may
be involved in immune complex formation and the subsequent glomerulonephritis
(Rasooly and Pestka, 1992; 1994; Rasooly et al., 1994). The observation that dietary VT
induces murine IgA nephropathy is important because foodbome VT might be an

etiological factor in the human form of the disease.

1 5
WWW

Autoimmunity is a pathological process whereby the body’s immune system
reacts against it’s own tissue. There are more than 40 known autoimmune diseases,
which range ﬁom organ speciﬁc to systemic and affect 5-7% of the human population,
making them a major cause of chronic illness (Scott et al., 1994). Systemic autoimmune
diseases affect multiple systems of the body and share in some of the properties in the
organ speciﬁc disorders. Systemic Lupus Erythematosus (SLE), rheumatoid arthritis, and
Sjogren’s syndrome are all examples of systemic autoimmune diseases. These
multisystem diseases have widespread target tissues and involve the skin, kidneys,
muscles, and joints. These diseases are also characterized by polyclonal B- and T-cell
activation and the presence of autoreactive antibodies to several tissues (Scott et al.,

1 994).

There are several murine models that present a unique opportunity to not only
study systemic immunity, but to study the effect that VT has on individuals with aberrant
immune systems. Speciﬁcally, there are murine models of SLE which develop a
pathology that resembles the human form of the disease (Theoﬁlopoulos and Dixon,
1985). Such mice exhibit B cell hyperactivity, elevated serum IgM and IgG, increases in
production of autoantibodies reactive to widely distributed antigens, and increases in
circulating IgG immune complexes. These eﬁ‘ects culminate in a fatal, immune complex-
mediated glomerulonephritis (Umland et al., 1989). SLE disorder was speciﬁcally
chosen as a study target because of it’s similarity to VT-induced IgA nephropathy. Both

disorders are immune complex diseases which reveal themselves pathologically not as the

16

direct eﬁ‘ect of autoantibodies to speciﬁc tissue but as immune complexes that are
deposited in renal mesangium which leads to glomerulonephritis. The major diﬂ‘erence
between them is that VT induces a IgA pathology whereas SLE is a IgG mediated
disease.

Female models of SLE are important since the incidence of SLE in women is nine
times greater than in men (Dubois, 1974). This is due primarily to the fact that women
typically generate a stronger immune response and thus are potentially more susceptible
to autoimmune disease (TheoﬁIOpoulos and Dixon, 1985). Two of the models used to
study systemic autoimmune disease are female NZBW/F, and MRL/lpr mice. These two
female murine models of SLE may be useful in studying the toxic eﬁ‘ects of VT in mice
with aberrant immune systems since females make up such a high percentage of
individuals with autoimmune disease and the eﬁ’ect VT has on this subpopulation has not
yet been evaluated. The other model, male BXSB mice, is also an important model to
study systemic autoimmune disease because it is a male only model of SLE. The male
BXSB murine model may also be an important model for studying the toxic eﬂ‘ects of VT
in mice with aberrant immune systems because previous studies have shown a male
predilection to VT-induced IgA nephropathy in the B6C3F, mouse (Greene et al., 1994a;
1994b; 1 995).

The female NZBW/F , mice were developed by mating the NZB mouse, which
comes from an rmknown background, with the phenotypically normal NZW strain. The
NZW strain donates modiﬁer genes to the F 1 hybrid which exacerbates the autoimmune

disease and low titers of autoantibodies found in the NZB strain to create F , hybrids with

1 7
intensiﬁed production of autoantibodies, increased circulating immune complexes, and
severe glomerulonephritis (Theoﬁlopoulos and Dixon, 198 5). While the disease in these
mice is clearly genetic in origin, and multiple genes appear to be involved, the genetic
basis for their disease remains elusive (Raveche et al., 1978; Miller et al., 1984).

The female MRL/lpr mice were developed by a series of crosses involving strains
AKR/J, C57BL/6J, C3H/Di and LG] performed by Murphy and Roths (1979) at Jackson
Laboratories. These mice develop SLE if they carry the lymphoproliferation (Ipr) gene
and exhibit increased lymphoproliferation and immune complex glomerulonephritis
(Theoﬁlopoulos and Dixon, 1985). The disease in these animals has been tracked to the
Ipr gene, which encodes the Fas antigen. A retrotransposon has been formd inserted into
the fas gene, resulting in impaired Fas synthesis and expression, and since the normal Fas
cell surface molecule mediates a signaling pathway that initiates programmed cell death
(apoptosis), self-reactive lymphocytes are not eliminated (Steinberg, 1994).

The male BXSB mice were developed through a series of crosses between male
SB/Le mice (carrying the satin and beige genes) and female C57BL/6J mice by Murphy
and Roths (1 97 9). These mice develop SLE with lymph node and spleen enlargement,
increased lymphoproliferation, and immune complex glomerulonephritis that is evident
primarily in males, because the SB/Le males donate the Yaa (Y chromosome-linked
autoimmune accelerator) gene that is found on the Y chromosome (F ossati et al., 1995).
The Yaa gene can not by itself cause autoimmune diseases in mice that are not
predisposed to them, but merely accelerates autosomal autoimmune induction genes

already present in lupus-prone mice.

1 8
W
The toxic and immunologic effects of continuous dietary VT exposure have been

well characterized for the B6C3F1 murine model by numerous studies. This chronic VT
exposure leads to reduced body weight gain, elevated in vitro and in viva production of
IgA, increased IgA immune complexes, mesangial IgA accumulation and persistent
glomerulonephritis. The dietary vomitoxin eﬁ‘ects seen in this murine model were
recognized as being highly analogous to the human disease IgA nephropathy. However,
several limitations exist when employing the continuous dietary VT exposure B6C3F ,
murine model which suggest that it may not accurately reﬂect all human and animal VT
exposure. First, infestation of cereal grains by Fusarium and other molds occurs in “hot
spots” depending on the immediate micro-environment. Thus, VT and other mycotoxins
are likely to be unevenly distributed throughout the grain. Also, because human diet is
not normally composed of a single food source, continuous consumption of a food

source contaminated with VT or other mycotoxin is not likely. It would be more realistic
to see intervals between consumption of mycotoxin contaminated food products based on
the average human diet. Secondly, there already exist many similarities between
systemic lupus erythematosus and VT-induced IgA nephropathy with the primary
diﬂ‘erence being the Ig isotype of the diseases. Although toxic and immunologic effects
of dietary VT exposure have been well characterized in the B6C3F, mouse model, a

question arises as to what eﬂ‘ects VT has on murine models with aberrant immune

systems.

 

 

g1 t
wi
“’2

eit

ga:
m“
Re;
inn
dur
lev
“an
1 3 -

deg

 

Chapter 2. The Effect of Intermittent Vomitoxin

Exposure on Body Weight Gain and the Serum
Immunoglobulin Proﬁle of the B6C3F1 Mouse

 

ABSTRACT

Continuous dietary exposure of female B6C3F 1 mice to the trichothecene VT
results in reduced body weight gain, elevated production of serum IgA, and
glomerulonephritis . To assess whether intermittent consumption of dietary VT, as seen
with most human exposures, has similar eﬁ‘ects as continuous consumption, a comparison
was made between these two different feeding regimens. Female B6C3F, mice were fed
either a semipuriﬁed AIN-76A diet containing 20 ppm VT continuously or 20 ppm VT
intermittently (every other week) for 13 wks. The eﬂ‘ect these diets had on body weight
gain, serum 1g proﬁle, urinalysis, and mesangial deposition were assessed and compared
with each other as well as with mice fed a clean control diet. Reduced weight gains in the
treatment groups were seen as early as 2 wks compared to control. By wk 4 the
intermittent group’s mean weight became successively higher than the continuous group
during wks when it was on a clean diet (through wk 12), but returned to continuous group
levels during the wks it was exposed to VT. Serum IgA levels in the intermittent group
remained at control levels and signiﬁcantly lower than the continuous group during the
13-wk study, but serum IgG and IgM levels for the intermittent group were signiﬁcantly

decreased compared to control and thus mimicked those of the continuous group.

19

20

Hematuria was signiﬁcantly greater in both treatment groups compared to control at wks
5 and 13 when the intermittent group was on the VT containing diet, but hematuria in the
intermittent group dissipated at wk 10 when it was on clean diet. Mesangial IgA
deposition was signiﬁcantly lower in the intermittent group compared to the continuous
group and had levels comparable to mice on the control diet. The results suggest that
intermittent exposure to VT had reduced toxic effects in B6C3F, mice as compared to the
continuous model. Animal models for mycotoxin exposure may need to be reevaluated

to account for sporadic and intermittent exposure patterns.

 

 

 

 

species
produce
to exce-
the pres
probler

process

Synthes

 

1983).
includi
Rotter
humor

1987).

elevat
muco:
produ.

COnCL

Pe

2 1
INTRODUCTION

Vomitoxin (deoxynivalenol or VT), is a trichothecene mycotoxin produced by
species of the fungal genus, F usarium. VT (MW 296) is a secondary metabolite
produced during conditions favoring fungal colonization. Environmental crop stress due
to excessive rainfall and cold weather as well as improper storage conditions can lead to
the presence of VT in cereal grains (Abouzied et al., 1991; Tanaka et al., 1988). The
problem is further exacerbated by VT’s resistance to inactivation during milling and
processing which allows it to contaminate grain-based food products worldwide.

Tricothecenes, including VT, are among the most potent inhibitors of protein
synthesis known and have been implicated in both animal and human toxicoses (U eno,
1983). Dietary VT has been shown to have chronic effects on experimental animals
including reduced body-weight gain, feed refusal, and vomiting (F orsell et al., 1986;
Rotter et al., 1996). Sublethal levels of ingested VT can also alter host resistance,
humoral immunity, and cell-mediated immunity (Pestka and Bondy, 1990; Pestka et al.,
1987).

One of the most notable immunological eﬂ‘ects of extended VT exposure is
elevated production of immunoglobulin A (IgA), which is symptomatic of an aberrant
mucosal immune response (F orsell et al., 1986). The net result of this elevated
production is that IgA becomes the major isotype of immunoglobulin (lg) in the serum.
Concurrent with this is an elevation of IgA immune complexes (IgA-1C), mesangial IgA
accumulation and persistent glomerulonephritis accompanied by hemattuia (Dong and

Pestka, 1993; Dong et al., 1991). These VT-induced effects closely mimic a human

22

disease known as IgA nephropathy, which is the most common form of
glomerulonephritis worldwide (D’Amico, 1987). The ability to use a VT-induced mmine
IgA nephropathy model is important because foodbome VT might be an etiological factor
in the human form of the disease and because it allows further study of the development
of IgA-mediated glomerulonephritis.

The toxic and immunologic effects of continuous dietary VT exposure have been
well characterized for the B6C3F, murine model by numerous studies. However, a
limitation of models employing continuous dietary exposure to the toxin, is that it may
not accurately reﬂect human or animal VT exposme. Infestation of cereal grains by
Fusarium and other molds occurs in “hot spots” depending on the immediate micro-
environment. Thus, VT and other mycotoxins are likely to be rmevenly distributed
throughout the grain. Also, because human diet is not normally composed of a single
food source, continuous consumption of a food source contaminated with VT or other
mycotoxin is not likely. It would be more realistic to see intervals between consumption
of mycotoxin contaminated food products based on the average human diet. The purpose
of the following study was to characterize the effects of intermittent exposure to VT on

body weight gain and various immunologic parameters in the B6C3F, mouse model.

23
MATERIALS AND METHODS

W. All chemicals (reagent grade or better), immunological

reagents and media components were purchased from Sigma Chemical Co. (St. Louis,
MO, USA) unless otherwise noted. Twenty-ﬁve B6C3Fl (C57BL/6 x C3H/HeN) female
mice (7-8 wks old, Harlan/Sprague-Dawley, Indianapolis, MO, USA) were randomly
divided into three groups. The mice were housed as pairs in environmentally protected
cages (Nalgene, Rochester, NY, USA) which consisted of a transparent polycarbonate
body with a ﬁlter bonnet, stainless steel wire lid, and a raised ﬂoor above a layer of heat-
treated hardwood chips. Upon arrival, the mice were introduced to a semi-pmiﬁed
powdered A1N-76A diet (ICN Nutritional Biochemical, Cleveland, OH, USA). The feed
was changed every 3-4 days and ﬂesh water was provided ad libitum The mice were
then allowed to acclimate for at least 7 days to their new housing, regulated temperatme
(72 °F), feed, 12-hr light/dark cycle, and to a negative-pressure ventilated area before
feeding regimens began.

The VT used in this study was produced in Fusariwn graminearum R6576
cultures and puriﬁed via the water-saturated silica gel chromatography method of
Witt et al. (1985). Puriﬁed VT was then mixed into the powdered A1N-76A diet as
described previously by Pestka et al. (1987). The control group (n=8) was fed a diet of
powdered AlN-76A containing no VT, the continuous group (n=9) was fed a
diet containing 20 ppm VT, and the intermittent group (n=8) was alternatively fed the

control or 20 ppm VT diet everyday dming one week intervals for 13 wks starting with

24

control diet at wk 0.

The weight of each mouse was measured weekly throughout the study. The mice
were also subjected to blood and mine collection at varying intervals during the 13-wk
study. For the intermittent groups, at least one blood and urine sample collection was
performed while the group was fed control diet and at least one blood and urine sample
collection while the group was fed the 20 ppm VT diet.

Approximately 42 hrs before the conclusion of the study, the immediate eﬂ'ect of
VT exposure on mesangial deposition of IgA, IgG, and C, was tested. All mice had their
feed withheld for 24 hrs but, were still supplied with ﬂesh water. At the end of the 24 hr
fast, 4 mice from each group had clean diet given to them and the other 4 mice ﬁ'om each
group were given 20 ppm VT diet for 18 hrs. The study was concluded by euthanizing the
mice in each group. Kidneys of each mouse were then removed for
immunohistochemistry.

ELISA. At wks 0, 4, 9, and 12, the mice were anaesthetized with ether and bled
through the retro-orbital plexus. The serum collected was used for lg quantiﬁcation and
to analyze for antibodies to oral- and self-antigens.

Serum IgA, IgG, and IgM were measured using an enzyme-linked immrmosorbent
assay (ELISA) which was previously described by Pestka et al. (1990 a). Brieﬂy,
Immulon II Removawell microtiter strips (Dynatech Laboratories, Chantilly, VA, USA)
were coated with 50a] of goat anti-mouse IgA, IgG, or IgM (Cappel Laboratories,
Malvern, PA, USA) that had been diluted 1:1000 in 0.1 M bicarbonate buffer (pH 9.6)

and these were incubated overnight at 4°C. Strips were then washed three times with

25

0.01 M phosphate buffered (pH 7.5) saline (PBS) containing 0.2% Tween 20 (PBS-
Tween). Each well was then ﬁlled with 300 pl of 1% (w/v) bovine serum albumin (BSA)
in PBS (1% BSA-PBS) and incubated for 30 minutes at 37°C to reduce non-speciﬁc
protein binding. The strips were then washed four times with PB S-Tween. A standard
curve was then developed for each plate using mouse immunoglobulin reference serum
(Bethyl Laboratories, Montogomery, TX, USA) to quantify Ig concentrations for each
serum sample. The mouse immunoglobulin reference serum and the serum samples were
diluted with 1% BSA-PBS, and 50 al was added to the appropriate wells. The plates
were sealed with paraﬁlm and aluminum foil and were incubated at 37 °C for 1 hr. The
plates were then washed ﬁve times with PBS-Tween. 50 pl of goat anti-mouse Ig bound
to horseradish peroxidase (HRP) (a-, y- or a-chain speciﬁc, Cappel Laboratories,
Malvern, PA, USA), diluted 1:500 in 1% BSA-PBS, was added to each well. Plates
were rescaled, incubated at 37°C for 30 minutes, and washed six times with PBS-Tween.
Bound peroxidase was determined with 100 ul/well 2, 2-azino-bis (3-ethylbenzoline—6-
sulfonate) (ABTS) substrate [0.4 mM ABTS, 50 mM citrate buffer (pH 4.0), and 1.2 mM
hydrogen peroxide] and allowed to develop color at room temperature (15 min for IgA, 5
min for IgG, and 8 min for IgM) before the reaction was stopped with 100 141 0.3 M citric
acid monohydrate containing 0.02% sodium azide. The absorbance was read at 405 nm
with the aid of the Molecular Devices kinetic microplate reader (Molecular Devices,
Menlo Park, CA, USA) and 1g quantity determined using Soﬁmax 2.32 version cm've

ﬁtting software (Molecular Devices).

26

Reactivity of serum IgA and IgG to model oral- and self- antigens was assessed
by solid phase antigen binding ELISA. TNP-B SA (trinitrophenylated bovine serum
albumin) was previously prepared by Linda Rasooly by trinitrophenylating BSA with
picric acid to yield a ratio of 53 moles TNP per mole of BSA (Good et al., 1980). Casein
was purchased from ICN Nutritional Biochemical in a puriﬁed form. Salmon sperm
DNA, which was purchased ﬁom Sigma, was extracted with a phenol/chloroform mixture
(Ausubel et al., 1990). IgA-MOPC 315 was purchased from Sigma. Immulon II
Removawell microtiter strips were coated with 50 all well TNP-BSA, casein, salmon
sperm double-stranded DNA, or IgA-MOPC 315 at a concentration of 10 ug/ml. TNP-
BSA, casein, and IgA-MOPC 315 were diluted with 0.5M bicarbonate buffer (pH 9.6),
covered with paraﬁlm and aluminum foil, and incubated overnight at 4°C. Salmon sperm
DNA was diluted in 0.1 M PBS (pH 7.5), left unwrapped, and dried overnight at 40°C
(Rasooly and Pestka, 1992). To determine IgA and IgG reactivity to speciﬁc antigens,
ELISAs were run exactly like those to quantify serum Igs. The only change made was
exclusion of mouse immunoglobulin reference serum to develop a standard curve, so IgA
and IgG concentrations are reported as absorbance units (Rasooly and Pestka, 1994).
BSA-PBS (1%) was used to dilute reagents and block background binding in the ELISA,
which inhibits binding of BSA-speciﬁc antibodies, yielding a zero background (Klinman,
1 992).

W. Mice were placed, individually, in metabolic cages overnight to
collect urine during a 12-hr period (~ 2 ml/mouse) at wks 5, 10, and 13 ofthe study. The

samples were centrifuged at 450 x g for 10 minutes. The supernatant was pipetted off,

27
leaving only the sediment behind. The sediment was vortexed and 50 pl of the

suspension was pipetted onto a microscope slide. Erythrocytes in 3 random microscopic
ﬁelds (X400) were counted and averaged (Dong et al., 1991).

WWW At wk 13. the mice were
euthanized with C02 and their kidneys were removed. Kidneys were cut in halfand
immediately frozen in liquid nitrogen. Each kidney was sectioned to 7 am with a
cryostat (Reichert-Jung, Cambridge Instruments, Buffalo, NY, USA) and stained for lg or
complement deposition with FITC-labeled goat anti-mouse IgA, IgG, or C3 (Cappel) as
previously described by Valenzuela and Deodhar (1981). Sections ﬁ'om each animal
were viewed under a Nikon Labophot epiﬂuorescence microscope. Fluorescent
intensities of 10 glomeruli ﬁom each section were measured using an ITM densitometric
video camera (W altham, MA) and JAVA image analysis system (Jandel Scientiﬁc, San
Rafael, CA, USA). The Java system generates a quanitative value ﬁ'om an encircled
immunoﬂuorescent stained glomerulus in a ﬁozen ﬂame and calculates the average
brightness for the circled area based on each pixel of the screen included in the circle.
The pixels in the circled area were measured on a grayness scale that ranged ﬁ'om 0
(black) to 255 (white).

Safety. Face masks and vinyl gloves were used for preparation of the VT.
Concentrated toxin solutions were handled in a fume hood. Any labware that was
contaminated with mycotoxin was detoxiﬁed by soaking overnight in 10% sodium

hypochlorite (Thompson and Wannemacher, 1984).

28
W. All animal handling was conducted in strict accordance with

regulations established by the National Institutes for Health Experiments were designed

to minimize numbers of animals required to adequately test the pr0posed hypothesis and

approved by Michigan State University Laboratory Animal Research committee.
W. The data were analyzed using the SigmaStat computer

program (Jandel Scientiﬁc) and reported as mean i SEM. Signiﬁcant differences

(p < 0.05) between groups were analyzed by Kruskal-Wallis and Dunn’s method for

multiple comparisons of nonparametric data Parametric data were analyzed by the

AN OVA and by Student-Newman-Keul tests for multiple comparisons.

29
RESULTS

Bony Weight

Body weights of mice in the control, continuous, and intermittent groups were
monitored weekly throughout the 13-wk feeding period (Figure 2.1). Body weights of all
three groups started out as statistically identical at wk 0 (~22 g). The mean body weight
of the continuous group remained fairly constant throughout the experiment, whereas
the mean body weight of the control group became signiﬁcantly higher than the
continuous group as early as 2 wks into the feeding regimen and remained so throughout
the study. Mean weight of the intermittent group remained consistent with the results of
the continuous group through wk 3, but as wk 4 began a new trend appeared. Mean body
weights progressively increased on even wks when the group was fed unspiked diet but
decreased to the level of the continuous group when fed the VT—spiked diet. At wk 12,
the body weight of the intermittent group was signiﬁcantly different ﬁom the continuous
group. The substantial recovery of the intermittent group on even wks of unspiked diet
never reached the mean weight gain of the control group and beginning at wk 2, was

signiﬁcantly less than the control group at all wks but wk 10.

Serum Igs

Serum Igs were monitored during the 13-wk feeding period to determine how the
different feeding regimens aﬁ‘ected isotype distribution. Serum samples taken on wks 0,

4, 9, and 12 were tested for IgA, IgG, and IgM. Serum IgA concentration in the

30

Figure 2.1. Effect of intermittent and continuous dietary vomitoxin exposure on
body weights of B6C3F 1 mice compared to control. Data are means d: SE (8-9
mice/group). The intermittent group was fed 20 ppm vomitoxin on odd weeks and a
clean diet on even weeks. ‘Control group signiﬁcantly different than both intermittent
and continuous groups at p s 0.05. t’Indicates group is signiﬁcantly different than the
continuous group at p s 0.05.

WEIGHT (9)

4O

31

 

35

30 -

25—

 

l

 

CONTROL
INTERMHTENT
I CONDNUO S,

.34

/

 

b

/’ k n
a N At‘
./ ' i I

 

 

menﬁLl

10

12

14

32

continuous group was signiﬁcantly higher than both the control and intermittent groups at
wks 4, 9, and 12 (Figure 2.2), whereas there was no difference between the intermittent
and continuous group. Serum IgG concentration in the continuous group was
signiﬁcantly higher than both the control and intermittent groups at wk 0, but became
signiﬁcantly lower than the control group by wks 9 and 12 (Figure 2.3). Serum IgG in
the continuous group was signiﬁcantly lower than the intermittent group only at wk 9.
The intermittent group became signiﬁcantly lower than the control group by wk 9 and 12.
Serum IgM concentrations for the intermittent and continuous groups became

signiﬁcantly lower than the control group at wk 12 of the study (Figure 2.4).
Serum Autoantibodies

Igs ﬁ'om the serum sample drawn on wk 12 were tested for their ability to bind to
speciﬁc self and non-self antigens. Serum IgA and IgG were tested for speciﬁcity to
_ TNP-BSA, casein, and dsDNA, and IgG was tested for speciﬁcity to IgA (Table 2.1).
IgG reactivity for TNP-BSA, casein, and dsDNA and for IgA speciﬁcity to casein, in the

continuous group was signiﬁcantly lower than both the intermittent and control groups.
Urinalysis

Hemattuia was monitored at wks 5, 10, and 13 ofthe study as a measrue for
glomerulonephritis severity (Figure 2.5). Erythrocyte counts for both the intermittent and

continuous groups were signiﬁcantly higher than the control at wks 5 and 13.

33

Figure 2.2. Eﬁect of intermittent and continuous dietary vomitoxin exposure on
serum IgA levels of B6C3F1 mice compared to control. Data are means :I: SE (8-9
mice/group). 'Indicates signiﬁcantly different than control group at p s 0.05.
l’Indicates signiﬁcantly different than other treatment group at p s 0.05.

 

       

will §§§§
S\§ w.

w, §§§

§§9

 

 

34

 

 

        

§4

 

§
§0

1- CONTROL
INTERMITTENT

 

 

2500

O

2000 — | CONTINUOUS

1500 -
IOOO -
500 e

geese <9

WEEKS

Figure 22

35

Figure 2.3. Effect of intermittent and continuous dietary vomitoxin exposure on
serum IgG levels of B6C3F1 mice compared to control. Data are means :h SE (8-9
mice/group). ‘Indicates signiﬁcantly different than control group at p s 0.05. bIndicates
signiﬁcantly different than other treatment group at p s 0.05.

36

 

 

 

 

l- CONTROL

INTERMITTENT
[ CONTINUOUS

 

 

 

a §§

a §\\\\\\§ w.

   
       

§\\\\\\§ 9

 

a

‘\\\\\.\k 4

m \\\\\\\\\\\\\\\\\\\\\\\\\\\\.
S\\\LO

 

 

4000

3500 - ‘

3000 -

O

2500 -
2000 -
1500 -
1000 -
500 -

A_E\miv Om.

WEEKS

Figure 2.3

37

Figure 2.4. Effect of intermittent and continuous dietary vomitoxin exposure on
serum IgM levels of B6C3F 1 mice compared to control. Data are means t SE (8-9
mice/group). ‘Indicates signiﬁcantly different than control group at p s 0.05.

38

 

 

 

 

 

 

 

             

a§

 

 

1100

A_E\m§.v 29

§\\\\\\\\\\\\§ 9
we §
mmw INNS...
mmm
NEN
wmm
V \\\\\\\\\\\\\\\\\\\
us® §\\\\ho
M. ... ... ... ... ... ... ... ... ... o
O 0 O O O O O O O O
m 9 8 7 6 5 4 3 2 1

a NNNN w.

WEEKS

Figure 2.4

39

.36 W a “a use» “5830.: ..oﬁo 55 320%.. hue—gouge assume—Q.

.36 w a 3 95cm 35:8 55 Engage maﬁa—mama 838mg?

.352» .89 8:: 035 Mm a 238 8a 8.5..
date.— 25» 25m 05 mam—.6 .3? :26 go

Home 520 a can .9? Eco co H> Egg on to.“ 83 95% 353E035 05. 5.3 2 can Anson—£58 H> .8 8% 35:8 B.“ 803 032.

 

 

 

rune 232 2.33 ea <2
”oxen :32 Seen <3
....reo. :3: area we <zee
38...me games €4.28 <2
range 3.38 :48 as case
Sana 288 wanes <3
:23: are: area me $25
.5 S .3255.
95352.30 «23:53:: 35:00 0333“ newt:
$.25 .... ado

 

.mumeoeuﬁouﬁx 880m Lo 5:885 05 co C3 Exo=Eo> E05 .«o 80mm .ﬁm 035.

40

Figure 2.5. Effect of intermittent and continuous dietary vomitoxin exposure on
hematuria of B6C3F1 mice compared to control. Data are means t SE (8-9 mice/group).
‘Indicates signiﬁcantly different than control group at p s 0.05. l’Indicates signiﬁcantly
different than other treatment group at p s 0.05.

41

 

 

a,b

 

S\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

. X
a ‘ 3

  

1

 

  

 

 

 

[- CONTROL

INTERMITTENT

 

 

 

m CONTINUOUS

 

 

O 5 O 5

Dig... mud mu§00mIgmm
mo mumEDZ mo<mm><

 

Figure 2.5

42

At wk 10, there was a decline in the number of erythrocytes per microscopic ﬁeld for the

intermittent group which made it signiﬁcantly lower than both the control and continuous

groups.

Mesangial 1g and C3 deposition

Mesangial IgA, IgG, and C3 were measured by immunoﬂuorescence in
conjrmction with immunoﬂuorescence microscopy in (Table 2.2). The most prominent
difference observed was elevated I gA deposition in the continuous group compared to

what was observed in both the intermittent and control groups.

43

.36 W a an macaw 3:25.35 .850 ~85 «caobmv .3ngme Bungee—o

.8... w a a Saw .880 55 aeoee $583... .382.
E: 3 he 3:. H> can cm 05 52w 203 028 v 35¢ 05 e5 3% 35:8 53m 203 9.2m
some Eon.“ 3:: v .33.“ ..E em 05 come. See _ Bu 038 =« 89¢ 20:55 33 team .eomaEES 283 .0: Ne beagxoanﬁ.

doc—.808 23—3083» non 8235 no 33

05 E eomﬁog .329 33:29: con mmmé go 0.3» Bog—Em a no 332: Be eoBmauE =EoE£w a: mm a 838 0.3 San.

._ 038. E eoetomoe we we.“ 825 8:2.

 

 

333.2 edges no
...233 snares 02
$63.3 333.8 <3 88528
330.8 5.3.3 .0
E33 393% we
33.: 363.8 <3 avenues
3.3.3 235 nO
33$ 33$ cs
23.? 4.93.3 <3 .880
09.80 .9595 09C.
35 h> $5 .9980 328qu .5832...

 

 

.no 93 .03 .<w_ no 53839 REESE :o 058an 0.3 53:89» 505 E0... meoq Ea ouameoaﬁm no “comm .m.m mqmﬂn

44
DISCUSSION

The B6C3Fl mouse has been used extensively as a model for dietary VT exposure
by our research group. The use of the B6C3Fl murine model for this study was crucial
because of the ability for comparison it provided. One concern with these previous
studies is that they focused on the long term effects of continuous dietary exposure to VT
on the B6C3Fl mouse. While beneﬁcial for indicating the toxic potential that VT can
possibly cause, continuous exposure may not accurately mirror human exposure in
developed countries. This is because the human diet is not normally composed of a
single food source, but is made up of a wide variety of foods that contribute as a whole to
human health. Also, VT and other mycotoxins occur as “hot spots” in animal feed. Thus,
rather than being exposed to a uniform mycotoxin level, farm animals might also be
exposed to intermittent exposures. The purpose of this research was to evaluate the eﬂ‘ect
of intermittent exposure to dietary VT using the well—deﬁned B6C3F, murine model. To
best mimic human exposure, the mice in the intermittent exposure group were given a
diet of 20 ppm VT on alternate weeks. Dietary VT exposure at the 20 ppm concentration
has been previously shown to be Optimal for inducing increased IgA levels in the B6C3F1
mouse (Pestka et al., 1989; Greene et al., 1994a). The intermittent exposure of the mice
to this concentration had the most profOtmd effects on body weight, serum Ig, and
hematm'ia.

Reduced body weight observed in the continuous dietary VT group is similar to
what has been reported previously in the B6C3F, model (Forsell et al., 1986). The mean

body weight of this group remained constant throughout the l3-week study compared to

45

the control group whose mean body weight surpassed that of the continuous group at wk
2 and rose progressively throughout the study. The reduced weight gain seen by the
continuous VT exposure group has been previously explained to result from reduced feed
conversion efﬁciency and feed refusal (F orsell et al., 1986; Rotter et al., 1992). Notably,
the weight gain of the intermittently fed group directly correlated to whether animals
were on the control or VT diet. Dm'ing the weeks when this group was on the 20 ppm VT
diet, their mean body weight consistently matched that of the continuous group whereas,
when this group was on the control diet their mean body weight increased remarkably.
Analogous effects were found by Dong and Pestka (1993), when the mean body weight
of B6C3F1 mice recovered after 8 wks exposure to VT and subsequent removal of the
VT-spiked diet.

Continuous daily exposure to dietary VT at a concentration equivalent to 1/20 of
the LD50 mg/kg body weight/day for the B6C3Fl mouse can dramatically change the
normal serum Ig proﬁle in these mice (F orsell et al., 1986). Dietary VT exposure
increases total serum IgA levels and the polymeric/monomeric IgA ratio and decreases
total serum IgG levels (Pestka et al., 1989). The intermittent group’s serum IgA levels
remained at control levels throughout the study and was signiﬁcantly less than the
continuous group as early as wk 4. However, intermittent dietary VT exposure had the
same effect as continuous exposure which caused decreased serum IgG and IgM levels.

These results show that even intermittent exposure to VT can lead to dysregulation of Ig

isotypic proﬁle.

46

Continuous dietary VT exposure has also been implicated in increased serum IgA
speciﬁc to intestinal and self antigens and a decrease in IgG speciﬁc antibodies (Pestka et
al., 1990 b; Rasooly and Pestka, 1992; Rasooly and Pestka, 1994). Here, continuous
dietary exposure for 13 wks caused a signiﬁcant decrease in serum IgG speciﬁc to TNP-
BSA, casein, and ds DNA and serum IgA speciﬁc to casein compared to both the control
and intermittent groups. Thus, intermittent exposure to dietary VT was not similar to
continuous exposure and the decrease in IgG speciﬁcity to these antigens results from
prolonged exposure to dietary VT.

Decreased kidney damage to the intermittent group was also reﬂected in the levels
of accumulated mesangial IgA in the glomeruli. Dietary VT has been previously shown
to cause accumulation of IgA into the glomerulus of kidneys (Pestka et al., 1989). The
continuously exposed group from this study was shown to be signiﬁcantly higher than
both the control and intermittent group, suggesting that IgA accumulation in glomeruli
. was less severe in the intermittent group.

Hematuria was monitored as an indication of kidney damage. It has been shown
that animals subjected to continuous VT exhibit increased hematuria as early as wk 4
(Greene et al., 1994b). Intermittent exposure to dietary VT had a similar effect as was
seen in the weight gain of these mice. During wk 10 when the intermittent group was on
the control diet, the hematuria level was decreased. During wks 5 and 13 when the group
was on the VT diet, the hematm'ia level was signiﬁcantly higher than the control group.
It is also notable that at wk 13 the intermittent group, though higher than the control

group, was signiﬁcantly less than the continuous group. These results suggest that severe

47

hematuria and presumably, kidney dysﬁmction, is a cumulative effect. Intermittent VT
exposure can lead to increased hematuria during exposure, but is reversible during times
of non exposure. Previous work has shown that once B6C3F, mice are removed from a
continuous VT diet after 8 wks, their hematuria levels become signiﬁcantly less than
mice which remain on the VT-containing diet (Dong and Pestka, 1993). This explains
why the wk 13 results show the continuous group signiﬁcantly higher than the
intermittent group. The recovery time during a control feeding wk reversed a lot of the
kidney damage, allowing a less severe hematuria during a wk of continuous VT
ingestion.

Taken together, the results presented here suggest the type of dietary exposure
regimen is critical in determining the extent of toxic eﬂ‘ects induced by VT and perhaps,
other mycotoxins. Continuous dietary VT exposure may not be the ideal model for
predicting human and animal toxicoses. The B6C3Fl intermittent VT dietary model
more closely resembles human VT exposure and has shown itselfto be quite diﬂ'erent
ﬁom the aforementioned model. Understanding the effects of intermittent VT exposure
will be necessary in determining the potential effects of this and other mycotoxins in

human diseases.

 

Chapter 3. Differential Effects of Subchronic Dietary
Vomitoxin Exposure on Murine Models for Systemic
Lupus Erythematosus

 

ABSTRACT

Dietary exposure to the trichothecene VT results in reduced body weight gain,
elevated production of serum IgA, terminal differentiation of Peyer’s Patch B cells to IgA
secreting plasma cells, and elevated kidney mesangial IgA accumulation in the BéC3F1
mouse. These eﬂ‘ects closely mirror human glomerulonephritis IgA nephropathy. Using
autoimmune—prone NZBW/F ,, MRL/lpr, and BXSB mouse strains as models, I assessed
whether consumption of diet containing 5 ppm or 10 ppm VT will similarly aﬂect mice
with aberrant immune systems. Reduced weight gains were seen in NZBW/F, and
MRL/lpr mice fed both doses of VT within 2-3 wks. In contrast, VT had little eﬂ‘ect on
BXSB weight gain. Serum lg levels in all three strains generally did not diﬂ’er from
control levels throughout the study. Hematuria was signiﬁcantly increased when all three
strains were fed l0 ppm VT. In the NZBW/F. Peyer’s patch cultures stimulated with
LPS, prior exposure to 10 ppm VT signiﬁcantly increased the IgG and IgM secretion but
had no eﬂ‘ect on IgA. In MRL/Lpr Peyer’s patch cultures stimulated with LPS, exposure
to 10 ppm VT increased IgA secretion but not IgM or IgG. BXSB Peyer’s patch cultures
prepared from both VT treatment groups produced signiﬁcantly more IgA than control

when stimulated with LPS or Con A. Mesangial deposition of IgA and IgG was

48

49
signiﬁcantly loWer in the treatment groups of NZBW/F 1 and MRL/lpr mice compared to
control. BXSB mice had signiﬁcantly higher IgA, IgG, and complement (C3) deposition
in the 10 ppm VT group as compared to both the 5 ppm VT and control groups. The

results suggest that VT differentially affected mice with aberrant immune systems.

50
INTRODUCTION

Vomitoxin (deoxynivalenol or VT), a trichothecene mycotoxin, is a toxic,
secondary metabolite produced by F usarium graminearum that is frequently found in
cereal grains and grain-based food products (Abouzied et al., 1991; Tanaka et al., 1988).
The problem of VT contamination in such food staples as corn, wheat, barley, rice, and
oats is further exacerbated by it’s resistance to inactivation during milling and processing
which allows it to contaminate food products worldwide.

Trichothecenes, such as VT, are sesquiterpenoid compormds that are among the
most potent inhibitors of protein and DNA synthesis known and that have been
implicated in both animal and human toxicoses (McLaughlin et al., 1977; Ueno, 1983).
Chronic dietary VT has been shown to cause a variety of toxic eﬁ‘ects in experimental
animals including vomiting, feed refusal, and reduced body weight gain (Forsell et al.,
1986; Rotter et al., 1996). Subchronic dietary VT ingestion has also been shown
experimentally to impair host resistance, humoral immunity, and cell-mediated immune
function (Pestka and Bondy, 1990; Pestka et al., 1987).

Besides the immunosuppressive eﬂ‘ects, dietary VT ingestion has been shown to
be simultaneously immmostimulatory in experimental animal models (Pestka and
Bondy, 1994). The most notable stimulatory effect of extended VT exposure is elevated
production of immunoglobulin A (lgAXT-‘orsell et al., 1986), which is symptomatic of an
aberrant mucosal imm1me response. This elevated production results in IgA becoming the

major isotype in the serum. Dietary VT also increases circulating IgA immtme

5 1
complexes, spontaneous and mitogen-stimulated IgA production by isolated
lymphocytes, hematuria, and mesangial IgA deposition in experimental mice (Dong and
Pestka, 1993; Dong et al., 1991). These VT-induced effects closely parallel the human
autoimmune disease known as IgA nephropathy, which is the most common form of
human glomerulonephritis worldwide (D’Amico, 1987). The observation that dietary VT
induces murine IgA nephropathy is important because foodbome VT might be an
etiological factor in the human form of the disease and because it can be used as a model
for ﬁrrther study of the development of IgA-mediated glomerulonephritis.

Although toxic and immunologic eﬂ‘ects of dietary VT exposure have been well
characterized in the BGC3F, mouse model, a question arises as to what eﬁ‘ects VT has on
murine models with aberrant immune systems. Autoimmunity is a pathological process
whereby the body’s immune system reacts against it’s own tissue. There are more than
40 known autoimmune diseases, which range ﬁom organ speciﬁc to systemic and aﬂect
5-7% of the human population, making them a major cause of chronic illness (Scott et al.,
1994). Systemic autoimmune diseases aﬂ'ect multiple systems of the body and share in
some of the properties in the organ speciﬁc disorders. Since it is of particular interest to
study the potential effect dietary VT has on such individuals, it would be useful to
employ murine models of systemic autoimmunity.

The female MRL/lpr, female NZBW/F ., and male BXSB mouse strains are
murine models of systemic lupus erythematosus (SLE) that develop a pathology which
resembles the human form of the disease (Theoﬁlopoulos and Dixon, 1985). These mice

exhibit B cell hyperactivity as evidenced by lymphadenopathy and splenomegaly,

52

elevated serum I gM and IgG, increases in production of autoantibodies reactive to widely
disuibuted antigens, and increases in circulating IgG immune complexes. These effects
culminate in a fatal, immune complex-mediated glomerulonephritis (U mland et al.,
1989). There are many similarities between systemic autoimmunity and VT-induced IgA
nephropathy with the primary difference being the Ig isotype of the diseases. The
purpose of the following study was to characterize the effect of subchronic exposure to
VT on immunologic parameters associated with the autoimmune murine models, and to
determine what effect genetic predisposition to a systemic autoimmune disease has on

previously documented effects of VT induced IgA nephropathy.

53
MATERIALS AND METHODS

W. All chemicals (reagent grade or better), immunological

reagents and media components were purchased ﬁ'om Sigma Chemical Co. (St. Louis,
MO, USA) unless otherwise noted. All animal handling was conducted in strict
accordance with regulations established by the National Institutes for Health.
Experiments were designed to minimize numbers of animals required to adequately test
the proposed hypothesis and approved by Michigan State University Laboratory Animal
Research committee.

Twenty NZBW/F 1 female mice (6 wks old, Jackson Laboratory, Bar Harbor, ME,
USA), twenty-one MRL/lpr female mice (5 wks old, Jackson Laboratory, Bar Harbor,
ME, USA), and twenty-one BXSB male mice (6 wks old, Jackson Laboratory, Bar
Harbor, ME, USA) were each randomly divided into three groups. The mice were housed
' as pairs in environmentally protected cages (N algene, Rochester, NY, USA) which
consisted of a transparent polycarbonate body with a ﬁlter bonnet, stainless steel wire lid,
and a raised ﬂoor above a layer of heat-treated hardwood chips. Upon arrival, the mice
were introduced to a semi-puriﬁed powdered AIN-76A diet (ICN Nutritional
Biochemical, Cleveland, OH, USA). The feed was changed every 3-4 days and fresh
water was provided ad Iibitum. The mice were then allowed to acclimate for at least 7
days to their new housing, regulated temperatm'e (72 °F), feed, 12-hr light/dark cycle, and

to a negative-pressure ventilated area before feeding regimens began.

54

The VT used in this study was produced in Fusarium graminearum R6576
cultures and puriﬁed via the water-saturated silica gel chromatography method of
Witt et al. (1985). Face masks and vinyl gloves were used for preparation of the VT.
Concentrated toxin solutions were handled in a fume hood. Any labware that was
contaminated with mycotoxin was detoxiﬁed by soaking overnight in 10% sodium
hypochlorite (Thompson and Wannemacher, 1984). Puriﬁed VT was then mixed into the
powdered AIN-76A diet as described previously by Pestka et al. (1987). The control
groups (n=6-7) of all three experimental strains were fed a diet of powdered AIN-76A
containing no VT. The two treatment groups (n=7) of all three experimental strains were
fed a diet containing 5 ppm and 10 ppm VT, respectively. The NZBW/F, strain was
subjected to this feeding regimen for 11 weeks, the MRL/lpr strain for 9 wks, and the
BXSB strain for 14 wks (termination timetable was determined by the health of the
strains during the study).

The weight of each mouse was measured weekly throughout the study. At wks 0,
4, 8, and 11 for the NZBW/F, strain; wks 0, 5, and 9 for the MRL/lpr strain; and wks 0, 5,
11, and 14 for the BXSB strain, the mice were anaesthetized with ether and bled through
the retro-orbital plexus. The serum collected was used for lg quantiﬁcation and to
analyze for antibodies to oral- and self-antigens. Mice were also placed, individually, in
metabolic cages overnight to collect urine during a 12-hr period (~ 2 ml/mouse) at wks 5
and 10 of the study for all strains of mice except the MRL/lpr strain which only had
urinalysis performed at wk 5 since the study was terminated at wk 9. The studies were

concluded by euthanizing the mice in each group. Spleens (SP) and Peyer’s patches (PP)

55

were removed for culture and the kidneys of each mouse were removed for
immunohistochemical examination.

W. At wk 11 for the NZBW/F, strain, wk 9 for the MRL/lpr
strain, and wk 14 for the BXSB stain, the mice were euthanized with CO, and their PP
and SP were removed. The PP were pooled (2-3 mice/pool) and teased apart in Hanks
balanced salts media and then passed through an 85-mesh stainless steel screen to
separate the lymphocytes from the membrane debris. The cells were centiﬁrged at 450 x
g for 10 min, washed with 10 ml RPMI 1640 medium supplemented with 100 U/ml
penicillin, 100 ug/ml steptomycin, 5 x 10" M 2-mercaptoethanol, 25 mM Hepes buﬁ‘er,
lmM sodium pyruvate, and 0.1 mM non-essential amino acids (RPMI), and centrifuged a
second time. SP were also teased apart in RPM, cellular membranes were removed, and
the lymphocytes were centrifuged at 450 x g for 10 min. The cells were then washed
with 10 ml RPM] and centrifuged a second time, and resuspended in RPMI supplemented
with 10% (v/v) fetal calf serum (F CS; Gibco Laboratories, Chagrin Falls, IL, USA).

Lymphocyte cormts were determined using a Neubauer hemacytometer
(American Optical, Buﬂ‘alo, NY, USA) and cell viability ascertained by typan blue dye
exclusion. PP lymphocytes (1 x 105 cells/250 pl well) and SP lymphocytes (5 x 10’
cells/250 [.11 well) were cultured in 96-well tissue culture plates containing 250 pl RPMI
supplemented with 10% F CS in the absence of a mitogen (unstimulated), in the presence
of the B cell mitogen Salmonella whimun'um lipopolysaccharide (LPS; 25 ug/ml), and
in the presence of the T cell mitogen concanavalin A (Con A; 5 ug/ml). After incubation

for 7 days at 37°C in a humidiﬁed 7% C02 atmosphere, culture supernatants were

56

assayed for lg concentations by ELISA.

1W. Serum and supernatant IgA, IgG, and IgM were measured using
a modiﬁcation of the ELISA previously described by Pestka et al. (1990 a). Brieﬂy,
Immulon H Removawell microtiter ships (Dynatech Laboratories, Chantilly, VA, USA)
were coated with 50/11 of goat anti-mouse IgA, IgG, or IgM (Cappel Laboratories,
Malvern, PA, USA) that had been diluted 1:1000 in 0.1 M bicarbonate buffer (pH 9.6)
and these were incubated overnight at 4 °C. Strips were then washed three times with
0.01 M phosphate buffered (pH 7.5) saline (PBS) containing 0.2% Tween 20 (PBS-
Tween). Bach well was then ﬁlled with 300 pl of 1% (w/v) bovine serum albumin (BSA)
in PBS (1% BSA-PBS) and incubated for 30 minutes at 37°C to reduce non-speciﬁc
protein binding. The strips were then washed four times with PBS-Tween. A standard
curve was prepared for each plate using mouse immunoglobulin reference serum (Bethyl
Laboratories, Montogomery, TX, USA) diluted with 1% BSA-PBS. The plates were
sealed with paraﬁlm and aluminum foil and were incubated at 37 °C for 1 hr. The plates
were then washed ﬁve times with PBS-Tween. Goat anti-mouse Ig horseradish
peroxidase (HRP) conjugate (a—, y- or a-chain speciﬁc, Cappel Laboratories, Malvern,
PA, USA) (50 ul) , diluted 1:500 in 1% BSA-PBS, was added to each well. Plates were
rescaled, incubated at 37 °C for 30 minutes, and washed six times with PBS-Tween.
Bound peroxidase was determined with 100 al/well 2,2-azino—bis (3 -ethylbenzolin-6-
sulfonate) (ABTS) substate [0.4 mM ABTS, 50 mM citrate buffer (pH 4.0), and 1.2 mM
hydrogen peroxide] and allowed to develop color at room temperatme (15 min for IgA,

5 min for IgG, and 8 min for IgM) before the reaction was stopped with 100 a1 0.3 M

57

citric acid monohydrate containing 0.02% sodium azide. The absorbance was read at 405
nm with the aid of the Molecular Devices kinetic microplate reader (Molecular Devices,
Menlo Park, CA, USA) and lg quantity determined using Softmax 2.32 version curve
ﬁtting software (Molecular Devices).

Reactivity of serum IgA and IgG to model oral- and self- antigens was assessed
by solid phase antigen binding ELISA using pre-termination serum samples. TNP-BSA
(trinitophenylated bovine serum albumin) was prepared by trinitophenylating BSA with
picric acid to yield a ratio of 53 moles TNP per mole of BSA (Good et al., 1980).
Puriﬁed casein was purchased from ICN Nutritional Biochemical. Salmon sperm DNA,
was purchased ﬁom Sigma and extacted with a phenol/chloroform mixture (Ausubel et
al., 1990). IgA-MOPC 315 was purchased ﬁom Sigma. Immulon II Removawell
microtiter strips were coated with 50 all well TNP-B SA, casein, salmon sperm double-
standed DNA, or IgA-MOPC 315 diluted to 10 ,ug/ml. TNP-BSA, casein, and IgA-
MOPC 315 were diluted with 0.5M bicarbonate buffer (pH 9.6), covered with paraﬁlm
and aluminum foil, and incubated overnight at 4°C. Salmon sperm DNA was diluted in
0.1 M PBS (pH 7.5), and dried overnight at 40°C (Rasooly and Pestka, 1992). To
determine IgA and IgG reactivity to speciﬁc antigens, ELISAs were run exactly like
those to quantify serum Igs. The only change made was exclusion of mouse
immunoglobulin reference serum to develop a standard curve, so IgA and IgG
concentations are reported as absorbance units (Rasooly and Pestka, 1994). BSA-PBS
(1%) was used to dilute reagents and block background binding in the ELISA, which

inhibits binding of BSA-speciﬁc antibodies, yielding a zero background (Klinman, 1992).

58

11mm. Urine samples were centrifuged at 450 x g for 10 minutes. The
supernatant was pipetted off, leaving only the sediment behind. The sediment was
vortexed and 50 ,ul of the suspension was pipetted onto a microscope slide. Erythrocytes
in 3 random microscopic ﬁelds (X400) were counted and averaged (Dong et al., 1991).

MW. At experiment termination,
kidneys of each euthanized mouse were also removed, cut in half and immediately ﬂown
in liquid nito gen. Each kidney was sectioned to 7pm with a cryostat (Reichert-Jung,
Cambridge Instruments, Buﬁ‘alo, NY, USA) and stained for lg or complement deposition
with F ITC-labeled goat anti-mouse IgA, IgG, or C3 (Cappel) as previously described by
Valenzuela and Deodhar (1981). Sections from each animal were viewed under a Nikon
Labophot epiﬂuorescence microscope. Fluorescent intensities of 10 glomeruli from each
section were measured using an ITM densitometric video camera (W altham, MA) and
JAVA image analysis system (Jandel Scientiﬁc, San Rafael, CA, USA). JAVA generates
a quantitative value from an encircled immunoﬂuorescent stained glomerulus in a frozen
ﬂame and calculates the average brightness for the circled area based on each pixel of the
screen included in the circle. The pixels in the circled area were measured on a grayness
scale that ranged from 0 (black) to 255 (white). Representative photographs were taken
on the same microscope using Kodak Tri-X Pan B&W ﬁlm.

W. The data were analyzed using the SigmaStat computer
program (J andel Scientiﬁc) and reported as mean :t SEM. Signiﬁcant differences
(p < 0.05) between groups were analyzed by Kruskal-Wallis and Dunn’s method for

multiple comparisons of nonparametric data. Parametric data were analyzed by the

59

AN OVA and by Student-Newman-Keul tests for multiple comparisons.

60
RESULTS

Body Weight

The body weight gain of the contol, 5 ppm VT, and 10 ppm VT groups were
monitored weekly for all three stains during their respective feeding regimens.
NZBW/Fl mice weights were monitored for 10 wks (Figure 3.1). The mean body weight
gain of the contol group was signiﬁcantly higher than either teatment groups as early as
wk 3 and remained so throughout the study. The mean body weight gain of the 5 ppm
group became signiﬁcantly higher than the 10 ppm group beginning at wk 8.

MRL/lpr mice weights were monitored for 8 wks (Figure 3.2). The mean body
weight gain of the contol group became signiﬁcantly higher than both teatment groups
at wk 2 of the feeding regimen and remained so for both groups through wk 6. At wk ‘7,
the 5 ppm treatment group was no longer signiﬁcantly different than the contol group
and only the 10 ppm teatment group remained signiﬁcantly lower than the contol group
' for the rest of the study.

BXSB mice weights were monitored for 14 wks (Figure 3.3). The mean body
weight gain of the contol, 5 ppm VT, and 10 ppm VT groups were not signiﬁcantly
diﬂ'erent during the feeding regimen except at wk 5 where the contol group was

signiﬁcantly higher than the 10 ppm VT group and at wk 6 where both the contol and 5

ppm groups were signiﬁcantly higher than the 10 ppm group.

61

Figure 3.1. Effect of subchronic vomitoxin exposure on weight gain by NZBW/Fl
female mice. Data are means :l: SE (3-7 mice/group). Data was analyzed by Student-
Newman-Keuls method following one-way analysis of variance (ANOVA). ‘Contol
group signiﬁcantly different than both teatment groups at p s 0.05. bIndicates group is
signiﬁcantly different than 10 ppm teatment group at p s 0.05.

 

WEIGHT GAIN (g)

16

14

12

O

62

 

 

 

 

 

 

 

Figure 3.1

a
_ O CONTROL
V 5 ppm
I 10 ppm 3 b
_ l
O
a/l
b
_ a a i
a
— b/ T
_ , '\ I
/ .
f l l l l l
2 4 6 8 1O

 

63

Figure 3.2. Eﬁect of subchronic vomitoxin exposure on weight gain by MRL/lpr
female mice. Data are means :1: SE (6-7 mice/group). Data was analyzed by Student-
Newman-Keuls and Dunn’s methods following one-way analysis of variance (AN OVA).
'Control group signiﬁcantly different than both teatment groups at p s 0.05. bIndicates
group is signiﬁcantly different than the 10 ppm teatment group at p s 0.05.

 

 

 

WEIGHT GAIN (g)

12

 

10—

CD
I

 

 

O CONTROL
V 5 ppm b b

 

 

 

l/i
/3/ W;
,/

 

Figure 3.2

 

65

F igtte 3.3. Effect of subchronic vomitoxin exposure on weight gain by BXSB
male mice. Data are means :r: SE (3-7 mice/group). Data was analde by Student-
Newman-Keuls method following one-way analysis of variance (AN OVA). ‘Indicates
group is signiﬁcantly different than the 10 ppm teatment group at p s 0.05. bIndicates
group is signiﬁcantly different than the other treatment group at p s 0.05.

 

 

WEIGHT GAIN (g)

12

10

(I)

66

 

 

 

O CONTROL
V’S mwn
I 10 mmn

 

 

 

 

 

Fumnﬁk3

 

67

Serum Igs

The serum Igs of the NZBW/F 1, MRL/lpr, and BXSB stains were monitored
during the 11-wk, 9-wk, and l4-wk feeding periods, respectively to determine the effect
VT exposure had on IgA, IgG, and IgM. Serum lg concentations in the NZBW/Fl
(Figure 3.4), and MRL/lpr (Figure 3.5) strains showed no signiﬁcant diﬁ‘erence between
the contol and teatment groups during their feeding regimens. Serum lg concentations
in the BXSB strain showed no signiﬁcant difference between the contol and teatment
groups except for IgM (Figure 3.6). Serum IgM in the 5 ppm VT group was signiﬁcantly
higher than both the control and 10 ppm VT groups at wk 14. The 10 ppm VT group’s
serum IgA concentation did show a two-fold increase over both the control and 5 ppm

VT groups at wk 14.

Serum Autoantibodies

lgs ﬁ'om the ﬁnal serum samples drawn from each of the three mouse stains were
' tested for their ability to bind to speciﬁc self and non-self antigens. Serum IgA and IgG
were tested for speciﬁcity to TNP-BSA, casein, and ds DNA, and serum IgG was tested
for speciﬁcity to IgA. IgG reactivity to TNP-BSA was signiﬁcantly lower in the serum
sample drawn on wk 11 from the 10 ppm group of the NZBW/F, stain compared to both
the control and 5 ppm groups (Table 3.1). In MRL/lpr mice, IgG reactivity to casein and
ds DNA as well as IgA reactivity to casein were signiﬁcantly lower in the serum sample
drawn on wk 9 ﬁom the 10 ppm group compared to both the contol and 5 ppm groups
(Table 3.2). IgG reactivity to IgA was also shown to be signiﬁcantly lower in both the 5

ppm and 10 ppm groups compared to that of the contol. The serum sample tom the

68

Figure 3.4. Effect of subchronic dietary vomitoxin exposure on serum IgG, IgA,
and IgM levels of NZBW/Fl female mice compared to contol. Data are means :h SE
(3-7 mice/group). Data was analyzed by Dunn’s method following one-way analysis of
variance (AN OVA). ‘Indicates signiﬁcantly different than other teatment group at
psODi

 

69

 

 

 

       

 

§§

 

§§
a‘\\\‘\‘\\t‘r‘m 8

 

 

 

      

S“\\L 4

‘ 1O ppnr

- CONTROL

-5T>pm
\\\\\\\

a§

SN 0

 

 

 

 

6000

O

1000 E

_
O
O
O
2

5000 -
4000 - ‘
3000 H

A

_E\@3 09

 

 

 

              

§
S“\‘L H

§
“V“Nm

                  

11

§
§\\L

\\\\\\\\\\\\\\\\\\\§
‘\\\\‘\L 8

V\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\s

V‘“‘r‘“§

§\“‘\§ 4

V‘\“\.‘L O

 

.\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\.

S\~“\r“\k O

_

 
 
 
 

 

_ _ _

 

700

600 ~

500 —

O O O O O
O O O O
8 6 4 2

_E\©3 29

1200

_
O
O
0

WEEKS

70

Figure 3.5. Effect of subchronic dietary vomitoxin exposure on serum IgG, IgA,
and IgM levels of MRL/lpr female mice compared to contol. Data are means :t SE
(7 mice/group). Data was analyzed following one-way analysis of variance (ANOVA).

 

 

 

71

 

 

       

           

 
 

IE9

§ \Eg

g9

         

 

 

   
 

 

Es §s gs
momm
mm...“ NNNR NR NNNNR
5R So R0 E0

 

 

 

 

45000

 

 

 

_ _ _
O O O O
O O 0

30000 - ‘

_ _ _ — _ _ _ _ _ _ _ P _ _
O O O O O O O O O O O O O O O O O O
O O O O O O O O O O O O O O O O
O O O O O O O 5 O 5 O 5 O 5 5 4 3 2 ..I
O 5 5 O 5 O 5 3 3 2 2 1.. 1
4 3 2 2 «I 1.
Activiv a
A_E\®3,V 09 A_E\@iv <9 _ _>_ _

WEEKS

72

Figure 3.6. Effect of subchronic dietary vomitoxin exposure on serum IgG, IgA,
and IgM levels of BXSB male mice compared to control. Data are means i SE (3-7
mice/group). Data was analyzed by Student-Newman-Keuls method following one-way
analysis of variance (AN OVA). ‘Indicates signiﬁcantly different than contol group at p
s 0.05. t’Indicates signiﬁcantly different than other teatment group at p s 0.05.

73

 

- CONTROL 1

-5T>pm

 

10 ppm

 

 

 

 

 

 

  

  

\NN‘NNNR M

    

‘NNNNNNRS. m

§§

 

30000

 

 

 

N\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

 

 

6000

 

 

   

b NNNNR4
...TIF\\\\\ \RS4 1

 

N\\\\\\\\\\\\\\\R

 

1600

74

.36 m a «a 92w 2.2585 .856 55 8ng magma? 82865..

.36 w a 3 98m .9280 55 “game 3585:»? 8332:»
.358» .2— 038 5-9 Mm a 288 2a Sons
.33 = .8.“ H> .3 “0% .268 wow 98? 8&2.

 

 

 

.93: $62 an: 6! <3
33:. :33 $33 <2
9&2 2.39 3.2.8 02 <75 a
24% as man: <2
tam: 2am: $5 03 £236
~32 38m 85m <3

3.3.2 Sag :aa 02 <3?

.3322
H> Eng 2 H> Eng m 35:00 2683— some:
$.25 .... ado

 

.85 6332 025m 5 3685532 8.55 .8 8882.. as 8 EC 5x232, been do seem .2 REE

75

.36 w a «a 9.2» 80:52 .850 55 «game 3506?»? 86835..

.36 w a 8 nap—m 35:8 55 Each? 3585:»? 833.659
.358» Be 3m:— 3 mm a 888 2a Sana
.9? m .8.“ ._.> 8 name 35:8 no.“ 295 8:2.

 

 

 

.538 35% Ra: m 03 <3
~33 3.3.32 Stems <3
3233 3.2: Seems o3 <za a
2:43 23% «N3; <3
rage Sana 33mm 03 535
:3: :«5 «$85 <3
83: 3.3% 3&3 e3 $25
@832
H> Eng 6— ._.> San m .9980 033.3% 592.2
.5 so .... ado

 

.032 25.22 255m 5 momeonuqaeedw Baum mo cameos—68m 05 so FCC 5x382; .0505 me 80mm .~.m mama?

76

.36 w a «a 32» Eon—~33 .850 05 55 EBobB 323322» «33:25.
.36 w a “a .38» 35:8 55 380.9“. magma? moagmvur
.338» .5.— oomﬁ 5.8 mm a" ESE 0.3 Sam;

.33 E he h> .8 6% 35:8 we.“ 803 8g.

 

 

 

32 San: $62 03 <3
3“: mm quotas 8am? <3

$23. 83% $33. 03 38%
333% 3&8 a 32m <3

«#2 3:: $2: 03 588
09:38 053% was: <3

8&3 was. $32 03 $329

@332
H> Egg 2 H> Eng m .8250 0.363% sewage
3.85 g. a. no

 

.032 mme 032 E momvopzcaous< Baum mo 5:335 2: so Ckc 5x815 > 505 go 80mm .m.m 033—.

77

BXSB strain drawn on wk 14 showed signiﬁcantly increased IgA reactivity to TNP-BSA
in the 5 ppm and 10 ppm groups compared to control (Table 3.3). Signiﬁcantly increased
IgA reactivity to ds DNA was also seen in the 5 ppm group compared to both the control
and 10 ppm groups. IgA reactivity to casein was also shown to be signiﬁcantly lower in

the 10 ppm group compared to both the control and 5 ppm groups.

Ig Secretion in Lymphocyte Cultures

To evaluate the effect dietary VT has on Ig isotype distribution in mucosa
associated lymphoid tissues, PP lymphocytes from treated and control NZBW/F,,
MRL/lpr, and BXSB mice were cultured for 7 days and IgG, IgA, and IgM supernatant
levels were quantiﬁed. Cultures were assessed in the presence and absence of the
mitogens LPS and Con A to identify the speciﬁc eﬂ‘ects on B and T cells, respectively.
To evaluate the eﬁ‘ect dietary VT has on systemic Ig isotype distribution, splenic
lymphocytes were tested in the same fashion.

Signiﬁcantly increased IgG and IgM was secreted by PP cultures ﬁom the 10 ppm
VT NZBW/F 1 group stimulated with LPS over both the 5 ppm VT and control groups
(Table 3.4). A signiﬁcant increase was also seen in IgM secretion from the 5 ppm VT
group stimulated with Con A over the 10 ppm VT group. In all cultures stimulated with
LPS, supernatants displayed higher Ig concentrations than unstimulated cultures or
cultures stimulated with Con A. When NZBW/Fi SP cells were cultured signiﬁcant
diﬂ‘erences were not detectable between any of the groups under any of the stimulation
conditions (Table 3.5). The results did show that IgM was the major systemic Ig isotype

in this strain. The results also showed a 6-8 fold increase in the IgM concentration of all

78

.36 w a .5 use» E32505 .850 E5 E238 3585:»? 588659
.36 w a «a use» .2289 55 Each? gamma? «88:55..

.993 s .8.“ 223383

Noo § 5 a 0.: a as»; c 82928 a 52.2 mom :2 a 333.8 2 x _ a 8530 so; as :85 E 8:: 35
8382983 :83 36>».— 332. no «Sagas.» 05 89a 355.3% mm an 2.38 25 San— ._ 055.". E costume—e 3 com 803 022.

 

 

 

$2538.? 2.32.2 «5.35.2 23

$8.335 2.385 9.352 02
3.355.: 2.3255 2.33.2 <2 9:
.3325 8.325 83:5 22
2.335 3.325 83:5 e2
8325 8.335 «5.322 <2 <80
8.385 55.385 3.325 23
2.385 :325 2.325 cm.
2.332 2.332 :33; <2 5325858:
H> Eng 2 h> 8a m 3280 258— Gomez—z

came 2

 

.852 .2332 225... do 83:5 52am £28 a 5588.5 2 5 £3 5358.5 .5505 .8 swam 55 mqmﬁ

.23 s é 2235255 N8 25 8 2 0.2 s 23: c 3&2? “_n 52.2 505 $2 a 333:3 2 x 5 a 52330 203 as 2:25
Hoe 028 h- 0 83829:». 252% no 2538995 05 89a 32:53.5 mm a 5:88 25 San ._ 035. a contact 8 no.“ 203 0.22.

 

79

 

 

3.33.: 3.3855 353555 22
8. 325 3. 325 55.325 02
8. 385 a. 385 8. 325 <2 2:
3.335 3.38.5 8.32.5 22
3.332 3.3555 3.322 02
55.332 2.3555 2.335 <2 < .80
2.385 2. 33... 2.325 22
2.3552 55.353 8.325 02
a. 335 a. 322 55.385 <2 5325558:
h> Eng 2 H> 8% m .9980 2592 522:2
2525 2

 

.83 .2332 0553 5o 52.26 can} 5 858525 2 8 EC 235:; 505 3 seem 55 355

80
the LPS stimulated cultures over that of both the unstimulated and Con A stimulated

cultures.

Increased IgA secretion was observed in unstimulated and LPS stimulated PP
cultures ﬁom the 10 ppm VT MRL/lpr group as compared to control (Table 3.6). The
LPS- stimulated cultures from the 5 ppm group also had signiﬁcantly higher IgA
concentrations then the above control group. Conversely, control PP cultures stimulated
with Con A produced signiﬁcantly more IgA then the 10 ppm VT group. A signiﬁcant
increase was also seen in IgG and IgM secretions ﬁom the LPS stimulated 5 ppm VT
group compared to both the control and 10 ppm VT groups. The results for the MRL/lpr
mice SP cultures (Table 3.7) were similar to those found with the NZBW/F1 strain. There
were no signiﬁcant diﬁemnccs between any of the groups under any of the stimulation
conditions. There was one difference between the strains though, in that the major
systemic Ig isotype of the female MRL/lpr strain is‘IgG.

In the BXSB strain, all treatment groups except the unstimulated 5 ppm VT group
were signiﬁcantly increased over control in IgA secretions (Table 3.8). In both the Con
A and LPS stimulated groups, the IgA secretion of the 5 ppm VT group was also
signiﬁcantly increased over that of the 10 ppm VT group. There were also signiﬁcant
increases in the IgG and IgM secretions of the unstimulated and Con A stimulated 10
ppm VT group compared to controls. The 10 ppm VT group was also signiﬁcantly
increased over the 5 ppm VT group for unstimulated IgG secretions. The 5 ppm VT
group also showed a signiﬁcant increase in Con A stimulated IgG and IgM secretions

compared to control. Signiﬁcant diﬂ'erences were seen in systemic IgA production of

81

.86 w a 8 use» 8553.: .850 55 85.83% $50858 838:5:
.36 w a 8 use» .253 55 38on 3:85:me 38085..

.236 N. com 803888

.8 Ox: 5 a 9.8 a 88: c 8&3? 3 =52 mom .2: a 832.8 .8 x _ a 823.8 a»: as 288 ea 88. 5-2
88.82383 :88 reckon vac—eon mo 3:5«833 05 Sat 35.58% mm um «:38 0.8 San .N 032—. 5 won—tong as com 203 022.

 

 

 

2.38.8 28.385 3.3858 82
8.32.8 2.8385 8.353 02
3.38.: an. 38.2 8.3555 <2 2:
8.38.8 8.38.8 838.8 82
8.33.8 8.32.8 8.38.8 o2
.8385 83:5 55. 38.8 <2 < 50
8.38.8 8.38.8 8.38.8 22
2.33.8 8.33.8 8.38.8 02
.3. 385 8.385 8.33.5 <2 Basses:
H> Eng 2 H> can m 85:00 0908— cowozz
852: 2

 

.022 25122 0880,.— uo 3.32:0 58am 233m 5 couosvoi 2 no 0.3 58828.5 505 .«o Seam 64m magi.

82

as 5 E 22585“ Non § 5 s 0.: a 93: c 8&2? a 3.2 Ba $2 5 :32? .2 x n .5 cease 203 as age»
009 038 5-8 moﬁoosqﬁb 3:030 mo 35808095 05 82m 3580053 um 0" 0.508 03 «an .N 035,—. E 03:00.0“. 8 v0.5 803 00:2.

 

 

 

2. 355. 55.385 3335 22

8.36.2 8.32.: £325 02

55.35? R3555 3.325 <2 2:
55.385 3.385 8.325 22

55.335 3.335 3.353 02

3.352 8.332 3.3:: <2 <80

2.33.5 3.335 2.325 22

a. 335 3.385 3.335 02

3.332 3.32.; 55. 32: <2 333858:
H> Eng 2 h> can m 35:00 2503 53:2

9525 2

 

.032 REE 0380."— .«o 000330 03015 5 530305 w— .5 C.>v 5.8388» 505 .«o 80am 5m magi.

83

.36 w a 8 anew .8558. 85c 55 896%“. magma»? 883qu

.36 w a 8 983 .838 .85 523% mun—50mins 53868..

.33 h 8% 22.8955

N8 .58 5 8 no: s 5.31 c 8823 a 985 mo". .82 5 50328 .2 x _ a 52330 22> as :85 .2. 88. 5-8
moiooaq§_ :83 9.8%.:— @208 .«o 8:52:33. 05 Bow 335.8% mm n." 888 98 85 .m 035,—. 5 eon—€83 3 v0.“ 803 3:2.

 

 

 

8.38.: 8538.8 8.38.: :3
8385 8.38.: 8.385 cm:
.8385 28. 385 8.385 <2 83
98.385 .8385 8.385 22
58.325 58.335 8.385 62
8.385 385.85.: 8.385 <2 <50
..8385 8.385 8.385 25.
2.83:5 8.385 8.385 02
.8385 8.385 2.352 <2. 838888:
H> can 3 H> Ema m .9980 25.3 smog—2
288.5 5.

 

.082 mmxm 032 .«o 35.30 :83 {0on 5 832.85 S :o CKC £583.55 505 go «comm .w.m mama;

84

.36 mm “a macaw Bog—«ob 85° 55 “manta magnum—awe 38065"
..3 a mm as anew 35:8 55 Soﬁa—v magma—m 38065..

as 5 é eoﬁSEa N8 .xs 5 a 9.: s in: c sumo? a 5.2 mom :2 a $323 é x m a 85% 203 .3 3:8»
“on 038 u- 8 moSooﬁ—Eh 032% we magma—03m 2: 89a toughen—V mm “m «:88 98 3.5 .m 035. 5 gunmen 3 com 203 022.

 

 

 

2.3;; e~.o«R.~ «332 29

$.53 8.3% 8.38.“ 02

8.33 8.33 Riga <3 9:
2.3%... 3.385 2.38 2a

2.53 2.933 2.38; we

$533 $.32; 3.38; <2 <96
2.3;; 3535 £532 23

3.33 8.32: 8.3”.” 93
22.136 «Sam: £533 «9 83358:
H> can 3 H> can m 3.980 unbomm annex—2

9833 m—

 

.ooaz mme 032 .«o 8.530 £53m E comedian 3 no but 5x339» 505 .«o Subm— dd mqmﬂw

85
unstimulated BXSB splenic cultures (Table 3.9). The 10 ppm VT group had signiﬁcantly

higher concentrations of IgA than both the control and 5 ppm VT group. The major
systemic Ig isotype of the BXSB strain is IgG except for a single aberration in the
unstimulated 10 ppm VT group where the IgA production was slightly higher than IgG

production.

Urinalysis

To determine glomerulonephritis severity, hematuria was monitored at wks 5 and
10 for the NZBW/Fl and BXSB strains and at wk 5 only for the MRL/lpr strain due to
study termination at wk 9 (Table 3.10). Erythrocyte counts in the 10 ppm VT groups of
all three strains were signiﬁcantly higher than the control groups at wk 5. The MRL/lpr
strain 10 ppm groupwasalso signiﬁcantlyhigherthanthe 5 ppm groupatwkS andthe
BXSB strain 5 ppm group was signiﬁcantly higher than the control at wk 5. NZBW/F,
and BXSB groups wk 10 hematuria measurements were too numerous to count (>80) for

all three treatment groups, making determination of signiﬁcant diﬁ‘erences impossible.

Mesangial Ig and C 3 deposition

Kidney mesangial IgA, IgG, and C3 deposition for all three mouse strains were
measured by immunoﬂuorescence microscopy. In the NZBW/F, mice, slightly decreased
IgA deposition was observed in both treatment groups compared to control (T able 3.11).
Signiﬁcantly decreased IgG deposition was fomd in the 10 ppm VT group when
compared to both the 5 ppm VT and control groups. The MRL/lpr strain also showed

slightly decreased IgA deposition in both treatment groups compared to control (Table

86

.36 w a “a 9.8» Bug 85¢ 55 8803“. baa—dogma 8332:,

.3... w a a 9.95 .288 55 53% again asses.

638838 NE? a .3 «20% 025 3 zoom moioogbm me 358 508 0.8 San;
.m c5 .m ._ 838. 5 @32on 3 B.“ 203 8:2.

 

 

8A 32 as ea 2
8A $2 {I Ea m mmxm
SA :3 380 oz:
a as? Baggage and. 5 ea 2
a as, Basque. mam". g 8% m .233
a as, Baggage 13m 380 omega
8A .mam t, 8% 2
3A «Ham .5 5.5m .EBmNz
8A 3: .880 03.5..“
...: #003 m x003 “Go-.5095. 0302
433%

 

.032 o:o£-o§8§o§< we «£395: go 9.3 £x§8o> .9305 no «oobm .2...” mama?

87
3.12). This strain also had signiﬁcantly less IgG deposition in both treatment groups
compared to control and signiﬁcantly decreased C3 deposition in the 5 ppm VT group
compared to control. In contrast to the two aforementioned strains, IgA deposition in
BXSB mice was markedly elevated in the 10 ppm VT group over both the 5 ppm VT and
control groups (Figure 3.7; Table 3.13). A signiﬁcant increase was also seen in IgG
deposition in the 10 ppm VT group compared to both the 5 ppm VT and control groups.
C3 deposition in both treatment groups was signiﬁcantly elevated over the control (Figure
3.8). The 5 ppm VT group also had a signiﬁcantly higher C3 deposition than the 10 ppm

VT group.

88

.36 w a 3 9.8» 60:58.: 65° 55 52on mango—mama «3865..
.36 w a 3 98m .8280 55 EDGE—u $5623..“ 828%59
683.38 33.682» 8m 3883 we 33
05 E vow—ﬂute ~33 32239: “on nmmé .«e 038 $3.35 a
co A0988 ..oa 3.5308 =EoEBm o 5 mm 4. 2on 0.8 8.5..
A 03am. 5 montage 3 com 203 032.

 

Qoﬂc. E. Bodﬂcée omdﬁcdm H> 8.5 c—
:«o. E. wdﬂcde ovdﬂodm H> Egan
563.3 093.8 «63.3 3.980

 

no em: <3 “5529

 

 

.83 £332 ”neon a ”.3 mo sausage Ewes»: 8 gmaxm ca unease, raga oaoassm .3 .85 a 3 Raﬁ

89

.36 w a 3 95% 35:8 55 «=95.va .3ngme «“3835»
.35308 333803 Ba 6835 we no.8
2: a Ewe»: Ea 3353 .2. $3 a. 28» maﬁa» a
no @968 “on 35808 35:83 o 3 mm a 838 2a 5.5..
.N 035. E Rim—8% 3 com 203 032.

 

 

53% 933.: 923.2 gang 2

023.: 5.96.2 0033“ .S 5%

23.3 33.: 23.2 3580
no 03 <2 “8’58;

 

 

.032 25122 oEEom E Mm— mo 52839 33532 no oamaxm 0.3 55:82, 505 oﬁoﬁﬁsm mo «comm N; ”.592.

90

.36 w a 3 98w 28830.5 .850 55 «scumbag 3:85am? 83325
.36 w a 3 95% 35:8 55 880%“. 3885:»? 830659
.02338 3:60:53 .59 3235 no 83
05 5 39826 Evan E=E>=u£ ~2— mmmé no 038 «mo—Saw a
so 332: “on 685308 230803 c: mm ﬂ 9805 03 8am.
.m 039—. E contomov ma vow 29$ 0&2.

 

 

033.3 333.2. 333$ ._.>8& 2
33.3 33.8 33.2 .5 5%
«5.2 <33}. 23.3 3580
no 03 <3 .8535

 

 

.022 988 on: a “.mzo 82839 Ewes»: 8 osmogxm EC 5338.5 505 022.3% mo .85 .Sa 33....

91

Figure 3.7. Mesangial IgA deposition after dietary VT exposure. Kidney sections
were prepared at wk 14 from male control and VT treated BXSB mice and stained with
FITC-labeled anti-mouse IgA. (a) control group (b) 10 ppm treatment group

 

93

Figure 3.8. Mesangial C3 deposition after dietary VT exposure. Kidney
sections were prepared at wk 14 from male control and VT treated BXSB mice and
stained with FITC-labeled anti-mouse C3. (a) control group (b) treatment group

 

95
DISCUSSION

VT is the trichothecene most often encountered in the US. food supply. Its
subchronic eﬂ‘ects have been studied extensively in our laboratory and others using
standard laboratory mice such as the B6C3F, strain (Rotter et al., 1996). This latter strain
is typically used in carcinogenicity and immunotoxicity studies because of its hardiness
and longevity (Cameron et al., 1985). While these traits are beneﬁcial for indicating the
toxic potential of VT, they may not be traits that accurately mirror exposures among
individuals who already have defective immune systems. One major class of immune
disorders is systemic autoimmune disease. An intriguing disorder of this class and the
one speciﬁcally modeled in this study is systemic lupus erythematosus (SLE). SLE was
chosen because it is a multi system, non-organ speciﬁc disorder, but more interestingly
because of it’s similarity to VT-induced IgA nephropathy. Both VT-induced IgA
nephropathy and SLE are immune complex diseases which reveal themselves
pathologically not as the direct eﬂ‘ect of autoantibodies to speciﬁc tissue but as immune
complexes (ICs) that are deposited in renal mesangium which leads to
glomerulonephritis. The major diﬁ‘erence between these two disorders is that VT induces
a IgA pathology whereas SLE is a IgG mediated disease.

Three murine models of SLE were used in this research. Theoﬁlopoulos and
Dixon (1985) outlined the history and notable histoimmunopathologic featm'es of SLE in
the three mm'ine models employed here. They also analyzed the serologic, histologic,

immunologic, virologic, genetic, and hormonal abnormalities that are expressed by these

96

mice. These collected data were used as a basis for comparison of the normal
characteristics seen in these mice. Two of the models used were female NZBW/F, and
MRL/lpr. Female models in SLE are important because females typically generate
stronger immune responses and thus are potentially much more susceptible to most
autoimmune diseases (Theoﬁlopoulos and Dixon, 1985). In fact, the incidence of SLE in
women is nine times greater than in men (Dubois, 1974). The other model, male BXSB
mice, was chosen not only because it is a male only model of SLE, but because previous
studies from this lab have shown a male predilection to VT-induced IgA nephropathy in
the B6C3F, mouse (Greene et al., 1994 a; Greene et al., 1994 b; Greene et al., 1995). The
ptu-pose of this research was to evaluate the eﬂ‘ect long term exposure to dietary VT
would have on autoimmune-prone mun'ne models. Based on previous studies showing
that VT exposure between 2 and 25 ppm concentration is optimal for inducing increased
IgA levels in the BGC3F, mouse (Pestka et al., 1989; Greene et al., 1994 a),
concentrations of 5 and 10 ppm were chosen. The exposure of all three strains to VT
resulted in variable effects on weight gain and the immunologic parameters tested here.
Reduced body weight gain was seen in the VT treated groups in both the female
NZBW/F, and the MRL/lpr models. These results are similar to what has been reported
earlier in the BGCBF, model (F orsell et al., 1986). The reduced weight gain seen in the
dietary VT B6C3F, model has been previously suggested to result ﬁom reduced feed
conversion emciency and feed refusal (F orsell et al., 1986; Rotter et al., 1992).
Interestingly, reduced body weight gain was not seen in VT-fed male BXSB mice. These

results are the exact opposite of what is seen in the BGC3Fl model. It is further notable

97

that, VT treated male B6C3F, mice have a greater net weight loss than females (Green et
al., 1994 a; Greene et al., 1994 b). These results show that VT had no effect on the
growth response in these mice. Mechanisms for this ﬁnding might include increased
metabolism and clearance of the toxin in BXSB mice. Theoﬁlopoulos and Dixon (1985)
reported that the average body weight for both female and male BXSB mice at 5 months
of age is 21 g. The average body weight of all three treatment groups (data not shown) of
the BXSB mice were actually 24.1 to 26.5 g at 5 months of age. These diﬁ‘erences might
be due to different diet compositions in the two studies.

Continuous exposure of the BGC3FI model to dietary VT has been shown to
dramatically increase total serum IgA levels and the polymeric/monomeric IgA ratio and
decrease total serum IgG levels (Pestka et al., 1989). This eﬁ'ect of VT posed a very
intriguing question as to what would happen in the SLE murine models whose very
pathology is IgG mediated. In all three models, IgG remained the dominant isotype in the
serum. IgA was the second most prevalent isotype in both MRL/lpr and BXSB, whereas
IgM was in the NZBW/Fl mice. The only major signiﬁcant diﬂ‘erence seen between
treatment groups was the BXSB 5 ppm VT group’s increase in serum IgM over both the
control and 10 ppm VT group at wk 14. The BXSB also showed a slight trend toward
increasedserumIgA betweenwks 11 and 14 ofthe study. Theseresultsasawhole show
that exposure to VT had very little, if any, eﬁ‘ect on the Ig isotypic proﬁle of SLE mice

and neither down regulated IgG or IgM nor upregulated IgA.

98

Subchronic exposure to dietary VT has been implicated in increased serum IgA
speciﬁc to intestinal and self antigens and decreased IgG speciﬁc antibodies (Pestka et
al., 1990 b; Rasooly and Pestka, 1992; Rasooly and Pestka, 1994). On the other hand, the
SLE-prone murine models characteristically have increased serum IgG and IgG
autoantibody levels which correlate inversely with sm'vival, and the particular IgG
isotypes (IgG 2a and 2b) predominate in the serum and IC’s (Theoﬁlopoulos and Dixon,
1985; Umland et al., 1989). My results show that VT did not hyper elevate serum IgA
speciﬁc for oral and self antigens in the female SLE models as has been reported
previously in the B6C3F 1 mouse, but did show the previously seen down-regulation of
speciﬁc IgG (Rasooly and Pestka, 1992). While the BXSB model did not exhibit down
regulation of speciﬁc IgG, it did show hyper elevated levels of antigen-speciﬁc IgA
against all antigens tested except casein. This too differs somewhat ﬁ'om the B6C3Fl
mouse, where it was reported IgA speciﬁc for casein is increased whereas IgG is
depressed (Pestka et al., 1990 b). It is important to point out that only the MRL/lpr strain
on the 10 ppm VT diet showed a signiﬁcant decrease in serum IgG to ds DNA. Elevated
serum IgG to nuclear antigens, especially ds DNA, is indicative of a SLE disease state in
both humans and mice (Tan, 1982). The BXSB strain on the 5 ppm VT diet did show a
signiﬁcant increase in serum IgA to ds DNA, but this is most likely an eﬁ‘ect of VT IgA
dysregulation on this strain.

Previous studies have shown, that VT-induced IgA hyperproduction seen in the
B6C3Fl murine model involves the secondary lymphoid organs that are associated with

both the mucosal and systemic immune systems (Pestka et al., 1990 c; Bondy and Pestka,

99

1991). Besides elevated serum IgA levels, VT has been shown to increase lymphocyte
proliferation, terminal differentiation to IgA-producing cells, and germinal center
development in PP, mesenteric lymph nodes, and SP (Bondy and Pestka, 1991; Pestka et
al., 1990 a, c). Speciﬁcally, quantitative changes were detected in the lymphocyte proﬁle
during VT exposure in the B6C3F, mice. These changes are demonstrated by increased
B cells in PP and by increases in IgA+ cells, T cells and CD4+ cells and in the CD4” :
CD8+ T cell ratio in PP and the SP (Pestka et al., 1990 a). This IgA secreting lymphocyte
p0pulation developed in the PP, later migrates to the systemic compartment of the
immune system, which includes the SP, resulting in IgA overtaking IgG as the major
isotype in the serum of B6C3F1 mice (Rasooly and Pestka, 1992).

IgG remained the dominant Ig in the serum for the experimental groups of all
three mouse strains in this study with low levels of IgA also being detected. In the PP of
NZBW/FI mice, no signiﬁcant diﬁ‘erences were seen in the unstimulated and Con A
stimulated cultures in all three treatment groups, but IgA was the primary secreted Ig.
The LPS-stimulated cultures also showed no treatment eﬂ‘ect on IgA secretion, but there
was a higher ratio ofIgA to IgG secretion. 10 ppm VT did have an eﬂ'ect on both IgG
and IgM secretion with IgG being signiﬁcantly higher than both the control and 5 ppm
VT groups and IgM having almost a 4-fold increase over both. In the LPS-stimulated PP
cultures, IgAwastheprimary secretedlgwhereasin SP, itwasIgM. Inthe PP of
MRL/lpr mice, IgA was the primary secreted Ig in all three stimulation groups. The LPS
stimulated cultures showed both the 5 ppm and 10 ppm VT groups with signiﬁcantly

elevated IgA levels over control. The Con A stimulated PP lymphocytes showed the

100

exact opposite pattern with the contol group having signiﬁcantly more IgA than the 10
ppm VT group. Once again the results in the PP cultures did not correlate with what is
seen in the SP cultures. The SP results show no signiﬁcant differences, but in all of the
cultures IgG was the primary secreted Ig. IgA levels are higher in the unstimulated and
Con A stimulated cultures, IgM was higher with LPS stimulation. While the serum in the
BXSB mouse was primarily IgG throughout the study, as with the MRL/lpr model, IgA
concentration climbed throughout the study and was always higher than IgM
concentrations. The potentiative effects of VT on Ig secretion in the PP cultures from
these mice were remarkable with effects being observed with either LPS or Con A
stimulation. Again, the potentiative eﬁ‘ects were largely not observed in SP cultures.
Taken together, these results suggest that VT did eﬁ‘ect the lymphocyte population in the
PP, especially increasing IgA secretion in both the MRL/lpr and BXSB stains, but these
effects were not apparent in the systemic immune system except for an increase in
secreted IgA in unstimulated SP cells in BXSB mice. The results further suggest that in
both the NZBW/F , and Wm mice that the VT targeted B cells whereas, the BXSB
strain showed similar effects that were described for the B6C3F, stain, where both B and
T cells were effected by VT based on mitogen stimulation results.

Hematuria was monitored as an indication of kidney damage. Previous work on
other murine stains has shown that dietary exposure to VT can result in increased
hematuria in as early as 4 wks (Greene et al., 1994 b). Both teatment groups and
contols of all three stains studied exhibited this eﬂ‘ect at wk 5. In fact, the contol group

ofallthreestainsactuallyhadhematuriagreaterthanorequaltothatoftheVTteated

101

mice at wk 4 in the work mentioned above. Nevertheless, VT teatment had a signiﬁcant
potentiative effect in these stains at wk 5. Both the NZBW/Fl and BXSB stains were
monitored again at wk 10 (the MRL/lpr mice were terminated at wk 9), the red blood
cells for all three teatment groups in both stains were too numerous to count and once
again showed hematuria greater than most stains treated with VT in the aforementioned
study. These results suggest that while severe hematuria is a natural effect in these mice,
VT exposure caused an earlier onset. Ftn'thermore, the BXSB stain is more susceptible
to the immunological eﬁ‘ects of VT even at lower doses than the other two stains.

In contast to B6C3Fl mice, NZBW/F , and MRL/lpr teatment groups exhibited
less mesangial IgA deposition than contols. Increased kidney damage was also seen in
the levels of accumulated IgA, IgG, and C3 in the glomeruli of the BXSB mouse. Dietary
VT has been previously shown to cause mesangial accumulation of IgA (Pestka et al.,
1989). The 10 ppm VT group in the BXSB stain was shown to be higher than both the 5
ppm VT and contol groups in IgA and IgG accumulation, whereas both the teatment
groups were higher in C3 accumulation than contol. It is also interesting to point out that
the 5 ppm VT group had signiﬁcantly higher C3 accumulation than the 10 ppm VT group.

Studying dietary VT exposure in the B6C3F, murine model has been important
for understanding the potential toxic eﬂ‘ects of VT in humans and animals under normal
immune conditions. The results presented here suggest that mice predisposed to SLE-like
immune dysﬁmction are diﬂ‘erentially affected by VT. While VT did not potentiate the
various SLE hallmarks such as serum immunoglobulins, it had subtle eﬂ‘ects on weight

gain, hematuria, and mesangial deposition. Further study of dietary VT exposure in

102

autoimmune- prone mice will help to deﬁne the effect VT has on this sub-population
which as seen ﬁom this study is much diﬂ‘erent than the aforementioned model.
Understanding the eﬂ‘ects of VT exposme on these autoimmune-prone models can also

provide insight into VTs mode of action by examining subtle immrme aberrations.

 

Chapter 4. Summary

 

Trichothecenes, such as VT, are sesquiterpenoid compounds that are among the
most potent inhibitors of protein and DNA synthesis known and these have been
implicated in both animal and human toxicoses (McLaughlin et al., 1977; Ueno, 1983).
Chronic dietary VT has been shown to cause a variety of toxic effects in experimental
animals including vomiting, feed refusal, and reduced body weight gain (Forsell et al.,
1986; Rotter et al., 1996).

One of the most notable immunological effects of extended VT exposure is
elevated production of IgA, which is symptomatic of an aberrant mucosal immune
response (Forsell et al., 1986). The net result of this elevated production is that IgA
becomes the major isotype in the serum. Concurrent with this is an elevation of IgA
immune complexes, spontaneous and mitogen-stimulated IgA production by isolated
lymphocytes, mesangial IgA accumulation and ﬁnally persistent glomerulonephritis
accompanied by hematuria (Dong and Pestka, 1993; Dong et al., 1991). These VT-
induced eﬂ'ects closely parallel the human autoimmrme disease known as IgA
nephropathy, which is the most common form of human glomerulonephritis worldwide
(D’Amico, 1987). The observation that dietary VT induces murine IgA nephropathy is
important because foodbome VT might be an etiological factor in the human form of the

disease.

103

104
The B6C3F, mouse has been used extensively as a model for chronic dietary VT

exposure by our research group and the toxic and immunologic effects of this exposure
have been well characterized by numerous studies. While these studies have been
beneﬁcial for indicating the toxic potential of VT, continuous dietary exposure to B6C3F.
mice may not accurately reﬂect normal human and animal exposure or be juxtaposed to
individuals who already have defective immune systems.

In the ﬁrst study of this thesis, intermittent consumption of dietary VT, as seen
with most human exposures, was assessed and compared to the eﬁ’ects seen during
continuous consumption. Reduced weight gains in the teatnent groups were seen as
early as 2-wks compared to contol. By wk 4 the intermittent group’s mean weight
became successively higher than the continuous group when it was on a clean diet
through wk 12, but returned to continuous group levels during the wks it was exposed to
VT. Serum IgA levels in the intermittent group remained at contol levels and

signiﬁcantly lower than the continuous group dming the 13-wk study, but serum IgG and
IgM levels for the intermittent group were signiﬁcantly decreased compared to contol
and thus mimicked those of the continuous group. Hematuria was signiﬁcantly greater in
both teatment groups compared to contol at wks 5 and 13 when the intermittent group
was on the VT containing diet, but hematuria in the intermittent group dissipated at wk
10 when it was on clean diet. Mesangial IgA deposition was signiﬁcantly lower in the
intermittent group compared to the continuous group and had levels comparable to mice
on the contol diet. The results suggested that intermittent exposure to VT had reduced

toxic effects in B6C3F1 mice as compared to the continuous model. These results are

105

important because the B6C3F 1 intermittent VT dietary model might more closely
resemble human VT exposure.

The purpose of the second study conducted in this thesis was to characterize the
effects of subchronic dietary VT exposure on immunologic parameters associated with
murine models of SLE and to determine what effect genetic predisposition to a systemic
autoimmune disease has on previously documented effects of VT induced IgA
nephropathy. Using autoimmune-prone NZBW/F ,, MRL/lpr, and BXSB mouse stains as
models, I assessed how the consumption of diet containing 5 ppm or 10 ppm VT would
affect these mice. Reduced weight gains were seen in NZBW/F , and MRL/lpr mice fed
both doses of VT within 2-3 wks. In contast, VT had little effect on BXSB weight gain.
Serum Ig levels in all three strains generally did not differ ﬁ'om contol levels throughout
the study. Hematuria was signiﬁcantly increased when all three stains were fed 10 ppm
VT. In the NZBW/Fl Peyer’s patch cultures stimulated with LPS, prior exposure to 10
ppm VT signiﬁcantly increased the IgG and IgM secretion but had no effect on IgA. In
MRL/lpr Peyer’s patch cultures stimulated with LPS, exposure to 10 ppm VT increased
IgA secretion but not IgM or IgG. BXSB Peyer’s patch cultures prepared ﬁom both VT
teatment groups produced signiﬁcantly more IgA than contol when stimulated with LPS
or Con A. Mesangial deposition of IgA and IgG was signiﬁcantly lower in the teatment
groups of NZBW/F , and MRL/lpr mice compared to contol. BXSB mice had
signiﬁcantly higher IgA and IgG deposition in the 10 ppm VT group as compared to both
the 5 ppm VT and contol groups and both VT teated groups had more complement (C3)

deposition than contol. The results suggest that mice predisposed to SLE-like immune

106

dysfunction are differentially affected by VT. While VT did not potentiate the various
SLE hallmarks such as serum Igs, it had subtle effects on weight gain, hematuria, and
mesangial Ig and C3 deposition.

Both studies in this thesis show that innovative approaches need to be developed
in order to fully understand the tue effects of human VT exposure. Epidemiology
studies need to be conducted to assess what a human exposure consists of on average.
Other models could be developed to reﬂect the variety in individuals that exist in our and
other animal populations, especially with traits that might exacerbate effects seen with
normal VT exposure.

Several follow-up studies need to be performed on the two models covered in this
thesis. First, a pair-fed B6C3F, mouse contol group needs to be incorporated into the
intermittent study to determine if the effects found in the previous study were the result of
VT exposure or reduced feed consumption. Secondly, other effects of dietary VT
exposure that have been previously reported in the continuous model need to be tested
for in the intermittent model to better understand the differences and similarities between
the two models. The autoimmune-prone murine models need to have a full scale dose
response study performed to determine the optimal concentation of VT to induce
pathologic and immrmotoxicologic changes over contols. These autoimmune-prone
murine models should then be fully studied in the same manner as the continuous and
intermittent models were. Finally, both of the models described in this thesis should be
combined together to see what effects intermittent VT exposure has on mice predisposed

to SLE.

LISIQEBEEERENCES

Abouzied M. M., Azcona J. I. , Braselton W.E. and Pestka J. J. (1991) Assessment of
mycotoxins in 1989 grain foods by monoclonal antibody based enzyme-linked
immunosorbent assay: evidence for deoxynivalenol (vomitoxin) contamination.
Applied and Environmental Microbiology 57, 672-677.

Ausubel F.A., Brent R., Kingston RE, Moore D.D., Seidman J .G., Smith J.A. and
Stuhl K. (Editors) (1990) Current Protocols in Molecular Biology. pp. 2.1.1.
Greene Publishing and Wiley-Interscience, New York.

Azcona-Olivera J. I., Ouyang Y.-L., Warner R. L., Linz J. E., and Pestka J. J. (1995)
Effects of vomitoxin (deoxynivalenol) and cyclohexamide on IL-2, 4, 5, and 6
secretion and mRN A levels in murine CD4+ cells. Food Chem T oxicol. 33:433-
441.

Bamburg J. R. (1983) Biological and biochemical actions of trichothecene mycotoxins.
In Progress in Molecular and Subcellular Biology. Edited by F. E. Hahn pp. 41-

110. Springer-Verlag, Berlin.

Betina V. (1989) Stucture-activity relationships among mycotoxins. Chem. Biol.
Interact. 71:105-146.

Bienenstock J. And Befus AD. (1980) Mucosal immunology. Immunology 41(2): 249-
270.

Bondy G. S. and Pestka J. J. (1991) Dietary exposure to the tichothecene vomitoxin
(deoxynivalenol) stimulates terminal diﬁ'erentiation of Peyer’s patch B cells to
IgA secreting plasma cells. T oxicol. Appl. Pharmacol. 108, 520-530.

Bradlaw J. A. , Swentzel K. C., Alterman E., and Hauswirth J. W. (1985) Evaluation of
puriﬁed 4—deoxynivalenol (vomitoxin) for unscheduled DNA synthesis in the
primary rat hepatocyte-DNA repair assay. Food Chem. Toxicol. 23:1063-1067.

Bunner D. L. and Morris E. R (1988) Alteration of multiple cell membrane ﬁmctions in
L6 myoblasts by T-2 toxin: an important mechanism of action. Toxicol. Appl.
Pharmacol. 92:113-121.

Cameron T. P., Hickman R. L., Komreich M. R., and Tarone R. E. (1985) History,
survival, and growth patterns of B6C3F1 mice and F344 rats in the National
Cancer Institute carcinogenesis testing program. Fundamental and Applied
Toxicology 5, 526—538.

107

108

D’Amico G. D. (1987) The commonest glomerulonephritis in the world: IgA
nephrOpathy. Quarterly Journal of Medicine 64, 709-727.

Dong W. and Pestka J. J (1993) Persistent dysregulation of IgA production and IgA
nephropathy in the B6C3Fl mouse following withdrawal of dietary vomitoxin
(deoxynivalenol). Fundamental and Applied Toxicology 20, 38-47.

Dong W., Sell J. E., and Pestka J. J. (1991) Quanitative assessment of mesangial
immunoglobulin A (IgA) accumulation, elevated circulating IgA immune
complexes, and hematuria during VT-induced IgA nephropathy. Fundam. Appl.
Toxicol. 17, 197-207.

Dubois E. L. (1974) The clinical picture of systemic lupus erythematosus. In Lupus
Erythematosus. Edited by E. L. Dubois. 2nd ed., pp. 232-242. Univ. Of Southern
California Press, Los Angeles.

Fossati L., Iwamoto R., Merino R., and Izui S. (1995) Selective enhancing eﬂ‘ect of the
Yaa gene on immune responses against self and foreign antigens. J Immunol.
25: 166-173.

Forsell J. H., Jensen R., Tai J.-H., Witt M., Lin W. S., and Pestka J. J. (1987) Comparison
of acute toxicities of deoxynivalenol (vomitoxin) and 15-acetyl deoxynivalenol in
the B6C3F, mouse. Food Chem. Toxicol. 25:155-162.

Forsell J. H., Witt M. F ., Tai J-H., Jensen R. and Pestka J. J. (1986) Eﬁ‘ects of 8-week
exposure of the B6C3F1 mouse to dietary deoxynivalenol (vomitoxin) and
zearalenone. Fd Chem. Toxic. 24, 213-219.

Good A.H., Wofsy L., Henry L. And Kimora J. (1980) Preparation of hapten-modiﬁed
protein. In Selected Methods in Cellular Immunology. Edited by BB. Mishell and
S. M. Shiigi. pp. 345-347. Freeman, San Francisco, CA.

Greene D. M., Azcona-Olivera J. I. and Pestka J. J. (1994a) Vomitoxin (deoxynivalenol)-
induced IgA nephropathy in the B6C3F1 mouse: dose response and male
predilection. Toxicology 92, 245-260.

Greene D. M., Azeona-Olivera J. 1., Murtha J. M., and Pestka J. J. (1995) Eﬁ‘ects of
dihydrotestosterone and estadiol on experimental IgA nephropathy induced by
vomitoxin. Fundamental and Applied Toxicology 26, 107-116.

Greene D. M., Bondy G. S., Azcona J. I., and Pestka J. J. (1994b) Role of gender and
stain in vomitoxin-induced dysregulation of IgA production and IgA nephropathy
in the mouse. Journal of Toxicology and Environmental Health 43, 37-50.

109

Jelinek C.F., Pohland A.E., and Wood GE. (1989) Review of mycotoxin contamination.
Worldwide occurrence of mycotoxins in foods and feeds. An update. J: Assoc.
017? Anal. Chem. 72:223-230.

Jone C., Erickson L., Trosko J. E., and Chang C. C. (1987) Eﬂ‘ect of biological toxins on
gap-junctional intercellular communication in chinese hamster V79 cells. Cell
Biol. T oxicol. 3:1-15.

Kagnoﬁ‘ M. F. (1981) Immunology of the digestive system. In Physiology of the
Gastrointestinal Tract. Edited by L. R Johnson. Raven Press, New York.

Klinman D. M. (1992) Similarities in B cell repertoire development between autoimmune
and aging normal mice. Journal of Immunolog 148, 1353-1358.

Marooka N., Uratsuji N., Yoshizawa T., and Yamamoto H. (1972) Studies on the toxic
substances in barley infected with Fusarium spp. J. Food Hjig. Soc. Jpn. 13:368-
375.

McLaughlin C. S., Wei C. M., Vaughan M. H., Stafford M. E., Campbell I. M., and
Hansen B. S. (197 7) Inhibitions of protein synthesis by tichothecenes. In
Mycotoxins in Human and Animal Health. Edited by J. V. Rodricks,

C. W. Hesseltine, and M. A. Mehlman. pp. 263-273. Chem-Orbital, Park Forest
South, IL.

Mestecky J. (1987) The common mucosal immrme system and current stategies for
induction of immune responses in external secretions. J. Clin. Immunol. 7 :265-
276.

Miller K. And Atkinson H. A. C. (1986) The in vitro eﬁ‘ects of tichothecenes on the
immune system. Food Chem. T oxicol. 24:545-549.

Miller M. L., Raveche E. S., Laskin C. A., Klinman D. M., and Steinberg AD. (1984)
Genetic studies in NZB mice. VI. Association of autoimmune taits in
recombinant inbred lines. J. Immunol. 133:1325-1331.

Mizel S. B. and Mizel D. J. ( 1981) Puriﬁcation to apparent homogeneity of murine
interleukin 1. J. Immunol. 126:834-837.

Murphy E. D. And Roths J. B. (1979) Autoimmunity and lymphoproliferation: Induction
by mutant gene lpr and acceleration by a male-associated facter in stain BXSB
mice. In Genetic Control of Autoimmune Disease. Edited by N. R. Rose, P.E.
Bigazzi, and N. L. Warner. pp. 207-220. Elsevier, Amsterdam.

110

Ouyang Y. -L., Azcona-Olivera J. I., Murtha J ., and Pestka J. J. (1996a) Vomitoxin-
mediated IL-2, IL-4, and IL-5 superinduction in murine CD4+ T cells stimulated
with phorbol ester and calcium ionophore: relation to kinetics of proliferation.
Toxicol. Appl. Pharmacol. 138:324-334.

Ouyang Y.-L., Li S., and Pestka J. J. (1996b) Effects of vomitoxin (deoxynivalenol) on
tanscription facter NF-xB/Rel Binding activity in murine EL-4 thymoma and
primary CD4+ T cells. T oxicol. Appl. Pharmacol. 140:328-336.

Pestka J. J. (1988) Enhanced surveillance of foodbome mycotoxins by immunochemical
assay. Journal of the Association of Oﬁ‘icial Analytical Chemists. 71(6): 1075-
1 08 l .

Pestka J. J. (1997) In response to an article published previously in MIU, Vol. 3, No. 3:
mechanisms of trichothecene-induced murine IgA nephropathy. Mucosal
Immunology Update. September 1997:49-52.

Pestka J. J ., and Bondy G. S. (1990) Alteration of immune function following dietary
mycotoxin exposure. Can. J. Physiol. Pharmacol. 66, 1009-1016.

Pestka J. J ., and Bondy G. S. (1994) Mycotoxin-induced immune modulation. In
Immunotoxicology and Immunopharmacology. Edited by J. H. Dean,
M. I. Luster, A. E. Munson, and I. Kimber. pp. 163-182. Raven Press, Ltd.,
New York.

Pestka J. J. and Casale W. (1990) Naturally occuring fungal toxins. Food Contamination
from Environmental Sources. 613-638.

Pestka J. J. and Dong W. (1994) Progressive serum IgE hyperelevation in the B6C3F 1
mouse following withdrawal of dietary vomitoxin (deoxynivalenol). Fundam.
Appl. Toxicol. 22:314-316.

Pestka J. J., Dong W., Warner R. L., Rasooly L. and Bondy G. S. (1990 c) Eﬁ‘ect of
dietary administation of the trichothecene vomitoxin (deoxynivalenol) on IgA
and IgG secretion by Peyer’s patch and splenic lymphocytes. Food and Chemical
Toxicology 28, 693-699.

Pestka J. J ., Dong W., Warner R. L., Rasooly L., Bondy G. S. and Brooks K. H. (1990 a)
Elevated membrane IgA“ and CD4” (T helper) populations in murine Peyer’s
patch and splenic lymphocytes during dietary administation of the trichothecene
vomitoxin (deoxynivalenol). Food and Chemical Toxicology 28, 409-420.

111

Pestka J. J ., Moorman M. A., and Warner R. L. (1989) Dysregulation of IgA production
and I gA nephropathy induced by the trichothecene vomitoxin. Food and
Chemical Toxicology 27, 361-368.

Pestka J. J. Moorman M. A., and Warner R. L. (1990 b) Altered serum immunoglobulin
response to model intestinal antigens during dietary exposure to vomitoxin
(deoxynivalenol). Toxicology Letters 50, 75-84.

Pestka J. J ., Tai J.-H., Witt M.F., Dixon DE. and Forsell J .H. (1987) Suppression of
immune response in the B6C3Fl mouse after dietary exposure to the Fusarium
mycotoxins deoxynivalenol (vomitoxin) and zearalenone. Food and Chemical
Toxicology 25, 297-304.

Pestka J .J ., Yan D., and King LE. (1994) Flow cytometric analysis of the eﬂ'ects of in
vitro exposure to vomitoxin (deoxynivalenol) on apoptosis in murine T, B and
IgA“ cells. Food and Chemical Toxicology 32, 1125-1136.

Rasooly L., Abouzied M. M., Brooks K. H., and Pestka J. J. (1994) Polyspeciﬁc and
autoreactive IgA secreted by hybridomas derived from Peyer’s patches of
vomitoxin-fed mice: Characterization and possible pathogenic role in IgA
nephropathy. Food Chem. T oxicol. 32:337-348.

Rasooly L. and Pestka J. J. (1992) Vomitoxin-induced dysregulation of serum IgA, IgM
and IgG reactive with gut bacterial and self antigens. Food and Chemical
Toxicology 30, 499-504.

Rasooly L. and Pestka J. J. (1994) Polyclonal autoreactive IgA increase and mesangial
’ deposition dming vomitoxin-induced IgA nephropathy in the BALB/c mouse.
Food and Chemical Toxicology 32, 329-336.

Raveche E. S., Steinberg A. D., Klassen L. W., and Tjio J. H. (1978) Genetic studies in
NZB mice. 1. Spontaneous autoantibody production. J. Exp. Med. 147:1487-
1502.

Robbana-Barnat S., Lafarge-Frayssinet C., Cohen H., Neish G.A., and Frayssinet C.
(1988) Immuno-suppressive properties of deoxynivalenol. Toxicology 48: 1 55-
166.

Roitt 1., Brostoff J ., and Male D. (1989) Molecules which recognize antigen. In
Immunology. pp. 5.1-5.12. The C. V. Mosby Company, St. Louis.

Rotter B. A., Prelusky D. B., and Pestka J. J. (1996) Toxicology of Deoxynivalenol
(V omitoxin). Journal of Toxicology and Environmental Health 48, 1-34.

112

Rotter B. A., Rotter R. 6., Thompson B. K, and Trenholm H. L. (1992) Investigations in
the use of mice exposed to mycotoxins as a model for growing pigs. Journal of
Toxicology and Environmental Health 37, 329-339.

Rotter B. A., Thompson B. K., Clarkin S., and Owen T. C. (1993) Rapid colorimetric
bioassay for screening of F usarium mycotoxins. Natural Toxins. 1:303-307.

Scott F. W., Cui J ., and Rowsell P. (1994) The role of food in autoimmunity. Trends
Food Sci. T echnol. 5, 111-116.

Scott RM. (1989) The natural occurrence of trichothecenes. In Trichothecene
Mycotoxicoses: Pathophysiologic Eﬁ’ects. Edited by V.R. Beasley. Vol I., pp.
1-26. CRC Press, Boca Raton, FL.

Scott RM. (1991) Possibilities of reduction or elimination of mycotoxins present in
cereal grains. In Cereal Grain. Mycotoxins, Fungi, and Quality in Drying and
Storage. Edited by J. Chelkewski. pp.529-572. Elsevier, Amsterdam.

Shah D. B., Kaumnan P.E., Boutin B. K., and Johnson CH. (1982) Detection of heat-
labile-enterotoxin-producing colonies of Escherichia coli and Vibrio cholerae by
solid-phase sandwich radioimmunoassays. J. Clin. Microbiol. 16:504-508.

Steinberg A. D. (1994) Systemic Lupus Erthematosus: Theories of pathogenesis and
approach to therapy. Clin. Immunol. Immunopath. 72(2): 171-176.

Tamrn C. (1977) Chemistry and biosynthesis of trichothecenes. In Mycotoxins in Human
and Animal Health. Edited by J. V. Rodricks, C. W. Hesseltine, and M.A.
Mehlman. pp. 209-228. Chem-Orbital, Park Forest South, IL.

Tan E. M. (1982) Autoantibodies to nuclear antigens (ANA): their immunobiology and
medicine. Adv. Immunol. 33, 167.

Tanaka T., Hasegawa A., Yamamoto S., Lee U.S., Sugiura Y. And Ueno Y. (1988)
Worldwide contamination of cereals by the Fusarium mycotoxins nivalenol,
deoxynivalenol, and zearalenone. Survey of 19 counties. Journal of Agriculture
and Food Chemistry 36, 979-983.

Theoﬁlopoulos A. N., and Dixon F. J. (1985) Murine models of systemic lupus
erythematosus. Adv. Immunol. 37:269.

Thompson W. L. and Wannemacher R. W., Jr. (1984) Detection and quanitation of T-2
mycotoxin with a simpliﬁed protein synthesis inhibition assay. Appl. Environ.
Microbiol. 48, 1 176-1 180.

113

Tryphonas H., Iverson F., So Y., Nera E. A., McGuire P. F., O’Grady L., Clayson D. B.,
and Scott P. M. (1986) Effects of deoxynivalenol (vomitoxin) on the humoral and
cellular immunity of mice. T oxicol. Lett. (Amsterdam), 30:137-150

Tryphonas H., O’Grady L., Arnold D. L., McGuire P. F., Karpinski K., and Vesonder
R. F. (1984) Effect of deoxynivalenol (vomitoxin) on the humoral immunity of
mice. T oxicol. Lett. 23:17-24.

Ueno Y. (Editor). (1983) General Toxicology. In T richothecenes: Chemical, Biological
and Toxicological Aspects. p. 135-146. Elsevier, New York.

Ueno Y. (1985) The toxicology of mycotoxins. In CRC Critical Reviews in Toxicology.
Edited by L. Goldberg. pp. 99-132. CRC Press, Boca Raton, FL.

Umland S. P., Go N. F., Cupp J. E., and Howard M. (1989) Responses of B cells from
autoimmune mice to IL-5. Journal of Immunolog 142(5) 1528-1535.

Valenzuela R., and Deodhar S. D. (1981) Interpretation of Immunoﬂuarescent Patterns in
Renal Disease, pp. 3-11. American Society of Clinical Pathology, Chicago, IL.

Vesonder R. F ., Ciegler A., and Jensen A. H. ( 1973) Isolation of the emetic principle
ﬁ'om F usarium-infected corn. Appl. Environ Microbiol. 26:1008-1010.

Vesonder R. F., Ciegler A., Jensen A. H., Rohwedder W. K., and Weisleder D. (1976).
Co-identity of the refusal and emetic principle from Fusarium infected corn.
Appl. Envir. Microbiol. 31, 280.

Vesonder R. F., Ciegler A., Rogers R. F., Burbridge K. A., Borhast R. J., and
Jensen A. H. (1978) Survey of 1977 crop year preharvest corn for vomitoxin.
Appl. Environ. Microbiol. 36:885-888.

Warner R. L., Brooks K., and Pestka J. J. (1994) In vitro eﬁ‘ects of vomitoxin
(deoxynivalenol) on T-cell interleukin production and IgA secretion. Food Chem.
T oxicol. 32:617-625.

Witt M., Hart LP. and Pestka J .J . (1985) Puriﬁcation of deoxynivalenol by water-
saturated silica gel chromatography. Journal of Agricultural and Food Chemistry
33, 745-748.

”‘ﬁlillllllllllltttlllES