TOXICOLOGICAL SCIENCES 73, 362–377 (2003)
DOI: 10.1093/toxsci/kfg074
Copyright © 2003 by the Society of Toxicology
Immunotoxicity of Aflatoxin B 1 in Rats: Effects on Lymphocytes and
the Inflammatory Response in a Chronic Intermittent Dosing Study
Dennis M. Hinton,* ,1 Michael J. Myers,† Richard A. Raybourne,* Sabine Francke-Carroll,* Rene E. Sotomayor,*
Joseph Shaddock,‡ Alan Warbritton,‡ and Ming W. Chou‡
*United States Food and Drug Association, Center for Food Safety and Applied Nutrition, Laurel, Maryland 20708, †United States Food
and Drug Association, Center for Veterinary Medicine, Laurel, Maryland 20708, ‡United States Food and Drug Association,
National Center for Toxicological Research, Jefferson, Arkansas 72079
Received November 26, 2002; accepted February 21, 2003
We investigated the effects of aflatoxin B 1 (AFB 1) on isolated
splenic lymphocytes and the histo-morphologic changes in the
spleens and liver of Fisher-344 male rats. Weaned animals were
fed chow diets that contained 0, 0.01, 0.04, 0.4, or 1.6 ppm AFB 1,
using an intermittent dosing regimen (4 weeks on and 4 weeks off
AFB 1), for 40 weeks. An additional group of animals was fed the
1.6 ppm AFB 1 diet continuously . The intermittent dosing regimen
was designed to evaluate effects of cumulative dose and exposure
for risk assessment comparisons. The percentages of T and B cells
were affected as shown by flow cytometric analysis after the dosing
cycles. The observed changes appeared to reverse or compensate to
some extent after the off cycles. Lymphocytes were stimulated in
culture for analysis of the production of IL-2, IL-1, and IL-6.
Significantly increased production of IL-1 and IL-6 was seen in the
second dosing cycle (12 weeks) and the second “off” cycle (16
weeks) at the higher doses. Inflammatory infiltrates were seen in
the liver after eight weeks of continuous and intermittent dosing
and were increased in size and number at 12 weeks in both 1.6
ppm dose groups correlating with the peak production of Il-1 and
IL-6. We concluded that AFB 1 effects on the immune system can
be either stimulatory or suppressive dependent on a critical exposure window of dose and time. Immune cells in spleen such as
T-lymphocytes and macrophages, both important mediators of
inflammatory responses to tissue damage, were affected differently
in the continuous and intermittent exposures to AFB 1.
Key Words: aflatoxin B 1; immunotoxicity; inflammatory response; intermittent dosing.
Aflatoxin B 1 (AFB 1), a secondary metabolite of the fungus
Aspergillus flavus, is a hepatocarcinogen in various animal
species, including fish, birds, rodents, and nonhuman primates
(Wogan, 1976, 1992, 2000). It is also a suspect human carcinogen and has been shown to play a role in human hepatocarcinoma (Dominguez-Malagon and Gaytan-Graham, 2001;
Wang et al., 2001). We present herein part of our immunotox1
To whom correspondence should be addressed at CFSAN, US FDA,
Module 1 Research Laboratories, Laurel, MD 20708. Fax: (301) 594-0517.
E-mail: [email protected].
362
icity study, which complements a larger, collaborative effort,
designed to assess potential biomarkers that may have a role in
the initiation and promotional stages of carcinogenesis (Morris
et al., 1999; Sahu et al., 1999; Sotomayor et al., 1999), and
which may be relevant in the “risk assessment” processs. In
particular, we were interested in the effects of AFB 1 on the
cells and mechanisms of cell mediated immunity (CMI), since
this had been implicated as the immune target (Pier et al.,
1977; Sharma, 1993), in relation to carcinogenesis. We were
interested also in any possible role of the immune system in
hepatotoxicity via the involvement of the inflammatory process. It is known that inflammatory mechanisms can cause liver
damage, e.g., in the case of alcohol-induced hepatitis (Batey
and Wang, 2002) and with 2,3,7,8-tetrachloro-dibenzo-p-dioxin (TCDD; Rosenthal et al., 1989). Batey and Wang (2002)
demonstrated the role of liver-associated T-lymphocytes in the
pathogenesis of alcohol related liver injury initiated by a variety of stimuli such as endotoxin (LPS), or Con-A. To our
knowledge there are only early reports (Butler, 1970) of associated inflammatory responses with the hepatotoxic effects of
AFB 1, mainly in chickens and turkeys. We hypothesized that
an inflammatory response could be associated with the hepatotoxic effects of AFB 1 in the rat. However, if the immune
system were also suppressed by AFB 1, to what extent would
this affect the inflammatory response?
Our objective was to evaluate any significant changes in the
relative proportions and functions of the main splenic lymphocyte classes and to compare these changes with the histopathology evaluations of the liver and spleen with emphasis on
the cell populations involved in the inflammatory response. An
intermittent exposure regimen was designed to simulate human
experience (Kodell et al., 1987; Murdoch et al., 1992) since
people are often exposed to an agent intermittently rather than
continuously. The focus of this study was how accumulated
dosing relates to the expression of particular immunologic
biomarkers and the formation of preneoplastic lesions in the
liver.
A considerable body of evidence exists suggesting that
IMMUNOTOXICITY STUDY OF AFLATOXIN B 1 IN F344 RATS
AFB 1 suppresses immune function by affecting T-cell dependent immunity in various animal species, in particular, cattle
(Bodine et al., 1984; Brown et al., 1981), chickens and turkeys
(Ghosh et al., 1990; Giambrone et al., 1985a,b), and swine (Liu
et al., 2002; Mocchegiani et al., 1998). Studies with laboratory
test species such as the mouse (Jakab et al., 1994; Reddy et al.,
1987), rat (Raisuddin et al., 1990, 1993), and rabbit (Venturini
et al., 1990) reinforce these findings. Immunosuppression by a
toxicant can result from various mechanisms such as decreased
protein and/or DNA synthesis, changes or loss in enzymatic
activity, and changes in metabolism or cell cycles, which may
result in apoptosis or necrosis. Immune mechanisms affected
by AFB 1, in addition to T-cell dependent immunity, include
reduced production of complement by the liver and decreased
phagocytosis by neutrophils and macrophage (Cusumano et al.,
1995, 1996; Dugyala and Sharma, 1996). Toxic effects on
T-lymphocytes (Dugyala and Sharma, 1996) and/or other lymphoid cells such as the cytotoxic T-cells and natural killer cells
(NK; Methenitou et al., 2001), which impair the function of
direct or indirect killing of tumor cells, can have pronounced
effects on tumorigenesis. Immunosuppression can result in a
greater rate of tumor progression (Raisuddin et al., 1991).
Moreover, cellular components of the immune system are
known to produce various cytokines, which play a key role in
host resistance and protection against tumor progression. These
same cytokines, however, are involved directly in the inflammatory mechanisms that are initiated when various organs have
been damaged by toxic assault (Batey and Wang, 2002.)
In this study, we present the results of the flow cytometric
analysis and assays of the functional, inflammatory cytokine
productive capacity of splenic lymphocytes in relation to the
histopathology evaluation of the liver and spleen for the highest dose groups of AFB 1. Our results complement the various
immune function studies of AFB 1 that have been reported,
since they relate to both immunotoxic and possibly the hepatotoxic effects of AFB 1. As far as we are aware, this is the first
report of an immunotoxicity study of AFB 1 wherein cycles of
feeding and rest were included in the study design.
MATERIALS AND METHODS
Chemicals. The following reagent grade chemicals were obtained from
Sigma Chemical Co. (St. Louis, MO): aflatoxin B 1 (CAS No 1162-65-8);
L-glutamine; gentian violet; glacial acetic acid; sodium chloride; and anhydrous and monohydrated sodium phosphates, used for white blood cell (WBC)
counts; and Histopaque ®, used for splenic lymphocyte purification. Wright’s
stain for the WBC differential counts and ethylenediaminetetraacetic acid
(EDTA) were obtained from Fischer Scientific Co. (Pittsburgh, PA). Ultraculture ® media, used for the splenic cell isolates, was obtained from BioWhittaker (Walkersville, MD). Alamar blue dye, used in the cytokine production
assays, was obtained from Alamar Biosciences, Inc. (Sacramento, CA). All
solutions were prepared using sterile, deionized, and distilled water.
Fluorescent antibodies, cell lines, and cytokine standards. All of the
fluorescent labeled antibodies used in the flow cytometry were obtained from
Pharmingen (San Diego, CA). These included: fluorescein isothiocyanate
(FITC)-labeled monoclonal antibodies to rat CD8a and rat CD45R, and the
363
FITC mouse IgG2a (used as a kappa isotype control); and phycoerythrin
(PE)-labeled monoclonal antibodies to rat CD3 and rat CD4 as well as PE
labeled mouse IgG1 used as the kappa isotype control.
The cytokine responsive cell lines, CTLL-2 and 7TD1, used in the splenic
lymphocyte stimulation assays were obtained from the American Type Culture
Collection (ATCC; Manassas, VA). The CTLL-2 line (ATTC number TIB214) is a mouse cytotoxic T-cell line, which was used in the IL-2 cytokine
assays as described by Lyte et al. (1987). The 7TD1 cell line, ATCC number
CRL-1851, is a mouse B-lymphocyte hybridoma that was used in the IL-6
cytokine assays. A D10.S cell line, a gift from Dr. Lawrence Shook of the
University of Illinois, was used in the IL-1 cytokine assays in a manner
described by Schook et al. (1992). The D10.S cell line is a mouse helper/
inducer T-lymphocyte line that is a sub-clone of the D10.G4.1 cell line, which
is commercially available from ATCC, number TIB-224.
Cytokine standards, human IL-1, IL-2, and IL-6 were obtained from BioSource (Camarillo, CA). Carrier free rat INF␥ was obtained from PBL Biomedical Laboratories (New Brunswick, NJ). Concanavalin A (Con-A) and
chromatographically pure Eschericia coli (E. coli) lipopolysaccharide, containing less than 1% protein, were obtained from Sigma Chemical Co.
Animals, diet, and study design. The life phase of the study was conducted at the National Center for Toxicological Research (NCTR) using
weaned male Fischer 344-N (F344) rats (21–24 days of age) obtained from the
NCTR breeding colony. The NCTR is fully accredited by the American
Association for Accreditation of Laboratory Animal Care (AAALAC). Animal
husbandry and all experimental procedures were reviewed and approved by the
NCTR Animal Care and Use Committee. The rats were maintained on a 12-h
light/dark cycle at a constant temperature of 22–24°C and humidity of 34 –
37%.
The control animals were fed certified NIH-31 meal diet (Purina, St. Louis,
MO), and the treated groups were fed AFB 1-NIH-31 diets that were prepared
at NCTR’s Dietary Preparation Facility. The AFB 1-NIH-31 meal diets were
prepared by mixing 40 kg of the NIH-31 meal with 64 mg of AFB 1 dissolved
in 500 ml of ethanol to obtain the high dose of 1.6 ppm aflatoxin. The ethanol
was removed by evaporation under reduced pressure. All other doses were
obtained by admixing the appropriate amount of NIH-31 meal with the high
dose AFB 1-NIH-31 preparation. The concentration of AFB 1 in each diet
preparation was measured by the method of Park et al. (1990). Separate groups
of rats were used to estimate food consumption. Four rats per dose group were
housed singly in hanging cages. Spilled food was collected and weighed. Food
consumption was corrected by adding the weight of wasted food to the weight
difference before and after feeding. From these data the amount of ingested
AFB 1 was calculated.
The animals were housed singly initially and provided NIH-31 diet and
water ad libitum for one week. After this acclimation period, the four-week-old
animals were randomly allocated to control and experimental groups. During
the study the animals were housed in pairs and were provided the appropriate
diets and water ad libitum. There were six dose groups consisting of animals
fed chow diet mixed with either 0.0, 0.01, 0.04, 0.40, or 1.6 ppm of AFB 1. The
study design is shown in Figure 1 and has been previously described (Morris
et al., 1999) for 20 weeks of the 40-week feeding study. Briefly, experimental
groups were fed diets containing AFB 1 for four weeks, then they were fed
chow diet without AFB 1 for another four weeks. The “on diet”/“off diet”
cycles, referred to as “intermittent” dosing, were continued up to 40 weeks.
One other group was included in the study design; this group of animals
received the 1.6 ppm diet continuously for 40 weeks.
For the immunotoxicity study, sets of 30 animals consisting of five male rats
per dose group (intermittent doses, 0.0, 0.01, 0.04, 0.1, and 1.6 ppm, and the
continuous dose, 1.6 ppm, designated 1.6C) were sacrificed at four-week
intervals corresponding to the on/off dosing cycles. Because both 1.6 ppm dose
groups were continuously fed the AFB 1-NIH diet during the first four weeks on
study, only five sets, 25 animals were euthanized at the end of the first dosing
cycle. All of the dose groups and sacrifices up to 20 weeks were included in
the immunologic study. Only the control group and the two high dose groups,
1.6 and 1.6C, were sampled at 40 weeks (i.e., termination of the feeding study,
364
HINTON ET AL.
assessed by admixing 0.1 ml of 10 6 splenic lymphocytes with either 0.1 ml of
LPS (1 ␮g/ml final concentration) or LPS with IFN␥ (100 U/ml final concentration) and then incubated for 24 h at 37°C. The resulting supernatant was
collected after centrifugation and stored at – 80°C until analysis for IL-1 and
IL-6 production.
FIG. 1. Schedule of the aflatoxin B 1 dosing and rest cycles in the “intermittent” study design.
equivalent to the fifth resting cycle) for evaluation of the certain of the flow
cytometric and histopathologic endpoints.
Hematology. A blood smear, approximately one cell layer thick, was
prepared from EDTA treated whole blood from terminal heart bleeds. Airdried smears were then treated with Wright’s stain for viewing. The WBC
differential counts were done by scoring 100 cells per slide for the various cell
types of interest, i.e., lymphocytes, segmented leukocytes, eosinophils, basophils, and monocytes (Creskoff et al., 1963). Total WBC counts were done
as previously described (Hinton et al., 1987) by counting acid-fixed and
gentian violet stained WBCs in a hemacytometer.
Histopathology. Animals were euthanized by CO 2 asphyxiation and then
necropsied. Parts of the organs were immediately placed into buffered-formalin and kept at room temperature for 24 h. The tissues were then removed from
the formalin and processed through a graded series of ethanol and xylene prior
to paraffin embedding.
Splenic cell isolation and purification. Approximately one-half of the
spleen was taken at necropsy for isolation of the splenic lymphocytes. A
portion of the spleen was minced immediately after necropsy in a culture disk
and then suspended in ice-cold Ultraculture media supplemented with 2 mm
L-glutamine. Ten ml of the cell suspension was layered on top of 5 ml of the
Histopaque in plastic tubes and then centrifuged at room temperature for 30
min at 2000 rpm. Portions of the purified lymphocytes were then used for flow
cytometry and splenic cell cultures for assessment of cytokine production.
Flow cytometry. Analysis of splenic lymphocyte populations was done by
fluorescent antibody cell sorting (FACS) analysis using an EPICS Elite flow
cytometer (Beckman/Coulter, Miami, FL). Splenic lymphocytes were suspended in phosphate buffered saline (PBS) containing 2% heat inactivated fetal
bovine serum and 0.05 % sodium azide (FACS diluent) at a concentration of
10 7 cells per ml. Monoclonal antibodies specific for rat cell surface antigens
were added to 50 ␮l of FACS diluent in wells of a 96 well microtiter plate to
achieve a predetermined optimal final concentration (0.1–1.0 ␮g/50 ␮l). Ten
␮l of spleen cell suspension was added to each well. Antibodies used for
immunofluorescent staining were directed against rat T-lymphocytes (CD3,
CD4, and CD8) or B-lymphocytes (CD45R). Direct FITC or R-phycoerythrin
conjugated antibodies were used for staining. Matched isotype control antibody conjugates were also used to determine background staining. Immunofluorescent staining took place for 30 min at 4°C. Samples were then washed
twice and resuspended in 100 ␮l FACS diluent. The FACS analysis was
conducted on the viable lymphocyte population as determined by forward light
scatter versus 90° light scatter gating. Five thousand cells were analyzed for
each antibody combination.
Splenic cell cultures. Splenic cell cultures, initiated by placing 1 ⫻ 10 6
lymphocytes in 0.1 ml of tissue culture media, were treated with 0.1 ml of
Con-A (5 ␮g/ml final concentration), and then incubated for 24 h at 37°C. The
resulting supernatant was collected after centrifugation and stored at – 80°C
until analyzed for IL-2 production. Production of either IL-1 or IL-6 was
Cytokine production bioassays. The assays for IL-1, IL-2, and IL-6 were
performed as previously described (Lyte et al., 1987 for IL-2 and Schook et al.,
1992 for IL-1 and IL-6) using cytokine responsive cell lines and as modified
by us (Myers et al., 1995, 1999). All assays were performed using complete
Ultraculture media, i.e., media supplemented with L-glutamine (2 mM, final
concentration), HEPES (50 mM, final concentration), gentamycin (50 ␮g/ml,
final concentration), and sodium bicarbonate (0.075%, final concentration).
Interleukin-2 activity was determined by adding 1 ⫻ 10 4 CTLL-2 cells in 100
␮l of media to an equal volume of either culture supernate or authentic IL-2
(for the standard curve). The cultures were incubated for 24 h at 37°C with 5%
CO 2. Four h prior to termination of culture, 20 ␮l of Alamar blue was added
to each well. The resulting fluorescence (560 nm excitation and 590 nm
emission) was determined using a CytoFluor 2350 (Millipore, Bedford, MA).
The IL-1 activity was determined by admixing 1 ⫻ 10 4 D10S cells (100 ␮l)
suspended in complete Ultraculture media with 100 ␮l of either the culture
supernatant or authentic human IL-1␤ (for the standard curve). The cultures
were incubated for 72 h at 37°C with 5% CO 2. Twenty-four h prior to
termination of culture, 20 ␮l of Alamar blue was added to each well. The
resulting fluorescence was then measured. The amount of IL-6 activity was
determined by adding 5 ⫻ 10 3 7TD1 cells (in 100 ␮l complete Ultraculture
media) to an equal volume of culture supernate or authentic human IL-6 (for
the standard curve). Twenty-four h prior to termination of culture, 20 ␮l of
Alamar blue was added to each well and the resulting fluorescence was
measured. Each microtiter plate had its own standard curve, which was used to
calculate the activity for the test samples on that particular microtiter plate
using the CytoCalc software program.
Cell lines used in bioassays. The CTLL-2 cells respond to only IL-2 and
murine IL-4; they do not measure rat IL-4. The D10S cells used for assessment
of IL-1 levels are a subclone of the D10.G4.1 cell line. The parent cell line
requires a source of murine IL-4 or IL-5 along with feeder cells and antigen for
propagation. It neither responds to nor produces IL-2. The D10S clone has the
advantage that it can be propagated in culture without the need for continual
antigen stimulation. It also does not respond to IL-2 but does respond to IL-1
(any species), murine IL-4, and murine IL-5. The 7TD1 cells used to measure
IL-6 may respond to murine IL-4, but only at very high levels.
Statistical analysis. All of the data generated were analyzed with validated
SAS PC (version 8.2) procedures. These included: means and error procedures
for general linear models; ANOVA, Dunnett’s multiple pairwise t-tests for
comparison of the dose groups to the control, 0.0 ppm group, and linear
regression models for evaluating dose responses. Differences between treated
groups and the control values that generated p-values equal to or less than 0.05
were considered statistically significant. For comparison of intermittent dosing
to the continuous dosing, Dunnett’s multiple pairwise t-test was applied also
using the continuous, 1.6 ppm dose as the statistical control, comparison group
for both the flow cytometry and the cytokine analyses.
RESULTS
Body Weights of Animals during the AFB 1 Feeding Study
(Data Not Shown)
Excluding a sporadic decrement (p ⱕ 0.05) in the weight of
the 0.01 ppm dose group after eight weeks on study, there were
no statistically significant differences between the body
weights of the control group and those groups fed 0.01, 0.04,
and 0.4 ppm AFB 1. In comparison to the control group, the
group intermittently fed 1.6 ppm AFB 1 showed a slight, but
IMMUNOTOXICITY STUDY OF AFLATOXIN B 1 IN F344 RATS
statistically insignificant, weight gain after four weeks on
study. This group consistently had a lower weight than the
control group after 8, 12, 16, and 20 weeks on study, although
the differences in weight between the treated and control group
were statistically significant (p ⱕ 0.05) only after 12 and 20
weeks on study. Likewise, the group continuously fed 1.6 ppm
AFB 1 consistently had a lower weight than the control group,
but the differences were statistically significant (p ⱕ 0.05) only
after 16 and 20 weeks on study. After 40 weeks on study, the
body weights of animals fed 1.6 ppm AFB 1 on either a continuous or an intermittent basis did not differ from that of the
control group.
Hematology
These data are shown in Figures 2a, 2b, and 2c, respectively.
Few significant changes were noted in the total WBC counts,
percentages of lymphocytes, and percentages of segmented
neutrophils during the study. In comparison to the control
group, there was an increase in the total WBC count (p ⱕ 0.05)
in the group continuously treated with 1.6 ppm AFB 1 after
eight weeks on study and in the group intermittently treated
with AFB 1 after 12 weeks on study. An increase in the percentage of lymphocytes (p ⱕ 0.05) and a concurrent decrease
in the percentage of segmented neutrophils (p ⱕ 0.05) were
observed in the group continuously treated with 1.6 ppm AFB 1
after 12 weeks on study.
Splenic Lymphocyte Subsets
The results of the flow cytometric analysis are presented in
Figures 3 and 4 respectively. The percentages of the T (CD3⫹)
and B (CD45R⫹) lymphocytes are presented in Figures 3a and
3b. The percentages of the T-lymphocyte helper (T-h, CD4⫹)
and suppressor (T-s, CD8⫹) subsets are shown in Figures 4a
and 4b. The “dynamics” of the immune system in the growing
rat are clearly demonstrated by the changes in the lymphocyte
populations that are seen in the untreated, control animals
between 8 and 24 weeks of age (corresponding to 4 and 20
weeks on study). The percentage of CD3⫹ lymphocytes (Fig.
3a) increased from approximately 28 to 57%, while the percentage of CD45R⫹ cells (Fig. 3b) decreased from approximately 58 to 29%. During the same time period, the percentage
of CD4⫹ cells increased from approximately 20 to 49% (Fig.
4a), whereas, the percentage of CD8⫹ cells (approximately
25%) remained unchanged (Fig. 4b).
The effects of AFB 1 were notable on both the T- and B-cell
populations in both the first and second dosing cycles. Percentages of T cells were significantly increased (p ⱕ 0.05) at the
0.4 and the 1.6 ppm continuous (C) and intermittent (I) dose
groups, while the percentages of B lymphocytes were significantly (p ⱕ 0.05) decreased at the mid (0.4 ppm) and the 1.6
ppm dose groups compared to the control (0.0 ppm) dose
groups. Only the percentage of the 1.6 ppm continuous dose
group was significantly different at eight weeks for T cells
365
compared to the control group. In contrast, no significant
differences for the percentage of B cells were seen during the
eight-week cycle. After 16 weeks on study, the effects of being
off the AFB 1 diet differed from those after eight weeks on
study. The percentages of T cells were significantly (p ⱕ 0.05)
decreased at the mid (0.4 ppm) and the 1.6 ppm dose groups
while the percentages of B cells were significantly (p ⱕ 0.05)
increased at the mid and high dose groups compared to the
control group. These results suggest not only a reversal of the
effects of AFB 1 but also the possibility of a compensatory
change at the end of the 16-week cycle. At 20 weeks, the
percentages of the two lowest dose groups (0.01 and 0.04 ppm)
were significantly (p ⱕ 0.05) increased for T cells and significantly decreased for B cells compared to controls. After 40
weeks the percentages of T-cells for the two 1.6 ppm dose
groups were not significantly different from the controls;
whereas the percentages of B cells for the two high dose groups
were significantly increased compared to the controls.
The percentages of either T-h or T-s were not statistically
different compared to controls for the first four-week period
(Figs. 4a and 4b). Statistically significant (p ⱕ 0.05) increases
in the percentages were seen for the T-h subset at the mid (0.4
ppm) and high (1.6 ppm) dose groups, while statistically significant decreased percentages were seen for the T-s subset at
eight weeks compared to the control groups. A statistically
significant difference was seen only at the 0.04 ppm dose group
for T-h cells while significant (p ⱕ 0.05) differences were seen
for the 0.04, 0.4, 1.6I, and 1.6C dose for T-s cells after 12
weeks. No significant differences in the percentages of the T-h
cells were seen for either the 16 or 20 week cycles. Significant
decreases in the percentages of the 0.4 and 1.6 ppm dose
groups were seen for the T-s cells after 16 weeks. Only the 0.4
ppm dose group of the T-s subset was significantly increased
after 20 weeks of study.
Cytokine Proliferation Assays
IL-2. This cytokine was included in our analyses since it is
an important indicator of proliferative capacity and may be
involved also in the inflammatory response. The IL-2 productive capacity of the splenic lymphocytes is shown in Figure 5.
From 8 (week 4 of the study) to 24 weeks of age (week 20 of
the study), the IL-2 productive capacity of the cells in the
maturing animal increased from 5 to 35 biological units in the
controls. Significantly decreased capacity was observed after
eight weeks for the highest doses of AFB 1 tested. No consistent
pattern of statistical significance, however, was seen at either
12, 16, or 20 weeks.
IL-1. We measured IL-1 since it is an indicator of a possible inflammatory response. The results for the determination
of IL-1 productive capacity are presented in Table 1. No
significant changes were seen at four weeks. Significant decreases were noted at all dose levels of AFB 1 compared to the
controls with LPS and INF␥ at eight weeks. There was a 4 to
366
HINTON ET AL.
FIG. 2. (a) Total white blood cell
counts, mean ⫾ SE. (b) The percentage
of the WBCs as lymphocytes from the
differential count. (c) The percentage of
the WBCs as segmented neutrophils
from the differential count. For each of
the time points, beginning at the first
dosing cycle at four weeks, the results
are plotted starting from the left side of
the cluster, solid black bar, 0.0 ppm dose
followed by the hatched and shaded bars
representing 0.01, 0.04, 0.4, and 1.6 ppm
in each 4-week cycle. For the cycles at 8,
12, 16, and 20 weeks, the last bar is the
continuous 1.6 ppm dose. For the 40week samples, the solid bar is the control, 0.0 ppm dose and the two bars from
left to right are the intermittent 1.6 ppm
dose and the 1.6 ppm continuous dose.
The asterisk (*) designations represent
statistically significant comparisons to
the control by Dunnett’s multiple t-test.
Legends are also presented in the figures.
5-fold reduction in IL-1 production at all dose levels with
splenocytes compared to that of the control splenocytes. The
most dramatic effects of AFB 1 on this end-point, however,
were observed at 12 weeks. The IL-1 productive capacity
was approximately 10-fold higher in the two highest, intermittent doses, 0.4 and 1.6 ppm; it was even higher (approximately, a 15-fold increase) at the 1.6C ppm dose. The IL-1
productive capacity was significantly increased at the middose levels, 0.04 and 0.4 ppm at 16 weeks. The productive
capacity returned to normal levels at the two highest dose
levels. There were no significant differences noted for the 20
week period.
IL-6. We measured IL-6 since it is also an indicator of a
possible inflammatory response. The results for the productive
IMMUNOTOXICITY STUDY OF AFLATOXIN B 1 IN F344 RATS
367
FIG. 3. (a) The % of the splenic cell
isolate as T-lymphocytes (CD3⫹),
mean ⫾ SE. (b) The % of the splenic cell
isolate as B-lymphocytes (CD45R⫹),
mean ⫾ SE. For each of the time points,
beginning at the first dosing cycle at four
weeks, the results are plotted starting
from the left side of the cluster, solid
black bar, 0.0 ppm dose followed by the
hatched bars representing 0.01, 0.04, 0.4,
and 1.6 ppm in each 4-week cycle. For
the cycles at 8, 12, 16, and 20 weeks, the
last bar is the continuous 1.6 ppm dose.
For the 40-week samples, the solid bar is
the control, 0.0 ppm dose and the two
bars from left to right are the intermittent
1.6 ppm dose and the 1.6 ppm continuous
dose. The asterisk (*) designations represent statistically significant comparisons to the control by Dunnett’s multiple
t-test.
capacity for IL-6 are presented in Table 2. There were no
significant differences seen at four weeks. The productive
capacity was significantly reduced at the two highest intermittent dose levels of 0.4 and 1.6 ppm at eight weeks. Again, as
seen for IL-1, IL-6 production was significantly increased (p ⱕ
0.05) approximately twofold higher with LPS and INF␥, compared to the controls at 12 weeks. No significant differences
were seen at 16 and 20 weeks respectively.
Histopathology
Histomorphologic changes in liver and spleen H&E-stained
sections following AFB1 exposure were evaluated in the high
dose continuous and intermittent groups in order to screen for
a possible inflammatory response. The rationale for this dosegroup selection was that changes are more likely to manifest in
the highest dose groups. Representative photomicrographs of
368
HINTON ET AL.
FIG. 4. (a) The % of the splenic
cell isolate as helper T-lymphocytes
(CD4⫹), mean ⫾ SE. (b) The % of the
splenic cell isolate as suppressor T-lymphocytes (CD8⫹), mean ⫾ SE. For each
of the time points, beginning at the first
dosing cycle at four weeks, the results
are plotted starting from the left side of
the cluster, solid black bar, 0.0 ppm dose
followed by the hatched bars representing 0.01, 0.04, 0.4, and 1.6 ppm in the
four-week cycle. For the cycles at 8, 12,
16, and 20 weeks, the last bar is the
continuous 1.6 ppm dose. For the 40
week samples, the solid bar is the control, 0.0 ppm dose and the two bars from
left to right are the intermittent 1.6 ppm
dose and the 1.6 ppm continuous dose.
The asterisk (*) designations represent
statistically significant comparisons to
the control by Dunnett’s multiple t-test.
liver (Figs. 6 and 7) and spleen (Fig. 8) sections have been
chosen to illustrate key observations.
Liver
High dose animals continuously treated for eight weeks
revealed a hepato-cellular cytoplasmic vacuolar change characterized by cell swelling and cytoplasmic membrane bound
clear space formation (Figs. 6a,b). Infrequent small inflamma-
tory cell infiltrates composed of lymphocytes, plasma cells,
mononuclear cells, and few segmented neutrophils appeared to
respond to degenerate vacuolated hepatocytes (Fig. 6b).
In contrast, the hepato-cellular vacuolar change was less
prominent in livers of the eight-week intermittently fed animals
(Figs. 6c,d). The focal inflammatory cell infiltrates aforementioned were observed more frequently however. Increased
numbers of activated Kupffer cells characterized by sinusoidal
369
IMMUNOTOXICITY STUDY OF AFLATOXIN B 1 IN F344 RATS
FIG. 5. Proliferative biological activity, mean ⫾ SE of stimulated splenic
cell isolates to concanavalin-a for measurement of IL-2 production. For each of
the time points, beginning at the first
dosing cycle at four weeks, the results
are plotted starting from the left side of
the cluster, solid black bar, 0.0 ppm dose
followed by the hatched bars representing 0.01, 0.04, 0.4, and 1.6 ppm in the
four- week cycle. For the cycles at 8, 12,
16, and 20 weeks, the last bar is the
continuous 1.6 ppm dose. For the 40
week samples, the solid bar is the control, 0.0 ppm dose and the two bars from
left to right are the intermittent 1.6 ppm
dose and the 1.6 ppmcontinuous dose.
The asterisk (*) designations represent
statistically significant comparisons to
the control by Dunnett’s multiple t-test.
cells with increased amounts of cytoplasm and vacuolated
nuclei as well as small foci of extramedullary hematopoiesis
(EMH) were also present.
At 12 weeks of continuous dosing, the severity of the vacuolar cytoplasmic change progressed to mild to moderate and
was diffuse in distribution (Fig. 6e). Mild billiary hyperplasia,
a proliferative change of the bile ducts was evident. Small
basophilic foci of hepato-cellular alteration were seen and were
characterized by focal proliferation of hepatocytes, assuming
either increased blue (basophilic) or red (eosinophilic) staining
properties (Fig. 6e). Activated Kupffer cells had a condensed
nuclear chromatin pattern consistent with pyknosis (Fig. 6f).
Loss of individual hepatocytes as well as apoptotic cell necrosis was seen infrequently. Foci of inflammation, described
TABLE 1
Splenic Lymphocyte Production of IL-1 after Stimulation with LPS or LPS and Interferon ␥
Week number in the intermittent feeding study
4
Dose AFB 1
(ppm)
0.0
0.01
0.04
0.40
1.6
1.6C
8
12
16
20
LPS
LPS ⫾ INF␥
LPS
LPS ⫾ INF␥
LPS
LPS ⫾ INF␥
LPS
LPS ⫾ INF␥
LPS
LPS ⫾ INF␥
14.7 ⫾ 4.9
1.8 ⫾ 0.7
5.1 ⫾ 1.4
5.9 ⫾ 2.4
15.0 ⫾ 5.6
ND
ND
ND
ND
ND
NA
45.1 ⫾ 13.1 a
36.3 ⫾ 3
69.4 ⫾ 14.2
30.0 ⫾ 11.7
26.7 ⫾ 6.2
75.6 ⫾ 14.9
24.0 ⫾ 6.9 b
84.5 ⫾ 0.5*
5.7 ⫾ 0.3*
3.8 ⫾ 0.3*
2.4 ⫾ 0.4*
5.6 ⫾ 1.1*
14.7 ⫾ 4.3
31.9 ⫾ 4.1
40.5 ⫾ 14.9
22.8 ⫾ 8.5
17.6 ⫾ 13.1
51.0 ⫾ 20.2
20.6 ⫾ 2.1 a
33.6 ⫾ 10.4
19.2 ⫾ 3.6
216 ⫾ 128*
200 ⫾ 75*
353 ⫾ 105*
25.8 ⫾ 20.4
32.5 ⫾ 16.7
54.0 ⫾ 9.0
64.0 ⫾ 16.9
57.2 ⫾ 6.7
45.6 ⫾ 14.4
18.7 ⫾ 3.1 a
20.6 ⫾ 2.1
207 ⫾ 62*
275 ⫾ 82*
14.6 ⫾ 1.8
14.9 ⫾ 14.4
3.5 ⫾ 0.1
3.4 ⫾ 0.1
48.2 ⫾ 4.0*
3.3 ⫾ 0.1
25.3 ⫾ 2.7
19.7 ⫾ 5.6
13.8 ⫾ 3.9
2.1 ⫾ 0.1
3.4 ⫾ 1.7*
74.9 ⫾ 18.6*
46.8 ⫾ 9.5*
37.8 ⫾ 12.9*
Note. Mean ⫾ SE. Figures shown indicate biological activity (x ⫾ SE). ND, not determined; NA, not applicable since 1.6C is equivalent to the 1.6 dose;
ANOVA (0.05 level).
a
p ⱕ 0.05.
b
p ⱕ 0.001.
*Dunnett’s T- statistic for comparison to the 0.0 ppm dose, p ⱕ 0.05. Data are the results of the bioassay of IL-1 production from splenocytes stimulated with
either LPS or LPS ⫹ INF␥.
370
HINTON ET AL.
TABLE 2
Splenic Lymphocyte Production of IL-6 after Stimulation with LPS or LPS and INF␥
Week number in the intermittent feeding study
4
Dose AFB 1
(ppm)
0.0
0.01
0.04
0.40
1.6
1.6C
8
LPS
LPS ⫾ INF␥
LPS
949 ⫾ 241
966 ⫾ 610
1266 ⫾ 229
491 ⫾ 28
719 ⫾ 200
ND
ND
ND
ND
ND
NA
823 ⫾ 137 a
954 ⫾ 157
1476 ⫾ 225*
527 ⫾ 73
495 ⫾ 50
1239 ⫾ 62
12
LPS ⫾ INF␥
LPS
LPS ⫾ INF␥
16
LPS
LPS ⫾ INF␥
20
LPS
1232 ⫾ 268 b 997 ⫾ 212 814 ⫾ 99
488 ⫾ 102 b
419 ⫾ 179 976 ⫾ 462 b
889 ⫾ 105 1072 ⫾ 156 948 ⫾ 178
105 ⫾ 24
230 ⫾ 23
421 ⫾ 25
1484 ⫾ 258 1457 ⫾ 307 1219 ⫾ 210
168 ⫾ 30
656 ⫾ 502 331 ⫾ 69
660 ⫾ 262* 720 ⫾ 250 1873 ⫾ 249* 1921 ⫾ 784 1502 ⫾ 259 1270 ⫾ 51
464 ⫾ 51*
621 ⫾ 295 1499 ⫾ 161* 1196 ⫾ 682
444 ⫾ 130 682 ⫾ 61
931 ⫾ 185 1133 ⫾ 573 1008 ⫾ 429 2106 ⫾ 1356 1314 ⫾ 566 578 ⫾ 96
LPS ⫾ INF␥
758 ⫾ 309
240 ⫾ 106
163 ⫾ 83
674 ⫾ 323
308 ⫾ 87
320 ⫾ 92
Note. Figures shown are IL-6 units (x ⫾ SE). ND, not determined); NA, not applicable since 1.6C is equivalent to the 1.6 dose; ANOVA (0.05 level).
a
p ⱕ 0.001.
b
p ⱕ 0.05.
*Dunnett’s T- statistic for comparison to the 0.0 ppm dose, p ⱕ 0.05. Data are the results of the bioassay of IL-6 production from splenocytes stimulated with
either LPS or LPS ⫹ INF␥.
above and responding to degenerate hepatocytes, increased
only slightly.
Most livers of the intermittent high dose group appeared to
be less affected by the vacuolar change after 12 weeks (Fig.
7a), while inflammatory cell infiltrates, Kupffer cell activation,
and billiary hyperplasia were slightly more prominent at this
time point compared to the continuously treated animals (Fig.
7b).
After 16 weeks of continuous dosing the severity of the
vacuolar hepato-cellular change had progressed to moderate.
The numbers of mixed inflammatory cell infiltrates in the
vicinity of degenerate hepatocytes and around vessels remained small however. The micro-architecture of the liver
sections was increasingly distorted due to focal areas of
hepato-cellular proliferation. Activated Kupffer cells, billiary
hyperplasia, and basophilic foci were prevalent. The vacuolar
change in animals of the intermittently fed group was slightly
less severe at 16 weeks while the inflammatory response was
comparable to continuously fed animals.
After 20 weeks (Fig. 7c), the hepato-cellular vacuolar
change, formation of basophilic and/or eosinophilic foci of
cellular alteration, billiary hyperplasia, and hepatocellular proliferation leading to distortion of the liver micro-architecture,
progressed while inflammation remained mild (Fig. 7d).
After 40 weeks of continuous and intermittent dosing the
normal hepatic architecture was moderately distorted (Fig. 7e),
due to progression of the proliferative and neoplastic changes
mentioned above. Within areas of hepatic neoplasia, larger
aggregations, consisting primarily of lymphocytes, could be
observed (Fig. 7f). These infiltrates differed in size and cell
composition from the small inflammatory cell foci responding
to degenerate vacuolated liver parenchyma, previously mentioned.
Spleen
(Extensive morphometric and immuohistochemical analyses
of various biomarkers in the spleen, thymus, and Peyer’s
patches are the subject of another report in progress.) Herein,
we present an evaluation of H&E-stained sections of spleen for
the two highest doses of AFB 1 in order to screen for the effects
on cells involved in inflammatory responses, i.e., macrophage,
neutrophils, and lymphocytes.
Evidence of inflammation in the spleen was not observed.
However, effects on the distribution of cells within their
splenic micro-compartments were evident. The cellularity of
the Mantel zones of control animals at four weeks on study was
low (Fig. 8a), indicating immunologic immaturity. Hemosiderinladen macrophages and scattered segmented neutrophils were
commonly seen in the outer rim of the Mantel zone bordering
the red pulp (Fig. 8b).
The Mantel zones of animals that had been on the highest
dose of AFB 1 for four weeks were, in contrast to controls,
denser and more cellular indicating immune stimulation. In
contrast to control animals, hemosiderin-laden macrophages
were not discernable in the high dose AFB 1-treated group.
Erythrophagocytosis was frequently seen in the AFB 1-treated
group however. Neutrophil numbers were slightly reduced
compared to the control animals.
Mantel zones of eight-week control animals had matured
and were more cellular therefore than at four weeks. Hemosiderin-laden macrophages and neutrophils were frequently observed in the outer rim of the Mantel zone. At eight weeks on
a continuous AFB 1 diet, Mantel zones of most animals appeared irregular. Their outlines were not well defined around
the follicles (Fig. 8c), blending into the bordering red pulp
and/or into the Mantel zone of the neighboring follicle. Focally, the Mantel zones were thin due to lesser cellularity or
IMMUNOTOXICITY STUDY OF AFLATOXIN B 1 IN F344 RATS
FIGURE 6
371
372
HINTON ET AL.
they were not discernable at all. Only few hemosiderin-laden
macrophages and neutrophils were seen in the outer Mantel
zone rim (Figs. 8c,d).
After eight weeks, the splenic micro-architecture of the high,
1.6 ppm intermittent dose group, with respect to the Mantel
zone, was similar to the eight-week control animals displaying
even cellularity. The number of hemosiderin-laden macrophages in outer rim of the Mantel zone, however, was reduced.
Neutrophil numbers were within limits of the controls.
Splenic mantel zones of animals that had been fed AFB 1
continuously and intermittently for 12 weeks were greatly
reduced in cellularity and width (Fig. 8e). The number of
hemosiderin-laden macrophages appeared to be only slightly
less compared to 12-week control animals (Fig. 8f), but were
more frequent in the intermittently fed animals compared to the
continuously fed animals. Neutrophil numbers were within
limits of the controls.
Animals that were on a continuous diet for 16 weeks presented with spleens similar to 16-week control animals with
respect to the Mantel zone micro-architecture (sections not
shown). The periarterioalar lymphocyte sheaths (PALS) were
of variable size, indicating a decrease of lymphocytes. The
cellularity of Mantel zones surrounding small PALS was
greatly reduced. Numbers and distribution of hemosiderinladen macrophages and neutrophils were within limits of the
controls. The spleen micro-architecture of the high-dose, intermittent group, at 16 weeks was similar to that of controls.
Splenic Mantel zones of the high, intermittent dose group at
20 weeks were less cellular compared to 20-week control
animals. Hemosiderin-laden macrophages and segmented neutrophils were as frequently seen as in control animals (sections
not shown). Most spleens, either of animals fed AFB 1 for 40
weeks continuously or for 40 weeks intermittently, were comparable in their micro-architecture.
DISCUSSION
The design of the “intermittent” dosing study with AFB 1
was originally conceived to address certain aspects of risk
assessment with regard to carcinogens such as dose and dura-
tion of exposure in relation to the expression of biomarkers
such as DNA adducts (Gaylor et al., 1992). These authors
concluded that, although promising, more data are needed to
judge the usefulness of DNA adduct concentrations to predict
cancer incidence across species. In addition to DNA adducts
formed as a result of AFB 1 exposure, it was known that AFB 1
affects immune function in various animal species. This aspect
of AFB 1 toxicity could be important also for risk assessment
extrapolations, if the immune system, via the inflammatory
process or other mechanisms, is involved in hepatotoxicity
and/or carcinogenicity. Thus, in the study presented herein, we
investigated the effects of AFB 1 on the main cellular targets by
histopathology, flow cytometric analysis of the proportions of
the various splenic lymphocyte subsets, and their functional
capacity to produce the inflammatory cytokines, IL-1 and IL-6
as well as IL-2. A preliminary report of the effects of AFB 1 on
other immunotoxicity biomarkers has been published (Hinton
et al., 2001).
There were early reports (Butler, 1970) of “slight” inflammatory responses in rats due to AFB 1-induced injury in the
liver. The researcher reported, however, that in the chicken,
which shows relatively slight cellular degeneration and almost
no necrosis, large lymphoid follicles appear in the areas of fatty
change. We believe that inflammatory and other possible immune mechanisms in relation to AFB 1-induced hepatotoxicity
and/or carcinogenicity in the F344 rat needed to be investigated
further. There are a number of reports that F344 rats are more
susceptible to chemically induced liver injury than other strains
(Kuester et al., 2002). In addition, we found no reports on the
immunotoxic effects of AFB 1 in an intermittent exposure regimen wherein there would be sufficient “resting” periods in
order for the immune system to recover, i.e., either reverse or
compensate for the effects of AFB 1. We chose to measure IL-1
and IL-6 since these cytokines are increased in an inflammatory response. We measured IL-2 as a measure of lymphocyte
proliferative capacity as well as its possible involvement in an
inflammatory response. Since there was a suggestion from the
cytokine analyses that an inflammatory response may be associated with AFB 1 toxicity, we evaluated the histopathology of
the liver at the high intermittent and continuous doses for the
FIG. 6. Photomicrographs of hematoxylin and eosin stained rat liver sections at the intermittent and high dose levels at eight weeks continuous, eight weeks
intermittent, and 12 weeks continuous dosing cycles. (a) Eight-week continuous dosing, magnification x100. Note the distribution and severity of the focal
vacuolar change (small arrows) at this early time point of continuous AFB 1 feeding. A small focus of mixed inflammatory cells (long arrow) is located in the
vicinity of the vacuolar change. (b). Eight-week continuous dosing, magnification x200. The hepatocellular vacuolar change (small arrows) is characterized by
cell swelling and poorly demarcated intra-cytoplasmic clear space formation leading to hepatocellular degeneration, necrosis and loss of individual hepatocytes.
The inflammatory infiltrate (long arrow) consists of a mixed cell population surrounding degenerate hepatocytes. (c) Eight-week intermittent dosing,
magnification x100. Note, nearly normal hepato-cellular morphology and scattered individual cells show minimal cytoplasmic clear space formation. (d)
Eight-week intermittent dosing, magnification x400. Notable only at a higher power, hepatocytes show a vacuolar change (small arrows) similar to the change
observed after eight weeks of continuous dosing. Inflammatory cell aggregates appeared to be slightly more common in intermittently dosed animals of this time
period. A mixed inflammatory cells focus is shown centering on a degenerate hepatocyte (larger arrow). (e) Twelve-week continuous dosing, magnification x200.
Most of the hepatocellular parenchyma is affected diffusely by the vacuolar change (small arrows). Note early basophilic focus (large arrow) comprised of
hepatocytes with homogeneously basophilic cytoplasm contrasting to the surrounding parenchyma cells. (f) Twelve-week continuous dosing, magnification x400.
Activated Kupffer cells (arrows) with elongated nuclei and some with condensed chromatin pattern that may indicate pyknosis are shown.
IMMUNOTOXICITY STUDY OF AFLATOXIN B 1 IN F344 RATS
FIGURE 7
373
374
HINTON ET AL.
various periods in the study in order to discern if there was any
involvement of the immune system. When we consider all of
the cellular data, the time cycle most indicative of possible
immune effects was at the 12 week, i.e., the second dosing
period. The hematology data also support this observation.
Although the hematology data are not generally sensitive indicators of immunotoxic effects (Hinton, 2000), the total WBC
count for the high, intermittent, 1.6 ppm dose was significantly
different (p ⱕ 0.05) from the control at 12 weeks. There were
also some indications from the high, continuous, 1.6 ppm dose
for the WBC differential count that significant immune effects
were occurring at 12 weeks as well. We showed in both the
flow cytometric splenic subset analysis and the cytokine bioassays that the immune system also is changing as the animal
matures from 4 to 24 weeks of age, i.e., during the important
phases of the initial dosing cycles of the feeding study. Thus,
we needed to compare the dosed groups to the control for each
time period. Results from all of the analyses support the
conclusion of significant immune effects at 12 weeks into the
study, i.e., after the second dosing cycle. There were suggestions from the flow cytometric analyses that the different
lymphocyte populations may compensate or reverse, to some
extent, the effects of AFB 1 during the “off” or resting cycles.
The histopathology evaluation of the spleen presented herein
suggested that continuous exposure resulted in cumulative
effects on T-lymphocyte cellularity in the PALS and that
macrophage function, from analysis of the hemosiderin-laden
cells, is suppressed. The histopathology evaluation of the liver
demonstrated that AFB 1 caused damage to hepatocytes as
exemplified by the vacuolar formation. This change was more
severe after eight weeks of continuous dosing compared to the
intermittent high dose group, but was equivalent in both dose
groups after the 12-week dosing cycle. Mixed inflammatory
cell infiltrates formed after four weeks of dosing but were more
abundant in the intermittent dose group after eight weeks
compared to the eight week and the 12-week intermittent dose
group, likely as a result of a degree of recovery from immunosuppression and in response to the degenerate hepatocytes.
This inflammatory process was most pronounced at 12 weeks
however. At 12 weeks of continuous dosing there were early
preneoplastic lesions. After 40 weeks of continuous dosing
there were defined inflammatory infiltrates/immune responses
to the damaged liver suggesting that even after this length of
duration of exposure at the highest dose, the immune system
had enough reserve capacity to function to some extent. It was
also of interest that there were more inflammatory infiltrates in
the intermittent dose groups compared to the continuous at 8,
12, 16, and 20 weeks. Suppression of the inflammatory response via suppression of Kupffer cell activation in the liver by
AFB 1 is in agreement with suppression of macrophage function as seen in the splenic histopathology. In order to correlate
the histopathology results with the flow cytometric and cytokine proliferative responses, we compared the statistical significance of the intermittent dosing (both the flow cytometric
and the cytokine measurements) using the 1.6 ppm continuous
dose group as the statistical control (data not shown). (This was
in addition to the statistics presented herein which used the 0.0
ppm dose group as the statistical control.) Significant statistical
differences were prevalent beginning with the eight-week period for many of the immune parameters at low doses. There
was almost a complete absence, however, of statistically significant differences when the continuous 1.6 ppm dose group
was compared to the same intermittent dose group. This suggests that, at least at the 1.6 ppm dose, the immunotoxic effects
are cumulative and are either not repaired after sufficient time
of exposure or are repaired slowly within the time frames used
as the resting cycle in this study.
As mentioned previously, the initial rationale for this study
was with regard to risk assessment of the hepatocarcinogenic
potential of AFB 1. The intermittent dosing regimen is also a
more realistic approach to exposure and allows for accumulated dose extrapolations. There are various reports (Henry et
al., 2002) where the risk of hepatocarcinoma is greatest in
those regions of the world, e.g., Africa and China (Wang et al.,
2001) where there are both high percentages of hepatitis B
(HB) and C (HC) infections and contamination of foodstuffs by
aflatoxin B1. One of the major debates in hepatocellular carcinogenesis (Kew, 1992) is whether the HB and HC viruses are
directly carcinogenic or exert their effect indirectly by causing
chronic necro-inflammatory hepatic disease, which in turn is
responsible for malignant transformation of hepatocytes. In
other words, HB and HC viruses as well as AFB 1 acting alone
could lead to hepatocarcinoma, provided that the critical doses
and damage to the liver are sufficient to induce immunologic
FIG. 7. Photomicrographs of hematoxylin and eosin stained rat liver sections at 12 weeks of intermittent dosing, 20 and 40 weeks of intermittent and
continuous dosing. (a) 12-week intermittent dosing cycle, magnification x100. The low power distribution of the hepatocellular vacuolar change is multifocal
(arrows) resembling the distribution pattern of the eight-week continuous dosing cycle. (b) 12-week intermittent dosing, magnification x200. Note most livers
appeared to be less affected by the vacuolar change (small arrows), while inflammatory cell infiltrates (large arrow), Kupffer cell activation, and billiary
hyperplasia were slightly more prominent at this dosing regimen compared to the continuously treated animals. (c) 20-week intermittent dosing, magnification
x100. Note the progression to a diffuse widespread distribution pattern of the vacuolar change (small arrows), affecting the entire hepatic parenchyma. (d)
20-week intermittent dosing, magnification x400. Shown is a small focal area of inflammation (large arrow) at the border of hepatocytes, which are moderately
affected by the progressing vacuolar change (small arrows). (e) 40-week continuous dosing, magnification x200. Note the distortion of the hepatic parenchyma
(small arrows) resulting secondary to hyperplastic lesions (nodular proliferation, foci of cellular alteration, benign and/or malignant neoplasia). (f) 40-week
continuous dosing cycle, magnification x200. Anti-neoplastic inflammation (long thin arrows) within a cholangioma (benign bile duct tumor). Note the larger
size of the inflammatory cell infiltrate compared to the occasional small lymphocytic foci observed within the liver parenchyma affected by the vacuolar change
during the course of dosing.
IMMUNOTOXICITY STUDY OF AFLATOXIN B 1 IN F344 RATS
FIGURE 8
375
376
HINTON ET AL.
mediated necrosis in the liver. In our study this point appears
to be after the second dosing cycle, at least for the high doses
tested.
An analogy to the mechanism of AFB 1 possible involvement
in carcinogenesis is dimethynitrosamine (DMN). First, DMN
also forms adducts (methyl-and hydroxyl guanyl) in the liver
(and other tissues). Cirrhosis in the early phases is accompanied by inflammatory filtrates (Mancini et al., 1991) composed
mainly of T-cytotoxic-inducer/T-suppressor cells. There are
progressive stages of immune mediated damage to the liver
resulting in hepatocarcinoma.
In conclusion, the immune system of the rat is both a target
for the toxic effects of AFB 1 and a participant in immune
mediated inflammatory reactions/immune responses in the
liver of the F344 rat. Thus, the effects that were seen in the
“intermittent” dosing study were complex suggesting that the
immune system may compensate or reverse, at least partially,
the toxic effects during the “resting cycles” at doses lower than
1.6 ppm. After the second resting cycle at 16 weeks, it appeared that the percentages of T and B cells were being
reversed in comparison to the 8- and 12-week results (Figs.
3a,b). The shift in T-h and T-s cells seen in Figure 4 during the
resting cycles, i.e., increased T-h (Fig. 4a) and decreased T-s
(Fig. 4b), suggests a compensatory change in response to the
down regulation of the percentages of B-lymphocytes after
four weeks (Fig. 3b). These results are likely a reflection of the
direct effects of AFB 1 on lymphocyte proliferation and function during the dosing cycles and the compensatory/recovery
efforts when AFB 1 is not included in the diet. The most
significant results from the cytokine analysis were the correlations with the histopathology evaluation of the liver. When the
results of the IL-1 and IL-6 were considered together, they
were suggestive of the induction of an inflammatory response
occurring after the second dosing cycle at 12 weeks. The
histopathology evaluation, however, demonstrated that the immune system had enough reserve capacity even at the high, 1.6
ppm, continuous dose to be associated with the hepatotoxic/
hepatocarcinogenic processes at the end of the 40-week
study.
ACKNOWLEDGMENTS
The authors wish to express their sincere appreciation for excellent technical
assistance from Mr. Michael Scott and Mr. Randolph Jackson in the hematological analysis, Mr. Elmer Bigley in flow cytometry, Mr. Andrei Perlloni in
some of the computer analyses, Dr. Curtis Barton for review of the statistical
analysis, and Drs. Thomas Collins and Carol Mapes for in-depth reviews of the
manuscript.
REFERENCES
Batey, R. G., and Wang, J. (2002). Molecular pathogenesis of T lymphocyteinduced liver injury in alcoholic hepatitis. Front. Biosci. 7, 1662–1675.
Bodine, A. B., Fisher, S. F., and Gangjee, S. (1984). Effect of aflatoxin B1 and
major metabolites on phytohemeagglutinin-stimulated lymphoblastogenesis
of bovine lymphocytes. J. Dairy Sci. 67, 110 –114.
Brown, R. W., Pier, A. C., Richard, J. L., and Krogstad, R.E. (1981). Effects
of dietary aflatoxin on existing bacterial intramammary infections of dairy
cows. Am. J. Vet. Res. 42, 927–933.
Butler, W. H. (1970). Liver injury induced by aflatoxin. Prog. Liver Dis. 3,
408 – 418.
Creskoff, A. J., Fitz-Hugh, Jr., T., and Farris, E. J. (1963). Hematology of the
rat—methods and standards. In The Rat in Laboratory Investigation, 2nd ed.
(E. J. Farris and J. Q. Griffith, Eds.), pp. 104 –160. Hafner Publishing, New
York.
Cusumano, V., Costa, G. B., Trifiletti, R., Merendino, R. A., and Mancuso, G.
(1995). Functional impairment of rat Kupffer cells induced by aflatoxin B1
and its metabolites. FEMS Immunol. Med. Microbiol. 19910, 151–155.
Cusumano, V., Rossano, F., Merendino, R. A, Arena, A., Costa, G. B.,
Mancuso, G., Baroni, A., and Losi, E. (1996). Immunobiological activities
of mould products: Functional impairment of human monocytes exposed to
aflatoxin B1. Res. Microbiol. 147, 385–391.
Dominguez-Malagon, H., and Gaytan-Graham, S. (2001). Hepatocellular carcinoma: An update. Ultrastruct. Pathol. 25, 497–516.
Dugyala, R. R., and Sharma, R. P. (1996). The effect of aflatoxin B1 on
cytokine mRNA and corresponding protein levels in peritoneal macrophages
and splenic lymphocytes. Int. J. Immunopharmacol. 18, 599 – 608.
Gaylor, D. W., Kadlubar, F. F., and Beland, F. A. (1992). Application of
biomarkers to risk assessment. Environ. Health Perspect. 98, 139 –141.
Ghosh, R. C., Chauhan, H. V., and Roy, S. (1990). Immunosuppression in
broilers under experimental aflatoxicosis. Br. Vet. J. 146, 457– 462.
Giambrone, J. J., Diener, U. L., Davis, N. D., Panangala, V. S., and Hoerr, F. J.
(1985a). Effects of aflatoxin on young turkeys and broiler chickens. Poult.
Sci. 64, 1678 –1684.
Giambrone, J. J., Diener, U. L., Davis, N. D., Panangala, V. S., and Hoerr, F. J.
FIG. 8. Photomicrographs of H&E-stained splenic sections. (a) H&E-stained section of spleen from the four-week control, 0.0 ppm AFB 1, dose group,
magnification x100. The micro-architecture (compartmentalization) of the spleen is shown. From left to the right: the dark periarteriolar sheets (PALS), comprised
of splenic T and B cell follicles, the Mantel zone (M), and the outer rim of the Mantel zone (O) bordering the red pulp (R) with hemosiderin laden macrophages.
(b) H&E-stained section of spleen from the four-week control, 0.0 ppm AFB 1, dose group, magnification x600. The outer rim of the Mantel zone (O) bordering
the red pulp (R) with hemosiderin laden macrophages (small arrows) and few hyper segmented neutrophils (long arrows) are shown. (c) H&E-stained section
of spleen from the eight-week continuous, high dose, 1.6(C), group, magnification x100. The micro-architecture (compartmentalization) of the spleen with the
darkly stained PALS and the Mantel zones are shown. Note the confluence of the Mantel zones occupying most of the red pulp and the paucity of discernable
hemosiderin-laden Macrophages in the outer rim of the Mantel zone. (d) H&E-stained section of spleen from the eight-week continuous, high dose, 1.6(C), group,
magnification x600. Shown is the outer rim of the Mantel zone and the red pulp lacking hemosiderin laden macrophages. Few segmented neutrophils are present.
(e) H&E-stained section of spleen from the 12-week continuous, high, 1.6(C) ppm dose group, magnification x100. Note the irregularity in shape and the
decreased size of the Mantel zone comprised of cells surrounding smaller PALS compared to the eight-week, Figure 8c. (f) H&E-stained section of spleen from
the 12-week continuous, high dose, 1.6(C), group, magnification x600. The outer rim of the Mantel zone bordering the red pulp with lesser numbers of
hemosiderin-laden macrophages and few hyper segmented neutrophils compared to the control, 0.0 ppm dose group are shown.
IMMUNOTOXICITY STUDY OF AFLATOXIN B 1 IN F344 RATS
(1985b). Effects of purified aflatoxin on broiler chickens. Poult. Sci. 64,
852– 858.
Henry, S. H., Bosch, F. X., and Bowers, J. W. (2002). Aflatoxin, hepatitis and
worldwide liver cancer risks. Adv. Exp. Med. Biol. 504, 229 –233.
Hinton, D. M. (2000). U.S. FDA Redbook II immunotoxicity testing guidelines
and research in immunotoxicity evaluations of food chemicals and new food
proteins. Toxicol. Pathol. 28, 467– 478.
Hinton, D. M., Jessop, J. J., Arnold, A. A., Albert, R. H., and Hines, F. A.
(1987). Evaluation of immunotoxicity in a subchronic feeding study of
triphenyl phosphate. Toxicol. Ind. Health 3, 81– 89.
Hinton, D. M., Myers, M. J., Perlloni, A., Hines, F., Raybourne, R. B.,
Sotomayor, R. E., Shaddock, J., Warbritton, A., Chou, M., and Pestka, J. J.
(2001). Enhanced histopathology with morphometry, immuno-and in-situ
staining in immunotoxicity studies of the mycotoxins, aflatoxin B1 (AFB 1)
and deoxnivalenol (DON). Toxicologist 60, 1947A.
Jakab, G. J., Hmieleski, R. R., Zarba, A., Hemenway, D. R., and Groopman,
J. D. (1994). Respiratory aflatoxicosis: Suppression of pulmonary and systemic host defenses in rats and mice. Toxicol. Appl. Pharmacol. 125,
198 –205.
Kew, M. C. (1992). Tumours of the liver. Scand. J. Gastroenterol. Suppl. 192,
39 – 42.
Kodell, R. L., Gaylor, D. W., and Chen, J. J. (1987). Using average lifetime
dose rate for intermittent exposures to carcinogens. Risk Anal. 7, 339 –345.
Kuester, R. K., Waalkes, M. P., Goering, P. L., Fischer, B. L., McCuskey,
R. S., and Sipes, I. G. (2002). Differential hepatotoxicity induced by
cadmium in Fischer 344 and Sprague-Dawley rats. Toxicol. Sci. 65, 151–
159.
Liu, B. H., Yu, F. Y., Chan, M. H., and Yang, Y. L. (2002). The effects of
mycotoxins, fumonisin B1 and aflatoxin B1, on primary swine alveolar
macrophages. Toxicol. Appl. Pharmacol. 180, 197–204.
Lyte, M., Blanton, R. H., Myers, M. J. and Bick, P. H. (1987). Effect of in vivo
administration of the environmental pollutant benzo(a)pyrene on interleukin-2 and interleukin-3 production. Int. J. Immunopharm. 9, 307–312.
Mancini, R., Paolucci, F., Svegliati Baroni, G., Jezequel, A. M., and Orlandi,
F. (1991). Phenotypic analysis of inflammatory infiltrate in rats with dimethylnitrosamine-induced cirrhosis. Int. J. Exp. Pathol. 72, 119 –128.
Methenitou, G., Maravelias, C., Athanaselis, S., Dona, A., and Koutselinis, A
(2001). Immunomodulative effects of aflatoxins and selenium on human
natural killer cells. Vet. Hum. Toxicol. 43, 232–234.
Mocchegiani, E., Corradi, A., Santarelli, L., Tibaldi, A., DeAngelis, E.,
Borghetti, P., Bonomi, A., Fabris, N., and Cabassi, E. (1998). Zinc, thymic
endocrine activity and mitogen responsiveness (PHA) in piglets exposed to
maternal aflatoxicosis B1 and G1. Vet. Immunol. Immunopathol. 62, 245–
260.
Morris, S. M., Aidoo, A., Chen, J. J., Chou, M. W., and Casciano, D. A.
(1999). Aflatoxin B1-induced Hprt mutations in splenic lymphocytes of
Fischer 344 rats. Results of an intermittent feeding trial. Mutat. Res. 423,
33–38.
Murdoch, D. J., Krewski, D., and Wargo, J. (1992). Cancer risk assessment
with intermittent exposure. Risk Anal. 12, 569 –577.
Myers, M. J., Farrell, D. E., Baker, J. D., Evock-Clover, C. M., and Steele,
N. C. (1999). Endotoxin challenge differentially affects cytokine production
and metabolic status of growing and finishing swine. Domest. Anim. Endocrinol. 17, 345–360.
377
Myers, M. J., Farrell, D. E., Evock-Clover, C. M., Cope, C. V., Henderson, M.,
and Steele, N. C. (1995). Effect of recombinant growth hormone and
chromium picolinate on cytokine production and growth performance in
swine. Pathobiology 63, 283–287.
Park, D. L., Nesheim, S. Trucksess, N. W., Stack, M. E., and Newell, R. F.,
(1990). Liquid chromatographic method for determination of aflatoxins B 1,
B 2, G 1, and G 2, in corn and peanur products: A collaborative study.
J.A.O.A.C. Int. 19, 157–170.
Pier, A. C., Fichtner, R. E., and Cysewski, S. J. (1977). Effects of aflatoxin on
the cellular immune system. Ann. Nutr. Aliment. 31, 781–788.
Raisuddin, S., Singh, K. P., Zaidi, S. I., Paul, B. N., and Ray, P. K. (1993).
Immunosuppressive effects of aflatoxin in growing rats. Mycopathologia
124, 189 –194.
Raisuddin, S., Singh, K. P., Zaidi, S. I., Saxena, A. K., and Ray, P. K. (1990).
Effects of aflatoxin on lymphoid cells of the weanling rat. J. Appl. Toxicol.
10, 245–250.
Raisuddin, S., Zaidi, S. I., Singh, K. P., and Ray P. K. (1991). Effect of
subchronic aflatoxin exposure on growth and progression of Ehrlich’s ascites tumor in mice. Drug Chem. Toxicol. 15, 185–206.
Reddy, R. V., Taylor, M. J., and Sharma, R. P. (1987). Studies of immune
function of CD-1 mice exposed to aflatoxin B1. Toxicology 43, 123–132.
Rosenthal, G. J., Lebetkin, E., Thigpen, J. E., Wilson, R., Tucker, A. N., and
Luster, M. I. (1989). Characteristics of 2,3,7,8-tetrachlorodibenzo-p-dioxin
induced endotoxin hypersensitivity: Association with hepatotoxicity. Toxicology 56, 239 –251.
Sahu, S. C., Chou, M. W., Sotomayor, R. E., Hinton, D. M., Barton, C. N., and
O’Donnell, M. W. (1999). Effects of intermittent exposure of aflatoxin b1 on
hepatic and testicular glutathione s-transferase in rats. J. Appl. Toxicol. 20,
215–219.
Schook, L. B., Lockwood, J. F., Yang, S.-D., and Myers, M. J. (1992).
Dimethylnitrosamine (DMN)-induced IL-1␤, TNF␣ and IL-6 inflammatory
cytokine expression. Toxicol. Appl. Pharmacol. 116, 110 –116.
Sharma, R. P. (1993). Immunotoxicity of mycotoxins. J. Dairy Sci. 76,
892– 897.
Sotomayor, R. E., Sahu, S., Washington, M., Hinton, D. M., and Chou, M.
(1999). Temporal patterns of dna adduct formation and glutathione stransferase activity in the testes of rats fed aflatoxin B1: A comparison with
patterns in the liver. Environ. Mol. Mutagen. 33, 293–302.
Venturini, M. C., Perfumo, C. J., Risso, M. A, Gomez, C. M., Piscopo, M. V.,
Sala de Miguel, M., and Godoy, H. (1990). Effect of aflatoxin B1 on
resistance induced by Bordetella bronchiseptica vaccine in rabbits. Vet.
Microbiol. 25, 209 –216.
Wang, J. S., Huang, T., Su, J., Liang, F., Wei, Z., Liang, Y., Luo, H., Kuang,
S. Y., Qian, G. S., Sun, G., et al. (2001). Hepatocellular carcinoma and
aflatoxin exposure in Zhuqing Village, Fusui County, People’s Republic of
China. Cancer Epidemiol. Biomark. Prev. 10, 143–146.
Wogan, G. N. (1976). The induction of liver cell cancer by chemicals. In Liver
Cell Cancer (H. M. Cameron, D. A. Linsell, and G. P. Warwick, Eds.), pp.
121–151. Amsterdam: Elsevier/North Holland Pub. Co.
Wogan, G. N. (1992). Aflatoxins as risk factors for hepatocellular carcinoma
in humans. Cancer Res.52(Suppl.), 2114 –2118, 121–151.
Wogan, G. N. (2000). Impacts of chemicals on liver cancer risk. Semin. Cancer
Biol. 10, 201–210.