Focal granulomatous inflammation developed in the livers of five 10-week-old male Sprague-Dawley rats. The characteristic features of this lesion were the presence of foreign body multinucleated giant cells engulfing calcium deposits and site-specific development in a fissure formed in a sub-lobation in the left lobe or interlobar fissure of the medial lobe of the liver. To clarify the pathogenesis of this lesion, rat livers showing abnormal sub-lobation or lobar atrophy, rat livers in an acute dermal toxicity study and guinea pig livers in a skin sensitization test were also examined histologically. Consequently, the present lesion was considered to be a reactive change against calcium that was dystrophically deposited in the area of hepatocellular necrosis due to delayed circulatory disturbance caused by external pressure or extension force. Granulomatous lesions like in the present cases should be differentiated from those caused by evident exogenous pathogens such as chemicals or microorganisms.
Focal granulomatous inflammation developed in the livers of five 10-week-old male Sprague-Dawley rats. The characteristic features of this lesion were the presence of foreign body multinucleated giant cells engulfing calcium deposits and site-specific development in a fissure formed in a sub-lobation in the left lobe or interlobar fissure of the medial lobe of the liver. To clarify the pathogenesis of this lesion, rat livers showing abnormal sub-lobation or lobar atrophy, rat livers in an acute dermal toxicity study and guinea pig livers in a skin sensitization test were also examined histologically. Consequently, the present lesion was considered to be a reactive change against calcium that was dystrophically deposited in the area of hepatocellular necrosis due to delayed circulatory disturbance caused by external pressure or extension force. Granulomatous lesions like in the present cases should be differentiated from those caused by evident exogenous pathogens such as chemicals or microorganisms.
In laboratory animals, including mice, rats, dogs and primates, focal aggregates of
inflammatory cells such as microgranulomas are common spontaneous lesions in the
liver, irrespective of age
. However, the cause of this focal inflammatory change remains to be
clarified, although the participation of agents from the intestinal tract via the
portal blood flow, such as bacteria and endotoxins, is suspected. With aging,
spontaneous occurring microgranulomas in rat livers increase in size, show a diffuse
distribution and finally appear as granulomatous inflammation in severe cases. On
the other hand, larger-sized focal granulomatous lesions as in the present cases are
rare, and their possible cause is difficult to clarify. Chemicals, including carbon
tetrachloride, are known to induce granulomas in the livers of laboratory
animals
,
. Since it is difficult to distinguish morphologically between spontaneous
and induced granulomatous lesions, it is important to clarify the possible cause of
spontaneous lesions.In regard to hepatocellular necrosis caused by circulatory disturbance, the
occurrence of infarct in the liver is usually rare because the liver has two sources
of blood supply, the hepatic artery and portal vein
. As one of the possible causes of circulatory disturbance, wrapping of the
torsos of rats has been reported to produce liver injuries, including centrilobular
coagulative necrosis
, inflammatory cell infiltration, fibrosis and granulomatous
inflammation
. Similar lesions are known to occur in women wearing severely restrictive
corsets, and this is referred to as corset liver
. Experimental investigations on the effects of wrapping the torso with a
bandage covering the entire body have been carried out using rats
, guinea pigs
and dogs
. In general, wrapping or bandaging seems to cause congestion in the livers
of humans
and animals
,
,
and ischemia following the congestion induces hepatocellular necrosis and
the accompanying changes.The focal granulomatous inflammation in the liver of the rats in the present cases
was characterized by histological findings such as the presence of foreign body
multinucleated giant cells engulfing calcium deposits. This lesion was found in 5
animals, and all of their regional development sites were at the tip of a fissure (1
case with a fissure in a sub-lobation and 4 cases with an interlobar fissure in the
medial lobe) in the liver. Considering the site-specific occurrence, circulatory
disturbance due to physical external pressure or extension force is suspected to
play a possible important role in the pathogenesis of these lesions, because these
sites are easy to move by external pressure or extension force. The site of the
medial lobe is known to develop tension lipidosis frequently as a result of tension
from the ligamenta falciforme hepatis. To obtain clues about the
pathogenesis of this focal granulomatous inflammation in the liver, we examined
various materials histologically as follows: 1) rat livers with sub-lobation in the
left lobe (to examine the presence or absence of aberrant arteries at the site
showing sub-lobation), 2) a rat liver showing spontaneously occurring lobar atrophy
in the caudate lobe (to compare with histological characteristics in lesions induced
by hypoxia possibly due to circulatory disturbance in the portal vein), 3) livers of
rats in an acute dermal toxicity study that were subjected to wrapping of the torso
once with a bandage applied to the body (to examine the effect of external pressure
on the rat liver as a result of wrapping the torso once and to certify the presence
or absence of aberrant arteries at the tip of the interlobar fissure) and 4) livers
of guinea pigs in a skin sensitization test that were subjected once or multiple
times to wrapping of the torso with a bandage applied to the body (to examine the
effect of external pressure on the guinea pig liver as a result of wrapping the
torso once or multiple times). Based on the results of these observations, an
attempt was made to clarify the pathogenesis of the present five cases of foreign
body granulomatous inflammation.
Materials and Methods
The lesion was observed in five 10-week-old, male Sprague-Dawley rats (Charles River
Japan Laboratories, Atsugi, Japan) in the untreated control group of a toxicity
study. The information on each animal is presented in Table 1. Of these cases, Case No.1 was histologically examined
in detail using various methods, and the other cases were limited to
histopathological examinations using hematoxylin and eosin (HE) stained sections
only.
Table 1
Summary of the Animals Bearing a Focal Hepatic Granulomatous
Inflammation
After a macroscopic examination, the livers of the 5 cases were fixed in 10% neutral
buffered formalin solution, trimmed, processed routinely and embedded in paraffin.
Block specimens were sectioned at 3 μm thickness and stained with
HE; the special stainings for Case No. 1 were as follows: von Kossa’s silver method,
periodic-acid Schiff (PAS) reaction, Schmorl’s reaction, Perl’s Prussian blue stain
and elastica Van-Gieson (EVG) stain. In addition, immunohistochemical stainings were
also performed by the labeled streptavidin-biotin method with several antibodies,
including anti-porcine vimentin antibody (Dako A/S, Glustrup, Denmark), α-smooth
muscle actin (α-SMA) antibody (Dako A/S), anti-rat macrophage antibody (clone: ED-1,
Serotec Ltd., Oxford, England) and anti-rat proliferating cell nuclear antigen
(PCNA) antibody (Dako A/S) as the primary antibodies. Pretreatment was conducted by
microwave oven heating for vimentin and PCNA and pepsin digestion for ED-1.
Immunoreaction with the primary antibodies was allowed to occur for 12 hours at 4°C.
All procedures, except for application of the primary antibody, were carried out in
a moist chamber at room temperature in the following sequence. The staining was
performed according to the manufacturer’s instructions (LSAB2 kit; Dako Japan, Co.,
Ltd., Kyoto, Japan). The reaction was visualized with 3,3’-diaminobenzidine in 0.1 M
tris HCl buffer (pH 7.6) plus hydrogen peroxide. Sections were washed with TBS-T
(Tris buffer solution plus Tween 20) at each step. Finally, all the sections were
counterstained with Mayer’s hematoxylin.For the following items (1 to 4), other livers than those from the present cases were
re-examined grossly depending on the cases, and a histological examination of the
liver-tissue specimens (HE stain), including semiserial sections, was performed to
clarify the pathogenesis of the present cases.1) Rat liver tissue showing abnormal sub-lobationLiver tissues showing abnormal sub-lobation in the left lateral lobe similar
to that in Case No. 1 were obtained from two 8-week-old, male Wistar Han
rats (CLEA Japan, Inc., Japan). To ascertain the aberrant arterial vessels
at the tip of a fissure forming an abnormal sub-lobation, histopathological
examination using semiserial sections was also performed.2) Rat liver tissue showing lobar atrophy of the caudate
lobeThis case was a 10-week-old male Sprague-Dawley rat (Charles River Japan
Laboratories) in the untreated control group of a toxicity study. Grossly,
both the cranial and caudal parts of the caudate lobe were atrophied. The
portal veins could not be examined grossly or microscopically because the
materials were already fixed with formalin after necropsy at another
laboratory.3) Livers from rats in an acute dermal toxicity studyThe torsos of ten 10-week-old Wistar Han rats (CLEA Japan, Inc.), both males
and females, were wrapped on with a surgical tape (3 M) after dermal
exposure to a test substance placed as a patch 4 × 5 cm). At 24 hours after
a single exposure, the patch and residual test substance were removed, and
the skin application site was cleaned using distilled water. The test
substance did not induce any toxicity in the rats. The gross and
histological findings of the effects, if any, of wrapping the torso with a
bandage and the presence of aberrant arteries at the tips of the fissures
found in the medial lobe of the livers were examined.4) Livers from guinea pigs used in a skin sensitization
testThe torsos of six-week-old Hartley strain guinea pigs (Japan SLC Inc.,
Japan) were wrapped with an elastic bandage (3 M); there were 4 animals with
1 exposure for 6 hours (single application) and 20 animals with 4 exposures
for 6 hours each (multiple applications). The animals were euthanized by
exsanguinations under ether anesthesia at 2 weeks after the final treatment.
Seven untreated, age-matched guinea pigs were used as controls.The procedures for handling and care of the animals used in the present study
were approved by the Animal Care Committee of BSRC.Case No. 1: a) The lesion was composed of a fibrous band connecting the
diaphragmatic and visceral surfaces of the liver (arrows) and was demarcated
clearly from the surrounding liver tissue. HE stain, ×36. b) Higher
magnification of the portion in the open square in Fig. 1 a). The lesion consisted of many multinucleated
giant cells engulfing basophilic material (arrows). HE stain, ×360.
Fig. 1
Case No. 1: a) The lesion was composed of a fibrous band connecting the
diaphragmatic and visceral surfaces of the liver (arrows) and was demarcated
clearly from the surrounding liver tissue. HE stain, ×36. b) Higher
magnification of the portion in the open square in Fig. 1 a). The lesion consisted of many multinucleated
giant cells engulfing basophilic material (arrows). HE stain, ×360.
Case No. 1: a) Area outside of the lesion: No remarkable change. EVG stain,
×180. b) Area of the lesion: Fragmentation (arrows) and/or aggregation of
elastic fibers in the arterial vessels. C: central vein. P: portal area. EVG
stain, ×180.Case No. 1: Macrophages (a) and myofibroblasts and arterial vessels (b:
arrows) seen in the lesion. Immunostaining with anti-ED-1 antibody (a: ×180
) and anti-αSMA antibody (b: ×110).Case No. 1: Observations in serial sections. HE stain. a) Focal
hepatocellular necrosis (a: asterisks), ×36. b) High magnification of the
portion in the square in Fig. 4a. Focal
area of hepatocellular necrosis (asterisk) surrounded by multinucleated
giant cells with a foamy appearance (arrows), ×180. c) Dystrophic
calcification in the focal area of hepatocellular necrosis (arrows),
×90.
Fig. 4
Case No. 1: Observations in serial sections. HE stain. a) Focal
hepatocellular necrosis (a: asterisks), ×36. b) High magnification of the
portion in the square in Fig. 4a. Focal
area of hepatocellular necrosis (asterisk) surrounded by multinucleated
giant cells with a foamy appearance (arrows), ×180. c) Dystrophic
calcification in the focal area of hepatocellular necrosis (arrows),
×90.
Case No. 2: HE stain. a) Fibrous area with foci of multinucleated giant
cells (arrows), ×36. b) High magnification of the portion in the rectangle
in Fig. 5a; an artery (arrows) and vein
(arrowheads) are indicated within the fibrous area, ×110.
Fig. 5
Case No. 2: HE stain. a) Fibrous area with foci of multinucleated giant
cells (arrows), ×36. b) High magnification of the portion in the rectangle
in Fig. 5a; an artery (arrows) and vein
(arrowheads) are indicated within the fibrous area, ×110.
Results
Only the lesion of Case No. 1 was found grossly to be a white, elastic lesion formed
at a fissure in a sub-lobation in the left lateral lobe of the liver at necropsy.In regard to the histological findings for Case No. 1, the lesion was clearly
demarcated from the surrounding liver tissue and was composed of a fibrous band
containing arterial vessels (Fig. 1a). The
fibrous band showed characteristic inflammatory changes with infiltration by
macrophages with or without pigments, lymphocytes and a few eosinophils, and bile
ductular proliferation. The most characteristic finding was the presence of many
macrophages forming foreign body-type multinucleated giant cells engulfing a
basophilic substance (Fig. 1b).The special stains revealed deposition of hemosiderin and lipofuscin in the
macrophages, calcium deposits in the foreign body giant cells and fragmentation
and/or aggregation of elastic fibers in the arteries showing an abnormal
distribution (Figs. 2a and 2b: portal region outside of the lesion, and portal region in
the lesion, respectively). Immunostainings demonstrated macrophages (Fig. 3a) and myofibroblasts as the main cellular
components of the lesions, although their activities were weak as evidenced by PCNA
staining. In addition, branching arteries were observed in the lesion (Fig. 3b). The bile duct cells within the lesion
were vimentin-positive and weakly positive by PCNA staining.
Fig. 2
Case No. 1: a) Area outside of the lesion: No remarkable change. EVG stain,
×180. b) Area of the lesion: Fragmentation (arrows) and/or aggregation of
elastic fibers in the arterial vessels. C: central vein. P: portal area. EVG
stain, ×180.
Fig. 3
Case No. 1: Macrophages (a) and myofibroblasts and arterial vessels (b:
arrows) seen in the lesion. Immunostaining with anti-ED-1 antibody (a: ×180
) and anti-αSMA antibody (b: ×110).
Based on these findings, the present lesion was considered to be a pathological
reaction against calcium deposits. To investigate the source of the calcium
deposits, serial sections from the same paraffin block were prepared. Examination of
the serial sections revealed areas of focal hepatocellular necrosis (Fig. 4a) surrounded by active histiocytic
reactions, including multinucleated giant cell formation with a foamy appearance
(Fig. 4b) and calcium deposition in the
necrotic areas of the hepatocytes (Fig. 4c).
The foreign body giant cell formation was considered to be a reactive change against
necrotized hepatocytes, and the calcium deposits were considered to be derived from
the calcium that was dystrophically deposited in the necrotized hepatocytes.In regard to the other 4 cases, histopathological examinations of liver specimens
stained with HE were conducted. Foreign body-type multinucleated giant cells
engulfing calcium deposits (Fig. 5a) were seen
at the tip of the interlobar fissure of the medial lobe in all cases. The lesions
found in 2 out of the 4 cases contained arteries that are not usually present at the
tip of the interlobar fissure (Figs. 5 a, b).Liver showing lobar atrophy in the caudate lobe. Time course of the
calcified lesion, HE stain. a) Early stage: ×140, b) Middle stage: ×180, c)
Late stage: ×360.At the tip of the fissure (asterisk) of the medial lobe of the normal rat
liver. An interlobular artery (small arrows), vein (large arrow) and bile
duct (arrowheads) can be seen. HE stain, ×36.Liver of a guinea pig used in a skin sensitization test. Area of focal
hepatocellular necrosis. a) Early stage: hepatocytes showing granular
basophilic cytoplasm (arrows) and scant cellular reaction. HE stain, ×140.
b) Late stage: individual hepatocytes having homogeneous basophilic
cytoplasm accompanied by an evident cellular reaction including
multinucleated giant cells (arrows). HE stain, ×140. c) Early stage:
argyrophilic (arrows) basophilic substance seen in hepatocytes. Von Kossa
reaction, ×140.1) Rat liver tissue showing abnormal sub-lobationIn 2 of the cases examined, no aberrant arteries were observed at the tip of
the fissure that was formed by abnormal sub-lobation in the left lobe of the
liver.2) Rat liver tissue showing lobar atrophy of the caudate lobeAreas of hepatocellular necrosis with dystrophic calcification were
distributed multifocally in various stages of progression in the atrophic
caudal lobe (Figs. 6a-c). The necrotic
areas were surrounded by macrophages with deposits of pigments, lymphocytes
and fibrosis, and the late stage lesions contained multinucleated giant
cells (Fig. 6c). The later stage
lesions showed more condensed and homogenous calcium deposits. The hepatic
cell cords in the surrounding liver tissue were distorted and intermingled
with proliferating bile ductules.
Fig. 6
Liver showing lobar atrophy in the caudate lobe. Time course of the
calcified lesion, HE stain. a) Early stage: ×140, b) Middle stage: ×180, c)
Late stage: ×360.
3) Livers from rats used in an acute dermal toxicity studyThere were no gross or histological abnormalities in the livers of any of
the animals, including in the tip of the interlobar fissure of the medial
lobe. One case showed liver lesions containing an artery and vein
originating from the adjacent portal triads (Fig. 7).
Fig. 7
At the tip of the fissure (asterisk) of the medial lobe of the normal rat
liver. An interlobular artery (small arrows), vein (large arrow) and bile
duct (arrowheads) can be seen. HE stain, ×36.
4) Livers from guinea pigs used in a skin sensitization testGrossly, white patches of varying size, ranging from less than 1 mm to 8×6
mm, were observed in the livers of most of the guinea pigs examined,
including the untreated controls. There were no differences in the lobar
distribution of the white patches or in incidence between the untreated and
treated animals. Histologically, the white patches consisted of areas of
focal hepatocellular necroses at varying stages of progression. Coagulative
necrosis of hepatocytes with granular or homogeneous basophilic cytoplasms
with a multinucleated giant cell reaction was seen (Figs. 8a, b). The basophilic cytoplasm of the
hepatocytes showed positive staining with von Kossa’s silver stain (Fig. 8c). However, no aberrant vessels
were observed at the tip of the interlobar fissure of the medial lobe in
these guinea pig livers.
Fig. 8
Liver of a guinea pig used in a skin sensitization test. Area of focal
hepatocellular necrosis. a) Early stage: hepatocytes showing granular
basophilic cytoplasm (arrows) and scant cellular reaction. HE stain, ×140.
b) Late stage: individual hepatocytes having homogeneous basophilic
cytoplasm accompanied by an evident cellular reaction including
multinucleated giant cells (arrows). HE stain, ×140. c) Early stage:
argyrophilic (arrows) basophilic substance seen in hepatocytes. Von Kossa
reaction, ×140.
Discussion
Based on similar findings seen in the livers of animals with a variety of
pathological conditions, the granulomatous inflammation in the liver in the present
cases was suggested to represent a reactive change to the calcium deposited in the
necrotizing hepatocytes in the infarct zone induced by circulatory disturbance. The
site-specific occurrence of this lesion may be a flexible or easy to move by an
external pressure or extension force. Therefore, we suspected that external pressure
or extension force is the trigger of delayed circulatory disturbance, as seen in
animals with an enwrapped torso exhibiting the effect of external pressure or like
the medial lobe infarct of rats caused by an extension force from the
ligamentum falciforme hepatis.In regard to the absence of hepatocellular necrosis in the rats from the acute dermal
toxicity study, the reasons for the differences in the findings from reported by
Nyska et al.
are unclear. Our cases were relatively older, being about 10 weeks old,
compared with the rats they used, and the total application time (6 hours, 1 time)
was shorter compared with used for their rats (6 weeks at age, 6 hours/ day/2
weeks). The development of the lesions may depend on the time and duration of
bandage application and the rat’s age. On the other hand, histological examination
of the white patches found on gross examination in both treated and untreated guinea
pig livers revealed hepatocellular necrosis. However, granular or homogeneous
deposits of calcium in the foci of necrotized hepatocytes and in the subcapsular
area were noted only in the treated animals. The pathogenic mechanism underlying the
calcium deposition in the necrotized hepatocytes may be accelerated hypoxia by
external pressure because, in general, calcium deposits occur under hypoxic
conditions
. Hepatocellular necrosis possibly developing as a result of external
pressure was also found in the subcapsular area of the guinea pig livers used in the
skin sensitization test
.The site-specificity of development of the present lesion suggested that the role of
aberrant arteries in inducing the lesions, in addition to the physical external
pressure. Although the presence of aberrant arteries at the tip of the interlobar
fissure or in the portion of the rat liver showing an abnormal sub-lobation was
rarely found, the combination of hypoxia due to circulatory disturbance and aberrant
arteries located at the tip of the interlobar fissure may be important in the
pathogenesis of this focal granulomatous inflammation of the liver.The changes found in the caudate lobe showing atrophy in the rat liver may be a
result of hypoxia by incomplete occlusion of the portal vein. A series of similar
changes, including hepatocellular necrosis, dystrophic calcification and foreign
body reactions, were also observed in this atrophied caudate lobe, suggesting a
similar delayed circulatory disturbance, as seen in this lesion, in pathogenesis.Several results and findings could be obtained from the present cases. Necrotic areas
in the liver will be gradually repaired, if not completely. The present cases would
equate to incomplete reparative changes. The cause of the granulomatous inflammatory
changes seen in the present cases may have role in the gradual hypoxic conditions
that allow calcium influx into cells.In conclusion, the granulomatous inflammation in the liver in the present study
probably represents a reparative change of focal hepatocellular necrosis induced by
hypoxia due to circulatory disturbance. The presence of aberrant arteries and
granulomatous lesions showing the various reparative stages in the narrow and
flexible areas suggested that the essential factor in their induction was their
gradual and multiple developments by external pressure or extension force. Although
evident exogenous factors such as chemicals and microorganisms are also known to
induce granulomatous lesions in the livers of laboratory animals, endogeneous
factors stimulating delayed circulatory disturbance as seen in this lesion should be
considered in differentiating among focal granulomatous lesions found in the livers
of laboratory animals.