M1/M2-macrophage Polarization-based Hepatotoxicity in d -galactosamine-induced Acute Liver Injury in Rats

Abstract

d-galactosamine (d-GalN) is a well-known hepatotoxic agent that causes liver injury. Conversely, hepatic macrophages play a crucial role in maintaining liver tissue integrity. Macrophage functions were investigated in hepatic lesions induced by a single intraperitoneal injection of d-GalN (800 mg/kg body weight [BW]) in 6-week-old F344 rats. Blood and liver samples were examined at 8 hr and on 1, 2, 3, and 5 days postsingle injection (PSI). Hepatic lesions consisting of degeneration/sporadic foci of coagulation necrosis, inflammatory cell reaction, and reparative fibrosis were seen on PSI days 1 and 2, reflected by significantly increased serum levels of aspartate transaminase and alanine transaminase and upregulation of CD68 M1 (tumor necrosis factor-α, interleukin [IL]-6, and interferon-γ) and CD163 M2 (transforming growth factor-β1, IL-10, monocyte chemoattractant protein-1, and IL-4) macrophage-related factors. Double immunofluorescence staining on PSI day 2 demonstrated that 82% of hepatic macrophages expressed of CD163/CD68 simultaneously; 65–75% of MHC class II macrophages showed co-expression of CD163 or CD68 and 95% CD204-expressing macrophages reacted to CD163 or CD68. These findings showed that both M1- and M2-macrophages contributed to the development of hepatic lesions induced by d-GalN and provided information about macrophage activation, indicating the importance of analysis of macrophage phenotypes for hepatotoxicity based on M1/M2-polarization.

The liver is one of the vital organs in mammals, playing roles such as nutritional metabolism and detoxification. Acute liver failure or fulminant hepatic failure is a drastic, unpredictable syndrome in which the severe deterioration of liver function rapidly occurs; such conditions may be typically followed by hepatic encephalopathy, coagulopathy, and, in some cases, progressive multiorgan failure (Bernal et al. 2010; Wlodzimirow et al. 2012). It has been reported that several chemical substances including acetaminophen, thioacetamide (TAA), and carbon tetra chloride (CCl₄) can cause hepatic disorders, and these chemicals have been used to develop liver lesions as animal models (Lee et al. 2008).

d-galactosamine (d-GalN) is a hexosamine-derived from galactose and constituent of some glycoprotein hormones such as follicle-stimulating hormone and luteinizing hormone. The d-GalN is also a hepatotoxic agent and has been well-studied as a model of human viral hepatic disease (Sugiyama et al. 1998; He, Noda, and Sugiyama 2001). Human viral hepatitis has emerged as a major public health problem throughout the world. The acute viral hepatitis may lead to chronic infection resulting in the development of cirrhosis and hepatocellular carcinoma (Wasley et al. 2008). Histopathologic findings of d-GalN-induced liver injury in rats resemble those of human acute viral hepatitis (Keppler et al. 1968). Functionally, d-GalN inhibits the synthesis of RNA and protein through a decrease in cellular uridine diphosphate concentration, which leads to focal necrosis and subsequent infiltration of inflammatory cells (Kmiec et al. 2000). It has also been reported that d-GalN causes inability of the cell to synthesize critical cell membrane components culminating in hepatocyte death (Decker and Keppler 1974; Shinozuka et al. 1973).

Macrophages are immune cells that are widely distributed throughout tissues in the body; out of them, hepatic macrophages including Kuppfer cells and interstitial dendritic cells in the Glissons’s sheath play a crucial role in regulating liver homeostasis and hepatic injury (Varol, Mildner, and Jung 2015). Following the exposure of toxicants, hepatic macrophages can show hypersensitivity, so that abundant release of mediators including pro-inflammatory and anti-inflammatory factors occur, which can foster acute liver injury or tissue repair (injury resolution/reparative fibrosis; Varol, Mildner, and Jung 2015; Ju and Tacke 2016; Ide et al. 2003). Hepatic macrophages in normal condition are divided into three types—exudate macrophages, resident macrophages, and antigen-presenting macrophages according to their morphology and immunophenotypes (Takahashi, Naito, and Takeya 1996; Hume 2008). The massive infiltration of macrophages is also a hallmark of acute and chronic liver injury (Antoniades et al. 2012; Liaskou et al. 2013). Hepatic macrophages in pathologic condition can change microenvironments, and the microenvironments may influence functions of hepatic macrophages. Therefore, hepatic lesions/hepatotoxicity may be evaluated based on the concept of M1/M2-macrophage polarization; M1-macrophages express CD68 antigen that shows high phagocytosis and cytotoxicity, whereas M2-macrophages express CD163 antigen that may play a role in tissue repair and reparative fibrosis (Duffield et al. 2005; Martinez et al. 2008). The properties of hepatic macrophages and M1/M2-polarization in d-GalN-induced rat liver injury have not yet been clarified.

The aims of the present study were to investigate the immunophenotypic characteristics of hepatic macrophages and also to focus on the functions of macrophages that take part in d-GalN- induced hepatotoxicity, on the basis of the M1/M2-macrophage polarization concept.

Material and Method

Animal Model

Twenty four, six-week-old F344 male rats (110–120 g BW) were obtained from Charles River, Yokohama, Japan, and used in this experiment after one-week acclimatization. Rats were housed in an animal room maintaining at 20–22°C temperature with 12-h light–dark cycle. The animals were fed a standard diet (DC-8; CLEA Japan, Tokyo, Japan) and tap water ad libitum. Rats were injected intraperitoneally with d-GalN (Sigma Aldrich, St. Louis, MO, dissolved in 0.9% NaCl at a dose of 800 mg/kg BW). Four control rats were injected with an equal volume of 0.9% NaCl through the same route (intraperitoneally). Four treated rats were examined histopathologically on each time point of PSI at 8 hr and on 1, 2, 3, and 5 days and controls rats were examined histopathologically on PSI day 1. One hour before euthanasia, all rats were injected intraperitoneally with bromo-2-deoxyuridine (BrdU; Sigma Aldrich Co., St. Louis, MO) dissolved in 0.9% NaCl at 50 mg/kg BW. After collecting blood, all rats were euthanized by deep isoflurane anesthesia, and liver samples were collected for further analyses. The blood was allowed to clot after collecting from the abdominal aorta. Then, serum was separated and analyzed for aspartate transaminase (AST) and alanine transaminase (ALT) with a clinical analyzer (SRL Inc., Tokyo, Japan). The animal experiments were conducted for animal care (no. 29–5) under the institutional guidelines approved by the ethical committee of Osaka Prefecture University, Japan.

Histopathology and Immunohistochemistry Staining

Liver tissues were collected from the left lobes of liver sections and fixed in 10% neutral-buffered formalin (NBF) or periodate-lysine-paraformaldehyde (PLP) solution which was further processed by PLP-Amex (acetone, methyl-benzoate, and xylene) method as described (Golbar et al. 2011). NBF-fixed tissues were dehydrated, embedded in paraffin, and then cut at 4 μm thickness. Then, sections were stained with hematoxylin and eosin for histopathological examinations.

PLP-fixed tissue sections were used for immunohistochemical analyses with primary antibodies listed in Table 1. After pretreatments, Histostainer (Histofine, Nichirei Bioscience Inc., Tokyo, Japan) was used for tissue staining. Briefly, the tissue sections were incubated with 5% skimmed milk in PBS for 10 min, followed by 1 hr incubation with primary antibodies. Then, the sections were treated with 3% H₂O₂ in PBS for 15 min and were incubated for 30 min with horseradish peroxidase-conjugated secondary antibody (Histofine simple stain MAX PO; Nichirei Inc., Tokyo, Japan). Positive reactions were observed with 3,3’-diaminobenzidine (DAB; Nichirei Inc., Tokyo, Japan) for 10 min and the slides were counterstained lightly with hematoxylin. As negative controls, tissue sections were treated with mouse nonimmunized serum instead of the primary antibody.

Table 1.

Primary Antibodies Used for Immunohistochemistry.

Antibody	Sample Type	Host	Fixative	Dilution	Pretreatment	Source
CD68	Paraffin	Mouse	PLP	1/300	MW in citrate buffer 20 min	Dako, Denmark, A/S
CD163	Frozen	Mouse	PLP	1/100	Acetone: methanol (1:1), 10 min	AbD Serotec, Oxford, UK
MHC class II	Paraffin	Mouse	PLP	1/10,000	100 μg/ml proteinase K, 10 min	BIO-RAD
CD204	Paraffin	Mouse	PLP	1/1,000	MW in citrate buffer 20 min	Transgenic Inc., kumamoto, Japan
BrdU	Paraffin	Mouse	PLP	1/500	HCl, 30 min and 100 μg/ml proteinase K, 10 min	Dako, Crop, Glostrup, Denmark

Note. PLP = periodate-lysine-paraformaldehyde; MW = microwave; BrdU = bromo-2-deoxyuridine.

Terminal Deoxyribonucleotide Transferase (TdT)–mediated Deoxyuridine Triphosphate Nick End Labeling (TUNEL)

A standard in situ TUNEL method (Appop Tag^R Peroxidase In situ Apoptosis Detection Kit, Millipore, Bedford) was used for the examination of apoptotic cells according to the manufacturer’s instructions. The formalin-fixed tissue sections were dewaxed, washed with PBS, and then treated with proteinase K (100 μg/ml) for 15 min at room temperature. Afterward, tissue sections were treated with 3% H₂O₂ for 5 min to inactivate endogenous peroxidases. Sections were incubated with TdT enzyme for 60 min at 37°C and the reaction was stopped by applying stop buffer for 10 min in room temperature. Then, sections were treated with horseradish peroxidase-conjugated anti-digoxigenin for 30 min at room temperature. The sections were then visualized with DAB. Negative control sections were incubated with distilled water instead of TdT enzyme.

Double Immunofluorescence Staining

To find out the CD68 M1- and CD163 M2-macrophage polarization, co-expression of cell specific antigens was investigated by using double immunofluorescence with combination of CD163/MHC class II, CD163/CD204, CD68/MHC class II, CD68/CD204, and CD163/CD68. Fresh frozen liver tissue sections (10-μm thick) from control and d-GalN-treated rats on PSI day 2 were used. The sections were fixed with acetone/methanol (1:1) for 10 min at 4°C, air dried for 30 min, and washed with PBS. Sections were blocked with 10% normal goat serum for 30 min and treated with primary antibodies overnight at 4°C. Then, sections were washed and allowed to react with secondary antibody goat anti-mouse IgG conjugated with Alexa 488 or Alexa 568 (Invitrogen, Carlsbad, CA) for 45 min at room temperature. The sections were incubated with second primary antibody followed by fluorescent dye-conjugated secondary antibody. Alexa 568 nonlabeled CD163 (AbD Serotec, Oxford, UK) was used for combination of CD163/CD68, CD163/CD204, and CD163/MHC class II. Alexa 568 nonlabeled CD68 (Dako, Denmark A/S) was used for combination of CD68/CD204 and CD68/MHC class II. The slides were visualized with Fuluoro-KEEPER antifading reagent containing 4’, 6-diamidino-2-phenylindole (Nakarai Tesque, Kyoto, Japan) for nuclear staining and analyzed by a virtual slide scanner (VS-120, Olympus, Tokyo, Japan).

Real-time Reverse Transcriptase Polymerase Chain Reaction (RT-PCR)

Liver samples were immersed in RNA later^R (Qiagen GmbH, Hilden, Germany) and stored at −80°C until use. Total RNA from the liver was extracted by using SV total RNA isolation system kit^R (Promega, Madison, WI) according to manufacturer’s instruction. The concentration of RNA was determined by using a nanodrop 1000^™ spectrophotometer (Thermo Scientific, Wilmington, DE) and 2.5 μg of total RNA was reverse-transcribed to cDNA by using a superscript VILO reverse transcriptase (Invitrogen). Real-time PCR was conducted by using TaqMan gene expression assays (Life Technologies, Carlsbad, CA) in a PikoReal Real-time PCR System (Thermo Scientific). Probes used are shown in Table 2. The mRNA expression was normalized against that of 18 s rRNA as an internal control. The data were calculated using the comparative C_t method (^ΔΔC_t method).

Table 2.

Real-time RT-PCR Probes.

Probes Name	Assay ID
IL-6	Rn 01410330_m1
IL-4	Rn 01456866_m1
IL-1β	Rn 00580432_m1
IL-10	Rn 00563409_m1
TNF-α	Rn 01525859_g1
IFN-γ	Rn 00563409_m1
TGF-β1	Rn 00572010_m1
MCP-1	Rn 00580555_m1
18S rRNA	Hs99999901_s1

Note. IL-6 = interleukin-6; IL-4 = interleukin-4; IL-1β = interleukin-1β; IL-10 = interleukin-10; TNF-α = tumor necrosis factor-alpha; IFN-γ = interferon-gamma; TGF-β1 = transforming growth factor-β1; MCP-1 = monocyte chemoattractant protein-1; and 18S rRNA = eukaryotic 18S rRNA.

Data Analyses

Cells expressing CD68, CD163, MHC class II, and CD204 were counted per 10× field, from randomly selected five areas in the perivenular (PV) portion and periportal (PP) portion including Glisson’s sheath. BrdU and TUNEL-positive cells were counted per 40× field and were compared with the total number of hepatocytes. Quantitative data were shown as mean ± standard deviation, and statistical analysis was performed using Dunnett’s multiple comparison tests. Significance was considered at p < .05.

Results

d-GalN-induced Hepatic Lesions and BrdU Immunohistochemistry

The serum levels of AST and ALT were significantly increased on PSI days 1 and 2 with a peak on PSI day 1, in contrast to those of the control groups where serum samples were collected on PSI day 1. These levels returned quickly to control levels on PSI day 3 onward (Figure 1A and B).

Figure 1.

Biochemical analyses on the serum levels of aspartate transaminase (AST) (A) and alanine transaminase (ALT) (B) in control and d-galactosamine (d-GalN)-injected rats. Compared to those of control, the levels of AST and ALT are significantly increased on postsingle injection days 1 and 2 in d-GalN-injected rats; n = 4; *versus control at p < .05 by Dunett’s test.

Histological lesions were shown in Table 3. Hepatic lesions induced by d-GalN were characteristically seen as degeneration and coagulative necrosis (as a single cell or multifoci of several cells) in both the PV and PP areas (Figure 2A), but prominent lesions were seen in the PP areas with scattered involvement of the midzonal and central lobular areas. Hepatic lesions developed on PSI days 1 and 2, and the most pronounced lesions were seen on PSI day 2. Additionally, infiltration of inflammatory cells was seen in the hepatic lesions (Figure 2C), and mesenchymal cells (apparently fibroblasts) with collagen deposition, demonstrable with azan-Mallory stain, appeared in and around these lesions (Figure 2D). On PSI days 3 and 5, the affected areas were replaced by regenerating hepatocytes. BrdU-positive hepatocytes (Figure 3A) were seen throughout hepatic lobules, with a significantly increased number on PSI day 2 (Figure 3C), indicating the regeneration of hepatocytes.

Table 3.

Time-dependent Histological Changes in the Liver after d-GalN-induced Injury.

Time Point	Histological Changes in the Liver after d-GalN-induced Injury
8 h	Sporadic single-cell necrosis
Day 1	Degeneration, coagulative necrosis (as single cells and foci of several cells) in both PV and PP areas, infiltration of inflammatory cells, and bile duct hyperplasia
Day 2	Degeneration, coagulative necrosis (as single cells and foci of several cells) in both PV and PP areas, infiltration of inflammatory cells, mesenchymal fibroblast reaction, and bile duct hyperplasia
Day 3	Minimal single-cell necrosis and regeneration
Day 5	Disappearance (complete recovery) of lesions

Note. PV = perivenular; PP = periportal.

Figure 2.

Histopathological changes of livers in d-galactosamine-injected rats on postsingle injection (PSI) day 2 (A, B, and C). (A) Degeneration, coagulative necrosis (circle), inflammatory cell reaction (arrow), and bile duct proliferation are seen in both perivenular and periportal areas. (B) Coagulative necrosis (as single cells indicated by arrows and foci of several cells noted by circles) is present in the areas of injury. (C) Infiltration of inflammatory cells is seen in the injured area. (D) Collagen deposition with mesenchymal filtration reaction (asterisks) is observed in and around the injured area on PSI day 2. (A–C) hematoxylin and eosin stain; (D) azan-Mallory stain. CV = central vein; GS = Glisson’s sheath. Bar = 200 μm (A) and 50 μm (B, C, and D).

Figure 3.

Proliferating marker bromo-2-deoxyuridine (BrdU) (A and C) for proliferating cells, as well as terminal deoxyribonucleotide transferase–mediated deoxyuridine triphosphate nick end labeling (TUNEL) method for apoptosis (B and D) in d-galactosamine-injected rats on postsingle injection (PSI) day 2. Reacting cells for BrdU (A, arrows) and TUNEL (B, arrows) are seen in the injury areas. BrdU (C) and TUNEL (D)-positive cells are seen significantly increased on PSI days 2 and day 1, respectively. Counterstained with hematoxylin (A–B); n = 4; *versus control at p < .05 by Dunett’s test.

TUNEL Method for Apoptosis

There were a few TUNEL-positive cells in control livers. TUNEL-positive cells, indicative of apoptotic cells and hepatocyte necrosis, were seen in the injured areas (Figure 3B), and a significantly increased number of positive cells were seen on PSI day 1 compared to that of controls (Figure 3D), indicating the involvement of apoptosis in the d-GalN-induced hepatic lesions.

Immunophenotypes of Hepatic Macrophages

For immunophenotypical analyses, we used four different antibodies against CD68 (for M1-macrophages), CD163 (for M2-macrophages), MHC class II, and CD204. The kinetics of macrophages positive for these antibodies is shown in Figure 4A–D.

Figure 4.

Kinetics of macrophages reacting to CD68 (A), CD163 (B), MHC class ΙΙ (C), and CD204 (D) in PV and PP areas including Glisson’s sheath of livers of control and d-GalN-injected rats; n = 4; * = perivenular area; ^† = periportal area at p < .05 by Dunett’s test.

CD68 and CD163 Immunostaining

The number of CD68-positive M1-macrophages in the PV and PP areas was gradually increased, showing a significant increase on PSI day 2, and then, returned to the control level on PSI day 3 onward (Figure 4A). CD163-positive M2-macrophages were seen along the sinusoids in control livers, indicative of Kupffer cells (Golbar et al. 2012); the number of cells reacting to CD163 in the PV and PP areas was significantly increased on PSI days 1 and 2 with a peak on PSI day 2, and the positive cells returned to control levels on PSI day 3 onward (Figure 4B). The presence of both CD68- and CD163-positive cells was frequently seen in the injured areas of PV and PP areas (Figure 5B and D; Online Supplemental Figure 1B and D). Morphologically, CD68-reacting cells showed round to spindle configuration (Figure 5B, inset), whereas CD163-positive cells were large round to oval in shape (Figure 5D, inset). The numbers of CD163 cells were consistently greater than that of CD68 cells (Figure 4A and B).

Figure 5.

Immunoreactivity in the perivenular (PV) areas of livers of control (A, C) and d-GalN-injected (B, D) rats on postsingle injection day 2. Macrophages reacting to CD163 (C) in controls are located along the sinusoids, indicating Kupffer cells. Increased number of macrophages reacting to CD68 (B), CD163 (D) is seen in PV areas of d-GalN-injected rats. Arrows indicate immunopositive cells. Respectively, CD68 and CD163-positive cells show round to spindle (B, inset) and round to oval (D, inset) shapes. Counterstained with hematoxylin, CV = central vein, bar = 100 μm.

MHC Class II and CD204 Immunostaining

In control livers, a small number of MHC class II–positive cells was seen in the PV and PP areas. In d-GalN-injected rats, the positive cells began to be increased gradually from 8 hr in PV areas and, on PSI days 1 and 2, showed a statistical increase in both areas (Figure 4C; Figure 6B and Online Supplemental Figure 2B); the positive cell numbers returned to control levels on PSI day 3 onward. Morphologically, MHC class II–positive cells were spindle to dendritic (Figure 6B, inset).

Figure 6.

Immunoreactivity in perivenular area of livers of control (A, C) and d-GalN-injected (B, D) rats. Macrophages reacting to CD204 (C) in controls are located along the sinusoids, indicating Kupffer cells. Macrophages reacting to MHC class ΙΙ (B) and CD204 (D) are seen in injured areas. MHC class ΙΙ and CD204-positive cells appear spindle to dendritic-shaped (B, inset) and round to oval (D, inset), respectively. Counterstained with hematoxylin, CV = central vein, bar = 100 μm.

Cells reacting for CD204 were frequently seen in control livers. The number of positive cells in d-GalN-injected rats on PSI days 1 and 2 showed a significant increase in both PV and PP areas, and the number returned to control level by PSI day 3 (Figure 4D; Figure 6D and Online Supplemental Figure 2D). CD204-positive cells showed round to oval shape (Figure 6D, inset).

Double Immunofluorescence Staining for CD68 M1-macrophages or for CD163 M2-macrophages

Double immunofluorescence staining with combination of CD163/CD68 (Figure 7A and B), CD163/MHC class II (Figure 8A and B), CD68/MHC class II (Figure 8C and D), CD163/CD204 (Figure 9A and B), and CD68/CD204 (Figure 9C and D) was conducted using samples obtained in control and PSI day 2. In control livers, 63% of CD163-positive Kupffer cells reacted simultaneously to CD68 (Figure 7A; Online Supplemental Figure 3); the percentage of double-positive cells for CD163/CD68 on PSI day 2 was 82% (Online Supplemental Figure 3). Similarly, a small number of MHC class II–positive cells in control livers co-expressed CD163 (9%) and CD68 (7%; Figure 8A and C; Online Supplemental Figure 4A and B); the percentages of double-positive cells for CD163/MHC class II and CD68/MHC class II on PSI day 2 were 65% and 75%, respectively (Figure 8B and D; Online Supplemental Figure 4A and B). These findings indicated an increase in double-positive cells for CD163/CD68 and CD163/MHC class II or CD68/MHC class II after d-GalN injection. On the other hand, CD163- or CD68-positive cells reacting to CD204 were frequently seen in control livers (84% and 85%, respectively; Figure 9A and C; Online Supplemental Figure 4C and D) and approximately 95% of CD163- and CD68-positive cells reacted to CD204 on PSI day 2 (Figure 9B and D; Online Supplemental Figure 4C and D); there was no clear difference in CD163/CD204 cells or CD68/CD204 cells between controls and d-GalN-treated livers.

Figure 7.

Double immunofluorescence staining for CD163/CD68 in livers of control (A) and d-GalN-induced lesions (B) on postsingle injection day 2. Arrows indicate double-positive cell reaction. 4’, 6-diamidino-2-phenylindole (DAPI) for nuclear stain, bar = 50 μm.

Figure 8.

Double immunofluorescence staining for CD163/MHC class ΙΙ (A and B), CD68/MHC class ΙΙ (C and D) in livers of control, and d-GalN-injected rats on postsingle injection day 2: (A and C) images from control livers; (B and D) images from d-GalN-injected livers. Arrows indicate double-positive cell reaction; 4’, 6-diamidino-2-phenylindole (DAPI) for nuclear stain. Bar = 50 μm.

Figure 9.

Double immunofluorescence staining for CD163/CD204 (A and B) and CD68/CD204 (C and D) in livers of control and d-GalN-injected rats on postsingle injection day 2: (A and C) images from control livers; (B and D) images from d-GalN-injected livers. Arrows indicate double-positive cell reaction; 4’, 6-diamidino-2-phenylindole (DAPI) for nuclear stain. Bar = 50 μm.

mRNA Expressions of Inflammatory (M1-macrophages)/Anti-inflammatory (M2-macrophages) Cytokines

M1-macrophage-related cytokines (tumor necrosis factor [TNF]-α, interleukin [IL]-6, and interferon-γ) were significantly increased on PSI days 1 and 2 (Figure 10A–C) compared to those of controls. However, IL-1β was decreased significantly at PSI 8 hr and on days 3 and 5 (Figure 10D).

Figure 10.

mRNA expression of M1-macrophage-related inflammatory cytokines such as tumor necrosis factor-α (A), interleukin (IL)-6 (B), INF-γ (C), and IL-1β (D) in control and d-GalN-injected rat livers at postsingle injection (PSI) 8 hr and on PSI days 1, 2, 3, and 5. Data were normalized to 18S mRNA level; n = 4; *versus control at p < .05 by Dunett’s test.

M2-macrophage-related cytokines such as transforming growth factor (TGF)-β1, monocyte chemoattractant protein (MCP)-1, and IL-4 were significantly increased on PSI days 1 and 2 (Figure 11A, C, and D), and IL-10 expression showed a peak on PSI day 2 with a significant change (Figure 11B).

Figure 11.

mRNA expression of M2-macrophage-related anti-inflammatory cytokines such as transforming growth factor-β1 (A), interleukin (IL)-10 (B), monocyte chemoattractant protein (MCP)-1 (C), and IL-4 (D) in control and d-GalN-injected rats livers at postsingle injection (PSI) 8 hr and on PSI days 1, 2, 3, and 5. Data were normalized to 18S mRNA level; n = 4; *versus control at p < .05 by Dunett’s test.

Discussion

d-GalN-induced hepatic lesions in rats were characterized by degeneration/multifocal coagulation necrosis and inflammatory cell reactions; they were not restricted to specific areas such as PV or PP areas. Acetaminophen-, TAA-, and CCl₄-induced hepatic lesions were exclusively seen in the PV areas (Ide et al. 2003; Hinson, Roberts, and James 2010; Hung et al. 2006), whereas allyl alcohol developed hepatic lesions mainly in the PP areas (Li, David, and Stewart 1999). Hepatic lesions in human virus hepatitis are seen as multifocal necrosis in the lobules, which may be accompanied by inflammatory cells depending on degrees and stages (Boyer and Klatskin 1970; Popper 1983). Therefore, it has been considered that d-GalN-induced hepatic lesions in laboratory animals resemble those of human virus hepatitis histopathologically.

Interestingly, the dose (800 mg/kg BW) used in this study developed hepatic lesions on PSI days 1 and 2; then, the lesions quickly recovered on PSI day 3 onward, regardless of multifocal necrosis. The lesion development was reflected by significantly increased levels of serum AST and ALT on PSI days 1 and 2. Interestingly, along with multifocal necrosis and inflammatory cell reactions, reparative fibrosis (characterized by the appearance of mesenchymal fibroblasts and collagen deposition) occurred simultaneously on PSI days 1 and 2. In addition, TUNEL-positive cells (indicative of single-cell death) on PSI day 1 and regenerating hepatocytes (demonstrable with BrdU staining) were seen on PSI day 2, with a statistically significant change. In TAA-induced hepatic lesions seen in the PV areas that were reported previously in our laboratory, the reparative fibrosis was followed by coagulation necrosis on subsequent days with still increased values of AST and ALT (Ide et al. 2003).

Generally, M1-macrophages are regarded as promoter of injury by producing cytotoxic/pro-inflammatory factors such as INF-γ, TNF-α, and IL-1β (Martinez 2011; Pello et al. 2011), whereas M2-macrophages may be related to tissue repair/reparative fibrosis by producing IL-4, IL-10, MCP-1, and TGF-β1 (Sica and Mantovani 2012; Njoku et al. 2009). In response to cytokines, macrophages can express specialized and polarized functional properties (Gordon and Taylor 2005). These M1- and M2-macrophages could modify hepatic lesions induced by hepatotoxicans (Wijesundera et al. 2014). It is interesting to evaluate the hepatotoxicity based on M1- and M2-macrophage functions, along with direct influences on hepatocytes by hepatotoxicans.

Usually, in immunohistochemical analyses, M1-macrophages are identified by the expression of CD68, whereas M2-macrophages by CD163. In the present d-GalN-induced rat hepatic lesions consisting of degeneration/multifocal necrosis of hepatocytes and inflammation as well as reparative fibrosis, there was no marked difference in the numbers of CD68 M1-macrophages and CD163 M2-macrophages between the PV and PP areas. CD68 M1-macrophages and CD163 M2-macrophages showed a simultaneous significant increase on PSI day 2; more interestingly, CD163 M2-macrophages appeared on PSI day 1 with a significantly increased number. In normal rat liver tissues, it is reported that Kupffer cells express CD163 as M2-macrophages (Golbar et al. 2012), although the cells are regarded as resident macrophages. Because d-GalN-induced rat hepatic lesions were characterized histopathologically by degeneration/multifocal necrosis in hepatic lobules, the preexisting Kupffer cells might react quickly to lesion development. It has been generally considered in TAA-induced centrilobular lesions that CD68 M1-macrophages take part in lesion development before the appearance of M2-macrophages in pathological lesions (Wijesundera et al. 2013, 2014); these findings are different from the kinetics of hepatic macrophages in d-GalN-induced rat hepatic lesions. Therefore, the reverse appearance of CD68 M1-macrophages and CD163 M2-marophages might be considered as a characteristic of d-GalN-induced rat hepatic lesions.

Additionally, in double immunofluorescence on PSI day 2, approximately 82% of hepatic macrophages expressed simultaneously with CD163/CD68, in contrast to controls (63%); the increased percentage of CD163/CD68-double-positive macrophages might indicate that CD163-expressing Kupffer cells became CD68-expressing M1-macrophages after d-GalN injection. Generally, it has been considered that CD68 M1-macrophages seen in PV areas injured by TAA may be derived mainly from blood monocytes. Preexisting Kupffer cells might be responsible for the appearance of hepatic macrophages, which quickly responded to d-GalN injection.

As mentioned above, CD163 M2-macrophages appeared mainly on days 1 and 2, and CD163-expressing Kupffer cells might become CD68 M1-macrophages on PSI day 2. Therefore, M1-macrophage-related factors (TNF-α, IL-6, and INF-γ) and M2-macrophage-related factors such as TGF-β1, IL-10, MCP-1, and IL-4 were increased significantly on PSI days 1 and 2; the simultaneous co-expressions of these factors of M1- and M2-macrophages might be reflected by histopathological findings such as degeneration/multifocal necrosis accompanied with reparative fibrosis on PSI days 1 and 2.

An interesting finding of the present study was a significant reduction of IL-1β mRNA level at PSI 8 hr and on PSI days 3 and 5 after d-GalN-injection. IL-1β is the IL-1 family members and secreted by Kupffer cells as a pro-inflammatory cytokine that often alter the hepatic microenvironment (Zheng et al. 2012). A significant reduction of IL-1β in serum level was reported in d-GalN-induced acute rat liver failure at late stages, suggesting an important role of IL-1β in liver regeneration (Wang et al. 2013). The reduction of IL-1β level in the present study might be related to the quick recovery of hepatocellular architecture, particularly on PSI days 3 and 5.

In double immunofluorescence analyses, it was reported in TAA-induced hepatic lesions of the PV areas that the majority of MHC class II–expressing macrophages became CD68-expressing M1 macrophages, whereas CD204-expressing macrophages are regarded as CD163-expressing M2-macrophages; MHC class II- and CD204-expressing macrophages are considered to be polarized toward M1- and M2-macrophages, respectively (Wijesundera et al. 2014). In the present double immunofluorescence staining in d-GalN-induced rat hepatic lesions on PSI day 2, many MHC class II macrophages showed a simultaneous co-expression with CD163 (65%) or CD68 (75%), whereas a small number MHC class II–reacting macrophages in control livers co-expressed with CD163 (9%) or CD68 (7%). MHC class II macrophages might have functions of both CD68 M1-macrophages and CD163 M2–macrophages after d-GalN-injection. In addition, macrophages expressing MHC class II are related to the induction of complicated immune response inducing Th1 lymphocytes (Perrigoue et al. 2009). It is interesting to investigate hepatic lesions induced by d-GalN, on the basis of immunoreaction in relation to M1- and M2-macrophage polarization.

CD204 is a scavenger receptor for lipid metabolism, and its expression may also be related to phagocytosis (Tomokiyo et al. 2002). Approximately 95% of CD204-expressing macrophages co-expressed with CD163 or CD68 on PSI day 2, in contrast to controls (85% for CD68 and 84% for CD163); these findings indicated that there was no clear differentiation in CD163/CD204 and CD68/CD204-double-positive cells between control livers and d-GalN-treated livers; no clear polarization in CD204-expressing macrophages toward M1- or M2-macrophages was seen.

In conclusion, the present study showed that the d-GalN-induced rat hepatic lesions were characterized by degeneration/multifocal necrosis with an inflammatory cell reaction (mainly macrophages), as well as concurrent occurrence of reparative fibrosis. Therefore, CD68 M1-macrophages and CD163 M2-macrophages concomitantly appeared in the lesions. Additionally, MHC class II–expressing macrophages and CD204-expressing macrophages appeared with both M1- and M2-macrophage characteristics. These findings indicated that M1- and M2-macrophages contribute to the development of hepatic lesions induced by d-GalN. Under depletion of hepatic macrophages, which was induced with injection of dichloromethylene diphosphonate clodronate, it was found in TAA-induced PV lesions that the hepatic lesions were aggravated because of deficient functions of M1-macrophages (tissue injury and removal of cell debris) and of M2-macropahges (reparative fibrosis; Golbar et al. 2016). Finally, immunophenotypic characteristics of hepatic macrophages, especially on the basis of M1- and M2-macrophage polarization, may be needed to study hepatotoxicity.

Supplemental Material

Supplemental Material, DS1_TPX10.1177_0192623318801574 - M1/M2-macrophage Polarization-based Hepatotoxicity in d-galactosamine-induced Acute Liver Injury in Rats

Supplemental Material, DS1_TPX10.1177_0192623318801574 for M1/M2-macrophage Polarization-based Hepatotoxicity in d-galactosamine-induced Acute Liver Injury in Rats by Nahid Rahman, Munmun Pervin, Mizuki Kuramochi, Mohammad R. Karim, Takeshi Izawa, Mitsuru Kuwamura, and Jyoji Yamate in Toxicologic Pathology

Supplemental Material

Supplemental Material, DS2_TPX10.1177_0192623318801574 - M1/M2-macrophage Polarization-based Hepatotoxicity in d-galactosamine-induced Acute Liver Injury in Rats

Supplemental Material, DS2_TPX10.1177_0192623318801574 for M1/M2-macrophage Polarization-based Hepatotoxicity in d-galactosamine-induced Acute Liver Injury in Rats by Nahid Rahman, Munmun Pervin, Mizuki Kuramochi, Mohammad R. Karim, Takeshi Izawa, Mitsuru Kuwamura, and Jyoji Yamate in Toxicologic Pathology

Supplemental Material

Supplemental Material, DS3_TPX10.1177_0192623318801574 - M1/M2-macrophage Polarization-based Hepatotoxicity in d-galactosamine-induced Acute Liver Injury in Rats

Supplemental Material, DS3_TPX10.1177_0192623318801574 for M1/M2-macrophage Polarization-based Hepatotoxicity in d-galactosamine-induced Acute Liver Injury in Rats by Nahid Rahman, Munmun Pervin, Mizuki Kuramochi, Mohammad R. Karim, Takeshi Izawa, Mitsuru Kuwamura, and Jyoji Yamate in Toxicologic Pathology

Supplemental Material

Supplemental Material, DS4_TPX10.1177_0192623318801574 - M1/M2-macrophage Polarization-based Hepatotoxicity in d-galactosamine-induced Acute Liver Injury in Rats

Supplemental Material, DS4_TPX10.1177_0192623318801574 for M1/M2-macrophage Polarization-based Hepatotoxicity in d-galactosamine-induced Acute Liver Injury in Rats by Nahid Rahman, Munmun Pervin, Mizuki Kuramochi, Mohammad R. Karim, Takeshi Izawa, Mitsuru Kuwamura, and Jyoji Yamate in Toxicologic Pathology