Participation of Functionally Different Macrophage Populations and Monocyte Chemoattractant Protein-1 in Early Stages of Thioacetamide-induced Rat Hepatic Injury

Abstract

Macrophages are crucial in hepatic fibrogenesis. In acute hepatic necrosis induced in rats by a single injection of 300 mg/kg body weight (BW) of thioacetamide (TAA), macrophage properties were investigated using single or double immunohistochemistry. Macrophages reacting with anti-CD68, anti-CD163, or major histocompatibility complex (anti-MHC) class II antibody appeared in injured centrilobular areas on days 1-5 after injection. Increased expression of CD68 and CD163 reflect phagocytosis and production of pro-inflammatory factors, respectively. There were also macrophages double-positive to CD68/CD163, CD68/MHC class II, or CD163/MHC class II; of these, macrophages double-positive to CD68/MHC class II were most frequent, indicating that macrophages with enhanced phagocytic activity came to express MHC class II. The appearance of these macrophages corresponded to increased expression of mRNAs of monocyte chemoattractant protein-1 (MCP-1), a chemokine, on day 1, and TGF-β1, a fibrogenic factor, on day 3. Some hepatic stellate cells (HSCs) in injured areas reacted with anti-MCP-1 antibody. To investigate the effects of MCP-1, we added MCP-1 to HS-P, a rat macrophage line. Addition of MCP-1 increased immunoexpression for CD68 and CD163 and up-regulated TGF-β1 mRNA expression. Collectively, macrophages in acute hepatic necrosis may express different properties such as phagocytosis, MHC class II expression, and TGF-β1 production; such expression may be influenced by MCP-1 produced by HSCs.

Introduction

Hepatic fibrosis is provoked following hepatocyte necrosis, which may be induced by causes such as hepatitis virus infection, alcohol abuse, excess intake of hepatotoxicants, or metabolic diseases (obesity or diabetes) (Bataller and Brenner 2005; Grossi et al. 2008; Svirtlih et al. 2007; Yeh and Brunt 2007). Fibrosis is a chronic event that may progress to cirrhosis, which may result in hepatic failure (Pinzani et al. 2008). A variety of inflammatory cells and growth factors/chemokines have been reported to participate in the development of hepatic fibrosis (Bataller and Brenner 2005; Pinzani et al. 2008; Wynn 2008). In particular, transforming growth factor-β1 (TGF-β1) plays an important role in the induction of myofibroblastic cells responsible for the production of extracellular matrices such as collagens and fibronectin (Bataller and Brenner 2005; Pinzani et al. 2008). Transforming growth factor-β1 is produced mainly by macrophages, suggesting that macrophages are a crucial cell contributing to the development of hepatic fibrosis (Guo et al. 2006; Hines et al. 1993; Ide et al. 2003; Wang 1999; Zhang et al. 2004). Besides cytokine production, macrophages have heterogeneous functions, such as phagocytosis and antigen presentation, depending on microenvironmental conditions (Takahashi et al. 1996; Yamate et al. 2000); however, detailed macrophage functions and factors influencing the functions have not yet been completely clarified.

Thioacetamide (TAA) is a hepatotoxicant relating to elevation of oxidative stress (Bruck et al. 2004; Lu et al. 1999). Thioacetamide-induced rat acute hepatic lesions are useful for studies on the pathogenesis of early events in hepatic fibrosis (Ide et al. 2003). In the present study, using this TAA acute rat model, the participation of macrophages that may exhibit functionally different properties was investigated by means of single or double immunohistochemical stainings using three different antibodies (ED1, ED2, and OX6). The antibody ED1 recognizes CD68 located on lysosomal membranes, particularly phagosomes, and thus, the expression level implies enhanced phagocytosis (Damoiseaux et al. 1994; Dijkstra et al. 1985; Suda et al. 1998). CD/63, recognized by ED2, is a member of the macrophage scavenger receptor family, and increased expression may be related to production of pro-inflammatory factors (Polfliet et al. 2006). Major histocompatibility complex class II (MHC class II) molecules (rat Ia), labeled by OX6, play important roles in immunological reactions after tissue injury (Ide et al. 2005; Yamashiro et al. 1994). Monocyte chemoattractant protein-1 (MCP-1) is a well-known chemokine that has a significant influence on recruitment of monocytes and macrophages into injured tissues (Schecter et al. 1997; Yamashiro et al. 1994). Additionally, it was reported that MCP-1 stimulated collagen production via up-regulation of TGF-β1 expression in cultured fibroblasts (Gharaee-Kermani et al. 1996). Along with expression of MCP-1 in hepatic lesions, influence of MCP-1 on macrophage functions was investigated by using a rat macrophage line (HS-P) (Yamate, Maeta, Benn et al. 2001; Yamate, Maeta, Tsukamoto et al. 2001).

Materials and Methods

In Vivo Study

Animals and TAA Injection:

Male F344 rats at the age of five weeks were obtained from Charles River Japan (CRJ, Hino, Shiga, Japan). These animals were housed in an animal room at a controlled temperature of 21°C ± 3°C and with a twelve-hour light-dark cycle; they were provided a standard diet for rats (MF; Oriental Yeast Co, Ltd., Tokyo, Japan) and tap water ad libitum. After a one-week acclimatization period, they were divided into controls (six rats) and TAA injection groups (thirty-six rats). The rats in the TAA group were given a single intraperitoneal injection of TAA (Wako Pure Chemical Industries, Ltd., Osaka, Japan) dissolved in physiological saline at a dose of 300 mg/kg body weight (BW). Six rats were euthanized by exsanguination under ether anesthesia on each of post-injection (PI) days 1, 3, 5, 7, 10, and 15. Control rats received an equivalent volume of physiological saline and were euthanized immediately after the injection (PI day 0). After euthanasia, the livers were removed and used for the following examinations. These experiments were in compliance with our institutional guidelines for animal care.

Histopathology and Immunohistochemistry:

Several strips, 4 mm in thickness and cut from the left external lobe of the liver of all animals, were fixed in 10% neutral buffered formalin or Zamboni’s fixative (0.21% picric acid and 2% paraformaldehyde in 130 mM phosphate buffer, pH 7.4) (Ide et al. 2003). These samples were dehydrated and embedded in paraffin. Tissue sections cut at 3 μm in thickness were stained with hematoxylin and eosin (HE) for histopathology or by the Azan-Mallory method for collagen (Ide et al. 2003). Zamboni’s solution-fixed, deparaffinized liver tissue sections, which were prepared from three rats at each examination point, were evaluated for immunohistochemistry with monoclonal antibodies of ED1 (×500; Chemicon International Inc., Temecula, CA, USA), ED2 (×50; Serotec, Oxford, UK) and OX6 (×100; Serotec) (Ide et al. 2003; Kawashima et al. 2003). The ED1, ED2, and OX6 antigens are CD68, CD163, and MHC class II molecules, respectively (Polfliet et al. 2006; Suda et al. 1998; Yamashiro et al. 1994). After pretreatment with proteinase K (Dako, Corp., Carpinteria, CA, USA) for ten minutes, tissue sections used for each antibody were treated with 3% H₂O₂ in phosphate buffered saline (PBS) to quench endogenous peroxidase, and then with 5% skimmed milk in PBS to inhibit nonspecific reactions. The sections were incubated with each primary antibody for fourteen to twenty hours at 4°C, followed by reaction with the second antibody using the avidin-biotin complex (ABC) method (Vectastain ABC Kit; Vector Laboratories, Inc., Burlingame, CA, USA). Positive reactions were visualized with 3,3′ -diaminobenzidine (DAB). Non-immunized mouse IgG served as negative controls. Sections were counterstained lightly with hematoxylin.

The double-immunohistochemical staining was carried out according to methods described previously (Kawashima et al. 2003). Tissue sections of livers were prepared from three rats at each examination point. The sections, which had been reacted with a primary antibody and visualized with DAB as described above (the first stain), were washed in running water for ten minutes, and then incubated with 0.1 M glycine-hydrochloric acid buffer for one hour to completely remove the excess primary antibody at the first stain. After being washed in PBS and incubated with skimmed milk, the sections were reacted with a primary antibody for the second stain for fourteen to twenty hours at 4°C. At the second stain, the Envision System-AP (Dako) was applied to the sections for one hour at room temperature, and positive reactions were visualized (red in color) by the Fuchsin Substrate-Chromogen System (Dako). Finally, sections were counterstained lightly with hematoxylin. Combinations in the dual immunohistochemistry were anti-CD68/anti-CD163, anti-CD68/anti-MHC class II, and anti-CD163/anti-MHC class II; the primary antibody used for the first stain in each combination was anti-CD163, anti-MHC class II, and anti-CD163 antibodies, respectively.

In the single immunohistochemistry, cells showing a distinct immunopositive reaction to CD68, CD163, or MHC class II were counted in five randomly selected areas (0.13 mm²/area) in the centrilobular area of the hepatic lobule at a magnification of ×200. In the dual immunohistochemistry, double- or single-positive cells were counted using the same method described above.

For detection of cells reacting to MCP-1 or desmin, fresh-frozen tissue sections (10 μm in thickness) were prepared from one or two rats at each examination point. Desmin expression is a marker of hepatic stellate cells (HSCs) in rat liver (Makino et al. 2006). After being dried, sections were fixed with 4% par-aformaldehyde (PFA) for twenty minutes at 4°C, and then incubated with 20% normal goat serum (Sigma Aldrich Inc., Japan) to suppress nonspecific reaction. Sections were incubated for one hour at room temperature with anti-rat polyclonal MCP-1 (2 μg/mL; PeproTech Inc., Rocky Hill, NJ, USA) or anti-desmin monoclonal antibody (×200; Dako). The secondary antibody in the Histofine simple stain kit (Histofine Multi Kit for rats: Nichirei Corp., Tokyo, Japan) was applied to the sections, and positive reactions were visualized with DAB. Nonimmunized rabbit IgG for MCP-1 stain or mouse IgG for desmin stain served as negative controls. The sections were counterstained lightly with hematoxylin. To identify cells reacting to both MCP-1 and desmin, we used mirror-image sections.

Immunofluorescent Staining:

Fresh-frozen liver tissue sections, prepared from one or two rats on PI days 0 and 3, were fixed for ten minutes in cold acetone and used for indirect immunofluorescent staining with OX62 (×100; Serotec); OX62 monoclonal antibody labels alpha E2 integrin and is used as a marker of dendritic cells in rats (Brenan and Puklavec 1992). After incubation with 20% normal donkey serum (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) for thirty minutes, the sections were reacted with OX62 for fourteen to twenty hours at 4°C, and then with FITC-conjugated affinipure F(ab′)₂ fragment donkey anti-mouse IgG (H+L) (Jackson Immunoresearch Laboratories, Inc., West Grove, PA, USA) for one hour at room temperature. Nonimmunized mouse IgG was used as a negative control. After being washed in PBS, the sections were covered with mounting medium containing 4′6 diamidino-2-phenylindole (DAPI) for nuclear staining (Vector laboratories Inc., CA, USA) and observed under a fluorescence microscope.

Reverse Transcriptase Polymerase Chain Reaction:

Three rats at each examination point were used. Total RNA was isolated from liver samples by Trizol reagent (Invitrogen Corp., Carlsbad, CA, USA), and then reverse-transcribed to cDNA with the Super Script Preamplification System (Invitrogen) (Ide et al. 2003). The cDNA was amplified by the PCR with Taq DNA polymerase and each of the specific primers for rat MCP-1, TGF-β1, and β-actin (control). The following conditions were used for the amplification: for MCP-1 (Kawashima et al. 2003), after five minutes of denaturation at 94°C, twenty-seven cycles of fifteen seconds of denaturation at 94°C, thirty seconds of annealing at 53°C, and thirty seconds of extension at 72°C; for TGF-β1 (Ide et al. 2005), after five minutes of denaturation at 94°C, 25 cycles of thirty seconds of denaturation at 94°C, one minute of annealing at 60°C, and thirty seconds of extension at 72°C; and for β-actin (Ide et al. 2005), after five minutes of denaturation at 94°C, twenty cycles of thirty seconds of denaturation at 94°C, thirty seconds of annealing at 59°C, and thirty seconds of extension at 72°C. The oligonucleotides used for PCR were as follows: MCP-1 sense primer 5′-ATG-CAGGTCTCTGTCACG-3′ and antisense primer 5′-CTAGTTCTCTGTCATACT-3′ (Kawashima et al. 2003); TGF-β1 sense primer 5′-CACCATCCATGACATGAACC-3′ and antisense primer 5′-GTTGGACAACTGCTCCACCT-3′ (Ide et al. 2005); and β-actin sense primer 5′-TAAAGACCTC-TATGCCAACAC-3′ and antisense primer 5′-CTCCTGC TTGCTGATCCACAT-3′ (Ide et al. 2005). The polymerase chain reaction (PCR) products were electrophoresed in 1% agarose gel, and DNA was stained with ethidium bromide on the gel. Using image analysis software (NIH Image 1.61, Bethesda, MD 2002), the band of MCP-1 or TGF-β1 was analyzed semiquantitatively in contrast to that of β-actin (Ide et al. 2005; Kawashima et al. 2003).

In Vitro Study

Macrophage Line (HS-P):

A macrophage line established from a rat histiocytic sarcoma, HS-P, has been useful for studies on macrophage function (Yamate, Maeda, Benn et al. 2001; Yamate, Maeda, Tsukamoto et al. 2001). Cells from the HS-P line were cultured in RPMI1640 medium (Nissui, Tokyo, Japan) supplemented with 10% inactivated fetal bovine serum (FBS; Bioserum, UBL, Japan), 0.03% L-glutamin (Nissui), penicillin (100 U/mL) and streptomycin (100 μg/mL), as well as 2% NaHCO₃ in a humidified 5% CO₂ atmosphere at 37°C. The cells were subcultured by treating with a mixture of 0.1% trypsin and 0.02% ethylenediaminetetraacetic acid in PBS.

Immunofluorescent Staining:

Cells of HS-P (1× 10⁵/mL) were seeded in slide chambers (1 mL), and grown in 10% FBS-containing RPMI1640 medium for twenty-four hours. After being washed in PBS, cells were incubated in 1% FBS-containing RPMI1640 medium for twenty-four hours, and then MCP-1 (MCAF, recombinant rat MCP-1; US Biological, Edouard-Michelin, Canada) diluted in 1% FBS-containing RPMI1640 medium was added at a concentration of 0, 10, or 100 ng/mL. After twenty-four or forty-eight hours of incubation, the cells were washed in PBS and fixed for ten minutes in acetone at 4°C. The cells were then treated for forty-five minutes in 0.5% skimmed milk added to 0.1% NaN₃ and incubated with anti-CD68, anti-CD163, or anti-MHC class II antibody at 4°C for fourteen to twenty hours. For the indirect immunofluorescent staining, the cells were reacted with FITC-conjugated affinipure F(ab′)₂ fragment donkey anti-mouse IgG (H+L) (Jackson) for one hour at room temperature. Nonimmunized mouse IgG was used as a negative control. The chamber slides were covered with mounting medium containing DAPI, and observed by a fluorescent microscope. The reactivity was evaluated as follows: −, negative; +, slight; 2+, moderate.

Reverse Transcriptase Polymerase Chain Reaction:

The HS-P cells grown in 10% FBS-containing RPMI1640 medium were incubated for twenty-four hours in 1% FBS-containing RPMI1640 medium. Then, recombinant rat MCP-1 (US Biological) was added to the culture at a concentration of 0, 10, or 100 ng/mL. After twenty-four hours of incubation, the cells were harvested and total RNA was extracted using Trizol reagent (Invitrogen), and the cells were then reverse-transcribed to cDNA with the Super Script Preamplification System (Invitrogen). cDNA was amplified by the PCR with Taq DNA polymerase and the specific primers for rat TGF-β1. Primers and conditions used for the amplification of TGF-β1 and β-actin were the same as described in the in vivo study. The band of TGF-β1 was analyzed semiquantitatively in contrast to that of β-actin. The measurements were conducted using three different samples.

Statistical Evaluation:

Obtained data represented mean ± standard deviation (SD), and statistical analysis was performed using analysis of variance (ANOVA). Significance was accepted at p < .05.

Results

In Vivo Study

TAA-induced Hepatic Lesions:

No significant changes were seen in the liver of control rats (Figure 1a). Thioacetamide-induced hepatic lesions were generally similar to those reported previously (Ide et al. 2003). In TAA-injected rats, on PI day 1, coagulation necrosis of hepatocytes was seen mainly around centrilobular vein (centrilobular area) of the hepatic lobules, accompanied by a few neutrophils and hemorrhage (Figure 1b); neither fatty change nor granular degeneration was seen in hepatocytes. On PI day 3, infiltration of mononuclear cells was observed in the affected centrilobular area (Figure 1c). Infiltrating cells were regarded as macrophages, because the majority reacted to CD68, CD163, and MHC class II, as described below; lymphocytes were rarely seen in the injured areas. On PI day 5 (Figure 1d) and subsequent days, histological changes of hepatocyte necrosis and mononuclear cell infiltrate were gradually decreased; on PI day 15, the affected areas were recovered by replacement with regenerated hepatocytes. The TAA-induced hepatic lesions were characterized by hepatocyte necrosis and cell infiltration, which were most prominent on PI days 1 and 3, respectively.

Collagens, demonstrated by the Azan-Mallory stain, were not seen in the centrilobular area in controls (Figure 1e) and among necrotic hepatocytes on PI day 1. Deposition began to be seen among infiltrating cells in the centrilobular area on PI day 3, and the degree peaked on PI days 5 and 7 (Figure 1f). Thereafter, deposition gradually disappeared on PI days 10 and 15. Formation of myofibroblasts in the fibrotic area, presumably derived from HSCs, has been confirmed by the immunohistochemical staining with an anti-α smooth muscle actin antibody (data not shown; Ide et al. 2003). The myofibroblasts have been regarded as cells producing extracellular matrices, leading to hepatic fibrosis (Guo et al. 2006; Ide et al. 2003).

Immunohistochemistry for Macrophages:

Based on histopathological characteristics, it was found that hepatocyte necrosis on day 1 was followed by cell infiltration on day 3 and fibrosis on days 5 and 7. This finding suggests that macrophages have important roles in fibrosis after hepatocyte necrosis. We pursued macrophage properties by performing immunohistochemistry with different antibodies.

In control livers, Kupffer cells located along sinusoids reacted with anti-CD68 and anti-OX163 antibodies; on the other hand, no Kupffer cells reacted to MHC class II. In the affected centrilobular area, the kinetics of cells reacting with anti-CD68, anti-CD163, or anti-MHC class II antibody was evaluated (Figures 2a and 2b). The number of anti-CD68-positive cells began to significantly increase as early as PI day 1 (p < .05); it became greatest on PI day 3, and thereafter, the number was gradually decreased, but remained significantly elevated up to PI day 15 (Figure 2a) (p < .05). Anti-CD68-positive cells were round or spindle shaped (Figure 3a). Anti-CD163- or anti-MHC class II-positive cells showed a significantly increased number only on PI day 3 (p < .05); however, there was no significant change on PI days 1, 5, 7, 10, and 15 (Figure 2a). Anti-CD163-positive cells were mainly round or oval in shape (Figure 3b); anti-MHC class II-positive cells showed large round or stellate configuration and were located mainly in the periphery of the injured centrilobular areas (Figure 3c). As shown in Figure 3 (a, b, and c), there was somewhat of a difference in the distribution between anti-CD68, anti-CD163, and anti-MHC class II-positive cells.

Therefore, we investigated co-immunoexpression between anti-CD68, anti-CD163, and anti-MHC class II-positive cells. The double immunohistochemistry revealed that there were cells double positive to CD68 and CD163, CD68 and MHC class II, or CD163 and MHC class II (Figure 2b). The number of cells reacting with both anti-CD68 and anti-MHC class II (Figure 3d) antibodies was significantly increased on PI days 1, 3, and 5 (p < .05), with a peak on PI day 3 (Figure 2b). Cells double-positive to CD68 and CD163 or CD163 and MHC class II showed the significantly increased number only on PI day 3 (Figure 2b) (p < .05). Cells reacting with both anti-CD68 and anti-MHC class II antibodies were most frequent (Figure 2b). The percentage of anti-CD68/anti-MHC class II double-positive cells to anti-CD68 single-positive cells was approximately 32%, 19%, and 18% on PI days 1, 3, and 5, respectively; the percentage of anti-CD163/anti-MHC class II double-positive cells to anti-CD163 single-positive cells was less than 5% on days 1, 3, and 5. These findings indicated that anti-CD68-positive cells could express MHC class II molecules rather than anti-CD163-positive cells.

Dendritic cells are antigen-presenting cells capable of expressing MHC class II molecules (Gao et al. 2007; Zimmerli and Hauser 2007). In the injured centrilobular areas on PI day 3, dendritic cells, demonstrable by the OX62 immunohistochemistry, were sporadically seen (3–6 cells/area at ×400) (Figure 4a); the distribution did not correspond to that of anti-MHC class II-positive cells (Figure 3c). In the intact Glisson’s sheath, OX62-reacting cells were occasionally seen, indicating the presence of interstitial dendritic cells (Figure 4b).

Immunohistochemistry for MCP-1 and Desmin:

Monocyte chemoattractant protein-1 is a well-known chemokine playing an important role in recruitment of macrophages (Schecter et al. 1997; Yamashiro et al. 1994). We explored cells capable of producing MCP-1 in the injured centrilobular areas. In control liver tissues, HSCs along the hepatic cords showed a positive reaction to desmin. No anti-MCP-1–positive cells were seen in control liver tissues or hepatic lesions on PI days 3 to 15. On the other hand, on PI day 1, anti-MCP-1-positive cells were detected (Figure 5a) in the injured centrilobular areas, and anti-desmin-positive cells were also seen among necrotic hepatocytes (Figure 5b). On mirror image sections, some cells reacting with anti-MCP-1 antibody corresponded to anti-desmin-positive cells (Figures 5a and 5b; arrows indicate representative cells reacting with both anti-MCP-1 and anti-desmin antibodies).

Expression of mRNAs of MCP-1 and TGF-β1:

Monocyte chemoattractant protein-1 and TGF-β1 are well-known factors contributing to macrophage induction and fibrogenesis, respectively (Gharaee-Kermani et al. 1996). We analyzed mRNA expression levels of these factors. In agreement with the appearance of anti-MCP-1-immunopositive cells, the expression of MCP-1 mRNA was seen only on PI day 1; on subsequent observation days (PI days 3 to 15) and in controls, MCP-1 mRNA expression level was under the detection limit (Figure 6). As compared with expression level of TGF-β1 mRNA in controls, the level was significantly increased on day 3 (p < .05); on other examination days (PI day 1 and on days 5 to 15), no significant increase was seen, but there was a tendency toward up-regulation (Figure 7).

In Vitro Study

Influences of MCP-1 on Macrophage Properties Using HS-P Cells:

Because MCP-1 expression on PI day 1 was followed by macrophage infiltration on PI day 3, we investigated the effects of MCP-1 on immunoexpression of macrophages. Cultured in the absence of MCP-1 (0 ng/mL), HS-P cells showed a slightly positive reaction to CD68 (+) (Figure 8a), whereas HS-P cells incubated for twenty-four and forty-eight hours with MCP-1 (10 and 100 ng/mL) reacted moderately to CD68 (2+) (Figure 8b). No cells reacting to CD163 were seen in MCP-1-untreated HS-P cells (0 ng/mL) (-) (Figure 8c), whereas cells reacting with anti-CD163 antibody were developed in HS-P cells by forty-eight-hour incubation with MCP-1 (10 and 100 ng/mL), although the reactivity was slight or moderate (+ or 2+) (Figure 8d). Regardless of the absence or presence of MCP-1, anti-MHC class II-positive cells were not seen in HS-P cells.

Transforming growth factor-β1 mRNA was dose-dependently up-regulated in MCP-1-added HS-P cells, with a significant increase at 100 ng/mL (p < .05) (Figure 9).

Discussion

Macrophages differ in morphology, immunophenotypes, or functions, depending on different milieus, such as inflammation and tumors; they are divided mainly into blood monocyte-derived macrophages, resident macrophages (Kupffer cells), and dendritic cells (Takahashi et al. 1996; Yamate et al. 2000). In this study, we used three different antibodies against CD68, CD163, and MHC class II molecules. Generally, CD68 is expressed in blood monocyte-derived infiltrating macrophages and resident macrophages; the expression implies enhanced phagocytosis (Damoiseaux et al. 1994; Dijkstra et al. 1985; Suda et al. 1998). Anti-CD163 antibody labels with resident macrophages, and enhanced expression relates to production of pro-inflammatory factors (Dijkstra et al. 1985; Polfliet et al. 2006). MHC class II molecules are expressed in mature dendritic cells and activated macrophages (Ide et al. 2005; McMaster and Williams 1979; Yamashiro et al. 1994).

The present study showed that the number of anti-CD68-, anti-CD163- or anti-MHC class II-positive cells increased in the centrilobular areas in very early phases of injured rat livers, with a peak on PI day 3; then, the numbers gradually decreased. Among these macrophages, the number of anti-CD68-positive cells was greatest, followed by anti-CD163- and anti-MHC class II-positive cells. Furthermore, the dual immunohistochemical stainings revealed that there were macrophages double-positive to CD68 and CD163, CD68 and MHC class II, or CD163 and MHC class II. These findings suggest that macrophages appearing in injured livers have functionally different properties demonstrable by these immunophenotypes; additionally, it is likely that these macrophages have some functions in common. Among macrophage populations, the number of cells reacting simultaneously with anti-CD68 and anti-MHC class II antibodies was greatest, showing a significant increase on PI days 1, 3, and 5 (p < .05). Major histocompatibility complex class II molecules are expressed in activated macrophages and dendritic cells (Yamashiro et al. 1994; Zhang et al. 1993; Zhao et al. 2006). Dendritic cells consist of the interstitial dendritic cells, interdigitating follicular cells of the spleen and lymph nodes, and Langerhans cells of the epidermis; the interstitial dendritic cells are widely distributed in the connective tissues, including the Glisson’s sheath (Takahashi et al. 1996).

To determine whether anti-CD68- and anti-MHC class II double-positive cells are derived from interstitial dendritic cells, OX62, a monoclonal antibody highly specific for rat dendritic cells (Brenan and Puklavec 1992), was applied to liver tissues. In the injured centrilobular areas, there was a very small number of OX62-reacting cells, and the distribution did not coincide with that of anti-MHC class II-positive cells. It was considered that anti-CD68 and anti-MHC class II double-positive cells were not generated from interstitial dendritic cells, but from cells such as monocyte-derived macrophages and resident Kupffer cells. Because the amount of CD68 expression depends on phagocytic activity (Damoiseaux et al. 1994), after taking up cell debris, macrophages with enhanced phagocytic activity might come to express MHC class II molecules. MCH class II-expressing cells present antigens to unprimed T cells, and subsequently Th1/Th2 lymphocyte commitment is determined (Banchereau et al. 2003; Gao et al. 2007; Zimmerli and Hauser 2007). The expression of MHC class II molecules is induced through the complicated immune system; it is interesting to pursue the functional roles of macrophages expressing both CD68 and MHC class II to clarify the complicated pathogenesis of progressive hepatic fibrosis.

TGF-β1 is the most important factor in the fibrogenic mechanism and may be produced by macrophages, hepatocytes, lymphocytes, or HSCs (Bataller and Brenner 2005; Guo et al. 2006; Kim et al. 1998; Pinzani et al. 2008; Qi et al. 1999; Zhang et al. 2004). In agreement with the increased number of macrophages on PI day 3, TGF-β1 mRNA expression was significantly up-regulated (p < .05). This finding suggests that macrophages appearing in hepatic lesions are an important cell for TGF-β1 production. TGF-β1 is responsible for induction of myofibroblasts capable of releasing collagens; in fact, collagens were increased in the injured areas on PI days 3–7. Recently, it has been reported that CD163 functions as a scavenger receptor for hemoglobin-haptoglobin complexes, and that the engagement of CD163 induces the production of pro-inflammatory mediators such as IL-6 and TNF-α (Polfliet et al. 2006). Anti-CD163 single-positive macrophages were the second greatest in number on PI day 3. As discussed below, HS-P cells with enhanced expression of CD163 in the presence of MCP-1 came to up-regulate TGF-β1 mRNA. Likely, CD163-reacting macrophages in injured livers might have attributed to the production of fibrogenic factors such as TGF-β1, although further examinations are needed to confirm TGF-β1 production by macrophages.

MCP-1 is a member of the C-C chemokine family and acts as a monocyte/macrophage chemotaxisis (Marra et al. 1998; Sakai et al. 2006; Schecter et al. 1997). In the present study, MCP-1 mRNA was expressed only on PI day 1 in TAA-injected rat livers, suggesting that a transient increase of MCP-1 at very early stages might induce macrophage infiltration into injured areas. Mirror-image sections showed that some cells in the injured centrilobular areas gave positive reactions to both desmin and MCP-1. Because rat HSCs are cells containing desmin (Makino et al. 2006), HSCs may be regarded as a cell population capable of releasing MCP-1 in injured livers (Marra et al. 1998).

Besides chemotaxisis, MCP-1 plays an important role in fibrogenesis (Gharaee-Kermani et al. 1996; Sakai et al. 2006; Tsuruta et al. 2004). This conclusion is based on in vitro findings that MCP-1-stimulated rat lung fibroblasts up-regulated endogenous TGF-β1 expression (Gharaee-Kermani et al. 1996), and that human peripheral CD14-positive monocytes showed enhanced TGF-β1 mRNA expression when MCP-1 was added (Sakai et al. 2006). In the present study, macrophage appearance and increased expression of mRNAs of MCP-1 and TGF-β1 were observed almost simultaneously in early phases (on days 1 and 3) in injured livers. As mentioned above, macrophages appearing in injured hepatic lesions showed heterogeneous functions. We pursued influences of MCP-1 on macrophage properties in vitro. Addition of MCP-1 increased immunoexpression of CD68 and CD163 in HS-P cells. Furthermore, HS-P cells to which MCP-1 was added increased mRNA expression of TGF-β1. Expression of CD68 is related to phagocytic activity (Damoiseaux et al. 1994), and CD163 expression implicates production of proinflammatory factors (Polfliet et al. 2006). Transforming growth factor-β1 is an important mediator for hepatic fibrogenesis (Bataller and Brenner 2005; Guo et al. 2006). Taken together, MCP-1 might be a factor capable of regulating phagocytosis and TGF-β1 production in macrophages through increased CD68 and CD163 molecules, respectively.

In conclusion, the present study has shown that macrophages emerging in early phases of TAA-induced rat liver lesions exhibit heterogeneous properties. It is particularly interesting to note that anti-CD68-positive macrophages with enhanced phagocytic activity had the capacity to express MHC class II molecules. Furthermore, it is likely that MCP-1, which may be produced by HSCs, influences macrophage properties such as phagocytic activity and TGF-β1 up-regulation. The participation of functionally different macrophages would be responsible for hepatic fibrosis after injury.