Expression of M-CSF, TNF-β, IFN-γ, and IL-10 in Rats with Liver Cirrhosis and Hypersplenism and its Significance

Objective: This study aimed to examine the expression of M-CSF, TNF-β, IFN-γ, and IL-10 in rats with liver cirrhosis and hypersplenism, and to investigate its significance. Methods: Seventy-two healthy male SD rats were randomly divided into a model group (n=60) and a control group (n=12). The rats in the model group were first gavaged with a 40% CCL4/peanut oil solution to develop cirrhosis and hypersplenism. The expression rates and intensities of the pro-inflammatory cytokines M-CSF, TNF-β, and IFN-γ, and the anti-inflammatory cytokine IL-10 in the spleen were measured and compared with the control group. The pro-inflammatory/anti-inflammatory cytokine ratio and its role were analyzed. Results: The positive expression rates of M-CSF, TNF-β, IFN-γ, and IL-10 in the model group (cirrhosis and hypersplenism) were 61.76%, 79.41%, 64.70%, and 88.24%, respectively, which were significantly different (P<0.05) from the 25%, 33.33%, 16.67%, and 50% in the control group. The relative protein expression intensities of M-CSF, TNF-β, IFN-γ, and IL-10 in the model group were 0.63 ± 0.58, 1.06 ± 0.49, 0.99 ± 0.38, and 1.12 ± 0.42, respectively, and were significantly different (P<0.05) from the 0.18 ± 0.12, 0.52 ± 0.27, 0.38 ± 0.28, and 0.60 ± 0.32 in the control group. The relative mRNA expression levels of M-CSF, TNF-β, IFN-γ, and IL-10 in the model group were 2.06 ± 0.11, 4.07 ± 0.19, 2.98 ± 0.11, and 7.94 ± 0.27, respectively, and were significantly different (P<0.05) from the 1.01 ± 0.05, 1.06 ± 0.11, 1.00 ± 0.31, and 1.02 ± 0.08 in the control group. Conclusion: The abnormally increased expression of M-CSF, TNF-β, IFN-γ, and IL-10 in rats with liver cirrhosis and hypersplenism, as well as the abnormal pro-inflammatory/anti-inflammatory cytokine ratio and regulation may play an important role in enhancing the phagocytosis of macrophages and causing peripheral cytopenias.


Introduction
Patients with splenomegaly due to cirrhotic portal hypertension are often complicated by mono-lineage cytopenias or multi-lineage peripheral cytopenias, which seriously affects prognosis [1,2]. The cause of this is related to hypersplenism resulting from increased activity of macrophages in the spleen. Studies have shown that in splenomegaly due to cirrhotic portal hypertension, macrophages are highly activated, with significantly enhanced phagocytosis, cytokine secretion, and antigen-presenting capability; a large number of blood cells in the spleen are engulfed and destroyed, which may be the cause of peripheral cytopenias in splenomegaly due to cirrhotic portal hypertension [3][4][5]. Cytokines that promote the activation of macrophages include macrophage colony-stimulating factor (M-CSF), tumor necrosis factor-β (TNF-β), interferon-γ (IFN-γ), and macrophage activating factor (MAF) [6]. Cytokines that inhibit macrophages mainly include interleukin-10 (IL-10), which helps to stabilize immune function [7]. Macrophage migration inhibitory factor (MIF) has both anti-inflammatory and pro-inflammatory functions [8], which are difficult to understand. Currently, M-CSF, TNF-β, IFNγ, and IL-10, have been extensively studied in lymphoma, atherosclerosis, and liver disease. However, only a single cytokine was analyzed in such studies, and multiple cytokines have rarely been investigated in a single disease, especially liver cirrhosis with hypersplenism. Therefore, the present study aimed to investigate the effects of these four cytokines on macrophage activity and peripheral cytopenias in hypersplenism.

Reagents and materials
Experimental animals: Seventy two specific pathogen-free (SPF) inbred SD rats, weighing 240-300 g, were purchased from Hunan SJA Laboratory Animal Co., Ltd. (China) (certificate number: 43004700011005). They were randomly divided into a liver cirrhosis and hypersplenism model group (n=60) and a control group (n=12). After quarantine inspection, the rats were raised in a separate room in the barrier area in Hainan Drug Safety Evaluation Center. For the model group, five rats were housed in one cage, and for the control group, six in one cage.
Reagents: Carbon tetrachloride (analytical reagent) was purchased from Guangdong Guanghua Sci-Tech Co., Ltd. (China) (batch number: 20140410). Commercially available Xiangmanyuan Brand peanut oil and Qianmen Brand Erguotou (alcohol content: 56% vol) were used. Pentobarbital sodium and deionized water used to prepare the anesthetic were provided by Hainan Drug Safety Evaluation Center.
Rabbit polyclonal antibodies for M-CSF, TNF-β, IFN-γ, and IL-10 were purchased from Santa Cruz Biotechnology, Inc. The Universal kit and DAB kit were purchased from Beijing Zhongshan Jinqiao Biological Technology Co., Ltd. (China). The enhanced chemiluminescence kit, developing and fixing kit, RIPA lysis buffer (medium), total RNA extraction kit, universal reverse transcription kit, and ComSYBRqPCR Mix (with ROX) were obtained from Shanghai Novland Co., Ltd. (China).

Experimental methods
Development of cirrhosis and hypersplenism: Rats were raised in the new environment for seven days after purchase for adaption, with an adequate supply of food and water. Rats in the control group were maintained on normal diet, without any treatment. Rats in the model group were gavaged a 40% CCL4/peanut oil solution at a dose of 3.0 ml/kg twice a week in accordance with the methods of Ye [9]. Gavage dose was adjusted according to weight gain in the previous week. The dose was not changed when the body weight increased or decreased by less than 5%, but was increased by 0.5 ml/kg with a weight gain of more than 5%, decreased by 0.5 ml/kg with a weight reduction of more than 5%, and discontinued with a weight reduction of more than 10%. After gavage administration, the water was replaced with a mixture of 10% liquor and deionized water in the first two weeks and with a mixture of 15% liquor and deionized water thereafter. At week 7, one randomly selected rat from the model group was anesthetized and sacrificed. The right liver lobe and spleen tissues were collected, then fixed in methanol, embedded in paraffin, and stained with hematoxylin-eosin (HE) to observe the pathological changes in the liver and spleen tissues under the microscope, in order to confirm the development of cirrhosis and hypersplenism. At week 8, when the development of cirrhosis and hypersplenism had been confirmed, 2 ml of orbital venous blood was collected from the rats in both groups for liver function and routine blood tests. After blood collection, all rats were sacrificed by cervical dislocation, and appropriate amounts of the right liver lobe and spleen tissues were collected, then fixed and preserved. The expression of M-CSF, TNF-β, IFN-γ, and IL-10 in the enlarged spleen of rats with cirrhosis was measured by immunohistochemistry. Another appropriate amount of spleen tissue was preserved at -80°C for later Western blotting and RT-PCR.
Criteria for development of cirrhosis and hypersplenism: a. monolineage or multilineage cytopenias (WBC, PLT, and RBC) as revealed by routine blood tests; b. abnormal liver function indexes; c. liver: visual observation of cirrhotic nodules and microscopic observation of pseudolobules; and d. spleen: splenomegaly compared with the control group, a significantly increased spleen index, and microscopic observation of red pulp expansion.
Immunohistochemistry: The fixed liver and spleen tissues were embedded in paraffin, and cut into 4 μm serial sections. One liver tissue section was stained with HE. One spleen tissue section was stained with HE, one was incubated with PBS in replacement of the primary antibodies and used as a negative control, and four were stained by immunohistochemical staining for M-CSF, TNF-β, IFN-γ, and IL-10, respectively. HE staining results were reviewed by two physicians from Hainan Provincial People's Hospital to confirm the diagnosis of cirrhosis and splenomegaly.
The same immunohistochemical procedure was performed for the four cytokines: Paraffin sections were dewaxed, dehydrated in gradient alcohol, placed in distilled water, transferred into pH-6.0 sodium citrate solution, and boiled at high pressure until 2 minutes and 30 seconds after valve rotation. The autoclave was then placed into a pot of cold water, with running water slowly rushing the surface. After cooling, the sections were placed into 3% H 2 O 2 for 5 minutes, then into distilled water for 1 minute, PBS I for 1 minute, and then PBS II, III, and IV, respectively, for 3 minutes each. After the diluted primary antibodies were added dropwise, the sections were incubated at 37°C for 1 hour, and then rinsed with PBS II, III, and IV, respectively, for 3 minutes each. After the immunohistochemistry solution was added dropwise, the sections were incubated at 37°Cfor 1 hour, and then rinsed with PBS V, VI and VII, respectively, for 3 minutes each. Afterwards, the sections were developed with freshly prepared diaminobenzidine (DAB) solution for 3-5 minutes under the microscope, and then rinsed with running water. Finally, the sections were counterstained, dehydrated in gradient alcohol, cleared with xylene, and mounted with gum.
Result judgment: M-CSF, TNF-β, IFN-γ, and IL-10-positive cells were stained mainly in the cytoplasm and membrane, which turned brownish yellow or brown. Five high-power (×400) fields were selected to calculate the percentage of positive cells and expression intensity. The positive expression intensity was scored as follows: no staining was scored as 0; faint yellow staining as 1; yellow staining as 2; and brown staining as 3. The percentage of positive cells rate was scored as follows: 5% was scored as 0; 5-25% as 1; 26-50% as 2; 51 -75% as 3; and >75% as 4. The final score was obtained by summing the above two scores: A score of 0 indicated negative expression (-); a score of 1 and 2 indicated weak positive expression (+);a score of 3-5 indicated medium positive expression (+); and a score of 6 and 7 indicated strong positive expression (+++).
Western blotting: Each tissue sample was treated with 300 μl of cell lysis buffer, and mixed with a pipette until complete lysis was observed. The lysate was transferred to a new centrifuge tube. Next, 10 μl of sample was mixed with 10 μl of 2×SDS-PAGE loading buffer, heated at 100°C for 5 minutes, cooled on ice, and centrifuged at 12000 g for 5 minutes. Insoluble precipitates were removed. The samples were separated by 10% SDS-PAGE at a volume of 20 μl per well. After electrophoresis, the PVDF membranes were immersed in methanol for 1 minute. Next, the gel, filter papers, and the PVDF membranes that had been immersed in methanol were immersed in transfer buffer at 4°C for 10 minutes. Transfer sandwiches were then prepared (Note: Semi-dry electrophoretic transfer was performed using a semi-dry cell at 30 mA for 60 min). The PVDF membranes were blocked with blocking buffer at 4°C overnight, washed for 15 minutes three times with 1 × TBST the next day, and incubated with diluted primary antibodies at 37°C for 2 hours. Then, the membranes were washed with 1 × TBST for 10 minutes four times, incubated with diluted secondary antibodies at 37°C for 2 hours and finally washed with 1 × TBST for 10 minutes four times. After chemiluminescent detection using the hypersensitive luminescence solution, the membranes were exposed to x-ray films. The films were developed, fixed, dried, and photographed by a gel imaging system. The images were analyzed by Gel-Pro Analyzer software. Reverse transcription was performed for 30 min at 42°C, and 10 min at 85°C. Cycling parameters used for quantitative PCR were: denaturation at 95°C for 3 min, 95°C for 12 seconds, and 62°C for 40 seconds for 40 cycles. Fluorescent signal was collected at the extension phase of each cycle. After amplification was completed, fluorescent signal in cycles 3-15 was set as baseline, a threshold fluorescence value was established where all PCR amplifications were within the exponential phase, and GAPDH was set as an internal reference gene to calculate Ct values for each sample and each gene. Relative quantitation (RQ) of gene expression was determined by the 2-Ct method.

RT-PCR
Statistical analysis: Test data were analyzed using the SPSS 19.0 software package. Normally distributed measurement data were expressed as x±S. The results of routine blood and liver function tests, Western blotting, and RT-PCR were compared using the independent samples t-test. Immunohistochemical positive expression was compared using the chi-square test. P<0.05 was considered statistically significant.

General conditions
Twenty six (43.3%) rats in the model group died from week 5 onward, and nine of them died instantly due to poor gavage technique. The 34 surviving rats had matted fur, severe fur loss, poor appetite, weight loss, and low activity. These rats were in poor spirits, relatively weak, insensitive to external stimuli, and had a black tongue, which was an obvious sign of intoxication. In contrast, no deaths occurred in the control group, and the rats had glossy fur and good appetite, and were in good spirits, sensitive to external stimuli, and lively.

Changes in the spleen index
Spleen index=spleen weight (mg)/body weight (g). The spleen index was 3.14 ± 0.98 for the model group, which was significantly different from the 2.33 ± 1.03 for the control group (t=-2.351, P<0.05).

Immunohistochemistry
In the model group, 61.76% (21/34, including 10 rats with a medium or higher level of positive expression) of the rats were M-CSF positive (a score of ≥ 2), which was significantly higher than the 25.00% (3/12) in the control group (X2=4.29, P<0.05). Furthermore, 79.41% (27/34, including 12 rats with a medium or higher level of positive expression) of the rats were TNF-β positive, which was significantly higher than the 33.33% (4/12) in the control group (X2=5.70, P<0.01).

Pathological changes
Liver pathological changes: The liver showed marked atrophy, a nodular surface, hyperplasia, and obvious periportal fibrosis, with pseudolobules of different sizes enveloping liver cells ( Figure 5).

Discussion
As a chemical hepatotoxicant, CCL4 has been widely used for its reliability to induce liver fibrosis and cirrhosis in experiments [10,11], despite its high mortality rate (43.3%). This study demonstrates that CCL4 can also be used to induce hypersplenism. In the model group, ALT, AST, and TBIL levels were significantly increased, while TP and ALB levels were significantly decreased, which indicated impaired liver function and was one of the criteria for diagnosis of liver fibrosis and cirrhosis [12,13]. For the model group, a nodular liver surface, an increased spleen index, and a significantly enlarged spleen were observed; pseudolobules in the liver tissue, and expanded red pulp, reduced white pulp, and increased fibrous tissue in the spleen were observed under the microscope, indicating the development of liver cirrhosis and hypersplenia in rats. Cirrhosis and hypersplenism often lead to peripheral cytopenias [14]. Compared with the control group, PLT and RBC counts were decreased significantly in the model group, which is similar to conditions in human cirrhosis and hypersplenism with peripheral cytopenias [15]. The significantly increased WBC count in the model group may be related to uncontrolled inflammation after modeling.
M-CSF is produced by activated macrophages, B cells, and T cells, and can promote transformation of monocytes to macrophages, increase the number and survival of macrophages [16], and stimulate macrophages to enhance phagocytosis, digestion, and cytotoxicity. It is also an important inflammatory cytokine that can induce monocytes to release tumor killers, such as IFN and TNF. TNF-β and IFN-γ are important proinflammatory cytokines generated by T cells after antigen or mitogen stimulation. They can promote macrophage activation through immunomodulation, and play important roles in strengthening and maintaining the activated state of macrophages [17]. IL-10, which is mainly secreted by Th2 cells and macrophages, is an anti-inflammatory cytokine secreted by a subset of regulatory T cells, and has an anti-hepatic fibrosis effect [18]. In addition to mediating immune response, it can also cause feedback inhibition of macrophages, and inhibit the expression of inflammatory cytokines, such as M-CSF, TNF-β, and IFN-γ [19].
As revealed by immunohistochemistry, the positive expression rates of M-CSF, TNF-β, IFN-γ, and IL-10 were significantly higher in the spleen of rats with liver cirrhosis and hypersplenism compared with normal spleens (P<0.05). Moreover, ≥ 43% of rats in the model group had medium or strong positive expression, while only partial medium expression was detected in the control group (no strong positive expression was detected). As determined by highly sensitive and specific Western blotting, the relative protein expression intensities of M-CSF, TNF-β, IFN-γ, and IL-10 were significantly higher in the model group than those in the control group (P<0.05). As demonstrated by RT-PCR, the relative mRNA expression levels of M-CSF, TNF-β, IFN-γ, and IL-10 were also significantly higher in the model group than those in the control group (P<0.05). The high expression of M-CSF, TNF-β, and IFN-γ in liver cirrhosis and hypersplenism can promote the transformation of blood monocytes to macrophages, increase and enhance macrophage activation [20], and enhance the phagocytosis and destruction of blood cells Eubank et al. [7], which may be an important cause of peripheral cytopenias. The number and activity of macrophages are known to be enhanced by the pro-inflammatory cytokines, M-CSF, TNF-β, and IFN-γ, and inhibited by the anti-inflammatory cytokine IL-10. These cytokines constitute a regulatory system to regulate the number and phagocytosis of macrophages. Although the level of the anti-inflammatory cytokine IL-10 was significantly increased in hypersplenism, it cannot antagonize the effects of pro-inflammatory cytokines and is not sufficient to control the progression of inflammation. The increased pro-inflammatory/anti-inflammatory cytokine ratio eventually leads to the occurrence and development of peripheral cytopenias in cirrhosis and hypersplenism. It is possible to recover the normal number and activity of macrophages and thereby treat peripheral cytopenias by treatment with drugs that inhibit M-CSF, TNF-β, and IFN-γ or promote the synthesis of IL-10.