Mem Inst Oswaldo Cruz, Rio de Janeiro, VOLUME 114 | JUNE 2019
Cannabinoid receptor-1 antagonism: a new perspective on treating a murine schistosomal liver fibrosis model
1University of Alexandria, Alexandria Faculty of Medicine, Medical Biochemistry Department, Alexandria, Egypt
2University of Alexandria, Alexandria Faculty of Medicine, Pathology Department, Alexandria, Egypt
3University of Alexandria, Alexandria Faculty of Medicine, Medical Parasitology Department, Alexandria, Egypt
BACKGROUND Formation of schistosomal granulomata surrounding the ova can result in schistosomiasis-associated liver fibrosis (SSLF). The current standard of treatment is praziquantel (PZQ), which cannot effectively reverse SSLF. The role of the cannabinoid (CB) receptor family in liver fibrosis has recently been highlighted.
OBJECTIVES This study aimed to assess the therapeutic effect of CB1 receptor antagonism in reversing SSLF in a murine model of Schistosoma mansoni infection.
METHODS One hundred male Swiss albino mice were divided equally into five groups: healthy uninfected control (group I), infected control (group II), PZQ treated (group III), rimonabant (RIM) (SR141716, a CB1 receptor antagonist)-treated (group IV) and group V was treated with combined PZQ and RIM. Liver sections were obtained for histopathological examination, alpha-1 smooth muscle actin (α-SMA) immunostaining and assessment of CB1 receptor expression using real-time polymerase chain reaction (RT-PCR).
FINDINGS The most effective reduction in fibrotic marker levels and granuloma load was achieved by combined treatment with PZQ+RIM (group V): CB1 receptor expression (H = 26.612, p < 0.001), number of α-SMA-positive cells (F = 57.086, p < 0.001), % hepatic portal fibrosis (F = 42.849, p < 0.001) and number of granulomata (F = 69.088, p < 0.001).
MAIN CONCLUSIONS Combining PZQ with CB1 receptor antagonists yielded the best results in reversing SSLF. To our knowledge, this is the first study to test this regimen in S. mansoni infection.
Schistosomiasis is a common parasitic digenetic trematode infection. It is a very prevalent tropical disease in several developing countries like Egypt. More than 240 million people are infected worldwide and nearly triple that number (700 million) are at risk of acquiring the infection. Schistosomiasis is responsible for around 300,000 deaths annually in Africa alone.(1)
The histopathological hallmark of schistosomiasis is the development of epithelioid granulomata surrounding the ova with an eosinophilic-rich inflammatory infiltrate in the portal tracts of the liver. Fibrous scar tissue forms around the granulomata and eventually replaces them over time, resulting in fibrous expansion of the portal tracts and inducing hepatic fibrosis known as schistosomiasis-associated liver fibrosis (SSLF). SSLF is a consequence of massive deposition of extracellular matrix (ECM) in the periportal spaces, which blocks portal veins and leads to portal hypertension, portocaval shunting, cirrhosis, splenomegaly and gastrointestinal varices.(2)
Two major events that generally characterise liver fibrosis are activation and proliferation of hepatic stellate cells (HSCs) and an increase in ECM deposition, especially collagen type I.(3) Increasing evidence has highlighted the role of activated myofibroblasts residing in the portal tracts encircling bile ducts and ductules and portal vein and hepatic artery radicals in the pathogenesis of portal fibrosis.(4) HSCs and myofibroblasts derived from different cell populations express alpha-1 smooth muscle actin (α-SMA) and synthesise fibrogenic cytokines [transforming growth factor (TGF-β1)], growth factors, chemokines, fibrosis components and inhibitors of matrix degradation.(5,6)
Despite the presence of efficacious schistosomicides like praziquantel (PZQ), it has been demonstrated that the granulomatous inflammatory reactions and hepatic fibrosis continue aggressively even with efficient treatment.(7) To date, several attempts have been made with other treatment modalities either alone or in combination with PZQ. However, there has been scant treatment for direct effect on fibrosis-forming cells in the case of schistosomiasis.(8)
It was recently shown that neuro-humoral signalling plays a role in HSC responses.Particularly, the endogenous lipidic cannabinoid (CB) ligands and their receptors, CB1 and CB2, have emerged as potent mediators of hepatic steatosis, stellate cell activation and hepatic fibrosis.(9) In addition, CBs provoke the hemodynamic alterations associated with advanced liver disease.(3) The two receptors, CB1 and CB2, exert opposing effects: CB1 activates the fibrogenic pathway and CB2 is antifibrotic.(10) AlthoughCB1 receptors are the most abundant receptors in the mammalian brain, they are also expressed at lower levels in a large number of peripheral tissues like various liver cell types. Based on these findings, antagonism of CB1 signalling in HSCs has emerged as a promising antifibrotic strategy.(6)
Here, we assessed the therapeutic effect of antagonising CB1 receptors (using CB1 receptor antagonist, rimonabant (RIM) (SR141716) in reversing SSLF for the first time in a murine model of Schistosoma mansoni infection.
MATERIALS AND METHODS
Animals, parasite and drugs –This study was carried out in the Parasitology, Medical Biochemistry and Pathology departments of Alexandria Faculty of Medicine, Alexandria University, Egypt.
The study used 100 male Swiss albino mice (four to six weeks old and about 20-25 grams weight) purchased from the animal house, Medical Parasitology Department, Faculty of Medicine, Alexandria University, Alexandria, Egypt. The mice were given tap water and a balanced ad libitum diet.
Ethical committee rules regarding animal housing and sacrifice were followed. All animal studies were approved by the local government based on national regulations for animal experimentation.
S. mansoni cercariae were shed from infected Biomphalaria alexandrina snails purchased from the Schistosoma Biological Supply Program (SBSP) in Theodor Bilharz Research Institute (TBRI), Imbaba, Giza, Egypt. They were used for infection of 80 mice with a dose of 100 freshly shed cercariae/mouse using the tail immersion technique.(11)
Two drugs were used: PZQ (Distocide) EIPICO, Cairo, Egypt, was purchased from a local pharmacy and RIM hydrochloride (SR141716, RIM), a CB-1 receptor antagonist, was purchased from Sigma Aldrich (catalogue number SML0800).
Experimental design –Mice were divided into five groups of 20 mice each; normal uninfected control (group I), infected control (group II), PZQ-treated (group III), RIM-treated (group IV) and group V treated with combined PZQ and RIM. All treatment regimens commenced at week 8 post-infection (pi). The timing was chosen because liver fibrosis is expected to be successfully established eight weeks after infection.(12) Drugs were administered once daily for two weeks.
PZQ was given to groups III and V at a dose of 300 mg/kg and RIM was given to groups IV and V at a dose of 10 mg/kg dissolved in 1 mL/kg of saline solution with a drop of Tween 80.(13)
Mice were sacrificed at week 10 pi. Liver specimens from sacrificed mice were divided. Part of each specimen was conserved in RNA later stabilisation reagent (Qiagen, USA, catalogue no. 76104) and stored at -80º C for later CB1 receptor expression assessment. The other part was placed in 10% buffered formalin for histopathological assessment.
Assessment of CB1 receptor expression using real-time polymerase chain reaction (RT-PCR) –On the day of the assay, 30 mg of liver tissue sections were transferred to RNase-free round-bottomed tubes on ice to be homogenised (0.6 mL of freshly prepared Qiazol lysis reagent) (Qiagen, Germantown, MD, USA; cat. no. 79306) containing 1% of 2-mercaptoethanol was added to each tissue sample. Homogenisation was performed using a rotor-stator according to the manufacturer’s protocol. Total RNA was extracted and purified using PureLink ® RNA Mini Kit (Invitrogen, Waltham, MA, USA; cat. no 12183018A). Concentration of total RNA was estimated using Nanodrop.
Purified RNA was reverse transcribed using Applied Biosystems (Waltham, MA, USA) High-Capacity cDNA Reverse Transcription Kits (cat. no. 4374966). Briefly, 2 μg of total RNA was used per 20 μL reaction. The thermal cycler was programmed as follows: 25ºC for 10 min, 37ºC for 120 min, 85ºC for 5 min and 4ºC until the removal of samples. A minus reverse transcription control was added in all experiments to rule out DNA contamination. Complementary DNA (cDNA) was stored at -20ºC until CB1 receptor expression assessment. RT-qPCR was performed using an Applied Biosystems Step-one Real-time system. For each sample, 1 µL of Taqman CB1 gene expression assay reagent [Thermo Fischer (Waltham, MA, USA) scientific assay no. Hs01038522_s1, cat. no. 4331182] was added to 10 µL of Taqman master mix and 5 µL of RNAse free H2O. Four microliters of the cDNA sample was added to complete the total volume to 20 µL. GAPDH (Taqman GAPDH control reagent, Thermo Fischer, cat. no. 402869) was used as the endogenous reference gene for data normalisation. RT-PCR settings were as follows: an initial two-minute hold cycle at 50oC, 10 min hold 95oC and 40 cycles of 15 sec at 95oC and 1 min at 60oC.
Histopathological studies –Liver biopsy specimens were fixed in 10% buffered formaldehyde for 24 h and processed routinely. The tissue samples were embedded into paraffin blocks and sectioned into 4 µm thick sections for staining with haematoxylin and eosin (H&E) and Masson’s Trichrome stain at three non-continuous levels.
Each sample was analysed histologically for the number of schistosomal granulomata, the constituent cells of the granulomata and presence of epithelioid and inflammatory cell infiltrates or myofibroblasts and fibrosis. Lobular necro-inflammatory activity was also assessed.
Immunohistochemical staining –Two consecutive four-micron-thick tissue sections of representative paraffin blocks were cut and mounted onto positively charged slides. Heat-induced antigen retrieval using citrate buffer at pH 6 was performed. Immunostaining using polyclonal α-SMA (Invitrogen #PA1-37024) primary antibody at a dilution of (1:100) was performed, followed by Streptavidin-HRP Conjugate (Thermo Scientific # D22187) and developed using DAB chromogen with Meyer’s haematoxylin as a counterstain to assess myofibroblastic cells.
Quantitative assessment of the α-SMA-positive portal myofibroblastic cells showing cytoplasmic positivity was performed using a mean value for the number of positive cells in ten high power fields (HPF).
Semi-quantifying the fibrosis extent –Quantitative analysis of the fibrosis area within the portal tracts of the liver was performed on the Masson’s Trichrome stained sections at three non-continuous sections using a morphometric analysis software (Olympus BX41, Tokyo, Japan).
Ethics –Ethical committee rules were followed regarding animal housing and sacrificing. All animal studies were approved by the local government based on national regulations for animal experimentation.
Statistical analysis of the data –Data were analysed using IBM SPSS software package version 20.0 (IBM Corp, Armonk, NY, USA). Qualitative data were described by number and percent. The Kolmogorov-Smirnov test was used to verify the distribution normality. Quantitative data were described using mean, standard deviation and median. Significance of the obtained results was judged at the 5% confidence level. F-test [analysis of variance (ANOVA)] was used for normally distributed quantitative variables to compare between more than two groups and post-hoc test (Tukey) was used for pairwise comparisons. Kruskal Wallis test was used for abnormally distributed quantitative variables to compare between more than two studied groups and post-hoc (Dunn’s multiple comparisons test) was used for pairwise comparisons. The Pearson coefficient was used to correlate between two normally distributed quantitative variables.
CB1 expression, fibrotic area extent, average number of α-SMA positive cells and average number of granulomata (Table) –We compared the uninfected control group (I) to the infected control group (II) to verify the increased CB1 expression, increased SMA-expressing cells, fibrosis extent and granulomata formation. All parameters were significantly increased in infected mice (p < 0.001 in all parameters).
To show the efficacy of different treatment modalities in reducing fibrotic markers, all treated groups were compared to each other and the control group II. Results showed that CB1 expression was mostly reduced in the combined treatment RIM+PZQ group [group V, median relative quantification (RQ) = 0.02], followed by group IV treated by RIM (median RQ = 0 .07). Group III treated with PZQ showed the smallest reduction in CB1 expression (median RQ = 0.31). Comparing all groups showed statistically significant differences in CB1 expression (H = 30.966, p < 0.0001).
The same pattern was followed regarding the extent of fibrotic areas. The mean fibrotic areas were measured in three non-consecutive sections of liver specimens and expressed as a percentage of the whole area of liver tissue. The mean fibrosis percentage was 21.8% in group V, 35% in group IV and 57.3% in group III compared to 86.7% in control group II. Comparing all the groups showed statistically significant differences in fibrosis extent (F = 42.849, p < 0.001).
Logically, the same applied to the mean number of α-SMA expressing cells, which was 14.9 in group V, 29.2 in group IV and 47.1 in group III compared to 74.7 in control group II. Comparing all the groups showed statistically significant differences in the average number of cells expressing SMA (F = 57.086, p < 0.001).
The pattern was slightly different regarding the average number of granulomata, where group V showed the greatest reduction in granulomata number (mean = 8.3), followed by group III (mean = 12.4) and group IV (mean = 14.9) compared to the control group II (mean = 17.5). Comparing all the groups showed statistically significant differences in the average number of granulomata (F = 69.088, p < 0.001)
CB1 expression was found to significantly correlate to each of extent of fibrosis measurement (r = 0.701, p < 0.001) (Fig. 1A) and the number of SMA-expressing cells (r = 0.686, p )t; 0.001) (Fig. 1B), but not significantly to the number of granulomata (r = 0.318, p = 0.076). The average number of SMA-express)g cells was significantly correlated to fibrosis extent (r = 0.903, p < 0.001) (Fig. 1C) and average number of granulomata (r = 0.710, p < 0.001) (Fig. 1D). Naturally, the average number of granulomata was significantly correlated to the fibrosis extent (r = 0.742, p < 0.001) (Fig. 1E)
Histopathologic examination – Control infected mice (group II)–Microscopic examination of the control mouse livers revealed marked expansion of the portal tracts by schistosomal granulomata centred around the schistosomal ova. The granulomata were composed of epithelioid cells, lymphocytes, plasma cells and eosinophils and surrounded by an outer zone of fibroblastic cells (Fig. 2A, B). The portal tracts were markedly expanded by heavy inflammatory infiltrates with spill-over into the hepatic lobules predominated by eosinophils. Hepatocytes revealed evidence of feathery degeneration and foci of hepatocyte necrosis with intralobular aggregates of lymphocytes and eosinophils (Fig. 2C). Sinusoidal dilation and intrasinusoidal lymphocytes were also evident. Kupffer cells were not evidently hyperplastic. Trichrome-stained sections revealed the notable expansion of portal tracts by fibrous tissue (Fig. 3A).
PZQ-treated mice livers (group III) –Examination revealed moderate expansion of the portal tracts by schistosomal granulomata with reduced amounts of inflammatory cells surrounding the granulomata, but portal fibroblastic cell proliferation was evident (Fig. 2D). Hepatocytes did not reveal morphologic signs of injury and only a moderate increase in the number of Kupffer cells was noted. Trichrome-stained sections revealed moderate expansion of the portal tracts by fibrous tissue (Fig. 3B).
RIM-treated mice livers (group IV) –Examination revealed a mild to moderate expansion of portal tracts by schistosomal granulomata comprising mainly inflammatory cellular infiltrate and minimal portal myofibroblastic proliferation (Fig. 2E). Trichrome-stained slides showed minimal fibroblastic expansion of the portal tracts in the study group (Fig. 3C). The hepatocyte lobules revealed unremarkable morphological changes in the hepatocytes, but marked Kupffer cell hyperplasia was noted (Fig. 2F).
PZQ+RIM treated mice livers (group V) –Examination revealed marked diminution of portal tract expansion by the schistosomal granulomata (Fig. 3D), a reduced number of inflammatory cells and reduced fibrosis area (Fig. 2G).
Alpha-smooth muscle actin immunohistochemistry staining –Examination of the study groups revealed prominent cytoplasmic staining of the portal myofibroblastic cells surrounding the schistosomal granulomata and within the portal tracts in the infected control group II (Fig. 4A-C). The number of α-SMA positive myofibroblastic cells was significantly reduced in the combined PZQ+RIM-treated group (Fig. 4G), followed by the RIM treated group (Fig. 4F) compared to controls. The group treated with only PZQ did not show a significant reduction in the number of α-SMA positive cells (Fig. 4D, E).