Figure 1.
Survival, body weight, and liver damage markers.
A Kaplan-Meier survival plot shows that no deaths occurred in Groups Control or MVC. In Groups CDE and CDE+MVC deaths were registered along the time of the experiment (A). Mean survival was 92 days for Group CDE whereas all other Groups did not reach that parameter. There were no statistically significant differences between Group CDE+MVC and those not receiving the CDE diet (Control and MVC). There was a very significant difference in survival between Group CDE and any of the other Groups. When body weight was measured (B), Groups which received the CDE diet (CDE and CDE+MVC) displayed a serious weight loss in the first week. This parameter recovered slowly in the following weeks. Nevertheless, the animals that were treated with the CCR5 inhibitor (CDE+MVC) recovered weight at a significantly higher rate than non treated animals (CDE). The markers of liver damage studied included transaminases (C), alkaline phosphatase (D), and bilirubin (E). Transaminase blood levels were measured at 4 time points during the experiment, whereas levels of AP and bilirubin were only measured at the end of the procedure (week 16). There was an abrupt increase of transaminases in animals that received the CDE diet during the first week, which diminished later but never reached the basal levels observed in the control diet Groups. Mice that received the CCR5 inhibitor (CDE+MVC) had significantly lower levels of transaminases than those who did not (CDE). The same pattern was observed for AP and bilirubin. Each bar represents the mean ± SEM of at least 8 animals. **p<0.01; ***p<0.001 with respect to control; &p<0.05; &&p<0.01; &&&p<0.001 with respect to CDE.
Figure 2.
Relative weight of the liver and the spleen.
A significant increase in the relative weight (weight of the organ divided by the body weight) of both organs was recorded in the animals receiving the CDE diet, compared with the control diet. Among the mice that received the CDE diet, those treated with MVC had a significantly smaller liver (A) and spleen (B) than those who were not treated. ***p<0.001 with respect to control; &p<0.05 with respect to CDE.
Figure 3.
Representative photographs and microphotographs of the liver.
A clear change in color and general texture was easily appreciate when comparing the liver of animals treated with control diet (A,B) or with the CDE one (C,D). The liver of animals in the CDE Group presents a large number of tumors (C). Tumors in Group CDE+MVC were less numerous and much smaller than those in the previous Group (D). Scale bar for A–D = 1 cm. Histological images were stained with hematoxylin-eosin (E–H), with the fluorescent TUNEL technique (I–L), anti Ki67 (M–P), or with anti-CCR5 antibody (Q–T). The first 2 Groups; Control and MVC (E,F) displayed a normal liver morphology. The liver of the CDE Group had numerous atypic cells and frank tumors (G). Animals treated with MVC had intermediate characteristics (H). Scale bar for E–H = 100 µm. The TUNEL technique detected few apoptotic cells in the liver of animals belonging to control Groups (I,J) but the number increased in animals treated with CDE (K) and was reduced by treatment with MVC (L). Scale bar for I–L = 200 µm. The proliferation marker Ki67 detected few cells in control animals (M,N) and great numbers of positive cells in the CDE Group (O). The number of proliferating cells was intermediate in the CDE+MVC Group (P). CCR5 expression was not detected in control Groups (Q,R) but was found in macrophages (arrowheads), HSC (arrows), and other cell types in the CDE (S) and CDE+MVC (T) Groups. Scale bar for M–T = 100 µm.
Figure 4.
Quantification of the number of tumors, apoptotic and proliferating cells.
The degree of tumor affectation was measured by counting the number of macroscopic tumors (A), the number of tumors seen under the microscope (B), and the diameter of the largest tumor on each animal (C). The number of apoptotic cells as determined by TUNEL (D), and the proliferation index, defined as the number of Ki67 positive cells divided by the total number of nuclei per field (E), were also quantified. The last bar, labelled “NT”, represents the non-tumoral fraction of the liver in the CDE+MVC Group. Bars represent the mean ± SEM of at least 7 animals and 5 photographs per animal (when appropriate). *p<0.05; **p<0.01; ***p<0.001 with respect to control; &p<0.05; &&p<0.01; &&&p<0.001 with respect to CDE.
Figure 5.
Determination of fibrosis in liver samples.
Picro Sirius red staining as viewed under bright light (A–D) and under polarized light (E–H) in animals of Groups Control (A,E), MVC (B,F), CDE (C,G), and CDE+MVC (D,H). Fibrosis was widespread in animals of the CDE Group (C,G) and was reduced after MVC treatment (D,H). Scale bar for A–H = 350 µm. The fibrotic index (I) was calculated from the polarized light images. Bars represent the mean ± SEM of at least 7 animals and 5 photographs per animal. ***p<0.001 with respect to control; &&p<0.01; &&&p<0.001 with respect to CDE. A representative Western blot for α-SMA was performed in liver extracts from animals of the 4 experimental Groups (J). An antibody against AKT was used as a loading control.
Figure 6.
Quantification of the expression of several chemokines both at the mRNA and protein level.
mRNA levels were quantified by qRT-PCR and corrected by the level of GAPDH on each sample as a house keeping gene (A). Proteins were analyzed by multiplex ELISA and are expressed as pg/ml (B). Protein assays for CCR5 were not available. Bars represent the mean ± SEM of at least 7 animals. *p<0.05; **p<0.01; ***p<0.001 with respect to control; &p<0.05; &&p<0.01 with respect to CDE.
Figure 7.
Quantification of the expression of several cytokines both at the mRNA and protein level.
mRNA levels were quantified by qRT-PCR and corrected by the level of GAPDH on each sample as a house keeping gene (A). Proteins were analyzed by multiplex ELISA and are expressed as pg/ml (B). Bars represent the mean ± SEM of at least 7 animals. *p<0.05; **p<0.01; ***p<0.001 with respect to control; &p<0.05; &&p<0.01 with respect to CDE.
Figure 8.
Signal transduction (p38 and ERK) in HSC.
Primary human HSC were incubated with 50 ng/ml human recombinant CCL5 in the presence and absence of 1.0 µM MVC. Western blot analysis showed that CCL5 induces phosphorylation of p38 and ERK whereas preincubation of these cells with MVC prevented this activation event.
Figure 9.
Human hepatoma cell line Hep3B was exposed to H2O2 in the absence and presence of MVC for different periods of time. Hydrogen peroxide induced apoptotic cell death in this cell line, which was prevented by MVC. Bars represent the mean ± SEM of 8 independent wells. *p<0.05; **p<0.01; ***p<0.001 with respect to cells treated with H2O2 but not with MVC.
Figure 10.
Schematic cartoon of the postulated mechanism by which MVC interferes with HCC progression.
The CDE diet damages resident cells of the liver parenchyma, mainly hepatocytes, inducing oval cell proliferation. As a response, these cells and some collaborating neighbors (such as Kupffer cells) secrete a number of cytokines and chemokines including TGF-β1, CCL3, CCL4, and CCL5. The chemokine cocktail promotes the activation of HSC from a quiescent state into a more aggressive phenotype, whereupon a number of chemokine receptors (CCR1, CCR3, CCR5) are expressed at the HSC membrane. Concomitantly, activated HSC secrete a large number of chemokines and cytokines, some of which perpetuate a positive feedback loop that maintain the activated phenotype of the HSC, whereas other molecules induce fibrosis and tumor progression. Maraviroc blocks the autocrine loop by interfering with CCR5, thus stopping HSC activation, fibrosis, and tumor progression.
Table 1.
Primers used for qRT-PCR.