Differential Expression of Hepatic Fibrosis Mediators in Sick and Spontaneously Recovered Mice with Experimental Biliary Atresia1

BackgroundHepatic fibrosis leading to cirrhosis is the major morbidity in patients with biliary atresia (BA). This fibrosis is due to an imbalance in extracellular matrix (ECM) breakdown and deposition. We have previously demonstrated increased mRNA expression for inhibitors of ECM breakdown without increased expression for mediators of ECM deposition in our animal model of BA by d 14. However, only a mild degree of hepatic fibrosis was seen at this time. We hypothesized that expression patterns for these proteins may change once more significant fibrosis had been established, and added resuscitation to the model to improve survival. Interestingly, we found that some mice spontaneously recovered at later time points with resuscitation, and thus compared expression for inhibitors of ECM breakdown and deposition in sick and recovered mice to determine the differences.MethodsNewborn Balb/c mice received an intraperitoneal injection 1.0 × 106 fluorescence forming units of rhesus rotavirus 24h after birth. Mice were monitored daily for weight gain, development of jaundice, acholic stools, and bilirubinuria. Fifty μL/g of 5% dextrose in normal saline were subcutaneously injected daily to each mouse starting on d 7 until sacrifice. Mice that survived past d 14 were sacrificed at d 21 after saline or RRV infection. Livers were then harvested post-injection d 21 for histologic and immunohistochemical analysis. RNA expression of known mediators of fibrosis was evaluated using quantitative real-time PCR. Protein expression was assessed using ELISA. Weights and normally distributed data were compared using Student's t test. Histologic findings were compared using Fisher's exact test. Comparisons of gene expression and skewed data were performed by the Mann-Whitney U test. Statistical significance was assigned to any P value less than 0.05.ResultsDaily resuscitation resulted in a 35% (24/68) survival rate to d 21 in our model. Mice that recovered were significantly heavier than those that remained ill on d 14 (6.15 ± 1.16 versus 4.94 ± 0.82, P = 0.02) and 21 (7.31 ± 1.41 versus 4.14 ± 0.53, P < 0.001) despite the fact that there was no difference between the groups with respect to weight on d 7 (4.29 ± 0.90 versus 3.89 ± 0.81, P = 0.32). We found that all (10/10) animals that displayed clinical signs of biliary atresia on d 21 had moderate or severe histologic findings, while only one (1/9) of the recovered animals had liver abnormalities at sacrifice (P < 0.001 versus sick group). We also found that the sick mice had statistically significant median fold-increases of mRNA expression for TIMP-1 (31.9 versus 9.1, P = 0.041), TIMP-4 (88.1 versus 1.8, P = 0.022), and MMP-7 (51.8 versus 11.9, P = 0.006) compared with those that recovered. There was a trend toward decreased mRNA expression for PAI-1, which did not reach statistical significance (median 27.7 versus 2.19, P = 0.066). Increased protein expression for TIMP-1 and PAI-1 were also found in the sick group. The mRNA expression for the fibrillar collagens, fibronectin-1, connective tissue growth factor, snail-1, TIMP-2 and -3, and MMP-2 and MMP-9 was not different in the sick and recovered groups 21 d after RRV infection, and was not elevated from baseline gene expression.ConclusionsWith resuscitation added to the animal model of BA, some mice spontaneously recover while others progress to more significant hepatic fibrosis. Mice with hepatic fibrosis have a continued increase in mRNA expression of TIMP-1, TIMP-4, and MMP-7, with a trend toward increased mRNA expression of PAI-1 on d 21. Protein levels for TIMP-1 and PAI-1 were also increased in the sick mice. Recovered mice display mild to no hepatic parenchymal disease and a normal pattern of mRNA expression for the mediators of fibrosis tested. No increase in mRNA expression for the mediators of ECM deposition was found in either group. These data further support the notion that inhibition of ECM breakdown alone is sufficient to induce hepatic fibrosis. Modulation of this process may be a putative target for preventing liver injury in patients with BA.