Date of Award

January 2020

Document Type


Degree Name

Medical Doctor (MD)



First Advisor

Guadalupe Garcia-Tsao


BACKGROUND: The progression of cirrhosis is comprised of two distinct clinical stages, an asymptomatic compensated stage followed by a decompensated stage defined by the presence of complications of cirrhosis, specifically ascites, variceal hemorrhage and hepatic encephalopathy. The main driver of decompensation is the severity of portal hypertension which results from increased intrahepatic resistance and a hyperdynamic circulatory state. A stage of further decompensation, characterized by additional complications of decompensation events, is driven by systemic inflammation and further vasodilatation, which are most probably the result of translocation of gut bacteria and microbial products to extra-intestinal sites. In fact, it has been proposed that even in the compensated stage, the hyperdynamic circulation is driven by bacterial translocation (BT). BT is a result of impaired immunity, increased permeability of the intestinal wall, and intestinal bacterial overgrowth and dysbiosis, or the imbalance of pathogenic and beneficial bacteria within the gut. We report a longitudinal study of the changes in the microbiome during the initiation and progression of liver fibrosis in a rat model of cirrhosis with ascites induced by inhalation of carbon tetrachloride (CCl4) gas. Through this study, we seek to characterize the timing of these microbiome changes during the development of cirrhosis as compared to age-related changes in healthy rats to further elucidate the relationship between intestinal dysbiosis and stage of liver fibrosis. We also correlate alterations in the microbiome to specific complications of cirrhosis, such as bacterial translocation and ascites formation.

METHODS: 90 male Sprague Dawley rats were induced to develop cirrhosis via inhalation of CCl4 gas three time per week until ascites became clinically apparent. Phenobarbital 0.35g/L was added to drinking water to accelerate cirrhosis development. Rats were individually housed. Stool was collected at weeks 1, 2, 4, and 7 during induction of cirrhosis as well as during a 3-week period following the formation of ascites. Three weeks after ascites development, animals underwent terminal laparotomy to collect stool from the ileum and descending colon and to sample mesenteric lymph nodes, ascites, and blood to assess the presence of bacterial translocation. Additionally, 12 healthy rats underwent stool collection at comparable time points as a control for changes observed in the intestinal microbiome of cirrhotic rats. Bacterial DNA was extracted and microbiome composition was determined using 16S rRNA PCR amplification and QIIME analysis.

RESULTS: Changes over time were observed in rats induced to develop cirrhosis (PERMANOVA r-squared 0.23, p-value<0.001) and in healthy rats (PERMANOVA r- squared 0.19, p-value<0.001). In cirrhotic rats, there was a decrease in several operational taxonomic units (OTUs) representing S24-7, an unspecified Bacteroidetes common in rodents, as well as an increase in Allobaculum and decrease in Clostridiales. These changes largely occurred between weeks 4 and 7, prior to the development of ascites (median 9 weeks). In healthy controls, we observed a decrease in OTUs representing Bacteroides, Bacteroidiales, Bacteroidetes, and Allobaculum as well as an increase in Lachnospiraceae. In rats with cirrhosis, there was a non-significant trend towards a decrease in alpha diversity over time. Comparison of cirrhotic rats with BT (n=13) and without BT (n=11) over time, BT status explained a significant amount of variation in microbiome (PERMANOVA r-squared 0.28, p-value<0.001). Compared to rats without BT, Rats with BT had an increase in multiple OTUs representing Bacteroides, particularly at time points following the appearance of ascites. Rats with BT also had a lower albumin (0.6 vs 0.8, p-value=0.025) and higher rate of colonic inflammation (61% vs 18%, p-value=0.031). We also compared cirrhotic rats with ascites to a subset that did not develop ascites (n=7). However, this analysis was limited by small sample size in the group without ascites.

CONCLUSIONS: While both cirrhotic and healthy rats had changes in intestinal microbiome over time, the specific bacteria driving these changes were distinct. This suggests that microbiome shifts occurring with the development/progression of cirrhosis are distinct from the age-related changes seen in healthy rats. Interestingly, the changes observed in cirrhotic rats appeared prior to the formation of clinically-detectable ascites, demonstrating that intestinal dysbiosis is present in the compensated stage of cirrhosis. Thus, the described dysbiosis could potentially contribute to bacterial translocation and hemodynamic abnormalities even before decompensation occurs. We also describe a difference in microbiome in cirrhotic rats with BT compared to those that do not have BT. Perhaps, these changes are associated with increased intestinal inflammation and worsened liver function, explaining their connect to BT. In the future, this distinction may allow us to predict which patients are at risk for BT and associated clinical complications.


This thesis is restricted to Yale network users only. This thesis is permanently embargoed from public release.