Dysregulated Gut Microbial Fermentation of Dietary Soluble Fiber Induces Cholestatic Hepatocellular Carcinoma in Mice
Open Access
- Author:
- Yeoh, Beng San
- Graduate Program:
- Immunology and Infectious Diseases
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- December 09, 2019
- Committee Members:
- Andrew David Patterson, Dissertation Advisor/Co-Advisor
Andrew David Patterson, Committee Chair/Co-Chair
Connie Jo Rogers, Committee Member
Margherita Teresa-Anna Cantorna, Committee Member
Kevin John Harvatine, Outside Member
Matam Vijay Kumar, Special Member
Margherita Teresa-Anna Cantorna, Program Head/Chair - Keywords:
- Gut microbiota
Innate immunity
Dietary fiber
Short chain fatty acids
Cholemia
Bile acids
Hyperbilirubinemia
Liver inflammation
Hepatocellular carcinoma - Abstract:
- Societal changes in dietary habits, which favor consumption of highly refined herbal and dietary supplements (HDS) in place of their natural counterparts, have been attributed to the rising incidences of HDS-induced jaundice and liver injury in both the United States and worldwide. Yet, the pathophysiology of HDS remains poorly understood due to the difficulty in identifying the causative agent and to the lack of a suitable animal model. The work done in this dissertation has led us to redefine inulin (a fermentable, dietary fiber) as a HDS capable of inducing hyperbilirubinemia, transaminitis and cholestatic liver injury in ~40% of Toll-like receptor 5 deficient (T5KO) mice within 2 weeks. When inulin feeding was prolonged to 24 weeks, the liver pathology evolved into icteric hepatocellular carcinoma (HCC) in a microbiota-dependent manner. Of note, the other ~60% of T5KO mice developed neither cholestasis nor HCC. Herein, we undertook this study with the overarching goal to characterize the atypical occurrence of cholestasis and HCC in mice fed dietary inulin and to demonstrate the utility of using this animal model to study HDS-induced liver diseases in humans. Chapter 1 provides a literature review examining several key concepts and potential mechanisms that accentuate the link between gut microbiota and metabolic diseases. Accumulated data from a variety of animal and human studies indicates that a dysbiotic microbiota plays a key role in instigating metabolic diseases via the following potential mechanisms: (i) increasing calorie extraction, (ii) producing obesogenic metabolites, (iii) causing metabolic endotoxemia-induced low-grade chronic inflammation, and (iv) reprogramming the host inflammatory/metabolic responses to favor the development of metabolic syndrome. We conclude the review by highlighting novel approaches that have been undertaken to harness the gut microbiota as a therapeutic target for treating a spectrum of metabolic diseases. This review establishes our rationale for feeding inulin to T5KO mice as the means for correcting their gut dysbiosis and metabolic syndrome. However, surprisingly, a ~40% subset of T5KO mice developed cholestasis and later HCC following feeding of dietary inulin. The initial goal of Chapter 2 was to investigate the metabolic abnormalities observed in ~40% of T5KO mice fed dietary inulin. At first, we deemed the reduction in body weight, adiposity, glucose and serum lipids in these subset of mice to be beneficial when compared to the other ~60% that retained their metabolic syndrome phenotype. However, these ~40% subset of mice eventually developed hyperbilirubinemia and icteric HCC. Accordingly, we evolved our focus to characterize our so-called ‘dietary inulin-induced HCC’ at the serological and histological levels. To address whether the susceptibility toward inulin-induced HCC is due to microbiota dysbiosis and thus not restricted to T5KO mice per se, we performed similar inulin feeding to other gut dysbiotic strains of mice, such Toll-like receptor 4-deficient (Tlr4KO) and Lipocalin 2-deficient (Lcn2KO) mice. Indeed, a subset of Tlr4KO and Lcn2KO mice also developed hyperbilirubinemia and HCC, albeit at a reduced incidence. Hence, in the subsequent chapters, we focused more on T5KO mice to study inulin-induced HCC due to their higher rate of HCC incidence and reproducibility. Chapter 3 follows our findings that hyperbilirubinemia could be harnessed not only to diagnose inulin-fed T5KO mice with active HCC, but also to predict which mice would develop HCC later. We demonstrate that hyperbilirubinemia could be detected within 2 to 4 weeks of inulin feeding and that 100% of mice displaying this condition would develop HCC 24 weeks later. Based on serum bilirubin levels, we segregated mice into either normal bilirubin (N-bili) or hyperbilirubinemia (H-bili) groups. To delineate the events initiating hepato-oncogenesis, we identified H-bili mice following 2 weeks of inulin-containing diet (ICD) feeding for detailed analysis. Our results collectively suggested that the disease progresses from early onset of cholestasis, hepatocyte injury/death, neutrophils infiltrating the liver, hepatic inflammation, liver fibrosis and cirrhosis, and finally HCC. The initial cholestatic event is also marked by >10-fold upregulation in serum total bile acids, which could be in part explained by dysregulated bile acid metabolism and a reduction in bile flow. By intervening with cholestyramine to block enterohepatic recycling of bile acids, we were able to prevent the HCC progression after 24 weeks of inulin feeding. Chapter 4 addresses whether inulin-induced HCC is dependent on the gut microbiota. First, we examined the microbiotal composition of T5KO H-bili mice fed inulin diet via 16S sequencing and observed that their gut microbiota substantially differed from both T5KO N-bili and WT mice. The linear discriminant analysis effect size (LEfSe) method examined the extent of microbiota dysbiosis and revealed that overabundances of Clostridia spp. and Proteobacteria were microbial signatures uniquely prominent in H-bili mice. The dysbiosis in H-bili mice was also denoted by increased total bacterial burden in the gut, loss in species richness and diversity, and elevated pro-inflammatory microbial ligands in the gut and systemic dissemination. Four key experiments performed herein attested the requirement of H-bili gut microbiota to potentiate inulin-induced HCC. These included (i) ablating the gut microbiota with broad-spectrum antibiotics or (ii) rendering T5KO mice as germ-free, which prevented inulin-induced HCC. Additionally, (iii) co-housing and (iv) cross-fostering WT mice with T5KO mice promoted inulin-induced HCC in WT mice that are refractory to the disease. In Chapter 5, the goal was to investigate whether the gut microbiota and/or their activity could be targeted to promote or demote inulin-induced HCC. We demonstrated that fermentable fibers other than inulin, such as pectin and fructooligosaccharide, could also potentiate hyperbilirubinemia and HCC in T5KO mice, albeit at a reduced incidence and severity compared to inulin. In contrast, feeding the non-fermentable fiber, cellulose, to T5KO mice did not result in either hyperbilirubinemia nor HCC. LC-MS metabolomics on cecal short-chain fatty acids revealed that butyrate was particularly elevated in inulin-fed H-bili T5KO mice. Feeding butyrate to T5KO mice for 36 weeks partially recapitulated the hyperbilirubinemia and liver dysfunction associated with inulin feeding. However, the lack of macroscopically-discernible tumor nodules in butyrate-fed mice suggests that factors other than butyrate may be required to facilitate the progression from cholestasis to HCC. Nevertheless, we demonstrated that inulin-induced HCC could be prevented by interventions that deplete butyrate-producing bacteria via metronidazole treatment or inhibit gut fermentation via β-acids treatment. In conclusion, our study was unprecedented in demonstrating that inulin —a fermentable fiber widely-held to be broadly health-promoting— could offset into cholestasis and HCC in a subset of gut dysbiotic mice. These findings caution that the beneficial effects of a HDS such as inulin may not be universal. We envision that adverse effects such as hyperbilirubinemia and cholemia could be used to identify groups that should abstain from HDS. Future studies are warranted to further investigate whether this phenomena of inulin-induced HCC in dysbiotic mice could be used as an animal model to study the HDS-associated liver pathology seen in humans. This material is based upon work supported by the National Institutes of Health under the R01 Award No. R01CA219144 to Dr. Matam Vijay-Kumar. Any opinions, findings, and conclusions or recommendations expressed in this dissertation are those of the author(s) and do not necessarily reflect the views of the National Institutes of Health.