(L-455) Elucidating the Synergistic Effects of Dietary Factors and Inflammation in a Microfabricated Model of Non-Alcoholic Fatty Liver Disease

Introduction:: Non-alcoholic fatty liver disease (NAFLD) is characterized by hepatic lipid accumulation that exceeds 5% of the liver’s weight, which can cause insulin resistance (diabetes), inflammation, liver fibrosis/cirrhosis, and cancer. Since animal models cannot fully recapitulate human NAFLD, in vitro human models are necessary to investigate NAFLD mechanisms and for high-throughput drug screening. However, current human NAFLD models either rely on functionally declining (insulin resistant) culture of primary human hepatocytes (PHHs) or do not incorporate liver nonparenchymal cells important for NAFLD progression or are too low throughput for routine drug screening. Here, we mitigate these limitations via the engineering of in vitro models of human NAFLD and elucidating the synergistic effects of key dietary factors and inflammation on hepatic functions.

Materials and Methods:: Multiwell plates were subjected to polydimethyl siloxane (PDMS)-based soft lithography to pattern collagen I into circular domains (500 µm diameter, 1200 µm center-to-center spacing). PHHs attached to the collagen domains and were surrounded by supportive 3T3-J2 fibroblasts or fibroblasts mixed with hepatic stellate cells (HSCs), termed either micropatterned co-cultures (MPCCs) or HSC-containing micropatterned tri-cultures (S-MPTCs). Concurrently, a microfluidic device was used to encapsulate PHHs into collagen-I droplets (~200 µm diameter), which were coated with endothelial cells (Fig. 1c). The liver models were treated with select dietary triggers, such as glucose, fructose, ethanol and saturated/unsaturated fatty acids, and in some cases with drugs or antibodies. Hepatic lipid accumulation, insulin resistance, and inflammation markers were assessed via published assays.

Results, Conclusions, and Discussions:: Ethanol (0.25%), fatty acids (FAs), and their combinations resulted in loss of insulin sensitivity in PHHs, with FAs causing the greatest effect (Fig. 1a). When MPCCs and S-MPTCs were treated with FAs for 3 weeks, lipid accumulation in S-MPTCs was significantly increased in comparison with non-treated S-MPTCs and MPCCs treated with FAs (Fig. 1b). It was hypothesized that increased steatosis in treated and non-treated S-MPTCs could be due to the production of interleukin 6 (IL-6) by myofibroblastic (activated) HSCs. Thus, exogenous IL-6 was added to MPCCs, which resulted in lipid accumulation at similar levels to S-MPTCs; similarly, when IL-6 neutralizing antibodies were added to S-MPTCs, steatosis was reduced to levels similar to non-treated MPCCs (not shown). When non-inflamed liver microtissues were treated with FAs, Nile red staining showed a significant increase in steatosis which could be ameliorated with obeticholic acid (OCA), a drug in clinical trials for NAFLD (Fig. 1d); such an outcome was due to the reduction in inflammation as assessed by hepatic C-reactive protein production (Fig. 1e).


We engineered in vitro human NAFLD models and used them to elucidate how dietary triggers and inflammation cause the key dysfunctions of clinical NAFLD, namely hepatic lipid accumulation and insulin resistance; we also showed in proof-of-concept studies that such dysfunctions could be alleviated using neutralizing antibodies or drugs. Future efforts are focused on incorporating additional liver non-parenchymal cells into the NAFLD models and further elucidating the cellular and molecular mechanisms that lead to detrimental outcomes in NAFLD.




Acknowledgements (Optional): :

References (Optional): :