Bioprinting Perfusion-Enabled Liver Equivalents for Advanced Organ-on-a-Chip Applications

Abstract

Many tissue models have been developed to mimic liver-specific functions for metabolic and toxin conversion in in vitro assays. Most models represent a 2D environment rather than a complex 3D structure similar to native tissue. To overcome this issue, spheroid cultures have become the gold standard in tissue engineering. Unfortunately, spheroids are limited in size due to diffusion barriers in their dense structures, limiting nutrient and oxygen supply. Recent developments in bioprinting techniques have enabled us to engineer complex 3D structures with perfusion-enabled channel systems to ensure nutritional supply within larger, densely-populated tissue models. In this study, we present a proof-of-concept for the feasibility of bioprinting a liver organoid by combining HepaRG and human stellate cells in a stereolithographic printing approach, and show basic characterization under static cultivation conditions. Using standard tissue engineering analytics, such as immunohistology and qPCR, we found higher albumin and cytochrome P₄₅₀ 3A4 (CYP3A4) expression in bioprinted liver tissues compared to monolayer controls over a two-week cultivation period. In addition, the expression of tight junctions, liver-specific bile transporter multidrug resistance-associated protein 2 (MRP2), and overall metabolism (glucose, lactate, lactate dehydrogenase (LDH)) were found to be stable. Furthermore, we provide evidence for the perfusability of the organoids' intrinsic channel system. These results motivate new approaches and further development in liver tissue engineering for advanced organ-on-a-chip applications and pharmaceutical developments.

Keywords: 3D cell-culture; bioink; bioprinting; drug development; in vitro testing; liver equivalent; stereolithography; tissue engineering; toxin testing.

Figure 1

Cellbricks printing process of liver model. (a) During the printing process, the virtual 3D model (top view) is translated into a cell-laden multi-material hydrogel; here shown in a 2× microscopic image; (b) The printing process is visualized schematically. (I) The bioink-cell suspension is filled in the printers bioink-reservoir. (II) Digital-light-processing (DLP) projection is done layer-by-layer through the transparent reservoir bottom onto the print head. (III) After each layer, the print head moves upwards for the next layer. (IV) Each layer was photopolymerized onto the previous layer. (V) After complete printing, the formed tissue model is removed from the print head, ready for cultivation.

Figure 2

Morphology of the printed tissue model. (a) Microscopic picture of liver model directly after printing. At 10× magnification, the channel-structure is clearly visible. After complete dissolution, the channel is a hollow structure, as seen by focusing on (b) the edges or (c) the channel bottom. The channel structure is hollow, as demonstrated by flushing with trypan blue dye (d,e).

Figure 3

TUNEL/Ki67 staining of printed liver constructs. Proliferative cells (Ki67 positive) are visible both at day zero and after 14 days of cultivation. Ki67 positive cells appear less at day 14 compared to day 0. Apoptotic cells (TUNEL positive) were observed in the positive DNase-treated control. The negative control did not show any unspecific staining.

Figure 4

Metabolic data of glucose consumption, lactate dehydrogenase (LDH) and lactate concentration over 14 days of cultivation. (a) Average glucose consumption ranged from 0.14 to 0.4 g/L. Cells treated with Triton X-100 consumed more than 1.95 g/L. (b) The LDH release started at about 100 mU/mL, but stabilized at around 70 mU/mL three days post-printing. Triton X-100 treated cells released 830 mU/mL in average. (c) Lactate concentration was measured between 3.7 and 6.4 mmol/mL. Cells treated with Triton X-100 showed 0.2 mmol/mL lactate concentration in the culture medium.

Figure 5

Gene expression results from bioprinted liver tissue compared to HepaRG monolayer cultures. In each graph, the qPCR results of albumin (a), CYP3A4 (b), ZO-1 (c) and MRP2 (d) of monolayer cultures and printed liver constructs at day 14 are compared to day 0, respectively. Results showed statistically significant differences. (*** p < 0.0005; **** p < 0.0001).

Figure 6

Immunohistochemistry staining of printed liver constructs showing expression patterns of vimentin (green) co-stained with cytokeratin 8/18 (red) after 14 days of cultivation shown in a complete section of the print (20× merged) (a). The expression of (b) albumin (Alb), (c) CYP3A4 and (f) ZO-1 is shown in a single staining after two weeks of cultivation. The location overview (d) shows where each staining is located within the construct. The negative control (e) verifies the results.