Precision-cut human liver slice cultures as an immunological platform

Abstract

The liver is the central metabolic organ in the human body, and also plays an essential role in innate and adaptive immunity. While mouse models offer significant insights into immune-inflammatory liver disease, human immunology differs in important respects. It is not easy to address those differences experimentally. Therefore, to improve the understanding of human liver immunobiology and pathology, we have established precision-cut human liver slices to study innate immunity in human tissue. Human liver slices collected from resected livers could be maintained in ex vivo culture over a two-week period. Although an acute inflammatory response accompanied by signs of tissue repair was observed in liver tissue following slicing, the expression of many immune genes stabilized after day 4 and remained stable until day 15. Remarkably, histological evidence of pre-existing liver diseases was preserved in the slices for up to 7 days. Following 7 days of culture, exposure of liver slices to the toll-like receptor (TLR) ligands, TLR3 ligand Poly-I:C and TLR4 ligand LPS, resulted in a robust activation of acute inflammation and cytokine genes. Moreover, Poly-I:C treatment induced a marked antiviral response including increases of interferons IFNB, IL-28B and a group of interferon-stimulated genes. Therefore, precision-cut liver slices emerge as a valuable tool to study human innate immunity.

Keywords: Antiviral; Hepatocyte; Innate immunity; Liver slice culture.

Fig. 1.

Metabolic activity of human liver slices after ex vivo culturing for a 15-day period. (A) MTS activity of liver slices normalized by the fresh cell weight of the liver slices. (B) Measurements of fresh weight of liver slices. Error bars indicate the 95% confidence interval of standard deviation of seven replicates at each time point. Asterisks indicate the statistically significant differences compared with time zero using the two-way ANOVA test (*, P < 0.05).

Fig. 2.

Expression of liver cell-type-specific genes in the ex vivo culture. Genes are classified as (A) Hepatocyte, (B) Kupffer cell, (C) LSEC, (D) HSC. Data are expression based on qRT-PCR and relative to time zero. Error bars indicates the standard deviation of three liver slices cultured from the same patients at each time point. Similar trend of expression changes was also observed in the other two analyzed patients.

Fig. 3.

Innate immune response following liver slicing. (A) Heat map showing gene expression changes of a panel of innate immune genes in liver slices after vibratome slicing and ex vivo culturing. The time points include day 0, 1 h and 12 h, days 1, 2, 4, 7, 10 and 15. Red color indicates increased abundance, and blue color indicates decreased abundance. Black bar means the gene was not detected. (B) Detailed time course changes of ACTB, IL-6, COL1A1, and TGFB1. Data of three individual donors are identified as PT_a, PT_b and PT_c. Error bars are the standard deviation of three liver slices cultured from the same patient at each time point. Also see Fig. S4. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 4.

Hematoxylin and eosin (H&E) stain of liver slices. (A–D) Morphology of liver slices at days 0, 2, 4 and 7. Scale bar represents 50 μm length. Injured hepatocytes appeared by day 2 but viable hepatocytes were also present at all time points. Also see Figs. S5–7.

Fig. 5.

Innate immune response of liver slices to Poly-I:C and LPS stimuli. Fold changes are referenced to the day 7 time zero slices. Red color indicates increased abundance, and blue color indicates decreased abundance. The PBS treatment is included as the control. Solid blocks of red indicate increased expression of chemokine genes and anti-viral genes, starting at 4 h after treatment. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 6.

Antiviral response preferentially induced by Poly-I:C compared with LPS. Box-and-whisker plot shows the inter-quartile range, the median value, and the 95% confidence interval of the standard deviation of data representing six patient samples. Responses were examined at four time points, including 4 h, 8 h, 12 h, 24 h. Fold changes are referenced to the time zero for each patient. A positive value indicates a more marked increase induced by Poly-I:C compared with LPS; conversely a negative value indicates a more marked increase induced by LPS compared with Poly-I:C. Statistical significance is indicated (*, < 0.05; **, < 0.01; two-way ANOVA test).