Regular article
Characterisation of some cytotoxic endpoints using rat liver and HepG2 spheroids as in vitro models and their application in hepatotoxicity studies. II. Spheroid cell spreading inhibition as a new cytotoxic marker

https://doi.org/10.1016/S0041-008X(03)00090-5Get rights and content

Abstract

Cells in liver spheroids and Hep G2 spheroids transferred from gyrotatory culture conditions and maintained in normal static culture conditions will spread out at the edges. Based on this observation, we developed a new test called the Spheroid Cell Spreading Inhibition Test (SCSIT) to screen hepatic cytotoxicity of xenobiotics and determine the spheroid cell spreading inhibition concentration (SCSIC) of test chemicals. Four model hepatoxicants, d-galactosamine, propranolol, diclofenac, and paracetamol, were studied with SCSIT in both rat liver and HepG2 spheroids. Both liver and HepG2 spheroids were prepared under gyrotatory culture conditions and used at 6 days in vitro. The results showed that all four hepatotoxicants tested inhibited cell spreading in liver spheroids (d-galactosamine at 20 mM, propranolol at 125 μM, diclofenac at 500 μM, and paracetamol at 25 mM) and HepG2 spheroids (d-galactosamine at 16 mM, propranolol at 125 μM, diclofenac at 500 μM, and paracetamol at 25 mM). The SCSIT results agreed with the conventional cytotoxic indicators, release of LDH and/or γ-GT and the inhibition of glucose secretion from rat liver spheroids. In conclusion, this study, for the first time, described the biological characteristics of liver and HepG2 spheroid cell spreading and demonstrates its application in hepatic cytotoxicity studies. This method may be used in testing in vitro “acute” toxicity, comparing relative cytotoxicity and generating reference concentrations for subsequent studies. Therefore, SCSIT could be a useful tool for screening hepatotoxicity relevant to preclinical lead optimization and compound library screening.

Introduction

The major concerns in adopting in vitro models to predict in vivo effects are data comparability and reliability. To increase the reliability and predictability of in vitro data to human and other animals, some fundamental aspects, such as similarity of cell functionality, comparability of “dosing” concentrations, and reliability of endpoints must be well defined or “framed.” The simple indicators, LD50 and LC50, play an essential role in animal experiments because they can be used as reference doses for subsequent studies and for comparing relative toxicities among different chemicals and across species. In most in vitro studies, while such a reference scheme has not been adopted, the EC50 or IC50 values indicate a specific endpoint Ponsoda et al 1995, Jinno et al 1997, Clemedson and Ekwall 1999, Ekwall 1999 and LD50 indicated by cell viability (Bironaite and Öllinger, 1997) has been used in some studies. As a result, a wide concentration range (5000-fold, highest/lowest), often with a 10-fold difference between two concentrations can be expected in some in vitro studies (Walker et al., 2000). A wide concentration range is neither a good approach for mechanistic or concentration–response studies nor an efficient practice. On the other hand, some in vitro studies tend to use very high concentrations, which may be difficult to reach in vivo. This may make the data less valuable to the in vivo situation because the data lack comparability in concentration between the two models. Therefore, a good in vitro model should adopt a “dosing” concentration reference system to make data comparable with in vivo data. To achieve this, we must find a way to link the concentrations between in vitro and in vivo models.

The similarity of liver spheroid models to intact tissues in structure and specific functions have been reported by many researchers and reviewed by Mueller-Klieser (1997). For example, under certain culture conditions, such as gyrotation, single liver cells can reorganize to form multicellular, three-dimensional spheroids, in which cells resume cell–cell contacts as seen in vivo Wu et al 1996, Yumoto et al 1996, Walker et al 2000, Tzanakakis et al 2001. The cells in spheroids survive and retain liver specific function much longer than cells cultured as monolayers Landry et al 1985, Lazar et al 1995, Eschbach et al 1997. Cells from adult rat pancreatic islets are capable of self-assembling to form aggregates that secrete insulin (Halban et al., 1987). Normal human adult breast epithelial stem cells formed three-dimensional “organoids” almost identical to those found in vivo (Chang et al., 2001). Mammary gland epithelial cells are able to reorganize into hollow, alveolar-like structures and retain the functions of secreting milk proteins (Barcellos-Hoff et al., 1989). Our recent work using magic angle spinning 1H-NMR, showed that the “finger print” (spectrum) of liver spheroids more closely resembles intact liver than isolated hepatocytes (Bollard et al., 2002). More and more evidence shows that, in some important aspects, a spheroid may resemble a viable, miniorgan that can be maintained long term. Therefore, the usefulness of liver spheroids and spheroids grown from hepatic cell lines as in vitro models is worthy of further characterisation.

This study investigated the cell-spreading behaviour of individual rat liver and HepG2 spheroids and their application to in vitro toxicology studies. Our early observations showed that liver cells in spheroids can spread out under certain culture conditions. This phenomenon may be exploited as a simple and useful endpoint to examine the in vitro “acute” cytotoxicity of a test compound to compare relative toxicity among different test agents. The utility of the cell-spreading phenomenon of rat liver and HepG2 spheroids was characterised using four model hepatoxicants, d-galatosamine, propranolol, diclofenac, and paracetamol, as tested in Part I of this study.

Section snippets

Chemicals and medium

l-Glutamine was obtained from GibcoBril. Foetal bovine serum (FBS), hepatocyte medium, Minimum Essential Medium Eagle (MEM), penicillin and streptomycin sulfate, toxicants, and other chemicals were purchased from Sigma (Sigma-Aldrich Company Ltd) unless otherwise indicated.

Liver spheroid culture

Liver cell preparation and spheroid culture were as stated in Part I.

HepG2 spheroid culture

The HepG2 cell line (Human Caucasian Hepatocyte Carcinomal cells from ECACC) was prepared as described in Part I.

Culture media

The culture medium used for cell-spreading

Spheroid cell spreading

After formation of spheroids from single liver cells under rotation-mediated culture conditions, the individual cells maintained in normal static culture conditions can spread out from the spheroids. Higher concentrations of FCS encouraged cell spreading. Cell spreading in the medium containing 10 % FCS was much better than 5 % FCS in terms of percentage, area, and speed of spreading. There was no significant difference between 10 and 15 % FCS in cell spreading (data not shown). Fig. 1 shows

Discussion

Cytotoxicity of a test compound is identified by a variety of cytotoxic endpoints. The commonly used in vitro cytotoxic endpoints include enzyme release such as LDH, γ-GT, GPT, and GOT Scaife 1985, Verhulst et al 1998, metabolic and functional parameters including neutral red uptake, dimethylthiazole-diphenyl tetrazolium bromide (MTT) metabolism (Sellers et al., 1994; Mosmann, 1983), and pyruvic acid consumption (Verhulst et al., 1998), cell viability (Jinno et al., 1997), and many other

Acknowledgements

The authors thank Dr. David Patton for his support in electron microscopy work.

References (25)

Cited by (36)

  • Critical analysis of 3-D organoid in vitro cell culture models for high-throughput drug candidate toxicity assessments

    2014, Advanced Drug Delivery Reviews
    Citation Excerpt :

    These data have been validated for both static dynamic and spheroid cultures created using Micro-Space Cell Culture plates [114,115,117,118]. Gene expression data were verified by direct exposure to cytochromes P450 (CYP)1A1, CYP1A2 and CYP3A-specific drugs in rocked cultures [114], CYP3A, CYP2C9 and UGT1A1 substrates in the Micro-Space Cell Culture plates [117] and induction of clinically relevant biomarkers such as gamma-glutamine transferase and lactate dehydrogenase release [112]. Results suggest that 3-D organoid hepatic spheroid models are capable of liver-specific metabolism-essential or liver toxicity assessment.

  • In-vitro maintenance and functionality of primary renal tubules and their application in the study of relative renal toxicity of nephrotoxic drugs

    2013, Journal of Pharmacological and Toxicological Methods
    Citation Excerpt :

    We studied the relative toxicity of different drugs using renal tubule aggregates to explore a way of their application in toxicity studies. The principle was based on the spheroid cell spreading inhibition test (SCSIT) as previously reported (Xu, Ma, & Purcell, 2003c). The renal tubule aggregates were used and renal toxic drugs were tested in a similar manner to SCSIT.

  • Development of a novel, physiologically relevant cytotoxicity model: Application to the study of chemotherapeutic damage to mesenchymal stromal cells

    2012, Toxicology and Applied Pharmacology
    Citation Excerpt :

    The co-culture model was validated for clinically relevant doses of several chemotherapeutic agents, including the prodrugs CY, DTIC, and the active drug VIN, using several methods to ensure adverse toxicity was not exerted on HepG2 spheroids. The Cell Spreading Inhibition Test (CSIT) is a previously published method enabling cellular toxicity to be assessed (Xu et al., 2003b). The method is based upon removing spheroids from their normal gyrotatory culture conditions, exposing them to a drug and maintaining at 37 °C but in stationary conditions.

View all citing articles on Scopus
View full text