Cytotoxicity of retinoic acid, menadione and aflatoxin B1 in rat liver slices using Netwell inserts as a new culture system

Toxic. in VitroVol. 9,No. 3,pp.291-298,1995 Copyright 0 1995Elsevier Science Ltd

0887-2333(%)oooo8-9

Printed in Great Britain. All rights reserved

0887-2333/95 $9.50+0.00

Cytotoxicity of Retinoic Acid, Menadione and Aflatoxin B, in Rat Liver Slices Using Netwell Inserts as a New Culture System W. R. LEEMAN,

I. A. van de GEVEL and A. A. J. J. L. RUTTEN

TN0 Nutrition and Food Research Institute, Division of Toxicology, PO Box 360, 3700 AJ Zeist, The Netherlands (Accepted 24 December 1994) Abstract-Precision-cut rat liver slices were used to develop a new dynamic incubation system in which histomorphology and measurement of the release of lactate dehydrogenase (LDH) and the conversion of MTT were applied to evaluate cytotoxicity. Liver slices, precision-cut using a Krumdieck tissue slicer, were cultured in a new system using 200~pm polyester mesh Netwell inserts in six-well cell-culture clusters on a rocker platform at 37°C and 40% 0,. The major advantage of this new culture system is the easy way in which slices can be manipulated and the culture medium be sampled or changed. Rat liver slices were exposed for 4 hr to retinoic acid (RA), menadione or aflatoxin B, (AFB,). Directly after treatment and after an additional 20-hr recovery period, histomorphological observations of slices were made, and LDH release and MTT conversion were measured. Slices exposed to RA showed dose-related cytotoxicity in the MTT assay only. The cytotoxic response to AFB, was more pronounced in the assay of LDH release than in the MTT assay. Histomorphology, LDH release and the MTT assay revealed cytotoxic effects induced by menadione. We conclude that culturing liver slices using Netwell inserts is a good alternative to other culture systems for testing non-volatile compounds.

INTRODUCTION

Precision-cut liver slices are currently used for studying the metabolism of xenobiotics in vitro. The most frequently used incubation system is the dynamic organ culture in sealed vessels as introduced by Smith et al. (1985). Good results were also described for other systems such as incubation in Conway units (Hart et al., 1983), the dynamic organ culture system based on the method of Trowel1 (Connors et al., 1990), or a simple, short-term incubation system as described by Dogterom (1993). Although the dynamic organ culture system is very useful for testing volatile compounds, it is laborious for testing nonvolatile compounds. Also, the high oxygen concentrations used in this culture system may give rise to oxidative damage (Chen et al., 1993a,b). Recently, lower oxygen concentrations were shown to give better results when extended culture periods are used (Wright and Paine, 1992). Metabolism in liver slices is considered to be more comparable to metabolism in situ than that in incubations with isolated hepatocytes or microsomes because of the undisrupted liver structure, intact membrane receptors, full differentiation of cells and

Abbreuiations: AFB, = aflatoxin B,; DMSO = dimethyl sulfoxide; FBS = foe.tal bovine serum; LDH = lactate PBS = phosphate buffered saline; dehydrogenase; RA = retinoic acid; WME = Williams’ medium E.

preserved cell
W. R. Leeman et al.

292

Paine, 1992). AFB, was selected to investigate whether biotransformation of AFB, occurs, because toxicity, which is induced by reactive metabolites of AFB,, is only detectable after prolonged culture periods (Hayes and Pickering, 1985). MATERIALS AND METHODS

Materials Williams’ medium E (WME) supplemented with Glutamax I, phosphate buffered saline (PBS) and gentamicin were obtained from Gibco BRL (Paisley, Scotland). Foetal bovine serum (FBS) was purchased from Integro B.V. (Zaandam, The Netherlands). AN-trans-retinoic acid, menadione, insulin and dimethyl sulfoxide (DMSO) were purchased from Sigma Chemical Company (St Louis, MO, USA). Netwell inserts (200 pm polyester mesh carrier) and six-well cell-culture clusters were obtained from Costar (Cambridge, MA, USA). Heraeus Kulzer (Wehrheim, Germany) supplied Technovit 7100 plastic resin. Aflatoxin B, was obtained from Aldrich Chemical Co. (Milwaukee, WI, USA). MTT was obtained from Aldrich Chemie (Steinheim, Germany). D-Glucose and toluidine blue 0 were purchased from Merck (Darmstadt, Germany).

Animals For preparation of liver slices, young adult male Wistar rats (Crl: WI(WU)BR; 250-300 g) were obtained from Charles River Wiga (Sulzfeld, Germany). For each time point one animal was used. Liuer slice culture Animals were killed by decapitation and 8-mm diameter liver biopsies were obtained using a disposable biopsy punch (Stiefel Laboratorium, Offenbach am Main, Germany). Liver slices (20&250pm) were prepared at room temperature with a Krumdieck tissue slicer (Krumdieck et al., 1980) using WME supplemented with D-glucose to a final concentration of 25 mM and pregassed with 95% 0,/5% COz. Slices were placed on 200 pm mesh Netwell inserts in six-well cellculture clusters (Fig. 1) containing 3.25 ml WME supplemented with D-glucose to 25 mM, 5% FBS, 0.1 pM insulin and 50 pg gentamicimml (culture medium). The tissue-culture clusters were placed in a humidified incubator at 37°C 40% O2 and 5% CO? on a rocker platform (approximately 10 cycles/min). The slices were immersed in culture medium for about half of the cycling time and in the

Fig. 1. Schematic drawing of the tissue culture system: the Netwell insert (A) was placed in a six-well culture cluster (B) with tissue slices (C) placed on the 200-mesh carrier. The cross-section shows tissue slices immersed in culture medium (D). Arrows indicate the movement of the rocker platform.

Cytotoxicity in rat liver slices

upper position the slice was exposed to air for the other half of the cycling time. Treatment

After 1 hr of preincubation the culture medium was replaced with medium containing the test compound. The slices were treated for 4 hr with either RA (final concentration 0.1, 10 or 1000 PM), menadione (final concentration 50, 100, 200 or 400 PM) or AFB, (final concentration 0.01, 0.1 or 1.OPM). Stock solutions were prepared with DMSO as solvent. The final DMSO concentration during treatment was 1% for all cultures. Immediately after 4 hr of treatment the culture medium was removed. One series of slices was processed immediately after treatment for the MTT assay, whereas another series of slices was cultured for an additional 20 hr in culture medium. All culture medium was collected immediately after treatment and 20 hr after treatment for measurement of LDH release. The slices were collected for the MTT assay and assessment of histomorphology. LDH leakage

Samples of culture medium were analysed for LDH content by the IFCC protocol at 37°C (Boehringer Germany) using the Hitachi 911 Mannheim, analyser. MTT assay

To detect chemically induced toxicity after RA, menadione or AFB, treatment, the slices were incubated after the treatment period with 1.2 mM MTT solution dissolved in WME. After a I-hr incubation at 37”C, 40% O2 and 5% COZ, the MTT solution was removed. Each slice was rinsed with cold PBS and the MTT-formazan product was extracted in the dark with 1 ml DMSO. After at least 1 hr of extraction, the MTTformazan extract was diluted with DMSO (1: 5, v/v) and absorbance was measured at a wavelength of 505nm. MTT-formazan absorbance was related to the dry weight of the slice. Dry weights were obtained by rinsing the slices twice with acetone, whereafter the slices were dried at room temperature. Histology

After treatment, slices were fixed in Bouin fixative, dehydrated, embedded in Technovit 7100 plastic resin, sectioned at 2 pm and stained with 1% toluidine blue. Histomorphological changes in the slices were evaluated by microscopical examination of the cross sections. Cytotoxicity was evaluated by scoring alterations in nuclear shape, staining intensity of the cytoplasm and visibility of the cell membrane. Statistical analysis

One-way analysis of variance (ANOVA) was used for statistical analysis. Comparisons were made using the Dunnett test for paired data and the Williams test for trend determination with P < 0.05 considered to be significant.

0

24

culture

48

72

period

(hr)

170

Fig. 2. MTT assay with rat liver slices cultured for up to 170 hr. Values are means k SD. Each bar represents the mean of three replicates from one experiment (Dunnett’s test: **P < 0.01). RESULTS MTT assay

As a preliminary tration

relationship

experiment, the time-concenfor the MTT assay using rat liver

slices was investigated. Incubations of rat liver slices with 1.2 mM MTT showed a linear MTT-formazan production up to 120 min. Histomorphoiogical examination and the results of the MTT assay (Fig. 2) indicated that the precision-cut liver slices cultured on Netwell inserts were viable for at least 72 hr. No significant changes were observed in MTT-formazan production during this period. Although a significant decrease of MTT-formazan production was observed after 170 hr of incubation, histomorphological examination indicated that the hepatocytes were still viable. Chemically induced cytotoxicity

All test compounds were dissolved in DMSO. The concentration of DMSO used in the experiments did not affect the LDH or MTT results (results not shown). On treatment with RA, no statistically significant effect on LDH release was observed (Fig. 3A), whereas a small, dose-related decrease in MTT conversion was observed immediately after treatment but not at 20 hr after treatment (Fig. 3B). In comparison with the control slices, no significant differences were observed in the histomorphology of RA-treated liver slices at either sampling time (Plate lB, F; Table 1). Menadione showed a dose-related cytotoxic effect as shown by LDH release and the MTT assay determined immediately after treatment (Fig. 3C, D). In addition, a dose-related effect was observed on

W. R. Leeman er al.

294

DISCUSSION

histomorphology. At a concentration of 100~~ menadione, no or only minor histomorphological changes were observed, whereas severe damage was observed at a concentration of 400 pM menadione (Plate lC, G; Table 1). Although slices exposed to 50 PM menadione showed an increase in LDH release 20 hr after treatment, no statistically significant effect was found. With respect to cytotoxic characteristics, histomorphology was less affected compared with, LDH release and the MTT assay, immediately after treatment. However, 20 hr after treatment changes in histomorphology were in agreement with the effects on LDH release and the MTT assay, showing severe damage at concentrations ranging from 100 to 400 PM (Plate lC, G; Table 1). Immediately after treatment with AFB, no toxic effects were observed, although a slight increase of LDH release was found at 0.01 PM AFB,. However, 20 hr after treatment AFB, induced a dose-related cytotoxic effect as indicated by both LDH release and the MTT assay (Fig. 3E, F). Only LDH release showed statistically significant differences after exposure to 0.1 or 1.0 pM AFB, when compared with the control. Marked histomorphological damage was observed only after exposure to l.OpM AFB,, 20 hr after treatment (Plate IH; Table 1).

In the present study we investigated a relatively easy-to-use culture system for the evaluation of chemically induced cytotoxicity in precision-cut rat liver slices. The dynamic organ culture system as developed by Smith et al. (1985) was modified using Netwell inserts in six-well culture clusters placed on a rocker platform. Under these conditions, precisioncut liver slices remained viable in culture for 72 hr, as evaluated by the MTT assay and histomorphology. The same prolonged culture times were demonstrated by Beamand et al. (1993) and Lake ef al. (1993) using the dynamic organ culture system (Smith et al., 1985). The MTT assay was initially developed for cell cultures and there have been only a few reports of its use with liver slices (Azri et al., 1992; Azri-Meehan et a/., 1992). We determined the optimal MTT incubation time and concentration to use in the present culture system. A concentration of 1.2 mM MTT and an incubation time of 60min was chosen for the establishment of cytotoxicity in the MTT assay. Although longer incubation periods, as well as higher concentrations of MTT, gave a higher MTT-formazan absorbance, our incubation conditions were selected to avoid the possible cyto-toxic effects induced by intracellular formazan crystals.

4

(E)

(A)

.*

t

_..t CD)

0

retinoic

acid (PM)

-I

-t (F)

so

100200400

menadione

(PM)

aflatoxin

B,

(I(M)

Fig. 3. Effects of (A) retinoic acid, (C) menadione and (E) aflatoxin B, on LDH release into the culture medium and in the MTT assay (B, D and F) immediately after treatment (W) and 20 hr after treatment (0). No slices were collected for the MTT assay immediately after treatment with 50~~ menadione. Values are means (n = 3 for LDH release; n = 4 for MTT assay) and range bars indicate the SD. (Dunnett’s test: *P < 0.05; **P < 0.01; Williams’ trend test: OP < 0.05).

Plate I. Photomicrographs of rat liver slices treated for 4 hr with test compounds. One series was fixed for histomorphological examination immediately after treatment: (A) DMSO; (B) 1000 PM retinoic acid; (C) 200 FM menadione; (D) I.0 PM agatoxin B,. Another series was cultured for a further 20 hr without test compounds before fixation: (E) DMSO; (F) 1000 pM retinoic acid; (G) 100 pM menadione; (H) 0.1 pM aflatoxin B, Toluidine blue staining, bar represents 50 pm. 295

Cytotoxicity in rat liver slices Table

I. Histomorphology acid,

of liver slices after treatment menadione

or afitoxin

with

ntinoic

B,

Histomorphology*

Treatment Control

Cone (DMSO)

Retinoic

(p

M)

-

++ ++

IO

++

++

1000

++

++

Menadione

50

nd

B,

changes;

minor,

+ = marked Two

+ _

100

++

200

f _

-

0.01

++

++

0.1

++

++

1.o

++

400

determined.

treatment

++

0.1

l+ + = no, or only

20 hr after

++

acid

Aflatoxin

0 hr after treatment

histomorphological damage;

-

= severe

liver slices were examined

_

f changes;

+ = clear

damage;

nd = not

for each experimen-

tal point.

Furthermore, 1.2 mM MTT is commonly used for detecting chemically induced cytotoxic effects on cell cultures. Under the experimental conditions described in this paper, we determined the cytotoxic effects of RA, menadione and AFB, . Immediately after treatment with RA, a dose-related decrease in MTT-formazan production was found. LDH release and histomorphology did not show any RA-induced effects. At 20 hr after treatment the effect of RA in the MTT assay was absent, which may indicate a recovery of the cells. Although the absence of RA-induced cytotoxic effects, as measured by LDH release, was also found in a study using hepatocytes (Budroe et al., 1987) an inhibition of intracellular LDH activity, as described by Tsao et al. (1984), can mask a possible cytotoxic effect. In this case the MTT assay may be more sensitive in detecting RA-induced cytotoxicity. Utley and Mehendale (1989) found that menadione-induced cytotoxicity was maximal at both 95% O2 (25 ppm) and 2 1% O2 (6 ppm) and was absent at 0% O2 in rat hepatocytes treated for 3 hr with menadione (10&400~~). This indicates that the threshold level of oxygen required for the expression of menadione toxicity in hepatocytes lies between 0 and 6 ppm. Wright and Paine (1992) using 21% 02, found that menadione was cytotoxic at a concentration of 200~~ in hepatocytes but not in liver slices. This effect was ascribed to non-absorption of menadione in the slices, a low intracellular oxygen concentration, or a more active detoxication of menadione in liver slices. Chan et al. (1992) found that menadione was cytotoxic at concentrations ranging between 100 and 300 PM in rat liver slices at 95% OZ. The results of the present study, using 40% 02, showed clear menadione cytotoxicity at concentrations ranging from 50 to 400 PM and therefore we suggest that the oxygen concentration during incubation may play an important role in menadione toxicity in rat liver slices. An important difference between culturing hepatocytes and slices with respect

297

to available oxygen is that hepatocytes are completely surrounded by tissue culture medium with a uniform oxygen concentration, whereas in precision-cut slices in culture only the outer cell layer is in direct contact with the culture medium. For other cells in the inner part of the slice, oxygen must be distributed through diffusion, which may result in a lower intracellular oxygen concentration. The absence of cytotoxicity in slices treated with 200 PM menadione and cultured at 21% OX(Wright and Paine, 1992), may indicate that the threshold oxygen level required for menadione toxicity had not been reached. Because of this, culturing slices at slightly higher oxygen concentrations may provide a system that is more similar to exposure in vivo, where oxygen is well distributed by the blood to each cell in the liver. In precision-cut liver slices exposed to AFB, no cytotoxicity was observed immediately after treatment, but cytotoxic effects of AFB, were observed 20 hr after treatment. The delayed cytotoxicity is probably related to biotransformation of AFB, into reactive metabolites. These findings were in agreement with those of studies of AFB, treatment of rat hepatocytes, as published by Hayes and Pickering (1985) and Hayes et al. (1986) who found a timeand concentration-dependent AFB, cytotoxicity as measured by LDH release in culture medium at treatment times between 18 and 48 hr. Although we found a dose-related AFB,-induced toxic effect in the MTT assay, LDH release was a more sensitive indicator of the toxic effect. Both LDH release into the culture medium and the MTT assay were used to study cytotoxic effects after chemical treatment in precision-cut liver slices. LDH release was a more sensitive indicator of AFB, toxicity than the MTT assay, but in contrast, the MTT assay was more sensitive in the case of RAinduced toxicity. This may indicate that both measurement of LDH release and the MTT assay are required for the overall detection of cytotoxicity induced by unknown compounds. The major advantage of this new culture system is its flexibility with respect to sampling and renewal of the culture medium, the latter being of importance for prolonged culture of slices. An oxygen concentration of 40% in this culture system seems to give a good oxygen supply because oxygen-driven metabolic pathways, as shown by menadione toxicity, did run properly. Furthermore, exposure to 95% 02, which is frequently used during culture, may result in oxidative damage of haem proteins in liver slices (Chen et al., 1993a,b). The 40% oxygen concentration used in this system is therefore considered to be more favourable than the high oxygen concentration of 95%. Couclusions

From the present experiments we conclude that Netwell inserts provide a relatively quick and easy model for prolonged culture of precision-cut rat liver

W. R. LeeInan et al.

298

slices. The most important improvement is the simple way in which the slices can be manipulated or culture medium can be sampled during culture. Furthermore, this model can be used for detecting cytotoxic effects of non-volatile compounds. Both release of LDH and the MTT assay are suitable for detecting cytotoxic effects in this new model system, and the sensitivity of these two indicators of cytotoxicity will depend on the chemical being tested. The use of a 40% oxygen concentration during culture is important for the detection of cytotoxic effects of chemicals that include the expenditure of oxygen during metabolic processes. Acknowledgement,s-The authors would like to thank Costar Europe Ltd for their generous gift of the Netwell inserts and in particular Mr P. Boerrigter MSc for his efforts in providing us with a preproduction of Netwell inserts. We thank Mr A. .I. M. Hagenaars MSc for statistical analysis, and Professor P. J. van Bladeren and Dr N. J. Snoeij for helpful discussions. The study was supported by the TN0 Network Alternatives to Animal Assays.

REFERENCES

Azri S., Mata H. P., Reid L. L., Gandolfi A. J. and Brendel K. (1992) Further examination of the selective toxicity of CCI, in rat liver slices. Toxicology and Applied Pharmacology 112, 81-86. Azri-Meehan S., Mata H. P., Gandolfi A. J. and Brendel K. (1992) The hepatotoxicity of chloroform in precision-cut rat liver slices, Toxicology 73, 239-250. Beamand J. A., Price R. J., Cunninghame M. E. and Lake B. G. (1993) Culture of precision-cut liver slices: effect of some peroxisome proliferators. Food and Chemical Toxicology 31, 137-147. Budroe J. D., Shaddock J. G. and Casciano D. A. (1987) Modulation of ultraviolet light-, ethyl methanesulfonate-, and 7,12-dimethylbenz[a]anthracene-induced unscheduled DNA synthesis by retinol and retinoic acid in the primary rat hepatocyte. Environmental and Molecular Mutagenesis 10, 1299139. Chan, H. M., Tabarrok R., Tamura Y. and Cherian M. G. (1992) The relative importance of glutathione and metallothionein on protection of hepatotoxicity of menadione in rats. Chemico-Biological Interactions 84, 113-124. Chen H., Pellett L. J., Andersen H. J. and Tappel A. L. (1993a) Protection by vitamin E, selenium, and /?carotene against oxidative damage in rat liver slices and

homogenate. Free Radical Biology and Medicine 14, 473-482. Chen H., Tappel A. L. and Boyle R. C. (1993b) Oxidation of heme proteins as a measure of oxidative damage to liver tissue slices. Free Radical Biology and Medicine 14, 509-517. Connors S., Rankin D. R., Gandolfi A. J., Krumdieck C. L., Koep L. J. and Brendel K. (1990) Cocaine hepatotoxicity in cultured liver slices: a species comparison. Toxicology 61, 171-183. Dogterom P. (1993) Development of a simple incubation system for metabolism studies with precision-cut liver slices. Drug Metabolism and Disposition 21, 699-704. Guguen-Guillouzo C. and Guillouzo A. (1983) Modulation of functional activities in cultured rat hepatocytes. Molecular and Cellular Biochemistry 53, 35-56. Hayes M. A. and Pickering D. B. (1985) Comparative cytopathology of primary rat hepatocyte cultures exposed to aflatoxin B,, acetaminophen, and other hepatotoxins. Toxicology and Applied Pharmacology 80, 345-356. Hayes M. A., Murray C. A. and Rushmore T. H. (1986) Influences of glutathione status on different cytocidal responses of monolayer rat hepatocytes exposed to aflatoxin B, or acetaminophen. Toxicology and Applied Pharmacology 85, I-10. Hart A., Mattheyse F. J. and Balinsky J. B. (1983) An organ culture of postnatal rat liver slices. In Vitro 19, 841-852. Krumdieck C. L., DOS Santos J. E. and Ho K. (1980) A new instrument for the rapid preparation of tissue slices. Analytical Biochemistry 104, 118-123. Lake B. G., Beamand J. A., Japenga A. C., Renwick A., Davies S. and Price R. J. (1993) Induction of cytochrome P-450-dependent enzyme activities in cultured rat liver slices. Food and Chemical Toxicology 31, 3777386. Mosmann T. (1983) Rapid calorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. Journal of Immunological Methods 65, 55563. Smith P. F., Gandolfi A. J., Krumdieck C. L., Putnam C. W., Zukoski C. F., Davis W. M. and Brendel K. (1985) Dynamic organ culture of precision liver slices for in vitro toxicology. I& Sciences 36, 1367-l 375. Tsao M. S., Nelson K. G. and Grisham J. W. (1984) Biochemical effects of 12-0-tetradecanoylphorpol-13acetate, retinoic acid, phenobarbital, and 5-azacytidine on a normal rat liver epithelial cell line. Journal of Cell Physiology 121, 16. Utley W. S. and Mehendale H. M. (1989) Phenobarbitalinduced cytosolic cytoprotective mechanisms that offset increases in NADPH cytochrome P450 reductase activity in menadione-mediated cytotoxicity. Toxicology and Applied Pharmacology 99, 3233333. Wright M. C. and Paine A. J. (1992) Resistance of precisioncut liver slices to the toxic effects of menadione. Toxicology in Vitro 6, 475481.