Pergamon

0887-2333(94)0020%

Toxic. in Vim Vol. 9, No. 2, pp. 133-138, 1995 Copyright 8 1995 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0887-2333195 $9.50 + 0.00

Effect of High Polyol Concentrations on the Neutral Red Absorption Assay and Tetrazolium-MTT Test of Rat Hepatocytes in Primary Culture
P. OLIVIER*,
*Roquette F&es,

P. TESTARD*,

D. MARZINT and D. ABBOTT$

Institute, 1 rue du Pr Calmette 41 I50 Rilly sur Loire, France

62136 Lestrem, tToxicology

BP 245, 59019 Lille Cedex and SDATOX

(Accepted

Department, Pasteur Consultants, La Halerie,

19 September 1994)

Abstract-The effects of three polyols, mannitol, sorbitol and xylitol, on primary cultures of rat hepatocytes were studied with the aid of two cytotoxicity markers, the MTT test and the neutral red assay. A dose-related increase in the uptake of neutral red (up to approximately 60%) was observed in comparison with untreated cells, whereas the MTT test was unaffected by concentrations of polyols up to 330 mM (620 mOsmol/litre). The increase in neutral red uptake does not indicate cellular toxicity as it was reversible and was shown to be due to an increase in the size of lysosomes. For compounds acting on cellular metabolism, such as phenol, mercuric chloride and pentobarbital, it was shown that the increase in neutral red uptake may minimize any cytotoxic effect of these compounds. The neutral red assay should therefore be used with caution when testing compounds that provoke lysosomal swelling. We recommend the use of multiple, mechanistically different, cytotoxicity markers for the evaluation of xenobiotic cytotoxicity

INTRODUCTION

Polyols are components present in numerous preparations including cosmetics, drugs and foods. These hydrogenated reducing sugars provide greater chemical stability and affinity for water than unmodified sugars without altering sweetness. Moreover, this chemical modification of sugars generally lowers their tendency to crystallize, makes their metabolism noninsulin-dependent, strongly depresses their cariogenicity, and reduces their metabolizable energy. The aim of the study reported here was to examine the eventual cytotoxic effects of three commonly used polyols: mannitol, which is poorly metabolized in mammals, including humans (Allison, 1979), and sorbitol and xylitol which are extensively metabolized (Sicard, 1982; Voedingsraad, 1987). Because the liver is the main organ involved in the biotransformation of these polyols, the effects of these compounds were examined in rat hepatocytes in primary culture, with the aid of two cytotoxicity markers. One of these
Abbreoiarions:

markers, the MTT assay, is an indicator of metabolic activity (Mosmann, 1983). This test is based on the reduction of soluble yellow MTT tetrazolium salt to a blue insoluble MTT formazan product by mitochondrial succinic dehydrogenase. The other marker, the neutral red (NR) assay, is an indicator of intact cell membrane integrity (Borenfreund and Puerner, 1984). This test is based on the uptake of neutral red, a supravital dye, with viable, uninjured cells accumulating neutral red in the lysosome. For the quantitation of viable cells, both assays are based on calorimetric measurements after incubation of the cells with test substances.

MATERIALS

AND METHODS

Chemicals

DMSO = dimethyl sulfoxide; NR = neutral red; NR,, = concentration resulting in a 50% decrease in absorption relative to controls; NR, = concentration resulting in a 10% decrease in absorption relative to controls: buffered PBS = phosphate saline; SDS = sodium dodecyl sulfate. 133

Mannitol, sorbitol and xylitol (purity > 98%) were obtained from Roquette (Lestrem, France), Neutral red was obtained from Fluka (St Quentin Fallavier, France) and MTT, sodium dodecyl sulfate (SDS) and phenol were from Sigma (St Quentin Fallavier, France). Other compounds used were mercuric chloride (HgCI,, RP Normapur; Prolabo, Paris, France), dimethyl sulfoxide (DMSO; Merck, Darmstadt, Germany) and sodium pentobarbital (Sanofi, Montpellier, France).

134 Hepatocyte preparation

P. Olivier e/ (I/. (w/v) CaCI,:OS% (v/v) formaldehyde], and then IO0 p I of a solution of I % (v/v) acetic acid : 50% (v/v) ethanol was added to each well to extract the dye. After rapid agitation on a microtitre plate shaker the absorbance was read at 540 nm. The mean absorption and the standard deviation were calculated for each concentration tested. Measurement of osmotic pressure The osmolarity was obtained by measuring the freezing point depression in an Osmometer (Fikes Associates, Science Park, Burlington, MA, USA). Osmotic pressure qfpolyol-containing media The presence of polyols increases the osmolarity of the culture medium. In order to maintain a constant osmotic pressure of the culture medium, Hams F12 medium was diluted with pyrogen-free water supplemented with the same additives, at the same concentration as present in the complete medium. The polyols were added at various concentrations to obtain a similar osmolarity as present in the complete medium (Table I).

Male Sprague-Dawley rats (I XL300 g) obtained from Iffa-Credo (IArbresle, France) were used for hepatocyte preparations. Hepatocytes were isolated according to the procedure described by Guguen et al. (1975) and modified by Seglen (1976). Culture conditions Hepatocytes were maintained in 200~1 of Hams Fl2 medium supplemented with 20 U penicillin/ml, 20 pg streptomycin/ml, 0.05 fig Fungizone/ml, 0.3 mg glutamine/ml, 2 mg NaHCO,/ml (Gibco, Cergy Pontoise, France) and 5% foetal calf serum, 5 pg insulin/ml, 0.2 mg bovine serum albumin/ml, hydrocortisone/ml (Sigma), (complete 3.5 pg medium) into 96-well microculture plates at a density of 35,000 cells/well in a humidified atmosphere of 5% CO,/95% air at 37C. Cytotoxicity assq After 4 hr, when the cells were attached to the culture plate, the medium was replaced by 200 ~1 of complete medium containing one of the following test substances (mannitol, sorbitol, xylitol, HgCl?, SDS, DMSO or phenol) at various concentrations. Each concentration was tested in six wells. Cells were incubated for 20 hr at 37C. Six wells containing complete medium without test substances were used as controls. Six wells containing complete medium but not cells were used as blanks. MTT assq~ The MTT assay was performed according to the method of Mosmann (1983) modified by Carmichael et al. (1987). Briefly, the test substance was carefully removed by inverting flicking, blotting the plate and washing the culture with phosphate buffered saline (PBS). I50 ~1 complete medium was added to each well together with 50 p I of a solution of 2 mg MTT/ml PBS. After incubation for 4 hr, supernatant was removed as previously described and the blue formazan product obtained was dissolved in 200~1 dimethyl sulfoxide. The plate was then shaken for 5 min on a microtitre plate shaker and the absorbance was read at 540 nm using an automatic kinetic microplate reader UVmax (Molecular Devices, Menlo Park, CA, USA). For each concentration (six wells) tested, the mean absorption and the standard deviation were calculated. Neutral red ussaJ The neutral red (NR) assay was performed according to the method of Borenfreund and Puerner (1984). Briefly, the test agent was removed by inverting the multiwell plate and washing the culture with PBS. After 3 hr of incubation with 200~1 of a solution containing 50 pg NR/ml complete medium, the cells were washed quickly with a fixative [I%

RESULTS

EfSect of mannitol as determined by MTT and NR assays The addition of mannitol to the culture medium increased the osmotic pressure in a concentrationdependent manner (Fig. 1). The MTT test showed that for doses ranging from 30 to 330 mM, the mitochondrial dehydrogenase activity was not altered (Fig. I). Between 310 and 635 mOsmol/litre of culture medium, the oxidation capacity of MTT was not modified (Fig. I). With the NR assay, significant increases in dye uptake occurred beginning at 30 mM mannitol (120% of control value) and reached a peak between 140 and 220 mM (160%). After this peak the uptake values fell, being 150% at 275 mM and 135% at 330 IIIM of mannitol (Fig. I). Effkct of sorbitol as determined by MTT and NR assays When sorbitol was added to the culture medium, similar results to those obtained with mannitol were seen (Fig. 2). At concentrations up to 80m~, a slight increase in mitochondrial reduction was seen (107% k 4;

Table obtain Hams dilution factor 4:s 3:4 2:3 I:2

I. Concentrations Ihe same osmotic

of polyols

added

to Hams

F I2 medun Hams

10 FI2

pressure as present in complete of polyol SorbikA 25 55 x0 I40 (IllM) -.-. Xylitol 35 65 100 130

Concn ----.-~ Mannitol 25 55 80 140

~~~~--- .--..--~

Osmolanly (mOsmol/litre) 295 295 300 300

Effect of
180 170 160 = :! Z 8 i; F$ z 2 ; : 130 120 110 150 140

polyols on two cytotoxicity

markers

135

650

100 350 90

/
80
0

I
25

I
50

I
75

I
100

I
125

I
150

I
175

I
200

I
225

I
250

I
275

I
300

I
325

350

300

Mannitol

concn (mM)

Fig. I. Effect of various mannitol concentrations on hepatocytes in primary culture after 24 hr as determined by the MlT (m) and neutral red (0) assays and osmotic pressure (0). Results of the viability tests are expressed as a percentage of the controls, arbitrarily set to 100%. Values are means & SD (n = 3) and range bars represent the SD. Symbols marked with asterisks differ significantly from the control (Mann-Whitney test): *P < 0.05; **P c 0.01.

P < 0.01). Above this concentration, up to 330 mM sorbitol, the MTT test remained around control values (Fig. 2). At 30 mM sorbitol the NR uptake was 125% of the control value. Similarly to the obser180

vations with mannitol, the maximum uptake for sorbitol occurred between 110 and 220 mM (around lSO%), and was followed by a slight decrease starting at 275 mM (140%) (Fig. 2). As observed for mannitol,
1
700

650 160

550

is tj 5 e i e p. .o j g

500

450

400

350

300 350

25

50

75

100

I2S

1.50

17.5

200

22.5

250

275

300

325

Sorbitol

concn (mM)

Fig. 2. Etrect of various sorbitol concentrations on hepdtocytes in primary culture after 24 hr as determined by the MIT(m) and neutral red (0) assays and osmotic pressure (0). Results of the viability tests are expressed as a percentage of the controls, arbitrarily set to 100%. Values are means k SD (n = 3) and range bars represent the SD. Symbols marked with asterisks differ significantly from the control (Mann-Whitney test): *P < 0.05; **P < 0.01.

136
160 150

P. Olivier et nl.
I-

700

650 140 130 = z 5 i; u 5 + _, 2 $ z 120 550 110 100 90 450 80 70 60 350 50 300 400 500 600 z .z P z $ 0 g 2 2 * : a .o 5 2

25

50

75

100

125

150

175

200 concn

225

250

275

300

325

350

375

400

Xylitol

(InM)

Fig. 3. Effect of various xylitol concentrations on hepatocytes in primary culture after 24 hr as determined by the MTT (m) and neutral red (0) assays and osmotic pressure (0). Results of the viability tests are expressed as a percentage of the controls, arbitrarily set to 100%. Values are means + SD (n = 3) and range bars represent the SD. Symbols marked with asterisks differ significantly from the control (Mann Whitney test): *P < 0.05: **P < 0.01.

the increase in the osmotic pressure medium reached 325 mOsmol/litre.

of the culture

Recersibility

of the efects

ofpolyols

in the NR assay

Efltict of xylitol us determined by MTT and NR ussu~s

Figure 3 shows the results obtained for xylitol. Up to concentrations of 130 mM xylitol. MTT oxidation remained comparable with that of the appropriate controls. Above that concentration it decreased significantly (Fig. 3). In the NR test, after increasing to 140% at 100 mM xylitol, the uptake began to decrease and fell below control values at a concentration of 330 mM. The highest concentration (395 mM) corresponded to an osmotic pressure of the culture medium of 700 mOsmol/litre.

The reversibility of the effects of polyols was determined after exposure of the hepatocytes to increasing concentrations of polyols over a period of 20 hr. Subsequently, the medium was replaced every 24 hr by fresh medium without polyols. NR assays were performed 24, 48 and 72 hr after cessation of exposure to sorbitol (Fig. 4). At both 0 and 24 hr. the uptake of NR was around 140% of the value observed for the controls and only at 48 hr did the uptake begin to decrease. After 72 hr in complete medium, NR uptake was similar to that observed in controls.
Table 2 Elects of polyols taken up in media with a constant osmotic pressure as determined bv an MTT test and a neulral red assav Assays resull (% of control)t Polyol MHlllltOl Concn (InM) 2s 55 80 140 25 5s 80 140 3s 65 100 130 MTT 103+4 101 i_ 5 95 f 6 74 * 4** 105 107 10s 90 + IO +4* * 7 5 4 NR 129 143 IS9 135 t + i_ * 8** x** 4** 7

Efltict of polyols taken up in media vi//l a constant osmotic pressure When the culture medium containing one of the three polyols was maintained at the same osmotic pressure as the complete medium, similar results to those described above were obtained (Table 2). However. when the results obtained with the NR assay under conditions of constant osmotic pressure. are compared with those obtained with increasing osmotic pressure (the polyol concentration being constant), an increase in NR uptake was seen while the oxidation of MTT remained unaffected. These differences can be explained by a decreased availability of nutrients at high concentrations, due to the dilution of the medium.

Sorbltol

147 i_ io** 162 + 14 IS4 f 7 ll7+6 IS4 i 13 136 t i7** I41 * 9** 79+9**

Xylilol

109 * 2** IOX * 3** II?? I** 96 + 4

tResulls are expressed as percenlages of the value obtained wllh control cells. Values are means k SD (n = 6) and those marked with asterisks differ signiticantly (Mann-Whitney test) from the corwols. (P < 0.05; **P < 0.01).

Effect of polyols on two cytotoxicity markers

137

80
O

25

50

75

100

125

150

175

200
(mM)

225

250

275

300

325

350

Sorbitol concn

Fig. 4. Effects on hepatocytes exposed to sorbitol for 24 hr and then incubated in fresh complete medium for 24 (m), 48(o) or 72 hr (0) as determined by the neutral red assay. Results are expressed as the percentage of the value obtained for control cells. Each point showed the mean and standard deviation

of six replicates.
Table 3. Effects of the addition of 55 rn~ mannitol to the culture medium on the cytotoxicity comoounds. as determined bv the neutral red assw Without Comoounds Sodium dodecyl sulfate Sodium pentobarbital Mercuric chloride Phenol Units PM rtM ItM
UM

of different

mannitol NR., 150*41 1013 k 27 18k 1 112+48

Mannitol NR, 23 & 8 1037 f 448 20 + 4. 6lk 35

(55 mM) NR, 133 1143 31 329 f k f f 53 28 6 66

NR, 22 f 2 796 f 49 8*5 l8f6

NR = neutral red The NR, and NR,, represent the concentration that resulted in a IO or 50% decrease, respectively, in absorption relative to the controls. Values are means + SD (n = 6) and those marked with asterisks differ significantly (Mann-Whitney test)
from the controls (**I < 0.01)

Comparatioe cytotoxicity efect of different compounds on hepatocytes

Table 3 shows the cytotoxicity of different compounds, with and without 55 IIIM mannitol in the complete medium, determined by NR assay. The concentration of the test compound that resulted in a 10 or 50% decrease in NR uptake (NR, and NR,, respectively) was determined for each compound studied. For the SDS, the addition of 55 mM mannitol did not significantly alter either the NR,, or the NR,,. However, both the NR,,) and the NR, for pentobarbital, mercuric chloride and phenol were significantly elevated in media containing mannitol.
DISCUSSION

By assessing the activity of mitochondrial dehydrogenase, the MTT test measures the resistance of hepatocytes to elevated osmotic pressure and our results showed that mitochondrial dehydrogenase

activity of cultured hepatocytes was not modified by polyol concentrations of up to 330 mM (620 mOsmol/ litre). Waymouth (1978) also demonstrated that an osmotic pressure of up to 620 mOsmol/litre remained tolerable for adult rat hepatocytes. The three polyols studied provoked a dose-related increase in uptake of NR even though the results of the MTT test were negative. This increase in NR uptake does not indicate a permanent cellular alteration as 72 hr after cessation of exposure to polyols, the cells capacity for NR uptake was very similar to that of the control. Taken together, these results indicate that there was no cytotoxicity. This phenomenon of increased NR uptake was due to the presence of the polyols themselves and not to variations in the osmotic pressure, because the cells that were exposed to mannitol, sorbitol or xylitol under iso-osmolar conditions showed similar effects. The NR assay is based on the fact that viable cells take up the dye and it has been shown that the dye

138

P. Olivier et u/

is almost exclusively sequestered in lysosomes (Allison and Young. 1964; De Duve et al.. 1974). Therefore, our results indicate that the increased uptake described is due to an increase in lysosomal content of the dye. De Duve et c-r/. ( 1974) reported mannitol to be lysosomototropic and used the term osmotic swelling to describe the phenomenon. The accumulation of polyol in the lysosome may cause an increase in the osmotic pressure within this organelle followed by the entry of water with subsequent swelling, and fusion with other small lysosomes to form vacuoles of increasing size. The enlargement of the lysosomes results in an increase in NR uptake which is visible on light microscopic examination and measurable by calorimetry. Several studies have demonstrated, in different cell types, the induction of cytoplasmic vacuoles by weakly basic substances (Ohkuma and Poole, I98 I : Rorig et al. 1987: Yang et (II.. 1965). Babich and Borenfreund (1992), in their study, attributed this vacuolization to lysosomal swelling. They observed an increase in NR uptake when both nicotine and phenyl propanolamine caused overloading of the lysosomes. Therefore this effect is not limited to compounds with a basic structure but it also occurs, in the case of polyols, whether they are metabolized or not. The NR test is used to determine cellular viability in the presence of xenobiotics, and thus an increase in NR uptake could mask a cytotoxic effect, For this reason we tested different substances in the presence of 55 mM mannitol. Compounds that act principally at the membrane level (SDS) showed cytotoxicity in the presence and absence of mannitol. On the other hand, compounds that act on cellular metabolism (mercuric chloride, phenol. pentobarbital) demonstrated that mannitol masked these cytotoxic effects. For example, our results show that in medium containing 55 mM mannitol. it was necessary to have three times the concentration of phenol to see the same NR,,, as in media without mannitol. This phenomenon of lysosomal swelling therefore leads to an underestimation of the cytotoxicity when the NR test is used. These results demonstrate that the NR test must be used with caution with compounds that cause lysosomal swelling. They confirm the necessity of using several, mechanistically differ-

ent, cytotoxicity markers xenobiotic cytotoxicity.

for

the

evaluation

of

REFERENCES

Allison A. C. and Young M. R. (1964) Uptake of dyes and

drugs by living cells in culture. L$e Sciences 3, 1407~1414. Allison R. G. (1979) Dietary sugars in health and disease. Federarion Proceedings 4, I -29. Babich H. and Borenfreund E. (1992) Cytotoxic and morphological effects of phenyl propanolamine. caffeine, nicotine and some of their metabolites studied in &a. Toxicology in Vitro 6, 493-502. Borenfreund E. and Puerner J. A. (1984) A simple quantitative procedure using monolayer cultures for cytotoxicity assays. Journal of Tissue Culrure Merhods 9, 7-9. Carmichael, J.. De Graff W. G., Gazdar A. F., Minna J. D. and Mitchell J. B. (1987) Evaluation of a tetrazoliumbased semiautimated calorimetric assay: assessment of chemiosensitivity testing. Cancer Research 47, 936-942. De Duve C., De Barsy T., Poole B., Trouet A., Tulkens P. and Van Hoof F. (1974) Lysosomotropic agents. Biochemicul Pharmacology 23, 2495-253 I. Guguen C.. Guillouzo A., Boisnard M., Le Cam A. and Bourel M. (1975) Etude ultra structurale de monocouches dhepatocytes de rat adulte cultivts en presence dhemissucinate dhydrocortisone. Biology Gastroenterology 8, 223-23 I. Mosmann T. (1983) Rapid calorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. Journal of Immunological Methods 65, 55-63. Ohkuma S. and Poole B. (1981) Cytoplasmic vacuolation of mouse peritoneal macrophages and the uptake into lysosomes of weakly basic substances. Journal afCe/l Biology 90, 656664. Rorig K. J., Ruben Z. and Anderson S. N. (1987) Structural determinants of cationic amphiphilic amines which induce clear cytoplasmic vacuoles in cultured cells. Proceedings afrhe Sociery for E.xperimenral Biology and Medicine 184, 165.-171. Seglen P. 0. (1976) Preparation of isolated rat liver cells. Methods in Cell Biology 13, 29983. Sicard P. J. (1982) Hydrogenated glucose syrups, sorbitol, mannitol and xylitol. Nufririre Sweeteners 8, 145170. Voedingsraad (1987) The energy value of sugar alcohols. Nutrition Council Advisor Board of the Minister of Welfare, Health and Cultural Affairs and to the Minister of Agriculture and Fisheries, Regarding Nutrition and Food Supply. Den Haag Voedingsraad 1455170. Waymouth C. (I 978) Osmolarity of mammalian blood and of media for culture mammalian cells. In Vitro 6. 1099127. Yang W. C. T.. Strasser F. F. and Pomerat C. M. (1965) Mechanism of drug-induced vacuolisation in tissue culture. E.rperimental Cell Research 38, 495506.