- Email: [email protected]
Expression of stress proteins in cultured HT29 human cell-line; a model for studying environmental aggression
Vol. 27, No. 4, Pp. 385-391, 1995 Copyright 0 1995 Elsevier ScienceLtd Printed in Great Britain. All rights reserved 1357-2725/95$9.50+ 0.00
Int. J. Biochem. Cell Biol.
OU20-711X(94)OOU69-7
Expression of Stress Proteins in Cultured HT29 Human Cell-line; a Model for Studying Environmental Aggression FLORENCE DELMAS, VERONIQUE JEAN-CLAUDE MURAT” INSERM
U317, Bat. L3, CHU-Rangueil,
TROCHERIS,
31054 Toulouse Cedex, France
The current study was undertaken to investigate the expression of stress proteins (HSP) in cultured human HT29 cells submitted to stressingevents under in vitro conditions. Heat shocks (45’C, for 15-60 mhr) or cold shocks(+ l°C for 4 hr) were found to modify cell growth (growth curves)and to enhanceHSP expression.In most cases,changesin HSP expressionare much more pronounced than changes in cell growth. Exposure to 8% ethanol for 15 min resulted in both growth inhibition and HSP overexpression. Propanol-1 was found to be more toxic since 5% concentration given for 15mh1 stops cell growth. 2.5% propanol-1 for 15min induces a slight reduction of cell growth but a clear-cut overexpression of stress proteins. We conclude that expression of stress proteins, especially those of the HSP68/70 family, coustitutes a more sensitiveresponse than changes in growth rate in case of external aggression. Tbis could make our model an interesting biological sensor to environmental physical or chemical pollutants. Keywords: Environmental aggression Stress proteins Heat shock Alcoholic shock HT29 cell-line Int. J. Biochem. Cell Biol. (1995) 27, 385-391
INTRODUCTION
Stress proteins, also named Heat Shock Proteins (HSP) since they were first observed in cells of Drosophila after exposure to hyperthermia (Ritossa, 1962), consist of several proteins, highly conserved throughout evolution, which appear ubiquitously after heat shock or other stressful events (for a review, see Welch, 1992). They are usually nomenclatured according to their molecular weight and the most commonly described are the HSP60, HSP70 and HSP90 families. It is now well documented that when cells experience adverse changes in their environment, they react by activating the genes coding for stress proteins (Lindquist, 1986). An important role of these proteins may be to provide the cells with a mechanism to handle increased *To whom all correspondence should be addressed. Received 25 May 1994; accepted 15 November 1994.
amounts of abnormally folded proteins in stressing or non-stressing conditions (Welch, 1992). It has been postulated that the accumulation of nonfolded polypeptides could be the signal for increased expression of HSP genes (Baler et al., 1992). Besides the case of hyperthermia, cellular stress evidenced by the expression of HSP has been observed to occur in several biological models after exposure to pollutants (Faber et al., 1993; Hatayama et al., 1993), ischemia (Novak, 1993), sodium arsenite or ethanol (Li, 1983), amino acids analogs (Hightower, 1980), heavy metals (Goering et al., 1993) and ionizing radiations (Sierra-Rivera et al., 1993). The wide variety of stressing events that stimulate such a response are thought to share the common property of either damaging cellular proteins or causing cells to synthesize abnormal proteins. Our aim was to investigate the stress response of cultured cells from human origin in order to 385
Florence
386
propose such a biological model as an in vitro sensor for environmental adversity. This system could have the advantage to react to low levels of aggression, to provide immediate data (within hours) in contrast to the long lasting exposition needed by most of the biological models (living animals or plants) used in field investigations and to offer the possibility to study a variety of suspected environmental aggressors (microwaves, low frequency electromagnetic fields, etc.) at the cellular and molecular levels in the laboratory. The HT29 cells were chosen to start our work because: (1) they are human cells deriving from gut epithelium, (2) they constitute a convenient model for long-lasting cultures and (3) several aspects of their metabolism and ultrastructure have been investigated. This first step of our research project was to verify the order of magnitude and characteristics of stress responses in these cells as a function of the intensity and/or nature of the aggression. MATERIALS
AND
METHODS
Cell culture
The HT29 cell line has been established in permanent culture from a human colon adenocarcinoma by Dr J. Fogh (Sloan Kettering Institute for Cancer Research, Rye, N.Y.) (Fogh er al., 1977). For routine culture, cells were seeded at a density of 3.4 x lo4 cells per sq cm in plastic flasks or Petri dishes and grown at 37°C under air/CO, (19: 1) atmosphere, in Dulbecco’s modified Eagle medium (DMEM), supplemented with 5% (v/v) fetal calf serum (FCS). Medium was renewed daily. Under such conditions, confluence is reached within 8 days. Estimation of growth rate
Growth curves were established by measuring the total protein content of the Petri dishes at given times, using the method of Bradford (1976). Protocols for stress induction
For evaluating their incidence on cell growth, heat shocks (45°C for 15, 30 or 60 min) were given once at day 3 of the culture by placing the dishes in a temperature regulated water bath set at 45°C. The dishes were then replaced in standard conditions for further growth.
Delmas
et crl
Cold shock, given at day 4, consisted 01 placing the dishes in a cold room at 1,‘C for 4 hr, then putting the dishes back into the incubator. These experiments were carried out in a closed flask, in order to avoid any dramatic change in atmosphere composition during exposure to cold. Acute exposure to alcohol was achieved at day 4 by adding the appropriate quantities of alcohol into the culture medium for 15, 30 or 60 min, then changing the medium by fresh standard DMEM and replacing the cell under standard conditions of culture. When given chronically, alcohol was merely added into the culture medium at the appropriate concentrations from the beginning to the end of the culture. Experiments on HSP expression were performed with post-confluent cell-layers (unless otherwise specified) grown at 37°C in Petri dishes 35 mm in diameter. Standard stress (heat shock) was achieved by putting the dishes in a water bath at 45°C for 30 min. Medium was then sucked out and the cell layer was rapidly rinsed with methionine-free DMEM containing 2% fetal calf serum. Labeling qf’ stress proteins
For labeling the de novo synthesized proteins, cell layer was immediately given 1 ml of DMEM containing 30 PCi of (35S)-methionine and allowed to stand at 37°C for 1,3 or 6 hr. Medium was then removed and layer was rinsed twice with 2 ml saline (9 g/l NaCl) and kept deep frozen at -80°C until used for electrophoresis. For protein separation, a frozen layer was scraped into 300 ~1 of 50 mM Tris buffer (pH 7.5). A 20~1 aliquot containing about 100 pg of proteins was mixed with 20~1 of loading buffer, boiled for 10 min and submitted to gel electrophoresis (SDS-PAGE; 10% acrylamide) analysis. Proteins were run on the gel together with molecular weight markers at 18 mA for about 6 hr. Gel was then colored for 30 min with Coomassie blue and rinsed in 5% methanol-10% acetic acid mixture and permanently shaken overnight to remove the excess dye. The gel was further rinsed in a 1% glycerol-20% methanol mixture for 10 min, vacuum-dried at 70°C for 2 hr and placed into a cassette with an X-ray film for two days at -80°C for revealing the 35S-radiolabeled proteins. In one experiment designed to ascertain the identity of HSP70 stress proteins, the gel was
Stress proteins in HT29 cell-line
blotted on nitrocellulose filter (Western blotting) for immunological identification using the corresponding commercial antibodies. Quantification of the stress proteins expression was achieved for each case by computer-assisted laser densitometric scanning of the autoradiograph of the electrophoretically Therefore, separated ‘*S-labeled proteins. changes in synthesis are expressed as the Area Under the Curve (AUC) for protein band intensity after standardizing gel lane background intensity at zero units.
5 Control
1;. 2
Growth curves
In Fig. 1 the effects are reported of three different heat shocks given at day 3 on the growth of HT29 cells in culture. It appears from statistical analysis that the 15 min shock did not influence cell growth, whereas 30 and 60min shocks seemed slightly effective in slowing down the growth rate, although not significantly at P < 0.05.
Figure 2 dealing with the effect of one cold shock given at day 4 shows that the growth rate is decreased when cells are submitted to 1“C for 4 hr (P < 0.01). No significant effect is obtained when cold shock was shorter in duration, i.e. 1 hr (not shown). Effects of chronic exposure of HT29 cells to ethanol, in concentrations ranging from 0.25 to 5%, are reported in Fig. 3. No effect was seen at 0.25 and 0.5% alcohol. By contrast, 1.25, 2.5
6 Days
.
, 6
.
, a
.
, 10
.
I 12
Days
and 5% ethanol induced a 30, 70 and 99% growth inhibition, respectively. Acute exposure was achieved by adding either 5 or 8% ethanol at day 4 of the culture for durations ranging from 15 to 60 min. No convincing effect on growth rate was observed at the lowest concentration (not shown). As seen in Fig. 4, 8% ethanol, given for only 15 min, was found to strongly inhibit cell growth (-40% at day 8). Under our experimental conditions, concentrations over 8% were found to be lethal. Similar experiments were performed using propanol-1 instead of ethanol. Chronic exposure to concentrations of 0.5% or above were found to fully inhibit cell growth (not shown). Study on acute exposure to propanol is
1-
0,25%
0
5
-
4
, 4
Fig. 2. Growth curves of HT29 cells submitted to a cold shock at 1°C for 4 hr at day 4 of the culture. Control cells were not submitted to the cold shock. Cell growth is expressed as mg total protein per dish. Results are the mean f. SEM of 4 separate experiments.
RESULTS
3
387
7
9
Fig. 1. Growth rate of HT29 cells submitted to a heat shock at 45°C for 1530 or 60 min, at day 3 of the culture. Control cells were not submitted to a heat shock. Cell growth is expressed as mg total protein per dish. Results are the mean + SEM of 4 separate experiments.
Control
10 Days
15
20
Fig. 3. Growth curves of HT29 cells submitted to chronic ethanol exposure. Cells were grown in the presence of concentrations ranging from 0.25 to 5% ethanol. Control cells were not given ethanol. Cell growth is expressed as mg total protein per dish. Results are the mean + SEM of 4 separate experiments.
Florence
-
Et-OH
Delmas
CI a/
8%
90 kD 2
3
4
5
6
7
8
70 kD 68 kD
9
Days Fig. 4. Growth curves of HT29 cells submitted to acute exposure to 8% ethanol for 15 min at day 4 of the culture. Control cells were not given ethanol. Cell growth is expressed as mg total protein per dish. Results are the mean f SEM of 4 separate experiments.
reported in Fig. 5: propanol-1 at 2.5% for 15 or 30 min fails to significantly reduce cell growth; by contrast, higher concentration (5%) exerts a strong inhibitory effect. Expression of stress proteins Under our experimental conditions, stress proteins were identified by their position on the molecular weight scale (Fig. 6). It is shown in Fig. 6 that four stress proteins are clearly overexpressed after 6 hr. We have verified that in all cases developed in the present work, the optimal labeling of HSP did occur 6 hr after the aggressive treatment. In Figs 7 and 8, the levels of expression of HSP after different aggressive treatments are summarized:
0 2
4
a
6
IO
12
Days Fig. 5. Growth curves of HT29 cells submitted to acute exposure to propanol-1. At day 4 of the culture, cells were given propanol-1 at concentrations ranging from 2.5 to 5%. Control cells were not given propanol. Cell growth is expressed as mg total protein per dish. Results are the mean & SEM of 4 separate experiments.
12
3
4
5
Fig. 6. Autoradiography of electrophoretically separated 35S-labeled proteins from HT29 cells submitted to heat shock under various conditions: control (lane l), 45°C for 30 min (lanes 2 and 5), 45°C for 45 min (lane 3), 45°C for 60 min (lane 4). Immediately after heat shocks, newly synthesized proteins were labeled for 6 hr with 35S-methionine. Experiments were carried out on confluent cells (lanes 1, 2, 3, 4) or exponentially growing cells (lane 5).
Heat shock. Unlike what is reported by Kim et al. (1993), who found an increase of HSP68 and HSP90 expression following a 45°C for 15 min heat shock, exposure to 45°C for 15 min failed under our experimental conditions to induce any change in HSP expression; cell growth is not affected either (see comments of Fig. 1). After exposure to 45°C for 30 min, HSPI 10, HSP90, HSP70 and HSP68 were overexpressed by 3.3-, 3.7-, 8.0- and 9.5-fold, respectively. After 45 min at 45°C results were that only HSP70 and HSP68 were overexpressed by 1.9and 4.4-fold factor, respectively. If the thermic treatment was prolonged for 1 hr, the synthesis of all proteins was practically suspended. As a consequence, the positive dontrols (exposure to standard aggression) in further experiments related to HT29 cells consisted of a 45”C, 30 min heat shock. Surprisingly, we have observed that HT29 cells during the exponential phase of the culture
Stress proteins in HT29 cell-line
behave differently to heat shock than the stationary cells. The rate of methionine incorporation was much reduced and, after 30 min at 45°C the HSP90, HSP70 and HSP68 were overexpressed by only 1.5-, 2.8- and 3.9-fold, respectively. Cold shock. If 4 hr at 1°C were needed to modify the growth rate (see above), 1 hr only of exposure at 1°C resulted in significant overexpression of HSP70 and HSP68 by 2.8- and 3.4-fold, respectively (not shown). Exposure to ethanol. Exposure of postconfluent cells to 5% ethanol for 48 hr (as an attempt to represent a chronic treatment) was found not to change the HSP expression under our experimental conditions. As seen in Fig. 3, true chronic exposure at this concentration of alcohol did not allow cell culture to progress, which rendered aimless the search for HSP. As seen in Fig. 8, when ethanol was given once (acute) at 5%, or below, no effect on stress proteins was observed. However, 8% ethanol, given for 15 min, induced an overexpression of HSP90, HSP70, and HSP68 by 2.5-, 3.1- and 4.8-fold, respectively. During the exponential phase of the culture (day 3), HT29 cells behave differently to ethanol acute exposure than do the stationary cells; with exponentially growing cells, the rate of proteins
2
:: 800 5 5 cn ; 600 5 : 5c B s g B
400 200 0 Cal
A Heat
0 shock
C
D
(45°C)
Fig. 7. Densitometric analysis of changes in de nouo synthesis of major stress and constitutive proteins in HT29 cells submitted to Heat Shock under different conditions: Con =controls; A =confluent cells heated at 45°C for 30min; B = confluent cells heated at 45°C for 45 min; C = confluent cells heated at 45°C for 60min; D = exponentially growing cells heated at 45°C for 30 min. Relative synthesis of HSP70, HSP90 and HSPl 10 proteins was determined for each case by computer-assisted laser densitometric scanning of the autoradiograph of the electrophoretically separated ‘Wabeled proteins. Changes in synthesis are expressed as the area under the curve (AUC) for protein band intensity after standardizing gel lane background intensity at zero units.
g
400
t 4 3i
300
; ‘i; p‘ : g t
389
200
100
g ; d
0 ConA
0 CD E Alcohol treatment
FG
Fig. 8. Densitometric analysis of changes in de novo synthesis of major stressand constitutive proteins in HT29 cells submitted to alcoholic shock under different conditions: Con = controls; A = confluent cells, 5% ethanol for 60 min; B = confluent cells, 8% ethanol for 15 min; C = confluent cells, 8% ethanol for 30 mitt; D = exponentially growing cells, 5% ethanol for 15 mitt; E = exponentially growing cells, 8% ethanol for 15 min; F = confluent cells, 2.5% propanol-I for 15 min; G = confluent cells, 2.5% propanol1 for 30min. Relative synthesis of HSP68, HSP70 and HSP90 proteins was determined for each case by computerassisted laser densitometric scanning of the autoradiograph of the electrophoretically separated 35S-labeled proteins. Changes in synthesis are expressed as the area under the curve (AUC) for protein band intensity after standardizing gel lane background intensity at zero units.
synthesis (methionine incorporation) was found to be preserved and, after 15 min in the presence of 8% ethanol, the HSP90 and HSP68 were overexpressed by 1.5- and 4.4-fold, respectively. Exposure to propanol- 1. No chronic exposure to propanol allowed cell growth or life processes, underscoring the high toxicity of this alcohol when compared to ethanol. When given once for 15 min at 2.5%, propanol induced an overexpression of HSP70 and HSP68 by 1.4- and 2.6-fold, respectively. If given for 30min at the same concentration, HSP70 and HSP68 were overexpressed by 1.4and 3.3-fold, respectively. At 5% for 15 min, propanol is found to totally inhibit the protein synthesis, which corroborates our results related to cell growth (see above). DISCUSSION
From the present results, we state that our biological model, i.e. HT29 cells, reacts to aggressive treatments by changing its growth rate and by producing stress proteins. However, these two parameters don’t appear to be strictly linked. In fact, it depends on the nature and on the level of aggression.
390
Florence
Delmas
If we compare the effects of stressing situations on either growth rate or HSP expression, it appears that weakest heat shock showing effectiveness (i.e. 45°C for 30 min) under our experimental conditions is found to affect both growth rate and HSP expression. However, the effect on HSP expression is much more evident than that on growth rate, for which a skilful statistical analysis is needed in order to provide reliable data, not to mention the fact that the establishment of a growth-curve over days is per se time-consuming. Excessive heat shock (i.e. 45°C for 60 min) does not modify cell growth so much, whereas it produces a clear-cut transient inhibition (over 6 hr) of protein synthesis. When considering the effect of cold shock, it appears that at least 4 hr at 1°C are needed to get a visible effect on growth curve, whereas 1 hr at 1°C is enough to give an obvious increase in HSP expression. Acute treatment with 5% ethanol for 60min modifies neither growth rate nor HSP expression. 8% ethanol for 15 min is required to induce a clear effect on both growth rate and HSP expression. If we consider the effects of propanol-1, it appears that the 2.5% for 15 or 30min acute exposure does not convincingly affect cell growth. By contrast, the HSP68 is clearly overexpressed under the same conditions. Differences in the sensitivity of the cells to express stress proteins as a function of the phases of culture could be due to the difference in degree of differentiation; it is known that HT29 cells are mostly undifferentiated when they divide actively, whereas some indices of differentiation appear in post confluent cells (Zweibaum et al., 1985). Paying attention to the different HSP families expressed in HT29 cells after aggression leads to the conclusion that some specificity exists in the overexpression of stress proteins. For instance, overexpression of HSPllO family appears only after heat shock. It is also clear that HSP68 is quantitatively the most overexpressed among HSP in our cellular model. We conclude that study of HSP expression gives faster and more obvious and clear-cut results than mere observation of growth curve for evaluation of the stressing power of an environmental aggression. This advantage, together with the development of an easy-to-use method to detect the HSP induction, could make our model a suitable biological sensor for environmental aggressions.
et al.
Acknowledgements-We thank Dr Siddharta Gautama for his wise contribution to the experimental work and MS Agnes Y&he for her kind help in preparing the manuscript. This research was made possible thanks to the fellowship from the ADEME to Florence Delmas.
REFERENCES Baler R., Welch W. J. and Voellmy R. (1992) Heat shock genes regulation by nascent polypeptides and denatured proteins: HSP70 as a potential autoregulatory factor. .I. Cell Biol. 117, 1151-1159. Bradford M. M. (1976) A rapid sensitive method for the quantitation of protein utilizing the principle of protein dye binding. Analyt. Biorhem. 72, 248-254. Faber F., Egli T. and Harder W. (1993) Transient repression of the synthesis of OmpF and aspartate transcarbamoylase in E. coli K12 as a response to pollutant stress. Microbial. Lett. 111, 189-196. Fischbach M., Sabbioni E. and Bromley P. (1993) Induction of the human growth hormone gene placed under human HSP70 promoter control in mouse cells: a quantitative indicator of metal toxicity. Cell Biol. Toxicol. 9, 177-188. Fogh J.. Fogh J. M. and Orfeo T. (1977) Hundred and twenty seven cultured human tumor cell-lines producing tumors in nude mice. J. Nat. Cancer Inst. 59, 221-226. Goering P. L., Kish C. L. and Fisher B. R. (1993) Stress protein synthesis induced by cadmium-cysteine in rat kidney. Toxicology 85, 25-39. Hatayama T.. Asai Y., Wakatsuki T., Kitamura T. and Imahara H. (1993) Regulation of HSP70 synthesis induced by cupric sulfate and zinc sulfate in thermotolerant HeLa cells. J. Biochem. 114, 592Z597. Hendrick J. P. and Hart1 F. U. (1993) Molecular chaperone functions of heat-shock proteins. Ann. Rev. Biochem. 62, 3499384. Hightower L. E. (1980) Cultured animal cells exposed to amino-acid analogs or puromycin rapidly synthesize several polypeptides. J. Cell Physiol. 102, 407427. Holland D. B., Roberts S. G., Wood E. J. and Cunliffe W. J. (1993) Cold shock induces the synthesis of stress proteins in human keratinocytes. J. Invest. llermatol. 101, 1966199. Kim S. H., Kim J. H., Erdos G. and Lee Y. J. (1993) Effect of staurosporine on suppression of heat shock gene expression and thermotolerance development in HT29 cells. Biochem. biophys. Res. Comm. 193, 759-763. Li G. C. (1983) Induction of thermotolerance and enhanced heat shock protein synthesis in Chinese hamster fibroblasts by sodium arsenite and by ethanol. J. CeN. Physiol. 115, 116122. Lindquist S. (1986) The heat shock response. Ann. Rev. Biochem. 55, 1151-1191. Novak T. S. (1993) Synthesis of heat shock/stress proteins during cellular injury. Ann. N. Y. Acad. Sci. 679, 142-156. Peiham H. R. B. (1989) Heat shock and the sorting of luminal ER proteins. EMBO J. 8, 3171-3176. Ritossa F. (1962) A new puffing pattern induced by a temperature shock and DNP in Drosophila. Experienfia 18, 571-573. Sierra-Rivera E., Voorhees G. J. and Freeman M. L. (1993) Gamma irradiation increases HSP70 in Chinese hamster ovary cells. Radiation Res. 135, 4045.
Stress proteins in HT29 cell-line Welch W. J. (1992) Mammalian stress response: physiology, structure/function of stress proteins, implications for medicine and disease. Physiol. Rev. 1063-1081. Zweibaum A., Pinto M., Chevalier G., Dussaulx
cell and 72,
E.,
391
Triadou N., Lacroix B., Haffen K. and Rousset M. (1985) Enterocytic differentiation of a subpopulation of the human colon tumor cell-line HT29 selected for growth in a sugar-free medium and inhibition by glucose. J. Cell, Physiol. 122, 21-28.
Int. J. Biochem. Cell Biol.
OU20-711X(94)OOU69-7
Expression of Stress Proteins in Cultured HT29 Human Cell-line; a Model for Studying Environmental Aggression FLORENCE DELMAS, VERONIQUE JEAN-CLAUDE MURAT” INSERM
U317, Bat. L3, CHU-Rangueil,
TROCHERIS,
31054 Toulouse Cedex, France
The current study was undertaken to investigate the expression of stress proteins (HSP) in cultured human HT29 cells submitted to stressingevents under in vitro conditions. Heat shocks (45’C, for 15-60 mhr) or cold shocks(+ l°C for 4 hr) were found to modify cell growth (growth curves)and to enhanceHSP expression.In most cases,changesin HSP expressionare much more pronounced than changes in cell growth. Exposure to 8% ethanol for 15 min resulted in both growth inhibition and HSP overexpression. Propanol-1 was found to be more toxic since 5% concentration given for 15mh1 stops cell growth. 2.5% propanol-1 for 15min induces a slight reduction of cell growth but a clear-cut overexpression of stress proteins. We conclude that expression of stress proteins, especially those of the HSP68/70 family, coustitutes a more sensitiveresponse than changes in growth rate in case of external aggression. Tbis could make our model an interesting biological sensor to environmental physical or chemical pollutants. Keywords: Environmental aggression Stress proteins Heat shock Alcoholic shock HT29 cell-line Int. J. Biochem. Cell Biol. (1995) 27, 385-391
INTRODUCTION
Stress proteins, also named Heat Shock Proteins (HSP) since they were first observed in cells of Drosophila after exposure to hyperthermia (Ritossa, 1962), consist of several proteins, highly conserved throughout evolution, which appear ubiquitously after heat shock or other stressful events (for a review, see Welch, 1992). They are usually nomenclatured according to their molecular weight and the most commonly described are the HSP60, HSP70 and HSP90 families. It is now well documented that when cells experience adverse changes in their environment, they react by activating the genes coding for stress proteins (Lindquist, 1986). An important role of these proteins may be to provide the cells with a mechanism to handle increased *To whom all correspondence should be addressed. Received 25 May 1994; accepted 15 November 1994.
amounts of abnormally folded proteins in stressing or non-stressing conditions (Welch, 1992). It has been postulated that the accumulation of nonfolded polypeptides could be the signal for increased expression of HSP genes (Baler et al., 1992). Besides the case of hyperthermia, cellular stress evidenced by the expression of HSP has been observed to occur in several biological models after exposure to pollutants (Faber et al., 1993; Hatayama et al., 1993), ischemia (Novak, 1993), sodium arsenite or ethanol (Li, 1983), amino acids analogs (Hightower, 1980), heavy metals (Goering et al., 1993) and ionizing radiations (Sierra-Rivera et al., 1993). The wide variety of stressing events that stimulate such a response are thought to share the common property of either damaging cellular proteins or causing cells to synthesize abnormal proteins. Our aim was to investigate the stress response of cultured cells from human origin in order to 385
Florence
386
propose such a biological model as an in vitro sensor for environmental adversity. This system could have the advantage to react to low levels of aggression, to provide immediate data (within hours) in contrast to the long lasting exposition needed by most of the biological models (living animals or plants) used in field investigations and to offer the possibility to study a variety of suspected environmental aggressors (microwaves, low frequency electromagnetic fields, etc.) at the cellular and molecular levels in the laboratory. The HT29 cells were chosen to start our work because: (1) they are human cells deriving from gut epithelium, (2) they constitute a convenient model for long-lasting cultures and (3) several aspects of their metabolism and ultrastructure have been investigated. This first step of our research project was to verify the order of magnitude and characteristics of stress responses in these cells as a function of the intensity and/or nature of the aggression. MATERIALS
AND
METHODS
Cell culture
The HT29 cell line has been established in permanent culture from a human colon adenocarcinoma by Dr J. Fogh (Sloan Kettering Institute for Cancer Research, Rye, N.Y.) (Fogh er al., 1977). For routine culture, cells were seeded at a density of 3.4 x lo4 cells per sq cm in plastic flasks or Petri dishes and grown at 37°C under air/CO, (19: 1) atmosphere, in Dulbecco’s modified Eagle medium (DMEM), supplemented with 5% (v/v) fetal calf serum (FCS). Medium was renewed daily. Under such conditions, confluence is reached within 8 days. Estimation of growth rate
Growth curves were established by measuring the total protein content of the Petri dishes at given times, using the method of Bradford (1976). Protocols for stress induction
For evaluating their incidence on cell growth, heat shocks (45°C for 15, 30 or 60 min) were given once at day 3 of the culture by placing the dishes in a temperature regulated water bath set at 45°C. The dishes were then replaced in standard conditions for further growth.
Delmas
et crl
Cold shock, given at day 4, consisted 01 placing the dishes in a cold room at 1,‘C for 4 hr, then putting the dishes back into the incubator. These experiments were carried out in a closed flask, in order to avoid any dramatic change in atmosphere composition during exposure to cold. Acute exposure to alcohol was achieved at day 4 by adding the appropriate quantities of alcohol into the culture medium for 15, 30 or 60 min, then changing the medium by fresh standard DMEM and replacing the cell under standard conditions of culture. When given chronically, alcohol was merely added into the culture medium at the appropriate concentrations from the beginning to the end of the culture. Experiments on HSP expression were performed with post-confluent cell-layers (unless otherwise specified) grown at 37°C in Petri dishes 35 mm in diameter. Standard stress (heat shock) was achieved by putting the dishes in a water bath at 45°C for 30 min. Medium was then sucked out and the cell layer was rapidly rinsed with methionine-free DMEM containing 2% fetal calf serum. Labeling qf’ stress proteins
For labeling the de novo synthesized proteins, cell layer was immediately given 1 ml of DMEM containing 30 PCi of (35S)-methionine and allowed to stand at 37°C for 1,3 or 6 hr. Medium was then removed and layer was rinsed twice with 2 ml saline (9 g/l NaCl) and kept deep frozen at -80°C until used for electrophoresis. For protein separation, a frozen layer was scraped into 300 ~1 of 50 mM Tris buffer (pH 7.5). A 20~1 aliquot containing about 100 pg of proteins was mixed with 20~1 of loading buffer, boiled for 10 min and submitted to gel electrophoresis (SDS-PAGE; 10% acrylamide) analysis. Proteins were run on the gel together with molecular weight markers at 18 mA for about 6 hr. Gel was then colored for 30 min with Coomassie blue and rinsed in 5% methanol-10% acetic acid mixture and permanently shaken overnight to remove the excess dye. The gel was further rinsed in a 1% glycerol-20% methanol mixture for 10 min, vacuum-dried at 70°C for 2 hr and placed into a cassette with an X-ray film for two days at -80°C for revealing the 35S-radiolabeled proteins. In one experiment designed to ascertain the identity of HSP70 stress proteins, the gel was
Stress proteins in HT29 cell-line
blotted on nitrocellulose filter (Western blotting) for immunological identification using the corresponding commercial antibodies. Quantification of the stress proteins expression was achieved for each case by computer-assisted laser densitometric scanning of the autoradiograph of the electrophoretically Therefore, separated ‘*S-labeled proteins. changes in synthesis are expressed as the Area Under the Curve (AUC) for protein band intensity after standardizing gel lane background intensity at zero units.
5 Control
1;. 2
Growth curves
In Fig. 1 the effects are reported of three different heat shocks given at day 3 on the growth of HT29 cells in culture. It appears from statistical analysis that the 15 min shock did not influence cell growth, whereas 30 and 60min shocks seemed slightly effective in slowing down the growth rate, although not significantly at P < 0.05.
Figure 2 dealing with the effect of one cold shock given at day 4 shows that the growth rate is decreased when cells are submitted to 1“C for 4 hr (P < 0.01). No significant effect is obtained when cold shock was shorter in duration, i.e. 1 hr (not shown). Effects of chronic exposure of HT29 cells to ethanol, in concentrations ranging from 0.25 to 5%, are reported in Fig. 3. No effect was seen at 0.25 and 0.5% alcohol. By contrast, 1.25, 2.5
6 Days
.
, 6
.
, a
.
, 10
.
I 12
Days
and 5% ethanol induced a 30, 70 and 99% growth inhibition, respectively. Acute exposure was achieved by adding either 5 or 8% ethanol at day 4 of the culture for durations ranging from 15 to 60 min. No convincing effect on growth rate was observed at the lowest concentration (not shown). As seen in Fig. 4, 8% ethanol, given for only 15 min, was found to strongly inhibit cell growth (-40% at day 8). Under our experimental conditions, concentrations over 8% were found to be lethal. Similar experiments were performed using propanol-1 instead of ethanol. Chronic exposure to concentrations of 0.5% or above were found to fully inhibit cell growth (not shown). Study on acute exposure to propanol is
1-
0,25%
0
5
-
4
, 4
Fig. 2. Growth curves of HT29 cells submitted to a cold shock at 1°C for 4 hr at day 4 of the culture. Control cells were not submitted to the cold shock. Cell growth is expressed as mg total protein per dish. Results are the mean f. SEM of 4 separate experiments.
RESULTS
3
387
7
9
Fig. 1. Growth rate of HT29 cells submitted to a heat shock at 45°C for 1530 or 60 min, at day 3 of the culture. Control cells were not submitted to a heat shock. Cell growth is expressed as mg total protein per dish. Results are the mean + SEM of 4 separate experiments.
Control
10 Days
15
20
Fig. 3. Growth curves of HT29 cells submitted to chronic ethanol exposure. Cells were grown in the presence of concentrations ranging from 0.25 to 5% ethanol. Control cells were not given ethanol. Cell growth is expressed as mg total protein per dish. Results are the mean + SEM of 4 separate experiments.
Florence
-
Et-OH
Delmas
CI a/
8%
90 kD 2
3
4
5
6
7
8
70 kD 68 kD
9
Days Fig. 4. Growth curves of HT29 cells submitted to acute exposure to 8% ethanol for 15 min at day 4 of the culture. Control cells were not given ethanol. Cell growth is expressed as mg total protein per dish. Results are the mean f SEM of 4 separate experiments.
reported in Fig. 5: propanol-1 at 2.5% for 15 or 30 min fails to significantly reduce cell growth; by contrast, higher concentration (5%) exerts a strong inhibitory effect. Expression of stress proteins Under our experimental conditions, stress proteins were identified by their position on the molecular weight scale (Fig. 6). It is shown in Fig. 6 that four stress proteins are clearly overexpressed after 6 hr. We have verified that in all cases developed in the present work, the optimal labeling of HSP did occur 6 hr after the aggressive treatment. In Figs 7 and 8, the levels of expression of HSP after different aggressive treatments are summarized:
0 2
4
a
6
IO
12
Days Fig. 5. Growth curves of HT29 cells submitted to acute exposure to propanol-1. At day 4 of the culture, cells were given propanol-1 at concentrations ranging from 2.5 to 5%. Control cells were not given propanol. Cell growth is expressed as mg total protein per dish. Results are the mean & SEM of 4 separate experiments.
12
3
4
5
Fig. 6. Autoradiography of electrophoretically separated 35S-labeled proteins from HT29 cells submitted to heat shock under various conditions: control (lane l), 45°C for 30 min (lanes 2 and 5), 45°C for 45 min (lane 3), 45°C for 60 min (lane 4). Immediately after heat shocks, newly synthesized proteins were labeled for 6 hr with 35S-methionine. Experiments were carried out on confluent cells (lanes 1, 2, 3, 4) or exponentially growing cells (lane 5).
Heat shock. Unlike what is reported by Kim et al. (1993), who found an increase of HSP68 and HSP90 expression following a 45°C for 15 min heat shock, exposure to 45°C for 15 min failed under our experimental conditions to induce any change in HSP expression; cell growth is not affected either (see comments of Fig. 1). After exposure to 45°C for 30 min, HSPI 10, HSP90, HSP70 and HSP68 were overexpressed by 3.3-, 3.7-, 8.0- and 9.5-fold, respectively. After 45 min at 45°C results were that only HSP70 and HSP68 were overexpressed by 1.9and 4.4-fold factor, respectively. If the thermic treatment was prolonged for 1 hr, the synthesis of all proteins was practically suspended. As a consequence, the positive dontrols (exposure to standard aggression) in further experiments related to HT29 cells consisted of a 45”C, 30 min heat shock. Surprisingly, we have observed that HT29 cells during the exponential phase of the culture
Stress proteins in HT29 cell-line
behave differently to heat shock than the stationary cells. The rate of methionine incorporation was much reduced and, after 30 min at 45°C the HSP90, HSP70 and HSP68 were overexpressed by only 1.5-, 2.8- and 3.9-fold, respectively. Cold shock. If 4 hr at 1°C were needed to modify the growth rate (see above), 1 hr only of exposure at 1°C resulted in significant overexpression of HSP70 and HSP68 by 2.8- and 3.4-fold, respectively (not shown). Exposure to ethanol. Exposure of postconfluent cells to 5% ethanol for 48 hr (as an attempt to represent a chronic treatment) was found not to change the HSP expression under our experimental conditions. As seen in Fig. 3, true chronic exposure at this concentration of alcohol did not allow cell culture to progress, which rendered aimless the search for HSP. As seen in Fig. 8, when ethanol was given once (acute) at 5%, or below, no effect on stress proteins was observed. However, 8% ethanol, given for 15 min, induced an overexpression of HSP90, HSP70, and HSP68 by 2.5-, 3.1- and 4.8-fold, respectively. During the exponential phase of the culture (day 3), HT29 cells behave differently to ethanol acute exposure than do the stationary cells; with exponentially growing cells, the rate of proteins
2
:: 800 5 5 cn ; 600 5 : 5c B s g B
400 200 0 Cal
A Heat
0 shock
C
D
(45°C)
Fig. 7. Densitometric analysis of changes in de nouo synthesis of major stress and constitutive proteins in HT29 cells submitted to Heat Shock under different conditions: Con =controls; A =confluent cells heated at 45°C for 30min; B = confluent cells heated at 45°C for 45 min; C = confluent cells heated at 45°C for 60min; D = exponentially growing cells heated at 45°C for 30 min. Relative synthesis of HSP70, HSP90 and HSPl 10 proteins was determined for each case by computer-assisted laser densitometric scanning of the autoradiograph of the electrophoretically separated ‘Wabeled proteins. Changes in synthesis are expressed as the area under the curve (AUC) for protein band intensity after standardizing gel lane background intensity at zero units.
g
400
t 4 3i
300
; ‘i; p‘ : g t
389
200
100
g ; d
0 ConA
0 CD E Alcohol treatment
FG
Fig. 8. Densitometric analysis of changes in de novo synthesis of major stressand constitutive proteins in HT29 cells submitted to alcoholic shock under different conditions: Con = controls; A = confluent cells, 5% ethanol for 60 min; B = confluent cells, 8% ethanol for 15 min; C = confluent cells, 8% ethanol for 30 mitt; D = exponentially growing cells, 5% ethanol for 15 mitt; E = exponentially growing cells, 8% ethanol for 15 min; F = confluent cells, 2.5% propanol-I for 15 min; G = confluent cells, 2.5% propanol1 for 30min. Relative synthesis of HSP68, HSP70 and HSP90 proteins was determined for each case by computerassisted laser densitometric scanning of the autoradiograph of the electrophoretically separated 35S-labeled proteins. Changes in synthesis are expressed as the area under the curve (AUC) for protein band intensity after standardizing gel lane background intensity at zero units.
synthesis (methionine incorporation) was found to be preserved and, after 15 min in the presence of 8% ethanol, the HSP90 and HSP68 were overexpressed by 1.5- and 4.4-fold, respectively. Exposure to propanol- 1. No chronic exposure to propanol allowed cell growth or life processes, underscoring the high toxicity of this alcohol when compared to ethanol. When given once for 15 min at 2.5%, propanol induced an overexpression of HSP70 and HSP68 by 1.4- and 2.6-fold, respectively. If given for 30min at the same concentration, HSP70 and HSP68 were overexpressed by 1.4and 3.3-fold, respectively. At 5% for 15 min, propanol is found to totally inhibit the protein synthesis, which corroborates our results related to cell growth (see above). DISCUSSION
From the present results, we state that our biological model, i.e. HT29 cells, reacts to aggressive treatments by changing its growth rate and by producing stress proteins. However, these two parameters don’t appear to be strictly linked. In fact, it depends on the nature and on the level of aggression.
390
Florence
Delmas
If we compare the effects of stressing situations on either growth rate or HSP expression, it appears that weakest heat shock showing effectiveness (i.e. 45°C for 30 min) under our experimental conditions is found to affect both growth rate and HSP expression. However, the effect on HSP expression is much more evident than that on growth rate, for which a skilful statistical analysis is needed in order to provide reliable data, not to mention the fact that the establishment of a growth-curve over days is per se time-consuming. Excessive heat shock (i.e. 45°C for 60 min) does not modify cell growth so much, whereas it produces a clear-cut transient inhibition (over 6 hr) of protein synthesis. When considering the effect of cold shock, it appears that at least 4 hr at 1°C are needed to get a visible effect on growth curve, whereas 1 hr at 1°C is enough to give an obvious increase in HSP expression. Acute treatment with 5% ethanol for 60min modifies neither growth rate nor HSP expression. 8% ethanol for 15 min is required to induce a clear effect on both growth rate and HSP expression. If we consider the effects of propanol-1, it appears that the 2.5% for 15 or 30min acute exposure does not convincingly affect cell growth. By contrast, the HSP68 is clearly overexpressed under the same conditions. Differences in the sensitivity of the cells to express stress proteins as a function of the phases of culture could be due to the difference in degree of differentiation; it is known that HT29 cells are mostly undifferentiated when they divide actively, whereas some indices of differentiation appear in post confluent cells (Zweibaum et al., 1985). Paying attention to the different HSP families expressed in HT29 cells after aggression leads to the conclusion that some specificity exists in the overexpression of stress proteins. For instance, overexpression of HSPllO family appears only after heat shock. It is also clear that HSP68 is quantitatively the most overexpressed among HSP in our cellular model. We conclude that study of HSP expression gives faster and more obvious and clear-cut results than mere observation of growth curve for evaluation of the stressing power of an environmental aggression. This advantage, together with the development of an easy-to-use method to detect the HSP induction, could make our model a suitable biological sensor for environmental aggressions.
et al.
Acknowledgements-We thank Dr Siddharta Gautama for his wise contribution to the experimental work and MS Agnes Y&he for her kind help in preparing the manuscript. This research was made possible thanks to the fellowship from the ADEME to Florence Delmas.
REFERENCES Baler R., Welch W. J. and Voellmy R. (1992) Heat shock genes regulation by nascent polypeptides and denatured proteins: HSP70 as a potential autoregulatory factor. .I. Cell Biol. 117, 1151-1159. Bradford M. M. (1976) A rapid sensitive method for the quantitation of protein utilizing the principle of protein dye binding. Analyt. Biorhem. 72, 248-254. Faber F., Egli T. and Harder W. (1993) Transient repression of the synthesis of OmpF and aspartate transcarbamoylase in E. coli K12 as a response to pollutant stress. Microbial. Lett. 111, 189-196. Fischbach M., Sabbioni E. and Bromley P. (1993) Induction of the human growth hormone gene placed under human HSP70 promoter control in mouse cells: a quantitative indicator of metal toxicity. Cell Biol. Toxicol. 9, 177-188. Fogh J.. Fogh J. M. and Orfeo T. (1977) Hundred and twenty seven cultured human tumor cell-lines producing tumors in nude mice. J. Nat. Cancer Inst. 59, 221-226. Goering P. L., Kish C. L. and Fisher B. R. (1993) Stress protein synthesis induced by cadmium-cysteine in rat kidney. Toxicology 85, 25-39. Hatayama T.. Asai Y., Wakatsuki T., Kitamura T. and Imahara H. (1993) Regulation of HSP70 synthesis induced by cupric sulfate and zinc sulfate in thermotolerant HeLa cells. J. Biochem. 114, 592Z597. Hendrick J. P. and Hart1 F. U. (1993) Molecular chaperone functions of heat-shock proteins. Ann. Rev. Biochem. 62, 3499384. Hightower L. E. (1980) Cultured animal cells exposed to amino-acid analogs or puromycin rapidly synthesize several polypeptides. J. Cell Physiol. 102, 407427. Holland D. B., Roberts S. G., Wood E. J. and Cunliffe W. J. (1993) Cold shock induces the synthesis of stress proteins in human keratinocytes. J. Invest. llermatol. 101, 1966199. Kim S. H., Kim J. H., Erdos G. and Lee Y. J. (1993) Effect of staurosporine on suppression of heat shock gene expression and thermotolerance development in HT29 cells. Biochem. biophys. Res. Comm. 193, 759-763. Li G. C. (1983) Induction of thermotolerance and enhanced heat shock protein synthesis in Chinese hamster fibroblasts by sodium arsenite and by ethanol. J. CeN. Physiol. 115, 116122. Lindquist S. (1986) The heat shock response. Ann. Rev. Biochem. 55, 1151-1191. Novak T. S. (1993) Synthesis of heat shock/stress proteins during cellular injury. Ann. N. Y. Acad. Sci. 679, 142-156. Peiham H. R. B. (1989) Heat shock and the sorting of luminal ER proteins. EMBO J. 8, 3171-3176. Ritossa F. (1962) A new puffing pattern induced by a temperature shock and DNP in Drosophila. Experienfia 18, 571-573. Sierra-Rivera E., Voorhees G. J. and Freeman M. L. (1993) Gamma irradiation increases HSP70 in Chinese hamster ovary cells. Radiation Res. 135, 4045.
Stress proteins in HT29 cell-line Welch W. J. (1992) Mammalian stress response: physiology, structure/function of stress proteins, implications for medicine and disease. Physiol. Rev. 1063-1081. Zweibaum A., Pinto M., Chevalier G., Dussaulx
cell and 72,
E.,
391
Triadou N., Lacroix B., Haffen K. and Rousset M. (1985) Enterocytic differentiation of a subpopulation of the human colon tumor cell-line HT29 selected for growth in a sugar-free medium and inhibition by glucose. J. Cell, Physiol. 122, 21-28.