- Email: [email protected]
The further development of rainbow trout primary epithelial cell cultures as a diagnostic tool in ecotoxicology risk assessment
Aquatic Toxicology 53 (2001) 279– 289 www.elsevier.com/locate/aquatox
The further development of rainbow trout primary epithelial cell cultures as a diagnostic tool in ecotoxicology risk assessment Kevin Dowling, Carmel Mothersill * Radiation Science Centre, Dublin Institute of Technology, Ke6in St., Dublin 8, Ireland
Abstract The use of short-term cytotoxicity assays for the initial screening of chemicals not only aids in establishing priorities for the selection of chemicals that should be tested in vivo, but also decreases the time in which potential toxicants can be valued. Rainbow trout primary skin epithelial cell cultures are one such assay. Rainbow trout primary skin cell cultures contain two cell types, keratinocytes and goblet mucus cells. Two aquatic pollutants, copper and prochloraz were screened using this cell system. The influence of media composition on the effects of the aquatic pollutants was also studied by testing the chemicals in both serum-containing and serum-free medium and the morphological changes that occurred within the cell cultures recorded. The concentration of copper that causes a reduction of 90% in the residual of day 3 growth of the primary cell culture system was found to be approximately 10 fold more than that of prochloraz. Prochloraz was found to cause a greater reduction in growth area when added to the primary cell culture system in serum-free medium than in serum-containing medium. Copper, in contrast, was found to exert reduced toxicity when added to the test cultures in serum-free medium compared with addition in serum-containing medium. Prochloraz was found to kill the epithelial cells by a process of necrosis. Copper, was found to kill the epithelial cells by both necrosis and apoptosis in a ratio of 2:1. It was also observed that as the dose of both chemicals increased, the number of goblet cells contained in the cell cultures decreased. A PAS stain was carried out to determine if the goblet cells were exocytosing their contents onto the cell culture surface. It was found that as chemical exposure increased the number of cells expressing positivity for mucus also increased. The results of this study add further evidence to support that primary cell cultures are a very appropriate model for toxicity risk assessment. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Rainbow trout primary epithelial cell cultures; Prochloraz; Copper; Serum-containing and serum-free media; Cell death; Goblet cells
1. Introduction
* Corresponding author. Tel.: + 353-1-4024558; fax: +3531-4756793. E-mail address: [email protected] (C. Mothersill).
Xenobiotics are chemicals which are foreign to biological systems and include industrial chemicals and pesticides (Nowak, 1997). Some naturally
0166-445X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 4 4 5 X ( 0 1 ) 0 0 1 7 2 - 2
280
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289
occurring substances have also been found to be toxic to aquatic life when additional quantities are introduced into the environment. As many xenobiotics enter into the aquatic environment, there is a need to develop rapid cost-effective bioassays to screen the vast number of chemicals for their potential ecotoxicology (Babich and Borenfreund, 1987a,b). In the broad perspective of environmental toxicology and biohazard assessment, epithelial cells are highly significant biological indicators (Bloom and Fawcett, 1970). Epithelial cells form the barrier between an individual and its environment, and are a primary target for pollution related carcinogenises (Langdon, 1983; Iger et al., 1994). Two xenobiotics, prochloraz and copper, were investigated under the direction of the European Diagnostic Ecotoxicology project (contract number ENV4-CT96-0223). Prochloraz (1-[N-propylN-2-(2,4,6-trichlorophenoxy)ethylcarbamoyl] imidazole) is an agricultural fungicide developed for use on cereals and oilseed rape (Wakerly and Russell, 1985) and has been found to be toxic to fish. (Rievere et al., 1984). Prochloraz also interacts with trout liver cytochrome P450 and was found to inhibit alderin epoxidase and EROD activity in vitro (Snegaroff and Bach, 1989). Although considered an essential metal for most, if not all, living organisms (Maage et al., 1989) copper is also a very toxic pollutant to aquatic biota (Hellawel, 1988) if it exceeds the standard non-toxic value of 5– 15 mg/l (Birge and Black, 1979). It has been shown to cause mutagenic effects by binding to DNA producing DNA strand breaks (Sagripanti et al., 1991). Copper has also been shown to depurinate DNA by releasing adenine. Depurination occurs concomitant with DNA strand scisson (Schaaper et al., 1987). Copper also catalyses peroxidation of membrane lipids (Chan et al., 1982) and causes a reduction in EROD activity in vitro (Stein et al., 1997). Mothersill et al. (1995) developed an in vitro technique for the culture of rainbow trout (Onchorynchus mykiss) skin epithelial cell cultures. Primary cells can be made to express a high number of differentiated properties. Rainbow trout primary skin epithelial cell cultures contain both epithelial cells and goblet mucus cells (Mothersill
et al., 1995). Dowling and Mothersill (1999) modified the technique established by (Mothersill et al., 1995) into a workable tool for in vitro toxicity assessment. However, in this study the influence of serum in the test medium was not considered when testing the toxicity of the aquatic pollutant in this study. The aim of this study is to further investigate the use of rainbow trout primary epithelial skin cell cultures, as an in vitro diagnostic tool for the screening of environmental contaminants. The toxicity of the aquatic pollutants, prochloraz and copper, on rainbow trout primary skin epithelial cultures was investigated using the method described in Dowling and Mothersill (1999). The influence of media composition on the toxicity of the substances was also studied by testing the chemicals in both serum-containing and serum-free medium. The ability of the goblet mucus cells to secrete their contents onto the cell culture surface, as suggested in Dowling and Mothersill (1999), was also investigated.
2. Materials and methods
2.1. Primary culture Rainbow trout were obtained from a commercial fish farm in Roscrea, Ireland. Epithelial cell outgrowths were obtained from explants using a modification of the technique earlier developed by Mothersill et al. (1995). A section of skin from the dorsal side was taken, cleaned of muscle tissue and chopped to 2 mm2 pieces. Individual tissue fragments were placed in 24 cm2 Nunc flasks (Biosciences, Dublin, Ireland) containing 2 ml of initiating medium containing serum as reported in Mothersill et al. (1995). Trypsin/collagenase was not used as reported in Mothersill et al. (1995). The cultures were incubated at 19°C in an atmosphere containing 5% CO2 in a refrigerated incubator (Leec, Nottingham, UK).
2.2. Chemical assays Prochloraz (Sigma Chemicals, London) were dissolved in a known volume of dimethylsulfoxide
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289
(DMSO). Sequential dilutions of this stock were made prior to addition to the cultures in volumes of 50 ml per flask; the control contained 50 ml DMSO (conc. 0.1% in solution). Cupric copper chloride (Sigma Chemicals, London) solution was prepared as an aqueous stock solution. Sequential dilutions of this stock were made prior to addition to the cultures in volumes of 50 ml per flask. The control contained 50 ml of water.
Fig. 1. Normal rainbow trout epithelial skin cell culture showing; (A) normal epithelial cells and (B) goblet cells ( × 20 Magnification).
Fig. 2. Rainbow trout epithelial skin cell culture showing necrotic cells (N) ( ×20 Magnification).
281
2.3. Cytotoxicity endpoints 2.3.1. Cell growth reduction The reduction of cell growth in primary cultures by the three test substances, was investigated. After 4 days post attachment the perimeter of the cultures was outlined on the bottom of the flask with permanent marker pen. Cells that have grown out from the explant form a whitish area that can be observed with the naked eye. The cultures were then treated for 24 and 48 h with sequential concentrations of the test substances. The experiment was repeated using two fish, three cultures per chemical concentration per fish. After exposure, the cultures were fixed in 10% buffered formalin and stained in haematoxylin. The outgrowth area from the explant was estimated using a 1 mm2 grid by counting the number of squares covered by the explant outgrowth. The area before treatment was also measured. The results are given as a percentage of the solvent control growth. The mean values and standard errors were calculated. This method takes into consideration that primary cultures from the same animal do not grow uniformly under the same conditions. The media in a number of the cultures was changed to serum-free Clonetics keratinocyte growth medium (KGM) (Clonetics Corporation, USA) prior to the addition of the test substances and the same procedure carried out as described above. 2.3.2. Morphology The types of cells present in each cell culture, treated in serum containing medium, were counted using a Leica image analyser (Leica Microsystems Holdings, Ernst-Leitz-Strasse, Germany). Five fields, each 0.55 mm2, were analysed from each of the explants described above, two from the inner region, two from the outer region and one from the mid-region. The cells were classified as (a) normal epithelia (Fig. 1), (b) goblet (Fig. 1), (c) Necrotic (Fig. 2) and (d) apoptotic (Fig. 3). Goblet cells are recognisable in the epithelium by their basophilic cytoplasm, their basally located nuclei, and the accumulation of mucus secretary granules that fill and distend their apex to give the cells their characteristic goblet shape (Weiss, 1988). Cells were classified as undergoing necrosis if they exhib-
282
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289
2.4. The PAS (Periodic-Acid Schiff ) stain for glycogen deposits
Fig. 3. Rainbow trout epithelial skin cell culture showing apoptopic cells (A). ( ×40 Magnification).
Cultures were fixed in 80% alcohol at 4°C for 30 min for glycogen identification. The samples were rinsed repeatedly in distilled water and covered in 1% periodic acid for 5 min. The cultures were then rinsed in distilled water and covered in Schiffs reagent (Sigma) for 10 min. The cultures were rinsed in several changes of distilled water then placed in running tap water for 10 min to aid colour development. The cell nuclei were counterstained using Mayer’s haematoxylin for 1 min and washed in hot flowing running tap water for 5 min to enhance blue staining. Coverslips were then attached to stained preparations with Kaiser’s glycerol jelly mounting media. The neutral mucins and glycogen stained magenta pink, the nuclei blue.
2.5. Scoring of PAS Stain for mucus results
Fig. 4. Rainbow trout epithelial cells stained negative for mucus. The goblet mucus cells (G) are stained intensely pink. ( × 20 Magnification).
ited the following features; (a) swelling of the cytoplasm and organelles, with only slight changes in the nucleus or (b) organelle dissolution and rupture of the plasma membranes resulting in leakage of the cellular contents into the extracellular space (Schwartzman and Cidlowski, 1993). Cells were classified as undergoing apoptosis if they showed evidence of two or more of the following, (a) cell volume shrinkage and picnotic nucleus (chromatin condensation), (b) blebbing of the cytoplasm, (c) nuclear fragmentation and (d) development into apoptotic bodies (Kerr and Harmon, 1991; Schwartzman and Cidlowski, 1993; Martin et al., 1994).
The epithelial cell surfaces were classed as being negative, weakly positive and intensely positive to mucus (Figs. 4 and 5). Cells were classed as weakly positive and intensely positive to mucus if they were stained light pink and dark pink, respectively. Negative cells stained blue. The mu-
Fig. 5. Rainbow trout epithelial cell cultures stained weakly positive (W) and intensely positive (I) for mucus. ( × 20 Magnification).
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289
283
Table 1 The percent residual of day 3 growth of rainbow trout primary epithelial cell cultures exposed to prochloraz for 24 and 48 h in serum-containing and serum-free mediaa Dose (mmol/l)
Residual of day 3 growth (%) 24 h
0 1 10 25 50 75
48 h
Serum-containing
Serum-free
Serum-containing
Serum-free
100 98.99 5.2 90.59 6.9 72.69 2.7 31.193.7 7.99 1.5
100 96.0 93.7 93.9 93.5 44.8 92.1 10.2 90.7 0.6 90.5
100 86.6 94.3 91.3 92.8 50.5 94.7 24.3 92.0 8.9 92.2
100 97.9 9 2.6 87.1 91.8 31.8 9 3.3 5.0 9 1.1 0.4 90.4
a The results are expressed as mean 9SEM for n= 6 replicate cultures. The values are based on the overall area of the cell culture inclusive of both epithelial cells and goblet cells.
3.1.1. Primary culture growth data
3.1.1.2. Serum-free medium. The primary cultures were found to exhibit a significantly different response to prochloraz when exposed in serum free media (Table 1). The dose of the pollutant required to kill 50% of the cells was found to be between 10 and 25 mmol/l for both 24 and 48 h exposures. The threshold level of response was found to be 10 mmol/l for both exposure periods. The solvent controls were found to exhibit a significantly less percent residual of day 3 growth (PB 0.009, for 48 h) than that of the cultures exposed to the test substance in serum-containing medium. The percent growth of the controls in serum-free medium was 68.291.0 and 73.89 1.2 for 24 and 48 h exposures, respectively.
3.1.1.1. Serum-containing medium. Prochloraz was found to kill 50% of the cells at a concentration of between 25 and 30 mmol/l after 24 h exposure (Table 1). The solvent control cell cultures were found to exhibit a percent residual of day 3 growth of 73.792.4 and 79.79 1.3 for 24 and 48 h exposure periods, respectively. Prochloraz tested on the primary cell system was found to have an acute lethal dose of 75 mmol/l for all exposure times. The threshold level of response, the level of exposure at which a significant decrease in survival begins to occur, is 10 mmol/l for both exposure times. The survival values are based on the overall area of the culture inclusive of both epithelial cells and goblet cells.
3.1.2. Primary culture cell death type (serum-containing medium) The average number of cells per field in the control cultures was 31.691.2 (Table 2). The amount of cells per field decreased as the chemical treatment dose increased. The number of cells per field in the cultures treated with 75 mmol/l prochloraz was 18.290.4. Table 2 shows that an increase in epithelial necrotic cells was observed with increasing prochloraz exposure in primary cell cultures. Epithelial cells located on the outside edge of the culture underwent necrosis primarily at lower doses. A significant increase (PB 0.0001) in those cells undergoing necrosis was observed at 25 mmol/l (0.1–5.19 1.4 cells per field) exposure
cous cells contained in the cell cultures reacted with intense positivity to PAS staining. The mucous cells were not scored in the results.
2.6. Statistics The unpaired t-test was used to compare the residual growth values and the percentage of cells expressing positivity to mucus. 3. Results
3.1. Prochloraz
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289
284
to the pollutant. Only a slight increase in apoptopic cell death was observed with increasing prochloraz exposure (0– 1.5 90.3 cells per field after 48 h exposure to 75 ml prochloraz).
19.69 2.2. The percent of cells stained weakly positive was 15.59 1.6, the percent stained with intense positivity was 4.29 1.4 (Table 3).
3.2. Copper 3.1.3. Relati6e effect on mucus cells 6ersus epithelial cells (serum-containing medium) A decrease in the number of goblet cells per field (from 7.09 1.7 to 0.99 0.5) was also observed as prochloraz dose increased (Table 2). This decrease was observed uniformly across the explant. The threshold level of this decrease was found to be 25 mmol/l. 3.1.4. PAS Stain for mucus The percent of cells covered with a mucus layer was found to increase significantly (P B 0.0001) with increasing exposure to prochloraz (Table 3). The percent of cells expressing positivity in the DMSO controls was 7.9%. The majority of this positivity was weak staining. The number of cells scored per explant in the controls was 4089 99. The percent of cells stained positive for mucus post exposure with 50 mmol/l prochloraz was
3.2.1. Primary culture growth data 3.2.1.1. Serum-containing medium. Copper was found to kill 50% of the cells, at a concentration of about 400 mmol/l for all exposure durations (Table 4). The solvent controls were found to exhibit a percent residual of day 3 growth of 101.491.4 and 101.295.2 for 24 and 48 h exposure periods, respectively. Copper tested on the primary cell system was found to have an acute lethal dose of greater than 1000 mmol/l for all exposure times. The threshold level of response, the level of exposure at which a significant decrease (PB 0.0001) in survival begins to occur, is 100 mmol/l for all exposure periods. The residual growth values are based on the overall area of the culture inclusive of both epithelial cells and goblet cells.
Table 2 Type of cells per primary epithelial cell culture post 48 h exposure to prochloraz in serum-containing mediuma Dose (mmol/l)
Number of normal epithelial cells/field
Number of goblet cells/field
Number of necrotic epithelial cells/field
Number of apoptotic epithelial cells /field
Control 1 10 25 50 75
24.5 9 2.1 24.8 91.7 21.5 91.6 12.5 9 2.9 5.8 9 2.7 3.1 91.3
7.09 1.7 8.2 9 1.6 5.99 0.8 5.8 9 1.1 1.3 9 1.4 0.99 0.5
0.1 9 0.1 0.1 90.1 1.4 9 0.6 5.1 91.4 11.5 9 2.4 12.7 9 2.4
0 0 0.5 9 0.3 0.6 90.3 2.19 0.7 1.5 90.3
a
Each value represents the mean 9S.E.M. for five (0.55 mm) grids counted in n =6 replicate cultures.
Table 3 The percent number of rainbow trout primary epithelial cells stained positive for mucus post treatment with prochloraz in serum-containing mediuma Dose (mmol/l)
Weak positivity (%)
Intense positivity (%)
Total positivity
0+DMSO 10 25 50
7.3 9 1.3 10.49 2.1 15.69 2.0 15.591.6
0.7 90.3 2.3 9 0.1 3.0 91.0 4.2 91.4
7.9 9 1.5 12.7 9 2.6 18.6 9 2.7 19.6 9 2.2
a
Each figure represents the mean 9S.E.M. for n = 9 cultures.
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289
285
Table 4 The percent residual survival of day 3 growth of rainbow trout primary epithelial cell cultures exposed to prochloraz for 24 and 48 h in serum-containing and serum-free mediaa Dose (mmol/l)
Residual of day 3 growth (%) 24 h
0 1 10 50 100 500 1000
48 h
Serum-containing
Serum-free
Serum-containing
Serum-free
100 101.192.7 102.391.8 96.892.5 93.693.4 35.09 4.1 10.891.5
100 99.49 5.6 95.9 92.4 85.5 94.5 92.49 5.4 77.6 93.6 10.7 91.2
100 101.4 93.4 108.1 92.1 101.9 9 2.8 107.7 9 2.6 38.2 95.2 8.0 9 1.3
100 101.9 9 2.9 98.6 9 4.6 92.3 9 2.8 85.2 9 5.6 67.6 9 4.5 10.3 9 1.2
The results are expressed as mean 9 S.E.M. for n = 6. The values are based on the overall area of the cell culture inclusive of both epithelial cells and goblet cells. a
Table 5 Type of cells per primary epithelial cell culture post 48 h exposure to copper in serum-containing mediuma Dose (mmol/l)
Number of normal epithelial cells/field
Number of goblet cells/field
Number of necrotic epithelial cells/field
Number of apoptotic epithelial cells /field
Control 10 50 100 500 1000
23.2 91.3 23.2 9 1.9 23.1 91.4 20.2 91.1 5.9 91.7 1.7 91.4
7.09 1.7 7.29 1.2 6.29 1.2 6.7 9 1.2 2.19 1.4 1.19 0.6
0.1 9 0.1 0.1 9 0.1 0.2 9 0.1 0.2 90.1 11.6 9 2.1 12.6 9 2.8
0 0 0 0.1 90.1 3.2 90.6 5.8 90.4
a
Each value represents the mean 9S.E.M. for five (0.55 mm) grids counted in n =6 replicate cultures.
3.2.1.2. Serum-free medium. The primary cultures were found to exhibit a significantly (P B 0.005) different response to copper when exposed in serum free media (Table 4). The dose of the pollutant required to reduce the growth area by 50% was found to be between 500 and 1000 mmol/l for both 24 and 48 h exposure periods. The threshold level of response was found to be 100 mmol/l for both exposure periods. The solvent control for 24 h were found to exhibit a significantly less percent residual of day 3 growth (P B 0.005) for 24 h, but not (P = 0.9) for 48 h, than that of the cultures exposed to the test substance in serum-containing medium. The percent residual of day 3 growth of the solvent control cultures in serum-free medium was 84.69 2.1 and 100.89 2.4 for 24 and 48 h exposures, respectively.
3.2.2. Primary culture cell death type (serum-containing medium) The average number of cells per field in the control cultures was 30.390.8 (Table 5). The amount of cells per field decreased as the chemical treatment dose increased. The number of cells per field in the cultures treated with 1000 mmol/l copper was 21.290.4. Table 4 shows that an increase in epithelial necrotic cells and apoptopic cells was observed with increasing copper exposure in primary cell cultures. Epithelial cells located on the outside edge of the culture underwent necrosis primarily at lower doses. A significant increase (PB 0.0001) in those cells undergoing necrosis was observed at 500 mmol/l (0.1–12.69 2.8 cells per field) exposure to the pollutant. A slightly less, but also significant, in-
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289
286
crease (P B0.0001) in apoptopic cell death was observed with increasing copper exposure (0– 5.8 90.4 cells per field after 48 h exposure to 1000 ml copper).
3.2.3. Relati6e effect on mucus cells 6ersus epithelial cells (serum-containing medium) A decrease in the number of goblet cells per field from 7.09 1.7 to 1.190.6) was also observed as copper dose increased (Table 5). This decrease was observed uniformly across the explant. The threshold level of this decrease was found to be 100 mmol/l. As observed in the cultures exposed to the other test substances, the goblet cells release their secretory contents onto the culture surface. The percentage of epithelial cells, per field, in the cultures was found to increase by 17.41% for 48 h exposure as the dose of pollutant increased. 3.2.4. PAS stain for mucus The percent of cells covered with a mucus layer was found to increase significantly (P B 0.0001) with increasing exposure to copper (Table 6). The percent of cells expressing positivity in the controls treated with water was 7.0%. The majority of this positivity was weak staining. The number of cells scored per explant in the controls was 4099 43. The percent of cells stained positive for mucus post exposure with 700 mmol/l copper was 17.99 3.0. The percent of cells stained weakly positive was 9.09 1.7, the percent stained with intense positivity was 8.89 2.6 (Table 6).
4. Discussion The toxicity of copper to the primary cell culture system was found to be approximately 10 fold less than that of the organic chemical studied. This is as one might expect as organic chemicals have been found in the past to be more toxic to fish cells than heavy metals (Babich et al., 1986). The toxicity of copper was observed to be less toxic to the primary epithelial cultures than that of cadmium, observed by Lyons Alacantara et al. (1996), and more toxic than that of nickel, observed by McSweeney (1998). Babich et al. (1986) observed a similar relationship between these metals when testing the in vitro cytotoxicity of metals to bluegill BF-2 cells. The toxicity curves of prochloraz and copper were similar to that obtained in Dowling and Mothersill (1999). An element that must not be neglected in any bioassay system, is the composition of the medium. The influence of medium composition on toxicity testing using the rainbow trout primary skin epithelial cell culture system was studied by exposing the cultures to the test substances in both serum-free and serum-containing medium. The process of changing the media to serum-free medium appeared to kill up to 20% of the control samples. Prochloraz proved to be more toxic when tested in serum-free medium (Table 1). A similar result was observed by Bertheussen et al. (1997) when testing the toxicity of the pesticides carbofuran, cypermethrin, lindane, glycophosphate and 2,4-D to the IE6 cell line. Lyons Ala-
Table 6 The percent number of rainbow trout primary epithelial cells stained positive for mucus post treatment with copper in serum-containing mediuma Dose (mmol/l)
Weak positivity (%)
Intense positivity (%)
Total positivity
0+Water 100 400 700
6.69 1.0 6.99 1.0 10.49 0.8 9.09 1.7
0.4 90.2 5.2 91.9 3.6 91.4 8.8 9 2.6
7.0 9 1.2 12.1 9 1.8 14.0 9 1.6 17.99 3.0
a
Each figure represents the mean 9SE.M. for n= 9 cultures.
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289
cantara et al. (1996) observed that both human primary urothelium cultures and fish skin primary epithelial cell cultures were more sensitive to cadmium in serum-free medium. Copper, in contrast, was found to be less toxic when added to the test cultures in serum-free medium compared with addition in serum-containing medium (Table 3). An important parameter of the cellular uptake and subsequent cytotoxicity of metal ions such as Cu++, is their binding to extracellular biological components and the resulting influence on their ability to interact with the cell membrane, enter the cell and exert intracellular cytotoxic effects. Whilst it appears that the serum components are preventing prochloraz from entering the cell, the serum components appear to be increasing the uptake of copper into the cell. Webb and Weinzierl (1972) suggested that nickel ions enter the cell after binding to low molecular weight serum components that actually serve to transport metallic ions into the cell. It appears that something similar may be occurring in the experiments carried out in this study. As exposure levels to the test substances increased, the quantity of goblet cells decreased. Goblet cells produce mucus. Mucus production is an established defence mechanism of epithelial surfaces against pathogens and pollutants (Shephard, 1994). The goblet cells release their secretary contents onto the culture surface forming a thick slimy protective coating in an attempt to prevent deleterious effects of the pollutant medium (Roy, 1988). It was observed using the PAS stain that the goblet cells were decreasing in number because they are exocytosing their contents. A similar observation was made by Dowling and Mothersill (1999) when studying the effects of nonoxynol on primary skin epithelial cell cultures. The results of this study demonstrate that in vitro rainbow trout primary epithelial cell cultures maintain the essential traits of the in vivo response of the skin to chemical stresses. The goblet cells contained in the epithelial cell cultures are retaining their in vivo functions in vitro. A decrease in goblet number and density in rainbow trout primary skin culture could be an important marker of toxicity of aquatic pollutants. An increase in necrotic epithelial cells was observed with increasing dose exposure of prochloraz
287
in serum containing media. The number of apoptotic epithelial cells remained extremely low and constant with dose exposure. These results suggest that prochloraz kills rainbow trout skin epithelial by the process of necrosis. Other organic and inorganic chemicals have been shown to induce similar responses in fish (Wester et al., 1988; Dietrich and Schlatter, 1989). In contrast, copper was found to kill rainbow trout primary skin epithelial by both necrosis and apoptosis in a ratio of approximately 2:1. Pelgrom (1995) observed that epithelial cellular responses of teleosts to Cu++ exposure include both increased necrosis and apoptosis of chloride cells and pavement cells. Carp exposed to 1.6 mmol/l Cu++ for 7 days were observed to show degenerative (apoptopic and necrotic) pavement cells in the skin (Iger et al., 1994). Other in vivo studies have shown that Tilapia exposed to Cu++ have a high incidence of apoptosis in the epidermis (Pelgrom, 1995). Copper shows a high specific binding to DNA producing DNA strand breaks in several biological systems (Sagripanti et al., 1991). Copper has been shown to act as a catalyst in the formation of reactive oxygen species (Chan et al., 1982). Evidence has shown that oxygen radicals may have a role to play in the initiation of apoptosis (Buttke and Sandstorm, 1994). The results of this study add further evidence to support that primary cell cultures are a very appropriate model for toxicity risk assessment. If cultured under the appropriate conditions the cells retain their full, in vivo, characteristics for up to 14 days, thus allowing sufficient time for investigation of the immediate in vitro effects of toxicants. The results of this study also clearly demonstrate that one should take into account the composition of the media, and the binding properties of the substance to be screened, when planning toxicity tests using this cell system. This is especially the case when screening compounds in relation to each other. The use of short-term cytotoxicity assays for the initial screening of chemicals not only aids in establishing priorities for the selection of chemicals that should be tested further in vivo, but also decreases the time in which potential toxicants can be evaluated (Babich et al., 1986). The primary cell culture system used in this study is an extremely
288
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289
useful tool for such assays for the reasons earlier stated. Work carried out in our group with this system has shown that changes in glycogen deposits, stress proteins and oncogene expression could also be studied using this system (Lyons Alacantara et al., 1996; McSweeney, 1998). The primary cell culture system could also be utilised for different areas of study. These include multiple assays/investigations on homologous cultures derived from in vivo exposure, which may result in a major step in the solving of the problem of comparing in vitro effects with that of in vivo. The cultures can also be infected with viruses and as such provide excellent opportunities for research into fish diseases.
Acknowledgements This work was supported by the European Communities Environmental Program: Diagnostic Ecotoxicology; Cell-Based Methodology to develop Markers for Early, Sublethal Effects Assessment: Contract ENVA-CT96-0223.
References Babich,H.,Borenfreund,E.,1987a.Invitrocytotoxicityoforganic pollutants to bluegill sunfish (BF-2) cells. Environ. Res. 42, 229 – 237. Babich, H., Borenfreund, E., 1987b. Fathead minnow FHM cells for use in in vitro cytotoxicity assays of aquatic pollutants. Ecotoxicol. Environ. Saf. 14, 78 –87. Babich, H., Puerner, J.A., Borenfreund, E., 1986. In vitro cytotoxicity of metals to Bluegill (BF-2) cells. Arch. Environ. Contam. Toxicol. 15, 31 –37. Bertheussen, K., Youseff, M.I., Figenschau, Y., 1997. A new sensitivecellculturetestfortheassessmentofpesticidetoxicity. J. Environ. Sci. Health B32, 195 – 211. Birge, W.J., Black, J.A., 19729. In: Nriagra, J.O. (Ed.), Copper in the Environment. Wiley-Interscience, New York, pp. 373 – 379. Bloom, W., Fawcett, D.W., 1970. A Textbook of Histology. WB Saunders, London. Buttke,T.M.,Sandstorm,P.A.,1994.Oxidativestressasamediator of apoptosis. Immunol. Today 15, 7 –10. Chan, P.C., Peller, O.G., Kesner, L., 1982. Copper(II)-catalyses lipid peroxidation in liposomes and erythrocyte membrane. Lipids 17, 331 – 337.
Dietrich, D., Schlatter, C., 1989. Auminium toxicity to rainbow trout at a low pH. Aquat. Toxicol. 15, 197 – 212. Dowling,K.,Mothersill,C.,1999.Theuseofrainbowtroutprimary epidermal cell cultures as an alternative to immortalised cell lines in toxicity assessment. A study with nonoxynol. Environ. Toxicol. Chem. 18 (12), 2846 – 2850. Hellawel, J.M., 1988. Toxic substances in rivers and streams. Environ. Pollut. 50, 61 – 85. Iger, Y., Lock, R.A.C., Jenner, H.A., Wendelaar Bonga, S.E., 1994. Cellular responses in the skin of carp (Cyprinus carpio) exposed to copper. Aquat. Toxicol. 29, 49 – 64. Kerr, J.F.R., Harmon, B.V., 1991. Definition and Incidence of Apoptosis. An Historical Perspective. Cold Spring Harbour Laboratory Press, New York. Langdon, J.S., 1983. Environmental management factors in diseases of fish. Fish Dis.: Proc. 106, 329 – 353. Lyons Alacantara, M., Tarazona, J.V., Mothersill, C., 1996. The differential effects of cadmium exposure on the growth and survival of primary and established cells from fish and mammals. Cell Biol. Toxicol. 12, 29 – 38. Maage, A., Sveir, H., Julshamn, K., 1989. A comparison of growth rateandtraceelementaccumulationinAtlanticSalmon(Salmo salar) fry fed four different commercial diets. Aquaculture 79, 267. Martin, S.J., Green, D.R., Cotter, T., 1994. Dicing with death: dissecting the components of the apoptosis machinery. TIBS 19, 26 – 30 Reviews. McSweeney, J.A., 1998. Investigation of the effects of Nockle on the growth and survival of primary cells and immortalised celllinesofhumanandteleostcelllines.PhDThesis,University of Dublin, Ireland. Mothersill, C., Lyng, F., Lyons, M., Cottell, D., 1995. Growth and differentiation of epidermal cells from the Rainbow Trout established as explants and maintained in various media. J. Fish Biol. 46, 1011 – 1025. Nowak, B.F., 1997. Biological significance of xenobiotic residues in fish. TEN 4, 148 – 154. Pelgrom, S.M.G.J., 1995. Interaction between copper and cadmium in fish: metal accumulation, physiology and endocrine regulation. PhD Thesis, University of Nimegen, The Netherlands. Rievere, J.L., Leroux, P., Bach, J., Gredt, M., 1984. Effects of some ergosterol biosynthesis inhibiting fungicides on sterols and cytochrome P-450 from the Japanese quail Coturnix coturnix. Pesticide Sci. 15, 317 – 323. Roy, D., 1988. Statistical analysis of anionic detergent induced changes in the goblet mucus cells of the opercular epidermis and gill epithelium of Rita rita (Ham)(Bargradal: pisces). Ecotoxicol. Environ. Saf. 15, 260 – 271. Sagripanti, J.L., Goering, P.L., Lamanna, A., 1991. Interaction of copper with DNA antagonism by other metals. Toxicol. Appl. Pharmacol. 110, 477 – 485. Schaaper, R.M., Koplitz, R.M., Theshelashvili, L.K., 1987. Metal-inducedlethalityandmutagenises:possibleroleofcupric intermediates. Mutat. Res. 177, 179 – 188. Schwartzman, R.A., Cidlowski, J.A., 1993. Apoptosis; the biochemistryandmolecularbiologyofprogrammedcelldeath. Endocr. Rev. 14, 133 – 151.
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289 Shephard, K.L., 1994. Functions for fish mucus. Rev. Fish Biol. Fisheries 4, 401 – 429. Snegaroff, J., Bach, J., 1989. Effects of the fungicide prochloraz on xenobiotic metabolism in rainbow trout: inhibition in vitro and time course of induction in vivo. Xenobiotica 19, 255– 267. Stein, X., Risso, C., Gnassia-Barelli, M., Romeo, M., Lafaurie, M., 1997. Effect of copper chloride in vitro and in vivo on the hepatic EROD activity in the fish Dicentrarchus labrax. Environ. Toxicol. Chem. 16, 214 – 219.
.
289
Wakerly, S.B., Russell, P.E., 1985. Prochloraz, a decade of development. Fungicides in crop protection. Proc. Bordeaux Mixture Centerary Meeting 2, 257 – 260. Webb, M., Weinzierl, S.N., 1972. Br. J. Cancer 26, 292. Weiss, L., 1988. Cell and Tissue Biology, Sixth edn. Urbam & Schwarzenberg, Baltimore. Wester, P.W., Canton, J.H., Dormans, J.A.M.A., 1988. Pathological effects in freshwater fish Poecilla reticulata (guppy) and Oryzias latipes (meduka) following methyl bromide and sodium bromide exposure. Aquat. Toxicol. 12, 323 – 343.
The further development of rainbow trout primary epithelial cell cultures as a diagnostic tool in ecotoxicology risk assessment Kevin Dowling, Carmel Mothersill * Radiation Science Centre, Dublin Institute of Technology, Ke6in St., Dublin 8, Ireland
Abstract The use of short-term cytotoxicity assays for the initial screening of chemicals not only aids in establishing priorities for the selection of chemicals that should be tested in vivo, but also decreases the time in which potential toxicants can be valued. Rainbow trout primary skin epithelial cell cultures are one such assay. Rainbow trout primary skin cell cultures contain two cell types, keratinocytes and goblet mucus cells. Two aquatic pollutants, copper and prochloraz were screened using this cell system. The influence of media composition on the effects of the aquatic pollutants was also studied by testing the chemicals in both serum-containing and serum-free medium and the morphological changes that occurred within the cell cultures recorded. The concentration of copper that causes a reduction of 90% in the residual of day 3 growth of the primary cell culture system was found to be approximately 10 fold more than that of prochloraz. Prochloraz was found to cause a greater reduction in growth area when added to the primary cell culture system in serum-free medium than in serum-containing medium. Copper, in contrast, was found to exert reduced toxicity when added to the test cultures in serum-free medium compared with addition in serum-containing medium. Prochloraz was found to kill the epithelial cells by a process of necrosis. Copper, was found to kill the epithelial cells by both necrosis and apoptosis in a ratio of 2:1. It was also observed that as the dose of both chemicals increased, the number of goblet cells contained in the cell cultures decreased. A PAS stain was carried out to determine if the goblet cells were exocytosing their contents onto the cell culture surface. It was found that as chemical exposure increased the number of cells expressing positivity for mucus also increased. The results of this study add further evidence to support that primary cell cultures are a very appropriate model for toxicity risk assessment. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Rainbow trout primary epithelial cell cultures; Prochloraz; Copper; Serum-containing and serum-free media; Cell death; Goblet cells
1. Introduction
* Corresponding author. Tel.: + 353-1-4024558; fax: +3531-4756793. E-mail address: [email protected] (C. Mothersill).
Xenobiotics are chemicals which are foreign to biological systems and include industrial chemicals and pesticides (Nowak, 1997). Some naturally
0166-445X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 4 4 5 X ( 0 1 ) 0 0 1 7 2 - 2
280
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289
occurring substances have also been found to be toxic to aquatic life when additional quantities are introduced into the environment. As many xenobiotics enter into the aquatic environment, there is a need to develop rapid cost-effective bioassays to screen the vast number of chemicals for their potential ecotoxicology (Babich and Borenfreund, 1987a,b). In the broad perspective of environmental toxicology and biohazard assessment, epithelial cells are highly significant biological indicators (Bloom and Fawcett, 1970). Epithelial cells form the barrier between an individual and its environment, and are a primary target for pollution related carcinogenises (Langdon, 1983; Iger et al., 1994). Two xenobiotics, prochloraz and copper, were investigated under the direction of the European Diagnostic Ecotoxicology project (contract number ENV4-CT96-0223). Prochloraz (1-[N-propylN-2-(2,4,6-trichlorophenoxy)ethylcarbamoyl] imidazole) is an agricultural fungicide developed for use on cereals and oilseed rape (Wakerly and Russell, 1985) and has been found to be toxic to fish. (Rievere et al., 1984). Prochloraz also interacts with trout liver cytochrome P450 and was found to inhibit alderin epoxidase and EROD activity in vitro (Snegaroff and Bach, 1989). Although considered an essential metal for most, if not all, living organisms (Maage et al., 1989) copper is also a very toxic pollutant to aquatic biota (Hellawel, 1988) if it exceeds the standard non-toxic value of 5– 15 mg/l (Birge and Black, 1979). It has been shown to cause mutagenic effects by binding to DNA producing DNA strand breaks (Sagripanti et al., 1991). Copper has also been shown to depurinate DNA by releasing adenine. Depurination occurs concomitant with DNA strand scisson (Schaaper et al., 1987). Copper also catalyses peroxidation of membrane lipids (Chan et al., 1982) and causes a reduction in EROD activity in vitro (Stein et al., 1997). Mothersill et al. (1995) developed an in vitro technique for the culture of rainbow trout (Onchorynchus mykiss) skin epithelial cell cultures. Primary cells can be made to express a high number of differentiated properties. Rainbow trout primary skin epithelial cell cultures contain both epithelial cells and goblet mucus cells (Mothersill
et al., 1995). Dowling and Mothersill (1999) modified the technique established by (Mothersill et al., 1995) into a workable tool for in vitro toxicity assessment. However, in this study the influence of serum in the test medium was not considered when testing the toxicity of the aquatic pollutant in this study. The aim of this study is to further investigate the use of rainbow trout primary epithelial skin cell cultures, as an in vitro diagnostic tool for the screening of environmental contaminants. The toxicity of the aquatic pollutants, prochloraz and copper, on rainbow trout primary skin epithelial cultures was investigated using the method described in Dowling and Mothersill (1999). The influence of media composition on the toxicity of the substances was also studied by testing the chemicals in both serum-containing and serum-free medium. The ability of the goblet mucus cells to secrete their contents onto the cell culture surface, as suggested in Dowling and Mothersill (1999), was also investigated.
2. Materials and methods
2.1. Primary culture Rainbow trout were obtained from a commercial fish farm in Roscrea, Ireland. Epithelial cell outgrowths were obtained from explants using a modification of the technique earlier developed by Mothersill et al. (1995). A section of skin from the dorsal side was taken, cleaned of muscle tissue and chopped to 2 mm2 pieces. Individual tissue fragments were placed in 24 cm2 Nunc flasks (Biosciences, Dublin, Ireland) containing 2 ml of initiating medium containing serum as reported in Mothersill et al. (1995). Trypsin/collagenase was not used as reported in Mothersill et al. (1995). The cultures were incubated at 19°C in an atmosphere containing 5% CO2 in a refrigerated incubator (Leec, Nottingham, UK).
2.2. Chemical assays Prochloraz (Sigma Chemicals, London) were dissolved in a known volume of dimethylsulfoxide
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289
(DMSO). Sequential dilutions of this stock were made prior to addition to the cultures in volumes of 50 ml per flask; the control contained 50 ml DMSO (conc. 0.1% in solution). Cupric copper chloride (Sigma Chemicals, London) solution was prepared as an aqueous stock solution. Sequential dilutions of this stock were made prior to addition to the cultures in volumes of 50 ml per flask. The control contained 50 ml of water.
Fig. 1. Normal rainbow trout epithelial skin cell culture showing; (A) normal epithelial cells and (B) goblet cells ( × 20 Magnification).
Fig. 2. Rainbow trout epithelial skin cell culture showing necrotic cells (N) ( ×20 Magnification).
281
2.3. Cytotoxicity endpoints 2.3.1. Cell growth reduction The reduction of cell growth in primary cultures by the three test substances, was investigated. After 4 days post attachment the perimeter of the cultures was outlined on the bottom of the flask with permanent marker pen. Cells that have grown out from the explant form a whitish area that can be observed with the naked eye. The cultures were then treated for 24 and 48 h with sequential concentrations of the test substances. The experiment was repeated using two fish, three cultures per chemical concentration per fish. After exposure, the cultures were fixed in 10% buffered formalin and stained in haematoxylin. The outgrowth area from the explant was estimated using a 1 mm2 grid by counting the number of squares covered by the explant outgrowth. The area before treatment was also measured. The results are given as a percentage of the solvent control growth. The mean values and standard errors were calculated. This method takes into consideration that primary cultures from the same animal do not grow uniformly under the same conditions. The media in a number of the cultures was changed to serum-free Clonetics keratinocyte growth medium (KGM) (Clonetics Corporation, USA) prior to the addition of the test substances and the same procedure carried out as described above. 2.3.2. Morphology The types of cells present in each cell culture, treated in serum containing medium, were counted using a Leica image analyser (Leica Microsystems Holdings, Ernst-Leitz-Strasse, Germany). Five fields, each 0.55 mm2, were analysed from each of the explants described above, two from the inner region, two from the outer region and one from the mid-region. The cells were classified as (a) normal epithelia (Fig. 1), (b) goblet (Fig. 1), (c) Necrotic (Fig. 2) and (d) apoptotic (Fig. 3). Goblet cells are recognisable in the epithelium by their basophilic cytoplasm, their basally located nuclei, and the accumulation of mucus secretary granules that fill and distend their apex to give the cells their characteristic goblet shape (Weiss, 1988). Cells were classified as undergoing necrosis if they exhib-
282
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289
2.4. The PAS (Periodic-Acid Schiff ) stain for glycogen deposits
Fig. 3. Rainbow trout epithelial skin cell culture showing apoptopic cells (A). ( ×40 Magnification).
Cultures were fixed in 80% alcohol at 4°C for 30 min for glycogen identification. The samples were rinsed repeatedly in distilled water and covered in 1% periodic acid for 5 min. The cultures were then rinsed in distilled water and covered in Schiffs reagent (Sigma) for 10 min. The cultures were rinsed in several changes of distilled water then placed in running tap water for 10 min to aid colour development. The cell nuclei were counterstained using Mayer’s haematoxylin for 1 min and washed in hot flowing running tap water for 5 min to enhance blue staining. Coverslips were then attached to stained preparations with Kaiser’s glycerol jelly mounting media. The neutral mucins and glycogen stained magenta pink, the nuclei blue.
2.5. Scoring of PAS Stain for mucus results
Fig. 4. Rainbow trout epithelial cells stained negative for mucus. The goblet mucus cells (G) are stained intensely pink. ( × 20 Magnification).
ited the following features; (a) swelling of the cytoplasm and organelles, with only slight changes in the nucleus or (b) organelle dissolution and rupture of the plasma membranes resulting in leakage of the cellular contents into the extracellular space (Schwartzman and Cidlowski, 1993). Cells were classified as undergoing apoptosis if they showed evidence of two or more of the following, (a) cell volume shrinkage and picnotic nucleus (chromatin condensation), (b) blebbing of the cytoplasm, (c) nuclear fragmentation and (d) development into apoptotic bodies (Kerr and Harmon, 1991; Schwartzman and Cidlowski, 1993; Martin et al., 1994).
The epithelial cell surfaces were classed as being negative, weakly positive and intensely positive to mucus (Figs. 4 and 5). Cells were classed as weakly positive and intensely positive to mucus if they were stained light pink and dark pink, respectively. Negative cells stained blue. The mu-
Fig. 5. Rainbow trout epithelial cell cultures stained weakly positive (W) and intensely positive (I) for mucus. ( × 20 Magnification).
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289
283
Table 1 The percent residual of day 3 growth of rainbow trout primary epithelial cell cultures exposed to prochloraz for 24 and 48 h in serum-containing and serum-free mediaa Dose (mmol/l)
Residual of day 3 growth (%) 24 h
0 1 10 25 50 75
48 h
Serum-containing
Serum-free
Serum-containing
Serum-free
100 98.99 5.2 90.59 6.9 72.69 2.7 31.193.7 7.99 1.5
100 96.0 93.7 93.9 93.5 44.8 92.1 10.2 90.7 0.6 90.5
100 86.6 94.3 91.3 92.8 50.5 94.7 24.3 92.0 8.9 92.2
100 97.9 9 2.6 87.1 91.8 31.8 9 3.3 5.0 9 1.1 0.4 90.4
a The results are expressed as mean 9SEM for n= 6 replicate cultures. The values are based on the overall area of the cell culture inclusive of both epithelial cells and goblet cells.
3.1.1. Primary culture growth data
3.1.1.2. Serum-free medium. The primary cultures were found to exhibit a significantly different response to prochloraz when exposed in serum free media (Table 1). The dose of the pollutant required to kill 50% of the cells was found to be between 10 and 25 mmol/l for both 24 and 48 h exposures. The threshold level of response was found to be 10 mmol/l for both exposure periods. The solvent controls were found to exhibit a significantly less percent residual of day 3 growth (PB 0.009, for 48 h) than that of the cultures exposed to the test substance in serum-containing medium. The percent growth of the controls in serum-free medium was 68.291.0 and 73.89 1.2 for 24 and 48 h exposures, respectively.
3.1.1.1. Serum-containing medium. Prochloraz was found to kill 50% of the cells at a concentration of between 25 and 30 mmol/l after 24 h exposure (Table 1). The solvent control cell cultures were found to exhibit a percent residual of day 3 growth of 73.792.4 and 79.79 1.3 for 24 and 48 h exposure periods, respectively. Prochloraz tested on the primary cell system was found to have an acute lethal dose of 75 mmol/l for all exposure times. The threshold level of response, the level of exposure at which a significant decrease in survival begins to occur, is 10 mmol/l for both exposure times. The survival values are based on the overall area of the culture inclusive of both epithelial cells and goblet cells.
3.1.2. Primary culture cell death type (serum-containing medium) The average number of cells per field in the control cultures was 31.691.2 (Table 2). The amount of cells per field decreased as the chemical treatment dose increased. The number of cells per field in the cultures treated with 75 mmol/l prochloraz was 18.290.4. Table 2 shows that an increase in epithelial necrotic cells was observed with increasing prochloraz exposure in primary cell cultures. Epithelial cells located on the outside edge of the culture underwent necrosis primarily at lower doses. A significant increase (PB 0.0001) in those cells undergoing necrosis was observed at 25 mmol/l (0.1–5.19 1.4 cells per field) exposure
cous cells contained in the cell cultures reacted with intense positivity to PAS staining. The mucous cells were not scored in the results.
2.6. Statistics The unpaired t-test was used to compare the residual growth values and the percentage of cells expressing positivity to mucus. 3. Results
3.1. Prochloraz
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289
284
to the pollutant. Only a slight increase in apoptopic cell death was observed with increasing prochloraz exposure (0– 1.5 90.3 cells per field after 48 h exposure to 75 ml prochloraz).
19.69 2.2. The percent of cells stained weakly positive was 15.59 1.6, the percent stained with intense positivity was 4.29 1.4 (Table 3).
3.2. Copper 3.1.3. Relati6e effect on mucus cells 6ersus epithelial cells (serum-containing medium) A decrease in the number of goblet cells per field (from 7.09 1.7 to 0.99 0.5) was also observed as prochloraz dose increased (Table 2). This decrease was observed uniformly across the explant. The threshold level of this decrease was found to be 25 mmol/l. 3.1.4. PAS Stain for mucus The percent of cells covered with a mucus layer was found to increase significantly (P B 0.0001) with increasing exposure to prochloraz (Table 3). The percent of cells expressing positivity in the DMSO controls was 7.9%. The majority of this positivity was weak staining. The number of cells scored per explant in the controls was 4089 99. The percent of cells stained positive for mucus post exposure with 50 mmol/l prochloraz was
3.2.1. Primary culture growth data 3.2.1.1. Serum-containing medium. Copper was found to kill 50% of the cells, at a concentration of about 400 mmol/l for all exposure durations (Table 4). The solvent controls were found to exhibit a percent residual of day 3 growth of 101.491.4 and 101.295.2 for 24 and 48 h exposure periods, respectively. Copper tested on the primary cell system was found to have an acute lethal dose of greater than 1000 mmol/l for all exposure times. The threshold level of response, the level of exposure at which a significant decrease (PB 0.0001) in survival begins to occur, is 100 mmol/l for all exposure periods. The residual growth values are based on the overall area of the culture inclusive of both epithelial cells and goblet cells.
Table 2 Type of cells per primary epithelial cell culture post 48 h exposure to prochloraz in serum-containing mediuma Dose (mmol/l)
Number of normal epithelial cells/field
Number of goblet cells/field
Number of necrotic epithelial cells/field
Number of apoptotic epithelial cells /field
Control 1 10 25 50 75
24.5 9 2.1 24.8 91.7 21.5 91.6 12.5 9 2.9 5.8 9 2.7 3.1 91.3
7.09 1.7 8.2 9 1.6 5.99 0.8 5.8 9 1.1 1.3 9 1.4 0.99 0.5
0.1 9 0.1 0.1 90.1 1.4 9 0.6 5.1 91.4 11.5 9 2.4 12.7 9 2.4
0 0 0.5 9 0.3 0.6 90.3 2.19 0.7 1.5 90.3
a
Each value represents the mean 9S.E.M. for five (0.55 mm) grids counted in n =6 replicate cultures.
Table 3 The percent number of rainbow trout primary epithelial cells stained positive for mucus post treatment with prochloraz in serum-containing mediuma Dose (mmol/l)
Weak positivity (%)
Intense positivity (%)
Total positivity
0+DMSO 10 25 50
7.3 9 1.3 10.49 2.1 15.69 2.0 15.591.6
0.7 90.3 2.3 9 0.1 3.0 91.0 4.2 91.4
7.9 9 1.5 12.7 9 2.6 18.6 9 2.7 19.6 9 2.2
a
Each figure represents the mean 9S.E.M. for n = 9 cultures.
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289
285
Table 4 The percent residual survival of day 3 growth of rainbow trout primary epithelial cell cultures exposed to prochloraz for 24 and 48 h in serum-containing and serum-free mediaa Dose (mmol/l)
Residual of day 3 growth (%) 24 h
0 1 10 50 100 500 1000
48 h
Serum-containing
Serum-free
Serum-containing
Serum-free
100 101.192.7 102.391.8 96.892.5 93.693.4 35.09 4.1 10.891.5
100 99.49 5.6 95.9 92.4 85.5 94.5 92.49 5.4 77.6 93.6 10.7 91.2
100 101.4 93.4 108.1 92.1 101.9 9 2.8 107.7 9 2.6 38.2 95.2 8.0 9 1.3
100 101.9 9 2.9 98.6 9 4.6 92.3 9 2.8 85.2 9 5.6 67.6 9 4.5 10.3 9 1.2
The results are expressed as mean 9 S.E.M. for n = 6. The values are based on the overall area of the cell culture inclusive of both epithelial cells and goblet cells. a
Table 5 Type of cells per primary epithelial cell culture post 48 h exposure to copper in serum-containing mediuma Dose (mmol/l)
Number of normal epithelial cells/field
Number of goblet cells/field
Number of necrotic epithelial cells/field
Number of apoptotic epithelial cells /field
Control 10 50 100 500 1000
23.2 91.3 23.2 9 1.9 23.1 91.4 20.2 91.1 5.9 91.7 1.7 91.4
7.09 1.7 7.29 1.2 6.29 1.2 6.7 9 1.2 2.19 1.4 1.19 0.6
0.1 9 0.1 0.1 9 0.1 0.2 9 0.1 0.2 90.1 11.6 9 2.1 12.6 9 2.8
0 0 0 0.1 90.1 3.2 90.6 5.8 90.4
a
Each value represents the mean 9S.E.M. for five (0.55 mm) grids counted in n =6 replicate cultures.
3.2.1.2. Serum-free medium. The primary cultures were found to exhibit a significantly (P B 0.005) different response to copper when exposed in serum free media (Table 4). The dose of the pollutant required to reduce the growth area by 50% was found to be between 500 and 1000 mmol/l for both 24 and 48 h exposure periods. The threshold level of response was found to be 100 mmol/l for both exposure periods. The solvent control for 24 h were found to exhibit a significantly less percent residual of day 3 growth (P B 0.005) for 24 h, but not (P = 0.9) for 48 h, than that of the cultures exposed to the test substance in serum-containing medium. The percent residual of day 3 growth of the solvent control cultures in serum-free medium was 84.69 2.1 and 100.89 2.4 for 24 and 48 h exposures, respectively.
3.2.2. Primary culture cell death type (serum-containing medium) The average number of cells per field in the control cultures was 30.390.8 (Table 5). The amount of cells per field decreased as the chemical treatment dose increased. The number of cells per field in the cultures treated with 1000 mmol/l copper was 21.290.4. Table 4 shows that an increase in epithelial necrotic cells and apoptopic cells was observed with increasing copper exposure in primary cell cultures. Epithelial cells located on the outside edge of the culture underwent necrosis primarily at lower doses. A significant increase (PB 0.0001) in those cells undergoing necrosis was observed at 500 mmol/l (0.1–12.69 2.8 cells per field) exposure to the pollutant. A slightly less, but also significant, in-
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289
286
crease (P B0.0001) in apoptopic cell death was observed with increasing copper exposure (0– 5.8 90.4 cells per field after 48 h exposure to 1000 ml copper).
3.2.3. Relati6e effect on mucus cells 6ersus epithelial cells (serum-containing medium) A decrease in the number of goblet cells per field from 7.09 1.7 to 1.190.6) was also observed as copper dose increased (Table 5). This decrease was observed uniformly across the explant. The threshold level of this decrease was found to be 100 mmol/l. As observed in the cultures exposed to the other test substances, the goblet cells release their secretory contents onto the culture surface. The percentage of epithelial cells, per field, in the cultures was found to increase by 17.41% for 48 h exposure as the dose of pollutant increased. 3.2.4. PAS stain for mucus The percent of cells covered with a mucus layer was found to increase significantly (P B 0.0001) with increasing exposure to copper (Table 6). The percent of cells expressing positivity in the controls treated with water was 7.0%. The majority of this positivity was weak staining. The number of cells scored per explant in the controls was 4099 43. The percent of cells stained positive for mucus post exposure with 700 mmol/l copper was 17.99 3.0. The percent of cells stained weakly positive was 9.09 1.7, the percent stained with intense positivity was 8.89 2.6 (Table 6).
4. Discussion The toxicity of copper to the primary cell culture system was found to be approximately 10 fold less than that of the organic chemical studied. This is as one might expect as organic chemicals have been found in the past to be more toxic to fish cells than heavy metals (Babich et al., 1986). The toxicity of copper was observed to be less toxic to the primary epithelial cultures than that of cadmium, observed by Lyons Alacantara et al. (1996), and more toxic than that of nickel, observed by McSweeney (1998). Babich et al. (1986) observed a similar relationship between these metals when testing the in vitro cytotoxicity of metals to bluegill BF-2 cells. The toxicity curves of prochloraz and copper were similar to that obtained in Dowling and Mothersill (1999). An element that must not be neglected in any bioassay system, is the composition of the medium. The influence of medium composition on toxicity testing using the rainbow trout primary skin epithelial cell culture system was studied by exposing the cultures to the test substances in both serum-free and serum-containing medium. The process of changing the media to serum-free medium appeared to kill up to 20% of the control samples. Prochloraz proved to be more toxic when tested in serum-free medium (Table 1). A similar result was observed by Bertheussen et al. (1997) when testing the toxicity of the pesticides carbofuran, cypermethrin, lindane, glycophosphate and 2,4-D to the IE6 cell line. Lyons Ala-
Table 6 The percent number of rainbow trout primary epithelial cells stained positive for mucus post treatment with copper in serum-containing mediuma Dose (mmol/l)
Weak positivity (%)
Intense positivity (%)
Total positivity
0+Water 100 400 700
6.69 1.0 6.99 1.0 10.49 0.8 9.09 1.7
0.4 90.2 5.2 91.9 3.6 91.4 8.8 9 2.6
7.0 9 1.2 12.1 9 1.8 14.0 9 1.6 17.99 3.0
a
Each figure represents the mean 9SE.M. for n= 9 cultures.
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289
cantara et al. (1996) observed that both human primary urothelium cultures and fish skin primary epithelial cell cultures were more sensitive to cadmium in serum-free medium. Copper, in contrast, was found to be less toxic when added to the test cultures in serum-free medium compared with addition in serum-containing medium (Table 3). An important parameter of the cellular uptake and subsequent cytotoxicity of metal ions such as Cu++, is their binding to extracellular biological components and the resulting influence on their ability to interact with the cell membrane, enter the cell and exert intracellular cytotoxic effects. Whilst it appears that the serum components are preventing prochloraz from entering the cell, the serum components appear to be increasing the uptake of copper into the cell. Webb and Weinzierl (1972) suggested that nickel ions enter the cell after binding to low molecular weight serum components that actually serve to transport metallic ions into the cell. It appears that something similar may be occurring in the experiments carried out in this study. As exposure levels to the test substances increased, the quantity of goblet cells decreased. Goblet cells produce mucus. Mucus production is an established defence mechanism of epithelial surfaces against pathogens and pollutants (Shephard, 1994). The goblet cells release their secretary contents onto the culture surface forming a thick slimy protective coating in an attempt to prevent deleterious effects of the pollutant medium (Roy, 1988). It was observed using the PAS stain that the goblet cells were decreasing in number because they are exocytosing their contents. A similar observation was made by Dowling and Mothersill (1999) when studying the effects of nonoxynol on primary skin epithelial cell cultures. The results of this study demonstrate that in vitro rainbow trout primary epithelial cell cultures maintain the essential traits of the in vivo response of the skin to chemical stresses. The goblet cells contained in the epithelial cell cultures are retaining their in vivo functions in vitro. A decrease in goblet number and density in rainbow trout primary skin culture could be an important marker of toxicity of aquatic pollutants. An increase in necrotic epithelial cells was observed with increasing dose exposure of prochloraz
287
in serum containing media. The number of apoptotic epithelial cells remained extremely low and constant with dose exposure. These results suggest that prochloraz kills rainbow trout skin epithelial by the process of necrosis. Other organic and inorganic chemicals have been shown to induce similar responses in fish (Wester et al., 1988; Dietrich and Schlatter, 1989). In contrast, copper was found to kill rainbow trout primary skin epithelial by both necrosis and apoptosis in a ratio of approximately 2:1. Pelgrom (1995) observed that epithelial cellular responses of teleosts to Cu++ exposure include both increased necrosis and apoptosis of chloride cells and pavement cells. Carp exposed to 1.6 mmol/l Cu++ for 7 days were observed to show degenerative (apoptopic and necrotic) pavement cells in the skin (Iger et al., 1994). Other in vivo studies have shown that Tilapia exposed to Cu++ have a high incidence of apoptosis in the epidermis (Pelgrom, 1995). Copper shows a high specific binding to DNA producing DNA strand breaks in several biological systems (Sagripanti et al., 1991). Copper has been shown to act as a catalyst in the formation of reactive oxygen species (Chan et al., 1982). Evidence has shown that oxygen radicals may have a role to play in the initiation of apoptosis (Buttke and Sandstorm, 1994). The results of this study add further evidence to support that primary cell cultures are a very appropriate model for toxicity risk assessment. If cultured under the appropriate conditions the cells retain their full, in vivo, characteristics for up to 14 days, thus allowing sufficient time for investigation of the immediate in vitro effects of toxicants. The results of this study also clearly demonstrate that one should take into account the composition of the media, and the binding properties of the substance to be screened, when planning toxicity tests using this cell system. This is especially the case when screening compounds in relation to each other. The use of short-term cytotoxicity assays for the initial screening of chemicals not only aids in establishing priorities for the selection of chemicals that should be tested further in vivo, but also decreases the time in which potential toxicants can be evaluated (Babich et al., 1986). The primary cell culture system used in this study is an extremely
288
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289
useful tool for such assays for the reasons earlier stated. Work carried out in our group with this system has shown that changes in glycogen deposits, stress proteins and oncogene expression could also be studied using this system (Lyons Alacantara et al., 1996; McSweeney, 1998). The primary cell culture system could also be utilised for different areas of study. These include multiple assays/investigations on homologous cultures derived from in vivo exposure, which may result in a major step in the solving of the problem of comparing in vitro effects with that of in vivo. The cultures can also be infected with viruses and as such provide excellent opportunities for research into fish diseases.
Acknowledgements This work was supported by the European Communities Environmental Program: Diagnostic Ecotoxicology; Cell-Based Methodology to develop Markers for Early, Sublethal Effects Assessment: Contract ENVA-CT96-0223.
References Babich,H.,Borenfreund,E.,1987a.Invitrocytotoxicityoforganic pollutants to bluegill sunfish (BF-2) cells. Environ. Res. 42, 229 – 237. Babich, H., Borenfreund, E., 1987b. Fathead minnow FHM cells for use in in vitro cytotoxicity assays of aquatic pollutants. Ecotoxicol. Environ. Saf. 14, 78 –87. Babich, H., Puerner, J.A., Borenfreund, E., 1986. In vitro cytotoxicity of metals to Bluegill (BF-2) cells. Arch. Environ. Contam. Toxicol. 15, 31 –37. Bertheussen, K., Youseff, M.I., Figenschau, Y., 1997. A new sensitivecellculturetestfortheassessmentofpesticidetoxicity. J. Environ. Sci. Health B32, 195 – 211. Birge, W.J., Black, J.A., 19729. In: Nriagra, J.O. (Ed.), Copper in the Environment. Wiley-Interscience, New York, pp. 373 – 379. Bloom, W., Fawcett, D.W., 1970. A Textbook of Histology. WB Saunders, London. Buttke,T.M.,Sandstorm,P.A.,1994.Oxidativestressasamediator of apoptosis. Immunol. Today 15, 7 –10. Chan, P.C., Peller, O.G., Kesner, L., 1982. Copper(II)-catalyses lipid peroxidation in liposomes and erythrocyte membrane. Lipids 17, 331 – 337.
Dietrich, D., Schlatter, C., 1989. Auminium toxicity to rainbow trout at a low pH. Aquat. Toxicol. 15, 197 – 212. Dowling,K.,Mothersill,C.,1999.Theuseofrainbowtroutprimary epidermal cell cultures as an alternative to immortalised cell lines in toxicity assessment. A study with nonoxynol. Environ. Toxicol. Chem. 18 (12), 2846 – 2850. Hellawel, J.M., 1988. Toxic substances in rivers and streams. Environ. Pollut. 50, 61 – 85. Iger, Y., Lock, R.A.C., Jenner, H.A., Wendelaar Bonga, S.E., 1994. Cellular responses in the skin of carp (Cyprinus carpio) exposed to copper. Aquat. Toxicol. 29, 49 – 64. Kerr, J.F.R., Harmon, B.V., 1991. Definition and Incidence of Apoptosis. An Historical Perspective. Cold Spring Harbour Laboratory Press, New York. Langdon, J.S., 1983. Environmental management factors in diseases of fish. Fish Dis.: Proc. 106, 329 – 353. Lyons Alacantara, M., Tarazona, J.V., Mothersill, C., 1996. The differential effects of cadmium exposure on the growth and survival of primary and established cells from fish and mammals. Cell Biol. Toxicol. 12, 29 – 38. Maage, A., Sveir, H., Julshamn, K., 1989. A comparison of growth rateandtraceelementaccumulationinAtlanticSalmon(Salmo salar) fry fed four different commercial diets. Aquaculture 79, 267. Martin, S.J., Green, D.R., Cotter, T., 1994. Dicing with death: dissecting the components of the apoptosis machinery. TIBS 19, 26 – 30 Reviews. McSweeney, J.A., 1998. Investigation of the effects of Nockle on the growth and survival of primary cells and immortalised celllinesofhumanandteleostcelllines.PhDThesis,University of Dublin, Ireland. Mothersill, C., Lyng, F., Lyons, M., Cottell, D., 1995. Growth and differentiation of epidermal cells from the Rainbow Trout established as explants and maintained in various media. J. Fish Biol. 46, 1011 – 1025. Nowak, B.F., 1997. Biological significance of xenobiotic residues in fish. TEN 4, 148 – 154. Pelgrom, S.M.G.J., 1995. Interaction between copper and cadmium in fish: metal accumulation, physiology and endocrine regulation. PhD Thesis, University of Nimegen, The Netherlands. Rievere, J.L., Leroux, P., Bach, J., Gredt, M., 1984. Effects of some ergosterol biosynthesis inhibiting fungicides on sterols and cytochrome P-450 from the Japanese quail Coturnix coturnix. Pesticide Sci. 15, 317 – 323. Roy, D., 1988. Statistical analysis of anionic detergent induced changes in the goblet mucus cells of the opercular epidermis and gill epithelium of Rita rita (Ham)(Bargradal: pisces). Ecotoxicol. Environ. Saf. 15, 260 – 271. Sagripanti, J.L., Goering, P.L., Lamanna, A., 1991. Interaction of copper with DNA antagonism by other metals. Toxicol. Appl. Pharmacol. 110, 477 – 485. Schaaper, R.M., Koplitz, R.M., Theshelashvili, L.K., 1987. Metal-inducedlethalityandmutagenises:possibleroleofcupric intermediates. Mutat. Res. 177, 179 – 188. Schwartzman, R.A., Cidlowski, J.A., 1993. Apoptosis; the biochemistryandmolecularbiologyofprogrammedcelldeath. Endocr. Rev. 14, 133 – 151.
K. Dowling, C. Mothersill / Aquatic Toxicology 53 (2001) 279–289 Shephard, K.L., 1994. Functions for fish mucus. Rev. Fish Biol. Fisheries 4, 401 – 429. Snegaroff, J., Bach, J., 1989. Effects of the fungicide prochloraz on xenobiotic metabolism in rainbow trout: inhibition in vitro and time course of induction in vivo. Xenobiotica 19, 255– 267. Stein, X., Risso, C., Gnassia-Barelli, M., Romeo, M., Lafaurie, M., 1997. Effect of copper chloride in vitro and in vivo on the hepatic EROD activity in the fish Dicentrarchus labrax. Environ. Toxicol. Chem. 16, 214 – 219.
.
289
Wakerly, S.B., Russell, P.E., 1985. Prochloraz, a decade of development. Fungicides in crop protection. Proc. Bordeaux Mixture Centerary Meeting 2, 257 – 260. Webb, M., Weinzierl, S.N., 1972. Br. J. Cancer 26, 292. Weiss, L., 1988. Cell and Tissue Biology, Sixth edn. Urbam & Schwarzenberg, Baltimore. Wester, P.W., Canton, J.H., Dormans, J.A.M.A., 1988. Pathological effects in freshwater fish Poecilla reticulata (guppy) and Oryzias latipes (meduka) following methyl bromide and sodium bromide exposure. Aquat. Toxicol. 12, 323 – 343.