Gill Reaction to Pollutants from the Tamiš River in Three Freshwater Fish Species, Esox lucius L. 1758, Sander lucioperca (L. 1758) and Silurus glanis L. 1758: A Comparative Study
Summary
The study evaluated the effects of waterborne pollutants from the Tamiš River on gill histology and possible differences in gill reaction patterns between three freshwater fish species, pike Esox lucius L. 1758, pike-perch Sander lucioperca (L. 1758) and wels catfish Silurus glanis L. 1758 from the Tamiš River. Gills from analysed fish species showed moderate to intense histopathological alterations. The most frequent progressive alteration was hyperplasia of epithelium, whereas the most frequent regressive alteration was epithelial lifting. Circulatory disturbances were most often manifested in the form of hyperaemia. During comparative analysis, differences in gill indices, reaction and alteration indices, as well as in gill and filament prevalence between analysed species, were observed. Although all analysed fish species did show both progressive and regressive alterations, there was a significant difference in the level of expression of these reaction patterns. Gill index obtained for pike clearly stands out as the lowest. Wels catfish showed the highest progressive reaction index, significantly higher in comparison with the other two species (P < 0.05), while pike-perch showed the highest regressive reaction index, also significantly higher in comparison with the other species (P < 0.001). These results may implicate species-specific gill reactions and thus present a useful tool for better understanding toxic mechanisms of various pollutants.
Introduction
The impact of water pollution and waterborne pollutants on human health and aquatic organisms has become a great concern and received much attention from researchers over the last few decades. Untreated industrial, agricultural and domestic effluents are the most common causes of pollution in aquatic ecosystems (Koca et al., 2005). These effluents usually contain mixtures of various pollutants potentially hazardous to aquatic organisms (Alberto et al., 2005; Camarago and Martinez, 2007).
Fish gills are in constant contact with the aquatic environment. Due to their delicate structure, the direct contact with the water and multiple important functions, gills are the first organ to be affected by pollutants and therefore are usually used as primary markers for aquatic pollution (Poleksić and Mitrović-Tutundžić, 1994; Heath, 1995; Bernet et al., 1999). They have a very thin epithelium and its total area is considerably larger than the total area of skin epithelium (Roberts, 1989), making this organ a suitable site for uptake of xenobiotics. Histopathological analyses of gills indicate that this organ may react non-specifically, and therefore, various pollutants can cause similar reactions and alterations of the gill tissue (Evans, 1987; Heath, 1995; Ferguson, 2006). Current literature predominantly deals with papers describing and quantifying fish reaction to specific pollutants or water pollution in general (Peuranen et al., 1994; Karan et al., 1998; Camarago and Martinez, 2007; Pandey et al., 2008; Liu et al., 2010; Wu et al., 2012; Barja-Fernández et al., 2013). However, only a few studies have involved comparative analysis of reaction of different fish species to waterborne pollutants (Coutinho and Gokhale, 2000; Nascimento et al., 2012).
The Tamiš River is located in the eastern part of the Vojvodina Province, The Republic of Serbia. It is the longest river in the Banat region and its basin covers a total of 10 352 km2. It is characterized by a very imbalanced flow as the water flow decreases to a minimum so that the bed is almost dried out in the dry season, while during the wet season, it tends to flood the surrounding area. Anthropogenic activity has had a significant impact on hydrological regime and water and sediment quality of the Tamiš River (Marković et al., 1998). Main sources of pollution are irrigation channels which are not maintained properly, sewage effluents and effluents from fish farms and livestock farms.
The aim of this paper was to assess possible differences in gill reaction patterns of three different freshwater fish species: wels catfish – Silurus glanis L., 1758, pike-perch – Sander lucioperca (L., 1758) and pike – Esox lucius L., 1758, to waterborne pollutants from the Tamiš River. Possible differences may implicate species-specific gill reaction and thus present a useful tool for better understanding toxic mechanisms of various pollutants.
Materials and Methods
Sampling sites
The Tamiš River originates in Romania, enters Serbia at the 222nd river kilometre and flows through Serbia to its confluence into the Danube River at its 340th kilometre. Sampling of fish was conducted along the Tamiš River at three sites (Fig. 1): (1) Sečanj (45°21′28.6″N, 20°46′22.2″E), (2) Banatski Despotovac (45°16′58.6″N, 20°37′51.7″E) and (3) Opovo (45°03′31.6″N, 20°25′00.6″E) sites. Sampling sites were chosen according to their strategic position along the river. The Sečanj site was the first site along the flow of the Tamiš River through Serbia at which the sampling of fish could be conducted. Area surrounding the Sečanj site is an agricultural area with presence of fish ponds and pig farms. Effluents from these farms, unlicensed waste disposal sites and faecal sewage system are considered to be the main polluters at this site (Aleksić, 2010). Banatski Despotovac site is situated a few kilometres upstream from the first lock on the Danube-Tisa-Danube (DTD) Canal. It is considered that the confluence of the DTD canal and construction of locks have lead to establishment of a new water regime of the Tamiš River, which is leading to higher organic production and deterioration of water quality (Marković et al., 1998). Opovo site was one of the most convenient sites for fish sampling in the second half of the river course. Main polluters in this area are also unlicensed waste disposals and sewage system.
Sample collection
Ichthyofauna samples were collected at three sites along the Tamiš River during August to October 2009 with standard electrofishing device and gill nets 37–100 m long, 3–5 m deep and having a mesh diameter 45–100 mm.
The sample included a total of 29 adult specimens of the following fish species: wels catfish – S. glanis L., 1758, pike-perch – S. lucioperca (L., 1758) and pike – E. lucius L., 1758 (Table 1). Sex determination was conducted by macroscopic observation of the gonads. Wels catfish individuals were pooled for further analysis due to small distance between the sampling sites (Fig. 1).
| Site | Trait | Wels catfish | Pike-perch | Pike | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Males | Females | Total | Males | Females | Total | Males | Females | Total | ||
| Site 1 | N | – | 4 | 4 | 6 | 2 | 8 | – | – | – |
| TL | – | 511.2 ± 131.8 | 511.2 ± 131.8 | 507.7 ± 68.1 | 377.0 ± 104.6 | 475.0 ± 92.4 | – | – | – | |
| W | – | 1064.5 ± 876.1 | 1064.5 ± 876.1 | 1013.3 ± 416.1 | 372.5 ± 286.4 | 853.1 ± 472.6 | – | – | – | |
| Site 2 | N | 3 | 4 | 7 | – | – | – | – | – | – |
| TL | 640.7 ± 19.0 | 632.7 ± 22.2 | 636.1 ± 19.6 | – | – | – | – | – | – | |
| W | 1510.7 ± 106.5 | 1552.7 ± 204.5 | 1534.7 ± 158.7 | – | – | – | – | – | – | |
| Site 3 | N | – | – | – | – | – | – | 8 | 2 | 10 |
| TL | – | – | – | – | – | – | 348.0 ± 57.3 | 283.0 ± 15.6 | 335.0 ± 57.7 | |
| W | – | – | – | – | – | – | 305.7 ± 190.6 | 155.0 ± 36.8 | 275.6 ± 180.1 | |
| Total | N | 3 | 8 | 11 | 6 | 2 | 8 | 8 | 2 | 10 |
| TL | 640.7 ± 19.0 | 572.0 ± 109.0 | 590.7 ± 97.0 | 507.7 ± 68.1 | 377.0 ± 104.6 | 475.0 ± 92.4 | 348.0 ± 57.3 | 283.0 ± 15.6 | 335.0 ± 57.7 | |
| W | 1510.7 ± 106.5 | 1038.6 ± 644.2 | 1363.7 ± 549.2 | 1013.3 ± 416.1 | 372.5 ± 286.4 | 853.1 ± 472.6 | 305.7 ± 190.6 | 155.0 ± 36.8 | 275.6 ± 180.1 | |
- Site 1, Sečanj; Site 2, Banatski Despotovac; Site 3, Opovo; N, number of individuals; TL, total length; W, weight.
Histology methods
The specimens were sacrificed with a quick blow to the head. Immediately after the fish death, the second gill arch from the left side of every fish was sampled and fixed in 4% formaldehyde. Fixed samples were decalcified, dehydrated in graded ethanol series, cleared in xylene and embedded into paraffin blocks. One block per individual was prepared. A total of nine sections (5 μm thin) were cut and placed onto three slides (each slide containing three sections). Slides were stained using standard haematoxylin and eosin (H&E) technique. One randomly selected section per slide was examined, and the mean score value (from three slides) for every alteration was used in further analysis. Sections were examined by Primo Star light microscope (Carl Zeiss, Heidenheim, Germany) and photographed by AxioCam MRc 5 digital camera (Carl Zeiss). All sections were coded and examined without the previous knowledge of the sampling site, fish species or sex.
Histological assessment of gill alterations
Ten randomly selected gill filaments per section were analysed. For description and classification of histological alterations, a slightly modified version of the protocol proposed by Bernet et al. (1999) was used (Table 2), with alteration, reaction and gill indices being calculated according to the same authors. In brief, pathological changes were classified into three reaction patterns: progressive changes, regressive changes and circulatory disturbances. The relevance and pathological importance of each alteration, that is, how it affects the organs functioning, was expressed by an importance factor ranging from 1 (minimal importance, that is, alteration is reversible when exposure to the irritants ends) to 3 (marked importance, that is, alteration is irreversible). A score value ranging from 0 (unchanged) to 6 (severe occurrence) was determined based on the extent of the alteration. Alteration index was calculated by multiplying the importance factor and score value for a given alteration. Reaction index presents the sum of alteration indices of a specific reaction pattern (i.e. circulatory disturbance reaction index presents the sum of hyperaemia, haemorrhage and telangiectasia alteration indices), while the organ index, in this case gill index, presents the sum of all alteration indices and represents the degree of damage.
| Reaction pattern | Alteration | Importance factor |
|---|---|---|
| Progressive changes | Epithelial hypertrophy | 1 |
| Epithelial hyperplasia | 2 | |
| Mucous cell hyperplasia | 2 | |
| Chloride cell hyperplasia | 2 | |
| Chloride cell hypertrophy | 1 | |
| Fusion of secondary lamellae | 1 | |
| Regressive changes | Atrophy | 2 |
| Epithelial lifting | 1 | |
| Necrosis | 3 | |
| Other structural alterationsa | 1 | |
| Circulatory disturbances | Hyperaemia | 1 |
| Haemorrhage | 1 | |
| Telangiectasia | 1 |
- a Other structural alterations include clubbing of distal parts of the lamellae, curving and branching of lamellae, rupture of epithelium; cited from Bernet et al. (1999), modified.
Filament prevalence (FP) presents the proportion of filaments (of 10 analysed filaments per slide) showing a specific type of alteration, while gill prevalence (GP) denotes the proportion of individual gills (of the total number of analysed gills, that is, individuals per species) showing a specific type of alteration.
Water quality data
Results of water quality studies of the Tamiš River for 2009 were taken from the Study of physical, chemical and biological status of the Tamiš river (Teodorović, 2010) and Hydrological yearbook of Republic of Serbia (RHSS, 2010). Water quality data are presented as mean values ± SE of parameter concentrations which were calculated from bimonthly analysis of water quality conducted by Republic Hydrometeorological Service of Serbia (RHSS) (Jaša Tomić station – 45°25′54.4″N, 20°51′29.3″E) (RHSS, 2010) and from two water quality analyses conducted in August and October 2009 (Jaša Tomić station and Jabuka station – 44°55′21.6″N, 20°36′50.5″E) (Dalmacija et al., 2010) (Table 3). Jaša Tomić station is located near the entry of the Tamiš River into Serbia (Fig. 1). It is a regular checkpoint for monitoring water quality of this river as it enables to assess the amount of pollution entering from Romania. Jabuka station is situated a few kilometres upstream from the confluence of the Tamiš and Danube rivers.
| Parameter/Substance | RHSS (2010) | Dalmacija et al. (2010) | |
|---|---|---|---|
| Jasa Tomić | Jasa Tomić | Jabuka | |
| Water temperature (°C) | 14.39 ± 2.59 | NA | NA |
| Alcalinity (mm) | 1.54 ± 0.1 | NA | NA |
| pH | 7.79 ± 0.065 | 7.55 ± 0.05 | 7.35 ± 0.05 |
| Electroconductivity (μS/cm) | 227.82 ± 25.51 | NA | NA |
| O2 (mg/l) | 9.55 ± 0.62 | 6.85 ± 0.35 | 5.85 ± 0.85 |
| Ammonia (mg/l) | 0.07 ± 0.01 | 0.075 ± 0.025 | 0.075 ± 0.025 |
| Orthophosphates (mg/l) | 0.03 ± 0.004 | 2.57 ± 0.16 | 2.65 ± 0.15 |
| Cl (mg/l) | 8.91 ± 1.14 | 5.5 ± 0.5 | 28.75 ± 1.25 |
| Fe (mg/l) | 1.35 ± 0.31 | NA | NA |
| Mn (mg/l) | 0.08 ± 0.02 | NA | NA |
| Zn (μg/l) | 38.98 ± 4.89 | NA | NA |
| Cu (μg/l) | 19.08 ± 3.85 | NA | NA |
| Cr (μg/l) | 2.38 ± 0.38 | NA | NA |
| Pb (μg/l) | 3.77 ± 1.26 | 20.25 ± 19.75 | 7.25 ± 6.75 |
| Cd (μg/l) | 0.77 ± 0.58 | 0.95 ± 0.45 | 0.75 ± 0.25 |
| Hg (μg/l) | <0.1 | 0.5 ± 0.00 | 0.5 ± 0.00 |
- Data are presented as mean ± SE. Values from the first column were calculated from bimonthly analysis of water quality conducted by Republic Hydrometeorological Service of Serbia (RHSS, 2010), while values from the second and third column were calculated from two water quality analysis conducted in August and October 2009 at two sites (Dalmacija et al., 2010). Parameters that were not assessed in the analyses are marked with ‘NA’.
Statistical analysis
The data on specific histopathological alteration measurements are presented as mean ± standard error. Shapiro–Wilk test was applied to determine the normality of data distribution. One-way anova followed by Tukey post hoc test was used to determine differences in the gill, reaction and alteration indices, as well as differences in GP and FP between species and differences in gill indices between sexes. Pearson's correlation index was used to determine the relationship between progressive and regressive reaction patterns. Significance was accepted when P < 0.05. All tests were carried out using statistica 10.0 software (StatSoft Inc., Tulsa, OK, USA).
Results
Gills from analysed fish species showed moderate to intense histopathological alterations. Hyperplasia of epithelium (Fig. 2a) was the most pronounced progressive alteration in wels catfish and pike-perch, while hypertrophy of epithelium (Fig. 2c) was the most expressed progressive alteration in pike (Table 4). Epithelial lifting (Fig. 2d) was the most pronounced regressive alteration in wels catfish and pike-perch. Although necrosis was the highest in pike-perch (Fig. 2e), it was the most expressed regressive alteration in pike (Table 4). Circulatory disturbances were most often manifested in the form of hyperaemia (Fig. 2f) in all analysed species. As for other, less frequent alterations observed, partial fusions of lamellae (Fig. 2a) and complete fusions of lamellae (Fig. 2b) were detected, as well as lamellar atrophy, clubbing of distal parts of the lamellae (Fig. 2c), folding (Fig. 2d) and disorganization (deformation) of lamellae, epithelial rupture, haemorrhage and telangiectasia (aneurysm) (Fig. 2f).
| Reaction pattern | Alteration | Wels catfish | Pike-perch | Pike | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| AI | FP (%) | GP (%) | AI | FP (%) | GP (%) | AI | FP (%) | GP (%) | ||
| Progressive changes | Epithelial hypertrophy | 1.94 ± 0.34a | 73a | 100a | 1.32 ± 0.27a | 80a | 100a | 1.99 ± 0.28a | 90a | 100a |
| Epithelial hyperplasia | 5.51 ± 1.23a | 69a | 82a | 1.40 ± 0.37b | 40a | 75a | 1.32 ± 0.51b | 31a | 70a | |
| Mucous cell hyperplasia | 0.60 ± 0.40a | 13a | 27a | 0.20 ± 0.13a | 5a | 25a | 0.40 ± 0.27a | 8a | 20a | |
| Chloride cell hyperplasia | 0.47 ± 0.31a | 13a | 36a | 0.00a | 0a | 0a | 0.08 ± 0.08a | 2a | 10a | |
| Chloride cell hypertrophy | 0.70 ± 0.25a | 38a | 64a | 0.00b | 0b | 0b | 0.04 ± 0.04b | 2b | 10b | |
| Fusion of secondary lamellae | 1.53 ± 0.49a | 50a | 73a | 0.00b | 0b | 0b | 0.29 ± 0.20b | 10b | 30a,b | |
| Regressive changes | Atrophy | 0.74 ± 0.38a | 18a | 45a | 1.35 ± 0.39a | 30a | 75b | 0.28 ± 0.17a | 6a | 30a |
| Epithelial lifting | 2.54 ± 0.30a | 85a | 100a | 3.55 ± 0.45a | 85a | 100a | 0.30 ± 0.08b | 29b | 80a | |
| Necrosis | 0.00a | 0a | 0a | 2.47 ± 0.56b | 40b | 100b | 1.69 ± 0.93a,b | 23b | 50c | |
| Other structural alterations | 0.42 ± 0.10a | 29a | 82a | 1.52 ± 0.06b | 75b | 100a | 0.95 ± 0.27a,b | 42a | 80a | |
| Circulatory disturbances | Hyperaemia | 1.11 ± 0.52a | 35a | 36a | 0.57 ± 0.24a | 35a | 50a | 0.48 ± 0.16a | 27a | 70a |
| Haemorrhage | 0.19 ± 0.08a | 15a | 45a | 0.00a,b | 0b | 0b | 0.00b | 0b | 0b | |
| Telangiectasia | 0.14 ± 0.05a | 16a | 73a | 0.06 ± 0.03a,b | 0b | 50a | 0.00b | 0b | 0b | |
| Gill index | 15.89 ± 2.10a | 12.46 ± 1.28a,b | 7.85 ± 1.52b | |||||||
- AI, alteration index; FP, filament prevalence; GP, gill prevalence.
- Data are presented as mean ± SE.
- Within rows (alterations), for a given parameter, different letters in the superscript indicate a significant statistical difference between species (Tukey post hoc test, P < 0.05).
During the comparative analysis, differences in gill and alteration indices, as well as in GP and FP, were observed between the analysed species (Table 4). A statistically significant difference between gill indices for pike and wels catfish (P = 0.0069) has been observed. The analysis also revealed great differences in reaction patterns between the fish species (Fig. 3). Wels catfish clearly showed higher, significantly different progressive reaction index as compared to pike-perch and pike (P = 0.0096 and P = 0.020, respectively). On the other hand, pike-perch showed higher, significantly different regressive reaction index as compared to wels catfish and pike (P = 0.00027 and P = 0.00018, respectively). Circulatory reaction indices were the lowest in all fish species as compared to the progressive and regressive reaction patterns (P < 0.05) (Fig. 3). Statistically significant difference in total gill indices between sexes was observed only in wels catfish (P = 0.011), while there were no statistically significant differences in pike-perch and pike (P = 0.439 and P = 0.573, respectively). Correlation analysis was also conducted to assess the relationship between progressive and regressive reaction patterns for the analysed species (Table 5). The analysis showed that the correlation index was statistically significant only in pike-perch (P = 0.037), indicating high correlation between progressive and regressive alterations in this species. Although correlation between progressive and regressive indices for wels catfish and pike-perch together was not statistically significant, it is indicative of a certain relationship: with an increase in the progressive index, the regressive index decreased and vice versa.
| Wels catfish | Pike-perch | Pike | Wels catfish and pike-perch | All species | |||||
|---|---|---|---|---|---|---|---|---|---|
| r | P | r | P | r | P | r | P | r | P |
| 0.156 | 0.646 | 0.737 | 0.037 | 0.342 | 0.333 | −0.36 | 0.13 | −0.11 | 0.57 |
Discussion
The results obtained by histopathological analysis of gills in fish from the Tamiš River suggest that, during the sampling period, the water contained pollutants which could induce observed histopathological alterations. Results of previous water quality studies from the Jaša Tomić and Jabuka sites (Table 3) can further help in interpreting histopathological observations by pointing out possible pollutants that could induce observed alterations. When compared with maximum admissible values proposed by Svobodova et al. (1993), the concentrations of pollutants obtained in previously conducted studies (Dalmacija et al., 2010; RHSS, 2010) have shown that values for iron, copper and chlorine were elevated and could possibly induce morphological changes in fish tissue. The large standard error for copper (Dalmacija et al., 2010) is the result of a very high copper concentration in August 2009 for both sites, while its concentration in October 2009 was much lower. As a depletion of copper concentration along the river was observed in August 2009, high concentration at Jaša Tomić site, located at the entrance of Tamiš into The Republic of Serbia, indicates that the plausible source of this substance was in the territory of The Republic of Romania near the Tamiš River. Concentrations for organic pollutants were mostly below detection limits according to the RHSS and therefore are not presented in this study.
Although the observed alterations are not pollutant-specific, they could be related to previously published results concerning gill alterations caused by experimental exposure to certain pollutants. Iron concentrations close to the values observed in the Tamiš River, in experimental conditions, induced gill damage in the form of hypertrophy, hyperplasia of epithelium, fusions, lifting and rupture of epithelium, haemorrhages and necrosis (Peuranen et al., 1994; Dalzell and Macfarlane, 1999; Pandey et al., 2008; Wu et al., 2012). Epithelial lifting and swelling, hyperplasia with lamellar fusions, hyperaemia, telangiectasia and alterations in chloride cells were common lesions of gills from fish experimentally exposed to copper (Karan et al., 1998; Arellano et al., 1999; Liu et al., 2010). Middaugh et al. (1980) observed that chlorine induced epithelial lifting, fusion of lamellae and telangiectasia. Moreover, higher accumulation levels of mercury and cadmium in Carassius gibelio (Bloch, 1782) muscle sampled during the same period as the present study (Dalmacija et al., 2010) may indicate a long-term exposure of fish to sublethal levels of these metals or biomagnification. The above-mentioned studies corroborate our hypothesis that pollutants in the reported concentrations from the Jaša Tomić site, which were elevated when compared to maximum admissible values proposed by Svobodova et al. (1993), could have induced the observed histopathological alterations.
Comparing the alteration index and GP and FP for a given alteration between the species, certain differences can be noticed (Table 4). In epithelial hyperplasia, no significant differences in GP and FP between the species were detected. In contrast, significant differences were observed in alteration indices. Similar distinction was observed in other structural alterations where there were no significant differences in GP but there were significant differences in alteration indices and FP. While all species displayed these alterations, there were significant differences in the level of expression.
Differences in reaction patterns between species may be induced by various environmental conditions and species-specific responses to pollution. Considering that not all species have been sampled at every sampling site, site-specific conditions may have influenced these differences to some extent, particularly in the case of pike which has been sampled only at the Opovo site. However, in the case of pike-perch and wels catfish, these two species have been sampled at the Sečanj and Banatski Despotovac sites. As these two sites are relatively close (Fig. 1) and without any significant polluters between them, differences in the environment and water quality are most likely negligible. This is why we suspect that species-specific differences are present. Furthermore, the medium negative correlation index between the reaction indices for pike-perch and wels catfish (Table 5) is indicative of a certain relationship and could contribute to our suspicion that observed differences in reaction patterns are most likely species-specific.
The exact reasons for species-specific differences in pathological response to pollutants are not very clear. These differences might be due to differences in life history, habitat, diet and the manner in which species metabolize pollutants (Stehr et al., 1998). For example, the studies on accumulation of metals in fish tissues indicate that pike-perch usually accumulates high amounts of metals (Falandysz et al., 2000; Milošković et al., 2013), while the wels catfish and pike are among the species that accumulate low amounts of metals (Mendil et al., 2005; Erdogrul and Ateş, 2006; Erdogrul and Erbilir, 2007; Milošković et al., 2013). The differences in accumulation might be attributed to detoxification systems, especially metallothioneins. Furthermore, trace metals induce oxidative stress, hence activating the antioxidant systems, in which depletion in activity of superoxide dismutase, catalase, glutathione S-transferase and glutathione in gills is usually noticed (Farombi et al., 2007). It has been observed that some regressive alterations in the form of rupture of epithelium and epithelial necrosis may be attributed to depletion in activity of these enzymes (Wu et al., 2012). Coutinho and Gokhale (2000) report distinct differences in gill reactions between common carp (Cyprinus carpio L., 1758) and Mozambique tilapia [Oreochromis mossambicus (Peters, 1852)] exposed to secondary sewage effluents. Common carp demonstrated extensive necrosis of epithelial, pillar and mucous cells and telangiectasis, while Mozambique tilapia demonstrated dilatations of lamellar capillaries, epithelial lifting and exfoliation of epithelial and pillar cells. In the same study, common carp showed an elevation of oxidative enzymes (isocitrate dehydrogenase and succinic dehydrogenase) and depletion of LDH, while Mozambique tilapia showed depletion of oxidative enzymes and increase in LDH. This study implicates differences in shifts of metabolic pathways between different fish species exposed to same pollutants and strongly corroborates results obtained in the present study. Moreover, the claim of Fijan (2006) that pike-perch is the most sensitive freshwater fish species to environmental hypoxia may imply high overall sensitivity of this species to unfavourable environmental conditions. This could also be observed in the present study as pike-perch had the highest regressive alteration indices. Although epithelial lifting is considered to be a defensive mechanism, highest indices for irreversible alterations, for example necrosis, could point out the higher sensitivity of this species to unfavourable environmental conditions.
It is presumed that epithelial hyperplasia and lifting are the two main defensive mechanisms of fish gills when exposed to xenobiotics. Epithelial hyperplasia (progressive alteration) reduces the respiratory surface, while epithelial lifting (regressive alteration) increases the water–blood barrier. Observed differences in gill reactions in the present study may be contributed by the ecology and bioenergetics of these fish species. Wels catfish is a bottom dweller that is static, mostly concealed in ‘resting places’ for extended periods of time with movements characterized by restricted mobility (Carol et al., 2007), while pike and pike-perch dwell in the water column and are active swimmers and predators (Chapman and Mackay, 1984; Vehanen and Lahti, 2003). It could be presumed that wels catfish has a lower metabolism rate than the two fast swimming fish species, thus consuming less oxygen throughout regular activities. In turn, this may allow the wels catfish to react with hyperplasia and fusions of lamellae (progressive changes), thus considerably reducing its respiratory surface and hindering the uptake of oxygen, while pike-perch has developed a defensive mechanism in the form of epithelial lifting (regressive alteration) which increases the water–blood barrier, effecting the uptake of oxygen to a much lesser extent.
Most studies report that fish gills react non-specifically, implying that various pollutants can cause similar reactions and alterations of the gill tissue, and thus, these gill alterations can be regarded as reflections of generalized stress response (Mallat, 1985; Poleksić and Mitrović-Tutundžić, 1994; Heath, 1995; Ferguson, 2006). However, the present study indicates that it is possible to detect certain differences in gill reaction to pollution between species. Although gills of all analysed fish species did show progressive, regressive and circulatory alterations, there was a significant difference in the level of expression of these reaction patterns. While wels catfish displayed significantly higher progressive alterations, mainly hyperplasia, pike-perch showed significantly higher regressive alterations, mainly epithelial lifting. These results demonstrate that mere description of histopathological alterations is not sufficient and that studies should also include quantification of these alterations. The inclusion of semi-quantitative (present study) or histometric analysis (measurements of specific structures, see Nero et al., 2006) can aid in understanding the severity of alterations and the extent to which xenobiotics affect the tissue. Furthermore, more detailed comparative studies might give a better insight in understanding fish reaction to pollution and toxic mechanisms of various pollutants. Finally, future research should also include studies on fish metabolism and bioenergetics, detoxification systems and gene expression with the aim of better understanding gill reaction to environmental pollution.
Acknowledgements
The research was financially supported by the funds of Ministry of Education and Science of the Republic of Serbia (project number: 143058) and the funds of European Union and the city of Pančevo (project number: 06SER02/03/007-8). Authors want to express their gratitude to the project coordinators prof. Dr. Radmila Kovačević and prof. Dr. Ivana Teodorović, prof. Dr. Vesna Rajković for support in statistical analyses and Dr. Ljiljana Knežević for improving the English. Finally, we would like to thank the anonymous reviewers for their valuable comments.
