Biomarker-Based Assessment of Four Native Fish Species in the Danube River Under Untreated Wastewater Exposure

Abstract

This study assessed the impact of untreated wastewater discharge in the Danube River on four native fish species: barbel (Barbus barbus), vimba bream (Vimba vimba), perch (Perca fluviatilis), and white bream (Blicca bjoerkna). Biomarkers of exposure and effect were evaluated, including metal and metalloid bioaccumulation in gills, liver, and gonads, DNA damage (comet assay), chromosomal abnormalities (micronucleus assay), liver enzyme activities (ALT, AST), and erythrocyte maturation. White bream showed the highest genotoxic damage (TI% = 22.57), particularly in liver tissue, indicating high sensitivity to pollution. Perch had elevated DNA damage in blood (TI% = 22.69) and strong biomarker responses, likely due to its predatory behavior. Barbel displayed notable DNA damage in gills (TI% = 30.67) and liver (TI% = 20.35), aligning with sediment exposure due to its benthic habits. Vimba bream had the lowest responses, possibly reflecting reduced exposure or resilience. Element accumulation varied across tissues and species, with perch showing the highest overall levels. Hepatic enzyme activities (highest values: ALT = 105.69 in barbel; AST = 91.25 in white bream) and changes in erythrocyte profiles supported evidence of physiological stress. Integrated Biomarker Response (IBR) analysis identified white bream as the most sensitive species, followed by perch and barbel. These results emphasize the value of multi-species biomonitoring and the importance of species-specific traits in freshwater ecotoxicology.

1. Introduction

Large river systems represent ecologically and socioeconomically significant ecosystems, delivering a broad spectrum of ecosystem services. These include potable water supply, fishery resources, irrigation water for agriculture, industrial water use, recreational and tourism opportunities, and vital functions as navigational and transport corridors [1]. They are the main recipients of wastewater, which significantly affects their water quality. The Danube is the second longest river in Europe (2857 km in length) and the most international river in the world, as it passes through 19 countries [2]. Over the last two decades, large investment projects have been implemented with the aim of wastewater treatment in the European Union (EU) member states. However, similar improvements in wastewater infrastructure have not occurred in non-EU member states in the middle and lower Danube River Basin, where a significant amount of untreated wastewater is still directly discharged into the Danube River [3]. In Serbia less than 13% of collected municipal wastewater is being treated before its release in the environment. This includes all municipal wastewater discharges from Belgrade (capital city of Serbia; population 1,700,000) [4,5].

Fish play a crucial role as bioindicators in assessing the impact of pollution on ecosystems. It is essential to investigate the effects on native species that are particularly vulnerable to different environmental stressors, as these impacts can lead to a decline in biodiversity [6]. Additionally, examining fish species commonly consumed by humans can reveal potential health risks associated with their consumption [7]. Gills, liver, and blood are important tissues that are often used in monitoring studies where fish are employed as bioindicators [8,9]. Gills represent a primary site of toxicant uptake, as they are in direct and continuous contact with the surrounding aquatic environment. The liver is a key organ for toxicological assessment due to its central role in xenobiotic metabolism and biotransformation processes. Peripheral blood serves as a minimally invasive matrix that enables access to circulating cells and offers insight into the systemic physiological condition of the organism. Given the ecological and toxicological relevance of specific fish organs and tissues, the selection of appropriate bioindicator species is essential for monitoring aquatic pollution. Native species include perch, Perca fluviatilis (L., 1758), vimba bream, Vimba vimba (L., 1758), white bream, Blicca bjoerkna (L., 1758), and barbell, Barbus barbus (L., 1758). These species have different ecological traits and trophic levels—ranging from benthic feeders to predators—and are therefore exposed to contaminants in different ways. Such diversity enables the assessment of species-specific physiological and biochemical responses to pollution, providing a broader and more reliable picture of ecosystem health. Studies have shown that these species display clear physiological and biochemical responses to contaminants, supporting their use in ecotoxicological assessments [10,11,12].

Metals and metalloids, the most dangerous environmental pollutants, are discharged into aquatic ecosystems through wastewater and enter food chains with the ability to bioaccumulate in different tissues of fish, negatively affecting their growth, reproduction, and physiology [13,14]. In addition, they pose a threat to public health [15]. They can compromise the integrity of DNA not only by generating reactive oxygen species but also by interfering with the biological functions of proteins that oversee the cell cycle, DNA maintenance, and apoptosis [16,17].

The alkaline comet assay is a highly sensitive method for assessing the response of living organisms to genotoxic substances [18]. It identifies current DNA damage that can potentially be repaired and can therefore be considered a biomarker for exposure [19]. To assess the impact of negative exposure at the chromosomal level, a micronucleus assay may be applied, which identifies the presence of micronuclei in the cell cytoplasm. A micronucleus is considered an irreparable lesion that results from the improper cleavage of an entire chromosome—aneugenicity—or parts of a chromosome—clastogenicity [20].

Liver transaminases, alanine aminotransferase (ALT) and aspartate aminotransferase (AST), are recognized as sensitive biomarkers of hepatic stress in fish exposed to aquatic pollution. Elevated activities of these enzymes indicate disturbances in liver function caused by contaminants such as metals and organic pollutants [21,22]. Several studies have confirmed their reliability as indicators of toxicant-induced stress and liver damage in fish from polluted rivers [23,24].

The objective of this study is to apply a battery of biomarkers—including metal and metalloid bioaccumulation, DNA integrity (comet assay), chromosomal integrity (micronucleus assay), liver enzyme activities, and erythrocyte maturation—in four autochthonous fish species from different trophic levels. The cumulative responses of these biomarkers across species and tissues are essential for differentiating the most sensitive species, organs, and biomarkers for ecological monitoring. Specifically, the study examines metal accumulation in gills, liver, and gonads; DNA damage in gills, liver, and blood cells; chromosomal abnormalities in erythrocytes; stress enzyme activity in liver; and erythrocyte developmental stages.

2. Materials and Methods

2.1. Sampling of Fish Specimens

All fish were obtained from commercial catches, conducted between 1162 and 1163 river kilometers, of the Danube River (Višnjica sampling site, municipality of Belgrade) (Figure 1). Fishing was performed using drifting gillnets. Sampling was conducted monthly from October to December 2018.

Sampled fish were anesthetized with 2-Phenoxyethanol (Merck, Darmstadt, Germany) at a concentration of 1 mL/L. Total length (TL), in centimeters, and total weight (W), in grams, were measured, and after that, fish were dissected with a plastic laboratory set. Fish tissue and blood samples were taken immediately after removal from the water. A total of 44 specimens were collected, including 9 barbel (Barbus barbus), 16 vimba bream (Vimba vimba), 5 perch (Perca fluviatilis), and 14 white bream (Blicca bjoerkna). For the analyses of metal bioaccumulation, liver enzyme activity, and erythrocyte maturation, all collected specimens were used. For the genotoxicological assays (comet and micronucleus tests), a subset of fish was analyzed, comprising 2 barbel, 6 vimba bream, 3 perch, and 9 white bream.

2.2. Determination of Metal and Metalloid Concentrations

After dissection of the specimens, samples of liver, gills and gonads were placed in separate plastic bags and stored at −20 °C in a freezer until preparation in the laboratory. After thawing in the laboratory, the samples were dried in a desiccator. After drying, the tissue samples were weighed on an analytical balance and placed in Teflon bottles with the addition of 6 mL of 65% H₂O₂ (Suprapur^®, Merck, Darmstadt, Germany) and 4 mL of 30% HNO₃ (Suprapur^®, Merck, Darmstadt, Germany). Digestion was performed according to the following temperature program: temperature increase to 180 °C over 10 min, maintaining the temperature at 180 °C for 20 min and cooling the vessels to room temperature (ETHOS EASY Advanced Microwave Digestion System 230 V/50 Hz, Milestone, Milan, Italy). After digestion, the dissolved samples were diluted with deionized water to a final volume of 25 mL. The prepared solutions were analyzed using ICP-OES. Concentrations of elements (Al, As, B, Ba, Cd, Hg, Pb, Cu, Fe, Mn, Mo, Se and Zn) were assessed by ICP-OES (Spectro Genesis EOP II, Spectro Analytical Instruments GmbH, Kleve, Germany). The following wavelength lines of ICP-OES were used: Al 396.152 nm, As 189.042 nm, B 249.773 nm, Ba 230.424 nm, Cd 214.438 nm, Co 228.616 nm, Cr 267.716 nm, Cu 324.754 nm, Fe 238.204 nm, Hg 184.950 nm, Li 460.289 nm, Mn 294.921 nm, Ni 231.604, Mo 202.095 nm, Pb 220.353 nm, Se 196.090 nm, Sr 460.733 nm, and Zn 213.856 nm. The detection limits for analyzed elements are Al 0.00158 μg/L, As 0.00282 μg/L, B 0.000931 μg/L, Ba 0.000531 μg/L, Cd 0.000132 μg/L, Co 0.00024 μg/L, Cr 0.000366 μg/L, Cu 0.000588 μg/L, Fe 0.000562 μg/L, Hg 0.00553 μg/L, Li 0.042 μg/L, Mn 0.000403 μg/L, Mo 0.000784 μg/L, Ni 0.00114 μg/L, Pb 0.00343 μg/L, Se 0.00197 μg/L, Sr 0.00138 μg/L, and Zn 0.000391 μg/L. The quality of the analytical process was controlled through analysis of BCR-185R reference materials of bovine liver and of IAEA-336 lichen reference material. The concentrations found were within 90–115% of the certified values for all measured elements. Concentrations of all elements are expressed as μg/g dry weight.

To evaluate the overall level of metal contamination in fish tissues, we calculated the Metal Pollution Index (MPI) using the following formula:

$$\mathrm{M}\mathrm{P}\mathrm{I}={({\mathrm{C}}_1\times {\mathrm{C}}_2\times {\mathrm{C}}_3\times \dots {\mathrm{C}}_{\mathrm{n}})}^{1/\mathrm{n}}$$

where C₁, C₂, C₃, …, C_n are the concentrations of individual metals in the sample, while n is the total number of metals considered [25].

2.3. Alkaline Comet Assay

The preparation of the cell suspension and comet assay procedure was performed based on our previous study [11] with minor modifications. After dissection, small liver and gill samples were placed in 3 mL of cold phosphate buffer solution (PBS). Blood was collected directly from the heart using a heparinized needle and syringe, with a drop placed in 1 mL of cold 1× PBS (pH 7.4). All samples were transported to the lab in a dark cooler (3–5 °C) and processed within an hour. The liver and gill tissues were minced in 300 µL of cold 1× PBS with two scalpels, followed by the addition of trypsin (final concentration 0.05%). After 10 min of incubation with shaking at room temperature, trypsin activity was halted by adding fresh PBS to reach a final volume of 10 mL. The supernatant containing single cells (approximately 8 mL) was transferred to a new tube and centrifuged at 2000 rpm for 10 min at 4 °C. The supernatant was discarded, and the cell pellet was resuspended in the remaining 1 mL of 1× PBS. Both cell suspensions and blood samples were further diluted to about 50,000 cells/mL. Prior to the comet assay procedure, cell viability was determined by differential staining with acridine orange/ethidium bromide dyes [26]. Next, 30 µL of the cell suspension was mixed with 70 µL of 1% LMP (Low Melting Point) agarose in microtiter plates and applied in duplicate onto a layer of 1% NMP (Normal Melting Point) agarose on glass slides pre-coated with 1% NMP. After gel solidification, coverslips were removed, and the samples were treated with a cold lysis solution (2.5 M NaCl, 100 mM EDTA, 10 mM Tris, 1% Triton X-100, 10% DMSO for blood samples, pH 10, 4 °C, 16–18 h). Following lysis, a denaturing solution was applied (300 mM NaOH, 1 mM EDTA, pH 13, 4 °C, 20 min), and electrophoresis was performed in the same solution (0.75 V/cm, 300 mA, 4 °C, 20 min). Samples were then neutralized with a neutralizing buffer (0.4 M Tris, pH 7.5, 4 °C, 15 min) and fixed in cold methanol (4 °C, 15 min). Comets were visualized with 20 μL of acridine orange dye (2 μg/mL) under a fluorescent microscope (Leica, DMLS, Wetzlar, Germany) at magnification ×400. DNA damage was quantified using the software Comet Assay IV, version 4.3 (Perceptive Instruments, Bury Saint Edmunds, UK) and reported by the TI% parameter.

2.4. Micronucleus Assay

A micronucleus assay was performed based on our previous study [27], with slight modifications. A drop of blood was placed on a clean glass slide to produce a thin smear, air dried for 1 h, and fixed with cold methanol for 20 min. Micronuclei were visualized with an acridine orange stain (final concentration 2 μg/mL) under a fluorescence microscope (Leica, DMLS, Wetzlar, Germany) at ×400 magnification. For each sample, 6000 randomly selected cells were analyzed, and the number of micronuclei was expressed in parts per thousand (‰).

2.5. Measurement of Hepatic Enzyme Activities (ALT, AST)

For the quantitative in vitro determination of Alanine Aminotransferase (ALT) and Aspartate Aminotransferase (AST) in the blood, Randox assays were used (AL3801, AS3804, Randox Laboratories Ltd., Antrim, UK). The experimental procedure was performed in accordance with the manufacturer’s instructions, and the obtained results were presented as units per liter of blood (U/L).

2.6. Determination of Erythrocyte Count and Maturation

Blood smears were prepared on microscopic slides, stained with Bio-Diff kit (Bio Optica, Milan, Italy) and observed under DMRB photomicroscope (Leica, Germany). Random fields of view were photographed on microscopic slides, and random erythrocytes within those fields were analyzed using the image analysis software (ImageJ, version 1.54n), which automatically calculated area, perimeter, and length of 50 erythrocytes for each individual. The shape factor of erythrocytes was calculated as:

$$\mathrm{Shape} \mathrm{factor}=(4\times \mathrm{\pi }\times {\mathrm{C}}_{\mathrm{A}})/{\mathrm{C}}_{\mathrm{L}}$$

where C_A is the cell area and C_L is the cell length. Development stages (immature, intermediary, and mature) were categorized based on the calculated shape factor [28], with immature erythrocytes having shape factor > 0.94, mature erythrocytes having shape factor < 0.91, and intermediary erythrocytes having shape factor in between these values.

2.7. Calculation of Integrated Biomarker Response (IBR Index)

To efficiently summarize the responses of the monitored biomarkers for each species and determine which species is the most sensitive, we applied the Integrated Biomarker Response (IBR). This approach was initially described by Beliaeff and Burgeot [29]. To conduct IBR analysis we used eight monitored biomarkers: metal pollution index in liver and gills (MPI liver, MPI gills), DNA damage measured by comet assay and expressed as TI (%) (TI blood, TI liver, TI gills), frequency of micronuclei in fish erythrocytes (MN), and liver enzymes (AST, ALT). The initial step in the calculation involved standardizing the data X to derive the Y value using the formula:

$$\mathrm{Y}=(\mathrm{X}-\mathrm{m})/\mathrm{s}$$

Here, X represents the individual biomarker data, m stands for the mean value, and s indicates the standard deviation.

Following this, we calculated the Z value. When there is inhibition of the biological response, we defined Z as Z = −Y; conversely, in the case of activation, Z equals Z = Y. The minimum value for each biomarker was then identified and added to the Z value, resulting in:

$$\mathrm{S}=\mathrm{Z}+\left|\mathrm{M}\mathrm{i}\mathrm{n}\right|$$

Using these S values, we generated radial diagrams to illustrate the biomarker results. To determine the total surface area of the specified field, we first calculated the areas of the triangles formed by two successive biomarker coordinates (S_i and S_i+1) utilizing the following formula:

$${\mathrm{A}}_{\mathrm{i}}={\mathrm{S}}_{\mathrm{i}}\times {\mathrm{S}}_{\mathrm{i}+1}\times \sin \left({\displaystyle \frac{2\mathrm{\pi }}{\mathrm{k}}}\right)/2$$

Here, k is the total number of biomarkers. Lastly, the IBR values for each species were calculated as a sum of A_i values [30]:

$$\mathrm{I}\mathrm{B}\mathrm{R}=\sum _{\mathrm{i}=1}^{\mathrm{k}}{\mathrm{A}}_{\mathrm{i}}$$

2.8. Statistical Analyses

Statistical analyses were performed in IBM SPSS (version 25.0). Normality of data for element concentrations was evaluated by Shapiro–Wilk test. In case of normal distribution, data was analyzed using ANOVA test, with subsequent Tukey HSD post hoc test. In case of a distribution that did not follow a normal distribution, Kruskal–Wallis H test was performed, with subsequent Mann–Whitney U tests. Pearson’s correlation analysis was performed to analyze the correlation between the element concentrations in analyzed tissues and total length (TL) and weight (W) of fish (p < 0.05). For comet assay, data distribution normality was assessed using the Kolmogorov–Smirnov test. Differences between species in the comet assay were analyzed with the Mann–Whitney U test, applying Bonferroni correction for multiple comparisons (p < 0.008). For the MN assay, significant differences were evaluated using Kruskal–Wallis ANOVA (p < 0.05). For determining differences in MPI, the homogeneity of variances was assessed, and either ANOVA or Welch’s ANOVA was applied as appropriate, followed by Tukey’s HSD or Games–Howell post hoc test, respectively. ALT and AST enzymes activities were analyzed using ANOVA test.

3. Results

3.1. Bioaccumulation of Metals and Metalloids in Tissues

All of the toxic elements (As, Cd, Hg, and Pb) and Al, B and Ba were below the detection limit in all samples.

In barbel, vimba bream, and perch, all of the analyzed trace elements were found in the gill tissue. Perch was the only sampled species in which every tissue had a detectable amount of trace elements. Liver tissue, in all species, was characterized by the highest concentrations of Cu and lowest concentrations of Mn (

Table 1. Element concentrations (mean ± SD; μg/g dw) in liver, gills, and gonads of barbell (Barbus barbus).

Species	Tissue	Cu	Fe	Mn	Mo	Se	Zn
Barbel	Liver	22.67 ± 15.6 a	97.77 ± 70.0 a	0.36 ± 0.4 b	nd	nd	44.28 ± 21.7 a
	Gills	1.75 ± 1.0 b	133.68 ± 25.4 a	7.76 ± 3.5 a	1.00 ± 0.5	5.29 ± 2.7	50.38 ± 5.8 a
	Gonads	nd	2.82 ± 4.4 b	nd	nd	nd	8.82 ± 23.0 b

^a,b Statistical differences between tissues for each element are indicated by letters (a > b).

,

Table 2. Element concentrations (mean ± SD; μg/g dw) in liver, gills, and gonads of vimba bream (Vimba vimba).

Species	Tissue	Cu	Fe	Mn	Mo	Se	Zn
Vimba bream	Liver	9.74 ± 5.4 a	268.55 ± 126.6 a	2.42 ± 1.9 b	0.21 ± 0.2 b	2.14 ± 2.3 b	35.40 ± 8.6 b
	Gills	1.10 ± 0.6 b	168.38 ± 37.6 b	20.09 ± 4.7 a	0.86 ± 0.2 a	5.34 ± 1.7 a	50.73 ± 3.1 a
	Gonads	nd	9.30 ± 17.0 c	nd	nd	nd	18.07 ± 49.0 c

^a,b,c Statistical differences between tissues for each element are indicated by letters (a > b > c), where “a” marks the highest concentration.

,

Table 3. Element concentrations (mean ± SD; μg/g dw) in liver, gills, and gonads of perch (Perca fluviatilis).

Species	Tissue	Cu	Fe	Mn	Mo	Se	Zn
Perch	Liver	59.43 ± 22.1 a	213.52 ± 194.4	1.98 ± 0.9 c	2.08 ± 0.3 b	2.10 ± 0.4 b	94.22 ± 15.8 a
	Gills	0.90 ± 0.5 b	198.44 ± 73.4	10.08 ± 4.4 b	3.25 ± 0.5 a	7.24 ± 1.9 a	55.05 ± 4.9 b
	Gonads	1.17 ± 0.5 b	78.65 ± 63.1	16.26 ± 2.0 a	2.30 ± 0.5 b	1.99 ± 2.2 b	70.40 ± 4.9 b

^a,b,c Statistical differences between tissues for each element are indicated by letters (a > b > c), where “a” marks the highest concentration.

and

Table 4. Element concentrations (mean ± SD; μg/g dw) in liver, gills, and gonads of white bream (Blicca bjoerkna).

Species	Tissue	Cu	Fe	Mn	Mo	Se	Zn
White bream	Liver	28.06 ± 24.1 a	109.49 ± 87.6 b	0.90 ± 0.8 b	nd	nd	63.98 ± 19.1 b
	Gills	1.08 ± 0.8 c	184.47 ± 43.7 a	6.77 ± 4.4 a	nd	nd	53.65 ± 9.1 b
	Gonads	10.73 ± 6.3 b	46.46 ± 18.8 c	7.99 ± 5.0 a	nd	nd	158.73 ± 74.5 a

^a,b,c Statistical differences between tissues for each element are indicated by letters (a > b > c), where “a” marks the highest concentration.

).

In barbel, statistical analyses revealed significant differences in Cu and Mn concentrations between the liver and gills, whereas Fe and Zn levels were significantly higher in both the gills and liver compared to the gonads. In vimba bream, all three tissues exhibited statistically significant differences in Fe and Zn concentrations; furthermore, significant differences were observed for the Cu, Mn, Mo, and Se between the liver and gills. In perch, liver concentrations of Cu and Zn were markedly higher than those in the gills and gonads, while Mn levels varied significantly among the three tissues, with the highest concentrations found in the gonads. Additionally, in this species, Mo and Se were present in significantly higher amounts in the gills compared to the other two tissues. For white bream, significant differences in Cu and Fe concentrations were noted across all three tissues, with the highest levels of Cu found in the liver and Fe predominating in the gills. Mn concentrations were significantly higher in both gills and gonads relative to the liver, while Zn levels were significantly elevated in the gonads compared to the other two tissues (Table 1).

Generally, statistically significant correlations between element concentrations in tissue and TL and W were few. In the liver, significant correlations (all negative) were detected only in barbel. In gills, there is only a single significant negative correlation detected in perch. In gonads, a single negative correlation was observed in perch, while in vimba bream, there were several positive significant correlations (

Table 5. Statistically significant correlations between element concentrations and fish total length (TL) and weight (W).

Species	TL, W	Element	Correlation Coefficient	p-Value
Barbel
Liver
	TL	Zn	−0.815	0.007
	W	Fe	−0.725	0.027
	W	Mn	−0.667	0.050
	W	Zn	−0.827	0.006
Vimba bream
Gonads
	TL	Fe	0.611	0.012
	TL	Zn	0.715	0.002
	W	Fe	0.500	0.049
	W	Zn	0.622	0.010
Liver	TL	Cu	−0.533	0.050
Perch
Gills	TL	Cu	−0.911	0.032
Gonads	TL	Zn	−0.914	0.030

). Correlation matrices for each species (and every organ) are presented in Supplementary Material.

3.2. Metal Pollution Index (MPI)

Results of the metal pollution index (MPI) calculations are presented in Figure 2. Among the studied species, white bream showed the highest MPI values in both liver and gill tissues, whereas vimba bream exhibited the lowest. Specifically, MPI values in the liver of white bream and perch were significantly higher than those in vimba bream. Similarly, in the gill tissue, white bream displayed significantly higher MPI values compared to both barbel and vimba bream.

3.3. DNA Damage Assessed by Alkaline Comet Assay

Results regarding DNA damage in blood, liver, and gill cells for the examined species are illustrated in Figure 3. In blood cells, perch exhibited the highest level of DNA damage, followed by white bream, while vimba bream and barbel showed the least damage. In liver cells, white bream and barbel had the highest levels of DNA damage, with perch and vimba bream following. For gill cells, barbel showed the highest level of DNA damage, followed by perch and white bream, while vimba bream had the lowest levels. Overall, vimba bream consistently showed the least DNA damage across all tissues. Tissue-specific patterns in species revealed the highest damage in perch blood, white bream liver, and gills of barbel and vimba bream.

3.4. Frequency of Micronuclei

Regarding the presence of MN in fish erythrocytes, the highest levels were found in white bream (0.15‰) and perch (0.11‰), followed by vimba bream (0.03‰). No MN were detected in the barbel. Additionally, there were no statistically significant differences observed between the studied species.

3.5. Liver Enzyme Activity

There were no significant differences in AST (H = 3.902; df = 3; p = 0.272) and ALT (H = 7.137; df = 3; p = 0.068) enzyme activity between sampled fish species. The highest concentration of ALT was detected in barbel, and the highest concentration of AST was observed in white bream (

Table 6. AST and ALT enzyme activity (mean ± sd; U/L) in sampled fish species.

Species	AST	ALT
Barbel	105.69 ± 59.0	72.43 ± 29.1
Vimba bream	71.85 ± 42.8	54.33 ± 33.7
Perch	57.67 ± 31.5	89.32 ± 49.8
White bream	89.15 ± 47.0	91.25 ± 45.6

).

3.6. Erythrocyte Count and Developmental Stage

Perch exhibited the highest proportion of immature erythrocytes and the lowest proportion of mature erythrocytes, followed by barbel. In contrast, vimba bream and white bream demonstrated a greater prevalence of intermediary erythrocytes relative to immature ones, as well as a higher percentage of mature erythrocytes compared to both barbel and perch (

Table 7. Percentage of immature, intermediary, and mature erythrocytes in sampled fish species.

Species	Immature (%)	Intermediary (%)	Mature (%)
Barbel	66.06	25.28	8.66
Vimba bream	39.11	44.62	16.26
Perch	82.04	11.84	6.12
White bream	40.41	44.86	14.73

).

3.7. IBR Index

The results of the IBR analysis are illustrated in Figure 4. Among the species studied, white bream displayed the highest levels of all tested biomarkers, except for TI blood and ALT. Perch had increased levels of AST, MN, and MPI in both liver and gills. Barbel recorded the highest ALT values, along with elevated levels of TI in both liver and gills. Increased levels of TI blood and ALT were noted in vimba bream as well.

Figure 5 shows the overall IBR values for each species, highlighting that white bream is the most sensitive, followed by perch, barbel, and vimba bream.

4. Discussion

One of the primary, and still present, challenges in ecotoxicology lies in assessing species-specific responses to chemical pollution and the subsequent effects on biodiversity [31]. With this in mind, we have selected a site downstream of Belgrade, which is impacted by untreated municipal and industrial wastewater, to investigate the biomarker responses in four indigenous and commercially significant species. This section of the Danube was previously examined during the extensive international expedition Joint Danube Survey 4 (JDS4), which underscored the high pollution pressures stemming from industrial compounds, pharmaceuticals, and personal care products [32].

4.1. Metal and Metalloid Concentrations in Fish Tissues

Barbel is a bottom-dwelling fish, feeding mainly on benthic invertebrates and occasionally small fish [33]. Their population has dramatically declined due to habitat loss and overfishing, as it is highly valued and attractive to anglers [34]. It is considered a reliable bioindicator species for metal pollution in sediments [10], especially for specific locations, due to its low mobility [35]. There are several ecotoxicological studies [10,35,36] on barbel in the Belgrade sector of the Danube River. Generally, in these studies, concentrations of Fe were lower and concentrations of Mn were higher in both liver and gills than those found in the barbel from the present study. For Mo, barbel from these studies had higher concentrations in the liver but lower concentrations in the gills. Based on the barbel as an indicator, the Višnjica sampling site could have a problem with higher Fe load in sediments.

Vimba bream is an adequate bioindicator species for metal and metalloid pollution in river ecosystems [37]. With some exceptions, vimba bream caught in the Danube in our previous studies [12,27,37] had similar or higher concentrations of analyzed elements in liver. There could be an excess of Mn and Se in the area around Višnjica, as vimba bream caught at this site had higher concentrations of these elements in their liver compared to those from site Novi Banovci, a location known as a regional collector of untreated wastewater outlet [27].

In the present research, perch is the only species in which all elements were above the detection limit in all tissues. This could imply that this species has higher bioaccumulation potential than other analyzed species and is more suitable as an indicator for ecotoxicological monitoring. Compared to element concentrations in perch, published by Subotic et al. [38], perch from Višnjica have lower element concentrations in liver (except for Cu and Zn) and higher element concentrations in gills (except for Mo and Zn). Perch in both studies were sampled in environments with various anthropogenic pollution pressures, so differences in element accumulation in tissues could be attributed to element loads in the environment before or during the sampling period.

Analysis of element concentrations in liver and gills of white bream in other studies from the Danube shows generally higher concentrations of elements in these tissues [27,39]. It should be noted that concentrations in gonads, presented by Lenhardt et al. [39], had lower element concentrations in gonads, compared to white bream from the present study. Franco-Fuentes et al. [40] reported that variations in element concentrations in gonads possibly occur due to abiotic and/or biotic factors and also have a seasonal aspect. Since white bream from Višnjica were caught in late autumn/winter, as opposed to late winter/spring fish analyzed in Lenhardt et al. [39], the seasonal aspect of element bioaccumulation in gonads could be the reason for higher element concentrations at the Višnjica site.

A few significant correlations between perch TL and W, and element concentration in liver and gills were also observed by Subotić et al. [38]. A low number of significant correlations between element concentration in liver/gills and fish body length/weight was also reported by Lenhardt et al. [39] and Milošković, A and Simić [41]. In fish from Višnjica, every significant correlation between element concentration in certain tissues and length was negative. Canli et al. [42] reported that this negative relationship could be a result of metabolic activity and fish age, with older fish having lower metabolic activity and smaller relative uptake of elements [43]. Since all sampled specimens from Višnjica were older specimens, this could be a plausible explanation for negative correlations between fish length and element concentrations. Significant positive correlations with fish size and weight were observed only for Fe and Zn in the gonads of vimba bream. Bang et al. [43] also reported a significant positive correlation between Zn concentration and fish size/weight, which could be a result of the importance of Zn in reproduction and sexual maturation.

4.2. Genotoxic and Mutagenic Potential—Comet and Micronucleus Assay

Regarding the comet assay, barbel had the highest levels of DNA damage in both liver and gills, but also the lowest DNA damage level in erythrocytes. These values, however, surpass those reported in our previous study on barbel erythrocytes conducted at a location several kilometers upstream from the site at Višnjica, as well as those reported at the reference site [36]. Given the benthic nature of the barbel, it is anticipated that this species would exhibit the highest levels of DNA damage due to its prolonged exposure to environmental pollutants from both water and sediments. Barbel’s benthic lifestyle and sediment-disturbing feeding behavior could also explain the higher gill damage, as sediment often acts as a sink for genotoxic compounds. This exposure is particularly pronounced in the gills, which remain in continual and direct contact with environmental contaminants [44]. Similar tissue-specific vulnerability has been reported by [45], who emphasized that gill tissues often reflect acute genotoxic insults due to their large surface area and thin epithelial layers. Furthermore, variations in the DNA damage response across different tissues are likely attributable to the distinct physiological functions of the respective organs, the rates of metabolic activation and detoxification processes, as well as the efficiency of DNA repair mechanisms [46]. Our results further support this, as barbel had the highest levels of DNA damage in liver and gills, but also the lowest DNA damage in erythrocytes, and absence of micronuclei, possibly reflecting tissue-specific differences in DNA repair efficiency. Boettcher et al. [47] employed the micronucleus test using barbel to evaluate the mutagenic potential of Danube River water across four different localities. Their findings indicate that barbel serves as a reliable biological indicator of environmental quality. The observed values ranged from 0.6‰ in the control group to 1.4–3.1‰ at sampling locations subjected to different environmental pressures. It is noteworthy that this investigation was conducted during the summer months, a period typically associated with heightened biomarker responses compared to winter [48,49].

Among the four species analyzed, vimba bream consistently exhibited the lowest levels of DNA damage across all three tissues, suggesting a potential resilience to genotoxic stressors in the sampled environment or a lower capacity for bioaccumulation of contaminants. In our previous investigation on vimba bream from the Danube River, we observed higher levels of DNA damage in liver, gill, and blood cells compared to the findings from the present study [27]. This discrepancy may be attributed to the different sampling season, specifically summer. Furthermore, in that study, the extent of DNA damage in liver and blood cells was notably higher in vimba bream compared to white bream [27] unlike the present results. In terms of micronuclei frequencies, vimba bream demonstrated the second lowest occurrence of MN among the species examined. Low MN frequency and higher sensitivity of white bream compared to vimba bream was reported previously [27], supporting the observed interspecies differences in genotoxic sensitivity.

In contrast, perch and white bream showed elevated levels of DNA damage in specific tissues, as well as the highest MN frequencies, indicating differential susceptibility or exposure. Perch exhibited the highest DNA damage in blood cells, consistent with multiple studies that highlight erythrocyte sensitivity to genotoxic stress, as evidenced by increased frequencies of micronuclei and other nuclear abnormalities in polluted environments [50,51,52]. These findings support our observation that perch, due to its higher metabolic activity and predatory behavior, is prone to a higher genotoxic burden.

White bream exhibited the highest level of DNA damage in liver cells. In our previous study on the Sava River bream species, a high seasonal impact on the response of different tissues was confirmed [11]. It was observed that, at least during spring and summer, gills had higher levels of DNA damage compared to liver [11,27], while during winter this difference was not present [11]. On the other hand, MPI index in the present study was highest in white bream liver, pointing to a possible cause of increased genotoxic response.

This study, consistent with many prior investigations, further confirms that the alkaline comet assay exhibits greater sensitivity than the micronucleus assay in assessing interspecies variation in response to pollution stress [27,48,53,54,55]. The low incidence of MN in vimba bream and barbel, along with their reduced sensitivity to pollution, as indicated by the IBR, suggests efficient DNA repair mechanisms or cell death prior to mitosis [56]. This particularly applies for barbel, which, despite exhibiting the highest levels of DNA damage in gills and liver, displayed a complete absence of MN.

4.3. Hepatic Enzyme Activities—AST and ALT

The analysis of hepatic enzyme activities—aspartate aminotransferase (AST) and alanine aminotransferase (ALT)—in the examined fish species reveals notable interspecies variations, likely reflecting differences in metabolic status, physiological condition, or exposure to environmental stressors. The highest mean AST activity was observed in barbel (105.69 ± 59.0 U/L), which may suggest elevated metabolic demands or possible cellular damage in tissues rich in AST, such as liver, heart, muscle, or kidneys. The considerable standard deviation indicates marked individual variability within the population, possibly due to differential sensitivity to environmental influences. In contrast, perch exhibited the lowest AST levels (57.67 ± 31.5 U/L) but simultaneously showed the highest mean ALT activity (89.32 ± 49.8 U/L). Given that ALT is more liver-specific than AST, this elevation may point to hepatocellular alterations even in the absence of significantly increased AST, suggesting a more localized hepatic response. White bream showed elevated levels of both AST (89.15 ± 47.0 U/L) and ALT (91.25 ± 45.6 U/L), potentially indicating a combined metabolic and hepatic response, or a more systemic reaction to environmental stimuli. A similar, though somewhat milder, pattern was observed in vimba bream. When comparing the AST/ALT activity ratio among species, barbel and vimba bream showed higher AST than ALT, which may suggest a predominance of extrahepatic sources of AST or relatively preserved hepatic function. Conversely, in perch and white bream, ALT levels exceeded AST, consistent with a possible hepatocellular stress response or adaptive metabolic shifts to particular environmental pressures. Overall, these findings support the use of transaminase activities as sensitive biochemical markers of hepatic function and general physiological status in fish. The observed interspecific variability underscores the importance of species-specific interpretation, as well as the potential of AST and ALT to reflect both metabolic and environmental influences.

4.4. Erythrocyte Morphometry

Erythrocyte morphometry is variable in fish, depending on the type of species, but also age or size of specimens within the species [57]. However, some studies also dispute that habitat and body size influence erythrocyte shape [58]. Erythrocyte morphometry is linked with environmental factors and can be a possible indicator for fish adaptations in ecosystems exposed to pollution [52]. Witeska et al. [59] reported that the frequency of immature cells is a more sensitive and reliable indicator of pollution than basic quantitative erythrocyte parameters, such as hematocrit, erythrocyte count or hemoglobin concentration. An increase in the frequency of immature cells in peripheral blood is an organism’s response to erythrocyte damage caused by environmental stressors [59]. A high percentage of immature erythrocytes can serve as an indicator of hypoxic events or problems in oxidative radical production [60]. At the Višnjica sampling site, both perch and barbel had higher immature erythrocyte percentages than the other two species. This could indicate that these species are better for this type of monitoring. On the other hand, a higher percentage of mature erythrocytes in vimba bream and white bream could be an indication of lower feeding efforts or starvation. Rios et al. [61] reported that a decreased number of immature erythrocytes is an indication of starvation, impairing erythrocyte production. Compared to previous research conducted on vimba bream and white bream [27], these two species from Višnjica have a lower percentage of both immature and mature erythrocytes. This could point to a more favorable environment, with less stress caused by pollution, than in Novi Banovci.

5. Conclusions

This study demonstrates significant interspecies and tissue-specific differences in genotoxic and biochemical responses to environmental contaminants among four native freshwater fish species. Each species represents a different ecological niche and feeding behavior, influencing their exposure and response to environmental stressors. The application of IBR effectively integrated multiple biomarker responses, providing a comprehensive assessment of the biological effects of pollution in studied species. These findings highlight white bream as the most sensitive bioindicator species, followed by perch and barbel, with vimba bream serving as a potential reference species.

Overall, incorporating multiple species and diverse tissue analyses enhances the ecological relevance and diagnostic power of biomonitoring programs in freshwater ecosystems. While the selected biomarker battery proved informative, it would have been valuable to also include endpoints for other rising environmental pollutants. Considering the strong tissue- and species-specific variability observed, this represents an important recommendation for future biomonitoring studies.