Effects of short-term dietary oxytetracycline treatment in the farmed fish Piaractus mesopotamicus

Abstract

Oxytetracycline (OTC), a key antibiotic used in global aquaculture, has still unclear ecotoxicological effects. In this study, freshwater fish Piaractus mesopotamicus were fed diets containing 750 mg kg⁻¹ of either pure OTC (ATB1) or commercial OTC (ATB2) for 10 days (treatment period-TP), followed by a 21-day withdrawal period (depuration period-DP). Fish fed with ATB2 showed decreased hematocrit (at DP) and increased glucose levels (TP and DP). In general, catalase activity increased in the liver, gills, and muscle of OTC-treated individuals at both TP and DP, particularly with ATB2. Similarly, glutathione S-transferase activity rose in the brain, gills, and muscle (TP and DP). Conversely, alkaline phosphatase activity in the liver decreased in both treated groups (TP and DP). Additionally, only ATB2 induced lactate dehydrogenase in fish muscle after 1-day depuration. Principal component analysis identified most antioxidant enzymes, hematocrit, weight, length, and mean corpuscular hemoglobin concentration as key biomarkers, distinguishing ATB2 from control fish. These results indicate that the dietary therapeutic dose of OTC caused adverse effects in P. mesopotamicus. Differences in biomarker responses between ATB1 and ATB2 might be linked to unknown compounds in the commercial formulation, potentially influencing biological responses or altering OTC bioavailability. Further research on the toxicity of antimicrobial impurities and degradation compounds should accompany enhanced quality control measures in aquaculture to guarantee sustainable and safe products.

Introduction

Antibiotics are commonly used in aquaculture facilities during the production cycle to treat and prevent bacterial diseases (Bojarski et al., 2020). Oxytetracycline (OTC) is one of the most used antibiotics in aquaculture worldwide, owing to its broad-spectrum activity, high potency, and low cost (Manna et al., 2021). It is within the limited number of pharmacologically active substances approved for aquaculture, including by the United States Food and Drug Administration and the National Service of Food Health and Quality in Argentina (U.S. Food and Drug Administration, 2022; Servicio Nacional de Sanidad y Calidad Agroalimentaria, 2010). In infectious diseases, OTC is usually administered via formulated feeds that contain therapeutic doses at a rate of 50–100 mg of OTC per kg of body weight per day for 3–21 days depending on the infection (Pês et al., 2018).

Despite the benefits of OTC, the toxicological effects of chronic or acute exposure are still not fully understood in fish. The results of OTC exposure vary not only with the antibiotic concentration but also with the exposure route, period, time since exposure, and fish species (Iida et al., 2022). Existing literature indicates that OTC could produce tissue alterations in liver, kidney, and gills, generate negative effects on the immune system, and cause oxidative stress and disturbances in the antioxidant system of fish (Krupesha Sharma et al., 2021; Pathak et al., 2024; Rodrigues et al., 2018). Therefore, the use of OTC could induce multiple side effects in farmed fish, which will ultimately affect production (Limbu et al., 2020). Examination of hematological, biochemical, and oxidative stress parameters could be used in aquaculture research for evaluating the well-being of the cultured organism (Kumar et al., 2022). Moreover, one aspect not covered is the presence of process impurities and degradation products in the active pharmaceutical ingredient even in small amounts, as these impurities may influence the efficacy and safety of the product and can be toxic to fish in particular (Phu et al., 2015).

Piaractus mesopotamicus, commonly known as pacú, is one of the most important farmed species in Latin America. The popularity of this freshwater fish in national and international aquaculture is due to favorable zootechnical characteristics, such as its hardy nature and rapid growth under a variety of conditions, and profitable breeding (Kuradomi et al., 2017). However, the inadequate practices in fish farms have caused several disease outbreaks that represent a serious risk to animal health and both production and consumption of fish products (Mastrochirico-Filho et al., 2019). Pacú is highly affected by Aeromonas hydrophila and OTC is one of the antibiotics available for its treatment (Farias et al., 2016).

The main goal of this study was to simulate the pharmacological treatment with OTC-containing feed in an experimental aquaculture system suitable for pacú, assessing the possible adverse side effects of this antibiotic in both pure and commercial (formulated) forms. Furthermore, the physiological and biochemical responses of the treated fish were assessed during the OTC withdrawal period. We hypothesized that fish treated with OTC would exhibit changes in morphometric, hematological, and biochemical parameters, with variations depending on whether the antimicrobial was in pure or formulated form.

Materials and methods

Fish

Healthy juvenile fish of the species P. mesopotamicus (n = 80; 8.37 ± 0.64 cm standard length; 20.85 ± 5.36 g initial body wt) were obtained from a local fish farm in Paraná, Argentina. The fish were acclimatized in 300-L tanks containing well-aerated dechlorinated water for 4 weeks. The physicochemical parameters of the water were controlled and recorded (temperature 23.4 ± 0.3 °C, dissolved oxygen 8.15 ± 0.29 mg/L, pH 7.52 ± 0.30, electrical conductivity 233.8 ± 7.7 μs/cm, and total ammonia nitrogen 0.17 ± 0.06 mg/L). Fish were fed daily ad libitum with commercial pellets (Avigan) up to 24 hr before the beginning of the exposure tests.

Experimental diets

The first diet was the commercial feed (Avigan) used as a control diet (CTRL). The second diet (ATB1) was formulated with CTRL and pure OTC 750 mg/kg (O5875, > 95% pure, Sigma Aldrich), and the third diet, with CTRL and commercial OTC 750 mg/kg (6% OTC hydrochloride, El Gigante S.R.L., Argentina, ATB2). This concentration was selected based on the recommended and standard therapeutic finfish dose ranging from 50–100 mg of OTC per kg of body weight and an average weight of 15 g (U.S. Fish and Wildlife Service, 2022). The chemical composition of the diets used was determined according to standard methods (Association of Official Agricultural Chemists, 1995) and is reported in Table 1. The OTC-medicated diets were obtained from the grinding and subsequent sieving of the commercial feed (≥ 50 µm). Starch and water (which contains commercial/pure OTC dissolved) were added. The resulting mixture was homogenized (planetary mixer, Erweka, Germany), extruded (extruder, Técnica Giraudo, Argentina), and spherized (spheronizer, Técnica Giraudo, Argentina). After pelleting, diets were dried at room temperature and stored at –20 °C until utilized. The control feed was prepared in the same way but without adding OTC.

An Orlando and Simionato (2013) based extraction strategy was used to quantify OTC in formulated diets. Briefly, 0.5 g of fish feed was combined with 5 ml of McIlvane buffer (pH = 4) and 240 µl of 0.01 M ethylenediaminetetraacetic acid (EDTA), followed by vortex mixing and centrifugation. The extraction procedure was repeated twice and the resulting supernatants were combined. After the addition of 1 ml trichloroacetic acid 20% the extract was mixed and centrifuged once more. The supernatant was loaded onto Strata X cartridges (500 mg/6 ml, Phenomenex; Torrance, CA, USA), which were conditioned with 5 ml methanol and 5 ml water. The samples were eluted with 8 ml methanol, evaporated to dryness in an automatic concentrator (Concentrator plus/Vacufuge plus, Eppendorf, Hamburg, Germany) and reconstituted for blank and control samples in 2 ml mobile phase, 10:90 (v/v) acetonitrile: aqueous 0.1% formic acid. Simultaneously, a subset of control samples (n = 3) was spiked with a solution of OTC at a final concentration of 1,500 mg/kg to verify recovery percentages (50% ± 13%). The extracts were analyzed in Waters Acquity Ultra-Performance liquid chromatography system coupled to a Xevo TQ-S tandem quadrupole mass spectrometer with electrospray ionization source (Waters Corporation, Milford, MA, USA) in accordance with Griboff et al. (2020). The injection volume was 10 µl. Online supplementary material Table S1 shows the transitions of OTC used for identification and quantification. Alpha-apo-OTC and beta-apo-OTC were also analyzed to assess the presence of impurities in the feed; however, neither of them was detected in the studied feeds. Oxytetracycline was below the detection limit in the CTRL diet (23.2 mg/kg). Concentrations of OTC in ATB1 and ATB2 diets were 747.26 ± 72.14 mg/kg dry weightand 660.89 ± 76.25 mg/kg dry weight, respectively. All analyses were conducted in triplicate.

Experimental design

Temperature and photoperiod were maintained like in the acclimation period. Individuals were randomly stocked into nine 300‐L tanks with 8–10 fish per tank. The diets were randomly assigned to triplicate tanks for animal treatment for 10 days. Fish were fed twice a day at a readjusted-restricted rate (5% biomass wt per day). Then, fish were maintained for 21 days following the last feeding of medicated feed (withdrawal period) as recommended for Terramycin 200 for Fish (Phibro Animal Health, 2023), while being fed with commercial feed (Avigan without OTC).

Five specimens of each treatment were randomly sampled at each time point: 0.25, 3, 10, 11 (1-day-depuration), and 31 (21-days-depuration) days. Additionally, for time 0, five specimens were used. Fish were anesthetized in benzocaine (100 mg/L), weighed (g ± 0.1 g), measured (total length, cm ± 0.1 cm), and euthanized by cervical transection. Blood was immediately sampled from the caudal vein with heparin as an anticoagulant and used to determine hematological parameters. Plasma was separated from the remaining blood by centrifugation and stored at 4 °C until analysis. Fish were dissected and the liver, gills, brain, and muscle were separated and stored at –80 °C until biochemical analysis. The Fulton condition factor (CF) was calculated following Goede and Barton (1990) using the formula CF = (BW/L³) × 100, where BW is the body weight (g) and L is the total length (cm).

Hematological and biochemical parameters

A Neubauer chamber was used for counting red blood cells (RBCs). Hematocrit (Ht) values and hemoglobin concentration (Hb) were measured by the micro-method and the cyanomethemoglobin method at a wavelength of 546 nm, respectively (Rossi et al., 2017). Additionally, hematimetric indexes (mean corpuscular volume [MCV], mean corpuscular hemoglobin, and mean corpuscular hemoglobin concentration [MCHC]) were calculated as described by Cazenave et al. (2005). Protein, cholesterol, triglycerides, and glucose levels were immediately determined in plasma (n = 5 per treatment) by enzymatic colorimetric methods (Wiener Lab commercial kits, Argentina).

Activity of enzymes

Enzyme extracts of the liver, gills, brain, and muscle tissues were prepared from each individual according to Cazenave et al. (2006). Briefly, organs were homogenized using a glass homogenizer and extraction buffer (0.1 M of sodium phosphate buffer pH 6.5, 20% (v/v) glycerol, 1 mM of EDTA, and 1.4 mM of dithioerythritol). All samples were centrifuged at 20,000 g for 30 min (4 °C) to collect the supernatant, which was further frozen in liquid nitrogen and kept at −80 °C for enzyme measurement. Glutathione-S-transferase (GST) activity was measured in all tissues using 1-chloro-2,4-dinitrobenzene as substrate at 340 nm, and the activity of catalase (CAT) at 240 nm using hydrogen peroxide as substrate (Cazenave et al., 2006). The alkaline phosphatase (APA) activity was determined in liver at 520 nm and lactate dehydrogenase (LDH) activity was estimated in muscle at 340 nm, using commercial kits from Wiener Lab (Argentina) for both. All the determinations were carried out in triplicate using a Synergy HT Multi-Mode Microplate Reader (BioTek Instruments, Winooski, USA). The enzymatic activity was calculated in terms of the protein content of the sample and is reported in nanokatals per milligram of protein (nkat/mg prot; Cazenave et al., 2006).

Data analysis

The results are expressed as mean ± SD. Shapiro-Wilks and Levene tests were used to assess normality and homogeneity of variances, respectively. Comparison of each biomarker among treatments and time points was performed using analysis of variance. A posteriori test, Fisher least significant difference test, was used to determine significant differences between the means. In addition, principal component analysis (PCA) was performed on standardized values, obtained by subtracting the mean and dividing by the SD of each variable. This approach provided a comprehensive view of the results and helped define the most important parameters involved in OTC toxicity (all timepoints together) by reducing, classifying, and grouping the variables. For all statistical analyses, Software InfoStat Ver. 2018 (Argentina) was used. A significance level of p < 0.05 was applied to all statistical analyses.

Results and discussion

Morphometric parameters

Fish promptly accepted diets, and no mortality occurred during either the feeding or the depuration trial. The main morphometric characteristics of fish are summarized in Table 2. The weight and length were relatively homogeneous and there were no significant differences between dietary treatments at the beginning of the experiment, after the 10-day feeding trial, or 21 days following the last feeding of medicated feed. However, on Day 10, the CF of specimens fed with ATB2 was significantly higher than those fed with CTRL and ATB1 (p = 0.0201). These results are consistent with the findings of another study, which showed that OTC does not promote growth in finfish species (Limbu et al., 2020).

Hematological and biochemical parameters

Data on the hematological response in fish after antibiotic treatment are scarce and ambiguous. For instance, Ambili et al. (2013) reported a decrease in Hb and Ht in OTC-treated Labeo rohita fish after 10 days and an increase after longer exposure periods (15–25 days). In our study, no major changes were observed in the values of hematological parameters among treatments (Table 2), except for Ht, which decreased after 1 day of depuration in ATB2-fed fish compared with CTRL and ATB1 (p = 0.0368). However, even when significant changes occurred, the mean values were within the normal reference ranges of hematological parameters reported for the species. In general, the range registered in this study was similar to results reported for pacú by Dal’Bó et al. (2015), who reported Ht values ranging from 32%–42%, Hb from 7.15–8.7 g/dL, RBC from 1.6–3.8 ×10⁶/μl, MCHC from 19.9%–25.3%, and MCV values ranging from 92.34–200.0 μm³. Therefore, the observed changes in blood component values may be attributed to genetic and physiological variations that could result from intraspecific differences within the species (Fazio, 2019).

Biochemical parameters such as glucose and proteins are the major energy suppliers of the organism, and their levels indicate the health status of the fish (Ramesh et al., 2018). During exposure, the ATB2-group showed an increase in glucose level at 3 (p = 0.0223), 10 (p = 0.0412), and 21-days depuration (p = 0.0284) and a decrease in plasmatic protein at 3 (p = 0.0179) and 1-day depuration (p = 0.0014; Table 2). The plasma glucose level is the most commonly measured indicator of secondary phase stress response in fish; its increase would be triggered by the metabolism of carbohydrates and energy reserves during a stressful situation (Pellegrin et al., 2023). In contrast, the triglycerides and cholesterol of fish fed with CTRL, ATB1, and ATB2 varied throughout the experiment, but no differences were observed among treatments. Although it is recognized that stress responses and the maintenance of homeostasis increase energetic costs (Rossi et al., 2017), in our study, plasma triglycerides and cholesterol were not required as extra energy sources by the OTC-exposed fish. This finding is in accordance with Xie et al. (2020), who demonstrated that the legal aquaculture dose of sulfamethoxazole in Micropterus salmoides did not significantly affect plasma triglyceride or cholesterol levels.

Enzyme activities

The CAT activity remained without changes over the experiment in liver, gills, brain, and muscle of fish in the CTRL group (see online supplementary material Figure S1 and Table 3). In the liver, the ATB1 and ATB2 diets caused a significant increase in CAT activity in P. mesopotamicus compared with the control group (CAT_ATB2> CAT_ATB1) 0.25 days after starting the experiment (p = 0.0008). At 21-days depuration, CAT activity (0.25 days) decreased 70% for ATB1 and 12% for ATB2 but never to the initial levels. Therefore, even when the withdrawal period finished, this antioxidant enzyme in the liver maintained its increased activity. Similarly, the gills of the fish fed with ATB2 had greater CAT activity when compared with ATB1 and CTRL diet at Day 10 (p = 0.0017) and 21-days depuration (p = 0.0106). Moreover, the brain CAT activity did not change significantly among the treatments, but there was some variability over time; in the ATB1 group the activity increased on Day 10 (p = 0.0003) and in ATB2 on Day 3 (p = 0.0014), maintaining these levels until the depuration was completed. Furthermore, the muscle of the fish fed with ATB2 had greater CAT activity on Day 10 (p = 0.0289), 1 day-depuration (p = 0.0022), and 21 days depuration (p = 0.0401), when compared with ATB1 and CTRL. Catalase is one of the enzymes that provides the first defense line against the production of reactive oxygen species and protects from tissue damage by hydrogen peroxide (Vasylkiv et al., 2011). Therefore, and considering our study’s results, oxidative stress with overproduction of hydrogen peroxide was established in P. mesopotamicus exposed to OTC. This result is consistent with lipid peroxidation and enzymatic and nonenzymatic antioxidant responses observed after treatment with OTC 0.1 mg/kg diet for 2 weeks in the silver catfish (Rhamdia quelen) by Londero et al. (2021). In particular, the elevation of CAT activity was found in fish exposed to antibiotics, including tetracycline in Gambusia holbrooki (Nunes et al., 2015) and norfloxacin in Pangasius sp. (Shreenidhi et al., 2023).

On the other hand, GST activity showed variable responses to OTC exposures in this species. The activity of this enzyme measured in the liver of the ATB2 fish group at 1-day depuration decreased significantly (p = 0.0407) when compared with the CTRL (see online supplementary material Figure S1 and Table 3). This decrease in liver GST activity could indicate oxidative damage caused by the 10-day treatment with ATB2. In fish gills, the activity of GST was significantly increased by treatment with ATB2 on 0.25 (p = 0.0015) and 3 days (p = 0.0313) and in the withdrawal after treatments with ATB1 and ATB2 on 1 (p = 0.0055) and 21-days depuration (p = 0.0443). In the brain, the ATB2 diet significantly increased GST activity, whereas the ATB1 diet caused an increase only at the beginning of the treatment phase, 0.25 (p = 0.0013) and 3 days (p = 0.0001), compared with the CTRL group. The highest values of this enzyme were observed during the withdrawal for both treatments; ATB1 at 1-day depuration (p = 0.0441) and ATB2 at 21-days depuration (p < 0.0001). The GST activity in muscle increased after exposure to OTC; with the highest activities at Day 0.25 for the ATB1 group (p = 0.0062) and at the rest of the time points for the ATB2 group, with significance levels of p = 0.0021 (3 days), p = 0.0003 (10 days), p = 0.0001 (1 day-depuration), and p = 0.0004 (21 days-depuration). The increase of GST activity in some tissues herein results in greater detoxification/excretion of xenobiotics by facilitating the conjugation of glutathione to inactivate highly reactive aldehydes produced from lipid peroxyl radicals (Götte et al., 2020). Therefore, elevated GST activity may indicate a mechanism to counteract the effects of OTC toxicity. When comparing CAT and GST activity across tissues in the OTC-treated groups, the most significant changes were observed in the liver. In this tissue, CAT activity showed variations at all exposure and depuration timepoints, whereas a similar trend was observed for GST activity. The liver appears to be more susceptible to excess free radical release and damage compared with other tissues (Rossi et al., 2017). The unchanged CAT activity in the gills and brain, along with the activation of GST, suggests that the phase II biotransformation pathway was activated to mitigate oxidative stress. However, although GST activity was activated and increased in the muscle, this response was insufficient to prevent the activation of CAT.

Significantly decreased APA values were observed in the groups fed with ATB1 and ATB2 when compared with CTRL at Day 3 (p = 0.0078), Day 10 (ATB1 and ATB2, p = 0.0023), and 21-days depuration (ATB2, p = 0.0114; see online supplementary material Figure S2A; Table 3). This result is in agreement with decreased APA activity detected in zebrafish gut after OTC exposure for 6 weeks (Zhou et al., 2018). The reduction in APA activity after OTC treatment may indicate a disruption in the membrane transport system (Khan and Khan, 2022).

Finally, the activity of LDH enzyme did not show significant differences among the times evaluated in each treatment (see online supplementary material Figure S2B and Table 3). The only significant difference for this enzyme was found at 1-day depuration, higher in the muscles of fish fed with ATB2 than in CTRL and ATB1 (p = 0.0113). Lactate dehydrogenase is an important enzyme in muscle physiology that can be increased by chemical stress, because a common response is metabolic hypoxia and a consequent increase in the anaerobic pathway of energy production (Oliveira et al., 2016). Therefore, it can be concluded that ATB2 induced anaerobiosis in muscle of P. mesopotamicus at 1-day depuration, which is coincident with the report of Rodrigues et al. (2018) in rainbow trout after 96 hr exposure to waterborne OTC.

Integrated analysis

The PCA developed for the enzymes, morphometric, hematological, and biochemical parameters produced two significant components, which together explained 100% of the total variability of the dataset (Figure 1; Table 4). The first axis (PC1) explained 77.5% of the total variance in the data, and it separated mainly CTRL and ATB2 treatments. Thus, on this first PC, ATB2 treatment (on the left) was clearly distinguished from CTRL (on the right), showing significant positive loadings for GST-liver [L], Ht, weight, and length of fish. In contrast, a negative correlation was found mainly for antioxidant enzymes (GST-brain [B], GST-gill [G], GST-muscle, CAT-L, and CAT-G) and mean corpuscular hemoglobin (MCH). The PC2 (22.5% of the total variance) differentiated fish fed with ATB1. This axis appears to be positively correlated with variables such as MCV, glucose, APA and condition factor; and negatively for Hb, MCH and CAT-B.

The diets used in this study were formulated with the same commercial feed and did not have significant differences in their macronutrient composition (Table 1). Therefore, the difference in responses observed between the control diet and the two treatments (ATB1 and ATB2) could be attributed to OTC. However, some biomarkers also showed different responses between ATB1 and ATB2 (Tables 2 and 3). These findings might be attributed to unknown compounds in the commercial formulation that trigger a biological response or modify the bioavailability of OTC. The European Pharmacopoeia states that OTC and its hydrochloride form may contain impurities such as 4-epi-OTC, tetracycline, and 2-acetyl-2-decarboxamido-OTC. Oxytetracycline hydrochloride may also contain anhydro-OTC, alpha-apo-OTC, and beta-apo-OTC (Khan et al., 1987). These impurities are allowed in concentrations between 0.5% and 2% because several of them are less potent than OTC (Lykkeberg et al., 2004). Unlike antimicrobial products used for human treatment, there is limited information on the quality of antimicrobial products used in aquaculture (Phu et al., 2015). Moreover, its potential toxicity to fish remains largely unknown. In this study, the three impurities analyzed in the diets were below the detection limit (see online supplementary material Table S1). Nevertheless, we cannot exclude the possibility of other impurities or degradation compounds present that were not included in the analysis. Further studies on the toxicity of antimicrobial impurities and degradation compounds should be conducted in parallel with stricter quality control of antimicrobials used in aquaculture.

Conclusion

Our results reveal that the dietary therapeutic dose of OTC, particularly the commercial formulation, causes significant alterations in the P. mesopotamicus enzyme system and, to a lesser extent, in biochemical and hematological parameters. This information would be valuable for aqua-farmers in enhancing fish health by preventing potential oxidative stress and other negative effects associated with OTC overuse. It is worth mentioning that the results obtained are from a specific treatment; however, the overuse or misuse of antibiotics in aquaculture could pose potential risks to human and environmental health, particularly by promoting the occurrence, selection, and spread of antibiotic-resistant bacteria and genes. Furthermore, water pollution in aquaculture areas may lead to exposure to multiple residues, potentially exacerbating the negative effects of therapeutic antibiotics and resulting in greater toxicity due to the presence of toxic mixtures. Therefore, the surveillance and quality control of antibiotics used in aquaculture, in addition to the evaluation of water quality, must be improved to guarantee both a sustainable use of antibiotics and safe aquaculture products for human consumption.

Supplementary material

Supplementary material is available online at Environmental Toxicology and Chemistry.

Data availability

The datasets generated during this study are available from the corresponding author on reasonable request.

Author contributions

Julieta Griboff (Formal analysis, Investigation, Methodology), Juan Cruz Carrizo (Formal analysis, Investigation, Methodology), Carla Bacchetta (Methodology, Validation), Andrea Rossi (Methodology, Validation), Daniel A. Wunderlin (Conceptualization, Funding acquisition), Jimena Cazenave (Conceptualization, Methodology, Supervision), and María Valeria Amé (Conceptualization, Funding acquisition, Investigation, Supervision)

Funding

This study has been co-financed by the Agencia Nacional de Promoción Científica y Técnica Argentina (FONCyT PICT-2019- N◦2892), the International Atomic Energy Agency (CRP: D52039, CN: 18849), and Secretaría de Ciencia y Técnica (SECyT, UNC, Res. 411/2018; 258/23).

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article.

Ethics statement

All procedures were previously approved by the Animal Ethical Committee at the National University of Córdoba (CICUAL/FCQ-UNC, RES. 735/2018).

Acknowledgments

The authors gratefully acknowledge CONICET (Consejo Nacional de Investigaciones Científicas y Técnicas-Argentina).

References