Crude extract of cyanobacterium (Radiocystis fernandoi) triggers primary osmoregulatory impairment in the gills of nile tilapia (Oreochromis niloticus)
Soares M.A.M., Pedrão P.G., Coelho J.P.S., Giani A., Fernandes M.N., Paulino M.G.
TL;DR
A cyanobacterium extract harms the gills of Nile tilapia by disrupting osmoregulation and causing structural damage, even without liver toxicity.
Contribution
This study reveals acute gill toxicity from waterborne exposure to a cyanobacterial extract containing MC-RR and MC-YR in Nile tilapia.
Findings
Gill osmoregulation was impaired, with decreased plasma chloride and epithelial disruption.
Mucous cell proliferation and pillar cell changes occurred after 12 and 96 hours of exposure.
No hepatotoxicity was observed despite gill structural and functional damage.
Abstract
Eutrophication, driven by microalgae proliferation, promotes the production of microcystins (MC), posing global risks to the environment and public health. In Brazil, the species Radiocystis fernandoi commonly produces variants such as MC-RR and MC-YR. However, most toxicological studies still emphasize the isolated hepatoxicity of the MC-LR variant, with limited data on the toxicity of other variants or water exposure of the entire cellular contents released during cyanobacterial lysis. This study assessed acute (12 h and 96 h) effects of waterborne exposure to a crude extract (CE) of R. fernandoi R28 containing MC-RR and YR mostly (10 µg MC-LReq L⁻¹) on juvenile Oreochromis niloticus, focusing on gill osmoregulation and gill and liver tissue integrity. Gill alterations included mucous cell proliferation (12 h) and epithelial disruption with pillar cell changes (96 h), increasing the…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7- —Universidade Federal Do Norte Do Tocantins
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAquatic Ecosystems and Phytoplankton Dynamics · Environmental Toxicology and Ecotoxicology · Aquaculture Nutrition and Growth
Introduction
The global expansion of eutrophication, largely driven by nutrient enrichment from anthropogenic sources since the mid-20th century (Tedengren, 2021), frequently culminates in harmful cyanobacterial blooms. These blooms pose a significant toxicological threat to aquatic ecosystems and public health due to the production of cyanotoxins, particularly microcystins (MCs) (Huo et al., 2021; Paulino et al., 2020). Environmental conditions prevalent in regions like Brazil (pH 6–9, 15–30 °C) further exacerbate the risk of blooms involving toxigenic species (Campos et al., 2024).
Among the diverse cyanotoxin producers, Radiocystis fernandoi is increasingly recognized in Brazilian freshwater systems (Lv et al., 2022). Notably, strains like R28 predominantly produce MC variants RR and YR, differing from the most studied MC-LR variant (Paulino et al., 2017, 2020). While MCs are known hepatotoxins, primarily acting via inhibition of protein phosphatases (PP1/PP2A) following uptake (Chen et al., 2016), their effects following direct exposure of aquatic organisms to contaminated water remain less understood, especially for variants other than MC-LR (Henri et al., 2021).
The gills are directly exposed to the aquatic environment and play essential roles in gas exchange and ion regulation. Due to this exposure, they are also a common site of interaction and uptake of waterborne toxicants (Fernandes et al., 2013; Zink & Wood, 2024). Consequently, exposure to waterborne toxicants, including MCs, can directly impair gill function, resulting in osmoregulatory disturbances, oxidative stress, and histopathological damage, potentially even without significant systemic accumulation or classic hepatotoxicity (Martins et al., 2019; Shingai & Wilkinson, 2023; Tavares et al., 2019). However, the specific impacts of environmentally relevant MC-RR and YR mixtures on gill osmoregulatory mechanisms following waterborne exposure require further investigation.
Oreochromis niloticus (Nile tilapia) a globally important aquaculture species known for its resilience, remains susceptible to environmental stressors, including cyanotoxins (Falfushynska et al., 2023; Wu et al., 2025). Understanding its physiological and toxicological responses to environmentally realistic cyanotoxin for fish exposure is crucial. Therefore, this study aimed to investigate the acute toxicological effects of waterborne exposure to R. fernandoi R28 crude extract (CE; containing MC-YR and MC-RR) on O. niloticus. Specifically assessing the temporal response (12 h and 96 h) of key biomarkers associated with gill osmoregulatory function and tissue integrity, thereby providing insights into the mechanisms of toxicity elicited by direct contact with these less-studied MC variants in a relevant fish species.
Materials and methods
R. fernandoi R28 cyanobacteria cultivation, extraction and quantification
The cyanobacterium R. fernandoi, strain R28, was isolated from the Furnas Hydroelectric Power Plant reservoir in Minas Gerais, Brazil (November 2006; 20°40’S; 46°19’W) and cultivated according to Pereira et al. (2012) at 25–30 mol m^− 2^ s^− 1^ irradiance under a 12-h light/12-h dark photoperiod and a constant temperature of 20 °C in the Phycology Laboratory of the Department of Botany at the Federal University of Minas Gerais. Crude extract was obtained from freeze-dried cyanobacteria and chemical characterization of this strain using HPLC-UV demonstrated the predominance of MC-RR and MC-YR and the absence of MC-LR, with only trace amounts of other variants (Paulino et al., 2017; Paulino et al., 2020). MCs in the CE were quantified as MC-LR equivalents (MC-LReq) using an Enzyme-Linked Immunosorbent Assay (ELISA) kit (Beacon Analytical Systems Inc., USA) and analyzed in a spectrophotometer (Molecular Devices Spectra, MAX GEMINI X, USA) at 450 nm absorbance.
Experimental design and tissue sample
Thirty-two juveniles of O. niloticus (body mass 47.54 ± 2.33 g and total length 14.73 ± 0.25 cm) were obtained from a fish farm in Santa Cândida (Santa Cruz da Conceição, São Paulo state, Brazil), and acclimated for 30 days in 1000 L tanks with natural photoperiod, constant aeration and continuous water flow at the Laboratory of Zoophysiology and Comparative Biochemistry, Department of Physiological Sciences, Federal University of São Carlos. Throughout acclimatization, temperature (24 ± 2 °C), dissolved oxygen (7.0-7.5 mg L^− 1^), pH (7.1–7.4), conductivity (125–130 µS cm^− 1^), alkalinity (35–43 mg L^− 1^ CaCO_3_), and hardness (39–50 mg L^− 1^ CaCO_3_) were monitored and maintained. The fish were fed ad libitum with commercial fish feed. After acclimatization, fish were randomly divided into four groups (n = 8) in 100 L aquariums: MC 12 h group (MC12), MC 96 h group (MC96), and two control groups in 12 h and 96 h (C12 and C96). Fish were fasted for 24 h prior to exposure and were not fed during the 96-h experimental period. The exposure was performed in a static system under a 12 h/12 h light–dark photoperiod using R. fernandoi crude extract at an environmentally relevant concentration of 10 µg MC-LReq L^− 1^, supported by reported microcystin levels in natural waters in Brazil and worldwide (Buley et al., 2021; Schneider et al., 2025). Throughout the exposure period, water quality parameters remained within the same ranges established during acclimation.
After exposure periods, fish were anesthetized with benzocaine solution (0.1 g L^− 1^). Blood samples (1 mL) were collected using heparinized syringes and were centrifuged under 10,000 rpm at 4 °C to separate plasma. The fish were euthanized by benzocaine overdose followed by spinal section, then gills and liver samples were immediately stored at -80 °C along with blood plasma for biochemical analyses. Subsamples of gills and liver were fixed in Bouin solution, for immunohistochemical analysis and in 2.5% glutaraldehyde for morphological analyses. For all subsequent analyses, a sample size of eight (n = 8) was employed. This project received approval University Federal of São Carlos Animal Ethics Committee (process 039/2014).
Histopathological analyses
Gill and liver samples were fixed in 2.5% glutaraldehyde in 0.1 M phosphate buffer (pH 7.3) were dehydrated in ethanol crescent series and embedded in historesin (Leica). Next, 3 μm-thick sections were made and stained with toluidine blue and basic fuchsin for histopathological analyses with methodology by Paulino et al. (2020), Histopathological analyses were conducted employing a randomized blind method with an Olympus BX51 microscope (Olympus, Denmark). This process included the evaluation of the tissue-specific alteration index (I_alt_), which is calculated by multiplying a numerical score (Sc), representing the frequency of alterations observed in the analyzed fields by a pathological importance factor (Fi), which reflects the severity of the organ lesion (I_alt_ = Sc x Fi). The histopathological index of the organ (I_org_) was determined by summing all the individual I_alt_ values (I_org_ = Σ I_alt_) to evaluate the morphofunctional impairment of the organ.
Enzymatic analyses
For normalization of enzymatic assays, the total protein content of the homogenates (mg protein^-1^ mL^-1^) was quantified using the Bradford method with Coomassie Brilliant Blue G-250 as the dye reagent (Bradford, 1976). Protein determination followed an adapted microplate protocol (Dynex Technologies Ltd., MRXTC, UK), described by Kruger (1994) with bovine albumin (1 mg mL^-1^) as a standard, was utilized. All samples were measured at 595 nm absorbance.
Endogenous liver PP1 and PP2A phosphatase activity
Liver samples were homogenized in an ice-cold buffer (40 mM Tris–HCl, 20 mM KCl, 30 mM MgCl_2_, pH 8.6 with 2 mM PMSF) at a ratio of 1:4 (w/v) using an Ultra Turrax homogenizer. The resulting homogenates were centrifuged at 12,000 g, 4 Cº for 15 min, with the obtained supernatants serving as sources for PP1 and PP2A enzymes. PP1 and PP2A activities (nmol min^− 1^ mg protein^− 1^) were assessed in a microplate reader utilizing p-nitrophenyl phosphate (pNPP) as a substrate. PP1 activity was determined according to Carmichael & An (1999) and PP2A activity was measured following the protocol of Heresztyn (2001), with adaptations described by Bieczynski et al. (2013). PP’s absorbance readings were taken at 405 nm, performed in triplicate at 10-minute intervals for 60 min at 25 ◦C.
Acid and alkaline plasmatic phosphatases activity
The activities of acid phosphatase (ACP) and alkaline phosphatase (ALP) were determined following the protocol described by Bergmeyer & Grassl (1983). Briefly, 100 µL of plasma were mixed with 100 µL of 120 mM p-nitrophenyl phosphate and incubated at 37 °C for 30 min. After incubation, the samples were centrifuged at 5000 g for 2 min. Subsequently, 400 µL of 2 M NaOH were added to stop the reaction. Enzymatic activity was measured at 405 nm absorbance using a spectrophotometer (DU^®^ 520, Beckman) and expressed as nmol p-nitrophenyl phosphate min⁻¹ mg protein⁻¹.
Na+/K+-ATPase and V(H+)-ATPase enzyme activity
The determination of Na^+^/K^+^-ATPase (NKA) and V(H^+^)-ATPase (VAT) (µATP mg protein-1 h^-1^) enzyme activity in gills followed the method of Gibbs and Somero (1989), adapted by Kültz and Somero (1995) and González-Cabo et al. (2005). Gill samples were homogenized in SEID buffer 0.5% (sucrose 150 mM; imidazole 50 mM; EDTA 10 mM; sodium deoxycholate 1%) and centrifuged at 3200 rpm for 7 min at 4 °C. The activity was determined by the total rate in relation to the difference in the rate of specific inhibitors (ouabain for NKA; N-ethylmaleimide for VAT). Readings were carried out with 5 µL of the sample homogenate and 200 µL of the reaction buffer (ATP 1 mM; NADH 0.2 mM; PK 3 U/mL; LDH 2 U/mL; fructose 0.1 mM; PEP 2mM), either pure or supplemented with ouabain 2 mM or N-ethylmaleimide 2 mM. Readings were taken at 340 nm at 28-second intervals for 10 min using a spectrophotometer (Molecular Devices Spectra, MAX GEMINI X, USA).
Carbonic anhydrase enzyme activity in the gills
The determination of the carbonic anhydrase (CA) enzyme activity in the total gill homogenate [(rate of catalyzed reaction / rate of non-catalyzed reaction) mg protein⁻¹] in gills followed the method of Henry (1991) adapted by Vitale et al. (1999). Gill samples were homogenized in a reaction buffer (mannitol 225 mM; sucrose 75 mM; tris base 10 mM; trisphosphate 10 mM, pH 7.4). After adding 50 µL of the homogenate to 7.5 mL of the reaction buffer and stabilizing the pH, the analysis was initiated by adding 1 mL of chilled saturated CO_2_ water (2.0 to 4.0 °C). Readings were taken every 4 s for 20 s on a pH meter (Quimis Q-400 A). CA activity was determined by the ratio between the slopes of linear regressions for non-catalyzed (blank) and catalyzed reactions.
Immunohistochemistry of gill ionocytes
Gills samples fixed in Bouin’s solution were dehydrated in an increasing ethanol series (50% − 100%) and clarification in 100% xylene before being embedded in histological paraffin. 8 μm-thick sections were cut using a microtome (MICRON HM 360). The slides were deparaffinized in xylene 100% (2x) for 5 min and hydrated in a decreasing ethanol series (100 − 50%) for 3 min. Afterward, the slides were rinsed (2x) for 10 min in tris-buffered saline with Triton 0.1%, pH 7.4 (TBS-T), diluted 1:10, with continuous agitation. A 20-minute pre-incubation in 20% normal goat serum (NGS - Gibco, Invitrogen) diluted in TBS-T preceded the incubation of the slides with the primary antibody (α5) - anti Na^+^/K^+^-ATPase, diluted in 0.1% TBS-T, overnight in a humid chamber at 20 °C.
Following incubation, the slides were rinsed (2x) for 10 minutes in diluted TBS-T and were then incubated with the secondary antibody (Goat Anti-Mouse Peroxidase - GAMPO - Chemicon International, USA), diluted 1:100 for 1 hour in a humid chamber. After incubation, washing was carried out with tris buffer, pH 7.4 (TB), diluted 1:10. The complex was visualized using DAB-Ni 0.5% (3,3’-diaminobenzidine + Nickel Ammonium Sulfate, Sigma-Aldrich) and hydrogen peroxide (H_2_O_2_). Staining was stopped in deionized water (2x) for 10 min. The slides were sealed with Entellan^®^ and examined using Olympus DP2 – B5W software on a light microscope (BX 51, Olympus, Denmark). The ionocytes counting in the filament and lamellae (number per mm^2^ of epithelium) was conducted in 20 random fields, according Paulino et al., (2012) using Motic Image Plus 2.0 software.
Plasmatic ions and total osmolality
Na^+^ (mEq L^− 1^) and K^+^ (mEq L^− 1^) concentrations were determined using a flame photometer (DIGMED DM-61), with a 1:100 ratio of plasma to deionized water. The concentration of Cl^−^ and Ca^2+^ was determined using commercial kits (LABTEST, ref.49 and ref. 95 respectively) at an absorbance of 490 nm for Cl^−^ and 505 nm for Ca^2+^ in a microplate reader (SpectraMax M5, Molecular Devices). Plasma osmolality (mOsmol Kg^− 1^) was determined using a semi-micro osmometer (µOSMETTE PRECISION SYSTEM), based on the freezing point.
Statistical analyses
All data were analyzed using GraphPad Prism 8.0 software. Normality was tested with the D’Agostino & Pearson test. Differences between the MC group and its respective control were assessed using the unpaired Student’s t-test (p ≤ 0.05) for parametric data and Mann-Whitney test (p ≤ 0.05) for nonparametric data. Group comparisons were analyzed with the ANOVA one-way test (p ≤ 0.05) with Tukey post-hoc for parametric data and Kruskal-Wallis test (p ≤ 0.05) with Dunns post-hoc for nonparametric data.
Results
During both exposures (12 and 96 h) of O. niloticus to R. fernandoi R28 CE, no mortality was observed, and the behavior of the exposed animals remained consistent with the control group.
Tissue damage analysis
Hepatic PP1 and PP2A activities exhibited no significant changes (Fig. 1). ALP activity showed no differences between groups. However, ACP showed an increase in 96-hour exposure when compared to 12-hour exposure (Fig. 2).
Fig. 1PP1 (A) and PP2A (B) activities in the liver of Oreochromis niloticus exposed to 10 µg MC-LReq L^-1^ in the crude extract of Radiocystis fernandoi R28 after 12 and 96 h. (C12) 12 h control group; (MC12) MC 12 h group; (C96) 96 h control group; (MC96) MC 96 h group. Lowercase letters indicate statistical differences between groups at the 12-hour, while uppercase letters indicate statistical differences between groups at the 96-hour (p < 0.05). Boxes represent the interquartile range (Q1–Q3), the central line indicates the median, whiskers represent minimum and maximum values, and dots indicate individual observations
Fig. 2ACP (A) and ALP (B) activities in the plasm of Oreochromis niloticus exposed to 10 µg MC-LReq L^-1^ in the crude extract of Radiocystis fernandoi R28 after 12 and 96 h. (C12) 12 h control group; (MC12) MC 12 h group; (C96) 96 h control group; (MC96) MC 96 h group. Lowercase letters indicate statistical differences between groups at the 12-hour, while uppercase letters indicate statistical differences between groups at the 96-hour (p < 0.05). Boxes represent the interquartile range (Q1–Q3), the central line indicates the median, whiskers represent minimum and maximum values, and dots indicate individual observations
The histopathological analyses (Table 1; Figs. 3 and 4) showed that 12-hours exposition can induce nuclear alterations in the hepatocytes, and in the gills showed mucous cell’s proliferation. Within 96 h exposure was observed alterations in pillar cell’s structure alteration, with reflects in gills I_org_ (Fig. 5).
Table 1. Mean ± standard error of tissue-specific index alteration (I_alt_) in gills and liver of Oreochromis niloticus after 12 and 96 h of exposure to 10 µg MC-LReq L⁻¹ from the crude extract of Radiocystis fernandoi R28Tissue-specific alteration index (I_alt_)C12MC12C96MC96 Gill pathology Aneurysm0.2 ± 0.2^a^0^a^0^a^0^a^Cellular Atrophy0.4 ± 0.4^a^0^a^0^a^0.8 ± 0.5^a^Congestion0.2 ± 0.2^a^0^a^0^a^0^a^Lamellar epithelium hypertrophy4.8 ± 0.8^a^4.8 ± 0.5^a^1.0 ± 0.6^a^4.0 ± 0.9^a^Epithelium detachment and edemas0.4 ± 0.2^a^0.4 ± 0.2^a^0.5 ± 0.3^a^0.6 ± 0.2^a^Total lamellar fusion0^a^0^a^0^a^0.8 ± 0.5^a^Partial lamellar fusion0.2 ± 0.4^a^0.4 ± 0.2^a^0.2 ± 0.2^a^0.6 ± 0.2^a^Lamellar epithelium hyperplasia3.2 ± 1.2^a^1.6 ± 0.4^a^1.5 ± 0.5^a^3.2 ± 0.8^a^Hypertrophy of ionocytes1.2 ± 0.5^a^2.0 ± 0.6^a^1.0 ± 0.6^a^1.6 ± 0.7^a^Mucous cell’s proliferation1.0 ± 0.3^a^ 3.6 ± 0.2 ^b^ 1.5 ± 0.3^a^2.2 ± 0.6^a^Rupture of epithelium1.0 ± 0.0^a^ 0.2 ± 0.2 ^b^ 1.5 ± 0.3^a^ 0.2 ± 0.2 ^b^ Pillar cell’s structure alteration1.8 ± 0.6^a^2.6 ± 0.2^a^0^a^ 1.2 ± 0.2 ^b^
Liver pathology Intracellular substance1.2 ± 0.2^a^1.0 ± 0.0^a^1.0 ± 0.4^a^1.4 ± 0.2^a^Nuclear alterations0.4 ± 0.4^a^ 1.6 ± 0.4 ^b^ 2.0 ± 0.0^a^1.6 ± 0.7^a^Hepatic architecture alteration1.0 ± 0.0^a^1.0 ± 0.0^a^0.7 ± 0.2^a^1.0 ± 0.0^a^Cellular atrophy0.4 ± 0.4^a^0^a^0.5 ± 0.5^a^0^a^Melano-macrophage centers0.4 ± 0.4^a^0^a^0.5 ± 0.5^a^0.5 ± 0.0^a^Cytoplasmic degeneration1.0 ± 0.0^a^1.2 ± 0.2^a^1.0 ± 0.0^a^1.0 ± 0.0^a^Necrosis1.8 ± 0.7^a^0.6 ± 0.6^a^0.7 ± 0.7^a^1.8 ± 0.7^a^Different letters indicate statistically distinct groups (p < 0.05), comparisons were made with the respective control groups: (C12) 12 h control group; (MC12) 12 h MC-exposed group; (C96) 96 h control group; (MC96) 96 h MC-exposed group
Fig. 3. Representative histopathological alterations in the gills and liver of Oreochromis niloticus exposed to 10 µg MC-LReq L^-1^ in the crude extract of Radiocystis fernandoi R28 after 12 and 96 h. (A) Normal gill morphology: pillar cells (arrows), pavement cells (arrowheads). (B) Disorganization of pillar cells (arrows). (C) Proliferation of mucous cells in the interlamellar regions (arrows). (D) Hypertrophy of pavement cells and epithelial rupture (arrows). (E) Normal liver morphology; “H” indicates hepatocytes, “S” indicates sinusoids. (F) Cytoplasmic degeneration (arrows) and increased intracellular inclusions (arrowheads). Staining: toluidine blue (A–D), toluidine blue and basic fuchsin (E–F). Scale bar: 20 μm
Fig. 4. Organ pathological index (I_org_) of (A) liver and (B) gills of Oreochromis niloticus exposed to 10 µg MC-LReq L^-1^ in the crude extract of Radiocystis fernandoi R28 after 12 and 96 h. (C12) 12 h control group; (MC12) MC 12 h group; (C96) 96 h control group; (MC96) MC 96 h group. Lowercase letters indicate statistical differences between groups at the 12-hour, while uppercase letters indicate statistical differences between groups at the 96-hour (p < 0.05). Boxes represent the interquartile range (Q1–Q3), the central line indicates the median, whiskers represent minimum and maximum values, and dots indicate individual observations
Osmoregulation and ionic balance
Plasma ion concentrations and total osmolality (Table 2) also demonstrated stability, except for a reduction in Cl^−^ concentration after both exposure durations. NKA activity (Fig. 6A) showed no discernible alterations following exposure, while AC and VAT activities in the total gill homogenate (Fig. 6B and C) increased after 12 h of exposure, but returned to normally in 96 h of exposure. However, VAT showed an increase in both control and exposed groups in 96 h of exposure.
Table 2. Mean ± standard error of osmolality and plasma ion concentrations of Oreochromis niloticus after 12 and 96 h of exposure to 10 µg MC-LReq L⁻¹ from the crude extract of Radiocystis fernandoi R28Osmoregulatory ParametersC12MC12C96MC96Total osmolality (mOsmol Kg^− 1^)307.8 ± 4.7^a^316.1 ± 3.5^a^308.1 ± 1.1^a^313.6 ± 2.7^a^Íons (mEq/L^− 1^)Na^+^197.2 ± 3.8^a^198.0 ± 2.8^a^181.6 ± 3.5^a^186.6 ± 2.8^a^K^+^1.33 ± 0.2^a^1.15 ± 0.2^a^1.15 ± 0.2^a^1.6 ± 0.1^a^Cl^−^136.1 ± 2.4^a^ 124.0 ± 3.0 ^b^ 132.7 ± 1.8^a^ 105.6 ± 9.5 ^b^ Ca^2+^15,80 ± 2,1^a^15,47 ± 2,8^a^15,36 ± 1,5^a^15,09 ± 2,0^a^Different letters indicate statistically distinct groups (p < 0.05), comparisons were made with the respective control groups: (C12) 12 h control group; (MC12) 12 h MC-exposed group; (C96) 96 h control group; (MC96) 96 h MC-exposed group
Fig. 5. Specific activity of (A) Na^+^-K^+^-ATPase and (B) Carbonic anhydrase and (C) V(H^+^)-ATPase in the gills of Oreochromis niloticus exposed to 10 µg MC-LReq L^-1^ in the crude extract of Radiocystis fernandoi R28 after 12 and 96 h. (C12) 12 h control group; (MC12) MC 12 h group; (C96) 96 h control group; (MC96) MC 96 h group. U mg protein⁻¹ = [(rate of the catalyzed reaction / rate of the non-catalyzed reaction) mg protein⁻¹]. Lowercase letters indicate statistical differences between groups at the 12-hour, while uppercase letters indicate statistical differences between groups at the 96-hour (p < 0.05). Boxes represent the interquartile range (Q1–Q3), the central line indicates the median, whiskers represent minimum and maximum values, and dots indicate individual observations
Immunohistochemical examination (Fig. 7) and quantification of NKA ionocytes per mm² in the gill filament (Fig. 7A) and lamella (Fig. 7B) of O. niloticus indicated an increase in ionocytes in the lamella after 12 h of exposure, while the filament exhibited no significant change in ionocyte number.
Fig. 6. Immunohistochemistry of gill NKA ionocyte of Oreochromis niloticus exposed to 10 µg MC-LReq L^-1^ in the crude extract of Radiocystis fernandoi R28 after 12 and 96 h. (A) Control Group 12 h. (B) Microcystin Group 12 h; (C) Control Group 96 h. (D) Microcystin Group 96 h. Arrows indicate an increase in the quantity of ionocytes in the lamellae after exposure. I = Ionocyte. F = Filament. L = Lamellae. CVS = Central Venous Sinus. Scale bar = 20 μm
Fig. 7. Ionocytes count in (A) filament and (B) lamellae of the gills of Oreochromis niloticus exposed to 10 µg MC-LReq L^-1^ in the crude extract of Radiocystis fernandoi R28 after 12 and 96 h. (C12) 12 h control group; (MC12) MC 12 h group; (C96) 96 h control group; (MC96) MC 96 h group. Lowercase letters indicate statistical differences between groups at the 12-hour, while uppercase letters indicate statistical differences between groups at the 96-hour (p < 0.05). Boxes represent the interquartile range (Q1–Q3), the central line indicates the median, whiskers represent minimum and maximum values, and dots indicate individual observations
Discussion
In a eutrophic environment, fish face potential exposure to MCs not only through the ingestion of toxic cyanobacteria or contaminated food but also passively, through the dissolved toxins in the water in direct contact with the gill epithelium (Shingai & Wilkinson, 2023). As a result, distinct toxic effects may be expected from those reported in the literature, due the fact that most studies describe MC toxicity in vitro, through intraperitoneal (i.p.) injections, or dietary routes (Falfushynska et al., 2023; Paulino et al., 2020). Nevertheless, this study provides relevant evidence on the toxic effects of exposure to cyanobacterial CE via the gills, highlighting their potential to disrupt osmoregulatory homeostasis in juvenile freshwater fish.
Gills are an important target tissue for early responses to environmental contaminants due to their direct and continuous contact with the surrounding water. Although the gastrointestinal tract is often considered the main route of MCs uptake when toxins are associated with ingested food or cyanobacterial cells (Falfushynska et al., 2023), the gill epithelium may also contribute to the uptake of dissolved, extracellular MCs under waterborne exposure scenarios (Mielewczyk et al., 2023). This pathway may be particularly relevant for freshwater teleosts such as O. niloticus, which maintain osmotic balance primarily by limiting voluntary water intake (Grosell & Oehlert, 2023). Experimental evidence indicates that waterborne MCs can interact with the gill surface and may enter branchial cells through organic anion transporting polypeptides (Oatps) expressed in the gill epithelium (Steiner et al., 2016). Moreover, exposure to the R. fernandoi R28 CE, may better reflect environmentally realistic conditions, where multiple MC variants typically coexist (Paulino et al., 2020). In this context, the gills appear to be a sensitive site of interaction with the crude extract, with observed alterations suggesting potential disturbances in osmoregulatory processes and tissue integrity, even under conditions of limited systemic uptake.
The acute exposure to R. fernandoi CE was associated with an early increase in mucous cell abundance in the gills of O. niloticus, particularly within the first 12 h. Such a response is commonly interpreted as an initial protective adjustment to environmental stressors, given that branchial mucus can act as a physical and biochemical interface that limits direct contact between contaminants and the respiratory epithelium (Mokhtar et al., 2023; Zink & Wood, 2024). This response may explain the relatively mild and predominantly reversible histopathological changes observed, which did not result in a significant increase in the I_org_. In contrast, more pronounced branchial lesions have been reported following i.p administration of microcystins (Tavares et al., 2019), a route that bypasses external branchial defenses and promotes direct systemic exposure. This contrast may indicate a potential involvement of the mucosal barrier during waterborne exposures, whereby localized branchial responses could be associated with a moderation of the immediate effects of the toxin on gill tissue.
Our study recorded a low incidence of liver histopathologies, no changes in I_org_ values, absence of PP1 and PP2A inhibition, and no significant differences in plasmatic ALP activity in both 12-hour and 96-hours exposure. These findings suggest that, under our experimental conditions, the defense mechanisms of O. niloticus were effective in limiting microcystin entry and preventing acute systemic toxicity, which may be associated with the reversible binding of MCs to PP1 and PP2A (Lin et al., 2021) and/or with the species’ efficient detoxification system, capable of metabolizing MCs into less toxic forms through phase I and II enzymes, such as cytochrome P4501A and glutathione S-transferase (Wang et al., 2006). This detoxification system has developed due to its planktivorous feeding habits, including cyanobacteria such as Microcystis aeruginosa (He et al., 1997).
Despite the activation of defense mechanisms, 96-hour exposure led to more pronounced toxic effects, evidenced by a significant alteration in the gill cellular architecture, resulting in elevated I_org_. The morphological changes observed at 96 h (total lamellar fusion, epithelial hyperplasia, and disruption of pillar cell structure) indicate a substantial reorganization of gill tissue and may reflect a plastic physiological adjustment that temporarily reduces contaminant uptake across the gill epithelium (Marinović et al., 2021). However, such structural modifications involve a clear trade-off, as the reduction in functional respiratory surface area is likely to compromise oxygen uptake and contribute to respiratory and osmoregulatory dysfunction (Tavares et al., 2019).
The osmoregulatory and acid–base regulatory processes in freshwater fish are tightly integrated and depend on the coordinated function of key enzymes, including NKA, VAT, and CA, which are localized in specialized cells known as ionocytes and whose activity is widely used as an important biomarker of fish health (Maraschi et al., 2024; Zink & Wood, 2024). In freshwater species, despite interspecific variability, ionocytes are typically distributed along both the filaments and lamellae, increasing the epithelial surface area in contact with the external environment and thereby, enhancing ion uptake (Maraschi et al., 2024). Within these cells, NKA is predominantly located in the basolateral membrane, whereas VAT and CA are generally localized in the apical membrane. Intracellular CA catalyzes the reversible hydration of CO₂, producing H⁺ and HCO₃⁻. These ions are then utilized by apical VAT to generate an electrochemical gradient that facilitates Na⁺ uptake into the cell (Maraschi et al.,2024).
Given that MCs can disrupt ionic homeostasis in freshwater fish (Falfushynska et al., 2023), modulations in ion-transporting enzymes are expected (Tavares et al., 2019). In this study, acute exposure to R. fernandoi crude extract induced an ionic disturbance, and a rapid yet transient osmoregulatory and acid–base response in the gills of O. niloticus, characterized by a significant reduction in plasma Cl⁻ concentrations at both 12 and 96 h, while plasma Na⁺, K⁺, and total osmolality remained stable. Simultaneously a transient increase in branchial CA and VAT activities at 12 h, whereas NKA activity remained unchanged throughout the exposure period. This temporal pattern is consistent with the main mechanism of MC toxicity, which peaks between 3- and 6-hours post-exposure, followed by a recovery period lasting up to 96 h (Tavares et al.,2019).
The maintenance of NKA activity indicates that basolateral cation transport and the electrochemical gradients required for cellular homeostasis were preserved during toxin exposure, consistent with the unchanged plasma Na⁺ and K⁺ levels. Moreover, NKA activity represents a relatively stable indicator of ionocyte functional capacity and typically requires longer time scales, involving transcriptional and translational regulation, for up-regulation to be reflected as a measurable change in total enzymatic activity (Bystriansky & Schulte, 2011). Importantly, the stability of NKA activity does not preclude the occurrence of an osmoregulatory disturbance; rather, it suggests that early compensatory responses were preferentially associated with anion transport and acid–base regulatory pathways, potentially mediated by CA and VAT activity.
The observed decrease in plasma Cl⁻ in the absence of changes in total osmolality or Na⁺ concentrations indicate the activation of compensatory mechanisms aimed at preserving electroneutrality and osmotic balance. This scenario can be attributed to the compensation by unmeasured osmolytes, specifically an increase in plasma HCO₃⁻ (Kreiss et al., 2015). In freshwater teleosts, Cl⁻ uptake is functionally coupled to HCO₃⁻ secretion via apical Cl⁻/HCO₃⁻ exchangers, with intracellular CA supplying the necessary H⁺ and HCO₃⁻ equivalents through CO₂ hydration (Kovac & Goss, 2024).
Accordingly, a reduction in Cl⁻ availability or changes in its transport kinetics are expected to modify the local acid–base environment at the gill epithelium, increasing the physiological demand for HCO₃⁻ production to sustain Cl⁻/HCO₃⁻ exchange and maintain systemic pH (Griffith, 2017; Kovac & Goss, 2024). The transient increase in branchial CA activity observed at 12 h is consistent with this requirement, as CA accelerates the intracellular generation of HCO₃⁻. Although plasma pH and HCO₃⁻ concentrations were not directly measured in the present study, Cl⁻/HCO₃⁻ exchange is a well-established and primary physiological mechanism for acid-base regulation in teleost when challenged by environmental or physiological stressors (Shartau et al., 2020; Montgomery et al., 2022). Simultaneously, the elevation of VAT activity supports this compensatory response by facilitating apical H⁺ extrusion, thereby sustaining the electrochemical gradients that indirectly favor Na⁺ uptake and overall ionoregulatory stability (Griffith, 2017; Kovac & Goss, 2024).
These findings indicate a compensatory activation of branchial ion transport and acid–base regulatory mechanisms following microcystin-induced gill impairment. Such impairment has been associated with cardiorespiratory dysfunction and oxidative damage in the gills, conditions that can secondarily disrupt acid–base balance and promote the development of respiratory acidosis in exposed fish (McArley et al., 2021; Svirčev et al., 2022). In this context, the increases in CA and VAT activity at 12 h are consistent with an early physiological adjustment aimed at supporting branchial ion transport and acid–base regulation during the acute phase of exposure (Maraschi et al., 2024).
The transient increase in lamellar ionocyte abundance observed at 12 h further supports this interpretation, suggesting a rapid modulation or redistribution of ionocytes, which constitute the primary sites of ionoregulation in freshwater fish (Zink & Wood ,2024). While these adjustments appear effective in maintaining overall osmotic balance in the short term, they are energetically demanding. Osmoregulation and acid–base regulation represent substantial components of the metabolic budget in freshwater fish, particularly under stress conditions, and sustained activation of these pathways may impose significant energetic costs (Su et al., 2022).
The substantial energy allocated to osmoregulation, particularly under stressful conditions like toxin exposure, can represent a significant portion of the fish’s total energy budget (Blondeau-Bidet et al., 2024). Such high energetic demands often necessitate physiological trade-offs, potentially diverting resources from other critical functions. For instance, increased osmoregulatory costs under environmental stress have been associated with compromised immune responses, as suggested by the downregulation of immune-related genes observed in fish acclimating to challenging salinities (Blondeau-Bidet et al., 2024). Furthermore, prolonged osmotic stress can negatively impact growth and reproduction (Agarwal et al., 2024). In the context of our findings, the compensatory activation of CA and VAT, alongside ionocyte proliferation observed at 12 h, undoubtedly incurred an energetic cost (Simó-Mirabet et al., 2025). While these mechanisms appeared effective in maintaining overall ionic balance initially, the sustained energetic drain might contribute to the increased tissue damage and stress indicators (e.g., elevated ACP, higher gill I_org_) observed at 96 h. This suggests that prolonged allocation of energy towards osmoregulation under MC stress can impair other vital physiological processes, ultimately impacting fish health and resilience in contaminated environments.
Conclusion
Our study demonstrates that acute waterborne exposure to a crude extract of Radiocystis fernandoi R28, containing the microcystin variants MC-RR/YR, induces significant toxicological responses in juvenile Nile tilapia, even in the absence of mortality or hepatotoxicity. The gills were identified as the primary target organ, showing time-dependent histopathological changes and functional osmoregulatory disturbances. These included persistent reductions in plasma Cl⁻ and transient compensatory responses, such as increased branchial carbonic anhydrase activity and lamellar ionocyte abundance. The findings underscore the vulnerability of gill tissue to direct toxicity from less-studied MC congeners and complex mixtures present in crude extracts, thereby challenging the traditionally liver-centered view of MC toxicity. The ability of R. fernandoi extract to impair gill function and disrupt ionic balance in a resilient species such as O. niloticus highlights the ecological risks posed by cyanobacterial blooms in eutrophic waters. Future research should focus on the molecular mechanisms underlying MC-RR/YR-induced gill toxicity and the chronic effects of exposure to these toxins and their complex mixtures in fish populations.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Bieczynski F, Bianchi VA, Luquet CM. (2013). Accumulation and biochemical effects of microcystin-LR on the Patagonian pejerrey (Odontesthes hatcheri) fed with the toxic cyanobacteria Microcystis aeruginosa. Fish Physiol Biochem. 39(5):1309–1321 10.1007/s 10695-013-9785-7
- 2Blondeau-Bidet E, Tine M, Gonzalez AA, Guinand B, Lorin-Nebel C. (2024). Coping with salinity extremes: Gill transcriptome profiling in the black-chinned tilapia (Sarotherodon melanotheron). Sci Total Environ. 929:172620. 10.1016/j.scitotenv.2024.172620
- 3Maraschi AC, Gutierre SMM, Prodocimo V. (2024). Osmoregulation and HSP 70 abundance in freshwater catfish Rhamdia quelen under acute salt exposure. Environ Biol Fishes. 10.1007/s 10641-024-01556-9
- 4Montgomery DW, Kwan GT, Davison WG, Finlay J, Berry A, Simpson SD, Engelhard GH, Birchenough SNR, Tresguerres M, Wilson RW. (2022). Rapid blood acid–base regulation by European sea bass (Dicentrarchus labrax) in response to sudden exposure to high environmental CO 2. J Exp Biol. 225 (2): jeb 242735. 10.1242/jeb.242735
- 5Paulino MG, Sakuragui MM, Fernandes MN. (2012). Effects of atrazine on the gill cells and ionic balance in a neotropical fish, Prochilodus lineatus. Chemosphere. 86(1):1–7. 10.1016/j.chemosphere.2011.08.033
- 6Paulino MG, Tavares D, Bieczynski F, Pedrão PG, Souza NES, Sakuragui MM, Luquet CM, Terezan AP, Fernandes JP, Giani A, Fernandes MN. (2017) Crude extract of cyanobacteria (Radiocystis fernandoi, strain R 28) induces liver impairments in fish. Aquat Toxicol. 182: 91–101. 10.1016/j.chemosphere.2017.02.089
- 7Pereira DA, Pimenta ADC, Giani A. (2012). Profiles of toxic and nontoxic oligopeptides of Radiocistis fernandoii (Cyanobacteria) exposed to three different light intensities. Microbiol. Res. 167(7):413– 421. 10.1016/j.micres.2012.02.007
- 8Simó-Mirabet P, Caderno A, Flores-Llano MJ, Silva EG, Schoenau W, Baldisserotto B, Martínez-Rodríguez G, Mancera JM, Martos-Sitcha JA. (2025). Evaluation of the sedative effect of limonene to reduce stress during transportation of gilthead seabream (Sparus aurata). Biology. 14(2):115. 10.3390/biology 14020115