Acute Effects of Microcystins MC-LR and MC-RR on Acid and Alkaline Phosphatase Activities and Pathological Changes in Intraperitoneally Exposed Tilapia Fish ( Oreochromis sp. )

Abstract

Microcystins (MC) are frequently present in cyanobacterial blooms in rivers and lakes, increasing the risk of toxicity to both animals and humans. There more than eighty reported microcystins, and the present study was undertaken to determine whether MC-LR and MC-RR can induce different enzyme alterations and histopathological changes in tilapia fish (Oreochromis sp.) exposed to a single intraperitoneal (i.p.) injection of the pure standards (MC-LR and MC-RR) at a dose of 500 μg/kg; the tilapia fish were then observed for seven days. The two MC variants caused significant changes in the activities of acid and alkaline phosphatases (ACP and ALP) in vital organs, showing a different response pattern. The livers and kidneys of fish injected with MC-LR were particularly affected. MC-RR induced a very pronounced increase of ACP in the kidney and a significant increase of ALP in the liver. Both MC variants caused pathological lesions in hepatic tissues, such as megalocytosis, necrotic process, and microvesicular steatosis, particularly in fish treated with MC-LR, and degenerative renal changes, glomerulopathy, were more severe in tilapias exposed to MC-RR. In addition, both microcystins also caused significant myopathy in the heart. In contrast, the gills did not show any change in enzyme activity or histopathological injury.

Introduction

The eutrophication of lakes and reservoirs leads to water blooms of cyanobacteria in many countries throughout the world. This is a great concern to society, because blooms not only decrease water quality, but also increase the risk of toxicity to both animals and humans because they produce microcystins (MCs) (Falconer 1999; Li et al. 2005). There are more than eighty MCs that have a ring structure of seven amino acids, which compose one unique phenyl decadienoic acid, four invariable D-amino acids, and two variable L-amino acids. MC-LR (MC lysine, arginine) is the most widely investigated cyanobacterial peptide toxin because it is frequently present in cyanobacterial blooms in rivers and lakes (Sivonen and Jones 1999). Other variants that also occur frequently are MC-RR (MC arginine, arginine), MC-YR, and MC-LA (de Figueiredo et al. 2004). Replacing hydrophobic L-Leu in the first variable position (position 2) with another hydrophobic L-amino acid (e.g., alanine, phenylalanine, or tryptophan) maintains toxicity, but replacing it with a hydrophilic amino acid (e.g., arginine) dramatically reduces it. Thus, MCs containing polar substitutions in both variable amino acid positions, such as MC-RR and MC-M(O)R (methionine sulfoxide, arginine) are the least toxic (Zurawell et al. 2005).

Particular attention has been paid to these hepatotoxins not only because of their ability to cause acute poisonings to those aquatic organisms, wildlife, domestic animals, and humans that drink or ingest the algae in the water (Carmichael 1996), but also because chronic exposure of humans to low microcystin concentrations may promote cancer. Indeed, the International Agency for Research on Cancer (IARC) has classified MC-LR as possibly carcinogenic to humans (Group 2B), although microcystis extracts are not (http:www.iarc.fr/).

It is well established that the main target of MCs in mammals is the liver, primarily because of toxin update by the bile acid transport system (Hermansky and Stohs 1991; Tencalla et al. 1994). Cell uptake studies with microcystins have demonstrated that they require cell-membrane–associated organic anion transporter proteins (OATP) (Fischer et al. 2005). On entering the cells, picomolar concentrations of the toxins bind to and inhibit protein phosphatases 1 and 2A, which affects the regulation of cell protein phosphorylation (Boaru et al. 2006; Maidana et al. 2006).

Aquatic animals such as zooplankton, fish, and mollusks consumed by humans have been reported to bioaccumulate MCs (Amorim and Vasconcelos 1999; Freitas de Magalhães et al. 2003; Williams et al. 1997). Cyanobacteria are an important dietary component of many tropical cichlids (e.g., tilapia, Oreochromis niloticus) and cyprinids (e.g., silver carp, Hypophthalmichthys molitrix) (Zurawell et al. 2005). Both tilapia and silver carp are filter-feeding fish and are especially important to humans because of their roles in aquatic ecosystems as direct consumers of phytoplankton, their importance as food fish, and their potential for the biological management of cyanobacterial blooms (Chen et al. 2006).

In general, the gross histopathological and ultrastructural changes in various fish species following acute exposure to either M. aeruginosa and/or purified MC analogs are similar in many respects to those seen in mammals (Zurawell et al. 2005). Nevertheless, several dissimilarities do exist. In addition to severe damage and dysfunction of the liver, MCs also induce pathological changes in the kidney, gills, and gastrointestinal tract (Carbis et al. 1997; Fischer and Dietrich 2000; Kotak et al. 1996; Rabergh et al. 1991), as well as cardiac alterations in fish (Best et al. 2001). However, species-specific differences in MC tolerance are indicated by the degree of liver damage and liver enzyme activities in the blood of exposed fish.

Previous studies in our laboratory showed that tilapia (Oreochromis sp.) exposed subchronically to cyanobacterial cells (60.0 μg MC-LR/fish per day) under laboratory conditions experienced changes in their enzymatic activities of acid and alkaline phosphatases (ACP and ALP) from liver, kidney, and gill tissues in response to MCs in a time-dependent manner. Simultaneously, histopathological changes were observed in these organs (Molina et al. 2005).

Therefore, and because to date no studies have evaluated the capacity of various forms of MCs to generate toxic effects in aquatic organisms, the present study was undertaken to determine whether MC-LR and MC-RR induce different alterations in enzymes such as acid phosphatase (ACP) and alkaline phosphatase (ALP) and different histopathological changes. The toxicological data obtained for both MC variants would greatly reduce the degree of uncertainty and improve the risk assessment process.

Material and Methods

Chemicals

The cyanobacterial hepatotoxins MC-LR and MC-RR (purity 95–99%) were supplied by Cyanobiotech (Berlin, Germany). Standard stock solutions (500 μg/mL) and working standard solutions of each toxin were prepared in saline (0.9% NaCl), and the correct dosing concentration was confirmed by high-performance liquid chromatography (HPLC) with diode array detection, using a Varian model 9012 chromatograph equipped with a Varian ProStar model 330 PDA detector. Chromatographic separation of MCs was performed on a 250 mmx 4.6 mm i.d., 5 μm LiChrosphere C18 column from Merck (Darmstadt, Germany). Chromatographic determination was performed under isocratic conditions with a mobile phase consisting of 38% MeCN in water with 0.05% TFA (Moreno et al. 2004). The wavelength was set at 240 nm, and the flow rate was 1 mL/min. In these conditions the limits of detection (LODs) were 0.3 μg/mL for MC-LR and 0.4 μg/mL for MC-RR. Methanol and ethanol were from Merck and bovine γ-globuline from Biomol (Plymouth, Pennsylvania, USA). Mannitol, tricaine methanesulfonate (MS-222), glutaraldehyde, HEPES, TRIS, and all other reagents used were from Sigma (Steinheim, Germany).

Animals

Studies were conducted using male Oreochromis sp. (Nile tilapia, Perciformes: Cichlidae) with a mean weight of 54 ± 7 g. Fish were obtained from a fish hatchery in Córdoba (Spain). The fish were acclimated to laboratory conditions for fifteen days before the experiments were begun, in aquariums (8 individuals/aquarium) with 96 L of fresh water. Filling the tanks at least three days before the fish were introduced minimized exposure to chlorine. Fish were kept at a water temperature of 21 ± 2°C in water under continuous filtration and aeration (Eheim Liberty 150 and Bio-Espumador cartridges [Bio-Espumador]). Dissolved oxygen values were maintained between 6.5 and 7.5 mg/L, and water quality was determined for nitrite (Nitrite Test Kit, Aquarium Pharmaceuticals, Inc., UK) at the beginning of the experimental period and once a week thereafter. Mean values for additional parameters of water quality were: pH 7.6 ± 0.2, conductivity 292 μS/cm, Ca²⁺ 0.60 mM/L, and Mg²⁺ 0.3 mM/L. Fish were fed once every morning with commercial fish food that was kindly supplied by the fish hatchery (Dibaq, Segovia, Spain).

Exposure

The experiment was carried out using three aquaria with eight fish in each. Tilapias were exposed intraperitoneally (i.p.) to a single dose of 500 μg/kg MC-LR or 500 μg/kg MC-RR in 0.5 mL of 0.9% (w/v) NaCl solution. The third group of fish received only the vehicle solution (0.5 mL of 0.9%, w/v, NaCl) and was used as a control. The dose was selected in accordance with those reported to produce acute toxic effects in fish after i.p. exposure to MCs (Kotak et al., 1996; Rabergh et al., 1991; Tencalla et al., 1994). Fish were observed daily and after seven days were sacrificed. At the end of the experiment, the fish were anaesthetized by immersion in 50 mg/L MS-222 solution for 5 to 10 minutes before they were killed by transaction of the spinal cord. The liver, kidney, and gills were removed, weighed, rinsed with ice-cold saline, and kept at −85°C until biochemical analysis. The rest of the tissue samples were fixed so that the histopathological study could be carried out.

Determination of the Enzyme Activity in Liver, Kidney, and Gills

For the estimation of enzyme activities, liver, kidney, and gill tissues were weighed, homogenized using an Ultra-Turrax tissue homogenizer at 4ºC in a buffer containing 100 mmol/L mannitol, 2 mmol/L HEPES/TRIS, pH 7.1, 0.1 mmol/L benzamidine, and 0.2 mmol phenylmethyl-sulfonyl fluoride. The use of ACP as a marker enzyme for lysosomal membranes and ALP as an apical membrane enzyme was analyzed using the methods of Pennington (1961) and Bretaudiere et al. (1977), respectively, previously optimized in our laboratory. The protein content of the tissues was quantified using Bradford’s method (1976) and with bovine γ-globuline as the standard. The optical density was measured spectrophotometrically at 400 nm (ALP) and 405 nm (ACP), using a Cary-100 (Varian, USA) spectrophotometer. Enzyme activity was expressed as nmol/mg protein per minute.

Light Microscopy and Electron Microscopy

Tissue samples for histological examination were taken from the livers, kidneys, gills, and heart of control and exposed fish. For light microscopy, samples were first fixed in 10% buffered formalin for twenty-four hours at 4ºC and then immediately dehydrated in graded series of ethanol, immersed in xylol, and embedded in paraffin wax using an automatic processor. Sections of 3–5 μm were mounted. After they had been deparaffinized, the sections were rehydrated, stained with hematoxylin and eosin, and mounted with Cristal/Mount.

Liver tissue sections were also stained with periodic acid Schiff (PAS) for glycogen content assessment.

For electron microscopy, liver and kidney samples were prefixed in 2% glutaraldehyde fixative (in pH 7.4 phosphate buffer for ten hours at 4ºC) and postfixed in 1% osmium tetroxide fixative (in pH 7.4 phosphate buffer for 0.5 hours at 4ºC). Subsequently, they were dehydrated in graded ethanol series and embedded in epon. Ultrathin sections, 50–60 nm, were cut with an LKB microtome. The sections were mounted on copper grid and stained with uranylacetate and lead citrate. The tissue sections were examined in a Philips CM10 electron microscope.

Statistical Analysis

Organ weights and enzymatic activities results were subjected to one-way analysis of variance (ANOVA) and represent mean ± standard error (SE) of eight animals per group. Differences in mean values between groups were assessed by the Tukey’s test and were considered statistically different from p < .05.

Results

The liver was affected in the case of MC-LR, and MC-RR increased the relative weight of the kidneys of exposed fish (Table 1). Protein content, expressed as mg prot/g organ weight, was significantly decreased in the liver and kidney of fish treated with both toxins. The decrease was particularly pronounced in the liver of the exposed fish (90%) in comparison to control fish. In the kidney, the effect of the MCs was similar, and protein content diminished by approximately 50% in comparison to control fish. In contrast, no significant changes were detected in the gills of the treated fish.

Effect of MC-LR and MC-RR on Enzymatic Activities ACP and ALP

ACP activity increased significantly in the liver and kidney of fish injected with MC-LR (Figures 1 and 2). The most noticeable change in fish receiving MC-RR was in the kidney. The ACP activity in this organ was approximately three times higher in fish treated with MC-RR than in the control group, and a twofold increase in ACP activity was found in the case of MC-LR treatment. MC-LR and MC-RR did not exert any significant effect on ACP activity in gills.

ALP activity increased significantly in the liver and the kidney of fish exposed to MC-LR, and particularly so in the case of the liver (up to 300%). The effect of MC-RR on this enzymatic activity was also significantly increased in the liver (166%), and although ALP activities were higher in the kidney of treated fish, no significant differences were observed. The highest values of ALP activity were found in the gills; however, for either of the MCs assayed, there were no changes in gill ALP activity as compared to controls.

Histopathological Study

Mortality and Macroscopic Observations

No fish died during the seven-day experiment. No gross pathological changes were observed in the kidneys and gills of fish treated with MC-LR and MC-RR, the appearance of which was macroscopically normal.

When compared to the control group, visual inspection of the gastrointestinal tract showed a yellowish discoloration in the livers of fish exposed to either MC-LR or MC-RR.

Light Microscopy

Mild to severe histopathological changes were present in the livers, kidneys, and heart (Table 2), but not in the gills or gastrointestinal tracts of tilapias treated with MC-LR or MC-RR. In control fish, no histological changes were observed in any of the studied organs (Figures 3A, 3B, 4A, 4B, and 5A).

The livers of fish exposed to 500 μg/kg MC-LR or 500 μg/kg MC-RR underwent structural alterations. In general, the pathological changes in the livers of the group treated with MC-LR appeared to be more severe than those in the MC-RR group.

The microscopic examination of the Hematoxylin and Eosin (H&E)-stained section revealed that the cord-like parenchymal architecture of the liver was lost, primarily in the perilobular regions. The hepatocellular injury by both MC analogs was characterized by condensed PAS-negative and microvesicular steatosis of hepatocytes, diffuse focal necrosis, pyknotic nuclei, and degeneration in the pancreatic acinus with necrotic cells (Figures 3C, 3E, 6A, and 6C).

Degenerative changes were observed in the kidneys of fish exposed to both MC analogs, and they were more severe in tilapias exposed to MC-RR than in those exposed to MC-LR. In the kidneys of the fish from the MC-RR group, glomerulopathy with a significant increase in mesangial cells and severe vacuolization of tubular epithelia and hyperemia in the loop of Henle and pelvis was observed (Figures 7A and 7C). Congestion and edema of the interstitial tissue were also observed (Figures 4C and 4E).

Heart lesions were characterized by myopathy, and numerous fibrolysis processes were detected in both groups of fish exposed to MC-LR or MC-RR. The cardiac fibers maintained their normal structure, and a generalized edema was shown (Figure 5B)

Ultrastructural Observations

In the livers of fish treated with 500 μg/kg MC-LR, the pathological changes in the hepatocytes were more evident than in the fish treated with MC-RR. The hepatocytes showed pyknotic nuclei and highly vacuolized cytoplasms, containing glycogen and microvesicular steatosis (Figures 3D and 3F). Swelling of the hepatocytic endomembrane system (rough endoplasmic reticulum [RER], mitochondria, and Golgi body), swelling nuclei, and numerous lysosomes were also observed (Figure 3F). Numerous pyknotic nuclei and autophagic vesicles containing remnants of organelles were also observed. The fish treated with 500 μg/kg of MC-RR showed slightly swollen mitochondria, their cytoplasm was highly vacuolized, and they had decreased glycogen content (Figures 6B and 6D). The stack of RER, some of which had lost their ribosomes and mitochondria, was dispersed in the cytoplasm. Some cells of the bile canaliculi showed fewer microvilli.

The lesions observed in the kidneys of fish exposed to MC-RR were considerably more severe than those in the fish exposed to MC-LR. A glomerulopathy with alteration of the basal membrane was observed. The proximal tubules in the kidneys of the fish exposed to both MC variants contained more lysosomes, and a considerable pleomorphic, generalized vacuolation of the epithelial cells and increased irregularity of the basal membrane of the capillaries were observed (Figures 4D and 4F). Unlike the fish treated with MC-LR, the fish treated with MC-RR often showed necrotic tubular epithelial cells adjacent to seemingly healthy tubules and loss of the uniformity of the basal membrane (Figures 7B and 7D).

Discussion

Fish can be exposed to MCs by different routes, and some reports implicate cyanobacterial hepatotoxins in fish mortality in bloom-prone lakes and reservoirs (Malbrouck and Kestemont 2006; Zurawell et al. 2005). Very few detailed studies have been carried out of toxic manifestations of microcystin variants, namely MC-LR and MC-RR, and of those that have, none has focused on fish. This is the first study to have been carried out in tilapia (Oreochromis sp.), an economically important freshwater fish.

The dose of 500 μg/kg MC-LR or MC-RR used in this study did not result in the death of any fish, though Rabergh et al. (1991) found that the LD₅₀ of MC-LR in common carp (Cyprinus carpio L.) by i.p. injection was between 300 and 550 μg/kg over seven days. Kotak et al. (1996) suggested that the LD₅₀ of MC-LR in rainbow trout was between 400 and 1000 μg/kg, and Carbis et al. (1996) found that an i.p. administration of 50 μg MCs/kg in carp was lethal within eight hours. Data on the LD₅₀ of MC-RR in fish are more scarce. These differences in response to MC, depending on the species, are probably the result of different detoxification mechanisms, since phytoplanktivorous fish are possibly more tolerant to high MCs (Xie et al. 2007).

Increase in liver weight is a characteristic toxic effect of MCs, and it has been reported in a number of animal models (Gupta et al. 2003). In the present study, this fact was observed in the action of MC-LR in the liver, and in the kidney by MC-RR. The increase in liver weight in the fish may have been caused by hepatocellular swelling (Kotak et al. 1996); in the kidney, the increase could be a result of the congestion and edema of the interstitial tissue. Moreover, the protein content of the liver and kidney decreased significantly in comparison to control fish. A previous study from our laboratory showed that MCs from cyanobacterial cells diminished the protein content of the liver and gills in orally exposed tilapia fish (twenty-one days) under laboratory conditions, but not of the kidney (Molina et al. 2005). A highly significant decrease in the total protein concentration was also observed in silver carp (H. molitrix) injected i.p. with pure MC-LR (Vajcova et al. 1998). As in in vivo experiments, pure MC-LR and MC-RR standards in vitro reduced the protein content when two fish cell lines (PLHC-1 and RTG-2) were acutely exposed to them (Pichardo et al. 2005, 2006). This result could be explained by the proteolysis and delay in protein synthesis produced by MCs, in concordance with the reduction in protein content observed in carps exposed to other pollutants such as pyretroids (Das et al. 2002).

The increased ACP activity observed in the liver and kidney of tilapia after acute exposure to MC-LR is in accordance with previous studies performed in tilapias exposed subchronically to a cyanobacterial bloom scum (Molina et al. 2005). In contrast, MC-RR induced ACP activity only in the kidney, perhaps because it is more hydrophilic than MC-LR, which facilitates its distribution to this organ. These increases indicate increased lysosomal activity in both organs because MCs interact with lysosomes (Ghorpade et al. 2002). Earlier in vitro investigations by our laboratory demonstrated that MC-LR exerted an extremely potent concentration-dependent stimulation on the lysosomal function in fish cell lines (Pichardo et al. 2005), demonstrating that MC-LR is much more toxic than MC-RR in both fish cell lines. This differential response of ACP activity in fish exposed to MC variants has also been demonstrated in other enzymatic activities, such as the induction of enzymatic antioxidant defenses in tilapia fish (Jos et al. 2005; Prieto et al. 2006), in which the liver was the most sensitive tissue to MC-LR and the kidney to MC-RR.

As far as the effect of MC variants on ALP activities is concerned, MC-LR proved to be more active than MC-RR, which induced ALP activity only in the liver. These increases indicate that the membrane properties are perturbed by interaction with MC variants, because alkaline phosphatases are intrinsic plasma membrane enzymes involved in membrane transport activities and in bone formation (Mazorra et al. 2002).

No significant changes were detected in ACP or ALP activities in the gills of fish treated with both MC variants. This result agrees with an earlier study in tilapia orally exposed to repeated doses of MCs (Molina et al. 2005).

Intraperitoneal exposure to MCs has different histopathological effects on liver, intestine, kidney, heart, spleen, and gills in such fish species as common carp (Carbis et al. 1996; Carbis et al. 1997; Rabergh et al. 1991), rainbow trout (Kotak et al. 1996), and hardhead catfish Arius felis and gulf killifish Fundulus grandis (Fournie and Courtney 2002). In this study, exposure to MC-LR and MC-RR caused severe pathological lesions in hepatic tissues. The main histological findings were loss of the parenchymal architecture, necrosis of hepatocytes with vacuolization of cytoplasms, presence of lipid droplets inducing steatosis, and proliferation of lysosomes. Very few in vivo studies have been made of the toxic effects of MCs on the ultrastructures of hepatocytes in fish, and the main changes induced in the omnivorous common carp and the carnivorous rainbow trout include swollen mitochondria, whirling of the RER, vacuolated cytoplasm, and condensed chromatin (Li et al. 2004; Li et al. 2005; Rabergh et al. 1991).

Vacuolization has been reported in isolated hepatocytes from common carp (Li et al. 2001) and in tilapias exposed orally to repeated doses of MCs. It might indicate an imbalance between the rate of synthesis of substances in the parenchymal cells and the rate of their release into the systemic circulation (Molina et al. 2005). The production of steatosis agrees with previous in vitro studies by our group, which found for the first time that MCs induced steatosis in PLHC-1 hepatoma cells after forty-eight hours of exposure to 50 μM, particularly with MC-RR, whereas general damage was more severe with MC-LR (Pichardo et al. 2005). The lysosomal proliferation observed in the livers of tilapia exposed to both MCs is in accordance with the results obtained from silver carp that were fed naturally toxic Microcystis blooms, and this lysosome activation could be an adaptative mechanism to eliminate or lessen cell damage caused by MC (Li et al. 2007).

In kidney, degenerative changes in tubuli, glomeruli, and interstitial tissue have been observed. Few papers in the literature show the effect of microcystin-LR on the renal system. Radbergh et al. (1991) have shown degenerative changes in the tubular epithelial cells, glomeruli, and interstitial tissue in the kidneys of carps exposed i.p. to MC-LR. These alterations are consistent with the ones observed in common carps exposed to MCs from cyanobacterial cells (Carbis et al. 1996) or gavaged with a single sublethal bolus dose of MC-LR (Fischer and Dietrich, 2000). Kotak et al. (1996) showed renal lesions in fish (Oncorhynchus mykiss) that consisted of coagulative tubular necrosis and dilatation of Bowman’s space. Two studies demonstrated that microcystin-LR could affect renal physiology by altering vascular, glomerular, and urinary parameters (Nobre et al. 1999; Nobre et al. 2001). To our knowledge, this is the first study to examine histopathological changes in fish kidney after exposure to MC-RR. Because of its greater hydrophilicity, the lesions in this organ were more intense than after exposure to MC-LR.

Although the lesions caused by the two MC variants were qualitatively identical, quantitatively there were considerable differences. MC-LR caused more severe general damage in the liver, whereas the effect of MC-RR was more severe in the kidney. Very few studies have focused on the comparative toxicity of microcystin variants (LR, RR, and YR). Gupta et al. (2003) studied mice and found that modifications at different sites of the microcystin molecule produced mild to marked changes in toxicity. They also found that the replacement of the second amino acid leucine in MC-LR with arginine (MC-RR) retained hepatotoxicity and could lead to greater damage to the kidney. The differences between hydrophilic and hydrophobic microcystins could result in changes in organotropism, toxicokinetics, and bioaccumulation (Pia and Meriluoto 2003).

In addition, this is the first study that illustrates histological changes in the hearts of tilapias exposed to MCs, for example, myopathy with processes of fibrolysis and generalized edema. The occurrence of myopathy could be attributed to the direct action of MCs on heart tissue, coupled with the indirect effects of both variants on the kidney (induced glomerulonephritis). Best et al. (2001) observed sublethal cardiac responses of brown trout alevins to aqueous extracts of Microcystis strains and to pure MC-LR at concentrations between 5 and 500 μg MC-LR equivalents/L.

In summary, our results indicate that the acute intraperitonal exposure of tilapia fish (Oreochromis sp.) to 500 μg/kg MC-LR or 500 μg/kg MC-RR induces toxic reactions in the liver and kidney, with subsequent disturbances in metabolic–enzymatic homeostasis. The most active variant in liver and kidney was MC-LR. MC-RR, on the other hand, showed significant changes only in the liver for ALP levels and induced ACP activity in the kidney. This differential response could be because MC-RR is less toxic and more hydrophilic than MC-LR. Histopathological changes were detected in the liver and kidney for both MC variants, mostly produced by MC-LR in the liver, and by MC-RR in the kidney.