Fish Toxicity of Commercial Herbicides Formulated With Glyphosate

The concept "sustainable development" is used worldwide since 1984, this represents the connection between development and several issues such as human resources, food, species, ecosystems, energy and industry. Argentina produces raw materials and agricultural inputs, being soy one of the most important. Much of this production is exported as grain and the Asian countries are the main buyers. Transgenic varieties, such as glyphosate resistant soy, are used to increase the yields and facilitate the agricultural practices; these particular advantages have significantly increased the use of glyphosate formulations in order to control weeds in the soybean crop cycle. This massive and uncontrolled use of this herbicide causes concern in urban and mainly rural population regarding the toxicological risks. Glyphosate is a highly water soluble substance (10,500 mg/L) with a half-life in water fluctuating between 3.5 and 70 days. Its herbicidal action is due to its ability to inhibit 5-enolpyruvylshikimate-3phosphatesynthase [1], an enzyme involved in the biosynthesis of aromatic amino acids in plants. Although this substance is an acid, it is commonly used as a salt, such as the isopropylamine salt (IPA) of N-(phosphonomethyl) glycine or the glyphosate isopropylamine salt. There are many commercial formulations commonly characterized by containing 480 g/l glyphosate IPA salt (48%) and the surfactant POEA (polyoxyethylamine), the latter belongs to the family of alkyl polyethoxylated compounds which are synthesized from fatty acids of animals and is thought to be the primarily responsible for the toxic effects of the formulations [2-4]. The strong risk that glyphosate and adjuvants contaminate surface water bodies and groundwater in areas of application constitutes a serious threat to the ecological balance of the regions affected because of their toxic potential to alter the food chain from the first link.


Introduction
The concept "sustainable development" is used worldwide since 1984, this represents the connection between development and several issues such as human resources, food, species, ecosystems, energy and industry. Argentina produces raw materials and agricultural inputs, being soy one of the most important. Much of this production is exported as grain and the Asian countries are the main buyers. Transgenic varieties, such as glyphosate resistant soy, are used to increase the yields and facilitate the agricultural practices; these particular advantages have significantly increased the use of glyphosate formulations in order to control weeds in the soybean crop cycle. This massive and uncontrolled use of this herbicide causes concern in urban and mainly rural population regarding the toxicological risks.
Glyphosate is a highly water soluble substance (10,500 mg/L) with a half-life in water fluctuating between 3.5 and 70 days. Its herbicidal action is due to its ability to inhibit 5-enolpyruvylshikimate-3phosphatesynthase [1], an enzyme involved in the biosynthesis of aromatic amino acids in plants. Although this substance is an acid, it is commonly used as a salt, such as the isopropylamine salt (IPA) of N-(phosphonomethyl) glycine or the glyphosate isopropylamine salt. There are many commercial formulations commonly characterized by containing 480 g/l glyphosate IPA salt (48%) and the surfactant POEA (polyoxyethylamine), the latter belongs to the family of alkyl polyethoxylated compounds which are synthesized from fatty acids of animals and is thought to be the primarily responsible for the toxic effects of the formulations [2][3][4]. The strong risk that glyphosate and adjuvants contaminate surface water bodies and groundwater in areas of application constitutes a serious threat to the ecological balance of the regions affected because of their toxic potential to alter the food chain from the first link.
The fish have been used for a long time as experimental biological models to measure the environmental impact of different substances [5,6]. Our research group has used fish to assess the potential toxicity of different natural and/or synthetic novel drugs [7][8][9].
The main objective of this study was to determine the acute and chronic toxicity toward fish of two commercial herbicides formulated with glyphosate. Two fish species frequently used in these kinds of tests [10], Poecilia reticulata "guppies" and Danio rerio "zebrafish" were selected as model organisms.

Formulations and solutions
We selected two commercial formulations of herbicides present in the local market whose active ingredient is glyphosate, the solution A (Glacoxan ® ) refers to an herbicide used in gardening; whereas solution B (Estrella ® ) is an herbicide used in extensive farming practices. Both products were purchased in agrochemistry shops in the province of San Luis, Argentina. The characteristics of these solutions are summarized in Table 1.

Acute toxicity assays
The technique recommended by the U.S. Fish and Wildlife Service [11] has been modified in order to employ a smaller amount of testing compounds as it was already reported by our group [12][13][14]. Organisms and solutions were placed in a 1 L capacity container. The water level was maintained during the measurements. The essay begins with an initial exposure to potentially toxic agents and continues for 96 h. Every 24 h the number of dead specimens in each container were counted and then removed. The mortality rate was documented at 96 h. The toxic effect was evaluated toward two fish species, D. rerio and P. reticulate.

Abstract
We report here the acute and chronic toxicity of two commercial formulations of herbicides whose active ingredient is glyphosate (Glacoxan ® and Estrella ® ). The acute toxicity was tested toward two fish species Danio rerio and Poecilia reticulata by evaluating the mortality. The chronic toxicity was assessed in D. rerio by measuring the biochemical parameters glutamic pyruvic transaminase (AST) and glutamic oxaloacetic transaminase (ASL). In addition the analysis of haematological parameters (morphological study) was carried out.
The results indicate that both herbicides produce acute toxicity toward the two tested species. In addition the biochemical parameters displayed high values as a sign of chronic toxicity. Thereby, indicating that both herbicides may produce environmental damage. Specimens were bred in our laboratory reaching a maximum size of 1 cm and 15 to 20 days old. Ten fish of each species were exposed to a solution of the tested compound starting from a maximum value of 100 μL/L. The lowest concentration of commercial formulation that produces 100% mortality (LC-100%M) and the maximum concentration that does not produce mortality (MC-0%M) were determined.

Chronic toxicity testing
Specimens of D. rerio of 5-6 months were placed in 6 L containers for a period of 21 days. Three specimens were placed in a container (1 specimen per 2 L of water) and kept at room temperature with controlled aeration and standardized food. Each group was exposed to sublethal doses considering as subletal doses the maximum concentration that does not produce mortality (MC-0%M) obtained from acute toxicity tests conducted previously.
A control group no exposed to herbicides was also tested. At the end of the evaluation period 3 specimens of each group were taken for biochemical testing. Fishes were sacrificed with scalpel and their wet weight was determined in precision analytical balance KERN 770. For each specimen a smear of fresh blood at the time of the sacrifice was made and then proceeded to carry out a dilution of 50% p/p (sample fish / saline solution). After that, a macerated in porcelain mortar Coors was performed and the macerated product was centrifuged for 10 min at 1500 rpm in a centrifuge ("Arcane" Macro TDL80-2B). The supernatant was sampled and the following studies were conducted: a) Biochemical parameters: a colorimetric method was used to determine the activity of glutamic pyruvic transaminase (AST) and glutamic-oxaloacetic transaminase (ASL) in serum using commercial enzymatic kits (transaminases 200, Wiener). b) Haematological parameters: a morphological study was performed from a blood smear cells using the coloration of May Grunwald-Giemsa [15].
In all cases the results were compared with the control group.

Statistical analyzes
The chi-square method was used in order to compare the mortality rates depicted in the acute toxicity tables, meanwhile the nonparametric Student test Mann-Whitne was used for comparisons between the LC-100%M and MC-0%M.
A nonparametric statistics using the Kruskal Wallis test was used to contrast chronic toxicity results (transaminases); a subsequent comparison using the Dunn's test for a significance level of 95% was also applied. The statistical software used was Statistx 8.0.

Acute toxicity
We have previously shown that differential effects in acute toxicity of commercial formulations of glyphosate and its pure salt against P. reticulata occurred [12].
In the present study we observed that for the two fish species, the lowest concentration of the commercial formulation of herbicide A that produce 100% of mortality (LC-100%M) was 100 μL/L and that 50 μL/L was the maximum concentration that does not produce mortality (MC-0%M) ( Table 2).
Values of 50 μL/L for the lowest concentration of commercial formulation producing 100% mortality (LC-100%M) and 25 μL/L for the maximum concentration that produce no mortality (MC-0%M) were observed when the herbicide B was tested toward D. rerio specimens.
It is evident that there is a differential sensitivity to the herbicide B in both species, being D. rerio the most affected; this result are in agreement with previous reports about the effect of other formulations toward different fish species [2,12].

Statistical analyzes
The acute toxicity difference is significant when compared the concentrations of 100 μL/L and 50 μL/L obtained for the two solutions tested with a p ≥ 0.0001. It is also significant (p ≥ 0.0022) the difference observed in the acute toxicity for both species D. rerio and P. reticulata (LC-0%M= 25 μL/L and 50 μL/L respectively).

Chronic toxicity
These studies were performed taking into account that various toxic effects in organisms do not occur immediately; but rather alter their biology, conditioning their survival chance.

Glutamic pyruvic transaminase (AST) and glutamic oxaloacetic transaminase (ASL)
Transaminases (AST and ASL) are enzymes widely distributed in the body. Their normal activity in serum is low or absent. The destruction of the tissues in which they are present leads to an increase of their presence in serum. In addition these enzymes are inducible by toxicity due to herbicides [16]. In particular, the pyruvic transaminase is an enzyme used to assess liver function, so its high concentration in blood confers some specificity in the diagnosis of toxic liver damage [17].
According to the data of Table 3 it can be concluded that, considering the time and concentration of herbicide exposure, both transaminase values are increased compared to controls although the values corresponding to the specimens treated with the solution of herbicide B are considerably higher than the ones obtained after treatment with herbicide A.
Since the levels of both enzymes can be considered as indicators of toxicity, it is clear that the exposure for a period of 21 days at sublethal doses (MC-0%M) produces chronic toxicity effects toward D. rerio, being the herbicide B the most toxic formulation in agreement with the results of acute toxicity tests above described.

Statistical analyzes
Significant differences (p ≥ 0.001) were observed in transaminase activity values. By subsequent comparisons using the Dunn's test, we found differences between the values of the solution prepared with the formulation A and B respect to the control.

Blood smear
This parameter has been reported by several authors as an indicator of toxic effects of the action of herbicides formulated with glyphosate. The study involves morphology analysis of blood cells through microscopic observation of colored preparations, in order to distinguish cytoplasmic structures such as granular appearance, as vacuoles and lysosomes [17].
The analysis of the blood smear performed from D. rerio specimens exposed to solutions of the herbicides A, B compared to the controls without herbicides do not show any significative celular morphological change. However, this observation contrasts with other authors reports about experiences performed with other commercial herbicides toward Cyprinus carpio haematopterus [12] and Carassius auratus [17].

Conclusions
Both tested herbicides formulations, Glacoxan ® and Estrella ® , present acute toxic effects toward P. reticulata, since the lowest concentrations that produce 100% of mortality and the maximum concentration that does not produce mortality are of 100 μL/L and 50 μL/L respectively. Estrella ® formulation shows a higher acute toxicity toward D. rerio since the lowest concentration that produces 100% mortality was 50 μL/L. Subletal doses of both formulations accuse chronic liver damage in D. rerio without morphological alterations at hematological level. In this case the formulation Estrella ® is also more toxic than Glacoxan ® in agreement with the differences observed for the acute toxicity.
Considering fish as experimental biological models to measure the environmental impact of different substances, these results indicate that both tested herbicides may produce potential environmental damage.