Toxicological, biochemical and morphophysiological effects of Serjania erecta leaf aqueous extract on Piaractus mesopotamicus

Abstract

This study was carried out to determine the toxicity and biochemical and morphophysiological changes caused by Serjania erecta leaf aqueous extract in pacu (Piaractus mesopotamicus). For acute toxicity testing (CL_50-4h), pacu juveniles were exposed during 4 h to Serjania erecta aqueous extract concentrations of 2.5, 12.5, 25, 50, 100, and 150 µg mL^-1, which were added directly to the water in the tanks. In the control group, the animals were kept in water free from aqueous extract. CL_{50-4 h} was estimated at 57.43 µg mL^-1. After exposure to the aqueous extract, the highest (P<0.05) glucose concentration and the lowest (P<0.05) plasma sodium level were when the fish were exposed to the S. erecta concentration of 50 µg mL^-1. Mortality occurred at S. erecta extract levels higher than 50 μg mL^-1, and all fish died at concentrations greater than 100 μg mL^-1. In addition, exposure to this extract caused severe histological changes in the gills and liver with higher prevalence of necrosis (30.2%), and fatty degeneration (77.4%) respectively. At the concentrations tested here, S. erecta aqueous extract causes morphofunctional alterations in this fish species.

Key words
fish; medicinal plant; morphophysiology; pacu; saponins

INTRODUCTION

Piaractus mesopotamicus, commonly known as “pacu”, is a fish of the family Serrasalmidae instead of Characidae (Holmberg 1887). It is a species of great importance to South American countries along the Paraná river basin such as Paraguay, Uruguay, Argentina, and Brazil (Urbinati & Gonçalves 2013). In the year 2017, pacu was one of the main native fish species farmed in Brazil (IBGE 2017).

There has been a growing interest in the effects of plant-derived bioactive compounds, or even medicinal plants, under the purpose of evaluating new natural products with potential for use in aquaculture; e.g. anesthetics, antiparasitic drugs, and antimicrobials (Kavitha et al. 2012, Adesina et al. 2013, Abalaka et al. 2015a, Andrade et al. 2016). This is a consequence of the expansion of this activity and the quest for achieving sustainable production through the use of biodegradable products, which cause the least impact on the environment and also on the health of consumers (Soares & Tavares Dias 2013).

Serjania erecta Radlk. (Sapindaceae), a native Brazilian plant popularly known as “cinco folhas” (“five leaves”) or “cipó-cinco-folhas” (“five-leaf vine”) (Guarim Neto et al. 2000, Pott et al. 2004), is spread mainly across tropical regions such as the cerrado biome area (Ferrucci 2004). Studies have shown that different species of the genus Serjania have antimicrobial (Lima et al. 2006, Cardoso et al. 2013), antiparasitic (De Mesquita et al. 2007, Hernández et al. 2012), antiinflamatory (Gomig et al. 2008), and antioxidant effect (Heredia-Vieira et al. 2015). The toxicological and environmental properties of a plant or compound should be investigated, since the fact that they are natural does not necessarily mean they are safer (Duke 1990). In this regard, studies on acute toxicity help in assessing the risk and determining safe doses (Syngai et al. 2016).

Researchers have examined the toxicity of different species of the genus Serjania sp. [e.g. Serjania marginata (Périco et al. 2015, Moreira et al. 2019), Serjania caracasana (Silva et al. 2017), and S. erecta (Castelo et al. 2009, Broggini et al. 2010)] in mice. However, no studies have evaluated the toxicity of S. erecta in aquatic organisms. The present study was thus undertaken to evaluate the CL_50-4h acute toxicity and possible biochemical and morphophysiological implications caused by S. erecta aqueous extract in pacu (P. mesopotamicus).

MATERIALS AND METHODS

Plant material, preparation, and chemical composition of S. erecta aqueous extract

The S. erecta leaves were collected in the municipality of Aquidauana - MS, Brazil (20° 27” 56.4’ S; 55° 47” 53.2’ W) and identified by Pott et al. (2004). A voucher specimen (HMS 8355) was deposited in the Herbarium of Embrapa Cattle in the state of Mato Grosso do Sul, Brazil.

The aqueous extract was obtained by the method described by Broggini et al. (2010), with adaptations. Leaves were dried in a forced-air circulation incubator with microprocessor and air renewal (Sterilifer - SXCR 40) at 37 ± 2 °C for 48 h and ground through a Wiley mill (TE-680- Tecnal) with a 10-mesh screen. The aqueous extract was obtained by maceration, with 1,000 mL of distilled water used for each 50.0 g of dried and ground plant material. The plant was kept in contact with distilled water for 24 h at room temperature (25 ± 2 °C). Next, the extract was filtered through filter paper and lyophilized for complete water removal.

An extract solution (concentration of 1000 µg mL^-1 in distilled water) was analyzed for flavonoid content by employing the method described by Lin & Tang (2007), and the result was expressed in milligrams of quercetin equivalents per gram of lyophilized extract. The phenolic compounds were assayed with the same samples used in the quantification of flavonoids, by applying the method described by Lin & Tang (2007)—results were expressed in milligrams of gallic acid equivalents per gram of lyophilized extract. The tannins were quantified using the vanillin reaction, following the method proposed by Broadhurst & Jones (1978) and adapted by Agostini-Costa et al. (1999)—results were expressed in milligrams of catechin equivalents per gram of lyophilized extract. All chemical composition assays were carried out in triplicate.

The presence of saponins in the aqueous extract was determined by qualitative techniques, by applying the foam test in accordance with the methods proposed by the Brazilian Pharmacopeia (2010).

Acute toxicity testing (CL_50-4h)

Seventy pacu (P. mesopotamicus) juveniles were acquired from a commercial fish farm located in Terenos - MS, Brazil (20°26’1.62”S; 55°17’10.87”W). The animals were kept in tanks with oxygenation at the Experimental Fish Farming Station at the Federal University of Mato Grosso do Sul (20°29’59.04”S; 54°36’52.59”W). Animal procedures were approved by the Ethic Committee on Animal Use of Federal University of Mato Grosso do Sul (CEUA/UFMS/976/2018).

For the acute toxicity test, the pacu juveniles (average weight 38.68 ± 3.00 g; average length: 13.34 ± 1.0 cm) were randomly distributed, from the same initial lot of fish into 50-L tanks. The fish—10 per experimental unit—were deprived of feed for 48 hours and kept in a static system with constant aeration, as recommended by OECD (1992). The fish were exposed to the following treatments: control (without S. erecta extract addition), containing only water from the growing tank; and different concentrations (2.5, 12.5, 25.0, 50.0, 100.0, and 150.0 µg mL^-1) of the extract, which were added directly to the tanks, per 4-hour exposure period. After the period of acute exposure to S.erecta extract five fish per treatment were used for morphophysiological analyzes.

Physicochemical parameters of water and behavioral alterations

Throughout the experimental period of four hours, the water quality parameters were measured with a YsiLife Sciences® multiparameter instrument and the toxic ammonia level was measured using the Alfakit® colorimetric reagent. The following mean values were obtained: dissolved oxygen = 5.0 ± 0.75 mg L^-1; pH = 7.68 ± 0.13; temperature = 27.46 ± 0.14 °C; ammonia = 0.07 ± 0.03 mg L^-1; and electric conductivity = 144.73 ± 10.75 μS cm−¹. These values are within the recommended range for tropical fish farming according to Boyd (1998).

The following behavioral alterations were observed: erratic swimming, loss of equilibrium, excessive mucus production, aerial respiration and lethargy. These observations were recorded every 30 minutes during the four hours period in which the fish were exposed to the concentrations of the aqueous extract of S. erecta. Dead fish were removed at each observation to avoid degradation of water quality parameters.

Biochemical analyses

After the period of acute exposure, five animals per concentration were anesthetized with eugenol (75 mg L^-1) and subjected to caudal vessel puncture using 10% EDTA-treated syringes for blood collection. After blood collection, the fish were sacrificed by a quick brain concussion, followed by biometry and necropsy for collection of organs for histological analysis. The animals that survived the different treatments were transferred to clean water and kept in the fish farming sector. The blood was centrifuged at 5000 g for 5 min to obtain the plasma (Fanem Centrimicro, Brazil), which was then stored at – Brazil (Winkaler et al. 2007). Subsequently, total plasma protein, albumin, globulin, chlorine (Cl–), total calcium, and ionized calcium were measured by spectrophotometry (1203 UV, Shimadzu, Japan), using a commercial kit (Labtest®). The sodium (Na⁺) and potassium (K⁺) ions were measured by flame photometry (910-M Analyzer) in accordance with the method described by Catani & Paiva Neto (1949). The plasma glucose concentration was determined in each animal, immediately after blood collection, using an Accu check® Advantage portable glucose meter (Roche Diagnóstica, Brazil).

Histological analysis

The fish sacrificed by a quick brain concussion, followed by a detailed macroscopic examination, necropsy, and observation of the cavity. The gills and the liver were dissected and weighed for morphological analysis. The following formula was used to determine the hepatosomatic index: HSI = LW/TW.100, where HSI = hepatosomatic index; LW = liver weight; and TW = total fish weight, according to Vazzoler (1996).

The second left branchial arch and a liver fragment were harvested from three fish per concentration and fixed in 10% formaldehyde buffered with monobasic and dibasic sodium phosphate at a pH of 6.9. The samples were dried in progressive alcohol series, cleared in xylol, and paraffin-embedded, as suggested by Steckert et al. (2018). Subsequently, they were sectioned to 5 μm and stained with Harris’ hematoxylin and eosin (HH&E) using a PAT-MR10 microscope (The Pathologist®, Brazil), in accordance with the method of Howard et al. (2004), mounted on Entellan® permanent slides, analyzed, and photographed by the DIC (Differential Interference Contrast) technique using an Axio Imager A2 light microscope (Zeiss®, Germany).

In addition to qualitative description, histological alterations in the gills and liver were assessed using the semi-quantitative method proposed by Steckert et al. (2018), which was adapted to a growing scale of mean alteration values (MAV), according to the degree of lesion severity, namely, 0 (no alterations), 1 (slight alterations or focal process), 2 (moderate alterations or multifocal process), and 3 (serious alterations or diffuse process). Based on this scale, a mean histological alteration value (MAV) was assigned for each lesion, ranging between mild (0.1 to 1.0), moderate (1.1 to 2.0), and intense (2.1-3.0). The prevalence of each lesion was also calculated in accordance with Steckert et al. (2018).

To classify the gill and liver-tissue injuries, a longitudinal section was made in the secondary lamellae and central lobe of the liver, using medium portions of the respective organs. Observations were made in a field covered by a 10x magnifying lens. The following tissue alterations were considered, for the gills: necrosis, vessel and filament dilatation, telangiectasia, epithelial hyperplasia, epithelial displacement, eosinophilic granular inflammatory infiltrate, interlamellar hyperplasia, cellular edema, and lamellar fusion and congestion; and for the liver: fatty degeneration, vacuolization, sinusoidal and vessel and congestion, necrosis, hepatocyte hypertrophy, leukocyte infiltrate, pyknotic nucleus, cellular peripheral nucleus, accumulation of hemosiderin pigments, cord disruption, and melanomacrophage centers.

Statistical analysis

The CL_50-4h value was calculated by the Trimmed Spearman-Karber method (Hamilton et al. 1978), with a confidence interval of 95%. All data were subjected to the Shapiro-Wilk and Levene tests to check for normality and variance homoscedasticity assumptions, respectively. The data which did not show homogeneity of variance were Log₂ (x+1) converted. If the assumptions were met, the data were subjected to single-factor ANOVA and means were separated by Tukey’s test. All tests were performed at the 5% significance level, using Statistica 10.0 software. The means were expressed as standard deviation.

RESULTS

Chemical composition of S. erecta leaf aqueous extract

Qualitative phytochemical analysis indicated presence of saponins. Quantitative analyses revealed the following amounts of phenolic compounds, flavonoids, and tannins: 386.78 ± 1.6 mg g^-1, 212.03 ± 1.2 mg g^-1, and 89.7 ± 0.3 mg g^-1, respectively.

Acute toxicity and behavioral alteration in Piaractus mesopotamicus exposed to S. erecta extract

Behavioral alterations such as erratic swimming, loss of equilibrium, excessive mucus production, aerial respiration, and lethargy were observed in the animals exposed to the concentrations of 25, 50, 100, and 150 µg mL^-1. Additionally, mortality occurred at S. erecta extract levels higher than 50 µg mL^-1, and all fish died at concentrations greater than 100 µg mL^-1 (Table I).

Table I
Mortality and behavioral alterations observed in Piaractus mesopotamicus (n=10) exposed to Serjania erecta aqueous extract for 4 h.

S. erecta extract concentration (µg ml– 1)	Percentage of dead animals	Erratic swimming	Loss of equilibrium	Mucus	Aerial respiration	Lethargy	Death
0	-	-	-	-	-	-	-
2.5	-	-	-	-	-	-	-
12.5	-	-	-	-	-	-	-
25	-	+ (n=3)1	+ (n=2)	+ (n=3)	+ (n=1)	+ (n=2)	-
50	30	++ (n=4)	+ (n=3)	+ (n=3)	++ (n=2)	+ (n=3)	+ (n=3)
100	100	++ (n=7)	++ (n=8)	++ (n=4)	++ (n=5)	++ (n=6)	+ (n=10)
150	100	++ (n=9)	++ (n=10)	++ (n=8)	++ (n=6)	++ (n=8)	+ (n=10)

• Behavioral alteration observed: + <60 min and ++ > 60 min.
  ¹Number of individuals engaging into the various behavioral alterations.

The concentrations of 100 and 150 μg mL^-1 caused 100% mortality in the fish within 4 h of exposure. Mortalities occurred at 175 and 158 minutes after exposure to the aqueous extract of S. erecta, respectively. At the concentration of 50 μg mL^-1, 30% mortality was observed with the occurrence of the first mortality at 200 minutes of exposure to the aqueous extract S. erecta.

According to the results, the concentration capable of causing mortality in 50% of pacu juveniles in four hours of exposure to the S. erecta aqueous extract was 57.43 µg mL^-1, with a lower limit of 46.98 µg mL^-1 and an upper limit of 70.21 µg mL^-1.

Biochemical parameters of Piaractus mesopotamicus exposed to S. erecta extract

It was not possible to collect blood from fish at concentrations of 100 and 150 μg mL^-1 for biochemical analysis because there was 100% mortality before completing 4 hours of exposure to S. erecta extract.

Plasma total protein, albumin, and globulin levels and the ionic parameters chlorine (Cl^-), potassium (K⁺), sodium (Na⁺), total calcium, and ionized calcium did not differ from control group (Table II). The lowest (P <0.05) glucose level was obtained in fish submitted to the concentration of 2.5 μg mL ^-1 of S. erecta. The lowest (P <0.05) plasma sodium level was obtained in fish submitted to concentration of 50 μg mL ^-1 of S. erecta (Table II).

Table II
Plasma biochemical parameters in Piaractus mesopotamicus exposed to different concentrations of Serjania erecta aqueous extract for 4 h.

Biochemical parameter	S. erecta aqueous extract concentration
	0 µg mL-1	2.5 µg mL-1	12.5 µg mL-1	25 µg mL-1	50 µg mL-1
Glucose (mg dL-1)	75.33 ± 5.03 ab	73.33 ± 11.68 b	84.00 ± 13.11 ab	131.67 ± 35.85 ab	145.33 ± 43.98 a
Total protein (mg dL-1)ns	3.22 ± 0.18 a	3.59 ±0.32 a	3.06 ± 0.45 a	3.17 ± 0.21 a	3.18 ± 0.13 a
Albumin (g dL-1) ns	1.75 ± 0.88 a	1.12 ± 0.03 a	1.08 ± 0.04 a	1.07 ± 0.08 a	1.07 ± 0.05 a
Globulin (g dL-1) ns	2.19 ± 1.06 a	2.83 ± 0.35 a	1.51 ± 0.44 a	2.2 ± 0.13 a	2.21 ± 0.10 a
Chlorine (mE dL-1) ns	41.51 ± 7.27 a	46.18 ± 5.96 a	45.02 ± 3.32 a	48.05 ± 4.42 a	50.65 ± 3.89 a
Calcium (mg dL-1) ns	9.47 ± 2.88 a	7.03 ± 2.28 a	7.11 ± 1.84 a	6.55 ± 0.32 a	7.64 ± 1.20 a
Ionized calcium (mg dL-1) ns	62.33 ± 17.29 a	47.66 ± 13.7 a	48.07±11.11 a	44.73 ± 1.96 a	51.26 ± 7.24 a
Sodium (mg mL-1)	3.73 ± 0.23 ab	4.53 ± 0.06 a	4.17 ± 0.58 ab	4.05 ± 0.15 ab	3.37 ± 0.48 b
Potassium (mg mL-1) ns	0.28 ± 0.18 a	0.22 ± 0.08 a	0.27 ± 0.03 a	0.28 ± 0.03 a	0.35 ± 0.05 a

• Mean ± standard deviation, different letters indicate significant differences between the treatments using one-way ANOVA and Tukey’s test (P <0.05). ^ns Not significant.

Organ morphometry of Piaractus mesopotamicus exposed to S. erecta extract

Fish weight and length did not differ significantly between the treatments. However, higher (P<0.05) liver weight and hepatosomatic index were obtained in the treatment with the extract concentration of 100 µg mL^-1 compared to control and the other concentrations (except 25 and 50 µg mL^-1). Heavier gills (P<0.05) were observed in the fish receiving concentrations greater than 12.5 µg mL^-1 (Table III).

Table III
Morphometric parameters of liver and gills of Piaractus mesopotamicus (n=5) exposed to different concentrations of Serjania erecta aqueous extract for 4 h.

Parameter	S. erecta aqueous extract concentration
	0 µg mL-1	2.5 µg mL-1	12.5 µg mL-1	25 µg mL-1	50 µg mL-1	100 µg mL-1	150 µg mL-1
Weight (g) ns	40.60 ± 6.05 a	36.13 ± 4.29 a	39.46 ± 1.72 a	43.40 ± 3.36 a	39.0 ± 6.20 a	38.06 ± 1.90 a	34.13 ± 6.11 a
Length (cm) ns	10.67 ± 1.15 a	9.83 ± 1.26 a	11.83 ±0.29 a	11.67 ±0.58 a	11.67 ± 1.26 a	11.67 ± 0.29 a	12.00 ± 0.50 a
*HSI	1.52 ± 0.34 b	1.67 ± 0.5 b	1.39 ± 0.23 b	2.51 ± 0.12 ab	2.63 ± 0.48 ab	4.46 ± 0.71 a	2.58 ± 1.88 b
Gills (g)	1.54 ± 0.16 bc	1.29 ± 0.15 c	1.88 ± 0.1 abc	2.29 ± 0.18 abc	2.2 ± 0.46 abc	2.89 ± 0.61 a	2.46 ± 0.51 ab

• Mean ± standard deviation, different letters indicate significant differences between the treatments using one-way ANOVA and Tukey’s test (P <0.05). *HSI - hepatosomatic index. ^ns Not significant.

Histopathology in Piaractus mesopotamicus exposed to S. erecta extract

The following gill alterations were observed: necrosis, dilation of vessels and filaments, telangiectasia, epithelial hyperplasia, epithelial displacement, eosinophilic granular inflammatory infiltrate, interlamellar hyperplasia, cellular edema, lamellar fusion, and congestion (Figure 1). The most prevalent of those was necrosis (30.2%), whereas the least frequent was congestion (1.6%). The mean histological alteration value (MAV) of necrosis was higher (P<0.05) only at the extract concentration of 150 µg mL^-1, whereas the MAV of eosinophilic granular inflammatory infiltrate increased (P<0.05) from 50 µg mL^-1. The MAV of epithelial hyperplasia was higher at the concentration of 12.5 µg mL^-1. The other alterations did not differ between the treatments (Table IV).

Figure 1
Gills of Piaractus mesopotamicus exposed to Serjania erecta extract. a = epithelial displacement (2.5 μg mL^-1); b = telangiectasia, lamellar fusion (12.5 μg mL – ¹ ); c = dilatation of vessels and filaments, interlamellar hyperplasia (25 μg mL^-1); d = necrosis (50 μg L^-1); e = hemorrhage, necrosis (100 μg mL^-1); f = leukocyte infiltrate, hemorrhage, necrosis (150 μg mL^-1). Hematoxylin Eosin - 400x. 10-μm scale.

Table IV
Prevalence of alterations in the gills of Piaractus mespotamicus exposed to concentrations of Serjania erecta aqueous extract.

	Prevalence	Serjania erecta aqueous extract concentration
Alteration index	(%)	0 µg mL-1	2.5 µg mL-1	12.5 µg mL-1	25.0 µg mL-1	50.0 µg mL-1	100.0 µg mL-1	150.0 µg mL-1
Necrosis	30.2	N.D	N.D	0.03 ± 0.02 a	0.75 ± 0.37 ab	1.15 ± 0.42 ab	1.26 ± 0.40 ab	1.44 ± 1.03 b
Telangiectasia ns	22.2	0.14 ± 0.03 a	0.11 ± 0.09 a	0.16 ± 0.07 a	0.71 ± 0.05 a	0.53 ± 0.46 a	0.74 ± 0.12 a	0.53 ± 0.46 a
Dilation of vessels and filaments ns	22.2	0.08 ± 0.03 a	0.05 ± 0.04 a	0.21 ± 0.06 a	0.28 ± 0.08 a	0.26 ± 0.05 a	0.68 ± 0.04 a	0.78 ± 0.10 a
Epithelial hyperplasia	17.5	0.15 ± 0.07 ab	0.44 ± 0.22 bc	0.59 ± 0.17 c	N.D	N.D	N.D	N.D
Epithelial displacement ns	14.3	0.17 ±0.04 a	0.31 ± 0.12 a	N.D	N.D	N.D	N.D	0.53 ± 0.46 a
Eosinophilic granular inflammatory infiltrate	14.3	N.D	N.D	N.D	N.D	0.70 ± 0.06 b	0.76 ± 0.12 b	0.86 ± 0.08 b
Interlamellar hyperplasia ns	12.7	N.D	N.D	0.21 ± 0.06 a	0.65 ± 0.04 a	0.30 ± 0.02 a	0.50 ± 0.44 a	N.D
Cellular edema ns	12.7	0.23 ±0.12 a	0.11 ± 0.09 a	0.11 ± 0.09 a	N.D	N.D	0.33 ± 0.08 a	0.29 ± 0.05 a
Lamellar fusion ns	4.8	0.03 ± 0.02 a	0.10 ± 0.07 a	0.11 ± 0.09 a	N.D	N.D	N.D	N.D
Congestion ns	1.6	0.03 ±0.02 a	N.D	N.D	N.D	N.D	N.D	N.D

• Mean ± standard deviation, different letters indicate significant differences between the treatments using one-way ANOVA and Tukey’s test (P <0.05). N.D. (not detected). ^ns Not significant.

The following hepatic alterations were observed: fatty degeneration, vacuolization, sinusoidal and vessel congestion, necrosis, hepatocyte hypertrophy, leukocyte infiltrate, pyknotic nucleus, peripheral cellular nucleus, accumulation of hemosiderin pigments, cord structure disruption, and melanomacrophage centers (Figure 2). The most prevalent of these conditions was fatty degeneration (77.4%), and the least prevalent was melanomacrophage (14.3%). The MAV of fatty degeneration and necrosis (P<0.05) from the aqueous extract concentration of 12.5 µg mL^-1, whereas nuclear deformity increased (P<0.05) only at levels equal to or higher than 50 µg mL^-1 and sinusoidal and vessel congestion prevalence increased (P<0.05) after 100 µg mL^-1. Alterations in vacuolization, hepatocyte hypertrophy, leukocyte infiltrate, pyknotic nucleus, and cord structure disruption were greater (P<0.05) than in control group only for the extract concentrations of 100.0, 25.0, 12.5, 25.0, and 50 µg mL^-1, respectively (Table V).

Figure 2
Liver tissue of Piaractus mesopotamicus exposed to Serjania erecta aqueous extract. a = sinusoidal congestion, peripheral cell nucleus (2.5 μg mL^-1); b = hepatocyte hypertrophy, fatty degeneration (12.5 μg mL^-1); c = leukocyte infiltrate, necrosis, vacuolization (25 μg mL^-1); d = fatty degeneration, loss of cell boundary, necrosis (50 μg mL^-1); e = sinusoidal congestion, vacuolization, inflammatory infiltrate (100 μg mL– ¹ ); f = necrosis, loss of cell boundary (150 μg mL^-1). Hematoxylin Eosin- 400x. 10-μm scale.

Table V
Prevalence of alterations in the liver of Piaractus mespotamicus exposed to concentrations of Serjania erecta aqueous extract.

	Prevalence	Serjania erecta aqueous extract concentration
Alteration index	(%)	0 µg mL-1	2.5 µg mL-1	12.5 µg mL-1	25.0 µg mL-1	50.0 µg mL-1	100.0 µg mL-1	150.0 µg mL-1
Fatty degeneration	77.4	0.03 ± 0.02 a	0.54 ± 0.36 a	1.77 ± 0.37 b	1.84 ± 0.77 b	2.27 ± 0.25 b	2.71 ± 0.07 b	2.73 ± 0.10 b
Vacuolization	45.2	N.D	0.29 ± 0.08 ab	0.94 ± 0.17 ab	1.06 ± 0.13 ab	0.85 ± 0.76 ab	1.44 ± 0.25 b	1.20 ± 0.60 ab
Sinusoidal and vessel congestion	42.7	N.D	0.24 ± 0.06 a	0.91 ± 0.47 ab	0.67 ± 0.38 ab	0.64 ± 0.56 ab	1.76 ± 0.54 b	1.71 ± 0.64 b
Necrosis	40.5	N.D	N.D	0.21 ± 0.19 b	0.53 ± 0.35 b	0.43 ± 0.16 b	1.35 ± 0.65 c	1.70 ± 0.69 c
Hepatocyte hypertrophy	39.3	N.D	0.44 ± 0.32 ab	0.81 ± 0.16 ab	1.55 ± 0.32 b	0.33 ± 0.17 ab	0.67 ± 0.40 ab	1.34 ± 0.78 ab
Leukocyte infiltration	32.1	N.D	0.17 ± 0.09 ab	1.23 ± 0.26 b	0.70 ± 0.22 ab	1.11 ± 0.19 ab	0.54 ± 0.17 ab	0.88 ± 0.06 ab
Pyknotic nucleus	25.0	0.10 ± 0.07 a	0.43 ± 0.04 ab	0.37 ± 0.31 ab	1.38 ± 0.70 b	N.D	0.16 ± 0.08 a	N.D
Peripheral nucleus ns	22.6	N.D	0.17 ± 0.03 a	0.31 ± 0.18 a	0.93 ± 0.24 a	0.35 ± 0.11 a	0.86 ± 0.76 a	0.16 ± 0.08 a
Nuclear deformation	22.6	N.D	0.08 ± 0.03 a	N.D	N.D	0.82 ± 0.37 b	1.09 ± 0.56 b	1.07 ± 0.12 b
Hemosiderosis ns	16.7	0.04 ± 0.02 a	N.D	0.33 ± 0.13 a	0.27 ± 0.24 a	0.17 ± 0.12 a	0.68 ± 0.19 a	0.48 ± 0.12 a
Cord structure disruption	15.5	N.D	0.03 ± 0.01 a	0.26 ± 0.24 a	N.D	1.44 ± 0.24 b	0.33 ± 0.18 a	N.D
Melanomacrophage ns	14.3	0.03 ± 0.02 a	0.09 ± 0.01 a	0.29 ± 0.16 a	0.15 ± 0.26 a	0.24 ± 0.12 a	0.33 ± 0.17 a	0.17 ± 0.09 a

• Mean ± standard deviation, different letters indicate significant differences between the treatments using one-way ANOVA and Tukey’s test (P <0.05). N.D. (not detected). ^ns Not significant.

DISCUSSION

The species of the family Sapindaceae are rich in secondary metabolites such as phenolic compounds, flavonoids, saponins, tannins, triterpenes, diterpenes, isoprenoids, polyphenols, lecithins, and hydrogels (Lima et al. 2006, Gomig et al. 2008, Broggini et al. 2010, Cardoso et al. 2013, Moreira et al. 2019). In the current study, the aqueous extract of S. erecta leaves obtained by cold pressing for 24 h exhibited phenolic compounds, flavonoids, tannins, and saponins. Similar results were obtained by Broggini et al. (2010) in S. erecta leaf aqueous extract obtained with boiling distilled water and kept at rest for 24 h in the dark. The authors identified saponins, tannins, and flavonoid glycosides. Gomig et al. (2008) showed the presence of saponins, flavonoids, triterpenoids, steroids, tannins, and catechins in methanolic extract obtained from stems and leaves of the same plant.

Studies on behavioral and acute toxicity caused by S. erecta aqueous extract administered orally and intraperitoneally to mice revealed low toxicity (Broggini et al. 2010). As for the methanolic and chloroformic extracts of S. erecta leaves, no acute toxicological effects were observed, either, when they were administered orally to male and female mice (Castelo et al. 2009). At high concentrations, flavonoids may generate reactive oxygen species by autoxidation and redox cycle (Hodnick et al. 1986, Yoshino et al. 1999), whereas tannins have high affinity for proteins in addition to complexing to metal ions (Monteiro et al. 2005). Furthermore, they may trigger hepatotoxic activities and antinutritional effects (Chung et al. 1998).

Studies with guppy (Poecilia reticulata) demonstrated the toxicity of serjanoside B at the concentration of 20 µg ml^-1 obtained from S. lethalis, which caused excitatory responses in the initial stage, followed by progressive hypoactivity and death after 5 h of exposure (Teixeira et al. 1984). In the present study, all fish exposed to the S. erecta aqueous extract concentrations of 100 and 150 µg mL^-1 died within 4 h of exposure. Experimental intoxication by saponins in Poecilia sp. and Geophagus sp. showed that the action of saponins is developed throughout five stages, whose duration varies according to the concentration of the toxic agent, species used, and temperature (Tabarelli Neto & Bonoldi 1945). Plants of the family Sapindaceae are characterized as abundant sources of saponins (Voutquenne et al. 2002), a compound class present in high amounts in different extracts of S. erecta leaves and stems (Gomig et al. 2008, Broggini et al. 2010). On this basis, it can be inferred that the toxic effects of S. erecta aqueous extract observed in the present animals are directly related to the concentration of extract used in the water and the presence of saponins in its composition.

Biochemical parameters are important indicators of physiological alterations in fish, and osmoregulation is characterized as an important target for toxic substances (Winkaler et al. 2007). The plasma ionic profile of the fish exposed to S. erecta extract did not show significant differences compared to control group, for any of the evaluated parameters. Similar results were obtained with neem (Azadirachta indica) leaf extract, which did not interfere with the osmoregulatory capacity of P. lineatus (Winkaler et al. 2007). Nevertheless, saponins have hemolytic activity and act on the cell membrane, with specific capacity to form pores and alter cell plasma membrane fluidity (Ma & Xiao 1998). In the present study, no alterations were observed in the blood variables except for glucose levels, which rose with the extract concentration, and plasma Na⁺ level, which declined as the plant extract concentration was increased. This increase in blood glucose can be seen as part of a stress response triggered by the presence of S.erecta extract in water. Hyperglycemia has also been found in P. lineatus exposed to 2.5 g L− ¹ neem leaf extract (Winkaler et al. 2007). This shows that the observed effects depend on the concentration of the toxic agent (Tabarelli Neto & Bonoldi 1945, Francis et al. 2002).

Researchers have related the toxicity of a plant extract to tissue alterations in gills (Temmink et al. 1989, Winkaler et al. 2007, Kumar et al. 2010, Abalaka et al. 2015a). When added to water in large amounts, saponins are highly toxic to fish due to damage caused to the respiratory epithelium of the gills, which stems from a reduction of water surface tension (Newinger 1994, Chen et al. 1996, Francis et al. 2002). In addition, they have hemolytic activity, releasing hemoglobin into the plasma, irritating the mucosae in general, and acting on the nervous system with an initial manifestation in the motor system and, later, on the sensory system (Tabarelli Neto & Bonoldi 1945).

In Heteropneusteus fossilis exposed to saponin, researchers observed hypertrophy of the branchial epithelium (Hemalatha & Banerjee 1997). The main mechanism of action of saponins is by reducing the surface tension between water and the gills, preventing oxygen absorption and leading the fish to a slow death from oxygen deprivation (Lamba 1970). When oxygen absorption is impeded, the animal metabolism initiates mechanisms to offset it, triggering numerous lesions which could be considered specific signs of the fish defense mechanism (Velasco-Santamarίa & Cruz-Casallas 2008). However, when not controlled, this mechanism may incorporate a degenerative nature, resulting in loss of morphofunctional efficiency of the branchial structures in the exposed fish (Hemalatha & Banerjee 1997), as observed in the present study.

Epithelial desquamation and lamellar fusion occur as defensive mechanisms to reduce the branchial surface area in contact with the aggressor toxic agent (Figueiredo-Fernandes et al. 2007). Such lesions occur whenever the osmoregulatory function of the gills is disturbed (Temmink et al. 1989) and the cell is not able to maintain homeostasis (Mcgavin & Zachary 2013). Similar lesions were reported in the gills of carp (Cyprinus carpio) exposed to tannins (Temmink et al. 1989) and in the gills of African sharptooth catfish (Clarias gariepinus) exposed to ethanolic extract from the stem-bark of A. obesum (Abalaka et al. 2015a). However, these alterations impair gas exchanges in the affected animals, reducing the surface area available for this process (Velasco-Santamarίa & Cruz-Casallas 2008), which results in respiratory discomfort and mortality.

The difficulty capturing oxygen leads to hypoxia, which increases respiratory frequency, resulting in behavioral changes (Tiwari et al. 2011) such as the aerial respiration observed in the current study. In an effort to limit the absorption of the toxic agent by the gills, the organism coats the body surfaces by increasing mucus release (Abalaka & Auta 2010), an adaptive response that may heighten the observed behavioral alterations.

The high reactivity of tannins to proteins (Gupta & Haslam 1980) renders these compounds toxic to aquatic organisms such as fish (Temmink et al. 1989). Excessive production of mucus and lesions, which result in secondary lamellae fusion, as observed here, suggests that tannins inactivate the transport of ATPase membranes in the gill epithelium. In C. carpio exposed to tannins, big holes were observed in the primary gill epithelium, which might be associated with the ability to bind to hydrogen and low penetration capacity of tannins of high molecular mass (Temmink et al. 1989).

The liver is an important organ in the metabolism of potentially toxic compounds (Moreira et al. 2019). Several plant metabolites, including tannins and flavonoids found in S. erecta extract, may be toxic to the liver, even at low concentrations (Chung et al. 1998, Galati & O’Brien 2004, Watjen et al. 2005, Tsuji & Walle 2008). Histological analyses of the liver tissue of the animals in the present experiment demonstrated that S. erecta aqueous extract is potentially hepatotoxic, which is reflected in the observed congestion, vacuolization, fatty degeneration, cellular infiltrate, and necrosis. Similar lesions were found in C. gariepinus exposed to ethanolic extract from A. obesum (Abalaka et al. 2015b). According to those authors, these are pathological responses which may be induced by metabolic disorders resulting from exposure to a toxic agent.

In an in vitro study with liver tissue cells from Oncorhynchus mykiss, flavonoids influenced hepatocyte cell growth, revealing cytotoxicity to those cells (Tsuji & Walle 2008). In rats, flavonoids induced cytotoxicity and DNA strand breaks (Watjen et al. 2005). According to Galati & O’Brien (2004), flavonoids are toxic due to their pro-oxidative capacity and because they induce mitochondrial dysfunction. This fact can be explained by the presence of activities similar to peroxidase in the hepatocytes of fish, which can activate flavonoids for a toxic species, damaging the cells (Tsuji & Walle 2008). In view of this, it may be inferred that S. erecta aqueous extract is potentially toxic to the fish.

CONCLUSIONS

The observed results demonstrate the importance of evaluating functional and morphological responses of fish exposed to plant extracts with different purposes for aquaculture. Serjania erecta extract at concentrations higher than 2.5 µg mL^-1 caused detrimental morphofunctional alterations in the gills and liver of P. mesopotamicus, discouraging the use of this extract at those concentrations for phytotherapeutic purposes in aquaculture.

REFERENCES

• ABALAKA SE & AUTA J. 2010. Toxicity of aqueous and ethanol extracts of Parkia biglobosa pods on Clarias gariepinus juveniles. J Anim Vet Adv 9: 1068-1072.
• ABALAKA SE, FATIHU MY, IBRAHIM NDG & AMBALI SF. 2015b. Liver histopathological changes in Clarias gariepinus exposed to ethanol extract of Adenium obesum stem bark. J Morphol Sci 32: 22-28.
• ADESINA BT, OMITOYIN BO & OGUNTUGA O. 2013. A Preliminary Toxicological Screening of Ichthyotoxic Compound of Moringa oleifera (LAM.) Hot Ethanolic Extract to Freshwater Fish, Oreochromis niloticus (L.) Fingerlings. Agrosearch 13: 256-265.
• AGOSTINI-COSTA TS, GARRUTI DS, LIMA L, FREIRE S, ABREU FAP & FEITOSA T. 1999. Avaliação de metodologias para determinação de taninos no suco de caju. Boletim CEPPA Curitiba 17: 167-176.
• BOYD CE. 1998. Water Quality for Pond Aquaculture. International Center OF Aquaculture and Aquatic Environments, Alabama Agricultural Experiment Station, Auburn University, 37 p.
• BRAZILIAN PHARMACOPEIA. 2010. 5a ed., Volume 1 e 2. Agência Nacional de Vigilância SumanEutumarEuuma[Eunternet] 2010. DEusponEuveeuem: <http://WWW.umanvEusuma.gov.br/hotsEute/cd_farmacopeia/index.htm >. Acesso em 10 de dezembro de 2018.
  » http://WWW.umanvEusuma.gov.br/hotsEute/cd_farmacopeia/index.htm
• BROADHURST RB & JONES WT. 1978. Analysis of condensed tannins using acidified vanillinz Journal of the Science of Food and Agriculture, Baffins Lane Chichester 29: 788-794.
• BROGGINI LSC, FERNANDES RS, NOGUEIRA T, SUZANO FR, CAETANO AL, BUCK HS, COUTO LB, FRANÇA SC & PEREIRA PS. 2010. Behavioral and enzymatic bioassays with Serjania erecta Radlk., Sapindaceae, correlated with cognitive dysfunctions. Braz J Pharmacog 20: 519-528.
• CARDOSO CAL, COELHO RG, HONDA NK, POTT A, PAVAN FR & LEITE CQF. 2013. Phenolic compounds and antioxidant, antimicrobial and antimycobacterial activities of Serjania erecta Radlk. (Sapindaceae). Braz J Pharm Sci 49: 775-782.
• CASTELO AP, ARRUDA BN, COELHO RG, HONDA NK, FERRAZOLI C, POTT A & HIRUMA-LIMA CA. 2009. Gastroprotective Effect of Serjania erecta Radlk (Sapindaceae): Involvement of Sensory Neurons, Endogenous Nonprotein Sulfhydryls, and Nitric Oxide. J Med Food 12: 1411-1415.
• CATANI RA & PAIVA NETO JK de. 1949. Dosagem do potássio e sódio pelo «fotômetro de chama”: Sua aplicação em análise de solo. Bragantia [online] 12: 175-183.
• CHEN JC, CHEN KW & CHEN JM. 1996. Effects of saponin on survival, growth, molting and feeding of Penaeus japonicus juveniles. Aquaculture 144: 165-175.
• DE MESQUITA ML, GRELLIER P, MAMBU L, DE PAULA JE & ESPINDOLA LS. 2007. In vitro antiplasmodial activity of Brazilian Cerrado plants used as traditional remedies. J Ethnopharmacol 110: 165-170.
• DUKE SO. 1990. Natural pesticides from plants. p. 511-517. In: Janick J and Simon JE (Eds), Advances in new crops. Timber Press, Portland, OR.
• FERRUCCI MS. 2004. Guías ilustradas de classes - sapindacea e juss. Flora.
• FIGUEIREDO-FERNANDES A, FERREIRA-CARDOSO J & SANTOS SG. 2007. Histopathological changes in liver and gill epithelium of Nile tilapia, Oreochromis niloticus, exposed to waterborne copper. Pesqui Vet Bras 27: 103-109.
• FRANCIS G, KEREM Z, MAKKAR HPS & BECKER K. 2002. The biological action of saponins in animal systems: a review Br J Nutr 88: 587-605.
• GALATI G & O’BRIEN PJ. 2004. Potential toxicity of flavonoids and other dietary phenolics: significance for their chemopreventive and anticancer properties. Free Radic Biol Med 37: 287-303.
• GOMIG F, PIETROVSKI EF, GUEDES A, DALMARCO EM, CALDERARI MT, GUIMARÃES CL, PINHEIRO RM, CABRINI DA & OTUKI MF. 2008. Topical anti-inflammatory activity of Serjania erecta Radlk (Sapindaceae) extracts. J Ethnopharmacol 118: 220-224.
• GUARIM NETO G, SANTANA SR & SILVA JVB. 2000. Notas etnobotânicas de espécies de Sapindaceae Jussieu. Acta Bot Bras 14: 327-333.
• GUPTA RK & HASLAM E. 1980. Vegetable tannins-- structure and biosynthesis. In Polyphenols in Cereals and Legumes (Edited by Hulse JH), International Development Research Centre, Ottawa, Canada, p. 15-24.
• HAMILTON MA, RUSSO RC & THURSTON RV. 1978. Trimmed Spearman–Karber method for estimating median lethal concentrations in toxicity bioassays. Environ Sci Technol 11: 714-719.
• HEMALATHA S & BANERJEE TK. 1997. Histopathological analysis of acute toxicity of zinc chloride to the respiratory organs of the air breathing catfish, Heteropneustes fossilis (Bloch). Vet Arh 67: 1-24.
• HEREDIA-VIEIRA SC, SIMONET AM, VILEGAS W & MACÍAS FA. 2015. Unusual C, O-Fused glycosyl apigenins from Serjania marginata leaves. J Nat Prod 78: 77-84.
• HERNÁNDEZ GP ET AL. 2012. In vitro and in vivo trypanocidal activity of native plants from the Yucatan Peninsula. Parasitol Res 110: 31-35.
• HODNICK WF, KUNG FS, ROETTGER WJ, BOHMONT CW & PARDINI RS. 1986. Inhibition of mitochondrial respiration and production of toxic oxygen radicals by flavonoids. A structure-activity study. Biochem Pharmacol 35: 2345-2357.
• HOLMBERG EL. 1897. Sobre algunos peces nuevos o poco conocidos de la República Argentina. Revista Argentina de Historia Natural 1: 180-193.
• IBGE - INSTITUTO BRASILEIRO DE GEOGRAFIA E ESTATÍSTICA. 2017. Pesquisa Pecuária Municipal 2017. Disponível em:< https://sidra.ibge.gov.br/Tabela/3940#resultado>. Acesso em:18 nov. 2018.
  » https://sidra.ibge.gov.br/Tabela/3940#resultado
• KAVITHA C, RAMESH M, KUMARAN SS & LAKSHMI SA. 2012. Toxicity of Moringa oleifera seed extract on some hematological and biochemical profiles in a freshwater fish, Cyprinus carpio. Exp Toxicol Pathol 64: 681-687.
• KUMAR A, PRASAD MR, SRIVASTAVA K, TRIPATHI S & SRIVASTAV KA. 2010. Branchial histopathological study of catfish Heteropneustes fossilis following exposure to purified nem extract, Azadirachtin. World J Zool 5: 239-243.
• LIN JY & TANG CY. 2007. Determination of total phenolic and flavonoid contents in selected fruits and vegetables, as well as their stimulatory effects on mouse splenocyte ploriferation. Food Chem 101: 140-147.
• MA LY & XIAO PG. 1998. Effects of Panax notoginseng saponins on platelet aggregation in rats with middle cerebral artery occlusion or in vitro and on lipid fluidity of platelet membrane. Phytother Res 12: 138-140.
• MCGAVIN MD & ZACHARY JF. 2013. Bases da patologia em veterinária, 5a ed., Mosby Elsevier, Rio de Janeiro, 1324 p.
• MONTEIRO JM, ALBUQUERQUE UP DE, ARAÚJO E DE L & AMORIM ELC DE. 2005. Taninos: uma abordagem da química à ecologia. Quím Nova 28: 892-896.
• NEWINGER HD. 1994. Fish poisoning plants in Africa. Bot Acta 107: 263-270.
• OECD. 1992. Test No. 203: Fish, Acute Toxicity Test: OECD Publishing.
• PÉRICO LL ET AL. 2015. Does the gastroprotective action of a medicinal plant ensure healing effects? An integrative study of the biological effects of Serjania marginata Casar. (Sapindaceae) in rats. J Ethnopharmacol 172: 312-324.
• POTT A, POTT VJ & SOBRINHO AB. 2004. Plantas úteis à sobrevivência no Pantanal. Corumba: Editora da Embrapa, 24 p.
• SILVA FL, SILVA JLV DA, SILVA JM, MARCOLIN LSA, NOUAILHETAS VLA, YOSHIDA M, VENDRAMINI PH, EBERLIN MN, BARBOSA-FILHO JM & MORENO PRH. 2017. Antispasmodic activity from Serjania caracasana fractions and their safety. Braz J Pharmacog 27: 346-352.
• SOARES BV & TAVARES-DIAS M. 2013. Espécies de Lippia (Verbenaceae), seu potencial bioativo e importância na medicina veterinária e aquicultura. Biot Amaz 3: 109-123.
• STECKERT LD, CARDOSO L, JERÔNIMO GT, PÁDUA SB DE & MARTINS ML. 2018. Investigation of farmed Nile tilapia health through histopathology. Aquaculture 486: 161-169.
• SYNGAI GG, DEY S & BHARALI R. 2016. Evaluation of toxicity levels of the aqueous extract of allium sativum and its effects on the behavior of juvenile common carp (cyprinus carpio l., 1758). J Pharm Clin Res 9: 417-421.
• TABARELLI NETO F & BONOLDI V. 1945. Da ação da Saponina sobre peixes: Guarus (Poecilia sp.) e acarás (Geophagus sp.). Braz J Vet Res An Sci 3: 1-2.
• TEIXEIRA JRM, LAPA AJ, CADEN S & VALLE JR. 1984. Timbós: Ichthyotoxic plants used by brazilian indians. J Ethnopharmacol 10: 311-318.
• TEMMINK JHM, FIELD JA, VAN HAASTRECHT JC & MERKELBACH RCM. 1989. Acute and sub-acute toxicity of bark tannins in carp (Cyprinus carpio l.) Water Res 23: 341-344.
• TIWARI M, NAGPURE NS, SAKSENA DN, KUMAR R, SINGH SP, KUSHWAHA B & LAKRA WS. 2011. Evaluation of acute toxicity levels and ethological responses under heavy metal cadmium exposure in freshwater teleost, Channa punctata (Bloch). Int J Aquat Sci 2: 36-47.
• URBINATI EC & GONÇALVES FD. 2013. Pacu (Piaractus mesopotamicus), In: Baldisserotto B and Gomes LC (Eds), Espécies nativas para piscicultura no Brasil. UFMS, Santa Maria, Cap. 8, p. 225-246.
• VAZZOLER AEA DE M. 1996. Biologia da reprodução de peixes teleósteos: teoria e prática. Maringá, EDUEM, 169 p.
• VELASCO-SANTAMARÍA YM & CRUZ-CASALLAS PE. 2008. Behavioural and gill histopathological effects of acute exposure to sodium chloride in moneda (Metynnis orinocensis). Environ Toxicol Pharmacol 25: 365-372.
• VOUTQUENNE L, KOKOUGAN C, LAVAUD C, POUNY L & LITAUDON M. 2002. Triterpenoid saponins and acylated prosapogenins from Harpullia austrocaledonica. Phytochemistry 59: 825-832.
• WATJEN W, MICHELS L, STEFFAN B, NIERING P, CHOVOLOU Y, KAMPKOTTER UM, TRAN-THI QH, PROKSCH P & KAHL R. 2005. Low Concentrations of Flavonoids Are Protective in Rat H4IIE Cells Whereas High Concentrations Cause DNA Damage and Apoptosis. Nutr Inter Toxic 135: 525-531.
• WINKALER EU, SANTOS TRM, MACHADO-NETO JG & MARTINEZ CBR. 2007. Acute lethal and sublethal effects of neem leaf extract on the neotropical freshwater fish Prochilodus lineatus. Comp Biochem Physiol C 145: 236-244.
• YOSHINO M, HANEDA M, NARUSE M & MURAKAMI K. 1999. Prooxi- dant activity of flavonoids: copper-dependent strand breaks and the formation of 8-hydroxy-2 -deoxyguanosine in DNA. Mol Genet Metab 68: 468-472.

Publication Dates

• Publication in this collection
  03 Sept 2021
• Date of issue
  2021

History

• Received
  24 Apr 2019
• Accepted
  27 July 2019