Antimicrobial and Toxic Activity of Citronella Essential Oil (Cymbopogon nardus), and Its Effect on the Growth and Metabolism of Gilthead Seabream (Sparus aurata L.)

Abstract

Aquaculture procedures usually induce stress that affects the physiological status of fish. For this reason, the inclusion of additives in fish feeds to palliate stress might be a good alternative. This study aimed to assess the antimicrobial activity of citronella (Cymbopogon nardus) essential oil (CEO) against bacterial pathogens and to determine its dietary impact on the growth performance of Sparus aurata. In vitro tests confirmed that CEO possesses antimicrobial activity against several fish-specific pathogens. For the in vivo tests, three experimental groups were fed for 60 days with different concentrations of CEO: CTRL (0 mL kg⁻¹ fish feed); CEO1 (1 mL kg⁻¹ fish feed); and CEO2 (2 mL kg⁻¹ fish feed). At the end of the experiment, the physiological status was characterized. Subsequently, the specimens of the CTRL and CEO2 groups were subjected to a challenge with an injection of Poly I:C for immune stimulation. Although S. aurata individuals tolerated CEO inclusion without compromising growth performance, it significantly reduced glycogen in the CEO2 group, concomitant to an increment of total peripheral leucocytes. Moreover, different hematological profiles’ responsive patterns against an inflammatory stimulus were observed. In conclusion, our results suggest that the use of CEO as a fish feed additive can prevent bacterial outbreaks and improve potential in vivo disease resistance in S. aurata without negatively affecting growth.

1. Introduction

Crowding densities, transportation, and handling cause major stress, altering the physiology and behavior of fish and ultimately causing significant economic losses produced by pathogen outbreaks [1,2]. Stress has a wide range of effects on overall physiology, involving several clinical manifestations related to metabolism, hydromineral balance, and/or immune system alterations [3,4].

Although antibiotics have been used to prevent and treat bacterial diseases in farmed fish, some pathogens have acquired resistance to many of these, including kanamycin, sulfonamide, tetracycline, ampicillin, chloramphenicol, florfenicol, and erythromycin [5]. Alternatives, such as medicinal plants in fish feed, have been a major focus recently [6,7]. Extracts and essential oils (EOs) of medicinal plants contain phenolics, polyphenols, alkaloids, terpenoids, and polypeptide compounds, which are natural and effective alternatives to antibiotics, agrochemicals, and other synthetic compounds [8,9]. These substances, which have growth-promoting properties, antimicrobial ability, and antistress properties, act as immune system enhancers and appetite stimulants [10,11]. Moreover, its minimal side effects on animal welfare [12] make their use an important means to improve farmed fish’s health and/or performance [11,13].

Citronella (Cymbopogon nardus) is a plant used to obtain an essential oil mainly composed of citronellal, citronellol, and geraniol [14]. These substances have been described as insect repellent, antiseptic, and bactericidal [15,16,17]. The toxicity of a substance can be measured via in vitro tests without the need for the use of experimental animals. One of the most commonly used in vitro toxicity tests is the Microtox test [18,19,20], where the decrease in the bioluminescence of the bacterium Vibrio fischeri is used as an indicator of acute toxicity. However, data regarding the dietary effects of citronella essential oil (CEO) on fish metabolism and its potential toxicity are scarce. CEO’s effects have been shown to depress the central nervous system in rodents and dogs. In fish, it has also been demonstrated that CEO promotes muscle relaxation and significantly modulates swimming behavior and cardioventilatory responses [21]. Remarkably, CEO may also induce lower cortisol levels in treated Cyprinus carpio than other EOs [22]. Furthermore, previous studies on the high inclusion of citronella (400 mg kg⁻¹ fish feed) in a diet showed that the fish had lower intestinal total bacteria while enhancing growth performance, feed utilization, and disease resistance [23]. Lastly, based on the available ecological data, exposure information, and the oil of citronella’s non-toxic mode of action, the uses of citronella oil pose minimal risk to nontarget wildlife and fish, according to the U.S. Environmental Protection Agency (EPA).

Gilthead seabream (Sparus aurata L.) is one of the most important Mediterranean aquaculture species [24]. However, intensive rearing conditions cause stress to fish, making them vulnerable to infectious diseases [3]. Some bacterial pathogens that affect gilthead seabream farming are Vibrio harveyi, Aeromonas hydrophila, and Photobacterium damselae subsp. piscicida [25]. Under this framework, CEO supplementation in fish feeds could be of interest as a preventive measure or as an alternative to antibiotic use in order to improve the immune status while not altering the growth and health of the gilthead seabream.

Thus, this work aimed (1) to determine the capacity of CEO to inhibit bacterial pathogens in S. aurata; (2) to assess the in vitro acute toxicity of CEO; (3) to evaluate the influence of dietary CEO on S. aurata growth and metabolism; and (4) to assess immune stimulation after challenging fish fed with CTRL and CEO2 to a subsequent viral infection-like treatment (i.e., Poly I:C).

2. Materials and Methods

2.1. In Vitro Antimicrobial Activity

According to Wei and Wee [16], citronella essential oil (CEO) was obtained from Cymbopogon nardus by steam distillation. Specifically, 600 mL of CEO was obtained from every 100 kg of green biomass from the plant. Its density at 20 °C was in the range of 0.87–0.89 g mL⁻¹. See Supplementary Materials for further information.

For the antimicrobial test, three fish pathogenic bacterial strains—Vibrio harveyi Lg16/00, Photobacterium damselae subsp. piscicida Lg41/01, and Aeromonas hydrophila Lg28/4—were used [26,27]. All of them were previously isolated from Solea senegalensis. The non-virulent bacterium Vibrio fischeri (isolated from seawater) was also used. Strains were cultivated in trypticase soy agar (TSA; Oxoid Ltd., Basingstoke, Hampshire, UK) with 2% NaCl (TSAs) for 24–48 h at 22 °C.

Aqueous and ethanolic dilutions of CEO were prepared using a saline solution and ethanol, respectively. The aqueous extracts were tested using the diffusion method, as Rattanachuay et al. [28] described. Bacterial colonies were suspended in a saline solution (2% of NaCl), matched to a turbidity of 0.5 McFarland standard (approximately 10⁸ CFU mL⁻¹). The bacterial suspension was spread over the surface of the TSA plates (with a 5 mm media thickness) to obtain an even inoculum. Wells with a 6 mm diameter were cut into the agar and then filled with 60 µL of different CEO concentrations. Saline solution was added as a negative control. A sterilized filter paper disc with a 6 mm diameter (Whatman no. 1 filter paper) was used for the ethanolic extracts. An inoculum of 10 μL for each tested concentration was transferred to each disc. As a negative control, 10 μL of ethanol was applied to the filter paper disc. The disks were dried at 37 °C for 15 min to evaporate the ethanol. The suspensions of microorganisms were inoculated into TSA plates as previously described, and then the disks were placed over the plates. The plates were incubated at 22 °C for 24–48 h. The antimicrobial activity was determined by the presence/absence of an inhibition area around each well or disc. The lowest CEO concentration that inhibited the growth of microorganisms was designated as the minimum inhibitory concentration (MIC).

2.2. Assessment of CEO Acute Toxicity

The acute toxicity test was performed by modifying the method described by Johnson [29] using the bioluminescent bacterium V. fischeri and the Microtox^® M500 test (Microbics Corporation, Carlsbad, CA, USA). Briefly, the CEO was diluted in serial fold 1:2 in 2% saline solution. Cuvettes with 1 mL of each dilution and control (2% saline solution) were maintained at 15 °C. V. fischeri, grown for 24 h in TSAs, was suspended in 2% saline solution at 0.6 optical density (600 nm), and maintained at 5 °C before use. Then, all the cuvettes were inoculated with 20 µL of the V. fischeri suspension. The samples were incubated at 15 °C for 15 min. After incubation, the bioluminescence of V. fischeri in each cuvette was measured by Microtox luminometer and relativized with those of the control, obtaining a percent decrease in bioluminescence. The luminescence inhibition was correlated with the toxicity of the concentration of CEO [30]. The percent reduction in the bioluminescence of V. fischeri produced by the CEO was recorded as the median and maximal effective concentration (EC₅₀ and EC₁₀₀) values.

2.3. Experimental Diets and Feeding Trial

A commercial isoproteic, isolipidic, and isoenergetic fish feed (57% crude protein, 18% crude fat, 10% ash, 1.6% phosphorus, and 19.5 mJ kg⁻¹ digestible energy, Skretting, Burgos, Spain) was used as a basis to prepare the experimental diets. Three concentrations of CEO were added: (i) CTRL (0 mL CEO kg⁻¹ fish feed), (ii) CEO1 (1 mL CEO kg⁻¹ fish feed), and CEO2 (2 mL CEO kg⁻¹ fish feed). All experimental feeds were elaborated by the Service of Experimental Diets from the University of Almería (Campus de Excelencia Internacional del Mar CEI·MAR, Almería, Spain). The addition of CEO to the feed was carried out by diluting this CEO in 100–150 mL of ethanol. Then, it was homogeneously sprayed onto the feed following a modified methodology described by Morgado et al. [31], i.e., 885 and 1770 mg of CEO kg⁻¹ fish feed. The feed was dried at 40 °C on a mesh to allow for the evaporation of ethanol and stored at −20 °C until use. Note that control pellets were treated with equal amounts of vehicle (ethanol).

Juvenile individuals of S. aurata (n = 162, 17.7 ± 1.5 g body weight, 10.2 ± 0.3 cm length) were provided by Servicios Centrales de Investigación en Cultivos Marinos (SCI-CM, CASEM, University of Cádiz, Puerto Real, Cádiz, Spain; Operational Code REGA ES11028000312) and were randomly distributed in a flow-through setup of nine 80 L tanks (n = 18 per tank). The three experimental dietary groups (CTRL, CEO1, and CEO2) were then established in triplicates. The fish were maintained under natural photoperiod (May–July 2017), temperature (18–19 °C), and salinity (39 g L⁻¹). Supplemental aeration was provided to maintain dissolved oxygen at 6.8 ± 0.4 mg L⁻¹. Ammonia (<0.1 mg L⁻¹), nitrite (<0.2 mg L⁻¹), and nitrate (<50 mg L⁻¹) were determined once weekly. Fish were fed twice per day (9:00 a.m. and 1:00 p.m.) at a rate of 2% of their body weight over the 60-day trial.

All the experimental procedures complied with guidelines of the University of Cádiz (Spain) and the European Union Council (2010/63/EU) for the use of animals in research. The experimental procedures were previously approved by the Spanish Government’s Ethics and Animal Welfare Committee (RD53/2013) and endorsed by the Regional Government (Junta de Andalucía reference number 28-04-15-241).

2.4. Injection of Poly I:C for Immune-Stimulation after Feeding Trial

At the end of the experiment, a total of 64 individuals (previously fed with CTRL and CEO2 for 60 days; 44.7 ± 0.8 g body weight) were distributed in a flow-through setup of three 400 L fiberglass tanks (n = 16 in each). Fish were anesthetized with 2-phenoxyethanol (0.5 mL L⁻¹ of 2-phenoxyethanol; Sigma P1126) before handling for injections. Fish were injected either intraperitoneally with 0.1 mL of Hank’s Balanced Salt Solution (HBSS) and were regarded as sham groups, or with 0.1 mL of the inflammatory agent poly I:C (50 μg mL⁻¹; Merck Chemicals P1038), a synthetic analog of double-stranded RNA that mimics a viral agent.

2.5. Sampling Procedures

All the fish from each experimental tank were slightly anesthetized with 2-phenoxyethanol, weighed, and measured at 0, 15, 30, 45, and 60 days of the experiment, and then returned to their respective tanks. These sampling points were used to adjust the feed ratio to maintain the initial 2% of their body mass during the experiment and record the biometric parameters for each tank and treatment. For each sampling point, the fish were fasted for 24 h before sampling and anesthetized with 2-phenoxyethanol (0.5 mL L⁻¹ of 2-phenoxyethanol; Sigma P1126) before handling.

At the end of the experiment (day 60), the fish were netted and euthanized with a lethal dose of 2-phenoxyethanol (1 mL L⁻¹ in seawater). All fish were weighed and measured, and six fish per tank were sampled (n = 18 per experimental diet). Blood was collected from caudal vessels with ammonium-heparinized syringes (Sigma-Aldrich H6279; 25,000 units in 3 mL of saline 0.6% NaCl). Plasma was separated from cells by centrifugation of whole blood (3 min, 10,000× g, 4 °C) and stored at –80 °C for further analysis. The fish were subsequently beheaded and opened. The livers were removed and weighted to determine the hepatosomatic index (HSI). Then, liver biopsy samples were frozen at −80 °C for analysis of metabolites.

Following injection with Poly I:C and HBSS, the fish were sampled for blood collection at 4 and 24 h after stimulation. Moreover, unstimulated individuals were also sampled on the day of injections and were regarded as time 0 h (control). For this challenge, parameters such as hematocrit, hemoglobin, as well as total red and white blood cell counts were taken in order to study the immune-modulatory effects of the experimental diets.

2.6. Growth Performance

Weight increase and feed consumption were used to calculate the following nutritional and biometrical indexes: weight gain (WG, g), weight gain (WG, %), condition factor (K, g cm³), specific growth rate (SGR), and hepatosomatic index (HSI, %). These indexes were calculated according to the following Formulas (1)–(5)

WG (g) = BMf (g) − BMi (g)

(1)

WG (%) = [(BMf (g) − BMi (g))/BMi (g)] × 100

(2)

K (g cm⁻³) = [100 × BM (g)/BL³ (cm)]

(3)

SGR (% day⁻¹) = 100 × [(ln BMf − ln BMi)/t (days)]

(4)

HSI (%) = (BL/BMf) × 100

(5)

where BL = fish body length; BM = fish body mass; BMf = final body mass; BMi = initial body mass.

2.7. Plasma and Hepatic Parameters

For the assessment of metabolite levels, previously frozen liver biopsies were finely minced on an ice-cold petri dish (~0.3 g) and subsequently homogenized by mechanical disruption (Ultra-Turrax, T25 basic, IKA^®-Werke, Stanger, Espelkamp, Germany) with 7.5 volumes (w/v) of ice-cooled 0.6 N perchloric acid, and neutralized after the addition of the same volume of 1 M KHCO₃. Before centrifugation, an aliquot of each homogenate was separated for the measurement of triglycerides (TAG). The remaining homogenates were then centrifuged (30 min, 13,000× g, 4 °C), and the supernatants were recovered in different aliquots and stored at −80 °C until used in a glucose–glycogen metabolite assay.

All plasma and hepatic metabolites were analyzed using specific commercial kits from Spinreact (Barcelona, Spain) (Glucose-HK: Ref. 1001200; Lactate: Ref. 1001330; Triglycerides: Ref. 1001311) adapted to 96-well microplates. Liver glycogen levels were assessed using the method from Keppler and Decker [32], in which the glucose obtained after glycogen breakdown (after subtracting free glucose levels) was determined using the commercial kit described above for glucose. All assays were run on an Automated Microplate Reader (PowerWave 340, BioTek Instrument Inc., Winooski, VT, USA), controlled by KCjunior™ software version 1.4. All standards and samples were measured in quadruplicate and duplicate, respectively.

2.8. Haematological Analyses

The determination of circulating total red blood cells (RBC) and white blood cells (WBC) was performed by diluting previously collected blood with heparinized HBSS (30 units mL⁻¹) at 1/200 and 1/20, respectively. Cells were counted using a Neubauer hemocytometer under a light microscope after 2 min from pipetting to let cells settle [33]. Hematocrit values were immediately determined as the percentage of total blood volume (%) after sampling by placing fresh blood in capillary glass tubes and then centrifuging them for 10 min at 10,000× g. Hematocrit readings were performed with the aid of a microhematocrit reader. Hemoglobin determination was performed using a specific commercial kit from Spinreact (Barcelona, Spain) (Hemoglobin, Ref. 1001230) following the manufacturer’s instructions and adapted to 96-well microplates.

Additionally, the mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), and mean corpuscular hemoglobin concentration (MCHC) were also calculated as follows (Equations (6)–(8)):

MCV (mm³) = (Ht/RBC) × 10

(6)

MCH (pg cell⁻¹) = (Hb/RBC) × 10

(7)

MCHC (g 100 mL⁻¹) = (Hb/Ht) × 100

(8)

where Ht = hematocrit; RBC = red blood cells count; Hb = hemoglobin.

2.9. Statistical Analyses

Results are presented as means ± standard error of the mean (SEM). Statistical comparison for all given results was performed using a one-way ANOVA. Previously, both normality and equal variance were confirmed by Shapiro–Wilk and D’Agostino–Pearson tests, respectively. In addition, for all data sets, outliers were identified by the ROUT method at Q = 1%. All one-way ANOVA analyses were followed by a Tukey’s post hoc test when significant differences were detected. Statistical analyses were performed using the GraphPad Prism^® (v.6.0b) software (GraphPad Software, La Jolla, CA, USA), and significance for all tests was set at p < 0.05.

3. Results

3.1. Antimicrobial Activity of CEO

CEO inhibited the bacterial growth of the microorganisms, both in saline and ethanolic solution (

Table 1. Inhibition of bacterial growth by CEO suspended in saline solution and dissolved in ethanol. The results were obtained in duplicate and expressed as the minimum inhibitory CEO concentration (MIC, mg mL⁻¹).

Extract	V. harveyi	A. hydrophila	P. damselae subsp. piscicida	V. fischeri
Saline solution	27.8	55.6	27.8	27.8
Ethanol	111.3	27.8	27.8	27.8

). In the tests with saline, CEO inhibited V. harveyi, P. damselae subsp. Piscicida, and V. fischeri at a lower concentration than required to inhibit the growth of A. hydrophila. In the case of CEO ethanol extract, the MIC was higher in V. harveyi with respect to the other bacteria assessed.

3.2. Acute Toxicity of CEO

Figure 1 shows the percentage of V. fischeri’s bioluminescence at different concentrations of CEO. The median effective concentration (EC₅₀) was 0.73 mg mL⁻¹, established by the interpolation of values fitted to the equation of a logarithmic regression line (y = −56.5 LN(x) + 39.022; R² = 0.94), and the maximal inhibition (EC₁₀₀) was obtained at 2.23 mg mL⁻¹.

3.3. Growth Performance

No significant mortality was recorded during the experimental procedure, and survival was above 95% in all tanks. Different growth performance parameters are shown in

Table 2. Biometrics, weight gain (WG, in g and %), specific growth rate (SGR), condition factor (K), and hepatosomatic index (HSI) in S. aurata individuals fed with different experimental diets (CTRL, CEO1, and CEO2) for 60 days. Data are presented as mean ± SEM (n = 9–54). Letters indicate significant differences among experimental diets (p < 0.05, one-way ANOVA followed by a Tukey’s post hoc analysis).

	CTRL	CEO1	CEO2
Initial Weight (g)	17.9 ± 0.3	18.0 ± 0.3	17.9 ± 0.3
Final Weight (g)	44.8 ± 0.9	44.5 ± 1.0	44.6 ± 0.8
Initial Length (cm)	10.4 ± 0.1	10.4 ± 0.1	10.3 ± 0.1
Final Length (cm)	13.3 ± 0.1	13.2 ± 0.1	13.2 ± 0.1
WG (g)	27.6 ± 6.8	25.9 ± 7.4	27.3 ± 6.2
WG (%)	163.3 ± 6.7	142.7 ± 7.1	165.5 ± 6.5
SGR (% day−1)	1.6 ± 0.0 ab	1.4 ± 0.0 a	1.6 ± 0.0 b
K (g cm−3)	19.1 ± 0.2	19.2 ± 0.1	19.4 ± 0.2
HSI (%)	1.3 ± 0.1	1.3 ± 0.1	1.2 ± 0.1

. All dietary treatments exhibited a 2.4-fold increase in their initial body weights after 60 days of the feeding trial. According to growth rates and feeding conditions, inhibition of around 20% of the total potential weight gain (WG in %) was observed in the CEO1 group as compared to the CTRL and CEO2 groups. Although the specific growth rate (SGR in % day⁻¹) of neither CEO1 nor CEO2 groups showed significant differences as compared to the CTRL group, the CEO2 group showed significantly higher SGR than the CEO1 group. No significant differences were detected, either in the condition factor (K) or the hepatosomatic index (HSI), while the experiment lasted.

3.4. Plasma and Liver Metabolites

Plasma glucose, triglycerides, and lactate values are shown in

Table 3. Plasma (mM) and liver (mmol/g wet weight) parameters in S. aurata individuals fed with different experimental diets (CTRL, CEO1, and CEO2) for 60 days. Data are presented as mean ± SEM (n = 9). Letters indicate significant differences among experimental diets (p < 0.05, one-way ANOVA followed by a Tukey’s post hoc analysis).

	CTRL	CEO1	CEO2
Plasma
Glucose (mM)	3.1 ± 0.2	3.6 ± 0.2	3.3 ± 0.2
Triglycerides (mM)	2.4 ± 0.1	2.1 ± 0.1	2.0 ± 0.2
Lactate (mM)	2.2 ± 0.3	1.4 ± 0.1	1.1 ± 0.1
Liver
Glucose (mmol g−1 w.w.)	36.3 ± 3.8	37.9 ± 4.5	41.2 ± 1.9
Glycogen (mmol g−1 w.w.)	285.4 ± 29.5 a	267.6 ± 20.5 ab	248.9 ± 20.6 b
Triglycerides (mmol g−1 w.w.)	7.2 ± 0.5	4.9 ± 1.0	5.1 ± 0.6

. The addition of CEO in experimental diets did not cause significant plasma glucose, triglycerides, or lactate changes. In the liver, the metabolic parameters assessed are shown in Table 3. Glycogen levels were significantly lower in the CEO2 group than in the CTRL group. No significant differences in hepatic free glucose nor triglycerides were detected among the experimental groups.

3.5. Hematological Parameters

The hematological profile is shown in

Table 4. Hematological parameters in S. aurata individuals fed with different experimental diets (CTRL, CEO1, and CEO2) for 60 days. Data are presented as mean ± SEM (n = 8–9). Letters indicate significant differences among experimental diets (p < 0.05, one-way ANOVA followed by a Tukey’s post hoc analysis).

	CTRL	CEO1	CEO2
WBC (×104 cell µL−1)	2.3 ± 0.1 a	2.5 ± 0.1 ab	3.0 ± 0.2 b
RBC (×106 cell µL−1)	1.6 ± 0.2 a	1.1 ± 0.1 b	1.5 ± 0.1 ab
Hematocrit (%)	32.2 ± 2.0	29.1 ± 1.3	28.3 ± 1.3
Hemoglobin (g dL−1)	3.1 ± 0.3	3.1 ± 0.2	3.2 ± 0.1
MCV (mm3)	209.9 ± 22.3 ab	256.6 ± 15.2 a	190.3 ± 10.0 b
MCH (pg cell−1)	21.0 ± 2.8	27.2 ± 4.4	22.1 ± 1.8
MCHC (g 100 mL−1)	9.7 ± 0.7	9.8 ± 1.5	11.5 ± 0.8

. Total WBC numbers increased significantly in those individuals fed with the highest CEO-supplemented diets (CEO2), while the CEO1 group had intermediate levels, in between the CTRL and CEO2 groups. In contrast, RBC decreased significantly in individuals fed with CEO1 when compared to the CTRL dietary treatment. Remarkably, the CEO2 group presented intermediate levels, in between the CTRL and CEO1 groups. Moreover, a significant reduction in the mean corpuscular volume (MCV) was detected in the CEO2 group as compared to the CEO1 group. The addition of CEO in experimental diets did not cause significant changes in hemoglobin, hematocrit, mean corpuscular hemoglobin (MCH), nor in the mean corpuscular hemoglobin concentration (MCHC).

3.6. Inflammatory Challenge after Feeding Trial

The in vivo experiment using the intraperitoneal administration of poly I:C in S. aurata juveniles previously fed with either CTRL or CEO2 was intended to study whether the CEO2 diet could modulate total WBC dynamics following an inflammatory stimulus. A significant progressive reduction in hematocrit was observed in the CTRL-fed individuals (0 h > 4 h > 24 h), injected either with sham or poly I:C (Figure 2A,B). No changes in this same parameter were observed for any of the CEO2-fed individuals. In contrast, the MCHC progressively increased just within CTRL-fed + poly I:C-injected individuals (Figure 2H). Note that this was not observed in any sham-injected group regardless of the diet (Figure 2G). Compared to CTRL-fed individuals, WBC were progressively reduced in CEO2-fed individuals (Figure 2D), both injected with poly I:C. In contrast, no similar responsive patterns were observed in WBC within sham-injected groups (Figure 2C). Additionally, MCV significantly increased in the CEO2-fed + poly I:C-injected individuals 24 h post-injection (Figure 2F), whereas no significant changes were observed in MCV within the sham-injected groups (Figure 2E).

4. Discussion

4.1. Antimicrobial Activity of CEO

The cytotoxicity and antibacterial effects of several herbs have been previously examined [34,35], highlighting their potential use for controlling bacterial diseases in cultured fish. In our study, CEO inhibited bacterial growth, both in saline and ethanolic solutions. The antimicrobial activity of C. nardus essential oils was first reported by Hammer et al. [36]. The CEO inhibited the growth of Acinetobacter baumanii, Candida albicans, Enterococcus faecalis, Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa, Salmonella typhimurium, Serratia marcescens, and Staphylococcus aureus. Recent studies have tested the antimicrobial activity of CEO against several fish pathogens [14,16]. The results achieved growth inhibition regardless of the bacteria being Gram-positive or Gram-negative. Specifically, MIC values ranged from 0.244 µg mL⁻¹ up to 0.977 µg mL⁻¹. Since MIC results were obtained using an aqueous solution instead of ethanol, the following experiments were performed using the CEO diluted in an aqueous solution.

4.2. Acute Toxicity of CEO

The measurement of acute toxicity by the inhibition of V. fischeri bioluminescence using Microtox^® is a well-established and standardized in vitro toxicity test, used in a variety of applications [37], including plant extract toxicity assessment [38,39,40]. Thus, the results obtained with this test can be extrapolated to the effect that the substances would have on animals. In our study, CEO showed an EC₅₀ of 0.73 mg mL⁻¹. Remarkably, this result is lower than those obtained by Skotti et al. [33] with different plant extracts, in which the toxicity (EC₅₀) of Melissa officinalis L., Origanum vulgare L., Origanum dictamnus L., Salvia officinalis L., and Hyssopus officinalis L. lay between 8.61 and 224.87 mg mL⁻¹. This implies that CEO should be used in fish diets at concentrations lower than those described for these plant extracts, which could have adverse effects. For example, oregano EO has been used in sea bass feed at a 200 ppm of feed concentration for 60 days without generating mortality or adverse effects [41]. Similarly, Ghafarifarsani et al. [42] showed that a 5% inclusion of garlic and oregano extract in carp feed produced no adverse effects. Moreover, EC₅₀ obtained for CEO in our study was higher (i.e., less toxic) than Greek herbs (e.g., Artemisia dracunculus, Achillea millefolium, Matricaria chamomilla, Salvia pomifera, S. officinalis, and S. fruticosa), which ranged from 0.005 up to 0.71 mg mL⁻¹ [40]. Thus, the toxic activity of CEO implies that its effect must be taken into consideration when administered in fish feed, which justifies the in vivo studies conducted in this work.

4.3. In Vivo Effects of CEO

Medicinal plants in different formats (e.g., extracts, EO, aqueous, alcoholic, etc.) are used in aquaculture as chemotherapeutics and feed additives [43]. Several physiological effects have been linked to medicinal plants, including growth performance, immune stimulation, and disease resistance against fish pathogens [44,45,46].

The present and previous studies have already shown the potential toxicity of Cymbopogon spp. EO in both aqueous and ethanolic solutions against microorganisms.

After validating CEO in the initial in vitro test to recognize antimicrobial properties, it was decided to move forward to an in vivo test throughout its supplementation of a commercial diet.

Although the addition of any concentration of CEO to a diet apparently did not interfere with S. aurata growth (CTRL vs. CEO1/CEO2), remarkably, the SGR was found significantly reduced in the CEO1 group than in the CEO2 group. Moreover, HSI was also not influenced by diets containing C. nardus EO. Al-Sagheer et al. [23] showed that the use of lemongrass (Cymbopogon citratus) EO at 200 or 400 mg kg⁻¹ of fish feed, respectively, improved growth performance, feed utilization, and the immunological parameters of Oreochromis niloticus. However, the employ of single or mixture EOs with no significant effects on the growth performance of S. aurata has also been previously reported [47,48,49].

Overall metabolism in S. aurata individuals fed with CEO dietary treatments remained unaffected, except for hepatic glycogen content, whose levels dropped in individuals fed with the highest dietary CEO level. Since those glycogen levels were in accordance with a normal range (200–300 mmol gw.w.⁻¹) observed in previous studies, and considering other harmful compounds that depleted the glycogen threshold content in the liver below 150 when supplied throughout the diet [50], it can be suggested that CEO at the levels tested in the present study appeared not to harm juvenile S. aurata energetic metabolism. Interestingly, a correlation between a drop in hepatic glycogen and the increase in WBC for the CEO2 group could be suggested since WBC have been shown to greatly augment the consumption of oxygen and sugar due to their intrinsic metabolism [51]. This evokes a prompt prophylactic reaction after 60 days of CEO2 feeding by an increase in leukocytes (since platelet counts were scarce) without significantly depleting the main metabolic tissue reservoirs.

The effect of diet on the immune system has been widely recognized in farmed animals, and previous studies have already reported improved total WBC in fish fed a number of dietary additives such as amino acids and medicinal herbs [52,53,54,55]. In fact, increasing attention has been paid to the use of plant-derived products as prophylactic strategies to control diseases in farmed fish, as already reviewed by Vallejos-Vidal et al. [56]. In the present study, total WBC augmented in a dose-dependent manner in S. aurata fed with dietary CEO. These data suggest a potential effect of CEO in stimulating cellular-mediated immunity. According to previous studies and review articles, an increase in circulating WBC numbers can translate into improved disease resistance against pathogens [55,56]. Therefore, there is potential to further explore the effects of dietary CEO under more challenging scenarios as a way to verify its putative immunomodulatory potential.

It is also remarkable that the RBC appeared to drop in levels at the intermediate CEO dose, which could be circumstantial and not necessarily linked to CEO ingestion. We also observed that MCV was significantly higher in the CEO1 group, whereas it decreased significantly in the CEO2 group. Usually, when MCV is high, the RBC are abnormally large, whereas a low MCV indicates that RBC are abnormally small. Interestingly, hemoglobin levels remained unaffected, whereas the MCH (mean corpuscular hemoglobin) appears to have a non-significant increase in the CEO1 group. This observation might indicate that changes evoked by dietary CEO1 involve an RBC volume increase correlated with an increase in the amount of hemoglobin that these cell types contain (as suggested by MCH). Note that changes in RBC volume might imply vesicle formation in the cytoplasm, but not necessarily changes in hemoglobin content [57], as indicated by our results. Therefore, if these biophysical changes in volume (increment) somehow disrupt oxygen transport in the blood, it could putatively slow down juveniles’ growth performance due to their more active daily metabolism, which could explain the significantly lower SGR observed within CEO1 as compared to the CEO2 group. However, we cannot rule out the possibility that some of the changes evoked by a potential immune response could be associated with changes in cell volume or plasma water content instead. Note that some works have addressed the relation between EO constituents (including the citronellal) and their effects on cardiovascular diseases [58].

In addition to studying the immunomodulatory effects of feeding S. aurata with CEO for 60 days, we also intended to verify the hematological response following an inflammatory insult by administering poly I:C. We observed that S. aurata fed with CEO2 supplementation seems to avoid some hematological responses only observed in CTRL-fed individuals, such as a progressive hematocrit reduction 24 h post-injection, as well as an increase in hemoglobin concentration specifically in poly I:C-injected specimens. In contrast, dietary CEO2 supplementation appears to evoke hematological and immune-related responses to the poly I:C injection, such as a progressive WBC reduction 24 h post-injection, and a significant increment in MCV, thus indicating an increase in RBC average volume. The decrease in peripheral WBC numbers is an important step of the innate immune response to different stimuli, and it has been observed as a result of WBC migration to the inflammatory site. For instance, Machado et al. [59] also observed a significant reduction in circulating WBC levels after stimulation with a phlogistic agent. Moreover, Engelsma et al. [60] described a similar WBC response for common carp (C. carpio L.) after acute stress.

In summary, the present results on the inclusion of CEO in the diet reaffirm this compound as a potential candidate in the prevention of pathological outbreaks due to its antimicrobial properties in vitro; this is also true for the prevention of potential pathogen infections in vivo, based on differential responsive patterns observed during the inflammatory poly I:C injection challenge. Moreover, this compound has been successfully included in a commercial fish feed with non-apparent adverse effects and concomitantly appeared to increase blood leukocyte cell population at the highest concentrations (2 mL of CEO kg⁻¹ fish feed) in just 60 days. At intermediate levels (CEO1), the specific growth rate of S. aurata juveniles seems to slow down. The lowest CEO concentrations might take longer to evoke the final physiological stage. Further research on supplying prophylactic fish feeds at different doses and for limited time windows should be considered in intensive aquaculture as an alternative to antibiotics or other more invasive treatments.

5. Conclusions

In conclusion, our results suggest that using citronella essential oil as a fish feed additive, at a rate between 1 and 2 mL per kg⁻¹ of fish feed, might prevent bacterial outbreaks and improve potential in vivo disease resistance in cultured S. aurata without negatively affecting growth.