Physiological responses of Rhamdia quelen ( Siluriformes : Heptapteridae ) to anesthesia with essential oils from two different chemotypes of Lippia alba

The aim of this study was to evaluate if Lippia alba has different chemotypes according to the chemical composition of the essential oil (EO) considering collection site, and if the EO may have different effects on blood and plasma parameters in silver catfish, Rhamdia quelen, during and immediately after anesthesia. The citral (EO-C) and linalool (EO-L) chemotypes were identified, and both presented similar anesthetic effects for silver catfish. Fish were exposed to two concentrations of each EO, which induced slow and fast anesthesia (100 and 300 μL L-1, respectively). Blood ions did not change at any time of anesthesia induction and recovery and, therefore, the electrolyte balance was not altered. Blood gases oscillated through all exposure and recovery times, but there was an increase in pO2 after 10 min recovery in fish anesthetized with EO-C. Glucose increased in fish exposed to both EOs when compared with the control group. Overall, exposure to both EOs (except 100 μL L-1 EO-L at most times) reduced plasma cortisol levels compared to the control and/or ethanol groups. However, as plasma creatinine levels in fish anesthetized with EO-C were higher than control fish, the use of EO-L is preferable.

Hematological and biochemical parameters of fish are valuable markers, since they can be used as indicators of physiological conditions, as well as in the control of diseases and stress manipulation (Aldrin et al., 1982;Tavares-Dias et al., 2008;Araújo et al., 2009).Plasma cortisol is one of the most used indicators to evaluate stress in fish (Wendelaar Bonga, 1997) and the two major actions of this hormone are the control of the ionoregulatory balance and energy metabolism (Liew et al., 2015).The electrolytic imbalance can be observed by changes in plasma or blood ions (McDonald, Milligan, 1997;Wendelaar Bonga, 1997;Takahashi et al., 2006).Glucose levels are also widely used as indicator of stress, hyperglycaemia being reported for several teleosts in this situation (Barton, Iwama, 1991).Stress also has an effect on other blood biochemical parameters such as levels of enzymes and substances with important metabolic functions, such as urea and creatinine, which indicate the overall health of the fish (Cnaani et al., 2004).
Physiological effects of the EO of L. alba cultivated in southern Brazil as anesthetic and sedative for silver catfish, Rhamdia quelen, was verified by many authors (Cunha et al., 2010;Heldwein et al., 2014;Toni et al., 2014;Salbego et al., 2014), but only the linalool chemotype.Therefore, it is of interest to investigate the anesthetic and physiological effects of EO obtained from other chemotypes of L. alba.Considering that a different chemotype of L. alba (myrcenecitral) cultivated in northern Brazil was identified by Oliveira et al. (2006), the aim of this study was to investigate a possible geographic effect in the EO composition (L.alba cultivated in northern and southern Brazil) and, if these EOs have different compositions, to evaluate their sedative and anesthetic effects in silver catfish, as well as their physiological effects on blood and plasma parameters.

Materials and Methods
Animals.One hundred sixty-eight juveniles silver catfish (Rhamdia quelen; 51.17 ± 1.69 g and 20.21 ± 1.40 cm) were obtained from a local fish farm and brought to the Fish Physiology Laboratory at the Universidade Federal de Santa Maria (UFSM).The species was identified at the Ichthyology Laboratory (Universidade Federal do Rio Grande do Sul) and a voucher specimen was deposited in this laboratory at number UFRGS 19612.The fish were maintained for one week in 250 L tanks (50 fish/tank) with continuous aeration; temperature 21 ± 2 °C; pH 6.5-7.5 and dissolved oxygen above 5.5 mg L -1 .The animals were fed once a day with commercial feed and kept fasted for a period of 24 h prior to the experiments.Botanical identification of L. alba linalool chemotype was made by Gilberto Dolejal Zanetti (Department of Industrial Pharmacy, UFSM) and a voucher specimen (SMDB 10050) was deposited in the herbarium of the Department of Biology (UFSM).The L. alba citral chemotype was identified by Dr. Fatima Salimena (Universidade Federal de Juiz de Fora) and a voucher was registered in the herbarium of this institution under number CESJ 65276.

e160083[3]
The essential oils were obtained by hydrodistillation of fresh leaves for 3h in a Clevenger apparatus (European Pharmacopoeia, 2007) and stored at -4 o C until utilization.The analysis of the EOs was performed by gas chromatographymass spectrometry-total ion chromatogram using an Agilent 6890 gas chromatograph coupled with an Agilent 5973 mass selective detector and employing a HP5-MS column (5% phenyl, 95% methylsiloxane, 30 m x 0.25 mm i.d.x 0.25 mm) as described by Silva et al. (2012).The constituents were identified by comparison of the Kovats retention index and their mass spectra with data from the mass spectral library (NIST, 2002) and the literature (Adams, 2001).
Experiment 1. Anesthetic induction and recovery times.Anesthesia induction and recovery were tested at concentrations of 25, 50, 100, 200, 300 µL L -1 for both EOs.Eight fish were used for each concentration tested, and each juvenile was used only once.Sedation was characterized by the decreased reactivity to external stimuli, and anesthesia by total loss of equilibrium and cessation of locomotion, according to Small (2003).The EOs were previously diluted in 95% ethanol (1:10).Ethanol at the highest concentration used does not have any anesthetic effect in silver catfish (Cunha et al., 2010).After induction, fish were transferred to anesthetic-free aquaria to measure anesthesia recovery time.The fish were considered to be recovered when they returned to normal swimming and reacted to external stimuli.
Experiment 2. Exposure to anesthetics for physiological evaluation.Silver catfish were individually placed in an 8 L aquarium containing one of the EOs at 100 µL L -1 for up to 5 min or 300 µL L -1 for up to 2 min.These concentrations led to sedation and deep anesthesia, respectively (Cunha et al., 2010).Afterwards, fish from all groups were transferred individually to 8 L aquaria with anesthetic-free water for up to 10 min for recovery.There were also groups subjected to 26700 µL L -1 ethanol (the concentration used to dilute the highest EO concentration) and water only (control), which were handled as outlined above.
Blood analysis.Blood was collected from the caudal vein of silver catfish in less than 30 s with heparinized syringes at 1, 2 and 5 min of exposure (groups exposed to the concentration of 300 µL L -1 were not assessed at 5 min) and 5 and 10 min of recovery (total of 30 fish per treatment, n = 6 for each EO, concentration and collection time, each fish was sampled only once).Control and ethanol exposed fish were held tightly for blood collection.An aliquot of this blood was used to measure Na + , K + , Ca 2 + , glucose, pH, partial pressures of O 2 (pO 2 ) and CO 2 (pCO 2 ) using the i-STAT ® portable clinical analyzer with CG8+ cartridges (Abbott Laboratories, Chicago, IL, USA).The sample temperature was corrected to match the experimental water temperature (Roth, Rotabakk, 2012).The use of i-STAT® and calculations for blood gases have been described for several fish species (Jacobs et al., 1993;Pidetcha et al., 2000;Harrenstien et al., 2005;Kristensen et al., 2010;Barbas et al., 2016).
Plasma analysis.The remaining blood collected was centrifuged (800 x g for 10 min) and the plasma was used for analysis of creatinine and urea using an automated Vitros 250 (Ortho -Clinical Diagnostics) and Johnson & Johnson dry chemistry kits.All tests were carried out in duplicate.
Plasma cortisol was also determined in duplicate using an enzyme-linked immunosorbent assay (ELISA) kit (Diagnostics Biochem Canada Inc., Canada).This analysis was previously validated (Souza et al., 2015).Absorbance was measured in a spectrophotometer at 450 nm, and intraand inter-assay coefficients of variation were 6.3% and 5.2%, respectively.
Statistical analysis.Data are reported as mean ± SE.The homogeneity of variances among groups was determined with the Levene test.All treatment groups were compared by two-way analysis of variance (time x concentration) and Tukey's test; or, when homogeneity of variances was not obtained, by the Scheirer-Ray-Hare extension of the Kruskal-Wallis test and the Nemenyi test.Analyses were performed using the STATISTICA software package, version 5.1 (StatSoft, Tulsa, OK, USA), and the minimum significance level was set at p < 0.05.

Results
Chemical composition of the essential oils.A total of 65 compounds were identified in the EO of L. alba collected in southern Brazil (EO-L) and 67 compounds in the EO of L. alba collected in northern Brazil (EO-C) (Tab.1).The major component in the EO-L was β-linalool (50.56%), while the major compounds in the EO-C were E-citral (29.84%) and Z-citral (24.41%).

Anesthetic induction and recovery times.
Both EO-C concentrations induced sedation faster than EO-L, but anesthesia was faster only at the lowest EO-C concentration.Fish anesthetized with EO-C took longer to recover than those anesthetized with EO-L (Fig. 1).Ethanol did not show any sedative or anesthetic effect.Blood analysis.Blood pH, Na + , K + and Ca 2+ did not differ between any of the treatments and overall means were: pH (7.32 ± 0.04), Na + (134.90 ± 0.89 mmol L -1 ), K + (2.89 ± 0.12 mmol L -1 ), Ca 2+ (1.18 ± 0.06 mmol L -1 ).
Overall, blood pO 2 levels of silver catfish anesthetized with both EOs and exposed to ethanol were lower than control fish, increasing after 10 min recovery (Figs.2a-b).In contrast, an increase in pCO 2 was observed for fish anesthetized with EO-C and no significant change was observed for silver catfish anesthetized with EO-L.At the end of 10 min of recovery fish anesthetized with EO-C still maintained blood PCO 2 levels higher than control fish (Figs.2c-d).Blood glucose levels increased in control and ethanol groups after 5 min, compared to initial values, returning to initial values at the end of the recovery period in the control group.Exposure to both EOs (except 300 µL L -1 EO-L) did not avoid this increase of blood glucose levels.
At the end of the recovery period, the blood glucose levels of silver catfish exposed to ethanol and both EOs (except 300 µL L -1 EO-L) remained higher than the initial values and higher than those of the control group (Figs.3a-b).
In the control group, plasma cortisol levels decreased after 2 min and remained low until the end of the recovery period.In fish exposed to ethanol, cortisol levels decreased up to 5 min after exposure and returned to the initial values after 5 min of recovery.Overall, exposure to both EOs (except 100 µL L -1 EO-L at most times) reduced plasma cortisol levels compared to the control and/or ethanol groups (Figs.3c-d).
Plasma creatinine values of control fish increased significantly after 5 min and remained high at the end of the recovery period.Fish exposed to ethanol and both EOs showed significantly higher creatinine levels 1 min after exposure compared to control fish and these levels returned to control values at the end of recovery only in those exposed to EO-L (Figs. 4a-b).Plasma urea levels in the control group remained constant at all evaluation times.Fish exposed to ethanol showed a significant increase in plasma urea levels when compared to control fish after 5 min of exposure and returned to control values at the end of the recovery period.Plasma urea was significantly higher with most anesthesia treatments and recovery times in fish exposed to both EOs when compared to the control and ethanol groups (Figs.4c-d).

Discussion
Since EOs represent a chemical interface between plant and the surrounding environment, their syntheses are often affected by environmental conditions, thus expressing the occurrence of chemotypes or chemical races in the producing plant species (Gobbo-Neto, Lopes, 2007).Although there are many examples of the occurrence of geographic variations of EOs chemical composition in several plants (Figueiredo et al., 2008), the distribution of chemotypes is often not locally limited.In some species, different chemotypes can grow side by side (Schmidt et al., 2004).
The present study demonstrates that the EO from L. alba cultivated by our group in southern Brazil has linalool as its main compound (50.56%), and so it belongs to chemotype linalool.On the other hand, L. alba collected in northern Brazil can be classified in the chemotype citral, once this is the major compound of its EO (54.26%).Some authors indicated that geographical distribution and exposure to different soil and weather conditions, season of collection can affect the chemical composition of L. alba EO (Pascual et al., 2001;Hennebelle et al., 2008;Maffei et al., 2011;Teles et al., 2012).However, specimens from the chemotypes citral, linalool and carvone, collected in different regions of Brazil, cultivated in similar conditions, maintained the same chemical composition, indicating that differences are due to genotypic variations (Tavares et al., 2005).
The EO-C anesthetized silver catfish within approximately 2 min at 300 µL L -1 , inducing anesthesia faster than EO-L.Anesthesia recovery was slower with EO-C, but it can be considered and adequate anesthetic for silver catfish, because an ideal anesthetic must induce anesthesia up to 3 min and enable the recovery in about 10 min (Park et al., 2008).The anesthetic effect of EO-L in silver catfish involves the modulation of the benzodiazepine (BDZ) site of the GABAergic system (Heldwein et al., 2012).The EO-C blocks the excitability of rat sciatic nerves (Sousa et al., 2015), but the anesthetic effect of the EO from Aloysia triphylla (which has citral as its major compound) in silver catfish is not related to a modulation of the BDZ site of the GABAA receptor (Santos et al., in press).
All parameters examined in this study are within the range previously observed for silver catfish (Barcellos et al., 2001;2004).Blood Na + , K + and Ca 2+ of silver catfish were not affected by anesthesia with either EO tested.Similar results were found in the blood of Amazon catfish (Leiarius marmoratus) anesthetized with 10-200 µL L -1 eugenol (Honorato et al., 2014) and in the plasma of silver catfish anesthetized with 150 and 300 mg L -1 MS-222 (Gressler et al., 2014).However, silver catfish anesthetized with 150-450 µL L -1 of Hesperozygis ringens and Lippia alba (EO-L) showed altered plasma Na + and K + between 30-240 min of recovery (Toni et al., 2014) and anesthesia of tambaqui with 20 mg L -1 jambu extract induced blood ionoregulatory changes 2 h after recovery from anesthesia and Na + levels did not return to control values, even after 72 h (Barbas et al., 2016).Apparently ionoregulatory effects of anesthesia in blood or plasma are significant only after a few hours of recovery, when they can be detected.
During fish anesthesia, opercular movement (respiration) generally decreases compared to conscious fish, explaining the lower blood pO 2 in silver catfish anesthetized with both EOs and the higher pCO 2 in those anesthetized with EO-C.Through anesthetic recovery from both EOs there was an increase in blood pO 2 levels and a reduction in pCO 2 levels in those exposed to EO-C, which is expected for the recovery period with normal return of opercular movements.These same oscillation patters in pO 2 and pCO 2 from fish anesthesia, were found for red "pacu" (Piaractus brachypomus) anesthetized with MS-222 (150 mg L -1 ) (Hanley et al., 2010) and "tambaqui" (Colossoma macropomum) anesthetized with waxy extract of "jambu" flowers (Spilanthes acmella) at 20 mg L -1 (Barbas et al., 2016).
An increase in plasma levels of glucocorticoids such as cortisol is one of the main responses to stress (Barton, 2002).Plasma cortisol increases significantly in juvenile R. quelen 5 -30 min after handling (Koakoski et al., 2012), but surprisingly, handling was not sufficient to raise the plasma cortisol in the control group of silver catfish in our study.Toni et al. (2015) observed no increase in plasma cortisol levels of silver catfish exposed for 6 h to 30 and 50 µL L -1 EO from Hesperozygis ringens and they proposed that the primary stress reaction only took place in the first minutes after contact with the EO, as observed in fish exposed to EO-L in the present study.
A study by Gesto et al. (2014) using stressed rainbow trout (Oncorhynchus mykiss) showed that when catecholamines were released in the blood no changes in cortisol levels were observed as glucose levels increased.As plasma cortisol levels did not increase significantly, we suppose that the increase of blood glucose in silver catfish observed in the present study might be due to catecholamine release.According to Morgan, Iwama (1997), an increase in blood glucose occurs in response to a stressor, in order to provide most of the energy demand to cope with this stress.Our results corroborate those obtained by Inoue et al. (2011) andHonorato et al. (2014), who observed that anesthesia with eugenol increased plasma glucose compared to the sham control in "tambaqui" (20 and 60 mg L -1 ) and Amazon catfish (10-200 µL L -1 ).Anesthesia with 20 mg L -1 "jambu" extract also increased blood glucose levels in "tambaqui" (Barbas et al., 2016).Several studies testing a variety of anesthetics on multiple fish species also demonstrated increased glycemia after anesthesia induction (Ortuno et al., 2002;Deriggi et al., 2006;Barbosa et al., 2007;Park et al., 2008).
An increase in plasma urea levels in silver catfish during anesthesia and recovery was observed for both EOs when compared to the control group.The same result was obtained for plasma creatinine levels, but these levels were much higher in silver catfish anesthetized with EO-C than in control fish, and these levels returned to control values after 10 min recovery in those exposed to EO-L.Anesthesia of goldfish (Carassius auratus) with 50 µL L -1 nanoencapsulated clove oil also increased serum urea levels (Gholipourkanani et al., 2015).Increases in urea and creatinine levels together are probably due to lesions caused in the kidney of fish (Das, Mukherjee, 2003).Studies conducted by Borges et al. (2007) showed an increase in urea and creatinine levels in the serum of silver catfish exposed to cypermethrin, suggesting that these analyses can be useful for early detection of intoxication in fish.However, as nitrogen compounds are excreted as ammonia mainly through the gills (Nawata et al., 2007), and time of exposure to the EOs was brief, the elevation of urea levels observed in both EO groups in our study may be due to changes in ammonia and creatinine gill excretion and not to renal lesions.
In summary, different chemotypes of L. alba were detected according to their place of cultivation.We suggest that the EO-L and EO-C can be safely used as anesthetics in silver catfish, because the alterations in most parameters returned to control values within 10 min.However, the EO obtained from different chemotypes of the L. alba presented different physiological responses in plasma creatinine and the use of EO-L is preferable because the high creatinine levels provoked by EO-C exposure.Additional studies with longer exposure and/or recovery times are necessary to improve our understanding of the effects of the EO of this chemotype on renal function.

Fig. 1 .
Fig. 1.Time required for silver catfish (Rhamdia quelen) anesthesia induction and recovery (n=8 for each concentration tested) using different concentrations of essential oils from the linalool (EO-L) and citral (EO-C) chemotypes of Lippia alba.Stages are defined according to Small (2003).Values are mean ± SEM.Different letters indicate difference between concentrations and EOs in the same anesthetic stage.Based on two-way ANOVA followed by the Tukey post hoc test (p < 0.05).

Fig. 2 .
Fig. 2. Partial pressures of a-b oxygen (PO 2 ) and c-d carbon dioxide (PCO 2 ) in silver catfish (n = 6) (Rhamdia quelen) submitted to different concentrations of essential oils from the linalool (EO-L) and citral (EO-C) chemotypes of Lippia alba.Values are mean ± SEM.Different letters indicate significant difference between times within the same EO concentration.* indicate significant difference from control, # indicate significant difference from ethanol.Two-way ANOVA and Tukey's test or Scheirer-Ray-Hare extension of the Kruskal-Wallis test and Nemenyi test.

Fig. 3
Fig. 3. a-b.Blood glucose (Glu) and c-d.plasma cortisol in silver catfish (Rhamdia quelen) (n = 6) subjected to different concentrations of essential oils from the linalool (EO-L) and citral (EO-C) chemotypes of Lippia alba.Values are mean ± SEM.Different letters indicate significant differences between times within the same treatment.* indicates significant difference from control, # indicates significant difference from ethanol.Two-way ANOVA and Tukey's test or Scheirer-Ray-Hare extension of the Kruskal-Wallis test and Nemenyi test were used to determine statistical significance.

Fig. 4
Fig. 4. a-b.Plasma urea and c-d creatinine in silver catfish (Rhamdia quelen) (n = 6) subjected to different concentrations of essential oils from the linalool (EO-L) and citral (EO-C) chemotypes of Lippia alba.Values are mean ± SEM.Different letters indicate significant differences between times within the same treatment.* indicates significant difference from control, # indicates significant difference from ethanol.Two-way ANOVA and Tukey's test or Scheirer-Ray-Hare extension of the Kruskal-Wallis test and Nemenyi test were used to determine statistical significance.