https://orcid.org/0000-0002-9804-4188
Figures
Abstract
Extracts made from the skin of dead Lithodytes lineatus frog individuals with the application of the benzocaine-based anesthetic gel, introduced into the oral cavity, were analyzed by 1H Nuclear Magnetic Resonance to investigate whether the application of this product (oral) can make studies that use extracts from the skins of these animals unfeasible. For comparison, we used skins of another species of anuran following the same death protocol. No trace of the benzocaine substance was found in the 1H-NMR spectra of the skin extracts from any of the tested anuran species. Still, using the hierarchical clustering model, it was possible to observe the formation of well-defined groups between the skin extracts of anurans and the anesthetic used to kill these animals. Our results suggest that the lethal dose of benzocaine in gel used inside the mouth of frogs may have no influence on potential results regarding the chemical composition or even bioassays using extracts made from the skin of these animals killed under this protocol since there was no detection of this substance for the analyzed samples.
Citation: Barros AdL, Lima AP, Fachin-Espinar MT, Nunez CV (2020) Evaluation of benzocaine-based anesthetic gel in anuran skins extracts: A case study using the frog Lithodytes lineatus (Anura: Leptodactylidae). PLoS ONE 15(12): e0243654. https://doi.org/10.1371/journal.pone.0243654
Editor: Benito Soto-Blanco, Universidade Federal de Minas Gerais, BRAZIL
Received: June 19, 2020; Accepted: November 20, 2020; Published: December 8, 2020
Copyright: © 2020 Barros et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: YES - Specify the role(s) played. The authors thanks to the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) granted the doctoral scholarship to ALB and financial support (Pro-Amazônia/CAPES - 23038.000738/2013-78) and the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) provide financial support (CT-Amazônia/CNPq - 405804/2013-0) and CVN fellowships (309654/2016-5 and 305177/2019-2) and to FAPEAM for financial support.
Competing interests: The authors have declared that no competing interests exist.
1. Introduction
Studies involving analysis of the chemical composition of the cutaneous secretion of several species of anurans have been widely carried out, because it is possible to find a large number of substances with bioactive properties in the skin of these animals that present, mainly, great antimicrobial potentials [1–6]. There are several methods to extract the substances present in the body of frogs. One of the most used is the preparation of extracts made from the integument [7–9]. The method using skin extracts requires greater care during treatment in order to avoid indirect contamination of the samples during handling.
Currently, there are several debates about the use of some types of death protocols for anurans, such as, for example, cooling and freezing [10–12] since the perception of pain in these animals is still not well understood [13]. Thus, the use of anesthetics as a “humane” way to kill individuals in this taxonomic group has been recommended [7, 12, 14–17]. Benzocaine-based anesthetics are highly effective for both, anesthesia and death in amphibians, not requiring a large amount of the product to be able to kill them [12, 18–21]. However, recently, Saporito and Grant [22] found traces of the benzocaine substance coming from the Orajel© (liquid) product in skin extracts made from individuals of the species Melanophryniscus moreirae and Lithobates clamitans killed with the oral application of this anesthetic and concluded that the application directly in the mouth in certain species of anurans, may invalidate potential studies on the chemical composition of the extracts of these animals, signalling false positives, such as inaccurate detection of substances and/or incorrect information about potential biological activities. The authors gave as an example the study by Amézquita et al. [23] who, through experiments, suggested that some populations of the frog Allobates femoralis showed higher toxicity in the extracts when used in mice and was considered by the authors to be from alkaloids present in the integument of this species. As A. femoralis belongs to a frog family that has no representative known for producing or sequestering diet alkaloids [24–28], this result was contested by Saporito and Grant [22] who described that the possible toxicity in A. femoralis is due to the presence of benzocaine substance that was converted to the animals' skin, detecting the presence of this substance experimentally using mass spectrometry (MS).
Studies showing the detection of substances used to kill anurans present in skin extracts of these animals are still scarce, but they are of great value since many experiments are conducted using different death protocols [12] and also, using anesthetics [7, 13–16].
In this study, we test whether the benzocaine used to kill individuals of the frog Lithodytes lineatus is transferred to the skin extracts of these animals.
2. Material and methods
2.1. Obtaining of the Lithodytes lineatus skin extracts and of the Benzotop© product
Individuals of L. lineatus collected in three locations in the Brazilian Amazon (Cruzeiro do Sul [n = 3], Chandless State Park [n = 5] and Rio Branco [n = 3], all in the state of Acre), were killed using 40 mg of benzocaine-based anesthetic gel (Benzotop© gel, DFL Indústria e Comércio S.A.) with direct application in the mouth to avoid contamination of the skins. Immediately after the deaths were confirmed the animals’ skins were removed through the inguinal incision from one end of the body to the other and sampled, by location, in microtubes containing methanol (100%) and prepared for extraction. Two substance extraction systems (10 mL / g of skin) were used based on differences in polarities between the solvents: 1) dichloromethane + methanol at a concentration 9:1 (v/v), and 2) methanol (100%). The use of different extracting solvents was made to be able to access different substances, potentially including, the benzocaine substance. Subsequently, in each extractor system, the skins were taken to an ultrasound bath (Unique©, model USC—1800, frequency 40Hz) for 20 minutes, and subsequently filtered on filter paper. This process was repeated three times, and the concentrates for each extraction, by location, packed in a single bottle. The extracts obtained were taken to the fume hood for total evaporation of the solvents.
Still, for comparison, we used skins of another species of anuran, Adelphobates galactonotus (n = 3), killed with the oral anesthetic application (based on benzocaine), however, disregarding the standardized amount (higher or lower) of the product used in L. lineatus individuals. The A. galactonotus skins used in this study were stored in the tissue bank of the Ecology Laboratory of the National Institute of Amazonian Research (INPA) and were used separately to assess whether there is an individual effect and / or the conservation time on the results. For analyzes that included this species of anuran, only methanolic extracts were used.
A sample containing 10 mg of the Benzotop© gel product flavoured Pina Colada was solubilized using dichloromethane (DCM) and left to dry in a fume hood for total solvent evaporation. Subsequently, 10 mg of the methanolic extracts from the skins of individuals of L. lineatus and A. galactonotus, as well as the diluted sample of Benzotop©, were solubilized in deuterated chloroform (CDCl3) containing tetramethylsilane (TMS) as a reference solvent to be analyzed by the method Hydrogen Nuclear Magnetic Resonance spectroscopic (1H-NMR / NMR: Bruker, model Fourier 300, magnet 300 SB UltraShieldTM, 7.05T, 300 MHz).
The search for the 1H-NMR spectrum of the benzocaine substance (standard) was performed in the SciFinder repository database (Fig 1). As benzocaine is a well-known substance, there was no difficulty in finding an available spectrum. Subsequently, the spectra referring to the tested samples were compared to the standard. All samples were solubilized in chloroform for 1H-NMR analysis and tetramethylsilane (TMS) was used as the reference solvent.
Withdrawal from and available at: https://scifinder.cas.org.
Mass spectrometry analysis–MS/MS, ESI+, 50 to 210 m/z amplification, positive mode (M+H)–was used to test the presence or absence of benzocaine substance in L. lineatus skin samples. The presence of benzocaine in the analysed samples is confirmed by detecting mass fragments of 166 m/z (benzocaine mass at positive mode) according to SciFinder repository database.
2.2. Chemometric analysis
The 1H-NMR spectra of extracts from frog’ skins and the Benzotop© product had both phase distortions and baselines adjusted in the TopSpin program (version 4.0.7, Bruker Biospin) where they were automatically converted to CSV files (Comma-separated values). Subsequently, the CSV files were analysed using the R Studio program (R Studio, version 3.3 [29]). We evaluated potential differences or similarities between the samples tested, selecting specific regions in the 1H-NMR spectra to be excluded from the analyses, to perform a comparison based on the chemical shifts of interest, and eliminate potential noise from the sample (Table 1). In this analysis, only the regions corresponding to the chemical shifts of benzocaine were maintained. In total, five regions were chosen for exclusion.
To highlight potential similarities or divergences between the different samples analysed, we used a Hierarchical Cluster Analysis (HCA) to assess whether there is the formation of groups between samples from Euclidean distances, the results of which are illustrated in a dendrogram. The analyzes were performed using the software R Studio.
2.3. Ethical approval
The National Institute of Amazonian Research approved the experiments with the frog species used in this study and the permissions for the use of animals were granted by the Ethics Committee for Research in the Use of Animals (CEUA) (INPA / CEUA, Protocol: 030 / 2014). The permissions for the animal collection were granted by the Chico Mendes Institute for Biodiversity Conservation (ICMBio / license number 57123–1). All experiments were carried out following the relevant guidelines and regulations.
3. Results
The signs referring to the benzocaine substance were detected in the sample made with the Benzotop© anesthetic and are formed by regions with signs between I) 1.33 and 1.37 ppm; II) 4.27 and 4.37 ppm; III) 6.61 and 6.66 ppm, and IV) 7.83 and 7.87 ppm (Fig 2). The other signals detected in the spectrum come from other substances that are part of the general composition of the product, besides the solvent signal.
Gray rectangles describe the signals belonging to the substance benzocaine. Above, enlargements of the signals from the highlighted regions.
3.1. Comparative analysis between the 1H-NMR spectra of individuals of Lithodytes lineatus and the Benzotop© product
The methanolic (MET) and dichloromethanic (DCM) extracts made with the skins of individuals of Lithodytes lineatus were compared with the dichloromethanic extract of the Benzotop© (BEN) product to verify the presence or absence of characteristic signs of the benzocaine substance (Fig 3).
In the spectrum: BEN corresponds to the extract of the Benzotop© product; DCM corresponds to the dichloromethanic extracts of L. lineatus; MET corresponds to the methanolic extracts of L. lineatus.
Through the analysis of the 1H-NMR spectra, it was possible to observe that MET presents a greater chemical complexity concerning DCM, clear by the superior amount of signals present in the spectra. Also, it was found that the benzocaine substance was not incorporated into the general composition of the skin extracts of L. lineatus individuals (Fig 4) killed with BEN, in neither of the two extractions, since the signs and consequently the characteristic chemical shifts of the substance in question were not detected.
In the spectrum: BEN corresponds to the extract of the Benzotop© product; DCM corresponds to the dichloromethanic extracts of L. lineatus; MET corresponds to the methanolic extracts of L. lineatus. Gray rectangles describe the signals belonging to the benzocaine substance.
To validate the experiment, a comparison was made between BEN with extracts from the skin of another frog, Adelphobates galactonotus (AG), and the methanolic extracts from L. lineatus (LL) (Fig 5). The use of only L. lineatus methanolic extracts was due to the complexity of the samples compared to the dichloromethanic extracts, previously observed in the 1H-NMR spectra (Fig 4).
In the spectrum: BEN corresponds to the extract of the product Benzotop©; LL corresponds to L. lineatus extracts; AG corresponds to the extracts of Adelphobates galactonotus.
There were also no signs of chemical shifts characteristic of the substance benzocaine in the 1H-NMR spectra of AG skin extracts. Thus, we found that there was no conversion of benzocaine to the skin extracts of either of the two tested species (Fig 6).
In the spectrum: BEN corresponds to the extract of the Benzotop© product; AG corresponds to the extracts of A. galactonotus; and LL corresponds to extracts of L. lineatus. Gray rectangles describe the signals belonging to the benzocaine substance.
3.2. Hierarchical Cluster Analysis (HCA)
The clusters analysis associated the samples by the similarity between the regions of interest (characteristic signs of the benzocaine substance). The formation of two well-defined clusters between the samples was evidenced, one formed by the skin extracts of Lithodytes lineatus (LL) individuals, in both extractions, and the other formed by Benzotop© (BEN) (Fig 7).
In the image: BEN = corresponds to the extract of Benzotop©; DCM = correspond to the dichloromethanic extracts of L. lineatus; MET = correspond to the methanolic extracts of L. lineatus.
Still, another analysis of clusters was made by adding AG and again showed that BEN formed a separate group in comparison to both samples, AG and LL (Fig 8). These results reinforce that one found in the analysis of the 1H-NMR spectra, where the presence of the benzocaine substance was not detected in any of the evaluated samples.
In the image: BEN = corresponds to the Benzotop© extract; LL = correspond to the L. lineatus extracts; AG = correspond to the A. galactonotus extracts.
As well as verified by 1H NMR and Cluster Dendrogram analysis, benzocaine was not detected in L. lineatus skin extract samples by using mass spectroscopy analysis (Fig 9).
Gray rectangles describe the region belonging to the benzocaine substance mass (166 m/z).
4. Discussion
The use of benzocaine-based anesthetics is considered a common practice and can be used for several purposes within herpetology. Kaiser and Green [30] observed that very low amounts of Orajel© and Anbesol© anesthetics applied to the heads of several anuran species cause temporary anesthesia, drastically reducing movement and can be used as a non-lethal method to facilitate “photographic tests” of these animals in field and laboratory. The use as an anuran death protocol is highly effective concerning lethality even when compared to other anesthetics [20, 21, 31, 32] besides to being considered a low-cost product and without restrictions on the purchase, facilitating access [12, 18]. On the other hand, the use of anesthetics to kill frogs that will subsequently have their skins removed for the preparation of extracts used in biological tests or investigation of the chemical composition present in the integument can be a problematic factor due to the potential contamination [22].
Some authors have used other anuran death protocols to avoid bias that can be caused by the use of anesthetics, such as, for example, cooling and freezing [9–11]. Despite eliminating the risk of skin contamination either by contact with the product or by the potential conversion of the anesthetic to the integument, the “cool and freeze” death protocol has generated many controversies in the scientific community, being considered by the American Association of Veterinary Medicine [12], as not humane and unacceptable.
The result of our study was contrary to that described by Saporito and Grant [22], in which large amounts of the benzocaine substance, of the product Orajel© liquid, were found in the skins extracts of the Melanophrynsicus moreirae and Lithobates clamitans species. We found that in none of the extracts analyzed by 1H-NMR, of both species of anurans, the presence of the benzocaine substance from the anesthetic Benzotop© in gel was detected, including in the skins (Adelphobates galactonotus) that were stored in tissue banks, whose lethal dosage used was indiscriminate. This allows us to infer that the use of Benzotop© in gel introduced inside the mouth (lethal dose) for anurans does not prejudice possible studies regarding the chemical composition and/or biological activity of the extracts since the conversion to the skin was not evidenced.
We also suggest, based on the study by Saporito and Grant [22], that the use of benzocaine in liquid form should be avoided for studies involving the use of anuran skins extracts in bioassays since there is a possibility of signalling false positives; besides allowing inaccuracies as to the general chemical composition present in the extracts (mainly in the case of the use of crude extracts and not of isolated substances). We reinforce that other substances detection techniques (such as NMR and chemometrics) are important tools for better robustness and understanding of the results in these types of studies and that there is a need to investigate other parameters involving the interaction between the use of anesthetics and anurans, for example, whether sex, size, and age of these animals can influence the speed of conversion of these products to the integument.
5. Conclusion
Our results showed that the lethal dose of benzocaine in gel (Benzotop© product) used inside the frogs' mouth was not converted to the integument. The benzocaine substance was not found in the extracts of both species of anurans evaluated, as evidenced by the 1H-NMR spectra. Also, in the Hierarchical Analysis of Clusters (HCA) it was possible to confirm the absence of benzocaine in the samples of the frogs' skin extracts.
Supporting information
S1 File. AG1_FID_CSV file: CSV file containing chemical shifts of the skin extract (DCM + MeOH) of Adelphobates galactonotus (specimen 01).
https://doi.org/10.1371/journal.pone.0243654.s001
(CSV)
S2 File. AG2_FID_CSV file: CSV file containing chemical shifts of the skin extract (DCM + MeOH) of Adelphobates galactonotus (specimen 02).
https://doi.org/10.1371/journal.pone.0243654.s002
(CSV)
S3 File. AG3_FID_CSV file: CSV file containing chemical shifts of the skin extract (DCM + MeOH) of Adelphobates galactonotus (specimen 03).
https://doi.org/10.1371/journal.pone.0243654.s003
(CSV)
S4 File. LL1_DCM+MeOH_FID_CH_CSV file: CSV file containing chemical shifts of the dichloromethanic skin extracts from the frog Lithodytes lineatus.
In the caption, CH corresponds to individuals collected at Chandless State Park.
https://doi.org/10.1371/journal.pone.0243654.s004
(CSV)
S5 File. LL2_DCM+MeOH_FID_CR_CSV file: CSV file containing chemical shifts of the dichloromethanic skin extracts from the frog Lithodytes lineatus.
In the caption, CR corresponds to individuals collected in Cruzeiro do Sul municipality.
https://doi.org/10.1371/journal.pone.0243654.s005
(CSV)
S6 File. LL3_DCM+MeOH_FID_UFAC_CSV file: CSV file containing chemical shifts of the dichloromethanic skin extracts from the frog Lithodytes lineatus.
In the caption, UFAC corresponds to individuals collected at the Federal University of Acre.
https://doi.org/10.1371/journal.pone.0243654.s006
(CSV)
S7 File. LL1_MeOH_FID_CH_CSV file: CSV file containing chemical shifts of the methanolic skin extracts from the frog Lithodytes lineatus.
In the caption, CH corresponds to individuals collected at Chandless State Park.
https://doi.org/10.1371/journal.pone.0243654.s007
(CSV)
S8 File. LL2_MeOH_FID_CR_CSV file: CSV file containing chemical shifts of the methanolic skin extracts from the frog Lithodytes lineatus.
In the caption, CR corresponds to individuals collected in Cruzeiro do Sul municipality.
https://doi.org/10.1371/journal.pone.0243654.s008
(CSV)
S9 File. LL3_MeOH_FID_UFAC_CSV file: CSV file containing chemical shifts of the methanolic skin extracts from the frog Lithodytes lineatus.
In the caption, UFAC corresponds to individuals collected at the Federal University of Acre.
https://doi.org/10.1371/journal.pone.0243654.s009
(CSV)
Acknowledgments
The authors thanks to the Instituto Nacional de Pesquisas da Amazônia (INPA) for provided laboratory infrastructure. Dr. William Magnusson helped with the English and made revisions and suggestions to the manuscript. Waldir (Dinho) Heinrichs and Jonas R. Gonçalves helped with the English.
References
- 1. Bevins CL, Zasloff M. Peptides from frog skin. Annu Rev. 1990; 59: 395–414. pmid:2197979
- 2. Daly JW. The chemistry of poison in amphibian skin. Proc Natl Acad Sci. 1995; 92: 9–13. pmid:7816854
- 3. Goraya J, Wang Y, Li Z, O’Flaherty M, Knoop FC, Platz JE, et al. Peptides with antimicrobial activity from four different families isolated from the skins of the North American frogs Rana luteiventris, Rana berlandieri and Rana pipiens. Eur J Biochem. 2000; 267: 894–900. pmid:10651828
- 4. Kawasaki H, Issacson T, Iwamuro S, Conlon M. A protein with antimicrobial activity in the skin of Schlegel´s green tree frog Rhacophorus schlegelii (Rhacophoridae) identified as histone H2B. Biochem Biophys Res Communications. 2003; 312, 1082–1086.
- 5. Nascimento ACC, Fontes W, Sebben A, Castro MS. Antimicrobial peptides from anurans skin secretions. Protein Peptide Lett. 2003; 10, 227–238. pmid:12871141
- 6. Shen W, Chen Y, Yao H, Du C, Luan N, Yan X. A novel defensin-like antimicrobial peptide from the skin secretions of the tree frog, Theloderma kwangsiensis. Gene. 2016; 576, 136–140. pmid:26456194
- 7.
Clark VC. Collecting arthropod and amphibian secretions for chemical analyses, in: Zhang W, Liu H. (Eds.), Behav Chem Ecol. Nova Science Publishers, Inc., U.K. Edition, pp. 1–46; 2009.
- 8. Mebs D, Yotsu-Yamashita M, Pogoda W, Alvarez JV, Ernst R, Köhler G, et al. Lack of alkaloids and tetrodotoxin in the neotropical frogs Allobates spp. (Aromobatidae) and Silverstoneia flotator (Dendrobatidae). Toxicon. 2018; 152: 103–105. pmid:30081062
- 9. Jeckel AM, Kocheff S, Saporito RA, Grant T. Geographically separated orange and blue populations of the Amazonian poison frog Adelphobates galactonotus (Anura, Dendrobatidae) do not differ in alkaloid composition or palatability. Chemoecology. 2019; 29: 225–234.
- 10. Shine R, Amiel J, Munn AJ, Stewart M, Vyssotski AL, Lesku JA. Is “cooling then freezing” a humane way to kill amphibians and reptiles? Biol Open. 2015; 00: 1–4.
- 11. Lillywhite HB, Shine R, Jacobson E, Denardo DF, Gordon MS, Navas CA, et al. Anesthesia and Euthanasia of Amphibians and Reptiles Used in Scientific Research: Should Hypothermia and Freezing Be Prohibited? BioScience. 2017; 67: 53–61.
- 12.
AVMA. Guidelines for the Euthanasia of Animals: 2020 Edition. American Veterinary Medical Association. 2020; 121 pp.
- 13. Guénette SA, Giroux MC, Vachon P. Pain perception and anesthesia in research frogs. J Exp Anim. 2013; 62, 87–92. pmid:23615302
- 14. Letcher J. Intracelomic use of tricaine methanesulfonate for anesthesia of bullfrogs (Rana catesbeiana) and leopard frogs (Rana pipiens). Zoo Biology. 1992; 11, 243–251.
- 15. Lafortune M, Mitchell MA, Smith JA. Evaluation of medetomidine, clove oil and propofol for anesthesia of leopard frog, Rana pipiens. J Herpetol Med Surg. 2001; 11, 13–18.
- 16. Mitchell MA. Anesthesic considerations for amphibians. Topics in Med Surg. 2009; 18, 40–49.
- 17. Goulet F, Hélie P, Vachon P. Eugenol Anesthesia in African Clawed Frogs (Xenopus laevis) of Different Body Weights. J Am Assoc Lab Anim Sci. 2010; 49(4): 460–463. pmid:20819393
- 18. Chen MH, Combs CA. A alternative anesthesia for amphibians: ventral application of benzocaine. Herpetol Rev. 1999; 30:34.
- 19. Brown HHK, Tyler HK, Mousseau TA. Orajel® as an Amphibian Anesthetic: Refining the Technique. Herpetol Rev. 2004; 35:3.
- 20. Crook AC, Whiteman HH. An Evaluation of MS-222 and Benzocaine as Anesthetics for Metamorphic and Paedomorphic Tiger Salamanders (Ambystoma tigrinum nebulosum). Am Midl Nat. 2006; 155: 417–421.
- 21. Cecala KK, Price SJ, Dorcas ME. A comparison of the effectiveness of recommended doses of MS-222 (tricaine methanesulfonate) and Orajel® (benzocaine) for amphibiam anesthesia. Herpetol Rev. 2007; 38(1): 63–66.
- 22. Saporito RA, Grant T. Comment on Amézquita et al. (2017): “Conspicuousness, color resemblance, and toxicity in geographically diverging mimicry: The pan-Amazonian frog Allobates femoralis”. Evolution. 2018; 72, 1009–1014. pmid:29524217
- 23. Amézquita A, Ramos O, Gonzales MC, Rodriguez C, Medina I, Simoes PI, et al. Conspicuousness, color resemblance, and toxicity in geographically diverging mimicry: the pan-Amazonian frog Allobates femoralis. Evolution. 2017; 71, 1039–1050. pmid:28067425
- 24. Daly JW, Garrafo HM, Spande TF, Jaramillo C, Rand S. Dietary source for skin alkaloids of poison frogs (Dendrobatidae)? J Chem Ecol.1994; 20: 4. pmid:24242207
- 25. Smith BP, Tyler MJ, Kaneko T, Garraffo HM, Spande TF, Daly JW. Evidence for biosynthesis of pseudophrynamine alkaloids by an australian myobatrachid frog (Pseudophryne) and for sequestration of dietary pumiliotoxins. J Nat Prod. 2002; 65: 439–447. pmid:11975476
- 26. Darst CR, Menéndez-Guerrero PA, Coloma LA, Cannatella DC. Evolution of dietary specialization and chemical defense in poison frogs (Dendrobatidae): a comparative analysis. Am Nat. 2005; 165:1. pmid:15729636
- 27. Saporito RA, Spande TF, Garraffo HM, Donnelly MA. Arthropod alkaloids in poison frogs: a review of the ‘‘dietary hypothesis”. Heterocycles. 2009; 79:277–297.
- 28. Saporito RA, Donnelly M, Spande TF, Garraffo HM. A review of chemical ecology in poison frogs. Chemoecology. 2012; 22: 159–168.
- 29.
R Core Team. R: A language and environment for statistical computing. Version 3.3.3. [software]. 2017. R Foundation for Statistical Computing, Vienna, Austria. Available from: https://www.R-project.org/.
- 30. Kaiser H, Green DM. Keeping the Frogs Still: Orajel® is a Safe Anesthetic in Amphibian Photography. Herpetol Rev. 2001; 32(2): 93–94.
- 31. Guénette SA, Lair S. Anesthesia of the Leopard Frog, Rana pipiens: A Comparative Study between four Different Agents. J Herpetol Med Surg. 2006; 16(2): 38–44.
- 32. Guénette SA, Hélie P, Beaudry F, Vachon P. Eugenol for anesthesia of African clawed frogs (Xenopus laevis). Vet Anaesth Analg. 2007; 34: 164–170. pmid:17444929