|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 EFFECTS OF NITROGREN AMMONIA AND MS-222 ON XENOPUS LAEVIS DEVELOPMENT AND FORAGING BEHAVIOR By J. KELLY BYRAM A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2008
2 2008 J. Kelly Byram
3 To John, Nathaniel, and Silvana
4 ACKNOWLEDGMENTS I thank the Cryptobranchid Interest Group, whose grant helped to fund th is research. I also thank the Dial Center for Written and Oral Comm unication for the assistantship that defrayed many of my educational costs. I thank the Reptile and Amphibi an Conservation Corps for its support. Special thanks go to my committee chair, Max A. Nickerson, and my committee members, Peter Frederick and Lou Guillette, fo r their guidance and support through the long and arduous process. This work was done with IACUC approval (p roject E976) and with the guidance and oversight of Drs. August Battles and Harvey Rami rez of Animal Care Services. This research could not have been completed without Sam Jones of the Department of Wildlife Ecology and Conservation. Thank you, Sam. Most importantly, I thank my family for their tolerance of my love of cold-blooded things that go bump in the night.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4 LIST OF TABLES ................................................................................................................ ...........6 LIST OF FIGURES ............................................................................................................... ..........7 ABSTRACT ...................................................................................................................... ...............8 CHAPTER 1 EFFECTS OF NITROGEN AMMONIA ON Xenopus laevis DEVELOPMENT ...................9 Introduction .................................................................................................................. .............9 Materials and Methods ......................................................................................................... ..11 Statistical Analyses .......................................................................................................... 14 Results ....................................................................................................................... ..............15 Discussion .................................................................................................................... ...........17 2 EFFECTS OF NITROGEN AMMONI A AND MS-222 ON DEVELOPMENT AND FORAGING BEHAVIOR IN Xenopus laevis ........................................................................27 Introduction .................................................................................................................. ...........27 Materials and Methods ......................................................................................................... ..28 Statistical Analyses .......................................................................................................... 30 Results ....................................................................................................................... ..............31 Discussion .................................................................................................................... ...........33 LIST OF REFERENCES ............................................................................................................ ...44 BIOGRAPHICAL SKETCH .........................................................................................................48
6 LIST OF TABLES Table page 1-1 Summary of mean ( SE) Nieuwkoop-Fabe r stage (tadpoles), total length (mm) (tadpoles), weight (g), and SVL (mm) for treatment (N) and control groups. ..................20 2-1 Summary of sample sizes for days 130131 for the nitrogen ammonia treatment (N) and control groups. ........................................................................................................... ..39
7 LIST OF FIGURES Figure page 1-1 Nieuwkoop-Faber stage by study day ................................................................................22 1-2 Total length by study day ................................................................................................. ..23 1-3 Total weight by study day ..................................................................................................23 1-4 SVL by study day...............................................................................................................24 1-5 Comparison of variance in treatment and control groups ..................................................26 2-1 Summary of the research design. ...39 2-2 Number of food pellets (kibbles) ea ten in 1 hour is a constrained value ...........................40 2-3 MS-222 affects foraging behavior in X. laevis ..................................................................41 2-4 Comparison of food pellet (k ibble) consumption patterns ................................................42 2-5 Reaction to treatment with buffered MS-222 ....................................................................43
8 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science EFFECTS OF NITROGREN AMMONIA AND MS-222 ON Xenopus laevis DEVELOPMENT AND FORAGING BEHAVIOR By J. Kelly Byram December 2008 Chair: Max A. Nickerson Major: Wildlife Ecology and Conservation MS-222 (tricaine), an anesthe tic, is widely used by biologi sts on amphibians in the field, even though field use of MS-222 on amphibians is not approved by the U.S. Food and Drug Administration (FDA). Previous studies have identified MS-222s impacts on vision, olfaction, stress, heart, and liver, and documented its leth ality to certain microbe s that commonly populate amphibian skin. This study examines the poten tial impacts of "off-label" use of MS-222 on a model aquatic amphibian, Xenopus laevis (African Clawed Frog). Animals were exposed to an environmentally relevant concentration of n itrogen ammonia, a pollu tant commonly found in U.S. waterways, and unbuffered MS-222 in a ma nner simulating typical field use of the drug. Half of the animals had the webbing between two toes clipped. The animals foraging success in the hour post-recovery was observed. The group of animals exposed to nitrogen ammo nia had an increased variance in weight and length when compared to their controls although an ANOVA reveal ed no statistically significant difference in their means. MS-222 imp acted foraging behavior, with animals exposed to MS-222 eating significantly more food pellets than the control animals ( P = 0.0134). Web clipping did not impact foraging success ( P = 0.3646).
9 CHAPTER 1 EFFECTS OF NITROGEN AMMONIA ON XENOPUS LAEVIS DEVELOPMENT Introduction The rivers and streams of Missouri have we ll-documented populations of the Hellbender ( Cryptobranchus alleganiensis ) that are studied intensively by multiple teams of researchers. Because C. alleganiensis individuals are long-lived (Taber et al., 1975) and their populations have been so well documented, their decline has been recorded in detail. Animal removal for scientific research a nd the pet trade are two documented cau ses for population decline, but are likely not the only causes (Byram and Nickers on, 2008). Hammerson and Phillips (2004), in their assessment of the Hellbender's decline for th e IUCN's Red List, cite habitat loss as the reason for the Hellbender's likely continued and swift decline. The Ozark Hellbender ( Cryptobranchus alleganiensis bishopi ) is a candidate for listing as endangered or threatened by the U.S. Department of the Interior (DOI) (U .S. Department of the Interior, 2001), and has already been listed as endangere d by the state of Missouri (Utr up and Mitchell, 2008). Recent anecdotal evidence relating increased algal growth compared to prior years indicates pollution and resulting eutrophication could be causing a decline in habitat quality and suitability for the Hellbender. Adult Hellbenders, a lthough they have lungs, are fully aquatic and use cutaneous respiration almost exclusively (Guimond a nd Hutchison, 1973), making them especially vulnerable to declining water quality. Naturally occurring ammonia plays an important role in aquatic environments. Aquatic animals excrete ammonia as a byproduct of meta bolism and, it is suspected, as a disturbance pheromone powerful enough to elicit an avoidan ce reaction in tadpoles (Manteifel, 2006). But eutrophication resulting from anthr opogenic activities has been impli cated as one factor in global amphibian decline (Nystrm et al., 2007) and pathology (Johnson et al ., 2007). Eutrophication
10 results from high levels of nitrogen (e.g., n itrates, nitrogen amm onia) or phosophorous compounds in an ecosystem. The algal blooms and increasing levels of nitrogen ammonia on the North Fork of the White River, some of which is flanked by agricultural land and increasingly used for leisure activities, i ndicate the source of pollution ca sing the eutrophication is likely human activities in and along th e river (Quinlan and Phlips, 2007). Solis and his colleagues (2007) measured the levels of chemicals in the Eleven Point River and the North Fork of the White River. These two rivers have historic ally been known for har boring large Hellbender populations. The median amounts of total nitrogen (T N) in the Eleven Point and North Fork were 0.53 mg/L and 0.68 mg/L, respec tively, between August 2003 and November 2004. While the median TN of the Eleven Point River had in creased only slightly from its 1993-1997 median concentration of 0.5 mg/L, the median concentrat ion of TN in the North Fork had risen by 0.13 mg/L. Quinlan and Phlips (2007) investigated periphyton in the North Fork in September 2006 and found mean concentrati ons of TN along the river ranging from 0.35-3.06 mg/L. Studies of the affects of ammonia (NH3) on embryos and larvae have revealed statistically significant findings. Jofre et al. (2000) studied the effects of concentrations of un-ionized NH3 up to 2 mg/l on green frogs ( Rana clamitans ) and leopard frog ( R. pipiens ). Hatching success declined when leopard frogs embryos we re exposed to concentrations of NH3 greater than 1.5 mg/l, and the frogs were more likely to be defo rmed. Green frogs experienced similar impacts at a lower concentration of 0.6 mg /l, and the green frogs displa yed the additional problem of decreased tadpole survival and gr owth at this concentration. This research was designed to study the possi ble effects of an environmentally relevant concentration of nitrogen ammonia on the development of the surrogate amphibian, Xenopus
11 laevis (Chapter 1), and to investigat e the effects, if any, of th is environmentally occurring concentration of nitrogen ammonia alone and in combination with MS-222 (Chapter 2). The latest assessment of the Hellbender's status by the IU CN was done 4 years ago, in 2004, at which point the animal wa s given the status of Near Th reatened. The researchers noted, however, that "because this spec ies is probably in significant dec line (but probably at a rate of less than 30% over three genera tions (assuming a generation lengt h to be approximately ten years) because of widespread habitat loss thr ough much of its range, thus making the species close to qualifying for Vulnerable" (Hammerson and Phillips, 2004). For this reason and in order to not disturb natural populations of any other species, a surrogate species that is laboratoryproduced and readily available, the African Clawed Frog ( Xenopus laevis ), was chosen for this study. While the differences between C. alleganiensis and X. laevis are multiple, their similarities matter more for the study at hand. Both species are amphibians, have a larval stage, and are fully aquatic and use aquatic respirati on as adults. The well-doc umented ontogeny of the larval stage of X. laevis allows for visual monitoring and easy tracking of limb development of study animals (Nieuwkoop and Faber, 1994). Materials and Methods X. laevis tadpoles ( n = 270) from one clutch produced on 12 December 2007 were purchased from Xenopus Express (Brooksville, Fl orida) and distributed among 4 10-gallon glass aquarium tanks: 2 control ( n = 136) and 2 treatment ( n = 134) tanks. Each tank was filled with 32 L tap water treated with 1 mL dechlorinator (Top Fin, Pacific Co ast Distributing, Inc., Phoenix, AZ). The 2 treatment tanks were dosed with NH3-N (Ricca Chemical Company, Arlington, Texas) to a concentration of 0.5 mL/L. Each 1.00 mL of NH3-N contained 1.00 mg N and 1.216 mg NH3. The concentration for the treatment e nvironments was calculated based upon the nitrogen content of the NH3-N. Concentration was verifi ed by testing ammonia (NH3/NH4 +)
12 levels in the tanks (API, Inc ., Chalfont, PA). Temperature, pH (5-in-1 strips, Hach Co., Loveland, CO), nitrite (nitrate/ nitrite test strips, Hach C o., Loveland, CO), and nitrate (nitrate/nitrite test strips, H ach Co., Loveland, CO), were also monitored. The tanks were aerated with 10 cm air stones (Rolf C. Hagen Corporatio n, Mansfield, MA) run off of one air pump and adjusted to provide the same amount of aeration to each tank (by visual estimation). Tanks were not heated or cooled. Air and wa ter temperature were monitored daily. Over the course of the study, the observed water temperature ranged be tween 16-23C, but the average observed daily water temperature was usually 20-22C. Tadpoles were fed a liquid diet of tadpole pow der (Xenopus Express, Brooksville, Florida) prepared with filtered tap water (Brita faucet mount filtration system, Brita Products Company, Oakland, California). The solution was mixed well and shaken as needed to assure the powder remained suspended and distribution of nutriti on was equivalent across the tanks. Additionally, food was added to the tanks in a uniform fashi on but in a random orde r, assuring that any variability in food distribution re lated to suspension of food was distributed evenly over time across the tanks. At approximately 55 days of development, an imals began to exhibit aggressive behavior and water quality had begun to decline, even th ough the water in the tank s was changed daily, so the animals were separated into individual habita ts. Only animals that had reached a stage where they could eat food pellets, Nieuwkoop-Faber stage 64 or greater, moved on to the next stage of the protocol. These juveniles ( n = 152) were given a unique identif ication number and placed in individual 1-gallon bowls marked with the animal 's number. The remaining animals that had not metamorphosed ( n = 106) did not continue in the prot ocol. These animals were adopted by
13 individuals through the University of Florida Department of Animal Care Service's laboratory animal adoption program. Bowls were housed in numerical order on sh elves and unused bench space throughout the laboratory. The placement of treatment and co ntrol animals was randomized to prevent confounding from environmental sources, includ ing the laboratory's heating and cooling systems, sunlight, sound, vibration, etc. This configuration was shifted 3 times during the experiment. Temperature was not maintained on an individu al basis and was dictated by the ambient temperature of the laboratory sp ace. Laboratory windows provided filtered light for the animals, which was dictated by local weather patterns, an d exposure to fluorescent artificial lighting was regular and incidental to peri ods of human occupation of the laboratory, which could not be regulated, since the lab was shared by multip le researchers on variable schedules. Animals were fed 3/32" (2.38 mm) floati ng frog food pellets (Xenopus Express, Brooksville, Florida) on a regime nted schedule, which varied as the animals progressed through development. Uneaten food was removed from the bowls. Water changes, first partial and then full, were performed on a regular basis with r oom-temperature tap water treated with Top Fin Tap Water Dechlorinator (Pacific Coast Distri buting, Inc., Phoenix, Arizona) for removal of chlorine, chloramine, and heavy metals. Development and growth was monitored th roughout the study. Tadpoles were randomly selected from the tanks for measurement and staging at several point s of the protocol by scooping or netting in a variable fashion throughout the tank ar eas. The tadpoles were measured and weighed, then viewed under a light microsc ope for staging. Animals were staged according to the Nieuwkoop-Faber table (Nieuwkoop and Fa ber, 1994). Once the animals metamorphosed
14 and were in their individual bowls, they were measured during the wa ter change process to minimize stress. These juveniles were measured from snout to vent (SVL) and weighed, then returned to their bowl. Dimorphism is not pr onounced enough to reliably sex animals as very young juveniles without dissection, so data on sex were not collected. These animals were used in the protocol desc ribed in Chapter 2. At the end of the study, the animals were adopted by individuals through th e University of Florida Department of Animal Care Service's laboratory animal adoption program, as the animals who left the study at day 55 had been. Statistical Analyses Randomly selected tadpoles we re collected, measured, and st aged during metamorphosis. After metamorphosis, once the animals were give n unique identifying numbers, all animals were examined, weighed, and measured 4 times (days 55, 96, 112-113, and 130-131). Length, weight, and stage data was compared for the treatmen t (nitrogen ammonia) and control groups. Data were analyzed with a nonparametric ANOVA (Kru skal-Wallis), because the data did not meet the assumptions of parametric analysis. Data for days 55, 96, 112-113, and 130-131 (the 4 times when all animals were weighed and measured) were analyzed further by comp aring variances of the treatment (nitrogen ammonia) and control groups. Orlando and Guille tte (2001) posit that examining data from pollution-exposed populations so lely in terms of central tendency (e.g., ANOVA) without studying the accompanying variance can result in Type II errors. Comparison of population means may not capture the variance of the data, which is indicative of the variation in individuals and individual responses to contaminant e xposure in the population (Orlando and Guillette, 2001).All statistical an alyses were performed with alpha = 0.05 using SAS software (v. 9.1, SAS Institute, Inc., Cary, North Carolina).
15 Results The mean length, weight, and Nieuwkoop-Fabe r stage for the nitrogen ammonia and control groups are summarized in Table 11. ANOVA (Kruskal-Wallis) found no statistically significant differences in length, weight, or Nieuwkoop-Faber stage between the nitrogen ammonia treatment group and control group, As the mean stage of the tadpoles surpasse d Nieuwkoop-Faber stage 56 (Figure 1-1), the animals' total length sharply decreased as the tail was absorbed (Figure 1-2). The mean length continued to decrease until the animals began to reach and surpass stage 60, at which point it began to increase again. The plot s of juvenile weight (Figure 1-3) and SVL (Figure 1-4) are similar for both groups, with the treatment group a ppearing marginally heav ier and longer in the graph, although this difference is not stat istically significant. These checks during metamorphosis were done to monitor animals' welfare and check for abnormalities, so the statistical power for the sample size of 40 is low for such small measurements. Although the ANOVA revealed no statistically significant differences between the means of the treatment (nitrogen amm onia) and control groups, the exam ination of variances revealed differences between the two groups (Figure 15). With the exception of the SVL on day 55, variation in the weight on day 55 and the SVL and weight on the subsequent dates analyzed revealed an increased variance in the nitrogen ammonia treatm ent population compared to the control population. Twenty-eight animals died during this study. Twelve tadpoles (4.4%) died before day 55, and 16 juveniles (10.5% of the an imals who metamorphosed and continued with the study) died between day 55 and the end of the study on day 131, although 12 of those deaths resulted from accidental overdoses by the laboratory technician. The overdose deaths were the only deaths to occur after the MS-222 exposure pr otocol detailed in Chapter 2.
16 One death was accidental and unrelated to tr eatmentan animal from the control group jumped out of a tank overnight and was found on the lab bench, dead from desiccation, the next day. Of the other 6 dead control animals, 1 tadpole appeared small and underdeveloped compared to its cohorts and 1 juvenile appear ed anorectic, but the others displayed no gross abnormalities internally or externally. With the exception of the single anorectic animal that died on day 59, all of the control animals that died did so by day 15. Conversely, all but 1 of the treatment animals survived the first 15 days. The 1 treatment tadpole death during that period was from unknown causes. After day 15, one deformed animal died on day 23 at stage 51 with 3 normally developed limbs, but a mass of orange cells under the skin where the fourth limb (a forelimb) shoul d have developed. This animal lagged behind its treatment peers in development (Nieuwkoop-Faber stage) by approximately 1 week, well outside the SE for its cohort. At 31 mm SVL, this anim al was longer than the average length (SE) its peers had achieved when they had been at the same stage (51) a week previous (Table 1-1). Four tadpole deaths occurred in a cluster, between days 52 and 55. The gills of these animals appeared clotted with f ood. The gills of 2 of these anim als also appeared tinged with blood intermingled with the food particles. Thes e animals were between stage 53 and 55 at their time of death, well behind the development curve of their peers in the treatment group by 22-32 days, on average (SE). Their total lengths (range: 12-16 mm, mean: 14.5 mm) were markedly smaller than the average total length for their pe ers at the same developmental stage (mean total length ranged from 35.65 1.37 mm to 45.65 1.04 for stages 53-55 in their cohort) (Table 1-1). One dead juvenile appeared anorectic, and the remaining dead treatment animals displayed no gross abnormalities.
17 The size of the frogs by the end of the study was still too small to make positive distinctions of sexual dimorphism, and other exte rnal physiological featur es were too immature to judge as well. Post-metamorphic animals that expired were dissected and examined, but no abnormalities were found internall y. Given the size of the animals, this implies no more than there were no obvious, internal gross abnor malities visible under a light microscope. Discussion Although amphibian development is filled with unique and notable exceptions, one may speak in general terms of an amphibian reliance upon a body of water, permanent or ephemeral, for reproduction. Generally speaking, embryos and larvae live aquatically, metamorphosis occurs in the water, and the animals become terrestria l as juveniles. Notable exceptions are aquatic species like C. alleganiensis and X. laevis which are fully aquatic throughout the lifespan, and salamanders that can move into an eft stage and return to the water as environmental conditions dictate. But this relationship between crucial amphibian developmental stages of embryo and larva and our waterways is where the impact of water pollution on amphibi ans potentially has its greatest impact. The malformation of the treatment frog that di ed on day 24 at stage 51 with 3 limbs and a retarded rate of development seems similar to th e findings of Jofre et al.'s (2000) study of the effects of un-ionized NH3 on R. clamitans and R. pipiens reviewed in the introduction. However, a review of the literature yi elded nothing describing anythi ng similar to the underdeveloped treatment subjects with clotted gills who died during the study, while th e animals were still housed in aquariums. X. laevis are obligate suspensionfeedin g larvae, and they entrap suspended food in mucus secreted by the gills. F ood entrapment is decreased when the animals' environment becomes hypoxic enough that the anim als are forced to use gills for ventilation (Boutilier et al., 1992). The ammoni a level had held steady in the period prior to and during the
18 cluster of deaths between days 52 and 55, but nitrite concentrati on had increased to measurable levels on day 47 (0.15 ppm), and the first observa tion of measurable nitrate levels occurred on day 50 (1.0 ppm). The water was changed daily, but the nitrite levels rebounded daily. Nitrate would dip below measurable levels during this time and sporadically rebound to a measurable level. Since none of the control an imals experienced this gill-clot ting problem, there are a couple of probable explanations. Oxygen levels were no t measured. Assuming the increased levels of nitrite and nitrate, in conjunc tion with an increase in oxygen demand by a growing population of animals, are indicative of the environmen t becoming hypoxic, it is possible that the NH3-N exposure during development affected the gene (s) modulating the production of mucus in the gills. As a result, the animals' food entrapment di d not decrease in response to hypoxia, but other physiological responses did modulate correctly, suffocating the animal. Conversely, the buildup of food could have been indicative of a genera l malfunction in the anim al's metabolic system, which would also account for the delayed deve lopment and the relatively small length of the animals. Using the Kruskal-Wallis ANOVA, this study did not find a statistically significant effect of an environmentally relevant concentra tion of nitrogen ammonia on the growth and development of X. laevis in the laboratory in terms of progression through the stages of development, weight, and length. This study di d, however, document an increased variance in weight and length in the treatment animals ve rsus their controls. Th is increased variance indicates possible effects from e nvironmentally relevant concentr ations of nitrogen ammonia on X. laevis development, contrary to th e conclusions of the ANOVAs. Orlando and Guillette (2001) conclude that "incr eased phenotypic variance may be an important early sign of perturbation in a population." The sm aller weight and length measurements seen in
19 the treatment population in this study are indicative in amphibi ans of decreased fitness and, therefore, reproductive success. In situations such as this, with each subsequent generation the decreased productivity of the population has an add itive effect. As the population size dwindles, the genetic diversity decreases. Studies in ecotoxicology tend to employ hi gh concentrations of contaminants, and extrapolating the effects of lower, environmentall y relevant concentrations of pollutants can be difficult. This study used a concentration of n itrogen ammonia found in a Missouri stream and discovered an increase in phenotypi c variance. This indicates that, even at such low levels, there may be implications of nitrogen ammonia e xposure for the wildlife populations dependent upon these streams, including the Hellbender.
20Table 1-1. Summary of mean ( SE) NieuwkoopFaber stage (tadpoles), total length (mm) (tadpoles), weig ht (g), and SVL (mm) for treatment (N) and control groups. Stage Total length (snout to end of tail) Weight SVL Age (days) N Control N Control N Control N Control 16 50.00 0.32 ( n = 20) 50.8 0.09 ( n = 20) 24.30 0.84 ( n = 20) 25.55 0.37 ( n = 20) 17 51.25 0.22 ( n = 20) 51.20 0.09 ( n = 20) 27.80 0.82 ( n = 20) 27.60 0.5 ( n = 20) 18 51.65 0.11 ( n = 20) 51.80 0.16 ( n = 20) 30.20 0.61 ( n = 20) 30.80 0.85 ( n = 20) 19 52.35 0.25 ( n = 20) 52.70 0.11 ( n = 20) 31.60 1.07 ( n = 20) 33.80 0.61 ( n = 20) 20 52.90 0.10 ( n = 20) 53.00 0 ( n = 20) 35.40 0.77 ( n = 20) 35.35 0.63 ( n = 20) 21 53.00 0 ( n = 19) 52.80 0.20 ( n = 20) 37.70 0.76 ( n = 20) 35.90 1.40 ( n = 20) 22 53.05 0.09 ( n = 19) 53.25 0.10 ( n = 20) 35.65 1.37 ( n = 20) 39.00 0.86 ( n = 20) 24 53.47 0.18 ( n = 19) 53.55 0.11 ( n = 20) 37.60 1.63 ( n = 20) 40.50 0.89 ( n = 20) 25 53.85 0.11 ( n = 20) 53.80 0.12 ( n = 20) 41.55 1.10 ( n = 20) 41.25 0.89 ( n = 20) 27 54.53 0.14 ( n = 19) 54.80 0.92 ( n = 20) 43.35 1.60 ( n = 20) 44.50 0.92 ( n = 20) 29 54.74 0.37 ( n = 19) 54.85 0.17 ( n = 20) 43.80 1.33 ( n = 20) 45.30 1.19 ( n = 20) 15.55 0.53 ( n = 20) 16.60 0.43 ( n = 20) 30 55.05 0.21 ( n = 20) 54.90 0.24 ( n = 20) 45.65 1.04 ( n = 20) 45.60 1.21 ( n = 20) 16.15 0.53 ( n = 20) 16.35 0.37 ( n = 20) 32 54.90 0.34 ( n = 20) 55.30 0.18 ( n = 20) 45.55 1.37 ( n = 20) 48.10 1.3 ( n = 20) 16.05 0.65 ( n = 20) 17.00 0.36 ( n = 20) 34 55.75 0.18 ( n = 20) 55.55 0.11 ( n = 20) 47.55 1.16 ( n = 20) 48.75 1.19 ( n = 20) 17.00 0.4 ( n = 20) 17.35 0.31 ( n = 20) 36 55.80 0.17 ( n = 20) 55.65 0.25 ( n = 20) 49.85 1.17 ( n = 20) 49.25 1.23 ( n = 20) 17.40 0.28 ( n = 20) 17.20 0.37 ( n = 20) 38 56.00 0.45 ( n = 20) 56.70 0.42 ( n = 20) 47.55 2.01 ( n = 20) 50.15 1.21 ( n = 20) 16.95 0.57 ( n = 20) 17.15 0.38 ( n = 20) 43 58.17 0.97 ( n = 12) 58.67 0.92 ( n = 12) 44.08 3.57 ( n = 12) 46.17 3.77 ( n = 12) 16.75 0.55 ( n = 12) 17.50 0.76 ( n = 12)
21Table 1-1. Continued Stage Total length (snout to end of tail) Weight SVL Age (days) N Control N Control N Control N Control 45 60.72 0.92 ( n = 18) 60.06 0.99 ( n = 18) 35.33 4.01 ( n = 18) 38.94 4.41 ( n = 18) 15.72 0.78 ( n = 18) 16.33 0.63 ( n = 18) 50 60.26 1.06 ( n = 19) 60.61 0.95 ( n = 18) 37.21 3.78 ( n = 19) 39.56 4.35 ( n = 18) 17.11 0.62 ( n = 19) 16.67 0.71 ( n = 18) 55 65.74 0.06 ( n = 74) 65.67 0.07 ( n = 78) 0.36 0.02 ( n = 74) 0.37 0.02 ( n = 78) 14.85 0.24 ( n = 74) 14.56 0.23 ( n = 78) 112 4.55 0.15 ( n = 31) 4.27 0.12 ( n = 44) 33.90 0.43 ( n = 31) 33.36 0.39 ( n = 44) 113 4.23 0.16 ( n = 41) 4.21 0.15 ( n = 33) 33.76 0.65 ( n = 41) 33.64 0.53 ( n = 33) 112-113 4.37 0.11 ( n = 72) 4.24 0.09 ( n = 77) 33.82 0.41 ( n = 72) 33.48 0.31 ( n = 77) 130-131 6.73 0.21 ( n = 59) 6.48 0.16 ( n = 77) 36.73 0.21 ( n = 59) 36.23 0.32 ( n = 77)
22 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 671 6 1 7 1 8 1 9 2 0 2 1 2 2 2 4 2 5 2 7 2 9 3 0 3 2 3 4 3 6 3 8 4 3 4 5 5 0 5 5Study dayStage Treatment group (nitrogen ammonia) Control group Figure 1-1. Nieuwkoop-Fa ber stage by study day
23 20 25 30 35 40 45 50 551 6 1 7 1 8 1 9 2 0 2 1 2 2 24 2 5 2 7 29 3 0 3 2 3 4 3 6 3 8 4 3 4 5 5 0Study dayTotal length (mm) Treatment group (nitrogen ammonia) Control group Figure 1-2. Total length by study day. Note that, as tadpoles develop into juveniles and their tails begin to disappear, their length decreases. Once the tail is gone, the juveniles continue to grow and their length be gins to increase once again. 0 1 2 3 4 5 6 7 855112*113*112 & 113* 130-131Study dayWeight (g) Treatment group (nitrogen ammonia) Control group Figure 1-3. Total weight by study day
24 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 3729 30 32 34 36 38 43 45 50 55 112 113 112 & 113* 130-131Study daySVL (mm ) Treatment group (nitrogen ammonia) Control group Figure 1-4. SVL by study day
25 A B C D E F
26 G H Figure 1-5. Comparison of variance in treatme nt and control groups. Orlando and Guillette (2001) propose that an increase or decrease in variance in populat ions exposed to sublethal levels of pollution compared to c ontrol groups reflects phenotypic variation among the exposed organisms. A comparison of the means (e.g., ANOVA) will not reveal heteroscedasticity indicative of population effects resulting from exposure, possibly leading to Type II errors. Here the variance for SVL and weight for the control (C) and nitrog en ammonia-exposed animals (T) are compared at day 55 (A and B), day 96 (C and D), days 112 and 113 (E and F), and 130 and 131 (G and H). Although there is not an obvious differen ce between the SVLs for treatment and control animals on day 55 (A), the incr eased variance of the treatment group measurements in weight on day 55 (B) and the SVL and weight on the other days (CH) is easily discerned.
27 CHAPTER 2 EFFECTS OF NITROGEN AMMONIA AND MS-222 ON DEVELOPMENT AND FORAGING BEHAVIOR IN XENOPUS LAEVIS Introduction MS-222 is widely used by biologists to seda te and euthanize amphibians, both in the laboratory and in the field. Although marketed as a sedative appropriate for use in the field (Western Chemical, Inc., undated; Argent Chem ical Laboratories, undated), the Food and Drug Administration (FDA) specifically restricts MS-222's use on am phibians to the laboratory, saying "In other fish and cold-blooded animals, the drug should be limited to hatchery or laboratory use" (FDA, 1998). Despite this clear re striction, the literature is replete with "offlabel" use (use of a drug in a manner not appr oved by the FDA) of MS-222 in field studies of amphibian populations (Byram and Nickers on, 2008). MS-222 is recommended for use on amphibians in veterinary and zoological publicati ons (Gentz, 2007) and for use in the field in government documents (Green, 2001). As Crook a nd Whiteman (2006) state in their own study of MS-222, "Typically the choice of anesthesia ha s been based on what other researchers have used rather than a critical ev aluation of different methods." The off-label use of MS-222 on amphibians in the field is potentially problematic because direct inquiry and anecdotal evid ence point to field techniques ve ry different from the suggested usage and safety guidelines, with field biologist s often mixing the anesth etic powder with water from a local body of water, as in structed by the package insert provided by Argent and Western (Western Chemical, Inc., undated; Argent Chem ical Laboratories, undated). MS-222 is intended for use with clean water, and with the decline of water quality in aquatic ha bitats (see Chapter 1) and the uncertainty of what chemicals and metals are in these habitats, th ere is the potential risk the anesthetic bath may be inappropriately prep ared. The Argent directions for Finquel read, "Do
28 not use...water containing chlorine, heavy me tals (copper, zinc, etc.), or other toxic contaminants" (Argent Chemical Laboratories, undated). Because of its use in the aquaculture indus try on fish consumed as food for humans, copious research on MS-222's effects on fish is available, but research on its effects on amphibiansespecially when it is used off-labelis comparatively scant. This research was designed to compare the effects of off-label fi eld use of MS-222 on amphibians to laboratory use. To mimic field use, the MS-222 bath was prepared with 0.5 mg/L nitrogen ammonia in dechlorinated tap water (see Chapter 1) with no buffer added, while the laboratory treatment animals were anesthetized in an MS-222 bath prep ared with dechlorinated tap water and a baking soda buffer. Half the animals in this protoc ol had the webbing between two toes clipped approximately 5 mm, half with MS-222, half w ithout, to determine web clipping's effects on foraging, if any. The research design is illust rated in Figure 2-1. After treatment, animals' foraging success over the next hour was observe d, and animals' post-treatment growth was measured. The use/omission of buffer was continge nt upon whether the treatment was "field use" (nitrogen ammonia in dechlorinated tap water) or "lab oratory use" (tap water alone), so this is an unbalanced design in that respec t. The additional number of an imals needed for statistical analysis had the design been balanced with regard to buffer was not justifiable since the effects of unbuffered MS-222 are well-documented in th e literature. An anal ysis of the various treatments' effects on regeneration of the webbing wa s planned, but the inability to create clips of uniform length rendered that pa rt of the protocol obsolete. Materials and Methods Half of the animals from the treatment a nd half from the control groups in the study described in Chapter 1 were randomly assigne d to the MS-222 treatment group by using a random-number generator ( www.random.org ) to pick animal numbers. The remaining half of
29 each group was assigned to the control group. Th en, half of each group (MS-222 and control) was randomly assigned to have their webbing cl ipped or not. In this fashion, the nitrogen ammonia and control animals were evenly disper sed among this study's treatments. As noted in Chapter 1, the tadpoles were produced on 12 Decem ber 2007, day 1. I performed this protocol on days 112 and 113. The anesthetic solution for the animals housed in nitrogen ammonia water was prepared with 1 L dechlorinated H2O at 23.3C, 0.50 mg NH3-N, and 1 g tricaine powder (Finquel, Argent Labs, Redmond, Washington). The anesthetic so lution for the control group (non-nitrogen ammonia) was prepared with 1 L dechlorinated H2O at 23.3C and 1 g tricaine powder. Unlike the solution for the nitrogen ammonia animals, this solution for the control animals was buffered with 1 g baking soda. Animals were removed from their individual tank s, weighed, measured, and then placed in the anesthetic bath. Animals were kept in th e anesthetic bath until they stopped swimming, at which point they were retrieved and gently placed on their backs on the researcher's palm. If the animal attempted to right itself, it was placed back in the anesthet ic bath and carefully monitored. When the animals ceased attempting to ri ght themselves in the researcher's hand, they were considered anesthetized. The time it took fo r the animals to reach the desired anesthetic plane was not recorded, and it vari ed widely from animal to animal. Typically animals ceased righting themselves within 3-10 minutes. Animals in the buffered anesthetic succumbed quicker than the animals who received the unbuffered MS-222. At this point, those in the group designated for web clipping were clipped. The web clipping control group was handled for a duration and in a fashion mimicking the handli ng of the web clipping group to control for handling stress.
30 Anesthetized animals were then placed direc tly in a recovery bat h. For animals in the nitrogen ammonia group, the bath was a mixtur e of 1 L 23.3C dechlorinated tap water and 0.50 mg NH3-N (the same concentration of nitrogen amm onia they had lived in since the beginning of the study). For the control animals, the bath omitted the NH3-N. Periodically, water was gently agitated by hand to create a flow of water on the skin of the animals to facilitate recovery. Once the animals recovered gross motor skills and were able to right themselves on the researcher's palm, they were returned to their individual ta nks, which had been cleaned and refilled with water in the interim. The tank wa s then placed in a lighted observation area, the time noted, and 8 pieces of the floating food pellets of the same size and brand they were fed throughout the study added to the water. Animals were observed in termittently from a distance of approximately 0.3 to 1.3 meters to assure their safety and rec overy. At one hour post-treatment, the number of food pellets eaten was noted, uneaten food pellets were removed from the tank, and the animal's tank returned to its assigned place in the laboratory space. Animals were weighed and measured on days 130 and 131. This ended the protocol, and all study animals were adopted through the Univ ersity of Florida Animal Care Service's laboratory animal adoption program. Statistical Analyses Statistical analysis was complicated by the mo rtality of treatment subjects due to causes unrelated to the experimental pr otocol. A graduate student with IACUC certification and years of experience working with animals who was briefly employed as a lab assist ant overdosed some of the treatment animals, presumably by a factor of two, which proved lethal. Therefore, the sample sizes of the groups for the measurement data of days 130-131 are not e qual. Group sample sizes are summarized in Table 2-1.
31 As in Chapter 1, the data for this protocol did not meet the assumptions for parametric analysis, and so the nonparameteric ANO VA (Kruskal-Wallis) was employed for ANOVA comparisons. Post-treatment growth (weight and length) were calculated by subtracting day 112 and 113 data from day 130 and 131 data. All statis tical analyses were performed with alpha = 0.05 using SAS software (v. 9.1, SAS Inst itute, Inc., Car y, North Carolina). Results The number of food pellets (kibble) eate n by each subject une xpectedly became a constrained value. The hypothesis was that an imals affected by MS-222 treatment would not successfully forage for food due to a disruption in vision and/or olfac tion and that control animals would eat the same, or if stressed, less (or none) of the food pellets. The MS-222 treatment in fact, ate more th an the control group, and food pelle ts became a constrained value. This resulted in an extremely non-normal distri bution (Figure 2-2). Addi tionally, the variances between the MS-222 and control gro ups were not equal. Data transformations were attempted to normalize the data's distribution and equalize th e variances, but both could not be achieved simultaneously; therefore, data were not transformed and were analyzed using a nonparametric ANOVA (Kruskal-Wallis). The research design called for the protocol to be performed on all of the animals in one day, having been fed the day prior to treatment. The protocol was performed on two subsequent days, instead, and the animals what were re-sch eduled for the second day received no food. The first and second day animals were randomly sp read over all treatment s, and a statistical comparison of the number of food pe llets eaten by animals on the firs t day to those on the second day confirmed behavior was not significantly impacted ( P = 0.0627). Therefore, data for the two days are treated as one set.
32 The animals treated with unbuffered MS-222 at e significantly more food pellets in the recovery hour after treatment than the control group did ( P = 0.0134) (Figure 2-3). Within-group analysis of the animals who received MS-222 tr eatment revealed no difference in food pellet consumption between the animals whose we bbing was clipped and their controls ( P = 0.0859), nor was there any difference between the animals ra ised in 0.50 mg/L con centration of nitrogen ammonia and their controls ( P = 0.7010). A visual comparison of the kibble consump tion data for the MS-222 and control groups reveals a notable difference between the two gr oups (Figure 2-4). The histogram for the food pellet consumption of all animals takes on the predic table form of a constrained value, with sharp peaks at the either end of the distribution. Separating the histog ram into the two groups reveals differences in the distributions. Roughly the same percentage of animals in each group ate all of the food pellets offered: 22.9% of the MS-222 group and 19.4% of the control group. However, the percentage of animals that ate nothing was vastly different between the two groups. Only 10% of the MS-222 animals ate nothing, while near ly one-third (31.9%) of the control animals did not consume any food pellets. Within the control group there were no statis tically significant differe nces in consumption between the animals with clippe d webbing and their controls ( P = 0.3646) or the nitrogen ammonia animals and their controls ( P = 0.4666). Animals anesthetized with unbuffered MS-222 e xhibited no difference in growth in terms of weight or length compared to cont rol subjects 18 days after treatment ( P = 0.1642 and P = 0.4667, respectively). Weight was likewise unaffected by the web clipping ( P = 0.2349), but web-clipped animals showed a statistically sign ificant increase in length compared to the unclipped animals ( P = 0.0109).
33 The mortalities ( n = 8) during the course of this pr otocol (post-day 111) were due to accidental overdose of nitrogen ammonia and not attributable to treatment with MS-222. Because the mortalities were al l in the nitrogen ammonia treatment group, the impact to the statistical power was great and the ability to differentiate decreased. Discussion The stress of toe-clipping ha s been a topic of hot debate in the herpetological community of late (Kinkead et al., 2006; Langkilde and Sh ine, 2006), but this study found no practical effect of clipping on foraging immediatel y after the procedure. In this study, the kinetics of foraging were not under investigation: the web, not a to e, was clipped. Rather, the study was designed to investigate the animals' reaction to the experience in terms of chemosenso ry acuity and clippinginduced stress during the recovery period, as m easured by the number of food pellets eaten (foraging behavior). All animal s were captured, weighed, and measured in a similar fashion, and animals whose webbing was not clipped were ha ndled in a way that simulated the handling experience of those being web-clipped (without the actual web clipping). However, getting the toes of the hindlimb to separate from each other and spread the webbing taught enough to accurately clip the web and cr eate a cut of uniform length proved difficult on some unanesthetized individuals. No difference in fo raging was found between those frogs and their anesthetized controls, so the effect of slightly prolonged ha ndling of these individuals (perhaps 5-10 seconds more) may have been negligible. Studies have used differences in eating pattern s to differentiate between the level of stress and pain. Carr et al. (2002) in their review of neuropeptides' ro les in prey-catching behavior, summarize the research on the relationship between stress and eating patterns across a range of animals, with chronic or severe stress typica lly inducing anorexia, a nd relatively minor stress (tail-pinching in rats) causing ove reating in response. The lack of difference in foraging between
34 the web-clipped and unclipped animals leads to the conclusion that be ing web-clipped during handling was no more stressful than the handling alone This is in line with the findings of recent research (Kinkead et al., 2006; Langkilde and Shine, 2006). Stress, pain, and distress are th e 3 elements of the USDA's classification status for animal research. Toe clipping is considered amputati on, and a protocol involvi ng toe clipping without anesthesia is considered a Category 3. Appr oval for Category 3 studies comes only after the close scrutiny of the IACUC members and a lot of additional work on the part of the investigator. It is for this reason, and perhaps the genuine be lief on the part of some researchers that if clipping a finger would cause them pain and stress the same must be true for amphibians, that MS-222 is used as a matter of course in amphibian research. With the publication of studies concluding that this anthropocen tric view of toe clipping is inaccurate (Kinkead et al., 2006; Langkilde and Sh ine, 2006), the appropriateness of anesthesia for protocols involving clipping wi thout other, more invasive pr ocedures, is arguable. Exposure to MS-222 has been shown to be more stressful than handling stress (Vethamany-Globus et al., 1977), but there appears to be no straightforward answer to the relationship between MS-222 and stress. Stressors in multiple iterations (pollution, predators, habitat loss) have been linked to immunosuppression (Belden and Ki esecker, 2005), declines in reproduction potential and survival (Edgington, 2003, Barbeau and Guillette, 2007) and synergistic toxi city with pesticide (Relyea and Mills, 2001). Dosage and length of exposure are two vari ables that figure predominantly in the stress res ponse of animals to MS-222 exposure, and the vast range of susceptibility to MS-222 that seems to vary by species and perhaps be modulated by repeated exposure (Zuccarelli and Ingermann, 2005), complicates the situation.
35 Part of the confusion arises from the apparent lack of a clear answer as to how to best physically measure the concept of stress. Vetham any-Globus et al. (1977) found that the stress response (as measured in bl ood glucose levels) in adult Notophthalmus viridescens exposed to MS-222 was greater than the res ponse to handling stress experi enced by their controls after 30 minutes of exposure, but that use of a highe r concentration and a shorter exposure time minimized the hyperglycemic response. Welker et al. (2007) also found the lowest glucose levels in their controls when measuring stress in Channel Catfish ( Ictalurus punctatus ). But their protocol measured stress with cortisol levels unde r the assumption that cort isol solely regulates hyperglycemia. They concluded animals treated with MS-222 had lower cortisol levels than their controls, but that cortisol levels increase d as MS-222 concentration increased and that hyperglycemia may not be solely regulated by cortisol. Welker et al. (2007) also found that bufferi ng MS-222 did not affect cortisol levels, glucose levels, induction, or recovery. However, it is generally accepted practice to buffer MS222 with NAHCO3 (baking soda) to minimize stress, and Zucccarelli and Ingerman (2005) observed faster i nduction when MS-222 is buffered for Oncorhynchus mykiss fry. Not buffering MS-222 solution in field conditions for the sake of expediency is a common practice, but not an acceptable one. Solutions are easily buffered by a dding baking soda, and the resulting solution is a comfortably neutral (pH 7) anes thetic bath. Unbuffered MS-222 dr opped to a pH of 4-5 in this study, and pH below 4-5 can cause death in am phibians (Boutilier et al., 1992). The negative effects of low pH on the amphibian skin are vari ous and well documented, but specific aspects of the topic directly apply to the question at hand. Exposure to acidic conditions has been shown to increase the toxicity of herbic ides as the environmental pH decreases (Edgington et al., 2003), and Rana pipiens exposed to acidic conditions experien ced pronounced infections of the spleen
36 from Gram-positive and Gram-negative bacteria (Simon, 2002). Increased glucocorticosteroids, a response to stress, was shown to be immunosupp ressive in larval treefrogs by Belden and Kiesecker (2005). However, while buffering mitigates the stress and potential damage to the epidermis and amphibian immune system, it does not completely negate them. During the recovery period, the skin on the back of one frog formed a white cottony layer buoyed by water bubbles (see Figure 2-5). It had been treated with buffered MS-222, and handled gently in gloved hands, yet clearly the epidermis had suffered trauma. Recently, Lauer et al. (2007) indicated th at "antifungal skin bacteria may form mutualistic ecological relationships with amphi bian species and help protect them from pathogenic fungi." More research needs to be done in this area before we can draw many conclusions, but we should consider the possibility that MS-222 is lethal to a certain subset of microorganisms that naturally occur on and in am phibians and upsets the be neficial balance of microorganisms on amphibian skin, potentially ne gatively impacting animals' respiratory and immune functions. Gram-negative bacteria are impacted by MS-222. Fedewa and Lindell identified 11 bacteria species with negatively impacted growth after exposure to MS-222 (2005). The effect was greater at highe r concentrations, and unbuffered solutions were more lethal starting at lower concentrations yet another indication of the importance of buffering solutions. Unfortunately, buffered MS-222 does not appear to kill Batrachochytrium dendrobatidis the fungus that causes chytridiomycosis in amphibi ans (Webb et al., 2005), although the effects of unbuffered MS-222 on B. dendrobatidis are unknown. MS-222, buffered and unbuffered, has been s hown to affect the amphibian heart and induce bradycardia (Cakir and Strauch, 2005). Wayson et al. (1976) hypothesize that amphibians'
37 slow metabolism is likely to render it more toxic to the liver, which is where amphibians hydrolize MS-222. Anesthesia's effect on the liver is well known, and a recent study found ammonia has "anesthetic propertie s," in part because high concentrations of ammonia cause fulminant hepatic failure in some mammals (Bro snan et al., 2007). While the study reported here was limited by its scope, timeline, and the conser vative level of nitrogen ammonia concentration used, it did attempt to discern whether a syne rgistic effect between environmental ammonia conditions and the effects of MS-222 existed through the foraging element of the protocol. Animals exposed to MS-222 who lived in nitroge n ammonia performed no differently in the foraging test from their clean-w ater MS-222-exposed peers. So, regardless of nitrogen ammonia exposure, all animals exposed to MS-222 ate significantly more food pellets in that hour postrecovery than did their controls. This behavi or was unexpected, and the use of 8 food pellets, which had seemed a generous amou nt, suddenly became a constraint. MS-222's hydrophobic alkyl chain "ta il" suppresses dark vision in amphibians (Bernstein et al., 1986) by inhibiting the pr oduction of rhodopsin. Bleaching of the visual system occurs whenever a dark-adjusted amphibi an is brought into the light, but in one study the control group of animals whose visual system was bleached by light exposure regained 100% of its rhodopsin by 1.5 hours after exposure to the light, while th e group exposed to light and MS-222 recovered less than half of its rhodopsin by this time (Hoffman and Basinge r, 1977). Lewis (1985) noted negative effects on the chemosensory discrimination of Channel Catfish ( Ictalurus punctatus Rafinesque) due to the loss of c ilia on the olfactory sensory ep ithelia which lasted 11-28 days following MS-222 exposure (Lewis et al., 1985). I did not expect th e same results. The recovery area in this study was lighted, and foraging succ ess, not differentiation between olfaction and visual acuity's role in the foragi ng, was the variable of interest.
38 Stress and chemosensory impairment were hypothesized to have an additive, if not synergistic, negative effect on the animals' intere st in and/or ability to forage. A number of plausible theories may explain the unexpected increase in foraging among the MS-222-treated animals. Future studies may want to explor e further MS-222's effects on sensory inputs. Undocumented visual and olfactor y impacts may have changed the tr eated animals' perception of the food or their surroundings. The animals' per ception of hunger may have been altered, or the mechanisms involved in the perception of satie ty might have been impacted. The exposed animals may experience a different type of stress than the animals that were not anesthetized. While chronic or severe stress causes anorexic behaviors, -endorphins ( E) are suspected of being responsible for stress-induced overeating in mammals, birds, and fish in response to irritation (Carr et al., 2002). With dwindling amphibian populations receivi ng more research attention, the question of how our interactions with our study populations impact their beha vior and health needs to be answered. A powerful chemical like MS-222 should not be applied to animals without a full appreciation for the implications of its use. This study's findi ng that the exposed animals ate more food pellets in the hour after treatment with and rec overy from MS-222 than their unexposed counterparts indi cates that the drug does have an effect on X. laevis after they have recovered their gross motor sk ills and appear "recovered." This study did not discern the underlying mechanism for that beha vior, but researchers exposing an imals to MS-222 in the field or the laboratory should take this informati on into consideration when designing amphibian research protocols.
39 Table 2-1. Summary of sample sizes for days 130-131 for the nitrogen ammonia treatment (N) and control groups. Web-Clipped Control N Control N Control MS-222 17 18 16 19 Control 10 20 16 20 MS-222 web clipped (unbuffered) not clipped MS-222 web clipped (buffered) not clipped no web clipped anesthesia not clipped no web clipped anesthesia not clipped Nitrogen Ammonia Dechlorinated Tap Water Figure 2-1. Summary of the research design
40 Figure 2-2. Number of food pelle ts (kibbles) eaten in 1 hour is a constrained value. Animals could eat no fewer than 0 food pellets a nd no more than the number offered, 8. In hindsight it is evident that some of them may have eaten more than 8, but while that would have altered the distribution, it would have offered very little additional insight into the behavior being studied.
41 Figure 2-3. MS-222 affects foraging behavior in X. laevis. Treated animals (1) ate more food pellets on average than the control animals (0).
42 Figure 2-4. Comparison of food pelle t (kibble) consumption pattern s. This histogram compares the number of food pellets (0 -8) eaten by the control group (the 9 left-most bars, labeled "0") and the MS-222 treatment group (the 9 right-most bars, labeled "1"). 31.9% of the control animals (group "0") at e none of the food pellets offered after treatment, compared to 22.9% the animals that had been anesthetized (group "1").
43 Figure 2-5. Reaction to treatment with buffered MS-222. This animal had no visible skin problems or defects prior to treatment with MS-222, but exhibited bubbling and sloughing in the hour post-treatment.
44 LIST OF REFERENCES Argent Chemical Laboratories (undated): Finquel MS-222 (instruction sheet). Redmond, Washington, Argent Chemical Laboratories. Barbeau, T.R., Guillette, L.J., Jr. (2007): Alte red endocrine parameters in female frogs ( Xenopus laevis ) exposed to environmentally relevant concen trations of nitrate. Journal of Herpetology 44 : 590-596. Belden, L.K., Kiesecker, J.M. (2005): Glucocortico steroid hormone treatment of larval treefrogs increases infection by Alaria sp. trematode cercariae. J. Parasitol. 91 (3):686-688. Bernstein, P.S., Fulton, B.S., Rando, R.R. (1986): M echanism of action of ar omatic amines that short-circuit the visual cycle. Biochemistry 25 : 3370-3377. Boutilier, R.G., Stiffler, D.F., Toews, D.P. ( 1992): Exchange of respiratory gases, ions, and water in amphibious and aquati c amphibans. In: Environmental Physiology of the Amphibians, p. 84. Feder, M.E., Burggren, W.W., Eds., Ch icago, IL, University of Chicago Press. Brosnan, R.J., Yang, L., Milutinovic, P.A., Zhao, J ., Laster, M.J., Eger, E.I., II, Sonner, J.M. (2007): Ammonia has anesthetic properties 104 (6):1430-1433. Byram, J.K., Nickerson, M.A. (in press): Th e Use of Tricaine (M S-222) in Amphibian Conservation. Gainesville, FL, Reptil e and Amphibian Conservation Corps. Cakir, Y., Strauch, S.M. (2005): Tricaine (MS-22 2) is a safe anesthetic compound compared to benzocaine and pentobarbital to in duce anesthesia in leopard frogs ( Rana pipiens ). Pharmacological Reports 57 :467-474. Carr, J.A., Brown, C.L., Mansouri, R., Venkate san, S. (2002): Neuropeptides and amphibian prey-catching behavior. Co mp. Bio. and Phys. Part B 132 :151-162. Crook, A.C., Whiteman, H.H. (2006) : An evaluation of MS-222 and benzocaine as anesthetics for metamorphic and paedomorphic tiger salamanders ( Ambystoma tigrinum nebulosum ). Am. Midl. Nat. 155 :417-421. Edgington, A.N., Stephenson, G.R., Sheridan, P. M., Thompson, D.G., Boermans, H.J. (2003): Effect of pH and Release on two life stages of four anuran amphibi ans. Environ. Toxicol. Chem. 22 (11):2673-2678. Fedewa, L.A., Lindell, A. (2005): Inhibition of growth for select Gram-negative bacteria by tricaine methane sulfonate (M S-222). J. Herpe. Med. Surg. 15 :13-17. Food and Drug Administration (1998): 21 CFR 529.2503 Tricaine methanesulfonate. ( www.absc.usgs.gov/research/v et/policies/21CFR530.9VL ). Gentz, E.J. (2007): Medicine and su rgery of amphibians. ILAR Journal 48 (3):255-259.
45 Green, D.E. (2001): Anesthesia of amphibians in the field. USGS Standard Operating Procedure. ARMI SOP No. 104. Guimond, R.W., Hutchison, V.H. (1973): Aquatic respiration: an unusual strategy in the Hellbender Cryptobranchus alleganiensis alleganiensis (Daudin). Science 182 : 1263-1264 Hammerson, G., Phillips, C. (2004): Cryptobranc hus alleganiensis. In: IUCN 2007. 2007 IUCN Red List of Threatened Species.
46 Orlando, E. F., Guillette, L. J., Jr. (2001): A re-examination of variation associated with environmentally stressed orga nisms. Hum Reprod Update 7 (3):265-272. Quinlan, E. L., Phlips, E. J. (2007): Preliminar y investigation of peri phyton in the North Fork Branch of the White River, Missouri. Unpublished report. Relyea, R.A., Mills, N. (2001): Predator-induced st ress makes the pesticide carbaryl more deadly to gray treefrog tadpoles ( Hyla versicolor ). PNAS 98 (5):2491-2496. Simon, M.P., Vatnick, I., Hopey, H.A., Butler, K., Korver, C., Hilton, C., Weimann, R.S., Brodkin, M.A. (2002): Effects of acid exposure on natural resistance a nd mortality of adult Rana pipiens J. Herpetol. 36 (4):697-699. Solis, M.E., Liu, C.C., Nam, P., Niyogi, D.K ., Bandeff, J.M., Huang, Y.-W. (2007): Occurrence of organic chemicals in two rivers inhabited by Ozark Hellbenders ( Cryptobranchus alleganiensis bishopi ). Arch. Environ. Contam. Toxicol. 53 :426-434. Taber, C.A., Wilkinson, R.F., Jr., Topping, M.S. (1975): Age and growth of hellbenders in the Niangua River, Missouri. Copeia 1975 (4):633-639. Utrup, J., Mitchell, K. (2008): The Ozark He llbender: Out from under a rock. Endangered Species Bulletin 33 (1):22-24. U.S. Department of the Interior ( 2001): Federal Regist er 50 CFR Part 17. 66 (210):54811. Vethamany-Globus, S., Globus, M., Fraser, I. (1977) Effects of tricaine methane sulfonate (MS222) on blood-glucose levels in adult salamanders ( Diemictylus viridescens ). Experientia 33 : 1027. Wayson, K.A., Downes, H., Lynn, R.K., Gerber N. (1976): Studies on the comparative pharmacology and selective toxicity of tricaine methanesulfonate : metabolism as a basis of the selective toxicity in poikilotherms. The Journal of Pharmacoloy and Experimental Therapeutics 198 (3):695-708. Webb, R., Berger, L., Mendez, D., Speare, R. (2005): MS-222 (tricaine methane sulfonate) does not kill the amphibian chytri d fungus Batrachochytrium dendr obatidis. Dis. Aquat. Organ. 68 :8990. Welker, T., Lim, C., Aksoy, M., Klesius, P. ( 2007): Effect of buffered and unbuffered tricaine methanesulfonate (Ms-222) at di fferent concentrations on the stre ss responses of channel catfish ( Ictalrus punctatus Rafinesque). Journal of Applied Aquaculture 19 (3):1-18. Western Chemical, Inc. (undated): Tricaine-S. ( http://www.wchemical.com/Assets /File/tricaineS _instructions.pdf ). Zar, J.H. (1999): Biostatistical Analysis, 4e. Upper Saddle River, New Jersey, Prentice Hall.
47 Zuccarelli, M.D., Ingermann, R. L. (2005): Influence of neutraliz ing agents on the anaesthetic efficacy of tricaine on Oncorhynchus mykiss (Walbaum) fry. Aquaculture Res. 36 :933-935.
48 BIOGRAPHICAL SKETCH J. Kelly Byram received her Bachelor of Ar ts from Indiana University, Bloomington, in 1990. She lives in Gainesville, Florida.