Response of the Italian Agile Frog (Rana Latastei) to a Ranavirus, Frog Virus 3: a Model for Viral Emergence in Naïve Populations

Home , Italian agile frog

Pearman, Peter B; Garner, Trenton W J; Straub, Monika; Greber, Urs F. Response of the Italian agile frog (Rana latastei) to a Ranavirus, frog virus 3: a model for viral emergence in naïve populations. J. Wildl. Dis. 2004, 40(4):660-9. Postprint available at: http://www.zora.unizh.ch University of Zurich Posted at the Zurich Open Repository and Archive, University of Zurich. Zurich Open Repository and Archive http://www.zora.unizh.ch

Originally published at: Winterthurerstr. 190 J. Wildl. Dis. 2004, 40(4):660-9 CH-8057 Zurich http://www.zora.unizh.ch

Year: 2004

Response of the Italian agile frog (Rana latastei) to a Ranavirus, frog virus 3: a model for viral emergence in naïve populations

Pearman, Peter B; Garner, Trenton W J; Straub, Monika; Greber, Urs F

Posted at the Zurich Open Repository and Archive, University of Zurich. http://www.zora.unizh.ch

Originally published at: J. Wildl. Dis. 2004, 40(4):660-9 Response of the Italian agile frog (Rana latastei) to a Ranavirus, frog virus 3: a model for viral emergence in naïve populations

Abstract

Ranavirus (family Iridoviridae) is a genus of pathogens of poikilotherms, and some ranaviruses may play a role in widespread mortality of amphibians. Ecology of viral transmission in amphibians is poorly known but can be addressed through experimentation in the laboratory. In this study, we use the Ranavirus frog virus 3 (FV3) as an experimental model for pathogen emergence in naive populations of tadpoles. We simulated emerging disease by exposing tadpoles of the Italian agile frog (Rana latastei), to the North American Ranavirus FV3. We demonstrated that mortality occurred due to viral exposure, exposure of tadpoles to decreasing concentrations of FV3 in the laboratory produced dose-dependent survival rates, and cannibalism of virus-carrying carcasses increased mortality due to FV3. These experiments suggest the potential for ecological mechanisms to affect the level of exposure of tadpoles to Ranavirus and to impact transmission of viral pathogens in aquatic systems. Journal of Wildlife Diseases, 40(4), 2004, pp. 660–669 ᭧ Wildlife Disease Association 2004

RESPONSE OF THE ITALIAN AGILE FROG (RANA LATASTEI) TO A RANAVIRUS, FROG VIRUS 3: A MODEL FOR VIRAL EMERGENCE IN NAIÈVE POPULATIONS

Peter B. Pearman,1,2,4 Trenton W. J. Garner,1,3 Monika Straub,1 and Urs F. Greber1 1 Zoologisches Institut, University of Zurich, Winterthurerstrasse 190, 8057 Zurich, Switzerland 2 Current address: Michigan Natural Features InventoryÐZoology and Department of Fisheries and Wildlife, 345 Giltner Hall, Michigan State University, East Lansing, Michigan 48824, USA 3 Current address: Wildlife Epidemiology, Institute of Zoology, Zoological Society of London, Regents Park, London NW1 4RY, UK 4 Corresponding author (email: [email protected])

ABSTRACT: Ranavirus (family Iridoviridae) is a genus of pathogens of poikilotherms, and some ranaviruses may play a role in widespread mortality of amphibians. Ecology of viral transmission in amphibians is poorly known but can be addressed through experimentation in the laboratory. In this study, we use the Ranavirus frog virus 3 (FV3) as an experimental model for pathogen emergence in naı¨ve populations of tadpoles. We simulated emerging disease by exposing tadpoles of the Italian agile frog (Rana latastei), to the North American Ranavirus FV3. We demonstrated that mortality occurred due to viral exposure, exposure of tadpoles to decreasing concentrations of FV3 in the laboratory produced dose-dependent survival rates, and cannibalism of virus-carrying carcasses increased mortality due to FV3. These experiments suggest the potential for ecological mechanisms to affect the level of exposure of tadpoles to Ranavirus and to impact transmission of viral pathogens in aquatic systems. Key words: Amphibian decline, cannibalism, emerging diseases, frog virus 3, iridovirus, IUCN red list, Ranavirus, Rana latastei.

INTRODUCTION effect of closely related iridoviruses (e.g., The genus Ranavirus (family Iridoviri- largemouth bass virus, Grizzle and Brun- dae) constitutes globally distributed path- ner, 2003) on species of economic impor- ogens that have been found in association tance. Ecologists interested in develop- with catastrophic mortality events in sala- ment of mechanistic ecological models manders and frogs (Cunningham et al., that express interactions between Ranavi- 1996; Jancovich et al., 1997) and may con- rus and their hosts are challenged to de- tribute to declines of amphibian popula- velop experimental systems that allow ma- tions (Daszak et al., 1999; Chinchar, 2002). nipulation of both host and pathogen. While the mechanisms of Ranavirus entry Frog virus 3 (FV3) is already used as a into cells, viral replication, and pathogen- model for the study of immunity to Ran- esis are increasingly understood (reviewed avirus (Gantress et al., 2003) and for the in Chinchar, 2002), Ranavirus host-path- study of mechanisms of Ranavirus viral ogen ecology has not been well studied. replication (Chinchar, 2002). We extend Little is known of the epidemiology of the use of FV3 as a model for host/path- Ranavirus, the biotic and abiotic factors ogen ecology and pathogen emergence. that influence transmission and host sus- Pathogens emerge when their geographic ceptibility, the means for viral persistence range expands, often due to anthropogenic in populations, and how the virus disperses influence, or when they invade new host among and spreads within host popula- species and populations (Daszak et al., tions. The importance of understanding 2001). Thus by definition, exposure to host-pathogen ecology of amphibians and emerging pathogens largely involves naı¨ve Ranavirus is highlighted by the potential hosts. We simulate disease emergence by for some anurans to harbor other patho- challenging tadpoles of a European frog, gens that may affect public health (Kozuch the Italian agile frog (Rana latastei), with et al., 1978; Kostiukov et al., 1985) and the a strain of Ranavirus from North America.

660 PEARMAN ET AL.ÐEXPERIMENTAL MODEL SYSTEM FOR EMERGING DISEASE ECOLOGY 661

While Ranavirus has a nearly global dis- METHODS tribution, differences exist among the Virus preparation strains found in amphibians; FV3 differs at the DNA sequence level from similar Ran- Frog virus 3 (provided courtesy of J. K. Jan- covich and E. W. Davidson and derived from avirus strains isolated from Europe (Hyatt the type strain) was cultured following a pro- et al., 2000) and in the principal lesions tocol developed for adenoviruses (Greber et al., observed in free-ranging and farmed pop- 1998; Suomalainen et al., 1999). The protocol ulations. In North America FV3 and close- was modified by substituting modified Eagle’s ly related viruses principally affect tad- medium for Dulbecco’s modified Eagle’s medium, using the epithelioma papillosum carp poles (Tweedel and Granoff, 1968; Green (EPC) cell line (courtesy of J. K. Jancovich and et al., 2002; but see Wolf et al., 1968), E. W. Davidson), and incubating the cultures while in Europe and Asia adults are com- at 18 C and 5% CO2. We extracted FV3 virus monly affected (Fijan et al., 1991; Cun- from infected EPC cells that showed cytopathic effect. A stock virus solution was prepared by ningham et al., 1996; Kanchanakhan, removing cellular debris through extraction 1998). with Tris saturated freon, centrifugation at Our aims were to study experimentally 1,000 ϫ G for 5 min, then removing the aque- ␮ the direct effect of viral dose on mortality, ous phase and passing it through a 0.2- m fil- and the additional effect of cannibalism on ter. Plaque assay showed this solution to have 5.5ϫ108 plaque forming units (pfu)/ml. We also mortality from FV3. Because FV3 and R. extracted uninfected cells in an identical man- latastei do not naturally co-occur, we ad- ner (i.e., a sham extract). dress whether mortality of tadpoles after exposure is an artifact resulting from virus Animals and husbandry preparation. We also used standardized bath exposure protocols that may facilitate Six populations of R. latastei were selected to span the species’ range. Two populations other studies of host-pathogen relation- were located near the western range limit (near ships of FV3 and related ranaviruses. Mor- Turin, Italy), three were in northeastern Italy, bidity and mortality of amphibians can be and one was in western Slovenia (Garner et al., induced by FV3 and similar viruses in the 2004). The east-west direction is the longest axis of distribution of R. latastei and the east- laboratory (Clark et al., 1968; Tweedel and west position of populations is the key variable Granoff, 1968; Wolf et al., 1968; Cullen et associated with genetic variation in R. latastei al., 1995). Nonetheless, few experiments (Garner et al., 2004). Twenty egg clutches from have addressed factors that influence ex- each population were collected between 10 and 17 March 2003. Multiple amplexus and pater- posure, transmission, and mortality. For nity are not reported for R. latastei, so we as- example, while little is known about trans- sume animals from a clutch are full-sib rela- mission of FV3 in nature, cannibalism has tives. All families were equalized as to devel- been hypothesized to increase transmis- opmental stage by moving hatching clutches to sion rates of diseases in fish and amphib- 4 C, a temperature that slows development but is well within the normal environmental range ians (Pfennig et al., 1991; Ariel and for this species. This is preferable to equalizing Owens, 1997) and to promote pathogen tadpoles by size because of ontogenetic chang- transmission in humans (Mead et al., es in the activity of the tadpole immune system 2003) and other vertebrates (e.g., Qureshi (Gantress et al., 2003). All tadpoles were free et al., 2000). While cannibalism may be swimming and began to exhibit feeding behavior within 24 hr of each other (Gosner stage sufficient for viral transmission in some 25, Gosner, 1960). This is the earliest stage at aquatic host-pathogen systems (e.g., Jan- which R. latastei tadpoles begin to feed. Tad- covich et al., 1997), an incremental effect poles were held in tap water aged in an open of cannibalism (or necrophagy for that container for 24 hr to allow removal of toxic volatiles (e.g., chlorine). All experiments used matter) on pathogen transmission has rare- water from a single batch of aged tap water. ly been demonstrated experimentally in Tadpoles were held on a 12 hr : 12 hr light–dark aquatic animals. cycle. 662 JOURNAL OF WILDLIFE DISEASES, VOL. 40, NO. 4, OCTOBER 2004

Experiment 1 pooling them as described above. To test for dose-dependent survival after FV3 exposure, When tadpoles reached Gosner stage 25 we observed tadpoles for the onset of feeding (Gosner, 1960) we randomly selected four fam- behavior (Gosner stage 25, Gosner, 1960) and ilies from each of the six populations. For a then assigned 120 developmentally equalized population, we pooled approximately 100 individuals randomly to 12 groups of 10 tad- seemingly healthy tadpoles from each of the poles. Two replicate groups were assigned at four families and from this pool chose 30 active random to each of the following six exposure tadpoles to represent the pooled families. We treatments: a virus solution of 4.5ϫ106 pfu/ml then randomly assigned 10 individuals to be ex- FV3 in aged tap water, one of four serial dilu- posed to FV3 and 10 to be exposed to an equiv- tions of this solution each an order of magni- alent dilution of the sham extract. The remain- tude lower in virus concentration, or a control ing 10 tadpoles were sacrificed in 0.2% aque- solution of aged water. We then conducted ex- ous MS222 (Fluka, Buchs, Switzerland) pre- posure, transfer to individual Petri dishes, and served in 10% formalin, and their feeding as described above. developmental stage (Gosner, 1960) was con- firmed by inspection with a dissecting micro- scope. We replicated this procedure for the re- Experiment 3 maining five populations. Feeder tadpoles: To examine the effect of Seventy milliliters of a solution used to ex- cannibalism of infected carcasses on infection pose the tadpoles were prepared by pipetting rates of susceptible tadpoles, we first created stock virus solution into aged tap water that was virus-exposed tadpole carcasses by allowing constantly stirred with a bar to a final virus con- randomly chosen groups of 10 developmentally centration of 2.25ϫ106 pfu/ml. Aliquots of 10 equalized tadpoles at Stage 25 (Gosner, 1960) ml were pipetted into 10-cm Petri dishes. The to swim in 10 ml of a solution of 4.5ϫ106 pfu/ groups of 10 experimental tadpoles from the six ml of FV3. Simultaneously, randomly chosen populations were transferred at random to groups of 10 tadpoles (to become control car- these Petri dishes using a piece of fiberglass casses) were held under identical conditions in screen. Three milligrams fish food mixture (a aged tap water. After 24-hr exposure, all tad- 50:50 mixture of finely ground Tetra Tabi Min poles were blotted of solution with a tissue and and Tetraphyll flakes, Tetra GmbH, Melle, transferred to new culture dishes with 20 ml Germany) was suspended in water was added fresh aged water. Virus-exposed tadpoles were to each Petri dish. An identical procedure was frozen at Ϫ80 C as soon as possible after death used for the sham exposure treatment except (generally within 6 hr). After 19 days the un- that an identical volume of sham extract was exposed tadpoles were flash frozen in liquid ni- used instead of FV3. All animals fed normally trogen and stored at Ϫ80 C for later use. during the exposure period. After 24 hr animals Experimental tadpoles: An experimental were removed from the solution and blotted group of developmentally equalized R. latastei with a disposable lab tissue, then transferred tadpoles from a single breeding site in Switzer- and held individually in 10-cm culture dishes, land (population PBPCH, Garner et al., 2003) each with 20 ml aged tap water. In this exper- was constituted by pooling siblings hatched iment and the ones that follow, animals in Petri from four egg clutches collected from nature. dishes were held on a single shelf of a climate- All individuals were at Gosner stage 25 (Gos- controlled room, at 14 C and 90% relative air ner, 1960). Fifty seemingly healthy tadpoles humidity, with a 12:12 hr dark:light schedule. were selected and housed individually in 0.5-l Tadpoles received 0.005 g fish food mixture in containers with 400 ml aged tap water, on a 12: aqueous suspension when their water was 12 hr dark:light schedule at 14 C and 90% rel- changed every 3 days. The difference in surviv- ative air humidity. Each tadpole was randomly al between the two treatments on the 30th day assigned to one of five treatments (Table 1), following exposure was tested for significance thereby creating 10 independent replicates of using a Wilcoxon signed-rank test (Siegel and each treatment. The treatments consisted of a Castellan, 1988). control for method and two crossed factors, Experiment 2 each factor having two levels. These factors were 1) exposure status of feeder tadpole car- Twenty egg masses were collected from a casses (virus exposed or freeze killed) and 2) single breeding site in southern Switzerland mode of presentation of the feeder tadpole car- (population PBPCH, Garner et al., 2003) on 15 cass to the experimental tadpole (i.e., in a cot- March 2003 and were held at ambient outdoor ton sack or free in the water). Thus, tadpoles temperature (ϳ12 C) until eclosion. We equal- assigned to the cannibal treatment had free ac- ized the developmental stage of four clutches, cess to carcasses upon which they could feed, PEARMAN ET AL.ÐEXPERIMENTAL MODEL SYSTEM FOR EMERGING DISEASE ECOLOGY 663

TABLE 1. Treatments of tadpoles in experiment 3. Each treatment was replicated 10 times in a 0.5 l container holding randomly assigned Rana latastei tadpoles. See text for additional details.

Treatment Location of code Treatment Exposure and food tadpole carcass Other comments

A Bag control Fish food No carcass Empty cotton bag B Cannibal control Frozen tadpole car- Free in container Empty cotton bag cass C Cannibal virus ex- Frog virus 3 exposed Free in container Empty cotton bag posed tadpole carcass D Isolated control Frozen tadpole car- Carcass in cotton bag cass and ﬁsh food E Isolated virus ex- Frog virus 3 exposed Carcass in cotton bag posed tadpole carcass and ﬁsh food

while tadpoles assigned to the isolated treat- were frozen promptly at Ϫ80 C. On day 22 sur- ment had a tadpole carcass suspended in their viving tadpoles were anesthetized in MS222 container in a cotton sack, which prevented and frozen. them from feeding on the carcass. An empty DNA preparation: Whole genomic DNA was bag (treatment A, Table 1) controlled for the extracted from tadpoles chosen from each effect of a submerged cotton bag on tadpole treatment in experiment 3 using the QIAamp௡ mortality. Comparison of mortality rates of an- DNA mini kit (Qiagen AG, Basel, Switzerland). imals that fed on freeze-killed carcasses versus Viral DNA was detected by polymerase chain those that ate fish food (treatments B versus D, reaction (PCR). Polymerase chain reaction mix- Table 1) controlled for the effect on mortality ture, FV3-specific primers, and conditions were of cannibalism per se. Comparison of morality as required for amplification of a portion of the rates of tadpoles exposed to an empty bag ver- FV3 major capsid protein gene (Mao et al., sus those exposed to bagged, freeze-killed tad- 1996; Gantress et al., 2003). Polymerase chain poles (treatments A versus D, Table 1) con- reaction was conducted using 35 cycles of 45 trolled for the effect on survival of having a sec denaturation at 95 C, 45 sec annealing at dead tadpole in the container. Comparison of 52 C, and 45 sec extension at 72 C. Samples mortality in groups exposed to either virus-ex- were then run on a 1% agarose gel and stained posed tadpoles or freeze-killed tadpoles (all in with Sybr Gold (Molecular Probes/Invitrogen, bags, treatments D versus E, Table 1) con- Basel, Switzerland). The size of the amplicon trolled for an effect on mortality of exposure to was determined by comparison with the Lamb- an FV3-exposed feeder tadpole carcass. Com- da—EcoRI/HindIII ladder (Qbiogene, Inc., parison of mortality in tadpoles with free access Basel, Switzerland). Frog virus 3–specific prim- to virus-killed tadpole carcasses versus mortal- ers were as follows: forward primer, 5Ј- ity when exposed to bagged, virus-killed feeder GTCTCTGGAGAAGAAGAA-3Ј, reverse prim- tadpole carcasses (treatments C versus E, Table er 5Ј-GACTTGGCCACTTATGAC-3Ј. Each re- 1) tested for cannibalism’s incremental effect action had a total volume of 30 ␮l and con- on mortality resulting from viral exposure. tained 1 ␮l of each primer (10 pmol), 2 ␮l Virus-exposed and control feeder tadpole genomic DNA (50–100 ng), 3 ␮l 1.25 mM of carcasses were randomly assigned to cannibal each deoxyribonucleotide triphosphate (Pro- and isolated treatments to eliminate the possi- mega, Wallisellen, Switzerland), 3 ␮l10ϫPCR bility of confounding the two factors. Tadpoles buffer,2UTaqDNApolymerase (Quantum- in the isolated and the bag control treatments Appligene, Illkirch, France), and 20 ␮l sterile received 0.005 g of fish food mixture every 2 double distilled water. Amplicons from eight days. All feeder tadpole carcasses, whether positive animals were sequenced at the core fa- bagged or free in containers, were removed cility for the Institutes of Zoology and Molec- (along with bags) and replaced with new ran- ular Biology, University of Zurich (Switzerland) domly chosen feeder tadpole carcasses every 2 using both forward and reverse primers. days. Experimental tadpoles were observed dai- ly, and the dates of deaths were recorded, along Statistical analysis with whether tadpoles had fed on carcasses or fish food. We observed the external appearance All randomizations conducted during the ex- of tadpoles for clinical signs. Dead tadpoles periments were done using SAS Proc Plan 664 JOURNAL OF WILDLIFE DISEASES, VOL. 40, NO. 4, OCTOBER 2004

FIGURE 1. Number of tadpoles surviving in 12 groups of 10 from six populations that span the range of Rana latastei. Each point refers to the number of surviving tadpoles alive in individual Petri dishes. (a) Tadpoles exposed to frog virus 3 for 24 hr and (b) tadpoles from the same populations when exposed to an equivalent dilution of extract from a sham extraction of uninfected cells. Source populations are shown in a legend as two-letter codes. Populations CN and SR are in northwestern Italy and have low genetic diversity at microsatellite loci. Populations SS, MT, and TG are in northeastern Italy, and GP is a population from Slovenia (see text).

(SAS Institute Inc., 1988). Selection of animals RESULTS in the studies, when not random, was haphaz- Experiment 1 ard, which, as used here, refers to selection without specific use of randomized numbers All experimental animals in the random- but without specific knowledge of the condition ly chosen sample of 10 preserved individ- of the units or animals. We used the Wilcoxon test for differences between survival curves, as uals were at Gosner stage 25. Tadpoles implemented in SAS Proc Lifetest, to examine from all populations had high mortality the statistical significance of differences be- rates when exposed to FV3 (Fig. 1a). In tween the survival of tadpoles that consumed contrast, tadpoles exposed to the sham ex- carcasses and tadpoles that did not (Allison, traction (Fig. 1b) survived significantly 1995). In analyzing data from experiment 3, we better until 30 days after exposure (Wil- tested for differences between survival in treat- ϭ ϭ ments C and E first in order to avoid problems coxon signed-rank test, n 6, c 21, associated with making multiple statistical tests Pϭ0.016). Tadpoles from two populations using the same data set. died at a faster rate than did tadpoles from the other four populations (Fig. 1a).

Experiment 2 Exposure to lower concentrations of virus resulted in increasing time to death across the range of doses used in this study (Fig. 2). The narrow standard errors sur- rounding the mean of the two replicates indicate that the exposure protocol pro- duces highly replicable mortality patterns.

Experiment 3 All tadpoles in cannibal treatments ac- FIGURE 2. Number of surviving tadpoles follow- tively fed upon feeder tadpole carcasses. ϭ ing 24-hr exposure to ﬁve concentrations (pfu Tadpoles exposed to infected carcasses plaque forming units) of frog virus 3 and one control solution of aged tap water. Each observation is the (treatments C and E) died at a higher rate mean of two replicates, with one standard error of than tadpoles in any of the control treat- the mean shown by bars. ments (Fig. 3). Cannibalism of virus-ex- PEARMAN ET AL.ÐEXPERIMENTAL MODEL SYSTEM FOR EMERGING DISEASE ECOLOGY 665

FIGURE 3. Number of surviving tadpoles under five experimental treatments. Each line represents the status of 10 individuals on successive days. Tad- poles in treatments B and C cannibalized carcasses, while tadpoles in treatments D and E were isolated from carcasses by cotton bags. Tadpoles in treatments C and E were exposed to carcasses of tadpoles that died following exposure to frog virus 3 (FV3), while tadpoles in treatments B and D were exposed to carcasses that were killed by flash freezing and had not been exposed to FV3. See text for further description of treatments. FIGURE 4. External gross lesions in tadpoles exposed to frog virus 3 (FV3). (a) Normal Rana latastei tadpole; (b) emaciation representative of that dis- posed carcasses increased the death rate of played by 12 of 20 tadpoles exposed to FV3; (c) ab- tadpoles compared to tadpoles isolated dominal edema displayed by two of 20 tadpoles exposed to FV3. from infected carcasses by cloth bags (SAS Proc Lifetest, Wilcoxon ␹2ϭ6.72, Pϭ0.01). Two clinical signs occurred in exposed tad- which clinical signs were not observed poles; they became emaciated (12 tad- (Fig. 5). Sequence of the PCR product poles) or developed edema (two tadpoles, was compared to sequences in Genbank Fig. 4). All affected tadpoles died while (http://www.ncbi.nlm.nih.gov/) using BLAST only four normal-appearing tadpoles died. and closely matched existing Ranavirus Of the 32 tadpoles alive at the end of the major capsid protein sequences. experiment, 31 appeared clinically normal DISCUSSION and one developed a condition in which the tail was flexed laterally. The association The results of our studies, including the of either of the two clinical conditions benign nature of exposure to products of (Fig. 4) with mortality was significant sham extractions (Fig. 1), FV3 dose-de- (Fisher’s exact test, two-tailed test, pendent mortality of R. latastei tadpoles PϽ0.001). (Fig. 2), transmission of FV3 from exposed When PCR primers specific for FV3 to unexposed individuals (Figs. 3, 5), and were used, DNA amplified from a sample clinical signs associated with mortality of animals exposed to FV3-exposed feeder (Fig. 4) indicate that FV3 is a transmissible tadpole carcasses (Fig. 5). Polymerase pathogen in R. latastei tadpoles. These re- chain reaction of a sample of tadpoles sults are consistent with other reports of from the control treatments did not pro- the effect of FV3 exposure on young tad- duce an amplification product when these poles (Tweedel and Granoff, 1968), evi- same primers were used. Polymerase dence of intracellular viral replication by chain reaction product was also obtained FV3 in other anuran species (Clark et al., from exposed tadpoles that died but for 1968), and reports of FV3 and closely re- 666 JOURNAL OF WILDLIFE DISEASES, VOL. 40, NO. 4, OCTOBER 2004

FIGURE 5. Agarose gel showing polymerase chain reaction (PCR) products using frog virus 3–specific primers to amplify from total DNA extractions from individual tadpoles in experiment 3 (see text and Fig. 3 for details). Tadpoles in successive lanes, from left to right, 1) treatment C, died with no visible signs of infection (i.e., a in Fig. 4); 2) treatment C, died with no signs; 3) treatment A, no signs; 4) treatment C, died with signs; 5) treatment C, died with signs; 6) treatment E, died without signs; 7) treatment E, died after developing edema; 8) treatment E, died with signs; 9) treatment E, died without signs; 10) treatment E, died after developing edema; 11) marker; 12) treatment D; 13) treatment B; 14) treatment A survivor; 15) treatment D survivor; 16) treatment B survivor; 17) treatment A survivor; 18) wild caught tadpole from Ticino, Swit- zerland; 19) marker; 20) negative PCR control. lated ranaviruses as the source of various anisms of transmission in nature will re- lesions of anurans in natural populations quire additional study, our results suggest and in aquaculture (Wolf et al., 1968; Kan- that ecological factors that influence con- chanakhan, 1998; Zhang et al., 2001). The centration of virus by an order of magni- mechanism(s) underlying pathogenesis in tude will affect the time course of tadpole the present study were not determined but mortality. Controlled exposure protocols may involve cytotoxicity, premature apo- will be essential for studying variation in ptosis, and/or cell lysis due to viral contact virulence among viral strains, variation in or replication (Chinchar, 2002; Chinchar immunocompetence as a function of pop- et al., 2003). ulation genetic composition or additional Our results (Fig. 1a) also suggest that environmental factors, and variation in populations vary in susceptibility to the ef- susceptibility among species. Understand- fects of FV3 exposure. Tadpoles from two ing these aspects of host-pathogen ecology Italian populations, CN and SR, survived will facilitate the study of Ranavirus-host less well when exposed to virus than tad- coevolution and the impact that emergent poles from the other four populations. No- pathogens may have on amphibian com- tably, these two populations are known to munity structure. have reduced levels of population genetic Use of FV3-specific primers resulted in variability in comparison with the other detection of a PCR product in exposed an- populations (Garner et al., 2003, 2004). imals in experiment 3 but not in unex- While lack of replication within populations posed ones, with confirmation of the iden- prohibits statistical inference regarding this tity of the amplicons by sequencing. Can- trend, further experiments will be conduct- nibalism upon tadpole carcasses exposed ed to explore the relationship between pop- to FV3 led to higher death rates than ex- ulation genetic diversity and resistance to posure to bagged carcasses of tadpoles that effects of novel exposure to FV3. died after virus exposure. Exposure to The second experiment (Fig. 2) dem- bagged, infected carcasses also increased onstrated that variation in levels of FV3 mortality in exposed tadpoles over back- virus exposure may result in different sur- ground rates. When considered together vival rates of tadpoles. This is important with the dose-dependency data, the results because it implies that ecological factors of these experiments suggest that mortality could affect viral acquisition and transmis- following FV3 exposure in nature may de- sion. While an understanding of the mech- pend on the degree of contact between in- PEARMAN ET AL.ÐEXPERIMENTAL MODEL SYSTEM FOR EMERGING DISEASE ECOLOGY 667 fected and uninfected individuals and may of amphibians to macroparasites may de- be a function of increasing environmental pend on environmental factors, such as concentrations of FV3. The effect of can- presence of agricultural toxins (Taylor et nibalism on disease transmission has been al., 1999; Kiesecker, 2002; Christin et al., hypothesized to influence the distribution 2003), suggesting that susceptibility to vi- of cannibalistic behavior in amphibian ral infection also may depend on the en- populations (Pfennig et al., 1991). vironment of the host. Further, our results suggest that FV3 Challenge studies that use viral strains transmission rates may vary with the com- that vary in virulence and host species that plexity of ecologic communities, where vary in susceptibility may be useful in iden- scavengers (such as crayfish or other in- tifying both molecular and epidemiologic vertebrates) or any number of environ- mechanisms that influence Ranavirus host- mental factors may act to decrease expo- pathogen ecology. The bath exposure that sure to virus-laden corpses, inactivate virus was used in experiments 1 and 2 was not particles, or concentrate them in a food invasive and did not in itself affect tadpole source for scavenging tadpoles. Virus may survival. Low variation in mortality among be spread in nature by consuming corpses replicates suggests that bath exposures or by contact with virus particles shed could be useful for assessing susceptibility upon decomposition. Other ecologic pro- to FV3 in quantitative genetic experiments. cesses may influence transmission rates by Finally, exposure protocols that produce affecting the degree to which susceptible low variance among independent replicates individuals aggregate with infected indi- are important for identifying the effects of viduals (e.g., Bjornstad et al., 2002). It is environmental factors that may influence not known whether tadpoles prefer unin- susceptibility to viral infection. Challenge fected carcasses over infected ones when protocols similar to those detailed here may feeding on conspecifics, as happens in be useful to investigate variation in suscep- some insects (Boots, 1998). tibility among populations and the genetic Challenge of a host with a viral strain basis of that variation. that does not naturally occur in the host population accurately simulates some as- ACKNOWLEDGMENTS pects of an anthropogenically mediated The authors extend their appreciation to J. emergent pathogen. Our approach here K. Jancovich and E. W. Davidson for providing does not, however, simulate aspects of access to FV3 and sharing protocols for raising pathogen emergence that result from evo- virus. K. Grossenbacher provided information on the location of R. latastei populations in lutionary changes in ranaviruses, such as Switzerland. D. Seglie and E. Marzona facili- specific mutations or selective pressures tated acquisition of R. latastei from herpetolo- that could lead to emergence in additional gists in Italy, and K. Pobljsaj facilitated work host species or result in increased viru- and collection in Slovenia. H.-U. Reyer provid- lence. These additional aspects may be ad- ed access to equipment for holding and raising tadpoles. T.W.J.G. was supported by Swiss Na- dressable in the future as the genetic basis tional Science Foundation grant 3100-64000.00 of differences in viral replication among to H.-U. Reyer. We thank members of the Gre- strains, and ultimately differences in path- ber laboratory for discussions and help with ogenicity, become better understood. Am- DNA sequencing. This work was supported by phibian species may differ in susceptibility the Swiss National Science Foundation and the Kanton Zurich, Switzerland (to U.F.G.). to FV3 (Clark et al., 1968), and some ev- idence suggests that within a single spe- LITERATURE CITED cies, susceptibility to FV3 may be related to host size, developmental stage (Tweedel ALLISON, P. D. 1995. Survival analysis using SAS: A practical guide. SAS Institute, Inc., Cary, North and Granoff, 1968), or genetic composi- Carolina, 304 pp. tion (Gantress et al., 2003). Susceptibility ARIEL, E., AND L. OWENS. 1997. Epizootic mortal- 668 JOURNAL OF WILDLIFE DISEASES, VOL. 40, NO. 4, OCTOBER 2004

ities in tilapia, Oreochromis mossambicus. Dis- 2003. Genetic depletion in Swiss populations of eases of Aquatic Organisms 29: 1–6. Rana latastei: Conservation implications. Biolog- BJORNSTAD, O. N., B. F. FINKENSTADT, AND B. T. ical Conservation 114: 371–376. GRENFELL. 2002. Dynamics of measles epidem- ,P.B.PEARMAN, AND S. ANGELONE. 2004. ics: Estimating scaling of transmission rates using Genetic diversity across a vertebrate species’ a time series SIR model. Ecological Monographs range. A test of the central-peripheral hypothe- 72: 169–184. sis. Molecular Ecology 13: 1047–1053. BOOTS, M. 1998. Cannibalism and the stage-depen- GOSNER, K. L. 1960. A simplified table for staging dent transmission of a viral pathogen of the In- anuran embryos and larvae with notes on iden- dian meal moth, Plodia interpunctella. Ecological tification. Herpetologica 16: 183–190. Entomology 23: 118–122. GREBER, U. F., M. Y. NAKANO, AND M. SUOMALAI- CHINCHAR, V. G. 2002. Ranaviruses (family Iridovir- NEN. 1998. Adenovirus entry into cells: A quan- idae): Emerging cold-blooded killers. Archives of titative fluorescence microscopy approach. In Virology 147: 447–470. Methods in molecular medicine, Vol. 21, W. S. ,L.BRYAN,J.WANG,S.LONG, AND G. D. M. Wold (ed.). Humana Press, Totowa, New Jer- CHINCHAR. 2003. Induction of apoptosis in frog sey, pp. 217–230. virus 3-infected cells. Virology 306: 303–312. GREEN, D. E., K. A. CONVERSE, AND A. K. SCHRAD- CHRISTIN, M. S., A. D. GENDRON,P.BROUSSEAU,L. ER. 2002. Epizootology of sixty-four amphibian MENARD,D.J.MARCOGLIESE,D.CYR,S.RUBY, morbidity and mortality events in the USA 1996– AND M. FOURNIER. 2003. Effects of agricultural 2001. Annals of the New York Academy of Sci- pesticides on the immune system of Rana pipiens ences 969: 323–339. and on its resistance to parasitic infection. En- GRIZZLE,J.M.,AND C. J. BRUNNER. 2003. Review vironmental Toxicology and Chemistry 22: 1127– of largemouth bass virus. Fisheries 28: 10–14. 1133. HYATT, A. D., A. R. GOULD,Z.ZUPANOVIC,A.A. CLARK, H. F., J. C. BRENNA,R.F.ZEIGEL, AND D. CUNNINGHAM,S.HENGSTBERGER,R.J.WHIT- T. KARZON. 1968. Isolation and characterization TINGTON,J.KATTENBELT, AND B. E. H. COUPAR. of the viruses from the kidneys of Rana pipiens 2000. Comparative studies of piscine and am- with renal adenocarcinoma before and after pas- phibian iridoviruses. Archives of Virology 145: sage in the red eft (Triturus viridescens). Journal 301–331. of Virology 2: 629–640. JANCOVICH, J. K., E. W. DAVIDSON,J.F.MORADO, CULLEN, B. R., L. OWENS, AND R. J. WHITTINGTON. B. L. JACOBS, AND J. P. COLLINS. 1997. Isolation 1995. Experimental infection of Australian an- of a lethal virus from the endangered tiger sala- urans (Limnodynastes terraereginae and Litoria mander Ambystoma tigrinum stebbinsi. Diseases latopalmata) with Bohle iridovirus. Diseases of of Aquatic Organisms 31: 161–167. Aquatic Organisms 23: 83–92. KANCHANAKHAN, S. 1998. An ulcerative disease of CUNNINGHAM, A. A., T. E. S. LANGTON,P.M.BEN- the cultured tiger frog, Rana tigrina, in Thailand: NETT,J.F.LEWIN,S.E.N.DRURY,R.E. Virological examination. Aquatic Animal Health GOUGH, AND S. K. MACGREGOR. 1996. Patho- Research Institute Newsletter 7: 1–2. logical and microbiological findings from inci- KIESECKER, J. M. 2002. Synergism between trema- dents of unusual mortality of the common frog tode infection and pesticide exposure: A link to (Rana temporaria). Philosophical Transactions of amphibian limb deformities in nature. Proceed- the Royal Society of London B 351: 1539–1557. ings of the National Academy of Sciences USA DASZAK, P., L. BERGER,A.A.CUNNINGHAM,A.D. 99: 9900–9904. HYATT,D.E.GREEN, AND R. SPEARE. 1999. KOSTIUKOV, M. A., Z. E. GORDEEVA,V.P.BULYCHEV, Emerging infectious diseases and amphibian N. V. NEMOVA, AND O. A. DANIOIROV. 1985. population declines. Emerging Infectious Dis- The lake frog (Rana ridibunda) one of the food eases 5: 735–748. hosts of blood-sucking mosquitoes in Tadzhiki- ,A.A.CUNNINGHAM, AND A. D. HYATT. stan—A reservoir of the West Nile Fever virus. 2001. Anthropogenic environmental change and Meditsinskaya Parizitologiya i Parazitarnye Bol- the emergence of infectious diseases in wildlife. ezni (3): 49–50. Acta Tropica 78: 103–116. KOZUCH, O., M. LABUDA, AND J. NOSEK. 1978. Iso- FIJAN, N., Z. MATASIN,Z.PETRINEC,I.VALPOTIC, lation of sindbis virus from the frog Rana ridi- AND L. O. ZWILLENBERG. 1991. Isolation of an bunda. Acta Virologica 22: 78. iridovirus-like agent from the green frog (Rana MAO, J., T. N. THAM,G.A.GENTRY,A.AUBERTIN, esculenta L.). Veterinarski Arhiv 61: 151–158. AND V. G. CHINCHAR. 1996. Short communica- GANTRESS, J., G. D. MANIERO,N.COHEN, AND J. tion: Cloning, sequence analysis, and expression ROBERT. 2003. Development and characteriza- of the major capsid protein of the iridovirus frog tion of a model system to study amphibian im- virus 3. Virology 216: 431–436. mune responses to iridoviruses. Virology 311: MEAD, S., M. P. H. STUMPF,J.WHITFIELD,J.A. 254–262. BECK,M.POULTER,T.CAMPBELL,J.B.UPHILL, GARNER, T. W. J., S. ANGELONE, AND P. B. PEARMAN. D. GOLDSTEIN,M.ALPERS,E.M.C.FISHER, PEARMAN ET AL.ÐEXPERIMENTAL MODEL SYSTEM FOR EMERGING DISEASE ECOLOGY 669

AND J. COLLINGE. 2003. Balancing selection at end-directed motilities are competing processes the prion protein gene consistent with prehistor- for nuclear targeting of adenovirus. Journal of ic kurulike epidemics. Science 300: 640–643. Cell Biology 144: 657–672. PFENNIG, D. W., M. L. G. LOEB, AND J. P. COLLINS. TAYLOR, S. K., E. S. WILLIAMS, AND K. W. MILLS. 1991. Pathogens as a factor limiting the spread 1999. Effects of malathion on disease suscepti- of cannibalism in tiger salamanders. Oecologia bility in Woodhouse’s toads. Journal of Wildlife 88: 161–166. Diseases 35: 536–541. QURESHI, T., R. E. LABES,M.LAMBETH,H.MONT- TWEEDEL, K., AND A. GRANOFF. 1968. Viruses and GOMERY,J.T.T.GRIFFEN, AND C. G. MACKIN- renal carcinoma of Rana pipiens. V. Effect of frog TOSH. 2000. Transmission of Mycobacterium virus 3 on developing frog embryos and larvae. bovis from experimentally infected ferrets to Journal of the National Cancer Institute 40: 407– non-infected ferrets (Mustelo furo). New Zealand 410. Veterinary Journal 48: 99–104. WOLF, K., G. L. BULLOCK,C.E.DUNBAR, AND M. SAS INSTITUTE INC. 1988. SAS/STAT User’s Guide, C. MCQUIMBY. 1968. Tadpole edema virus: A Release 6.03. SAS Institute Inc., Cary, North viscerotropic pathogen for anuran amphibians. Carolina. 1028 pp. Journal of Infectious Diseases 118: 253–262. SIEGEL, S., AND N. J. CASTELLAN,JR. 1988. Non- ZHANG, Q.-Y., F. XIAO, Z.-Q. LI, J.-F. GUI,J.MAO, parametric statistics for the behavioral sciences, AND V. G. CHINCHAR. 2001. Characterization of 2nd Edition. McGraw-Hill, New York, New an iridovirus from the cultured pig frog (Rana York. 330 pp. grylio) with lethal syndrome. Diseases of Aquatic SUOMALAINEN, M., M. Y. NAKANO,S.KELLER,K. Organisms 48: 27–36. BOUCKE,R.P.STIDWILL, AND U. F. GREBER. 1999. Microtubule-dependent plus- and minus Received for publication 21 November 2003.