Inactivation of Frog Virus 3 and Channel Catfish Virus by Esculentin-2P and Ranatuerin-2P, Two Antimicrobial Peptides Isolated from Frog Skin

Virology 288, 351–357 (2001) doi:10.1006/viro.2001.1080, available online at http://www.idealibrary.com on

V. G. Chinchar,*,1 Jun Wang,* Gopal Murti,† Cynthia Carey,‡ and Louise Rollins-Smith§

*University of Mississippi Medical Center, Department of Microbiology, Jackson, Mississippi 39216; †Department of Virology and Molecular Biology, St. Jude Children’s Research Hospital, Memphis, Tennessee 38105; ‡Department of Environmental, Population, and Organismic Biology, University of Colorado, Boulder, Colorado 80309; and §Department of Pediatrics and Department of Microbiology/Immunology, Vanderbilt University Medical Center, Nashville, Tennessee 37232 Received June 6, 2001; returned to author for revision June 26, 2001; accepted July 6, 2001

While it is clear that some amphibian populations have recently experienced precipitous declines, the causes of those die-offs are complex and likely involve multiple variables. One theory suggests that environmental factors may trigger events that result in depressed immune function and increased susceptibility to infectious disease. Here we examine one aspect of innate immunity in amphibians and show that esculentin-2P (E2P) and ranatuerin-2P (R2P), two antimicrobial peptides isolated from Rana pipiens, inactivate frog virus 3, a potentially pathogenic iridovirus infecting anurans, and channel catfish herpesvirus. In contrast to mammalian antimicrobial peptides, E2P and R2P act within minutes, at temperatures as low as 0°C, to inhibit viral infectivity. Moreover, these compounds appear to inactivate the virus directly and do not act by inhibiting replication in infected cells. This is the first report linking amphibian antimicrobial peptides with protection from an amphibian viral pathogen and suggests that these compounds may play a role in maintaining amphibian health. © 2001 Academic Press

INTRODUCTION stranded, ␤ sheet. (b) A second family (Bac-5, Bac-7, and PR-39) lacks disulfide bonds but is exceptionally proline/ There are over 100 known antimicrobial peptides arginine-rich, with these amino acids making up more (AMPs) synthesized by vertebrates (Lehrer et al., 1993; than 60% of residues. (c) The magainin-like family (bom- Nicolas and Mor, 1995; Boman, 1995; Simmaco et al., binins, magainins, and dermaseptins) contains linear, 1998). A variety of structurally diverse AMPs are present amphipathic, helical peptides lacking disulfide bonds. (d) in the skin secretions of several amphibian and fish A fourth family, encompassing brevinins, esculentins, species (Goraya et al., 1998, 2000; Simmaco et al., 1993). Among these, esculentin-2P (E2P) and ranatuerin-2P ranatuerins, and ranalexin, differs from defensin-like (R2P), recently isolated from Rana pipiens, were shown peptides in possessing a single disulfide bond involving to inhibit the growth of gram-positive (Staphylococcus closely spaced cysteines at the C-terminus of the pep- aureus) and gram-negative (Escherichia coli) bacteria as tide (Table 1). Although the inhibitory mechanism of well as Candida albicans (Goraya et al., 2000). In mam- AMPs is not completely understood, all four families are mals, AMPs are found in neutrophils and macrophages thought to function by forming pores within microbial (e.g., defensins) and within glands lining the digestive membranes and disrupting essential metabolic func- and respiratory tract. Similar to the presumed function of tions. However, despite their presumed common lytic E2P and R2P, mammalian AMPs are thought to function mechanism, the antimicrobial spectrum of these com- as a first line of defense against pathogenic micro- pounds varies widely. Some agents (e.g., mammalian organisms (Lehrer et al., 1993; Nicolas and Mor, 1995). defensins) show activity against multiple bacterial, fun- Collectively, AMPs are small (20–46 amino acids), ba- gal, and viral species, whereas others (e.g., bombins) sic (i.e., lysine- and arginine-rich), amphipathic peptides inhibit only a limited number of pathogens (Lehrer et al., that can be classified by structural features into at least 1993; Nicolas and Mor, 1995). Surprisingly, even struc- four distinct families. (a) The defensin-like family of turally related homologous peptides differ markedly in AMPs is distinguished by three intramolecular disulfide their activity toward a given pathogen (Lehrer et al., 1985; bonds that maintain the peptide as a rigid, triple- Goraya et al., 2000). In this report, the ability of E2P and R2P to inactivate two viruses, the iridovirus, frog virus 3 (FV3), and the herpesvirus, channel catfish virus (CCV), was investi- 1 To whom correspondence and reprint requests should be ad- dressed at Department of Microbiology, University of Mississippi Med- gated. FV3 was chosen as a representative of the family ical Center, 2500 N. State Street, Jackson, MS 39211. Fax: (601) 984- Iridoviridae, which has recently been linked to wide- 1708. E-mail: [email protected]. spread die-offs of frogs and salamanders (Cunningham

TABLE 1 As shown in Fig. 1, treatment of both FV3 and CCV with ␮ Primary Structure of E2P and R2P E2P or R2P at concentrations ranging from 5 to 500 M resulted in a marked reduction in infectivity as judged by Esculentin-2P plaque assay. CCV plaque formation was inhibited by GFS SIF RGV AKFASK GLG KDL ARL GVN LVA CKISKQ C Ͼ99% at E2P and R2P concentrations as low as 50 ␮M, Ranatuerin-2P whereas FV3 titers were inhibited by Ͼ90% at 500 ␮M GLM DTV KNV AKN LAG HML DKL KCK ITG C E2P or R2P. This level of inhibition was roughly equiva- Note. Basic amino acid residues are in boldface type. Cysteine lent to that achieved by mammalian AMPs when tested residues involved in disulfide bond formation are in italics. against a panel of enveloped viruses (Lehrer et al., 1985; Daher et al., 1986; Yasin et al., 2000). Consistent with the possibility that E2P and R2P are targeted to membranes, et al., 1996; Jancovich et al., 1997; Bollinger et al., 1999; CCV was greater than 10-fold more sensitive to these Daszak et al., 1999; Zhang et al., in press). CCV was compounds than FV3. Finally, preliminary evidence indi- chosen because this herpesvirus also infects only cold- cated that 500 ␮M E2P or R2P reduced the infectivity of blooded animals and, unlike FV3, is absolutely depen- Regina ranavirus, an iridovirus linked to widespread die- dent upon its membrane for infectivity (Abodeely et al., offs among tiger salamanders in western North America 1970; Roizman and Furlong, 1974; Davison, 1992; Booy et (Bollinger et al., 1999), by ϳ90% (data not shown). al., 1996). Previously we examined the ability of six pep- Several additional features of these compounds de- tides (dermaseptin, ranalexin, magainin-I and -II, CPF, serve comment. (a) In keeping with their assignment to and PGLa) from three amphibian species to block the the same peptide family, the inhibitory activities of E2P replication of fungal (Batrachochytrium dendrobatidis and R2P were found to be additive, rather than synergis- and Basidiobolus ranarum) and bacterial (Aeromonas hydrophila) agents associated with amphibian disease (Rollins-Smith et al., in press). Here we show that two other AMPs, E2P and R2P, markedly inhibited FV3 and CCV plaque formation (i.e., infectivity) at concentrations ranging from 5 to 500 ␮M. Furthermore, we provide evidence that the peptides directly inactivated virus rather than blocking replication in virus-infected cells. This is the first report demonstrating that amphibian AMPs inhibit the replication of viruses pathogenic to cold-blooded vertebrates and suggests that these peptides may play a role in protecting amphibians from viral disease.

RESULTS Effect of E2P and R2P on FV3 and CCV infectivity To determine whether one or both of the amphibian skin-derived peptides possessed antiviral activity, FV3 and CCV were incubated in the presence of various concentrations of E2P or R2P for1hat26°C. After1hthe samples were diluted 100-fold and the residual infectivity was measured by plaque assay. FV3 was chosen as a test virus because it is the type species of the genus Ranavirus (family Iridoviridae) (Willis et al., 1985; Williams et al., 2000). Several members of this genus have been implicated in die-offs of amphibians and fish (Daszak et al., 1999). However, because FV3 preparations are mix- tures of enveloped and nonenveloped virions, both of which are infectious (Braunwald et al., 1979), and be- FIG. 1. Effect of E2P and R2P on FV3 and CCV Infectivity. Ten- cause amphibian skin peptides such as E2P and R2P are microliter aliquots of FV3 (top) or CCV (bottom) were mixed with an thought to act by disrupting membranes rather than by equal volume of peptide or water at the indicated concentrations and incubated for1hat26°C. Subsequently, the virus titer was determined altering protein–protein interactions (i.e., viral capsids), by plaque assay and expressed as percentage of control. The data we also tested the effect of these compounds on CCV, a represent the mean of three to seven repetitions; error bars indicate 1 fish pathogen that requires a viral envelope for infectivity. standard deviation. ANTIMICROBIAL PEPTIDES ISOLATED FROM FROG SKIN 353

tic (data not shown). (b) Because some AMPs have cytolytic as well as an antimicrobial activity, the integrity of fathead minnow (FHM) and channel catfish ovary (CCO) cells was visually monitored after exposure to E2P and R2P. We found that neither FHM nor CCO cell monolayers were lysed by peptide concentrations as high as 10 ␮M, indicating that at inhibitory concentrations E2P and R2P were not cytolytic (data not shown). (c) To rule out that possibility that a random peptide might possess antimicrobial activity, we tested the inhibitory ability of ␣ two MAP peptides { MAP2, [(TNGTWPTNEYKEDA)2 K]2 ␤ KA and MAP, [(KLTIDEDNYETYV)2 K]2 KA} that contain epitopes for catfish T cell receptor ␣ and ␤ chains (Pos- nett et al., 1988). At a concentration of 75 ␮M, ␣MAP2 and ␤MAP lowered the CCV titer to ϳ60% of control. In ␮ FIG. 2. Effect of increased incubation time on inactivation induced by contrast, 50 M E2P reduced the titer of CCV to 0.2% of E2P and R2P. Ten-microliter aliquots of FV3 or CCV were incubated with the control value. (d) Finally, to determine whether virus E2P or R2P and incubated at 26°C for the indicated periods of time. that replicated after AMP treatment represented a pep- Residual infectivity was determined, and the results are expressed tide-resistant population, random plaques were selected relative to the titer of virus incubated in the absence of peptide (shown ␮ from assay plates containing virus exposed to 100 ␮M here as the 0-min time point). For FV3, treatment was with 50 M E2P or R2P; for CCV treatment was with 5 ␮M E2P or R2P. The data shown E2P and R2P and retested for peptide sensitivity. Virus depict a representative experiment. prepared from four plaques isolated after treatment with E2P, and from a single plaque isolated after treatment with R2P, was as sensitive to 100 ␮M peptide as the original virus stock. This result suggests that the residual cating that viral inactivation occurred over a wide tem- infectivity seen after treatment with E2P or R2P is not due perature range (i.e., 0–26°C). While it is likely that inac- to a minor population of genetically resistant virions, but tivation occurred at low temperatures it is also possible may be due to virus aggregation that prevents complete that binding of the peptide to the virus took place rapidly virus inactivation. in the cold, but that inactivation occurred only later when To determine whether the treatment conditions used the virus was added to the cells and exposed to higher above were optimal for virus inhibition, FV3 and CCV temperatures. were incubated with E2P and R2P for varying periods of To strengthen the suggestion that the peptides re- time (10–180 min) and over a temperature range of duced viral infectivity by interacting directly with the 0–26°C. In these studies, peptide concentrations were virion, rather than by influencing replicative events in ϳ chosen that resulted in an 50–80% reduction in infec- virus-infected cells, we determined whether pre- or post- ␮ tivity. As shown in Fig. 2, incubation of FV3 with 50 M treatment of virus-infected cells with peptide blocked ϳ E2P for 10 min at 26°C resulted in an 80% reduction in infectivity. Accordingly, CCO cell monolayers were incu- infectivity. Incubation for an additional 50 or 170 min bated with 10 ␮M E2P or R2P either for 1 h before resulted in at most a further twofold reduction in infec- addition of virus or 1 h after addition of virus. Although tivity. Approximately equal levels of inhibition were seen this peptide concentration is lower than those used in following treatment of FV3 with 50 ␮M R2P. Similar re- most of the experiments described above, it was chosen sults were seen with CCV where 5 ␮M E2P reduced CCV because it is equivalent to the highest concentration of infectivity by ϳ80% after 10 min and by ϳ90% after 180 peptide present when the diluted virus sample was min. Although in the experiment shown in Fig. 2 R2P added to cell monolayers during a plaque assay. If the resulted in only a 40% reduction in titer, levels of inhibition similar to those seen with E2P were observed in peptide perturbs the cell membrane in a manner that other experiments. Taken together these results suggest blocks virus entry, or if peptide–cell interaction triggers that E2P and R2P acted rapidly and that most of their events that inhibit virus replication, then treatment of antiviral effect was determined within the first 10 min of CCO cells prior to, or following, CCV adsorption should interaction. result in a lower level of virus replication. As shown in Because temperature influences the fluidity of mem- Table 2, pretreatment of cells prior to virus addition, or branes, the impact of temperature on inhibition was also posttreatment of cells following virus adsorption, had determined. CCV was incubated with 12.5 ␮M E2P or only a minimal effect on virus infectivity, suggesting that R2Pfor1hat0,18,and26°C. As shown in Fig. 3A, a the peptides exerted their inhibitory effect on the virus marked drop in CCV infectivity was seen at all tempera- directly, perhaps by disrupting the viral envelope, rather tures. Similar results were seen with FV3 (Fig. 3B), indi- than by altering virus replication in the infected cell. 354 CHINCHAR ET AL.

enveloped, indicating that the sensitivity of these preparations to inactivation was not due to an increased number of enveloped particles. These results are consistent with E2P and R2P interacting with, and disrupting, the CCV envelope and/or nucleocapsid and support the no- tion that AMPs directly inactivate viral infectivity.

DISCUSSION The results presented above demonstrate that E2P and R2P decrease FV3 and CCV infectivity. This is the first demonstration that amphibian skin peptides inactivate physiologically relevant amphibian viral pathogens, and it is one of the few studies showing that AMPs inactivate viruses. Our results are consistent with those using mammalian and amphibian AMPs to inactivate mammalian viruses (Lehrer et al., 1985; Daher et al., 1986; Egal et al., 1999; Yasin et al., 2000) and suggest that these compounds may play a role in preventing viral infections in amphibians. Moreover, in contrast to previous studies, we found that E2P and R2P were effective against both enveloped and nonenveloped virions and that they acted rapidly at both low and ambient temperatures. As shown in Fig. 1, CCV is 10–20 times more sensitive to peptide-induced inactivation than FV3. However, it is not known whether the increased sensitivity of CCV to E2P and R2P reflects inherent differences in the susceptibility of the two viruses or is a reflection of the high percentage of nonenveloped virions within the FV3 prep- FIG. 3. Effect of temperature on peptide activity. Equal volumes of aration that may be less susceptible to an agent that virus and peptide were incubated for1hattheindicated temperatures and residual infectivity was determined. FV3 was incubated with 500 primarily disrupts biological membranes. It is estimated ␮M peptide, whereas CCV was incubated with peptide at a final that 80% of FV3 virions remain cell associated and are concentration of 12.5 ␮M. The data shown depict a representative likely nonenveloped (Braunwald et al., 1979; Goorha and experiment.

Electron microscopy TABLE 2 To directly confirm the effect of E2P and R2P on viral Effect of Exposure to Peptide Pre- or Post-CCV Adsorption infectivity, CCV and FV3 were incubated with peptide as Virus titer (percentage of control) described above and visualized by transmission electron microscopy. In the presence of water alone (control), Agenta Treatment Experiment 1b Experiment 2 Expectedc numerous intact, enveloped CCV particles were observed (Fig. 4). Following treatment of CCV with either 50 Water Pre- 100% 100% 100% ␮M E2P or R2P, fewer particles, both total and intact, Water Post- 100% 100% — E2P Pre- 96% 89% 28% were detected, suggesting that AMPs adversely affected E2P Post- 78% 91% — viral integrity. From these results it appears that the R2P Pre- 95% n.d.d 43% AMPs lysed not only the viral envelope, but also affected R2P Post- 68% n.d. — the stability of the nucleocapsid. In contrast to the a Cell monolayers grown on 35-mm dishes were incubated with 0.1 marked effect of E2P and R2P on CCV, treatment of FV3 ml of DMEM containing 4% FCS and 10 ␮M E2P or R2P for 1 h either with 500 ␮M E2P or R2P appeared to have no detectable directly before (“pre-”) or immediately following (“post-”) virus adsorp- effect on virion morphology or integrity. Although subtle tion. Monolayers were subsequently overlaid with DMEM containing changes in the appearance of treated FV3 virions were Noble’s agar and incubated for 7 days to allow plaque development. b noted, their significance was unclear since similar The values shown here are virus titers expressed as a percentage of control (i.e., titer following incubation of virus with water alone). changes were also seen, albeit to a lesser extent, in c The value shown here is the expected titer following incubation of control reactions (data not shown). Furthermore, Ͼ90% virus and peptide for1hat26°C. of virions present within the FV3 preparation were non- d n.d. denotes “not determined.” ANTIMICROBIAL PEPTIDES ISOLATED FROM FROG SKIN 355

FIG. 4. Transmission electron microscopy of control and peptide-treated CCV. CCV was incubated for1hat26°C in 50 ␮M E2P or R2P, and virions were visualized by transmission electron microscopy as described under Materials and Methods. Magnification, 45,000ϫ.

Granoff, 1979). Thus the 90% reduction in FV3 plaque events in virus-infected cells and blocking one or more formation seen after treatment with 500 ␮M E2P is functions essential for virus replication, then AMPs higher than expected if the peptide were acting only added directly to the cells should be as inhibitory as against enveloped forms of the virus. In view of this, it is AMPs present in the virus plus peptide mixture. The possible that E2P disrupts the internal lipid membrane observation that the latter are more inhibitory than the that underlies the FV3 capsid (Willis and Granoff, 1974; former suggests that peptide–virus interaction leads to Williams et al., 2000). If this hypothesis is correct, the direct inactivation of virus infectivity. This result is con- higher susceptibility of CCV to peptide inhibitors may sistent with findings in a related system, wherein the reflect the ease with which its essential membrane (i.e., defensin HNP-1 was shown to directly bind HSV-1 (Daher the viral envelope) is degraded. In contrast, the relative et al., 1986). Electron microscopic investigation of pep- resistance of FV3 may be due to the decreased acces- tide-treated virions supports this view and shows degra- sibility of the inner membrane of FV3 and other iridovi- dation of the CCV envelope following peptide treatment. ruses. In contrast there were only minor changes in FV3 mor- In support of the above hypothesis, several observa- phology after peptide treatment, suggesting that if pep- tions are relevant. First, Braunwald et al. (1985) specu- tides disrupted the internal FV3 membrane, its dissolu- lated that the inner membrane of FV3 is essential for tion did not adversely affect virion morphology. Similar to entry by nonenveloped FV3 particles. Second, the inter- the assembly of African swine fever virus capsids, it is nal membrane is essential for infectivity since lipid sol- possible that the internal membrane of FV3 serves as a vents render virions noninfectious (Willis and Granoff, scaffold, or nucleation point, for capsid formation (An- 1974). Third, Broo et al. (2001) showed that alkylating dres et al., 1998). Alternatively, lipid and capsid protein agents were able to permeate the capsid of a nonenvel- interaction may result in the formation of complexes that oped virus, flockhouse virus (FHV), and inactivate viral self-assemble around viral DNA cores and produce ma- nucleic acid. This finding is consistent with work by ture FV3 virions (Willis and Granoff, 1974). Thus disrup- Bothner et al. (1998) showing that internal regions of the tion of the internal membrane within mature virions may FHV virion were readily susceptible to proteolytic degra- not markedly affect virion morphology, but may decrease dation. Taken together these results suggest that the virion infectivity. viral capsid is a dynamic structure in which internal In contrast to results with MCP-1 and MCP-2 (Lehrer et protein regions can translocate to the capsid surface and al., 1985), two peptides isolated from rabbit granulocytes small molecules can penetrate to the interior. If the same and alveolar macrophages, and HNP-1, a defensin iso- phenomena occur with FV3, then AMPs may be able to lated from human neutrophils (Daher et al., 1986), we did penetrate the viral capsid and disrupt the internal mem- not observe a marked decrease in viral infectivity with brane. As suggested by the plaque reduction (Fig. 1) and increased incubation time, nor did we see enhanced electron microscopic data, this may result in a loss of virus inactivation at higher temperatures (Figs. 2 and 3). infectivity, without an overt change in virion morphology. At present it is not known whether this reflects differ- Additional experiments, in which peptide was added ences in the peptides (i.e., defensin family vs esculentin/ to cells 1 h before infection, or 1 h after infection, also ranatuerin family) or in the target viruses. However, since support the view that E2P and R2P function by directly E2P and R2P function in amphibians, it would not be inactivating virions. If the peptides acted by altering host surprising if they retain activity at low temperatures. For cell membranes and disrupting virus entry or subsequent example, since amphibians are often at risk of infection replicative events, or if the peptides acted by perturbing during spring spawning, it would be reasonable that 356 CHINCHAR ET AL.

antimicrobial peptides maintain activity at temperatures Peptides lower than those optimal for mammalian-derived pep- The inhibitors E2P and R2P (Table 1) were originally tides. isolated from the skin of R. pipiens (Goraya et al., 2000). Coupled with previous work, the current study indi- The peptides used in this study were commercially syn- cates that E2P and R2P inactivated both enveloped and thesized (Sigma/Genosys, Woodland, TX), dissolved in nonenveloped viral pathogens in vitro. In the wild, sterile water at a concentration of 1 mM, and stored at aquatic animals are exposed to infectious agents by Ϫ70°C until use (Rollins-Smith et al., in press). ingestion, respiration, and/or skin contact. Whether antimicrobial peptides, either induced or constitutively Plaque inhibition assay present, are sufficient to block infectivity on mucosal and skin surfaces, and thus influence the outcome of infec- To determine whether E2P and R2P inactivated FV3 tion, remains to be determined. Tyler et al. (1992) showed and CCV, plaque reduction assays were performed. ␮ that mild electrical stimulation, or treatment with adren- Briefly, 10 l of either FV3 or CCV (containing between 6 8 ␮ ergic agents, led to the release of ϳ50 mg of peptides 10 and 10 PFU/ml) was mixed with either 10 l peptide ␮ from the skin of frogs. While these agents may be rapidly at the indicated concentration or 10 l sterile deionized diluted in the surrounding aquatic environment, it is pos- water (control) and incubated at 26°C. After 1 h, samples sible that their effective concentration on the surface of were diluted with 1 ml of DMEM containing 4% FCS. Subsequently, samples were serially diluted 10-fold in the skin or on mucosal membranes may be sufficient to DMEM/4% FCS, and 0.1-ml volumes of the appropriate inactivate pathogenic organisms. Furthermore, it is not dilutions were plated in duplicate onto FHM (for FV3) or clear whether environmental insults (e.g., spawning CCO (for CCV) monolayers grown on 35-mm tissue cul- stress, pollutants, UV irradiation, crowding) lower the ture dishes or 6-well tissue culture plates. Following 1 h level of antimicrobial peptides and thus increase sus- of incubation at 26°C to permit virus adsorption, the ceptibility to certain pathogens. For example, FV3 has monolayers were overlaid with DMEM containing 2% been viewed as a relatively nonpathogenic virus in wild FCS and 1% Noble’s agar and incubated for 5–7 days at R. pipiens (Goorha and Granoff, 1978). However, die-offs, 26°C. At that time, the cultures were stained with 2.5 ml attributable to an FV3-like virus, have recently occurred of DMEM containing neutral red and incubated at 26°C among cultured frogs (R. grylio and R. tigrina) in China for at least 6 h, and the plaques were counted. In some and Thailand (Kanchanakhan, 1998; Zhang et al., in experiments involving FV3, residual infectivity was deter- press), suggesting that an agent of low pathogenicity in mined by a TCID50 assay rather than by plaque assay. healthy animals can trigger high levels of morbidity and mortality in “stressed” animals. Clearly, elucidating inter- Transmission electron microscopy actions among hosts, pathogens, peptides, and environment will be required to determine the role of AMPs in Equal volumes of virus and peptide were mixed, incu- disease prevention. bated at 26°C for 1 h, and visualized by transmission electron microscopy. FV3 was exposed to 500 ␮M E2P or R2P; CCV was treated with 50 ␮M E2P or R2P. Treated MATERIALS AND METHODS virus preparations were adsorbed to freshly glow-dis- charged, carbon-coated grids, negatively stained with 2% Cells and virus phosphotungstic acid, and observed in a Phillips 301 FV3, originally obtained from Allan Granoff, St. Jude electron microscope at 60 kV (Coronell et al., 1999). Children’s Research Hospital (Memphis, TN), was propagated in FHM cells grown in sealed 175-cm2 ACKNOWLEDGMENTS

tissue culture flasks at 26°C in Eagle’s minimum es- This work was partially supported by an IRCEB award from the sential medium containing Hank’s salts (HMEM) and National Science Foundation (IBN-9977063; J.P. Collins, P.D.) and an 2–4% fetal calf serum (FCS). CCV (ictalurid herpesvirus award from the U.S. Department of Agriculture (99-35204-7844; V.G. 1) was purchased from the American Type Culture Chinchar and N.W. Miller, Co-PIs). Collection (VR-665) and grown in CCO cells in closed flasks at 26°C using Dulbecco’s modified Eagle’s me- REFERENCES dium (DMEM) with 4% FCS. For plaque assay, both Abodeely, R. A., Lawson, L. A., and Randall, C. C. (1970). Morphology FHM and CCO cells were grown at 26°C in a humid- and entry of enveloped and de-enveloped equine abortion (herpes)

ified CO2 incubator using DMEM with 4% FCS. FHM virus. J. Virol. 5, 513–523. cells were purchased from the American Type Culture Andres, G., Garcia-Escudero, R., Simon-Mateo, C., and Vinuela, E. Collection (CCL-42); CCO cells were obtained from Dr. (1998). African swine fever virus is enveloped by a two-membraned collapsed cisterna derived from the endoplasmic reticulum. J. Virol. L. Hanson, Mississippi State University, College of 872, 8988–9001. Veterinary Medicine. Bollinger, T. K., Mao, J., Schock, D., Brigham, R. M., and Chinchar, V. G. ANTIMICROBIAL PEPTIDES ISOLATED FROM FROG SKIN 357

(1999). Pathology, isolation, and preliminary molecular characteriza- Rana luteiventris, Rana berlandieri, and Rana pipiens. Eur. J. Bio- tion of a novel iridovirus from tiger salamanders in Saskatchewan. J. chem. 267, 894–900. Wildlife Dis. 35, 413–429. Jancovich, J. K., Davidson, E. W., Morado, J. F., Jacobs, B. L., and Collins, Boman, H. G. (1995). Peptide antibiotics and their role in innate immu- J. P. (1997). Isolation of a lethal virus from the endangered tiger nity. Annu. Rev. Immunol. 13, 61–92. salamander Ambystoma tigrinum stebbinsi. Dis. Aquat. Org. 31, 161– Booy, F. P., Trus, B. L., Davison, A. J., and Steven, A. C. (1996). The capsid 167. architecture of channel catfish virus, an evolutionarily distant her- Kanchanakhan, S. (1998). An ulcerative disease of the cultured tiger pesvirus, is largely conserved in the absence of discernable se- frog, Rana tigrina, in Thailand: Virological examination. AAHRI News- quence homology with herpes simplex virus. Virology 215, 134–141. lett. 7, 1–2. Bothner, B., Dong, X. F., Bibbs, L., Johnson, J. E., and Siuzdak, G. (1998). Lehrer, R. I., Daher, K., Ganz, T., and Selsted, M. E. (1985). Direct Evidence of viral capsid dynamics using limited proteolysis and inactivation of viruses by MCP-1 and MCP-2, natural peptide antibi- mass spectrometry. J. Biol. Chem. 273, 673–676. otics from rabbit leukocytes. J. Virol. 54, 467–472. Braunwald, J., Nonnenmacher, H., and Tripier-Darcy, F. (1985). Ultra- Lehrer, R. I., Lichtenstein, A. K., and Ganz, T. (1993). Defensins: Antimi- structural and biochemical study of frog virus 3 uptake by BHK-21 crobial and cytotoxic peptides of mammalian cells. Annu. Rev. Im- cells. J. Gen. Virol. 66, 283–293. munol. 11, 105–128. Braunwald, J., Tripier, F., and Kirn, A. (1979). Comparison of the prop- Nicolas, P., and Mor, A. (1995). Peptides as weapons against microor- erties of enveloped and naked frog virus 3 (FV3) particles. J. Gen. ganisms in the chemical defense system of vertebrates. Annu. Rev. Virol. 45, 673–682. Microbiol. 49, 277–304. Broo, K., Wei, J., Marshall, D., Brown, F., Smith, T. J., Johnson, J. E., Posnett, D. N., McGrath, H., and Tam, J. P. (1988). A novel method for Schneemann, A., and Siuzdak, G. (2001). Viral capsid mobility: A producing anti-peptide antibodies. J. Biol. Chem. 263, 1719–1725. Roizman, B., and Furlong, D. (1974). The replication of herpes viruses. In dynamic conduit for inactivation. Proc. Natl. Acad. Sci. USA 98, “Comprehensive Virology” (H. Fraenkel-Conrat and R. R. Wagner, 2274–2277. Eds.), Vol. 3, pp. 229–403. Plenum, New York. Coronell, E. C., Murti, K. G., Takimoto, T., and Portner, A. (1999). Human Rollins-Smith, L. A., Doersam, J. K., Longcore, J. E., Taylor, S. K., Sham- parainfluenza virus type 1 matrix and nucleocapsid protein genes blin, J. C., Carey, C., and Zasloff, M. A. Antimicrobial peptide defenses transiently expressed in mammalian cells induce the release of against pathogens associated with global amphibian declines. Dev. virus-like particles containing nucleocapsid-like structures. J. Virol. Comp. Immunol., in press. 73, 7035–7038. Simmaco, M., Mignogna, G., and Barra, D. (1998). Antimicrobial pep- Cunningham, A. A., Langton, T. E. S., Bennett, P. M., Lewin, J. F., Drury, tides from amphibian skin: What do they tell us? Biopolymers 47, S. E. N., Gough, R. E., and MacGregor, S. K. (1996). Pathological and 435–450. microbiological findings from incidents of unusual mortality of the Simmaco, M., Mignogna, G., Barra, D., and Bossa, F. (1993). Novel common frog (Rana temporaria). Philos. Trans. R. Soc. London B 351, antimicrobial peptides from skin secretion of the European frog Rana 1539–1557. esculenta. FEBS Lett. 324, 159–161. Daher, K. A., Selsted, M. E., and Lehrer, R. I. (1986). Direct inactivation Tyler, M. J., Stone, D. J. M., and Bowie, J. H. (1992). A novel method for of viruses by human granulocyte defensins. J. Virol. 60, 1068–1074. the release and collection of dermal, glandular secretions from the Daszak, P., Berger, L., Cunningham, A. A., Hyatt, A. D., Green, D. E., and skin of frogs. J. Pharmacol. Toxicol. Methods 28, 199–200. Speare, R. (1999). Emerging infectious diseases and amphibian pop- Williams, T., Chinchar, G., Darai, G., Hyatt, A., Kalmakoff, J., and Seligy, ulation declines. Emerging Infect. Dis. 5, 735–748. V. (2000). Family Iridoviridae. In “Virus Taxonomy: Classification and Davison, A. J. (1992). Channel catfish virus: A new type of herpesvirus. Nomenclature of Viruses, 7th Report of the International Committee Virology 186, 9–14. on Taxonomy of Viruses” (MHV van Regenmortel et al., Eds.), pp. Egal, M., Conrad, M., MacDonald, D. L., Maloy, W. L., Motley, M., and 167–182, Academic Press, San Diego. Genco, C. A. (1999). Anti-viral effects of synthetic membrane-active Willis, D. B., and Granoff, A. (1974). Lipid composition of frog virus 3. peptides on herpes simplex virus. Int. J. Antimicrobiol Agents 13, Virology 61, 256–269. 57–60. Willis, D. B., Goorha, R., and Chinchar, V. G. (1985). Macromolecular Goorha, R., and Granoff, A. (1979). Icosahedral cytoplasmic deoxyribo- synthesis in cells infected by frog virus 3. Curr. Top. Microbiol. viruses. In “Comprehensive Virology” (H. Fraenkel-Conrat and R. R. Immunol. 116, 78–106. Wagner, Eds.), pp. 347–399. Plenum, New York. Yasin, B., Pang, M., Turner, J. S., Cho, Y., Dinh, N., Waring, A. J., Lehrer, Goraya, J., Knoop, F. C., and Conlon, J. M. (1998). Ranatuerins: Antimi- R. I., and Wagar, E. A. (2000). Evaluation of the inactivation of infec- crobial peptides isolated from the skin of the American bullfrog, Rana tious herpes simplex virus by host-defense peptides. Eur. J. Clin. catesbeiana. Biochem. Biophys. Res. Commun. 250, 589–592. Microbiol. Infect. Dis. 19, 187–194. Goraya, J., Wang, Y., Zhihong, L., O’Flaherty, M., Knoop, F. C., Platz, J. E., Zhang, Q.-Y., Xiao, F., Li, Z.-Q., Gui, J.-F., Mao, J., and Chinchar, V. G. and Conlon, J. M. (2000). Peptides with antimicrobial activity from four Characterization of an iridovirus from the cultured pig frog (Rana different families isolated from the skins of the North American frogs grylio) with lethal syndrome. Dis. Aquat. Org., in press.