Elizabeth W. Davidson

Pathogenicity and transmission of chytridiomycosis in tiger salamanders (Ambystoma tigrinum)

Abstract A chytrid fungus, Batrachochytrium dendrobatidis, was found in salamanders, Ambystoma tigrinum stebbinsi, collected in southern Arizona, USA. The chytrid was isolated and cultured, and Kochs postulates were satisfied by infection of metamorphosed salamanders with pure culture and subsequent reisolation from these salamanders. We used the salamander strain and a strain isolated from lowland leopard frogs in Arizona, Rana yavapaiensis, to infect metamorphosed A. tigrinum, R. yavapaiensis, and R. boylii. All three species became infected, but none of the infected salamanders died within 60 days, and mortality of infected frogs did not differ significantly from controls, although sample size was small. Chytrid infection could not be detected by light histology in most of the infected frogs and one of the infected salamanders 60 days after infection. To date, there are three records of chytridiomycosis in salamanders on websites; ours is the first complete report of occurrence and pathology of chytridiomycosis in field-collected North American salamanders. Our results also demonstrate that chytridiomycosis does not always lead to mortality. Individuals within a species vary in susceptibility to infection, animals appear to recover from the infection, and syntopic salamanders and frogs may act as reciprocal pathogen reservoirs for chytrid infections.

Evidence for emergence of an amphibian iridoviral disease because of human-enhanced spread

Our understanding of origins and spread of emerging infectious diseases has increased dramatically because of recent applications of phylogenetic theory. Iridoviruses are emerging pathogens that cause global amphibian epizootics, including tiger salamander (Ambystoma tigrinum) die‐offs throughout western North America. To explain phylogeographical relationships and potential causes for emergence of western North American salamander iridovirus strains, we sequenced major capsid protein and DNA methyltransferase genes, as well as two noncoding regions from 18 geographically widespread isolates. Phylogenetic analyses of sequence data from the capsid protein gene showed shallow genetic divergence (< 1%) among salamander iridovirus strains and monophyly relative to available fish, reptile, and other amphibian iridovirus strains from the genus Ranavirus, suggesting a single introduction and radiation. Analysis of capsid protein sequences also provided support for a closer relationship of tiger salamander virus strains to those isolated from sport fish (e.g. rainbow trout) than other amphibian isolates. Despite monophyly based on capsid protein sequences, there was low genetic divergence among all strains (< 1.1%) based on a supergene analysis of the capsid protein and the two noncoding regions. These analyses also showed polyphyly of strains from Arizona and Colorado, suggesting recent spread. Nested clade analyses indicated both range expansion and long‐distance colonization in clades containing virus strains isolated from bait salamanders and the Indiana University axolotl (Ambystoma mexicanum) colony. Human enhancement of viral movement is a mechanism consistent with these results. These findings suggest North American salamander ranaviruses cause emerging disease, as evidenced by apparent recent spread over a broad geographical area.

INTRASPECIFIC RESERVOIRS: COMPLEX LIFE HISTORY AND THE PERSISTENCE OF A LETHAL RANAVIRUS

Virulent parasites cannot persist in small host populations unless the parasite also has a reservoir host. We hypothesize that, in hosts with complex life histories, one stage may act as an intraspecific reservoir for another. In amphibians, for example, larvae often occur at high densities, but these densities are ephemeral and fixed in space, whereas metamorphs are long-lived and vagile but may be very sparse. Parasite persistence is unlikely in either stage alone, but transmission between stages could maintain virulent parasites in seasonally fluctuating amphibian populations. We examined this hypothesis with a lethal ranavirus, Ambystoma tigrinum virus (ATV), that causes recurrent epidemics in larval tiger salamander populations, but which has no reservoir host and degrades quickly in the environment. Although exposure to ATV is generally lethal, larvae and metamorphs maintained sublethal, transmissible infections for >5 mo. Field data corroborate the persistence of ATV between epidemics in sublethally infected metamorphs. Three-quarters of dispersing metamorphs during one epidemic were infected, and apparently healthy metamorphs returning to breed harbored ATV infections. Our results suggest that larval epidemics amplify virus prevalence and sublethally infected metamorphs (re)introduce the virus into uninfected larval populations. Intraspecific reservoirs may explain the persistence of parasites in and declines of small, isolated amphibian populations.

Insecticidal activity of the crystalline parasporal inclusions and other components of the Bacillus sphaericus 1593 spore complex

Abstract Bacillus sphaericus 1593 spore complexes were disrupted by French pressure cell. Fractions recovered from centrifugation of these complexes on 10–50% NaBr gradients were assayed against mosquito larvae and examined using the electron microscope. Crystalline parasporal inclusions were concentrated in the fraction of highest insecticidal activity. The fractions containing sporangium, exosporium, and spores also were insecticidal at a lower level. These results indicate that the crystals are the major source of insecticidal toxin in strains of B sphaericus which produce them.

Variation in binding of Bacillus sphaericus toxin and wheat germ agglutinin to larval midgut cells of six species of mosquitoes

Bacillus sphaericus toxin labeled with fluorescein isothiocyanate was readily ingested by Culex pipiens, Aedes aegypti, Anopheles stephensi, Anopheles gambiae, Anopheles quadrimaculatus, and Anopheles albimanus larvae. Fluorescent toxin bound to the luminal cell surface in discrete regions of the posterior midgut and gastric caecum in C. pipiens. In Anopheles spp., toxin bound in a variable pattern to these structures and central and anterior midgut as well. The toxin did not bind to midgut cells of A. aegypti. The toxin was internalized in bright fluorescent vesicles in C. pipiens, but was not internalized in Anopheles spp. and appeared to be weakly bound in these larvae, leaking rapidly from the gut surface. The lectin, wheat germ agglutinin, which interferes with binding of the B. sphaericus toxin, bound to the posterior midgut and gastric caecum of all species, but was not internalized. These results suggest that the sugar moiety of the receptor is not solely responsible for specificity of this toxin, and that binding to Culex spp. midgut cells may be highly specific and of high affinity, whereas binding to Anopheles spp. cells may be nonspecific and/or of low affinity.

Isolation and assay of the toxic component from the crystals ofBacillus thuringiensis var.israelensis

The 25-Kdal fragment of the 28-Kdal toxic protein extracted fromBacillus thuringiensis var.israelensis crystals was found to be responsible for the insecticidal, cytolytic, hemolytic, and mouse-lethal activities of the crude toxin extract. This activity was found to have no relation to the hemolysin produced by other strains ofB. thuringiensis. This protein was rich in the amino acids Asp and Glu, but did not contain Cys.

Enzymatic activation of the Bacillus sphaericus mosquito larvicidal toxin

Bacillus sphaericus insecticidal toxin was activated by trypsin, α-chymotrypsin, and mosquito gut homogenates to a form which was cytotoxic to cultured mosquito cells. Gut extracts from highly resistant Aedes aegypti larvae were as effective in activating this toxin as extracts from the highly sensitive species Culex quinquefasciatus. Activation altered the apparent molecular weight of the toxin by ca. 2–4 kDa.

Pathogenesis of Bacillus sphaericus strain SSII-1 infections in Culex pipiens quinquefasciatus ( = C. pipiens fatigans) larvae

Abstract Numbers of viable bacteria in second instar Culex pipiens quinquefasciatus larvae were determined following ingestion of pathogenic strain SSII-1 and nonpathogenic Bacillus sphaericus . Numbers of nonpathogenic B. sphaericus recovered from larvae declined rapidly after cessation of feeding, as did numbers of pathogenic SSII-1 cells fed at LD 20 dosage. When pathogenic cells were fed at LD 70 dosage, the number of B. sphaericus in larvae increased following initial decline. When chloroformtreated SSII-1 cultures, in which all bacteria except spores were dead, were fed at LD 10 and LD 98 dosages, no viable B. sphaericus were recovered from larvae. In all SSII-1 treatments, other bacterial flora multiplied rapidly in larvae following onset of mortality; the role of this multiplication in the pathogenesis was not determined. It is proposed that toxic material is released when SSII-1 cells are digested and that multiplication of B. sphaericus in the larval gut is not essential in the pathogenesis. There appears to be no difference in the pathogenesis when differing numbers of B. sphaericus . i.e., LD 10–20 or LD 70–98 dosages, are ingested. Possible nature of the toxic material is discussed.

Culturable Bacteria associated with the Whitefly, Bemisia Argentifolii (Homoptera: Aleyrodidae)

Several different types of bacteria were cultured from surface-sterilized Bemisia argentifolii Bellows, Perring, Gill and Hedrick 1994 (Homoptera: Aleyrodidae) adults and nymphs, including Bacillus spp., Gram-variable pleomorphic rods and Gram-positive cocci. Two of the isolates were capable of being ingested by adults and passed into the honeydew. One of these, Enterobacter cloacae, was found within the gut cells of adult whiteflies and was mildly pathogenic. This isolate represents the first bacterium with potential as a pathogen of whiteflies. Bacteria which were not capable of being ingested, may have been located in structures which were protected from surface sterilization, such as the lingula or the female reproductive tract.

Ultrastructural effects of the Bacillus sphaericus mosquito larvicidal toxin on cultured mosquito cells

Abstract Crude Bacillus sphaericus extracts and purified toxin derived from these extracts caused very rapid changes in cultured Culex quinquefasciatus cells, including dilation of mitochondrial cristae, endoplasmic reticula, and Golgi secretory vesicles, and condensation of the mitochondria. The cell membrane gradually lost integrity as intoxication progressed. These observations are compared to the ultrastructure of the pathology due to Bacillus thuringiensis in cultured cells and larvae, and are discussed in relation to binding and internalization of the toxin.