No description available.

No description available.

No description available.

Many insects are important agricultural pests, and active control is necessary to keep them at abeyance. The naturally occurring entomopathogenic fungus Metarhizium is a promising tool to control pest insects, and its use avoids the well-known harmful side effects of chemical pesticides. Thousands of unique isolates of Metarhizium exist throughout the world. These isolates vary widely in their ability to cause infection and to tolerate stressful habitats. The research reported here tests the THESIS: A laboratory-based system can be devised that identifies, from among many Metarhizium isolates, those isolates with the greatest potential for successful biological control of pest insects in the field. The study was built on the testing of four hypotheses: (1) Laboratory bioassays using target pest insects will distinguish highly virulent strains of Metarhizium from less virulent strains, (2) Quantity and quality of mass-produced pathogenic fungi will vary among species and strains of Metarhizium, (3) The tolerance to ultraviolet radiation will vary among species and strains of Metarhizium, (4) The effect of temperature on growth rates and survival of both Metarhizium spores and hyphae will vary among isolates and species. These hypotheses test four field-relevant traits using a panel of ten isolates of Metarhizium isolates. Seven sets of laboratory experiments were devised to define the range of responses within the traits covered by the hypotheses. This series of general laboratory tests was developed to assist in identifying fungal isolates with high potential for field use. These tests included evaluation of each isolate's (a) insect pathogenicity, (b) mass–production capabilities, (c) tolerance to high temperatures, (d) tolerance to UV-B radiation, (e) rate of vegetative growth, (f) rate of spore germination, and (g) an evaluation of presence or absence of a post–stress growth inhibition. The application of this protocol to the isolates used in this study indicates that four isolates have high field potential, i.e., DWR 203, DWR 346, DWR 356 and ARSEF 324, and three of these were tested in a field trial. By following the procedures outlined in this thesis, selection of “good” isolates can be accomplished in the laboratory, and a successful isolate can be identified from the abundance of isolates present in nature.

DNA similarity studies were used to determine the species of thirty-one strains of bacteria isolated from the hemolymph of infected larvae from Mexico and throughout Central and South America. Twenty-one of the strains were determined to be <I>Paenibacillus popilliae</I> and ten were found to be more closely related to <I>Paenibacillus lentimorbus</I>. Only one of the <I>P. popilliae</I> strains, an isolate from Mexico, was resistant to the antibiotic vancomycin, a trait characteristic of <I>P. popilliae</I> strains from other geographic areas. As expected, all <I>P. lentimorbus</I> strains were sensitive to vancomycin. The polymerase chain reaction (PCR) was used to amplify a portion of a ligase gene necessary for vancomycin resistance in the Mexican strain. Sequencing of the amplicon revealed a sequence identical to that obtained from a North American strain of <I>P. popilliae</I> previously described. The ability of <I>P. popilliae</I> and the inability of <I>P. lentimorbus</I> to grow on medium supplemented with 2% sodium chloride has been used as a phenotypic trait for differentiating between the two species. Approximately 86% of the <I>P. popilliae</I> strains were capable of growth on medium supplemented with 2% sodium chloride and 60% of the <I>P. lentimorbus</I> strains were not capable of growth on this medium. Microscopic examination revealed that all of the Mexican and Central and South American strains of <I>P. popilliae</I> and <I>P. lentimorbus</I> produced a parasporal body. PCR was used to amplify two different regions of the <I>cry18Aa1</I> gene encoding the paraspore in all of the isolates. One primer pair, CryBP2, detected the <I>cry18Aa1</I> gene in 17 of the 21 <I>P. popilliae</I> strains and in all ten of the <I>P. lentimorbus</I> strains. The second primer pair, CryBP4, detected the parasporal gene in 20 of the 21 <I>P. popilliae</I> strains and in all ten of the <I>P. lentimorbus</I> strains. Thirty of the thirty-one <I>P. popilliae</I> and <I>P. lentimorbus</I> strains produced amplicons of approximately 616 bp with the CryBP4 primers. The CryBP4 primers did not detect the paraspore gene in one of the <I>P. popilliae</I> strains. The CryBP2 primer pair produced amplicons of three different sizes, indicating possible variability in the parasporal proteins of <I>P. popilliae</I> and <I>P. lentimorbus</I>. Eleven of the <I>P. popilliae</I> strains produced CryBP2 amplicons approximately 660 bp in size and six of the <I>P. popilliae</I> strains produced CryBP2 amplicons approximately 1100 bp in size. The <I>cry</I> gene was not detected in four of the <I>P. popilliae</I> strains with the CryBP2 primers. The <I>P. lentimorbus</I> strains produced CryBP2 amplicons approximately 750 bp in size. Three PCR products representing the variable CryBP2 amplicon sizes were sequenced and compared to the published <I>cry18Aa1</I> gene sequence. Sequencing data revealed that the Central and South American CryBP2 amplicons are similar to the published <I>cry18Aa1</I> sequence, however, the 1100 bp amplicon has a 453 bp insert that is not found in the published <I>cry18Aa1</I> gene sequence. Paraspores are produced by <I>P. popilliae</I> and <I>P. lentimorbus</I> and is not a reliable phenotypic trait for differentiation between the two species. The ability of Mexican and Central and South American strains of <I>P. lentimorbus</I> to produce paraspores supports the previous findings of a North American group of <I>P. lentimorbus</I> strains that were also capable of producing paraspores. Except for one Mexican strain of <I>P. popilliae</I>, the Central and South American strains of <I>P. popilliae</I> are sensitive to vancomycin. This was unexpected since all North American strains of <I>P. popilliae</I> are vancomycin resistant. Vancomycin resistance could be useful in identifying strains of <I>P. popilliae</I> from North America but not for identifying strains of <I>P. popilliae</I> from Central and South America. So far, no vancomycin resistant strains of <I>P. lentimorbus</I> have been identified. There was variability in the ability of these organisms to grow on medium supplemented with 2% sodium chloride so the usefulness of this trait is debatable. However, the majority of <I>P. popilliae</I> strains from Mexico and Central and South America will grow on medium supplemented with 2% sodium chloride and the majority of the <I>P. lentimorbus</I> strains from these same areas will not grow on this medium. North American strains of <I>P. popilliae</I> and <I>P. lentimorbus</I> also showed variability of growth on medium supplemented with 2% sodium chloride. / Master of Science