Laboratory studies assessing the microbial biocontrol potential of diverse strains of Beauveria bassiana isolated from coffee berry borer, with emphasis on strains from Hawai'i Island and comparisons to commercial strain GHA.

A series of experiments investigated genetically diverse strains of Beauveria bassiana (Bb) isolated from coffee berry borer (CBB). Objectives included assessment of their biocontrol potential, particularly in comparison to Bb commercial strain GHA currently applied for CBB control, and identification of various attributes potentially contributing to their comparatively greater epizootic potential in CBB populations. Bioassays identified one strain from Hawai'i Island and one from Puerto Rico with virulence greater than GHA based on equal weights of unformulated conidial powder (CP); however, the greater potency of the CPs was ultimately explained by their 2.4-fold greater conidial densities (ca. 3.1 vs 1.3 × 1011 viable conidia/g CP). Density was explained, in large part, by conidial size, but not by size alone. Bb-inoculated CBB held on moist filter paper were more susceptible to infection than those held on cooked green coffee bean (CGCB). A Bb strain representative of the most common Hawaiian haplotype produced 2.6x more conidia after death of CGCB-held beetles than GHA (19.1 vs 7.3 x106 conidia/cadaver). Following host death, no difference was observed in time to emergence and initial conidial production by GHA and a selected group of Hawaiian strains; however, mass sporulation was initiated within 2 days by strain GHA compared to 4-5 days by the Hawaiian strains. In a preliminary evaluation of conidial mass-production potential, CP yields of several strains were comparable to GHA on a weight basis and significantly greater than GHA on a conidial basis (1.3-1.6 vs 0.7 × 1013 viable conidia/kg barley substrate).

Collection, isolation, in vitro culture, and laboratory transmission of Hirsutella eleutheratorum (Hypocreales: Ophiocordycipitaceae) from coffee berry borer on Hawai'i Island.

The insect pathogenic fungus Hirsutella eleutheratorum was first reported as a pathogen of coffee berry borer (CBB) Hypothenemus hampei in Colombia in 1993. A similar CBB pathogen identified as Hirsutella sp. was reported also from Colombia in 2007; attempts at isolation and in vitro culture of this fungus were unsuccessful. During 2016 and 2017 on the island of Hawai'i, extensive sampling of CBB populations was conducted in coffee fields treated with Beauveria bassiana-based biopesticides and in untreated fields. Among the samples collected from two high-elevation sites in the district of South Kona were rare findings of adult foundress CBB infected with a species of Hirsutella fitting the description of H. eleutheratorum. Prevalence of the pathogen was, in all cases, very low (<1%), having no significant impact on pest populations, even under conditions supporting epizootics of B. bassiana. The fungus was readily isolated from freshly-killed CBB and cultured on potato dextrose agar (PDA). Molecular characterization identified the fungus as a member of the Hirsutella citriformis clade, which includes species recently placed in the genus Ophiocordyceps. Adult CBB exposed to fungus-killed beetles or to PDA cultures of the fungus succumbed to infection within 10-14 days. Under high-humidity laboratory conditions, the fungus emerged from the killed host and produced long, conidia-bearing synnemata characteristic of the species. To our knowledge, this is the first record of H. eleutheratorum from CBB in Hawai'i and the first account of isolation, in vitro culture, genetic characterization, host-to-host transfer, and culture-to-host transfer of this fungal pathogen.

Vegetative compatibility groups in indigenous and mass-released strains of the entomopathogenic fungus Beauveria bassiana: likelihood of recombination in the field.

Using nitrate non-utilizing (nit) mutants, we determined vegetative compatibility groups (VCG) among strains of Beauveria bassiana representing strains indigenous to North America, isolated from diverse insect hosts, and strains that have been mass released for insect control. Genetic similarity among these strains was analyzed using random amplified polymorphic DNA (RAPD) markers. Our data revealed 23 VCGs among the 34 strains tested, with most of these groups comprised of only a single strain. We also observed a VCG comprised of eight genetically similar strains isolated from Colorado potato beetles (CPB). Co-inoculation studies of CPB larvae with complementary nit mutants from the same or from different VCGs revealed heterokaryosis in four out of five same-VCG pairs, with only 5-17% of the sporulating cadavers generating few parasexual recombinants. In contrast, none of the infected beetles treated with non-compatible pairs generated recombinants. The large number of VCGs observed and the low frequency of in vivo recombination limited to vegetatively compatible strains indicate that this self/non-self recognition system may be an effective barrier preventing genetic exchange between dissimilar strains in the field.

Effect of relative humidity on viability of primary conidia of Zoophthora radicans.

Four isolates of the entomopathogenic fungus Zoophthora radicans were compared in a laboratory study to evaluate the effect of relative humidity (RH) on duration of primary conidial viability. Primary conidia were showered onto agar-coated glass microscope slides within an enclosed chamber equilibrated to one of five test RH levels (60, 75, 80, 95, or 100%). Target RH levels were achieved by recirculating air through a glycerin/water solution, of controlled specific gravity, contained in a reservoir within the chamber. Conidial samples of each isolate incubated for 5, 10, 30, 60, 120, 180, or 240 min at each RH were removed and inspected using a technique of simultaneous vital fluorochrome staining to determine percentage conidial viability. At 60% RH, isolates did not differ significantly and average viability dropped to less than 10% within the first 60 min. At 75% RH, viability did not change significantly over 4 h. However, average viabilities at 75% RH differed significantly for the four isolates and ranged from 24 to 63%. At 80% RH, viability differed significantly among isolates and declined slowly over time, remaining above 80% for 2 h and above 50% for 4 h. At 95 and 100% RH, average viability was near 95% and did not vary significantly with time or isolate. These data can be used to assist selection of appropriate isolates for biological control.