Polyclonal antibodies were produced in mice against Spodoptera exigua (beet armyworm) larval hemolymph and hemocytes and against cell wall surfaces of hyphal bodies and hyphae of the entomopathogenic hyphomycete Nomuraea rileyi. In addition to exhibiting strong activity against their original antigenic substrates, all of the antibodies cross-react extensively with other substrates. The hemolymph antibody binds to hemocytes and vice versa, and both antibodies cross-react to the insect fat body basement membrane (extracellular matrix (ECM) and to N. rileyi and Beauveria bassiana (another entomopathogenic fungus) cell wall surfaces (ECM). Likewise, the anti-fungal antibodies cross-react with S. exigua hemolymph and hemocytes, especially the granules that may contain ECM components, and with fat body basement membrane. These cross-reactivities are specific as indicated by negative controls in the microscopy and Western blotting assays. Parallel labeling experiments using Con A suggest that the reactive epitopes contain mannose; however, none of the antibodies bind to mannose residues of nonentomopathogenic Candida albicans or Saccharomyces cerevisiae yeast cells. Thus, these cross-reactivities suggest that the host mimicry expressed by surface components of entomopathogenic fungi represents an important pathogenic determinant.

Monoclonal antibodies (MAbs) were generated against epitopes on yeast-like hyphal bodies and hyphae of the entomopathogenic hyphomycete, Nomuraea rileyi. Two MAbs (4C10, 2H4) bind to epitopes common to both hyphal bodies and hyphae, whereas MAb 4E9 binds only to hyphal surfaces. 4C10 and 2H4 appear to be directed towards carbohydrate portions of cell surface mannoproteins, as evidenced by similarities in staining patterns between these MAbs and Concanavalin A on Western blots of N. rileyi cell wall extracts. These MAbs cross-react with antigens on blastospore and hyphal surfaces of two other entomopathogenic fungi, Beauveria bassiana and Paecilomyces farinosus in fluorescence microscopy assays, but do not cross-react with a non-entomopathogenic strain of Candida albicans or with Saccharomyces cerevisiae yeasts. MAb 4C10 also cross-reacts with immunocompetent granular hemocytes from Spodoptera exigua (beet armyworm) and Trichoplusia ni (cabbage looper) larvae and with S. exigua plasmatocytes. Electron microscopy revealed that this MAb binds to a component in cytoplasmic granules in the hemocytes, and that surface labeling may be due to the release of this MAb-positive component upon degranulation. MAb 2H4 does not cross-react with granular hemocytes, but does bind to plasmatocytes and hemocytes that tightly adhere to the substrate in monolayer assays. Additionally, MAb 4C10 specifically labels a basement membrane epitope on S. exigua fat body, suggesting that this antibody binds to mannose residues on extracellular matrix glycoproteins. Cross-reactivity of these N. rileyi MAbs with insect hemocyte and tissue components indicates that fungal surface epitopes can mimic host surface molecules, which could explain why N. rileyi hyphal bodies are not recognized by granulocytes and are able to circulate freely in the hemolymph without binding to basement membranes lining the hemocoel.

The entomopathogen Nomuraea rileyi is a dimorphic hyphomycete having invasive hyphal stages and a yeastlike vegetative stage. Each phenotype has unique characteristics that contribute to pathogenicity of the fungus towards insects. Growth both in vivo and in vitro is highly synchronous. In vitro, conidial germ tubes produce hyphal bodies (yeastlike cells) in complex mycological medium and, if transferred to a less complex medium (Vogel's), the hyphal bodies synchronously convert to mycelia. This transition can be reversed by transferring the mycelia back to the original medium. The hyphal body phase can be by-passed by placing germ tubes directly into Vogel's. The ability to regulate fungal development provides a means of comparing characteristics (e.g., cell surface epitopes, enzyme production) which differ according to phenotype, and to identify genes associated with phase transitions.

A 16-kDa protein toxin was purified from Hirsutella thompsonii var thompsonii and named hirsutellin A (HtA). At 0.5 and 5.0 microM concentrations, HtA caused detectable cytopathic effects on Spodoptera frugiperda cells (Sf-9) within 2-4 hr and completely inhibited Sf-9 cell growth at 4 days posttreatment. Electron microscope data showed that the HtA treated Sf-9 cells became hypotrophied and internal organelles and cell membranes were disrupted. At the same concentration, HtA effectively inhibited Brome mosaic virus protein synthesis of both rabbit reticulocyte and wheat germ in vitro translation system. The ribosomal RNA extracted from HtA treated Sf-9 cells produced a smaller RNA (approximately 528 bases) than untreated Sf-9 cells. In summary, HtA is the first mycotoxin of a invertebrate mycopathogen determined to possess ribosomal inhibiting activity and appears to possess some specificity to invertebrate cells.

The arrangement of carbohydrate molecules on surfaces of fungal cells may play an important role in nonself recognition of these microorganisms by potential invertebrate hosts. Changes in the ability of various galactose and mannose-specific lectins to bind to surface components on cell walls of the insect pathogen Paecilomyces farinosus were therefore examined during growth and differentiation of the fungus. Fluorescein isothiocyanate conjugates of concanavalin A (Con A, specific for alpha-D-mannose) and peanut agglutinin (PNA, beta-D-galactose) bound inconsistently to blastospores and weakly to mycelia except at apical regions where strong fluorescence was observed. Labeling patterns were similar on cells tested with a galactose-specific lectin purified from Spodoptera exigua (beet armyworm) hemolymph, but Bandeiraea simplicifolia lectin (BS-I alpha-D-galactose) bound only to mycelia. Electron microscopy using ferritin and gold probes showed that the galactomannans are located in a loosely bound coating on the cell wall surface. Variations in lectin binding patterns are apparently due to absence (e.g., by shedding) of the coat or to rearrangement of carbohydrate components in the coat. Staining of Western blots of dithiothreitol (DTT) cell wall extracts further indicated that the BS-I-binding entity is a unique component of the mycelial surface since, as in the fluorescence studies, blastospore preparations were not labeled. Staining of blastospore blots with other galactose-specific probes (e.g., PNA) was comparable to staining of mycelial blots.(ABSTRACT TRUNCATED AT 250 WORDS)

A double-stranded DNA virus was isolated from hyperplasic salivary glands of male and female houseflies, Musca domestica L. (Diptera: Muscidae), collected from a dairy in Alachua County, Florida, U.S.A. Sodium dodecyl sulphate (SDS)-polyacrylamide gel electrophoresis (PAGE) of this housefly salivary gland hyperplasia (SGH) virus revealed the presence of two major and eight minor structural polypeptides. Restriction endonuclease analysis indicated that the c. 137 kilobase pair DNA was double-stranded. Weekly, sweep-net sampling of the fly population throughout the season (May-October, 1991) showed that 1.5-18.5% of the dissected flies possessed hyperplasic salivary glands. The virus replicated within the nuclei of the salivary gland cells and was transmitted per os to newly-emerged healthy adult flies.

In vivo cells (hyphal bodies) of the hyphomycetous insect pathogen Beauveria bassiana collected from host Spodoptera exigua larval hemolymph were osmotically sensitive and lacked a well-defined cell wall. In light and electron microscope studies, a galactose-specific lectin purified from S. exigua hemolymph, concanavalin A (specific for alpha-mannose), and a polyclonal antibody to B. bassiana cell walls all bound to surfaces of in vitro-produced B. bassiana blastospores; however, none of these probes labelled the thin layer of extracellular material covering the plasma membranes of hyphal bodies. These cells were observed freely circulating in S. exigua hemolymph at 36 h postinfection, although immunocompetent hemocytes were known to be present. Additionally, association of hyphal bodies with hemocytes in monolayers was significantly less than for opsonized in vitro blastospores or submerged conidia. The absence of antigenically important galactomannan components on in vivo cells may therefore allow these cells to escape recognition and phagocytosis. Lack of structural components (e.g., chitin, as evidenced by the absence of binding of wheat germ agglutinin) may also be important with respect to evasion of host cellular defense mechanisms. Production of wall material resumed 48 to 60 h postinfection and therefore may coincide with loss of phagocytic capabilities of the hemocytes due to immunosuppressive effects of fungal metabolites. The protoplast-like cells may be formed by the action of hydrolytic enzymes in the hemocytes or by inhibition of fungal cell wall synthetases.

The Spodoptera exigua galactose binding lectin, extracted from hemolymph by affinity chromatography, was comprised of large molecular weight aggregates (100–700 kDa). SDS-PAGE of this lectin preparation revealed it to contain 2 subunits of 33.2 and 34.4 kDa in equimolar concentrations. These subunits had similar amino acid profiles, possessed identical N-terminal sequences and reacted equally to a bank of antilectin monoclonal antibodies. By staining Western blots with various lectin conjugate probes, we demonstrated that the 34.4 kDa subunit contains complex mannose residues, suggesting that this subunit is the glycosylated form of the 33.2 kDa subunit. The N-terminal sequence of the S. exigua lectin was distinct from other invertebrate galactose binding lectins. Light microscopy in combination with immunoelectron microscopy was used to localize the S. exigua lectin in the granules of the granulocyte class of hemocytes. Degranulation of these cells resulted in the release of the lectin. Isotope incorporation studies followed by immunoprecipitation with a S. exigua monoclonal antibody demonstrated that the fat body was the major site of lectin synthesis. Similar studies with hemocyte monolayers did not result in the production of detectable levels of 35S-labeled S. exigua lectin.

Several probes were used in this ultrastructural study to localize polysaccharides in cell walls on conidial germ tubes, hyphal bodies, and mycelia of the entomogenous hyphomycete Nomuraea rileyi. With the exception of galactose, labelling patterns did not vary from one morphological stage to another. Galactose, which was localized by using a monoclonal antibody to a galactose-specific lectin purified from insect larval hemolymph, was absent from cell walls of hyphal bodies and conidia but was present on germ-tube and mycelial surfaces. Chitin (N-acetylglucosamine), labelled with a wheat-germ agglutinin-ferritin conjugate, was present in the middle regions of lateral walls and septa, and β1-4 glucans were located in the middle and inner regions, as indicated by binding of a cellulase-gold conjugate. An anti-laminaribiose antibody was used to label β1-3 glucans present in the outer wall areas and inner regions near the plasmalemma. The location of mannose residues as indicated by concanavalin A - ferritin binding was similar to that of the β1-3 glucans; vesicle-like structures were also labelled. None of the probes labelled the outer conidial pellicle or exocellular sheath surrounding germ tubes, and labelling of mycelial sheath was inconsistent. The absence of galactose from Nomuraea hyphal body walls is discussed in terms of host-parasite interaction. Key words: Nomuraea rileyi, entomopathogenic fungi, glycoconjugates, lectins, monoclonal antibodies.

Representative species of all the major fungal groups have evolved the capability of infecting insect hosts (Samson et al., 1988). The general infection pathway, which includes attachment of propagules to the host cuticle, penetrant germ tube formation, vegetative growth, and production of progeny spores, requires the sequential development of unique fungal structures tailored to optimally utilize the insect as a nutritional substrate for the production of infectious spores. Deviations in the developmental sequence of a particular mycopathogen may expose the invasive fungal cell to the host defense reactions, aborting the infection process. The insect defense system is comprised of both passive mechanisms, such as the cuticular barrier, and active components, i.e., the humoral and cellular reactions. In the following discussion we will review the in vivo development of a “typical” insect pathogenic deuteromycete. It is important to realize that the fungal-insect interactions are as diverse as those found among vertebrate fungi. Emphasis will be placed on the composition of the fungal surfaces and with the interplay between the fungal cells and the host cell recognition mechanisms.

A non-occluded baculovirus was isolated from nymphs of the field cricket,Gryllus rubens. SDS-polyacrylamide gel electrophoresis revealed the presence of 6 major and 11 minor polypeptides in these particles. Restriction endonuclease analysis indicated that the genome, 87.0±1.8 kilobase pairs, was a closed circular DNA molecule. DNA-DNA hybridization in low strigency conditions revealed no homology with the genomes ofOryctes baculovirus orAutographa california NPV. The virus replicated in nuclei of fat body cells, and was transmitted per os to a small proportion of first instarG. rubens nymphs.

Blastospores of some entomopathogenic hyphomycetes (e.g. Paecilomyces farinosus, Beauveria bassiana) apparently become opsonized when incubated in a galactose-binding haemagglutinin (lectin) purified from Spodoptera exigua larvae. Labelling with fluorescein isothiocyanate-peanut agglutinin demonstrated the presence of exposed galactose residues on the blastospore cell walls of both P. farinosus and B. bassiana and on the outer membrane surface of granulocytes of S. exigua. In in vitro tests using haemocyte monolayers, the number of S. exigua granulocytes with attached or endocytosed (associated) opsonized P. farinosus blastospores may be above 80% depending upon the haemagglutination titre of the lectin. This association is specifically inhibited by galactose. The number of granulocytes with associated non-opsonized blastospores is approx 5–10%. Interaction between granulocytes and P. farinosus blastospores treated with S. exgua sera with known haemagglutinin activity is similar to that observed when purified agglutinin was used. Nomuraea rileyi blastospores (hyphal bodies), which do not have exposed galactose residues, do not bind to S. exigua granulocytes even if pre-treated with the lectin. In vivo studies show that injected P. farinosus cells having galactose receptors are rapidly cleared from S. exigua haemolymph, whereas N. rileyi blastospores are not cleared.

The attachment of the conidia of the insect-pathogenic fungi Nomuraea rileyi, Beauveria bassiana, and Metarrhizium anisopliae to insect cuticle was mediated by strong binding forces. The attachment was passive and nonspecific in that the conidia adhered readily to both host and nonhost cuticle preparations. The hydrophobicity of the conidial wall and the insect epicuticle appeared to mediate the adhesion process. Detergents, solvents, and high-molecular-weight proteins known to neutralize hydrophobicity reduced conidial binding when added to conidium-cuticle preparations. However, these chemicals did not remove the hydrophobic components from the epicuticle or from conidial preparations. The outer surface of the conidium consists of a resilient layer of well-organized fascicles of rodlets. Intact rodlets extracted from B. bassiana conidia bound to insect cuticle and exhibited the hydrophobicity expressed by intact conidia. Both electrostatic charges and various hemagglutinin activities were also present on the conidial surface. However, competitive-inhibition studies indicated that these forces played little, if any, role in the adhesion process.

A new iridovirus has been detected from diseased southern mole crickets, Scapteriscus acletus, collected in Brazil during the spring of 1986. This icosahedral virus measuring 146 nm (side-side) to 172 nm (apex-apex) has been purified via Ficoll gradient centrifugation and demonstrated to be infectious to 1st instar Scapteriscus vicinus nymphs. The cytopathology of this virus is typical of the pattern documented for other iridovirus isolates. Characterization of the structural polypeptides by SDS-polyacrylamide gel electrophoresis revealed an array of 3 major and 17 minor polypeptides ranging in molecular weight from 15.1 to 152.0 kDa. Electrophoresis in agarose gels of purified DNA revealed a single band of high molecular weight. Analysis of various restriction endonuclease (REN) digests of this DNA demonstrated it to have an approximate molecular weight of 144 kilobase pairs. Based on differences in the polypeptide profile and REN profiles we believe this virus is distinct from previously characterized invertebrate iridovirus isolates.

A heterologous rDNA probe was used to detect restriction fragment length polymorphisms in Entomophaga rDNA sequences. Six Canadian strains of Entomophaga aulicae, isolated from the spruce bud worm or hemlock looper in Ontario or Newfoundland, showed no detectable rDNA variation at 10 different restriction enzyme loci: BamHI, DraI, EcoRI, EcoRV, HindIII, HinfI, MspI, PstI, RsaI, or TaqI. A total of 14 isolates of E. aulicae representative of several different geographic and host origins were compared at the DraI and HindIII rDNA loci and two different banding patterns were detected. Of these, 12 showed the same patterns and were designated E. aulicae type 1. The two members which differed were designated E. aulicae type 2. The variations in rDNA restriction sites did not appear to be geographically dependent. Entomophaga maimaiga, a recently reclassified species from the E. aulicae complex, displayed an rDNA banding pattern clearly distinguishable from the E. aulicae patterns with DraI, EcoRI, EcoRV, or HindIII. Members of the E. grylli species complex exhibited patterns which clearly differed from the patterns seen with either E. aulicae or E. maimaiga isolates. However, members of the E. grylli species complex appeared to be more heterogeneous than those in the E. aulicae complex. Among four E. grylli members, three different rDNA banding patterns were detected with either HindIII or DraI. These were designated as E. grylli type 1, type 2, and type 3. An undesignated Entomophaga isolate from a dipteran host displayed rDNA polymorphisms not previously noted in either the lepidopteran or orthopteran isolates. Our results suggest that RFLPs in rDNA are useful in the delineation of genera and species within the Entomophthorales, but may not be as useful at lower taxonomic levels. These and other RFLPs can however provide useful information regarding the epidemiology of Entomophaga epizootics.

A complex of protease inhibitor activities has been detected in the hemolymph of the 6th instarAnticarsia gemmatalis larvae that are resistant to infection by the fungusNomuraea rileyi. A site-specific serine protease inhibitor extracted fromA. gemmatalis hemolymph inhibits both the germination ofN. rileyi conidia and subsequent germ tube development.

The hyphomycete Sorosporella sp. was isolated from the mole cricket, Scapteriscus vicinus collected in Alachua Co., Florida. Scanning (SEM) and/or transmission (TEM) electron microscopy were used to study the chlamydospores characteristic of this genus and the conidial production in the alternate state, Syngliocladium. Brick-red chlamydospores, which occur in clusters in insect cadavers, have a fibrous cell wall as revealed by TEM. Pores often occur between walls of adjacent cells. Lipids, including a large central lipid droplet and smaller droplets along the periphery of the cell, are abundant in the cytoplasm. Most organelles were difficult to distinguish due to the density of the cytoplasmic material. Chlamydospores germinate on water or media. On Sabouraud maltose agar, germinating chlamydospores produce a white mycelial mat; synnematous-type growth was sometimes observed. Mycelia may bear conidiophores and ellipsoid conidia.

Hemolymph from Spodopteraexigua larvae agglutinates rabbit and human 0 erythrocytes. The agglutinin appears to be naturally occurring i.e., injury (by injection) or injection of larvae with fungal cells does not induce an increase in titer of hemolymph samples. Hemagglutinin activity is destroyed by heat or EDTA, and galactosidic carbohydrates inhibit agglutination of both red blood cell types. The agglutinin was purified by affinity chromatography using an Affi-Gel ® ovalbumin column. Binding to the column is calcium (cation) dependent. SDS gel electrophoresis shows that the agglutinin is a minor component of whole hemolymph represented by two bands with molecular weights of 30,500 and 31,000 daltons. Fluorescence microscopy using rhodamine-labeled agglutinin indicates that the agglutinin binds to fungal cell wall surfaces known to have galactose residues (e.g., Paecilomycesochraceus), and that binding is specifically inhibited by galactose. There is no specific binding to fungal walls known to lack galactose residues (e.g., Paecilomycesochraceus, Nomuraearileyi). The agglutinin may be involved in the immune response of the insect e.g., by opsonization of microbial (fungal) surfaces which render the invading cells more susceptible to phagocytosis or agglutination.

The hyphomycetous insect pathogens Nomuraea rileyi, Paecilomyees farinosus, Metarhizium anisopliae and Beauveria bassiana were tested with fluorescein isothiocyanate (FITC)-labeled lectins in order to determine cell surface sugars. Concanavalin A, specific for glucose and mannose residues, bound to cell surfaces (hyphal body and/or mycelium) of all fungi examined. Soybean and wheat germ agglutinins (affinities for N-acetylgalactosamine and N-acetylglucosamine, respectively) bound inconsistently to some cell surface areas. Peanut agglutinin, specific for galactose residues, bound to walls of only P. farinosus and B. bassiana. Fungal surface sugar residues may be important with respect to insect defense mechanisms against invading cells. Lectins produced by insects (e.g. Anticarsia gemmatalis and Spodoptera exigua) are known to be specifically inhibited by galactose (or higher concentrations of glucose or mannose). Fungal cells (P. farinosus, B. bassiana) which have exposed galactose residues as detected by FITC-peanut agglutinin appear to be especially efficient in removing hemagglutinin activity from A. gemmatalis and S. exigua hemolymph indicating adsorption of the hemagglutinins to fungal surfaces. Such adsorption may render the fungal cells more susceptible to phagocytosis or other insect defense mechanisms.

Infection of insects by the entomopathogenic fungus Beauveria bassiana proceeds via attachment and penetration of the host cuticle. The outermost epicuticular layer or waxy layer of the insect represents a structure rich in lipids including abundant amounts of hydrocarbons and fatty acids. A member of a novel cytochrome P450 subfamily, CYP52X1, implicated in fatty acid assimilation by B. bassiana was characterized. B. bassiana targeted gene knockouts lacking Bbcyp52x1 displayed reduced virulence when topically applied to Galleria mellonella, but no reduction in virulence was noted when the insect cuticle was bypassed using an intrahemoceol injection assay. No significant growth defects were noted in the mutant as compared with the wild-type parent on any lipids substrates tested including alkanes and fatty acids. Insect epicuticle germination assays, however, showed reduced germination of ΔBbcyp52x1 conidia on grasshopper wings as compared with the wild-type parent. Complementation of the gene-knock with the full-length gene restored virulence and insect epicuticle germination to wild-type levels. Heterologous expression of CYP52X1 in yeast was used to characterize the substrate specificity of the enzyme. CYP52X1 displayed the highest activity against midrange fatty acids (C12:0 and C14:0) and epoxy stearic acid, 4–8-fold lower activity against C16:0, C18:1, and C18:2, and little to no activity against C9:0 and C18:0. Analyses of the products of the C12:0 and C18:1 reactions confirmed NADPH-dependent regioselective addition of a terminal hydroxyl to the substrates (ω-hydroxylase). These data implicate CYP52X1 as contributing to the penetration of the host cuticle via facilitating the assimilation of insect epicuticle lipids.