Root colonization by endophytic insect-pathogenic fungi.

Several ascomycetous insect-pathogenic fungi, including species in the genera Beauveria and Metarhizium, are plant root symbionts/endophytes and are termed as endophytic insect-pathogenic fungi (EIPF). The endophytic capability and insect pathogenicity of Metarhizium are coupled to provide an active method of insect-derived nitrogen transfer to plant hosts via fungal mycelia. In exchange for the insect-derived nitrogen, the plant provides photosynthate to the fungus. This symbiotic interaction offers other benefits to the plant-EIPF can improve plant growth, they are antagonistic to plant pathogens and herbivores and can enhance the plant tolerance to abiotic stresses. The mechanisms and underlying biochemical and genetic features of insect pathogenesis are generally well-established. However, there is a paucity of information regarding the underlying mechanisms in this plant-symbiotic association. Here we review five aspects of EIPF interactions with host plant roots: (i) rhizosphere colonization, (ii) signalling factors from the plant and EIPF, (iii) modulation of plant defence responses, (iv) nutrient exchange and (v) tripartite interactions with insects and other micro-organisms. The elucidation of these interactions is fundamental to understanding this symbiotic association for effective application of EIPF in an agricultural setting.

Insect Pathogenic Fungi as Endophytes.

In this chapter, we explore some of the evolutionary, ecological, molecular genetics, and applied aspects of a subset of insect pathogenic fungi that also have a lifestyle as endophytes and we term endophytic insect pathogenic fungi (EIPF). We focus particularly on Metarhizium spp. and Beauveria bassiana as EIPF. The discussion of the evolution of EIPF challenges a view that these fungi were first and foremost insect pathogens that eventually evolved to colonize plants. Phylogenetic evidence shows that the lineages of EIPF are most closely related to grass endophytes that diverged c. 100MYA. We discuss the relationship between genes involved in "insect pathogenesis" and those involved in "endophytism" and provide examples of genes with potential importance in lifestyle transitions toward insect pathogenicity. That is, some genes for insect pathogenesis may have been coopted from genes involved in endophytic colonization. Other genes may be multifunctional and serve in both lifestyle capacities. The interactions of EIPF with their host plants are discussed in some detail. The genetic basis for rhizospheric competence, plant communication, and nutrient exchange is examined and we highlight, with examples, the benefits of EIPF to plants, and the potential reservoir of secondary metabolites hidden within these beneficial symbioses.

Endophytic insect-parasitic fungi translocate nitrogen directly from insects to plants.

Most plants obtain nitrogen through nitrogen-fixing bacteria and microbial decomposition of plant and animal material. Many vascular plants are able to form close symbiotic associations with endophytic fungi. Metarhizium is a common plant endophyte found in a large number of ecosystems. This abundant soil fungus is also a pathogen to a large number of insects, which are a source of nitrogen. It is possible that the endophytic capability and insect pathogenicity of Metarhizium are coupled to provide an active method of nitrogen transfer to plant hosts via fungal mycelia. We used soil microcosms to test the ability of M. robertsii to translocate insect-derived nitrogen to plants. Insects were injected with (15)N-labeled nitrogen, and we tracked the incorporation of (15)N into amino acids in two plant species, haricot bean (Phaseolus vulgaris) and switchgrass (Panicum virgatum), in the presence of M. robertsii. These findings are evidence that active nitrogen acquisition by plants in this tripartite interaction may play a larger role in soil nitrogen cycling than previously thought.

Conidium production by insect pathogenic fungi on commercially available agars.

AIMS: Conidium production by three species of insect pathogenic fungi, Metarhizium anisopliae, Beauveria bassiana and Verticillium lecanii, was assessed on various depths and types of commercially available agars. METHODS: Conidium production was assessed after 14 d of growth on commercially available media as well as at three different agar depths. RESULTS: Metarhizium anisopliae and B. bassiana isolates showed greatest conidium production on potato dextrose agar (PDA) at a depth of 2 mm, whereas V. lecanii showed greatest conidium production on yeast extract-peptone-dextrose agar (YPDA) regardless of agar depth. Optimum conidium production for M. anisopliae and B. bassiana was not only dependent upon the isolate used but also on the medium type and agar depth. SIGNIFICANCE AND IMPACT OF THE STUDY: Conidia are the infective structures for insect pathogenic fungi and this study suggests a rationale basis for consistent conidium production for laboratory and commercial practices.

Habitat association in two genetic groups of the insect-pathogenic fungus metarhizium anisopliae: uncovering cryptic species?

Strains of insect-pathogenic fungi with high virulence toward certain pest insects have great potential for commercial biological control applications. Identifying such strains has been a central theme in using fungi for biological control. This theme is supported by a persistent paradigm in insect pathology which suggests that the host insect is the predominant influence on the population genetics of insect-pathogenic fungi. In this study, a population genetics analysis of the insect-pathogenic fungus Metarhizium anisopliae from forested and agricultural habitats in Ontario, Canada, showed a nonrandom association of alleles between two distinct, reproductively isolated groups (index of multilocus association = 1.2). Analyses of the mitochondrial DNA showed no differences between the groups. The two groups were associated with different habitat types, and associations with insect hosts were not found. The group from forested areas showed an ability for cold-active growth (i.e., 8 degrees C), while the group from the agricultural area showed an ability for growth at high temperatures (i.e., 37 degrees C) and resilience to UV exposure. These results represent a significant paradigm shift; habitat selection, not host insect selection, drives the population structure of these insect-pathogenic deuteromycetous fungi. With each group we observed recombining population structures as well as clonally reproducing lineages. We discuss whether these groups may represent cryptic species. Worldwide, M. anisopliae may be an assembly of cryptic species, each adapted to certain environmental conditions. The association of fungal genotypes with habitat but not with host insects has implications on the criteria for utility of this, and perhaps other, fungal biocontrol agents.

Genetic polymorphisms in three subtilisin-like protease isoforms (Pr1A, Pr1B, and Pr1C) from Metarhizium strains.

Restriction fragment length polymorphisms (RFLP) were examined in three isoforms of a gene family encoding subtilisin-like proteases (Pr1A, Pr1B, and Pr1C) in several isolates of the entomopathogenic fungus Metarhizium anisopliae. RFLP variation was not observed in any of the Pr1 genes from isolates within the same genetically related group. Between genetically related groups and between isolates from disparate geographical areas, the greatest variation in RFLP patterns was observed for Pr1A. When variation does occur at Pr1B and Pr1C, it was generally observed at an EcoRI site. Metarhizium anisopliae var. majus strain 473 and a M. flavoviride isolate were most dissimilar in RFLP patterns at all Pr1 genes when compared to the M. anisopliae strains. We suggest that Pr1 genes represent a gene family of subtilisin-like proteases and that the Pr1A gene encodes for the ancestral subtilisin-like protease which has subsequently duplicated and rearranged within the genome.

Bacterial fitness and plasmid loss: the importance of culture conditions and plasmid size.

Several pBluescript-derived plasmids of various sizes were constructed to study the effects of multicopy plasmid size on bacterial fitness and plasmid loss. Transformed and untransformed bacterial clones were grown in media with or without ampicillin. Bacterial fitness (measured by growth rate), plasmid presence or absence, and plasmid copy number were assessed during successive subculturings. In selective media (minimal medium or Luria Broth plus ampicillin), the clone transformed with the largest plasmid (pBluescript with a 9000-bp insert) had a significantly longer lag phase than all other clones. In nonselective media the rate of plasmid loss during successive subculturings was greatest in the clone with the largest insert. The clone with the largest insert displayed a lower plasmid copy number than clones with a small insert or no insert at all. Plasmid loss in the form of segregational instability and plasmid copy number reduction in nonselective environments are important to the understanding of the evolution of the bacteria-plasmid associations and the appreciation of the potential for altering the genetic properties of a clone maintained or subcultured on a standard medium.

Construction of an improved mycoinsecticide overexpressing a toxic protease.

Mycoinsecticides are being used for the control of many insect pests as an environmentally acceptable alternative to chemical insecticides. A key aim of much recent work has been to increase the speed of kill and so improve commercial efficacy of these biocontrol agents. This might he achieved by adding insecticidal genes to the fungus, an approach considered to have enormous potential for the improvement of biological pesticides. We report here the development of a genetically improved entomopathogenic fungus. Additional copies of the gene encoding a regulated cuticle-degrading protease (Pr1) from Metarhizium anisopliae were inserted into the genome of M. anisopliae such that Pr1 was constitutively overproduced in the hemolymph of Manduca sexta, activating the prophenoloxidase system. The combined toxic effects of Pr1 and the reaction products of phenoloxidase caused larvae challenged with the engineered fungus to exhibit a 25% reduction in time of death and reduced food consumption by 40% compared to infections by the wild-type fungus. In addition, infected insects were rapidly melanized, and the resulting cadavers were poor substrates for fungal sporulation. Thus, environmental persistence of the genetically engineered fungus is reduced, thereby providing biological containment.

Biochemical characterization and ultrastructural localization of two extracellular trypsins produced by Metarhizium anisopliae in infected insect cuticles.

Proteinase 2 (Pr2) is a fungal (Metarhizium anisopliae) serine proteinase which has a tryptic specificity for basic residues and which may be involved in entomopathogenicity. Analytical and preparative isoelectric focusing methods were used to separate two trypsin components, produced during growth on cockroach cuticle, with isoelectric points of 4.4 (molecular mass, 30 kDa) and 4.9 (27 kDa). The catalytic properties of the proteases were analyzed by their kinetic constants and by a combination of two-dimensional gelatin-sodium dodecyl sulfate-polyacrylamide gel electrophoresis and enzyme overlay membranes. Both Pr2 isoforms preferentially cleave at the carboxyl sides of positively charged amino acids, preferring arginine; the pI 4.4 Pr2 isoform also possessed significant activity against lysine. Compared with the pathogen's subtilisin-like enzyme (Pr1), the pI 4.4 Pr2 isoform shows low activity against insoluble proteins in a host (Manduca sexta) cuticle. However, it degrades most cuticle proteins when they are solubilized, with high-molecular-weight basic proteins being preferentially hydrolyzed. Polyclonal antibodies raised against each Pr2 isoform were isotype specific. This allowed us to use ultrastructural immunocytochemistry to independently visualize each isoform during penetration of the host (M. sexta) cuticle. Both isoforms were secreted by infection structures (appressoria) on the cuticle surface and by the penetrant hyphae within the cuticle. The extracellular sheath, which is commonly observed around fungal cells, often contained Pr2 molecules. Intracellular labelling was sparse.

Characterization and Ultrastructural Localization of Chitinases from Metarhizium anisopliae, M. flavoviride, and Beauveria bassiana during Fungal Invasion of Host (Manduca sexta) Cuticle.

Extracellular chitinases have been suggested to be virulence factors in fungal entomopathogenicity. We employed isoelectric focusing and a set of three fluorescent substrates to investigate the numbers and types of chitinolytic enzymes produced by the entomopathogenic fungi Metarhizium anisopliae, Metarhizium flavoviride, and Beauveria bassiana. Each species produced a variety of N-acetyl-(beta)-d-glucosaminidases and endochitinases during growth in media containing insect cuticle. M. flavoviride also produced 1,4-(beta)-chitobiosidases. The endochitinases could be divided according to whether they had basic or acidic isoelectric points. In contrast to those of the other two species, the predominant endochitinases of M. anisopliae were acidic, with isoelectric points of about 4.8. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis resolved the acidic chitinases of M. anisopliae into two major bands (43.5 and 45 kDa) with identical N-terminal sequences (AGGYVNAVYFY TNGLYLSNYQPA) similar to an endochitinase from the mycoparasite Trichoderma harzianum. Use of polyclonal antibodies to the 45-kDa isoform and ultrastructural immunocytochemistry enabled us to visualize chitinase production during penetration of the host (Manduca sexta) cuticle. Chitinase was produced at very low levels by infection structures on the cuticle surface and during the initial penetration of the cuticle, but much greater levels of chitinase accumulated in zones of proteolytic degradation, which suggests that the release of the chitinase is dependent on the accessibility of its substrate.

Fate of biological control introductions: monitoring an Australian fungal pathogen of grasshoppers in North America.

In North America there are two generally recognized pathotypes (pathotypes 1 and 2) of the fungus Entomophaga grylli which show host-preferential infection of grasshopper subfamilies. Pathotype 3, discovered in Australia, has a broader grasshopper host range and was considered to be a good biocontrol agent. Between 1989 and 1991 pathotype 3 was introduced at two field sites in North Dakota. Since resting spores are morphologically indistinguishable among pathotypes, we used pathotype-specific DNA probes to confirm pathotype identification in E. grylli-infected grasshoppers collected at the release sites in 1992, 1993, and 1994. In 1992, up to 23% of E. grylli-infected grasshoppers of the subfamilies Melanoplinae, Oedipodinae, and Gomphocerinae were infected by pathotype 3, with no infections > 1 km from the release sites. In 1993, pathotype 3 infections declined to 1.7%. In 1994 grasshopper populations were low and no pathotype 3 infections were found. The frequency of pathotype 3 infection has declined to levels where its long-term survival in North America is questionable. Analyses of biocontrol releases are critical to evaluating the environmental risks associated with these ecological manipulations, and molecular probes are powerful tools for monitoring biocontrol releases.

Pathotypes in the Entomophaga grylli species complex of grasshopper pathogens differentiated with random amplification of polymorphic DNA and cloned-DNA probes.

The zygomycetous fungus Entomophaga grylli is a pathogen that shows host-specific variance to grasshopper subfamilies. Three pathotypes of the E. grylli species complex were differentiated by three molecular techniques. In the first method, the three pathotypes showed different fragment patterns generated by random amplification of polymorphic DNA (RAPD). There was little or no interisolate variability in RAPD fragment patterns within each pathotype. Passage of an isolate of pathotype 3, originally from an Australian grasshopper (Praxibulus sp.), through a North America grasshopper resulted in no differences in the resultant RAPD fragment patterns. In the second method, polymorphic RAPD fragments were used to probe the genomic DNA from the three pathotypes, and pathotype-specific fragments were found. In the third method, restriction fragments from genomic DNA of the three pathotypes were cloned and screened for pathotype specificity. A genomic probe specific for each pathotype was isolated. These probes did not hybridize to DNA from Entomophaga aulicae or from grasshoppers. To facilitate the use of RAPD analysis and other molecular tools to identify pathotypes, a method for extracting DNA from resting spores from infected grasshoppers was developed. The DNA from the fractured resting spores was of sufficient integrity to be blotted and probed with the pathotype-specific DNA probes, thus validating the use of these probes for pathotype identification in field-collected grasshoppers.

Cloning of a cuticle-degrading protease from the entomopathogenic fungus, Beauveria bassiana.

A Beauveria bassiana extracellular subtilisin-like serine endoprotease is a potential virulence factor by virtue of its activity against insect cuticles. A cDNA clone of the protease was isolated from mycelia of B. bassiana grown on cuticle/chitin cultures. The amino acid sequence of this gene was compared to that of Metarhizium anisopliae Pr1, the only pathogenicity determinant so far described from an entomopathogenic fungus, and proteinase K, isolated from Tritirachium album, a saprophytic fungus. The cDNA sequence revealed that B. bassiana Pr1 is synthesized as a large precursor (M(r) 37,460) containing a signal peptide, a propeptide and the mature protein predicted to have an M(r) of 26,832.

Characterization of a novel carboxypeptidase produced by the entomopathogenic fungus Metarhizium anisopliae.

Preparative isoelectric focusing and gel filtration chromatography were used to purify a carboxypeptidase produced by the entomopathogenic fungus Metarhizium anisopliae during growth on cockroach cuticle. The enzyme was inhibited by diisopropyl fluorophosphate, implying involvement of a serine residue in catalysis. However, the M. anisopliae enzyme differed from most serine carboxypeptidases in also being inhibited by the metal chelator 1,10-phenanthroline and in being a small (30 kDa), basic (pI 9.97) protein with a neutral pH optima (pH 6.8). These properties resemble those exhibited by some metalloproteases but the enzyme is not inhibited by Cd2+; nor do Zn2+ or Co2+ restore activity in enzyme inhibited with phenanthroline. The amino-terminal sequence (22 residues) showed no similarity to other protein sequences. Unlike previously reported fungal carboxypeptidases, the M. anisopliae enzyme is powerfully inhibited by potato carboxypeptidase inhibitor. The carboxypeptidase shows a broad primary specificity toward amino acids with hydrophobic side groups in a series of N-blocked dipeptides, with substrates with phenylalanine being the most rapidly hydrolyzed. The S1 subsite also accommodated Glu, confirming its low selectivity. Proline at P1 or P1 resulted in a very poor substrate. The specificity of the carboxypeptidase complements that of the subtilisin-like protease (Pr1) of M. anisopliae. Both Pr1 and the carboxypeptidase are produced during carbon and nitrogen deprivation, which indicates that the exopeptidase functions with Pr1 to degrade peptides to supply amino acids during starvation and pathogenicity.

Isoforms of the cuticle-degrading Pr1 proteinase and production of a metalloproteinase by Metarhizium anisopliae.

The entomopathogenic fungus, Metarhizium anisopliae, produces three distinct types of proteinases during growth on cockroach cuticle. These were separated by analytical isoelectric focusing and characterized according to their substrate specificity and inhibition patterns as Pr1 subtilisin-like proteinases (four isoforms pI range approximately 9.3-10.2), a thermolysin-like metalloproteinase (pI approximately 7.3), and trypsin-like serine Pr2 proteinases (two major isoforms, pI approximately 4.4 and 4.9 and two minor isoforms, pI approximately 5.2). Preparative isoelectric focusing was used to separate the four Pr1(2) components produced during growth on cockroach cuticle with isoelectric points of 10.2 (m = 30.2 kDa), 9.8 (m = 28.5 kDa), 9.3 (m = 29.5 kDa), and 9.0 (m = 31.5 kDa). Two of the isoforms were also produced, at diminished levels, during growth on elastin or cellulose presumably as a result of carbon and nitrogen derepression. The pI 10.2 Pr1 differed from the other isoforms in preferring alanine over bulky hydrophobic groups at P2 and P3, in discriminating against proline at P2 and in its lack of sensitivity to tetra-butyl-oxycarbonyl-Gly-Leu-Phe-chloromethyl ketone. Differences in the N-terminal amino acid sequences confirmed that the four isoforms are related products of at least two distinct genes. The isoforms showed similar primary specificities, with the aromatic P1 phenylalanine being 10- to 16-fold more reactive than a P1 leucine residue reflected principally in Kcat. However, methionine (containing a long unsubstituted side chain) was also a good substrate for each isoform confirming the low selectivity of their S1 subsites. The isoforms all degraded a variety of solubilized cuticle proteins, with high-molecular-weight acidic proteins being preferentially hydrolyzed. The metalloproteinase is active against the Pr1 substrate succinyl-(Ala)2-Pro-Phe-7-amino-4-coumarin trifluoromethyl, but differs from the Pr1 isoforms in being inhibited by 1,10-phenanthroline and phosphoramidon. The potential role of the metalloproteinase in pathogenicity is discussed.

Differentiation of species and strains of entomopathogenic fungi by random amplification of polymorphic DNA (RAPD).

Polymerase chain reaction (PCR)-based technology, involving random amplification of polymorphic DNA (RAPD), was used to assess the genomic variability between 24 isolates of deuteromycetous fungi (Metarhizium anisopliae, Metarhizium flavoviride, unidentified strains of Metarhizium and Beauveria bassiana) which were found to infect grasshoppers or locusts. M. flavoviride showed little intraspecific variability in PCR-amplified fragments when compared to M. anisopliae. The high level of variability in PCR-amplified fragments contained within M. anisopliae was similar to the total variability between B. bassiana, M. anisopliae and M. flavoviride, and suggests that M. anisopliae may include a number of cryptic species. Four polymorphic RAPD fragments were used to probe the genomic DNA of the various species and strains. On the basis of these probes the fungi can be grouped into M. flavoviride, M. anisopliae, or B. bassiana. According to PCR-amplified fragments, previously-unidentified Metarhizium strains were characterized as M. flavoviride. There was little evidence that these fungi, all isolated from, or virulent towards, grasshoppers or locusts, showed host-selection in PCR-amplified fragments. Nor was geographical origin a criterion for commonality based on PCR-amplified fragments. PCR-fragment-pattern polymorphisms and the construction of probes from one or more of these fragments may provide a useful and rapid tool for identifying species and strains of entomopathogenic fungi.

Partial purification and characterization of two extracellular N-acetyl-D-glucosaminidases produced by the entomopathogenic fungus Beauveria bassiana.

Beauveria bassiana grown in a liquid medium containing N-acetyl-D-glucosamine and colloidal chitin produced two distinct N-acetyl-D-glucosaminidases, NAGase 1 and NAGase 2. NAGase 1 had a molecular weight of 97,000 and NAGase 2 was comprised of two subunits, of molecular weights 64,000 and 66,000. The optimal temperature and pH for NAGase 1 were 57 degrees C and pH 5 and for NAGase 2 they were 37 degrees C and pH 5. NAGase 1 was more thermostable than NAGase 2. The isolectric points of NAGase 1 and 2 were ca. pH 9.5 and 5.5, respectively. NAGase 1 and 2 were unaffected by a 10 mM concentration of chloride salts of the ions Ca2+, Mg2+, or Zn2+, by 10 mM EDTA, and by 0.25-1 mM of short chain fatty acids. Dithiothreitol caused some inactivation of NAGase 2 while stimulating activity of NAGase 1. NAGase 1 had a Km of 0.38 mM and a Kcat/Km of 3923.88 M-1.s-1.NAGase 2 had a Km of 2.095 mM and a Kcat/Km of 411.88 M-1.s-1 when p-nitrophenyl-beta-N-acetylglucosaminide was used as the substrate.

Regulation of extracellular N-acetyl-D-glucosaminidase production in the entomopathogenic fungus Beauveria bassiana.

The entomopathogenic fungus Beauveria bassiana produces two extracellular N-acetylglucosaminidases (NAGase) in liquid medium containing colloidal chitin as the sole source of carbon and nitrogen. To study the regulation of NAGase synthesis, N-acetyl-D-glucosamine (GlcNAc), glucose NH4NO3, or amino acids were added to the colloidal chitin medium and NAGase activity was measured. NAGase synthesis was (i) induced with GlcNAc, and no repression was observed with GlcNAc provided at 2% (w/v); (ii) repressed in the presence of glucose plus NH4NO3; (iii) partially repressed when glucose or NH4NO3 was provided; and (iv) repressed to levels that were < 40% of the control levels when glutamic acid, tyrosine, arginine, proline, valine, and histidine were provided to the colloidal chitin medium. Total NAGase activity levels were > 60% of the control activity when alanine, glycine, isoleucine, aspartic acid, and leucine were tested. It appears that synthesis of NAGase is sensitive to cell energy and the carbon and nitrogen requirements.

Carbohydrate Storage in the Entomopathogenic Fungus Beauveria bassiana.

The entomopathogenic fungus Beauveria bassiana was grown in 1% (wt/vol) gelatin-liquid media singly supplemented with a monosaccharide (glucose or fructose), a disaccharide (maltose or trehalose), a polyol (glycerol, mannitol, or sorbitol), or the amino sugar N-acetyl-d-glucosamine. The relative contributions of the carbohydrate, protein, and water contents in the fungal biomass were determined. Carbohydrates composed 18 to 42% of the mycelial dry weight, and this value was lowest in unsupplemented medium and highest in medium supplemented with glucose, glycerol, or trehalose. Biomass production was highest in liquid cultures supplemented with trehalose. When liquid cultures were grown in medium supplemented with 0 to 1% (wt/vol) glucose, trehalose, or N-acetyl-d-glucosamine, there was an increase in the biomass production and the contribution of carbohydrate to mycelial dry weight. Regardless of the glucose concentration in the culture, water content of the mycelia remained about 77.5% (wt/wt). Mycelial storage carbohydrates were determined by capillary gas chromatography. In gelatin-liquid medium supplemented with 1% (wt/vol) glucose, B. bassiana stored glycogen (12.0%, wt/dry wt) and the polyols mannitol (2.2%), erythritol (1.6%), glycerol (0.4%), and arabitol (0.1%). Without glucose, B. bassiana stored glycogen (5.4%), mannitol (0.8%), glycerol (0.6%), and erythritol (0.6%) but not arabitol. To our knowledge, this is the first report of carbohydrate storage in an entomopathogenic fungus, and the results are discussed in relation to other fungi and the potential implications to commercial formulation and insect-fungus interactions.

N-Acetyl-d-Glucosamine-Mediated Regulation of Extracellular Protease in the Entomopathogenic Fungus Beauveria bassiana.

The entomopathogenic fungus Beauveria bassiana GK2016 grown in a liquid medium incorporating gelatin as the sole carbon and nitrogen source produced an extracellular serine protease (molecular weight, 35,000; pI ca. 10). Without gelatin, B. bassiana could utilize N-acetyl-d-glucosamine (GlcNAc; 2-acetamido-2-deoxy-d-glucose) as the sole source of carbon and nitrogen, and GlcNAc availability increased the storage carbohydrate content in mycelia. Synthesis of protease was repressed in gelatin medium containing GlcNAc at levels of >1.07 mumol mg of fungal dry weight. At levels below this, protease synthesis was initiated; subsequently, free amino nitrogen appeared in the medium and diauxic growth was observed. Slow feeding with GlcNAc (35.34 mug ml h) did not repress protease synthesis nor did GlcNAc accumulate in the medium above 0.5 mg ml. Increasing the rate of release of GlcNAc (83.51 mug ml h) resulted in the accumulation of GlcNAc in the medium to 2.0 mg ml, a 45% increase in growth and a decrease in protease synthesis by about 81%. Free amino acids generated from the hydrolysis of gelatin did not repress protease synthesis. These data are interpreted in terms of known interaction of B. bassiana with insect cuticular components. We suggest that the action of extracellular chitinases synthesized by B. bassiana on insect cuticle, and pursuant release of GlcNAc, may have important consequences on the regulation of other extracellular catabolic enzymes such as the protease.