Complex chitinolytic system of Aspergillus fumigatus.

Native polyacrylamide gels incorporating a glycol chitin substrate were used to detect several chitinolytic enzymes in the culture filtrate and cell surface, wall and mixed membrane fractions of Aspergillus fumigatus during the exponential phase of growth. Much of the cellular chitinase activity did not bind to concanavalin A (Con A) matrix and was heat-sensitive. In contrast, almost all chitinases secreted appeared to be heat-stable glycoproteins. The heavily glycosylated molecules, in a Con A-binding fraction, were the most immunologically-reactive components, as judged by their binding to anti-Aspergillus antibodies, present in the serum of patients with aspergillosis. Most of the cellular chitinases of A. fumigatus mycelium bound to an insoluble chitin matrix while most of the secreted chitinases did not bind to chitin.

Analysis of peptidogalactomannans from the mycelial surface of Aspergillus fumigatus.

Peptidogalactomannans (pGMs) from mycelium of two strains of Aspergillus fumigatus were fractionated by Cetavlon precipitation and size exclusion chromatography and their carbohydrate structures analysed using methylation-fragmentation analysis, partial acetolysis and 13C-nuclear magnetic resonance spectroscopy. The most significant difference between the pGMs of the two strains was the degree of branching and the proportion of non-reducing ends of alpha-D-Manp and beta-D-Galf units. Methylation data showed that the pGM from AF 2109 contained alpha-D-Manp and beta-D-Galf non-reducing end units in a proportion of 3:1 while, in contrast, the proportion of these structures in pGM from AF 2140 was 7:1, resulting in a highly branched structure. The immunoreactivity of the pGM fractions was tested by indirect immunofluorescence. The fractions were also tested in an ELISA system with rabbit antiserum raised to whole cells of A. fumigatus NCPF 2140 and with serum from patients with either proven aspergilloma or ABPA. The carbohydrate moiety of the pGM appears to be responsible for the antigenicity. Periodate treatment, partial acid hydrolysis and beta-elimination removed most of the antibody binding capacity.

Effects of tunicamycin on glycoprotein antigens of Aspergillus fumigatus.

Tunicamycin, which inhibits N-glycosylation of proteins, was used as a tool to determine the type of linkage which occurs in glycoprotein antigens of Aspergillus fumigatus. When A. fumigatus extracts were electrophoretically separated and blotted then probed with anti-Aspergillus patients' sera, differences in antigenic profiles were noted when tunicamycin-treated samples were compared with controls. Tunicamycin had no detectable effect on the cellular proteinases of A. fumigatus, most of which are glycosylated. Some enzymatic components were lacking when extracellular proteinases were compared with those of control samples. The major catalase component of A. fumigatus is a concanavalin A (ConA)-binding glycoprotein. In cultures grown in the presence of tunicamycin, partially-deglycosylated catalase components were obtained which could be distinguished from the native catalase by altered mobilities in polyacrylamide gels. The effect of deglycosylation on catalase antigens was monitored using an antiserum raised to a ConA-binding fraction of A. fumigatus mycelium. These antibodies bound both to the native glycoprotein and the partially deglycosylated material. These latter two were largely unaffected when incubated with an antiserum raised to a non-ConA-binding fraction of A. fumigatus which is essentially carbohydrate free. The ability to produce partially-glycosylated antigens of A. fumigatus offers a model to study the effect of basic structural modifications on both the enzymatic and antigenic activities of these molecules.

Specific antibody detection in invasive aspergillosis by analytical isoelectrofocusing and immunoblotting methods.

Aspergillus fumigatus antigens have been tested to determine their potential as aids in the diagnosis of invasive aspergillosis (IA). Immunoglobulin G (IgG) antibodies to these antigens were detected by analytical isoelectrofocusing in conjunction with immunoblotting. A total of 12 antigenic fractions, including culture filtrates and surface and mycelial extracts of A. fumigatus, were investigated. Eleven were reactive with serum specimens from patients with aspergilloma, which served as positive controls for the evaluation of a specific IgG response. Eight of 12 antigens showed good responses with serum specimens from patients with allergic bronchopulmonary aspergillosis, which were used to assess the sensitivity of IgG detection. No measurable reactivity was detected in 18 negative control serum specimens, while 11 of 13 patients with proven, highly probable, or probable cases of IA had anti-Aspergillus IgG to multiple antigenic preparations. Patients with IA who were capable of mounting a substantial humoral response to Aspergillus antigens gave an antibody profile with five antigenic preparations which seemed to be characteristic of the disease. Data show that this method is highly sensitive and may allow the selection of fractions which are both highly antigenic and specific for the detection of antibodies to Aspergillus antigens. They also indicate that the use of a spectrum of antigenic molecules is advisable, given the variability observed in the immune responses of individual patients.

Chemical and immunological analysis of the Aspergillus fumigatus cell wall.

Hyphal-wall preparations of Aspergillus fumigatus have been analysed by sequential treatment with KOH, nitrous acid and again with KOH. By acidification of the alkali-soluble extract, a polyglucose was precipitated which showed an X-ray diffraction pattern similar to that of (1-->3)-alpha-glucan. The remainder of the alkali-soluble fraction was precipitated with ethanol; it contained all the mannose, galactose and protein of the wall and, in addition, 6.2% of the amino sugars. This wall-associated glycoprotein, following SDS-PAGE and immunoblotting, reacted with antisera raised against several mycelial extracts of A. fumigatus. Sera from patients with aspergilloma have antibodies which recognize components of this glycoprotein. The glycoprotein nature of these antigens was shown by their ability to bind Lens culinaris lectin. In addition, the antigen/antibody binding could be disrupted by exposure of antigen to periodate oxidation, hydrolysis with dilute acid or pretreatment with a large excess of an exo-beta-D-galactofuranosidase. The alkali-insoluble fraction consisted of a covalently linked glucan-chitin complex. Nitrous acid treatment, which specifically disrupts glycosidic linkages involving glucosamine, did not solubilize much material but changed the X-ray diffraction pattern from diffuse to a pattern showing the characteristic lines of crystalline (1-->3)-beta-glucan and chitin. Most of the glucan became alkali-soluble after this treatment, and the insoluble residue appeared to contain crystalline chitin.

A study of the alkaline proteases secreted by different Aspergillus species.

Strains from several species of Aspergillus were grown in the presence of soluble collagen, and the major secreted proteins present in the culture fluid were examined for proteolytic activity. The possibility of relatedness among the alkaline proteases secreted by Aspergillus was studied by probing extracts from the various species with polyclonal antisera raised to the isolated alkaline proteases of A. fumigatus and A. oryzae. The pathogenic species A. flavus, A. terreus and A. nidulans hydrolyse collagen and were found to secrete an alkaline protease related to that of Aspergillus fumigatus. In contrast, A. niger and non-pathogenic species such A. glaucus, A. versicolor and A. clavatus were unable to degrade collagen in vitro. These findings suggest a possible pathogenic role for the secreted alkaline proteases of Aspergillus species.

The effect of sulconazole on the antigenic composition of two dermatophyte species.

Trichophyton rubrum and Trichophyton interdigitale have been grown in liquid culture in the presence of sulconazole. The antigenic activity of detergent extracts of intact organisms was analysed following SDS-PAGE and the probing of Western blots with homologous antisera raised in rabbits and with sera from patients with dermatophyte infections. Differences in protein-band patterns were noted; some bands present in control samples were absent in azole-treated samples and vice versa. These differences were reflected in antigenic band patterns, especially among components of approximate molecular weight of 30-40, 50-60 and 92-100 kDa.

Specific recognition pattern of IgM and IgG antibodies produced in the course of experimental paracoccidioidomycosis.

Specific IgM and IgG responses to Paracoccidioides brasiliensis produced in resistant and susceptible mice during experimental paracoccidioidomycosis were examined by the immunoblotting procedure. Sera from infected mice recognized 51 antigen bands with apparent molecular masses from 8 to 86 kD. Sixteen of these were defined as major antigen bands because of almost universal presence of antibodies to them, and their intense staining. All sera, including those from normal control mice, tested for both IgM and IgG antibody reacted with the major E antigen which appeared as a large diffuse band from 43 to 47 kD. Comparisons between resistant and susceptible mice showed some significant differences in IgM responses to many antigen bands. While IgG responses were quite similar for both strains, differences were apparent in the response to the antigens at 62 and 68 kD.

Analysis of Aspergillus fumigatus catalases possessing antigenic activity.

Analysis of Aspergillus fumigatus water soluble fractions by electrophoresis on non-denaturing polyacrylamide gels (PAGE) showed the presence of at least three catalase bands. They were designated F, S1 and S2 in order of descending electrophoretic mobility with respect to the anode. The multiple enzyme forms appear to be distinct in their physicochemical properties. Enzyme bands S1 and S2 were simple catalases; the F band had an additional peroxidase function. All of the components were antigenic and differed in their binding to specific antibodies raised in rabbits with separate fractions of A. fumigatus mycelium. When serum from patients with aspergilloma, allergic bronchopulmonary aspergillosis, cystic fibrosis and chronic asthma were pre-incubated with A. fumigatus antigens and analysed by PAGE, 17 of 26 samples either abolished or reduced catalase activity. Enzyme F was a non-Concanavalin A (ConA)-binding antigen; the S1 and S2 enzymes were ConA-binding glycoprotein antigens. The major catalase band present in A. niger preparations represented only a minor component in A. fumigatus.

Immunolocalization of Aspergillus fumigatus mycelial antigens.

Specific immunoglobulin from the sera of patients with antibodies to Aspergillus and from antisera raised in rabbits to Aspergillus fumigatus fractions bound almost exclusively to the mycelial wall, as shown by immunogold labelling of ultra-thin sections. Different layers of the wall were labelled, depending on the source of the antigen used to produce antibody. Internal, cytoplasmic components were generally not labelled, except with antisera raised to crude wall material and to a Concanavalin A (ConA)-binding fraction of a water-soluble preparation.

Immunochemical studies of Aspergillus fumigatus mycelial antigens by polyacrylamide gel electrophoresis and western blotting techniques.

Differences were detectable among strains of the opportunist fungal pathogen Aspergillus fumigatus when water-soluble (WS) preparations were analysed by combined SDS-PAGE and Western blotting procedures. A wide range of molecules of apparent molecular masses from approximately 20 to greater than 100 kDa showed specific binding to antibodies raised in rabbits to A. fumigatus wall and cytoplasmic components. The ability to bind antibody was markedly reduced by treatment of these antigens with sodium periodate or with specific proteases or glucanases. Pretreatment of blotted antigens with either concanavalin A (ConA) or wheat germ agglutinin (WGA) did not, however, inhibit subsequent antibody binding. The antigens of subfractions prepared from a single strain of A. fumigatus WS material were also susceptible to periodate oxidation and enzymic hydrolysis. Slight cross-reactivity was apparent when crude preparations of cellular or culture filtrate antigens, used in this laboratory to detect antibodies to Candida albicans, Coccidioides immitis and Cryptococcus neoformans, were probed with hyperimmune rabbit antisera to A. fumigatus. Efforts were made to characterize the WS preparations of A. fumigatus, used as diagnostic antigens in many laboratories. The electrophoretically separated antigenic moieties were shown to be predominantly glycoproteins. Binding of cytoplasmic antigens to antibodies raised to wall material showed the presence of many common components in both wall and cytosol. Antiserum to wall components revealed most differentiation among A. fumigatus strains.

Structural analysis of water-soluble fractions obtained from Aspergillus fumigatus mycelium.

Fractions were prepared from the water-soluble components of Aspergillus fumigatus mycelium either by lectin-affinity chromatography or salt precipitation. While they varied considerably in their amino-acid composition, each contained a preponderance of aspartic and glutamic acids. 13C-NMR spectroscopy of these fractions, compared with that of polysaccharide obtained by alkaline extraction, indicated the presence of glycoproteins, the polysaccharide components of which contained beta-D-Galf units that are part of structures chemically different from those obtained by alkali treatment. In two of the three fractions examined, gas-liquid chromatography--mass spectrometry showed marked differences in the contents of non-reducing end-units of alpha-D-Manp and beta-D-Galf. Sodium dodecyl sulphate-polyacrylamide gel electrophoresis of the preparations revealed an array of components, which stained to differing extents with silver stain and with Coomassie Blue and many of which were bound by lectins with specificity for different sugars.

Immunoreactivity of a surface wall fraction produced by spherules of Coccidioides immitis.

The membranous spherule outer wall (SOW) isolated from liquid cultures of Coccidioides immitis has been shown to elicit reactivity with human anti-Coccidioides antibody by immunofluorescence and the immunodiffusion-tube precipitin assay. The serologically reactive components were extracted from SOW with the nonionic detergent N-octyl-beta-D-glucopyranoside (OG). The OG-soluble fraction of SOW was shown to be reactive with immunoglobulin G in 25 serum samples from coccidioidomycosis patients by an enzyme-linked immunosorbent assay. The isolated SOW and OG-soluble fraction of SOW were also demonstrated to be capable of eliciting lymphocyte blastogenesis. The antigenic and protein compositions of the OG-soluble fraction were examined by two-dimensional immunoelectrophoresis and sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), respectively. Two antigens which were extracted from SOW were identified as antigens 2 and CS on the basis of the coccidioidin-anticoccidioidin reference system. The latter was isolated earlier and shown to correspond to a molecular mass (Mr) of 19 X 10(3) by SDS-PAGE under reducing conditions. This same electrophoresis band was shown to be reactive with sera from coccidioidomycosis patients by immunoblot analysis. One other SDS-PAGE component of the OG-soluble fraction of SOW with an Mr of 66 X 10(3) was shown to be reactive with sera from patients by immunoblot analysis. The SOW of C. immitis represents an important reservoir of immunoreactive wall components which has not previously been reported.

Isolation and morphology of an immunoreactive outer wall fraction produced by spherules of Coccidioides immitis.

A previously undescribed, immunoreactive, membranous spherule outer wall (SOW) fraction produced by Coccidioides immitis (strains 634 and 735) grown in culture was isolated. Both this fraction and intact spherules were reactive with sera from coccidioidomycosis patients, as demonstrated by immunofluorescence microscopy. The serological activity of SOW was also demonstrated by its reactivity with human anti-C. immitis tube precipitin in a standardized immunodiffusion assay. Extraction of SOW with the nonionic detergent N-octyl-beta-D-glucopyranoside (OG) permitted the isolation of an OG-soluble fraction which was reactive in the immunodiffusion assay. Rabbit antisera raised against the OG-soluble fraction were used in immunofluorescence and immunoelectron-microscopic studies of the parasitic cycle to confirm that the immunoreactive components of the solubilized fraction of SOW were associated with the inner and outer layers of the spherule wall as well as with distinct cytoplasmic organelles observed in thin sections of spherules. The immunoreactivity of SOW with sera from patients suggested that infected individuals are exposed to this surface wall material isolated from in vitro-grown spherules.

Evaluation of a test to detect circulating Aspergillus fumigatus antigen in a survey of immunocompromised patients with proven or suspected invasive disease.

An ELISA for the detection and measurement of Aspergillus antigenaemia has been developed and evaluated by examining sera submitted over a 12-month period from immunocompromised patients with a likelihood of invasive aspergillosis. Results from proven cases of invasive aspergillosis confirmed at post-mortem and specimens from individuals with suspected disease showed that tests on single serum samples were often negative. Multiple specimens from the same patient greatly increased the frequency of detection. Repeated monitoring of sera from a single patient showed wide fluctuations in antigen level, which was considered to be due partly to the medical regimen to which the patient was subject. Control sera from healthy laboratory personnel were consistently negative, but a number of 'at-risk' patients without other evidence of invasive aspergillosis sometimes had low amounts of antigen. Concentrations of Aspergillus antigen of 100 ng ml-1 or higher were considered to be strongly suggestive of fungal invasion.

Cellular and humoral immune responses in experimental aspergillosis in mice.

Invasive aspergillosis was established in a naive murine model. Both humoral and cellular aspects of the immunological mechanism responded to invasion. Circulating immune complexes containing IgG in the infected group of animals were detected by both the C1q and conglutinin solid phase assays. Attempts to identify these complexes as Aspergillus antigen-specific complexes were unsuccessful. Cellular responses, as measured by blastogenic activity and inhibition of migration of sensitized cells were significantly elevated in the infected group.