Detection of Cell Wall Galactomannoprotein Afmp1p in Culture Supernatants of Aspergillus fumigatus and in Sera of Aspergillosis Patients

Abstract

Mannoproteins are important and abundant structural components of fungal cell walls. The AFMP1 gene encodes a cell wall galactomannoprotein of Aspergillus fumigatus. In the present study, we show that Afmp1p is secreted into the cell culture supernatant at a level that can be detected by Western blotting. A sensitive enzyme-linked immunosorbent assay (ELISA) developed with antibodies against Afmp1p was capable of detecting this protein from the cell culture supernatant of A. fumigatus. The anti-Afmp1p antibody is specific since it fails to react with any protein from lysates of Aspergillus flavus, Aspergillus niger, Aspergillus terreus, Penicillium marneffei, Candida albicans, Cryptococcus neoformans, Blastomyces dermatitidis, and Histoplasma capsulatum by Western blotting. In addition, this Afmp1p antigen-based ELISA is also specific for A. fumigatus since the cell culture supernatants of the other eight fungi gave negative results. Finally, a clinical evaluation of sera from invasive aspergillosis patients indicates that 8 of 15 (53%) patients are Afmp1p antigen test positive. Furthermore, an Afmp1p antibody test was performed with these serum specimens. The combined antibody and antigen tests for invasive aspergillosis carry a sensitivity of 86.7% (13 of 15). The specificities of the tests are high since none of the 138 control sera, including 100 from normal blood donors, 20 from patients with penicilliosis marneffei, 6 from patients with candidemia, 8 from patients with typhoid fever, and 4 from patients with melioidosis, was positive by either test. In conclusion, the combined Afmp1p antibody and antigen tests are highly sensitive and specific for A. fumigatus invasive aspergillosis.

Since the last decade, Aspergillus species have gained prominence as opportunistic pathogens. In immunocompetent hosts, Aspergillus species rarely causes serious illnesses, except for aspergilloma in patients with preexisting chronic lung diseases. On the other hand, invasive aspergillosis is one of the most important infectious causes of mortality in patients with hematological malignancies and bone marrow transplant (BMT) recipients, with an incidence of 6% in our recent study on 230 BMT recipients (18). Furthermore, up to 2.5% of solid organ transplant recipients, 12% of patients with AIDS, and 40% of patients with chronic granulomatous disease could be affected by this infection (8). The mortality rate in patients with invasive aspergillosis with pulmonary involvement and persistent neutropenia was 95% (6). Of all the known Aspergillus species, Aspergillus fumigatus is the most common species associated with human disease.

The successful management of invasive aspergillosis is hampered by difficulties in establishing diagnosis. The “gold standard” for making a diagnosis is to obtain a positive culture of A. fumigatus and to demonstrate histological evidence of mycelial invasion from tissue biopsy. Due to the very sick nature of these patients and often the presence of bleeding diathesis, tissue biopsy is often not possible or acceptable for patients. For serological diagnosis of invasive aspergillosis, although commercial kits for antigen detection assay using a monoclonal antibody against the galactomannan antigen extract are available for clinical use, no commercially available antigen or antibody detection kit based on recombinant antigens of Aspergillus is presently available. Recombinant antibody and antigen detection tests may offer a higher specificity and reproducibility. Moreover, recombinant antigens and the correspondingly generated antibodies are easy to standardize.

Recently, the cloning of the AFMP1 gene, which encodes an antigenic cell wall galactomannoprotein of A. fumigatus (Afmp1p), was described, and it was demonstrated that an enzyme-linked immunosorbent assay (ELISA) using purified recombinant Afmp1p was useful for specific antibody detection in patients with invasive A. fumigatus infection (5, 19). In this study, we report the development of an ELISA-based antigen test for the serodiagnosis of invasive A. fumigatus infection with polyclonal serum obtained from guinea pigs and rabbits immunized with purified recombinant Afmp1p protein. The sensitivity and specificity of such an assay and those of a combined Afmp1p antibody and antigen test in patients with aspergilloma and invasive aspergillosis are also compared.

A. fumigatus, Aspergillus flavus, Aspergillus niger, and Aspergillus terreus were clinical isolates from patients with invasive aspergillosis after BMT at Queen Mary Hospital, Hong Kong (18). Penicillium marneffei was a clinical isolate from a patient with systemic penicilliosis at Queen Mary Hospital. Candida albicans was a blood culture isolate from a patient with systemic candidiasis at Queen Mary Hospital. Cryptococcus neoformans was a blood culture isolate from a patient with systemic cryptococcosis at Queen Mary Hospital. Histoplasma capsulatum (ATCC 26032) and Blastomyces dermatitidis (ATCC 26199) were obtained from the American Type Culture Collection (Manassas, Va.).

A. fumigatus, A. flavus, A. niger, A. terreus P. marneffei, C. neoformans, H. capsulatum, B. dermatitidis, and C. albicans were grown first on Sabouraud agar plates at 37°C for 2 or 3 days to get single colonies. Broth cultures were obtained by inoculating fungal cells from plates into synthetic RPMI medium (Gibco-BRL, Gaithersburg, Md.) and further shaking at 37°C for 1 to 5 days to achieve a cell density of greater than 10⁵/ml of culture.

Human sera were obtained from patients with computed tomography- and culture-documented aspergilloma caused by A. fumigatus (n = 9; 6 from Queen Mary Hospital and 3 from Grantham Hospital). Sera were obtained from BMT recipients and patients with hematological maligancies with culture- and histology-documented invasive aspergillosis caused by A. fumigatus (n = 15; Queen Mary Hospital). Control sera were obtained from healthy blood donors (n = 100), patients with culture-documented typhoid fever (n = 8, Queen Mary Hospital), melioidosis (n = 4; Queen Mary Hospital), penicilliosis marneffei (n = 20; Queen Mary Hospital), and systemic candidiasis (n = 6; Queen Mary Hospital). Guinea pig and rabbit anti-Afmp1p antibodies were produced as described previously (5).

Fungal cells were collected by centrifugation and were resuspended in lysis buffer (25 mM Tris-HCl [pH 7.5], 100 mM NaCl, 0.1% Nonidet P-40, 1 mM EDTA, 0.5 mM dithiothreitol, and 1 mM phenylmethylsulfonyl fluoride). After disruption of the cells by sonication, the lysed cells were centrifuged at 13,000 rpm in a microcentrifuge (Eppendorf, Hamburg, Germany) for 15 min. The supernatants were collected as cell lysates.

To obtain culture supernatants of A. fumigatus, the cells were grown in 500 ml of RPMI medium to an optical density at 600 nm (OD₆₀₀) of 1. They were precipitated and resuspended in 20 ml of RPMI medium and were shaken at 37°C for an additional 2 h. After centrifugation, the culture supernatant was passed through a 0.45-μm-pore-size filter (Corning Inc., Corning, N.Y.). The proteins in the supernatant were precipitated by adding 80 ml of saturated (NH₄)₂SO₄, and the protein pellet was resuspended in 500 μl of H₂O.

For Western blot analysis, approximately 20 μg of proteins from the cell lysates or 5 to 10 μl of concentrated culture supernatant of A. fumigatus was loaded onto each lane of a sodium dodecyl sulfate-10% polyacrylamide gel and subsequently the proteins were blotted onto a nitrocellulose membrane (Bio-Rad, Hercules, Calif.). The blot was incubated with a 1:1,000 dilution of guinea pig anti-Afmp1p antibodies, and the proteins were then detected with an enhanced chemiluminescence fluorescence system (Amersham Life Science, Little Chalfont, Buckinghamshire, England) as described previously (2).

The Afmp1p antibody test for aspergillosis was previously described (5). To produce ELISA plates for the Afmp1p antigen test for aspergillosis, Nunc (Roskilde, Denmark) immunoplates were coated with a guinea pig anti-Afmp1p antiserum at a 1:5,000 dilution for 12 h and were further blocked in phosphate-buffered saline with 2% bovine serum albumin. The serological test was performed as described previously (2). Specifically, fixed amounts of purified Afmp1p proteins, diluted fungal culture supernatants, or 1:20-diluted human serum specimens were added to the wells and the plate contents were incubated at 37°C for 2 h. After the wells were washed, the rabbit anti-Afmp1p antiserum was added at a 1:500 dilution and the plate contents were incubated at 37°C for 1 h. After the wells were washed, 1:2,000-diluted alkaline phosphatase-conjugated goat anti-rabbit antibody was added. Detection was carried out with p-nitrophenyl phosphate substrate (Sigma ImmunoChemicals, St. Louis, Mo.). The cutoff OD₄₀₅ of the ELISA was defined as the mean plus 5 times the standard deviation of the OD₄₀₅ values of sera obtained from normal blood donors. Eight patients with culture-documented typhoid fever, 4 patients with culture-documented melioidosis, 20 patients with culture-documented penicilliosis marneffei, and 6 patients with culture-documented candidiasis were used as the disease controls.

To examine cell culture supernatants of A. fumigatus for the presence of the Afmp1p mannoprotein, the supernatant was concentrated and 10 μl (Fig. 1, lane 1) of the concentrated samples was loaded onto a sodium dodecyl sulfate-protein gel for Western blot analysis with specific antiserum against Afmp1p. As a positive control for Afmp1p, 20 μg of A. fumigatus cell lysate was also loaded onto the same gel (Fig. 1, lane 2). The Western blot was probed with a guinea pig anti-Afmp1p antibody. The results of the Western blot analysis presented in Fig. 1 revealed the presence of the Afmp1p protein with a molecular mass of about 60 kDa in both the cell lysate (Fig. 1, lane 2) and the concentrated cell culture supernatant (Fig. 1, lane 1). The size of the protein is significantly greater than the predicted molecular mass of 31.4 kDa on the basis of its amino acid sequence, and this is likely to be due to the glycosylation of Afmp1p.

FIG. 1.

Western blot analysis of Afmp1p in the culture supernatant of A. fumigatus. Lane 1, 10 μl of a concentrated culture supernatant; lane 2, 20 μg of a cell lysate.

To exclude the cross-reactivity between anti-Afmp1p antibody and proteins of other pathogenic fungi, cell lysates were made from A. flavus, A. niger, A. terreus, P. marneffei, C. albicans, C. neoformans, B. dermatitidis, and H. capsulatum. The cell lysates were analyzed by Western blot with a rabbit anti-Afmp1p antibody. The results in Fig. 2 indicated that only the lysate of A. fumigatus contains a reactive band of 60 kDa (lane 1), whereas none of the other fungal pathogens has cross-reacting protein (lanes 2 to 9). Thus, no cross-reacting protein from the cell lysates of eight medically important pathogenic fungi can be detected with the anti-Afmp1p antibody.

FIG. 2.

Afmp1p is specific for A. fumigtus cells. Approximately 20 μg of each of the cell lysates from A. fumigatus (lane 1), A. flavus (lane 2), A. niger (lane 3), A. terreus (lane 4), P. marneffei (lane 5), C. albicans (lane 6), C. neoformans (lane 7), B. dermatitidis (lane 8), and H. capsulatum (lane 9) was loaded into each lane for Western blot analysis with a specific rabbit anti-Afmp1p antibody.

The presence of Afmp1p in cell culture supernatants raised the possibility that the protein antigen could be detected in serum specimens from infected patients. To develop a sensitive test for the detection of Afmp1p for the diagnosis of aspergillosis, two types of polyclonal antibodies were obtained from both rabbits and guinea pigs after immunization with purified Afmp1p protein. A sandwich ELISA system was then generated with a guinea pig anti-Afmp1p antiserum as the capturing antibody and a rabbit anti-Afmp1p antiserum as the detection antibody.

By using a serial dilution of the purified recombinant Afmp1p protein, a standard curve for the Afmp1p antigen test was obtained, as shown in Fig. 3. Bovine serum albumin was used to establish the baseline for the test at an OD₄₀₅ of 0.124. The cutoff value was set to be 0.248, which is equal to twice the OD₄₀₅ for bovine serum albumin. The largest dilution of the purified recombinant Afmp1p protein gave a concentration of 8 pg/ml. By the ELISA the OD₄₀₅ at this dilution is 0.328, which is well above the cutoff of 0.248. Therefore, the lower limit of the detection sensitivity of the test is 8 pg/ml for the Afmp1p protein. The standard curve is linear for Afmp1p at concentrations between 0.1 and 1 ng/ml, therefore allowing the quantitation of the Afmp1p protein.

FIG. 3.

Standard curve for Afmp1p determined with a purified recombinant Afmp1p protein.

The result from a limited-dilution experiment of a culture supernatant of A. fumigatus with a density between 1 × 10⁶ and 2 × 10⁶ cells/ml is shown in Fig. 4. A positive ELISA signal can be detected with a 1:10 dilution but not with a 1:20 dilution. Thus, a positive ELISA signal could be expected from the cell culture supernatant of A. fumigatus with 2 × 10⁵ cells/ml.

FIG. 4.

Afmp1p antigen ELISA is specific for A. fumigatus. Cell culture supernatants were obtained from cultures of A. fumigatus, A. flavus, A. niger, A. terreus, P. marneffei, C. albicans, C. neoformans, B. dermatitidis, and H. capsulatum grown to densities of about 1 × 10⁶ to 2 × 10⁶ cells/ml and were subjected to the Afmp1p antigen ELISA. The positive ELISA values are restricted to the supernatants of A. fumigatus cells. Notice that the largest dilution for the A. fumigatus culture supernatant with a positive OD₄₀₅ is 1:20.

To determine the specificity of the Afmp1p ELISA, culture supernatants were obtained from important pathogenic fungi, A. fumigatus, A. flavus, A. niger, A. terreus, P. marneffei, C. albicans, C. neoformans, B. dermatitidis, and H. capsulatum, grown to densities between 1 × 10⁶ and 2 × 10⁶ cells/ml. Serial dilutions of the supernatants were produced and were analyzed by this test. The result of the ELISA is presented in Fig. 4. Only the A. fumigatus culture supernatant gave a positive signal. None of the other culture supernatants has OD₄₀₅ values greater than 0.2. In fact, none of those values is different from that for the culture medium alone. The results indicate that this ELISA-based antigen test is specific for A. fumigatus and has no cross-reactivity with other important pathogenic fungi.

A clinical evaluation of Afmp1p was carried out. To establish the baseline of this assay, serum specimens from 100 healthy blood donors were tested (Fig. 5). The mean OD₄₀₅ for these specimens as determined by the ELISA was 0.0728, with a standard deviation of 0.0266. The cutoff OD₄₀₅ of the ELISA was then defined as follows: cutoff = mean + 5 × standard deviation = 0.206. As invasive aspergillosis is an uncommon infection, the cutoff was set sufficiently high to eliminate the occurrence of false-positive results in the test.

FIG. 5.

Evaluation of sensitivity and specificity of the A. fumigatus antigen test for the detection of aspergillosis in patients.

Serum specimens were obtained from a total of 9 patients with culture- and histology-documented aspergilloma and 15 patients with invasive aspergillosis. The results of the Afmp1p antigen test presented in Fig. 5 indicate that, among the patients with aspergilloma, zero of nine patients (0%) tested positive, whereas among the patients with invasive aspergillosis, 8 of 15 patients (53%) tested positive.

In addition to the antigen test, Afmp1p antibody tests were also performed with all serum specimens listed in Fig. 5. The antibody test was done as described previously (5). A detailed analysis of both Afmp1p antibody and Afmp1p antigen test results is presented in Table 1. For patients with aspergilloma, all nine patients were positive according to the antibody test, but none was positive according to the antigen test. For patients with invasive aspergillosis, none of the 15 patients was positive by both the antibody and the antigen tests. However, 13 of the 15 (86.7%) patients were either antibody positive (5 out of 15) or antigen positive (8 out of 15). Therefore, the proportion of patients who are positive by at least one of the tests is 86.7%.

TABLE 1.

Performance of antigen and antibody ELISA for diagnosis of A. fumigatus invasive aspergillosis

Serological test resulta	No. of patients (%)
	Aspergilloma (n = 9)	Invasive aspergillosis (n = 15)	Negative controls (n = 138)
Ag+, Ab+	0 (0)	0 (0)	0 (0)
Ag+, Ab−	0 (0)	8 (53.3)	0 (0)
Ag−, Ab+	9 (100)	5 (33.3)	0 (0)
Ag−, Ab−	0 (0)	2 (13.3)	138 (100)
Positive by either testb	9 (100)	13 (86.7)	0 (0)

^aAg⁺, antigen test positive; Ab⁺, antibody test positive; Ag⁻, antigen test negative; Ab⁻, antibody test negative.

^bThe sensitivity, as indicated by a positive result by either test, is 86.7% (13 of 15), with a specificity of 100% (138 of 138).

The specificity of the Afmp1p antigen test is expected to be higher than that of a Western blot assay since the antigens need to be recognized by two different antibodies to produce a positive signal. In this study, the Afmp1p antigen test results with serum specimens from 100 healthy blood donors, 8 patients with culture-documented typhoid fever, 4 patients with culture-documented melioidosis, 20 patients with culture-documented penicilliosis marneffei, and 6 patients with culture-documented candidiasis indicated that none of the subjects in the negative control groups produced a positive signal, showing that the test has a high specificity of 100%.

We report the development of a recombinant antigenic protein-based ELISA for detection of specific antigens in patients with aspergillosis. The cloning and characterization of a highly antigenic cell wall mannoprotein (Mp1p) in P. marneffei have previously been described (1), and it has been shown that ELISA based on recombinant Mp1p is very useful for serodiagnosis of penicilliosis marneffei (2, 3). Since there is no commercially available recombinant antigen-based kits for serodiagnosis of A. fumigatus infections, it would be logical to search for the Mp1p homologue in A. fumigatus and to examine its potential for serodiagnostic purposes. Recently, the cloning of the AFMP1 gene, which encodes the Mp1p homologue of A. fumigatus, was reported (19). Indirect immunofluorescence and immunoelectron microscopy studies indicate that Afmp1p is specifically located in the cell walls of A. fumigatus. Western blot studies showed that patients with invasive A. fumigatus infections develop high levels of specific antibody against Afmp1p. Furthermore, we have demonstrated that an ELISA using purified recombinant Afmp1p was useful for specific antibody detection in patients with invasive A. fumigatus infection (5). In this study, we evaluated the clinical usefulness and the corresponding sensitivity and specificity of a sandwich Afmp1p ELISA-based antigen test for the serodiagnosis of aspergillosis.

A combination of the antigen and antibody tests was much sensitive than either test alone for serodiagnosis of A. fumigatus invasive aspergillosis. Most patients with invasive aspergillosis were severely immunocompromised. They were not able to generate the specific immune response against Afmp1p. Therefore, they would be Afmp1p antibody negative but Afmp1p antigen positive. On the other hand, in about one-third of the patients, they were still able to generate the specific immune response against Afmp1p. Therefore, they were able to clear the Afmp1p in their sera. Hence, they would be Afmp1p antigen negative but antibody positive. For the patients who were Afmp1p antigen and antibody negative, we speculate that they may have generated an immune response that was just sufficient to clear the Afmp1p antigen to an undetectable level or due to antigen-antibody complex formation but that there was no excessive antibody in their sera. Depending on the degree of immunosuppression, serial serum specimens from these patients will probably show that they were either antigen positive and antibody negative or vice versa in the course of their disease. Overall, the sensitivity of combined Afmp1p antigen and antibody testing was 86.7%, which is comparable to that of the capture ELISA for galactomannan antigen (92.6%), which was used for serial serum monitoring (9). Further studies on a combination of antigen and antibody testing on serial serum specimens from patients at high risk of invasive aspergillosis should be performed. They may reveal a higher sensitivity of the tests and will show whether the presence of Afmp1p antigen or antibody precedes the development of clinical disease, which is crucial for the management of invasive aspergillosis.

The Afmp1p antigen test is not useful for serodiagnosis of aspergilloma. None of the patients with aspergilloma had detectable levels of Afmp1p in their sera, but they were all antibody positive. The Afmp1p antibody testing is therefore the diagnostic test of choice for patients with suspected aspergilloma. This phenomenon of differential antigen and antibody response in patients with invasive aspergillosis and aspergilloma and in patients with invasive aspergillosis but with different degrees of immunosuppression is probably related to the difference in immune status of the two groups of patients and is analogous to the situation demonstrated in a study on antibody production in human immunodeficiency virus (HIV)-positive and HIV-negative patients with P. marneffei infections (16). In that study, Mp1p-based ELISA showed a low sensitivity for HIV-positive (more immunocompromised) and a high sensitivity for HIV-negative (less immunocompromised) patients. On the other hand, detection of serum Mp1p using polycloncal Mp1p antibody-based ELISA was able to achieve a higher sensitivity in HIV-positive patients but a lower sensitivity in HIV-negative patients with P. marneffei infections.

Cloning of the gene encoding the homologous protein of Afmp1p in A. flavus and other Aspergillus species would be helpful for the diagnosis of invasive aspergillosis and aspergilloma. Although A. fumigatus is the most common Aspergillus species that causes human disease, other Aspergillus species, such as A. flavus, A. niger, and A. terreus, are also associated with human disease (8, 18). Therefore the Afmp1p antigen and antibody tests may miss the aspergilloma and invasive aspergillosis caused by the other Aspergillus species, though they are good for serodiagnosis of A. fumigatus-related diseases. Cloning the genes encoding the corresponding antigens in these Aspergillus species and using ELISA based on a combination of Afmp1p and the homologous proteins in other Aspergillus species would probably benefit the serodiagnosis of aspergilloma and invasive aspergillosis as a whole.

In addition to laboratory diagnosis, the Afmp1p protein may have a potential use as an immunomodulating glycopeptide in patients at risk of developing invasive aspergillosis. Mannoproteins have been implicated both in activating nonspecific immunity (12, 13, 15) and in eliciting cell-mediated immunity (10, 14). Recently, we have demonstrated that Mp1p-based DNA vaccine is useful for prevention of P. marneffei infections (17). For extracellular opportunistic fungi, antibodies against mannan of C. albicans have been shown to protect against intravenously injected C. albicans cells (7). Similarly, monocloncal antibodies against the capsular polysaccharide glycomannan of C. neoformans prolonged survival when mice were inoculated with the fungus (4, 11). As for invasive aspergillosis, it is shown clearly that Afmp1p is closely associated with humoral immunity. Antibodies generated by Afmp1p-based vaccines could have neutralizing activity on the infectious conidia by preventing the adherence of the fungus onto the surface of the host cells that represents the very first step of fungal invasion. Furthermore, elevation of the antibody response might lead to lysis of the mold by stimulating the complement pathway and facilitate phagocytosis of the mold by opsonization, thereby preventing infection. Further studies with Afmp1p-based vaccines as well as passive immunization using humanized monoclonal antibodies against Afmp1p in patients at risk of developing invasive aspergillosis may shed light on the prevention of such a deadly infection in immunocompromised hosts.