Serologic Analysis of Antigen-Specific Reactivity in Patients with Systemic Candidiasis
Janice K. Au-Young, Frederic A. Troy, and Elliot Goldstein
Antibody responses to candidal polypeptides and mannans were studied in patients with systemic candidiasis, candiduria, and other fungal and bacterial infections, and in healthy laboratory personnel to determine the diagnostic value of these immunologic responses. When tested by immunoblot analysis, sera from 15 patients with systemic candidiasis frequently contained antibodies to three antigens: 15 of 15 sera from patients with invasive disease reacted to a molecular species having a molecular weight (Mr of 90-200 kd, 13 of 15 reacted with a 45-kd polypeptide, and 12 of 15 reacted with a 17-kd polypeptide. Lesser reactivity was observed in 11 of 15 sera with a 28-kd candidal antigen and in 9 of 15 to a 57-kd candidal antigen. Quantitation of antibody titers against the 45-kd candidal polypeptide demonstrated much higher immunoreactivity in patients with systemic candidiasis than in patients with superficial candidal infections, bacterial infections, other systemic mycoses, and healthy individuals. Antimannan antibody titers were measured by an enzyme-linked immunosorbent assay (ELISA ) and these titers were also higher in patients with systemic candidiasis than in patients in the other categories. These differences, however, were less than those observed with the anti-45- kd polypeptide antibody. Therefore, the ability to detect systemic candidiasis is improved by testing sera for immunoreactivity to polypeptide and to mannan antigens from Candida al- bicans. Detection af polypeptide antibodies improves the serodiagnosis of systemic candidiasis.
INTRODUCTION
Al though many strategies have been at tempted, the serodiagnosis of systemic can- didias is cont inues to be a perplexing medical problem. Some patients can be diag- nosed by detect ing cand ida l ant igenemia or metaboli tes (Kerkering et al., 1979; Kiehn et al., 1979; Marier et al., 1982; Meckstroth et al., 1981; Meunier-Carpent ier and Armstrong, 1981; Mil ler et al., 1974; Segal et al., 1979; Stevens et al., 1980; Warren et al., 1977; Weiner and Yount, 1976; Weiner and Coats-Stephen, 1979; Wong et al., 1982) but because of the presence of blocking ant ibody (Lew et al., 1982; Au-Young and Goldstein, 1982) and the short half-life of antigen wi th in the b loodst ream (Warren et al., 1978), ant igenemia is an inconsis tent finding. Similarly, ant ibody responses to candida l antigens are easi ly demonstrable, but the interpreta t ion of these results
Presented in part at the 23rd Interscience Conference on Antimicrobial Agents and Chemo- therapy, October 1983, Abstract No. 1033.
From the Division of Infectious and Immunologic Diseases, Department of Internal Medicine (J.K.A.-Y. and E.G.), and Department of Biological Chemistry (F.A.T.), School of Medicine, University of California, Davis, California.
Address reprint requests to: Dr. Elliot Goldstein, Division of Infectious and Immunologic Diseases, UCD Professional Building, 4301 X Street, Sacramento, CA 95817.
Received July 12, 1984; accepted October 30, 1984.
is equivocal since patients with superficial infection or colonization also have ele- vated antibody titers (Filice et al., 1977). Moreover, individuals with impaired im- mune function who develop systemic candidiasis often have low levels of antibody and therefore yield false-negative test results (Guinan et al., 1979).
Recently, more exacting methods using immunoblot analysis have been developed to study serologic responses to complex antigen systems (Towbin et al., 1979). This suggested to us that these techniques, which can detect antibodies to specific anti- genic components of C. albicans, might aid in differentiating patients with systemic infection from those with superficial or noncandidal infections. This study used ELISA and immunoblot analysis to assess the antibody reactivity of sera with specific candidal mannans and polypeptides. The serum samples evaluated were from pa- tients with systemic candidiasis, superficial candidiasis, noncandidal fungal infec- tions, or bacterial infections. This preliminary report shows that patients with sys- temic candidiasis often have high levels of antibody against specific candidal polypeptides and less often have high levels of antibody against candidal mannas.
MATERIALS AND METHODS
Patient Populations
Sera were collected from patients with candidiasis who were treated at the University of California at Davis Medical Center (UCDMC). Patients were classified as having systemic candidiasis according to the criteria of Filice et al. (1977): (a) gastrointestinal infection: histologically evident infection was present distal to the esophagus; (b) fungemia: Candida were isolated from at least two blood cultures and there were no clinically apparent sources at postmortem examinations; blood cultures were per- formed by culturing a 40-ml volume of sterile filtered trypticase soy broth with a 10- ml volume of whole blood, creating a 50-ml culture; (c) disseminated candidiasis: histologically proven infection was present in any internal organ. Sera from patients with coccidiomycosis were kindly provided by Dr. D. Pappagianis; sera from patients with other systemic mycoses and bacterial infections were obtained from individuals treated at the UCDMC. Normal human sera were collected from healthy laboratory personnel.
Antigens
Mannans were prepared according to the original method of Peat el al. (1961) using the modifications of Kocourek and Ballou (1969). Briefly, yeast phase organisms of C. albicans isolated during the stationary phase were suspended in 0.02 M citrate buffer, pH 7.0. The cell suspension was autoclaved at 121°C for 90 min, cooled, and then centrifuged. The supernatant was saved, and the pellet was resuspended in citrate buffer and reextracted as described above. The supernatant fractions were pooled. Mannans were precipitated as a mannan-copper complex from solution at 4°C using Fehling's solution (3.5% CuSO 4 5H20, 1.7% NaK tartrate, 5% NaOH). The mannan-copper complex was separated in 3 N HC1 and mixed again with metha- nol:acetic acid, dissolved in distilled water, recomplexed with Fehling's solution, dissolved in 3 N HC1, reprecipitated with methanol:acetic acid, and washed exten- sively over a sintered glass filter. The final product was air-dried.
French Press Extraction of C. albicans
Organisms were grown to midexponential phase at 37°C, pelleted, and washed three
Antigens of C. albicans 421
times in saline. For each gram of cells (wet weight), 0.02 g sodium deoxycholate was added. Prior to solubilizing, the microorganisms were suspended in 10 mM Tris HC1, 0.015 M NaC1, 0.02% NAN3, and 1 mM phenylmethylsulfonylfluoride (PMSF). Cells were disrupted by pressure disintegration in a French press cell at 6000 lb/in 2. Cell debris was pelleted at 1000 g for 15 min, and the supernatant fraction containing the candidal antigens was stored in aliquots at -20°C.
SDS-Polyacrylamine Gel Electrophoresis (SDS-PAGE) Gradient SDS-polyacrylamide slab gels of 5-15% or 8-15% were used in the dis- continuous buffer system of Laemmli (1970). Samples were resuspended in electro- phoresis sample buffer (62.5 mM Tris, 2% SDS, 5% 2-mercaptoethanol, 10% glycerol, 0.01% bromphenol blue (w/v), pH 6.8) heated at 100°C for 1.5 min then loaded onto a single 120-mm well or individual 3-mm wells. Electrophoresis was performed using the Hoefer No. 520 vertical slab gel apparatus or the Hoefer Protean slab gel apparatus with coolant refrigerated to 10°C. The voltage was set at 150 V for 6-8 h (12 cm running gels ) or for 30 h (25 cm running gels). Following electrophoresis, the gel was trimmed at the dye front and stained in 0.5% Coomassie blue R-250 in 45% methanol (v/v), 45% water, and 10% acetic acid for 3 h. The gels were destained successively with 45% methanol (v/v), 45% water, and 10% acetic acid and then with 10% methanol (v/v), 7.5% acetic acid, and 82.5% water, and then dried in a gel dryer.
Electroblotting of Proteins Protein transfer to nitrocellulose was performed according to the method of Towbin et al. (1979). Transfer occurred at 300 mA for 18 hr in 25 mM Tris, pH 8.3,192 mM glycine, and 20% (v/v) methanol. The region of the nitrocellulose with the molecular weight standards was cut and stained with 0.1% amido black in 45% methanol, then destained with 10% methanol and 7.5% acetic acid.
Antigenic Analysis Using Immunoblotting Unbound sites on the nitrocellulose were blocked with 3% BSA in Tris-EDTA-NaC1 buffer (0.05 M Tris HC1, pH 7.4, 0.001 M EDTA, 0.15 M NaC1, with 0.02% NAN3, 0.05% Tween 20). The nitrocellulose was cut into 8-mm strips, placed in the relevant serum diluted at 1:100 (unless otherwise noted) in Tris-EDTA-NaC1 buffer, and in- cubated for 2 hr with agitation at room temperature. Blots were washed three times in Tris-EDTA-NaC1 buffer, then incubated with 125I-goat antihuman IgG (120,000-280,000 dpm/ml) for a minimum of 1 hr with agitation. The immunoblots were washed three times to remove unbound antibodies, air dried, and exposed to Kodak X-Omat Ar film with an intensifying screen at -70°C for autoradiography.
Radiolabeling of Antiglobulin 125I-labeled antihuman IgG was prepared with purified goat antihuman IgG (Cappel Laboratories) and Na~25I using the chloramine T method (Greenwood et al., 1963). Purified antibody (1 mg) in 0.2 ml PBS was added to 1 mCi of carrier-free Na~25I (New England Nuclear) in a borosilicate test tube. Coupling of the radiolabel to the antibody was initiated by 20 ~L1 choloramine T (2.4 mg/ml in PBS), and it was ter- minated after 1.5 min by the addition of 0.1 ml of potassium metabisulfite (2.4 mg/ml in PBS). The excess iodine was separated from the mixture by passage through a Sephadex G-25 column (PD-10 column, Pharmacia). The void volume fractions con-
422 J.K. Au-Young et al.
taining the highest specific activity were pooled and stored at -20°C with BSA (10 mg/ml).
For quantitation of 125I goat antihuman IgG, the radioactivity contained on the nitrocellulose blot at 45 kd was measured by the method of Lasky and Troy (1984).
Lactoperoxidase Labeling of C. albicans
Cells were grown to midlogarithmic phase, washed several times, and resuspended to a concentration of 5 × 109 cells in 150 p3 of PBS. Lactoperoxidase (100 p./50 p.l) and I mCi of Na125I were added to the cells, then H202 (20 p,] of a 0.03% Iv/v] solution) was transferred to the suspension. After vigorous mixing at 30°C, an identical amount of H202 was added. The mixture was allowed to incubate at room temperature for 10 min. To stop the reaction, PBS containing 0.02% (w/v) sodium azide and 2 mM potassium iodide was added. Prior to extraction of cellular proteins, the labeled cells were washed extensively. TO the pelleted cells, 0.5 ml of extraction buffer containing 0.4% (w/v) sodium deoxycholate, 10 mM Tris HC1, 0.15 M NaC1, 0.02% NAN3, and I mM PMSF was added. The mixture was stirred for I hr at 37°C. After centrifugation at 6,000g, the radiolabeled supernatant and pellet were mixed with sample buffer for electrophoresis.
ELISA for Antimannan Antibody Titers
Antimannan antibody titers were determined by a noncompetitive, indirect ELISA (Voller et al., 1976) in which the wells of polystyrene micro-ELISA plates (No. 011- 010-33350, Dynatech Laboratories) were coated with mannan prepared as described. Each well was incubated with 100 ng/100 p~l of mannan in 0.1 M sodium bicarbonate buffer, pH 9.6, for 3 hr at 37°C. The wells were washed with distilled water. Serial dilutions of donor sera were made with Tris-EDTA-buffered saline containing 0.1% bovine serum albumin and 0.05% Tween 20, pH 7.4 (TEN buffer) and incubated with mannan-coated wells for one hour at 37°C. The wells were again washed, and per- oxidase-labeled goat antihuman immunoglobulin (igG + IgA + IgM) antiserum (Cappel Laboratories) diluted 1:400 in TEN buffer was added to each well for another 1 hr incubation at 37°C. After repeating the washing, 100 ~L1 of a solution containing 2 mM 2,2'-azino-di-(3-ethylbenzthiazoline) sulfonic acid, 0.08% hydrogen peroxide, and 0.05 M sodium citrate buffer at pH 4.0 was added to the wells. The color was allowed to develop for 10 min before the reaction was stopped with 0.2 N hydrogen fluoride at pH 3.3. The optical density of the colored product was measured at 405 nm by use of an ELISA microspectrophotometer (Dynatech Laboratories). The anti- body titers were defined by the highest dilution of each test serum that gave an absorbance greater than three standard deviations above the mean background optical density.
Protein Concentrations
Protein concentrations were determined by the Coomassie blue binding assay of Bradford (1976). Bovine plasma albumin and/bovine gamma globulin were included as standards each time the assay was performed.
RESULTS
The deoxycholate (DOC) solubilized extract of yeast phase Candida albicans, sero- type A, subjected to pressure shearing yielded 40-50 polypeptides upon SDS-
Antigens of C. albicans 423
acrylamide gel electrophoresis and Coomassie blue staining. Figure 1 shows the results of two solubilization experiments conducted under identical conditions.
Figure 2 shows an immunoblot in which DOC solubilized antigens were incubated with 15 serum samples from patients with systemic candidiasis (all sera were diluted 1:100). Detection of antibody reactivity was achieved with a radiolabeled second antibody and autoradiography. Detailed analysis of three strongly recognized antigens showed that 15 of 15 sera reacted with a molecular species having a diffuse Mr of 90-200 kd, 13 of 15 sera reacted at 45 kd, and 12 of 15 reacted at 17 kd. Frequent responses of less intensity were observed at 28 kd for 11 of 15 sera and at 57 kd for 9 of 15 sera.
Figure 3 shows the immunoblots from patients with candidal urinary tract infec- tions, with systemic mycoses other than candidiasis, with bacterial infections, and from normal individuals. Reactivity with the 90-200-kd antigen was widespread. All three patients with candidal urinary tract infections, three of five patients with bacterial infections, five of eight patients with noncandidal infections, and four of five normal individuals had antibodies reacting with this diffuse antigen. Addition- ally, recognition of the 45- and 17-kd antigens was frequent in the three samples from patients with candidal urinary tract infections (two of three responded to the 18-kd antigen and one of three recognized the 45-kd antigen). However, in the groups with noncandidal disease, only two of 18 individuals recognized the 45-kd antigen and six of 18 reacted with the 17-kd antigen.
The ability to quantitate the antibody response in immunoblots was established by demonstrating that the amount of radiolabeled antiglobulin on the nitrocellulose
12J : ~ ~ : : M R X 10 - 3
- - 2 0 0 , 0
_ _ 1 1 6 , 2 5
_ _ 9 2 , 5
_ _ 6 6 , 2
_ _ 45,0
__ 31,0
I , _ _ 21,5
__ 14,4
FIGURE 1. Yeast phase Candida albicans, serotype A, were sub- jected to pressure shearing in the presence of sodium deoxycholate. Approximately 300 p.g protein of the solubilized extract was loaded onto a denaturing, discontinuous 8-15% SDS-polyacrylamide gel, then stained with Coomassie blue. Lanes 1 and 2 are the results of two independent yeast fractionation experiments run under the same conditions (see Materials and Methods).
424 J.K. Au-Young et al
57,0 -- 45,0 --
28,0 --
-- 200.0
__ 116,25 92,5
__ 66,2
~ --- 45, 0
,~ _ _ 31,0
_ _ 21,5
d 14,4
FIGURE 2. Sera from 15 patients with systemic candidiasis tested by the immunoblot procedure, which utilized deoxycholate (DOC) solubilized antigens from C. albicans, serotype A. Each serum sample was diluted at 1:100. The bound IgG was detected by use of ~25I- labeled goat antihuman IgG antiserum.
iiiiiii i ii! e .... " . 0
~ x 10 -3
200.0
116.25
92.5
66,2
45,0
31,0
"~ _ _ 21,5 14.4 O
FIGURE 3. Serum from three patients with superficial candidiasis (lanes 16-18), from five patients with bacteremias (lanes 19-23), from eight patients with systemic mycoses other than candidiasis (lanes 24-31), and from five normal individuals (lanes 32-36) were reacted with the DOC-antigens from C. albicans.
Antigens of C. albicans 425
was proportional to the amount of primary antibody bound. Figure 4a represents the immunoblot from which the quantitation was standardized. Each strip contained the same profile of electroblotted antigens but was incubated with serial twofold dilutions of serum from a patient with systemic candidiasis, starting from 1:50 (strip 10) and ending at 1:25,600 (strip 1). The 45-kd antigen-antibody complex is stained contin- uously for all dilutions (strips 1-10). In Figure 4b, it can be seen that there is a linear relationship between the radioactivity (dpm/slice) and the amount of anti-45-kd an- tibody bound: The amount of radiolabeled goat antihuman Ig antibody was inversely proportional to the serum dilution.
Quantitation of the antibody response to the 45-kd antigen can be seen in Figure 5 for patients with systemic candidiasis (group 1), superficial candidiasis (group 2), bacterial infections (group 3), other mycoses (group 4), and for normal individuals (group 5). The sera from the 13 patients in group 1 reacting strongly to the 45-kd antigen had a radioactive index of 1> 2.9 [index = (dpm sample-dpm background)/ dpm background] while none of the individuals from the remaining groups reached this level of anti-45-kd antibody activity. Although one patient with cryptococcosis (patient 26) and one with a candidal urinary tract infection (patient 17) had moderate amounts of anti-45-kd antibody, the levels of antibody to the 45-kd candidal poly- peptide were much higher in serum samples from patients with systemic candidiasis than in samples from controls or individuals with other infections.
Tables 1 and 2 provide clinical information for each patient, the titer of anti- mannan antibodies as determined by ELISA, and the quantitation of anti-45-kd an- tibodies in immunoblots. (The numbered lanes in Figures 2 and 3 correspond to the patient numbers listed in Tables 1 and 2, respectively.) The antimannan titers range from low (1:20 for patient 4) to high (1:10,240 for patient 15) for the patients with systemic candidiasis. It is emphasized that patient 15 had one of the highest anti- mannan antibody titers but had a low level of anti-45-kd antibody. Immunoreactivity to mannans had no correlation with immunoreactivity to the polypeptide as patients with the highest responses to the 45-kd antigen had relatively low titers of antimannan antibody.
Eleven of the patients described in Table 1 were on chemotherapy for candidiasis. This did not appear to have a positive or negative effect on the humoral antibody response in the patients' sera studied.
Serial serum samples obtained from three patients with systemic candidal infec- tions and tested by immunoblotting with the DOC antigens show that antibodies to candidal polypeptides develop following infection (Figure 6). Patient X's serum con- tained little or no antibody on July 3, 1981. A positive blood culture for C. albicans was reported on July 5, 1981. Three weeks later, antibodies against antigens having molecular masses of 45, 30, and 26 kd were observed in immune blots. Approximately five months after the Candida blood culture was reported, immunoreactivity against three polypeptides had decreased in this patient.
A blood culture from patient Y on June 27, 1982 yielded C. albicans. Serum collected on July 1, 1982 contained antibody directed against antigens of molecular masses of 45, 41, 37, and 30 kd. The intensity of immunoreactivity had increased six days later (7/7/82) and persisted at this level for at least 14 days (7/20/82). Finally, patient Z became blood culture positive for C. albicans on August 15, 1981. His serum of August 21, 1981 showed antibody to the 45-kd antigen. Increased titers to the 45, 41, and 37 kd antigens were observed by September 8, 1981.
The antimannan antibody titers of these three patients' sera were determined. For patient X, the titer increased dramatically and then decreased as follows: 1:640 on July 3, 1981; 1:10,240 on July 24, 1981; 1:40,960 on August 5, 1981; and 1:2560 on December 1, 1981. The titers for patient Y rose slightly form 1:640 on July 1, 1982
FIGURES 4. (A) Quantitation of 125I-labeled antiglobulin bound to the 45-kd antigen. The bands on the nitrocellulose blot containing antibody to the 45-kd antigen were cut out. The radio- activity of 125I goat antihuman IgG antibodies was measured in dpm/slice (log dpm/slice) and plotted as a function of the serum dilution (1/log serum dilution). Correlation coefficient = 0.95.
M r x 10 .3
200 - -
116 -- 9 2 - -
6 6 - -
4 5 - -
31 - -
21 - - 14- -
1 2 3 4 5 6 7 8 9 10
'j - .,., o ,
I
FIGURE 4. [B) Serum titration of the 45-kdianti-45-kd antigen-antibody complex hy use of the sermn of one patient with systemic candidiasis. A constant amount of protein (40 pog) was present in each strip. After blotting, twofold dilutions of the serum were applied from 1:50 (lane 101 to 1:25,600 (lane 1). Development of the autoradiogram is described in Materials and Methods.
Antigens of C. albicans 427
.J WJ
W
m
0
tD
I
F- Z
1 2
!
• $
• $ 0 .O = I , , ,
1 2 3 4 5
GROUP CLASSIF ICATION OF SERUM SAMPLE
FIGURE 5. The anti-45-kd antibody levels of human sera calculated by quantitat ion of im- munoblot radioactivity. Serum samples were classified in the following groups: 1. sera from patients with systemic candidiasis; 2. sera from patients with superficial candidiasis; 3. sera from patients with bacteremias; 4. sera from patients with systemic mycoses other than can- didiasis: and 5. control human sera.
PATIENT
X PATIENT PATIENT
Y Z
"* ~51
" g g - - t - -
fl R X 10 -3
-- 200.0
__ 116.25 92,5
66,2
45.0
31,0
FIGURE 6. Serial serum samples from three pa- tients with systemic candidiasis were reacted against the DOC antigens from C. albicans. The nitrocel- lulose strips were incubated with a 1:100 dilution of each patient 's serum. Dates of serum collection are noted above each lane.
428 J.K. Au-Young et al.
0
<
T
<
[-.,
0
<
cO
<
~ m
~ ° ~ ' ~
u ~ ~ .~
o o
o
F o , ' ~
• ! ~.! .~.~ .~.~oo .~o ~ ~ 0~ .~o -!
®~ ~ ~ ~ < ~ < -
o~ o
=
o
e~
o
o
N
"il o "~
;i m e~
N e,
i R
~ X
~ N .e.~ ..~
~ g N ~ m
~ R ~ . ~
Antigens of C. albicans 429
TABLE 2. Comparison of A n t i m a n n a n Ant ibody Titers with Anti-45-kd Polypeptide Ant ibody Levels in Patients with Superficial Candidiasis, Bacteremias, Systemic Mycoses Other than Candidiasis, and Normal Individuals
Quantitation b of Patient Elisa results ° Anti-45-kd
no. Type of infection Antimannan antibody antibody
32 Control 640 1.91 33 Control 40 1.01 34 Control <20 0.69 35 Control 40 1.06 36 Control 20 1.02
aTiters are expressed as the reciprocal of the highest serum dilution resulting in an absorbance greater than three standard deviations above the mean background optical density.
bMeasurement of anti-45-kd antibody activity is expressed as
dpm sample - dpm mean background radioactivity dpm mean background radioactivity
cUTI = urinary tract infection.
to 1:1280 on July 7 and 20, 1982. Finally, the titers for patient Z decreased from 1:320 on August 21, 1981 to 1:80 September 8, 1981.
Lactoperoxidase-catalyzed iodinat ion of intact yeast cells was performed to label cell surface proteins. Log phase cells were labeled with 125I, solubilized with deoxy- cholate, then centrifuged. The soluble and particulate fractions were analyzed after SDS-polyacrylamide gel electrophoresis by Coomassie blue staining and autoradiog- raphy (not shown). The radiolabeled proteins had apparent molecular weights of >100, 78, 66, 53, 43, 35, 23, 21, and 15 kd. In the same gel, Coomassie blue staining revealed that several polypeptides had been extracted but were not radiolabeled: These polypeptides had apparent Mrs of 49, 45, 41, 38, 34, 32, and 30 kd.
DISCUSSION
These studies demonstrate that patients with systemic candidiasis who are tested by immunoblo t analysis have high levels of ant ibody to three molecular species having apparent Mrs of 90,000-200,000, 45,000, and 17,000, respectively. Also, these patients often have antibodies against candidal antigens of 28,000 and 57,000 daltons. Of
430 J.K. Au-Young et al.
particular interest was the finding that similar levels of antibody reacting with the 45-kd antigen are not observed in the serum of patients with candiduria, with other fungal and bacterial infections, or in controls. Quantitation of this antibody aids in identifying patients with systemic candidiasis. The serial studies of specimens from three patients with documented dates of infection correlates the increasing antibody response to the 45-kd antigen with systemic infection. The conclusion that patients with systemic candidiasis can be identified by demonstrating elevated levels of an- tibody to a polypeptide is corroborated by the recent study of Strockbine et al. (1984), who reported a 44-52-kd antigen. Although our method of antigen preparation was different, it appears likely that the 45-kd antigen reported in this study is equivalent to that previously reported.
The biochemical tests show that the 45-kd candidal antigen is probably a poly- peptide rather than a mannan or mannoprotein. This conclusion is based on the following: 1. the antigen stained with Coomassie blue; 2. it did not stain with periodic acid/Schiff reagent, nor did it bind to Concanavalin A (data not shown); 3. the in- ter~sity of the antibody response to the 45-kd antigen did not correlate with the titers of the antimannan antibody. The absence of radioactivity in the region of the 45-kd antigen when the yeast cell surface was labeled with 125I suggests that the antigen was not externally exposed. Since the 45-kd polypeptide appeared in the Coomassie blue-stained gel, which was later subjected to autoradiography, it was solubilized but not radiolabeled. Therefore, it was presumably located on the internal side of the membrane or within the cytoplasm.
The mannan of yeast cell walls is a collection of glycoproteins that have large mannose-rich oligosaccharides. Mannans are located at the cell surface and have a very small amount of associated protein compared to the amount of total carbohydrate (Ballou, 1976). The 90-200-kd antigen is likely to be mannan, because it was labeled upon catalytic iodination of the cell surface, it stained with the periodic acid/Schiff reagent (not shown), and it did not appear in gels stained with Coomassie blue, reflecting its low proportion of polypeptide. The widespread immunoreactivity to this diffuse antigen by human sera has been reported previously. Our studies also show that patients with systemic candidiasis are more likely to have elevated titers of antimannan antibodies (10/15 had titers of 1:640 or greater) than are patients with bacterial or other fungal infections in whom the titers are less than 640. This cor- relation is in agreement with the study of Greenfield et al. (1983), who reported that levels of antibody to mannan are elevated during episodes of invasive candidiasis. However, since two of three patients with candiduria and one of five controls had antimannan antibody titers of 1:640 or greater, measurements of antimannan antibody were less specific than measurements of antibody to the 45-kd protein in differen- tiating patients with systemic candidiasis from patients with candiduria or controls. However, one patient with disseminated candidiasis had a high antimannan antibody titer and low anti-45-kd antibody levels.
The ability to measure antibody to the 45- and 17-kd p olypeptides and to mannans should aid in diagnosing systemic candidal infections. Detection of elevated levels of antibody to the 45-kd candidal polypeptide suggests a recent systemic infection. If the level of antibody to the 45-kd polypeptide is equivocal or low and antibody to mannan is also low, a systemic infection is unlikely as this combination was not observed in any of the 15 patients with systemic candidiasis. Although testing for both antibodies separates patients with systemic candidiasis from patients with non- candidal infections, these tests do not separate patients with systemic infections from patients wtih candidal urinary tract infections. The latter can have high levels of antibody to candidal mannans and equivocal levels of antibody to the 45-kd antigen, thus rendering them indistinguishable from patients with systemic candidal infec- tion.
Antigens of C. albicans 431
REFERENCES
Au-Young JK, Goldstein E (1982) Detection of Candida antigen by enzyme linked immunosor- bent assay is blocked by anti-mannan antibodies. Abstract. Annual Meeting of the American Society of Mocrobiology.
Au-Young JK, Troy FA, et al (1983) Analysis of antigen-specific reactivity in sera from patients with systemic candidiasis. Abstract. Interscience Conference on Antimicrobial Agents and Chemotherapy.
Brandford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248.
Filice G, Yu B, et al (1977) Immunodiffusion and agglutination tests for Candida in patients with neoplastic disease: Inconsistent carrelation of results with invasive infections. J Infect Dis 135:349.
Greenfield RA, Bussey MJ, et al (1983) Serial enzyme-linked immunosorbent assays for antibody to Candida antigens during induction chemotherapy for acute leukemia. J Infect Dis 148:275.
Greenwood FC, Hunter WM, et al (1963) The preparation of 131I-labeled human growth hormone of high specific radioactivity. Biochem J 89:114.
Guinan ME, Portas MR, et al (1979) The Candida precipitin test in an immunosuppressed population. Cancer 43:299.
Kerkering TM, Espinel-Ingroff A, et al (1979) Detection of Candida antigenemia by counter- immunoelectrophoresis in patients with invasive candidiasis. J Infect Dis 140:659.
Kiehn TE, Bernard EM, et al (1979) Candidiasis: Detection by gas-liquid chromatography of D- arabinitol, a fungal metabolite, in human serum. Science 206:577.
Kocourek J, Ballou CE (1969) Method for fingerprinting yeast cell wall mannan. I Bact 100:1175.
Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacte- riophage T4. Nature 227:680.
Lasky RD, Troy FA (1984) Possible DNA-RNA tumor virus interaction in human lymphomas: Expression of retroviral proteins in Ramos lymphoma lines is enhanced after conversion with Epstein Barr Virus. Proc Natl Acad Sci USA 81:33.
Lew MA, Siber GR, et al (1982) Enhanced detection with an enzyme-linked immunosorbent assay of Candida mannan in antibody-containing serum after heat extraction. J Infect Dis 145:45.
Marier RL, Milligan E, et al (1982) Elevated mannose levels detected by gas-liquid chromatog- raphy in hydrolysates of serum from rats and humans with candidiasis. J Clin Microbiol 16:123.
Meckstroth KL, Reiss E, et al (1981) Detection of antibodies and antigenemia in leukemia patients with candidiasis by enzyme-linked immunosorbent assay. J Infect Dis 144:24.
Meunier-Carpentier F, Armstrong D (1981) Candida antigenemia, as detected by passive hem- agglutination inhibition, in patients with disseminated candidiasis or Candida colonization. J Clin Microbiol 13:10.
Miller GG, Witwer MW, et al (1974) Rapid identification of Candida albicans septicemia in man by gas-liquid chromatography. J Clin Invest 65:1235.
Peat S, Turvey JR, et al (1961) The polysaccharides of baker's yeast, Part IV Mannan. J Chem Soc (London) 28-35.
Segal E, Berg RA, et al (1979) Detection of Candida antigen in sera of patients with candidiasis by an enzyme-linked immunosorbent assay-inhibition technique. J Clin Microbiol 10:116.
Stevens P, Huang S, et al (1980) Detection of Candida antigenemia in human invasive candi- diasis by a new solid phase radio°immunoassay. Infection 8(Suppl 13}:$334.
Strockbine NA, Largen MT, et al (1984) Identification and molecular weight characterization of antigens from Candida albicans that are recognized by human sera. Infect Immun 43:715.
Towbin H, Staehelin T, et al (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets. Proc Natl Acad Sci USA 76:4350.
Voller A, Bidwell DE, et al (1976) Enzyme immunoassays in diagnostic medicine: Theory and practice. Bull WHO 53:55.
Warren RC, Bartlett A, et al (1977) Diagnosis of invasive candidosis by enzyme immunoassay of serum antigen. Br Med J 1:1183.
43Z J.K. Au-Young et al.
Warren RC, Richardson MD, et al (1978) Enzyme-linked immunosorbent assay of antigens from Candida albicans circulating in infected mice and rabbits: The role of mannan. Mycopathol 66:179.
Weiner MH, Yount MJ (1976) Mannan antigenemia in the diagnosis of invasive Candida in- fections. ] Clin Invest 58:1045.
Weiner MH, Coats-Stephen M (1979) Immunodiagnosis of systemic candidiasis: Mannan an- tigenemia detected by radioimmunoassay in experimental and human infections. ] Infect Dis 140:989.
Wong B, Bernard EM, et al (1982) The arabinitol appearance rate in laboratory animals and humans: Estimation from the arabinitol/creatinine ratio and relevance to the diagnosis of candidiasis. ] Infect Dis 146:353.
