INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Pulmonary alveolar proteinosis (PAP) is a rare lung disease characterized by excessive accumulation of surfactant proteins and phospholipids within the alveoli and terminal bronchioli (1, 2). The accumulation of surfactant proteins is believed to be caused by impaired clearance rather than excessive secretion (3). Because alveolar macrophages (AM) play an important role in the clearance of surfactant proteins and lipids (4- 6), it is plausible that impaired clearance of surfactant by AM causes PAP. PAP is a heterogeneous disorder, and has been classified into two groups; acquired PAP and congenital PAP (C-PAP). Acquired PAP may be further divided into two distinct forms, an idiopathic (I-PAP) and a secondary form (S-PAP). Most PAP patients are those with I-PAP, which affects predominantly adults and occurs in the absence of identifiable underlying diseases or inhalation (7).

Patients with PAP usually complain of dyspnea and cough, and some patients may improve spontaneously, whereas others progress to develop respiratory failure. Chest radiograph typically shows diffuse, finely nodular, soft infiltrates in a perihilar butterfly pattern, and in some cases accompanied by an interstitial pattern (8, 9). However, any other diseases that show a diffuse acinar-filling pattern on chest radiograph, such as pulmonary edema, idiopathic pulmonary fibrosis, and granulomatous lung diseases must be differentiated from PAP. Because whole lung lavage provides temporary symptomatic relief for patients with PAP (10), definitive diagnosis is crucial before severe respiratory failure occurs. At present, definitive diagnosis of PAP requires bronchoalveolar lavage, and in some problematic cases, transbronchial lung biopsy or open lung biopsy is also required (7). These diagnostic methods are invasive, and may be associated with risk of morbidity and even mortality in cases with severe respiratory failure. Several serological markers, such as surfactant protein (SP)-A, SP-D, and KL-6, have been reported to be elevated in PAP patients (7, 11). However, they are also elevated in the sera from patients with other lung diseases, and so far no specific marker for the serological diagnosis of PAP is established.

Recently, we reported the specific occurrence of neutralizing autoantibodies against granulocyte-macrophage colony-stimulating factor (GM-CSF) with IgG isotype in the bronchoalveolar lavage fluid (BALF) from 11 Japanese I-PAP patients, and in sera from five Japanese I-PAP patients (12). The autoantibody is considered to be a causative agent for I-PAP, because dysfunction of AM due to neutralization of endogenous GM-CSF bioactivity by the autoantibody should result in impairment of surfactant clearance. In this report, we confirm and extend our previous findings and show that the high titer autoantibody is present in sera from all 24 I-PAP patients in Japan, Australia, Switzerland, the United States, and New Zealand. Furthermore, in an effort to establish a convenient and less invasive procedure for the diagnosis of I-PAP, we developed and evaluated a new latex agglutination test for the detection of the autoantibody in sera.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

We obtained the sera from 24 I-PAP, four S-PAP, and two C-PAP patients, 40 patients with other lung diseases (10 cases each of idiopathic pulmonary fibrosis and acute respiratory distress syndrome, nine cases of sarcoidosis, five cases of collagen vascular lung diseases, and one case each of bronchiolitis obliterans organizing pneumonia, asbestosis, hemosiderosis, allergic bronchopulmonary aspergillosis, lymphangiomyomatosis, and pulmonary hyalinizing granulomatosis), and 40 normal subjects in Japan, Australia, Switzerland, the United States, and New Zealand (Table 1). The diagnosis of I-PAP was based on biochemical analysis of BALF (13 of 24, 54.2%) and/or histopathological findings of lung biopsy (transbronchial lung biopsy [12 of 24, 50%] and open lung biopsy [11 of 24, 45.8%]). The I-PAP patients did not have any concurrent infectious diseases or hematological disorders, and had no history of toxic inhalation. Two of four S-PAP patients had underlying infectious diseases, and the others developed PAP secondarily to hematological disorders. No subjects had been treated with GM-CSF in the past. Eleven of 24 I-PAP patients were enrolled on a study evaluating the efficacy of GM-CSF treatment in I-PAP after sample collection of sera for present study (J. F. Seymour and coworkers, submitted). All sera were stored at 70° C until assayed. Written informed consent to collect samples of sera was obtained from all subjects.

Blot Assay with ¹²⁵I-GM-CSF

The method of this assay has been described previously (12). Briefly, proteins in sera were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using gradient gel (2 to 15%) under a nonreducing condition. Separated proteins were transferred electrophoretically to a polyvinylidene fluoride (PVDF) membrane. The membrane was fixed with 10% (vol/vol) acetic acid and 50% (vol/ vol) methanol, stained with Coomassie brilliant blue solution, washed with methanol, and treated overnight at 4° C with a blocking reagent containing 1% (wt/vol) bovine serum albumin and 0.1% (vol/vol) Tween 20. The membrane was incubated with 0.16 nM ¹²⁵I-recombinant human (rh) GM-CSF (Escherichia coli derived; NEN Life Science Products, Boston, MA) for 1 h at room temperature. After washing, the membrane was exposed to X-ray film for 4 d.

Antigen Capture Assay

This assay was performed according to the method described previously (12). Briefly, sera were diluted 1,500-fold with phosphate-buffered saline (PBS). A volume of 50 µl of diluted sera was transferred to a plate coated with 1 µg/ml of rhGM-CSF (Escherichia coli derived; Kirin Brewery Co. Ltd., Gunma, Japan) and the plate was kept at room temperature for 1 h. After washing, autoantibodies captured by rhGM-CSF were detected by peroxidase-labeled anti-human IgA, IgD, IgE, IgG, or IgM antibody (Dako Corporation, Carpinteria, CA). After washing, color was developed using tetramethylbenzidine and the absorbance was measured at 450 nm.

3-[4,5-Dimethylthiazol-2yl]-2,5-Diphenyltetrazolium Bromide (MTT) Assay

The method of this assay has been described previously (12). In brief, TF-1 cells (1 × 10⁴ cells/well), a GM-CSF-dependent cell line (15), were incubated for 4 d at 37° C in 100 µl complete medium with 0.5 ng/ml of rhGM-CSF (Kirin Brewery Co. Ltd.) and 1 µl of sera. Then 5 µg/ml of MTT (Sigma, St. Louis, MO) was added and incubated for 3 h at 37° C. After formation of formazan crystals, acid isopropanol (0.04 mol/L HCl in isopropanol) was added to dissolve the crystals, and the absorbance was measured at 595 nm.

Latex Beads Coupled with rhGM-CSF

Latex beads of 1 µm in diameter (Polystyrene Polybeads, Polyscience, Inc., Warrington, PA) were coupled with rhGM-CSF (Kirin Brewery Co. Ltd.) using Glutaraldehyde Kit for Amino Beads and Blue Dyed Beads (Polyscience, Inc.) according to the manufacturer's instruction. As a result, 50 µg of rhGM-CSF was coupled to 1 ml of 1.25% (wt/ vol) latex beads.

Isolation of the Autoantibody Using the rhGM-CSF-coupled Latex Beads

A volume of 100 µl of 1.25% (wt/vol) rhGM-CSF-coupled latex beads and 50 µl of serum from an I-PAP patient were incubated at room temperature on a rotary shaker for 1 h. The mixture was centrifuged at 20,000 × g for 6 min, and the supernatant was removed. The pellet was washed 3 times with PBS containing 0.1% (wt/vol) Tween 20. The autoantibody was eluted from the latex beads by resuspending and shaking vigorously in 0.1 M sodium citrate, pH 2.5, for 10 min. The suspension was centrifuged and the supernatant was dialyzed against PBS. The concentration of the isolated autoantibody was determined by a sandwich ELISA with nonlabeled and peroxidase-labeled anti-human IgG antibodies (Dako Corporation), using human IgG (Dako Corporation) as a standard for quantitation.

Immunoblot Assay

The autoantibody isolated using the rhGM-CSF-coupled latex beads was subjected to SDS-PAGE under a nonreducing condition, transferred electrophoretically to a PVDF membrane, and the membrane was treated with a blocking reagent as described previously. To detect IgG, the membrane was incubated with 0.3 µg/ml of peroxidase-labeled anti-human IgG antibody for 1 h at room temperature. To determine the subclasses of the autoantibody, the membrane was incubated with 5 µg/ml of biotinylated anti-human IgG₁, IgG₂, IgG₃, or IgG₄ antibody (PharMingen, San Diego, CA) for 1 h at room temperature. After washing, the membrane was incubated with 0.3 µg/ml of streptavidin- peroxidase for 1 h at room temperature. Color was developed using diaminobenzidine.

Latex Agglutination Test

Fifty µl of 0.05% (wt/vol) rhGM-CSF-coupled latex beads and 10 µl of diluted serum with PBS were transferred to a round-bottom microplate and mixed well. The plate was kept at room temperature for 24 h. Agglutination was detected by visual inspection. The titer of the autoantibody was defined as the endpoint dilution rate of serum at which agglutination was observed.

Statistics

Data are shown as means ± SE. The comparison of data between two groups was performed using unpaired t test. The comparison of results between two assays was performed using paired t test. Analysis of the results of the latex agglutination test was performed using chi-square test. A p value less than 0.05 was considered to be significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Consistent Presence of the Neutralizing Autoantibody against GM-CSF in Sera from Patients with I-PAP

Previously we had reported the presence of neutralizing autoantibodies against GM-CSF with IgG isotype in sera from 5 Japanese I-PAP patients (12). To confirm the specific and consistent presence of such autoantibodies in sera from patients with I-PAP, we examined sera from donors including 24 I-PAP patients, in Japan (n = 10), Australia (n = 7), Switzerland (n = 3), the United States (n = 3), and New Zealand (n = 1). All specimens were assayed by the same operator who was blinded to the diagnosis of each donor. Sera from three of 10 Japanese I-PAP patients had been already examined by blot assay with ¹²⁵I-rhGM-CSF in the previous study (12). The blot assay gave characteristic bands with a molecular weight of 180 kDa in the sera from all patients with I-PAP that we had previously reported as the autoantibody (12). Such rhGM-CSF binding proteins were not detected in the sera from 4 S-PAP patients, 2 C-PAP patients, or 40 normal subjects. The autoantibodies captured by rhGM-CSF-coated plate reacted with only anti-human IgG antibody, but not with anti-human IgA, IgD, IgE, or IgM antibody (data not shown). All sera from I-PAP patients significantly inhibited GM-CSF-dependent cell growth of TF-1 cells (Figure 1). These results indicate that the neutralizing autoantibody against GM-CSF with IgG isotype is specifically and consistently present in the sera from all I-PAP patients in different countries.

View larger version (25K): [in this window] [in a new window]   Figure 1.   Inhibition of GM-CSF-dependent cell growth by the sera from patients with I-PAP. TF-1 cells, a GM-CSF-dependent cell line, were incubated in the medium containing rhGM-CSF with the sera from I-PAP patients or normal subjects. The vertical axis is the absorbance measured at 595 nm. Data are shown as means ± SE. All 24 sera from I-PAP patients inhibit the cell growth significantly.

Specific Binding of the Autoantibody to the rhGM-CSF-coupled Latex Beads

We made latex beads coupled with rhGM-CSF and evaluated their specific binding activity of the autoantibody in the sera from patients with I-PAP. The sera and the latex beads were incubated and the protein associated with the latex beads was eluted. As shown in Figure 2, the protein associated with the latex beads was IgG, and had ¹²⁵I-rhGM-CSF binding activity. In contrast, proteins unassociated with the latex beads contained IgG, but had no ¹²⁵I-rhGM-CSF binding activity. ¹²⁵I-rhGM-CSF binding protein was not detected in either fraction when serum from a normal subject was used (data not shown). These results indicated that the latex beads specifically bind the autoantibody in sera from I-PAP patients.

View larger version (94K): [in this window] [in a new window]   Figure 2.   Specific binding of the autoantibody in the serum from a patient with I-PAP to the rhGM-CSF-coupled latex beads. Serum from the I-PAP patient (lanes 1, 4, and 7 ), proteins associated with the latex beads (lanes 3, 6, and 9), and those unassociated (lanes 2, 5, and 8) were subjected to SDS-PAGE under a nonreducing condition. The proteins were stained with Coomassie blue (lanes 1, 2, and 3), or assayed for anti-human IgG antibody binding (lanes 4, 5, and 6), or 125I-rhGM-CSF binding (lanes 7, 8, and 9). Proteins associated with the latex beads are IgG (lane 6), and have GM-CSF binding activity (lane 9). Proteins unassociated with the latex beads contain IgG (lane 5), but have no GM-CSF binding activity (lane 8). The apparent molecular weight of the IgG seems to be larger than the usual weight for IgG, because proteins were subjected to SDS-PAGE under a nonreducing condition.

Further Characterization of the Autoantibody

Using the autoantibody eluted from the latex beads, we determined subclasses of the autoantibody by immunoblot assay with anti-human IgG subclass specific antibodies. In all specimens from 10 patients with I-PAP examined, the autoantibody reacted with both anti-human IgG₁ and IgG₂ antibodies, but not with anti-human IgG₃ or IgG₄ antibody (data not shown). The results indicated that the autoantibody formed in I-PAP patients is not monoclonal.

Quantitation of the Autoantibody in Sera from I-PAP Patients

To quantify the autoantibody in the sera from I-PAP patients, we employed two methods, antigen capture assay using anti-human IgG antibody and latex agglutination test using the latex beads. We performed the antigen capture assay using the autoantibody isolated from serum of a patient with I-PAP as a standard for quantitation (2 to 300 ng/ml), and the titer was calculated from the absorbance. For the latex agglutination test, we used 2-fold serially diluted sera. The mean titer of the autoantibody in the sera from 24 I-PAP patients was 180 ± 22 µg/ml (range, 35 to 430 µg/ml) by the antigen capture assay, and 5,400 ± 1,200 fold (range, 512- to 24,576-fold) by the latex agglutination test. As shown in Figure 3, a good correlation was found between the results of the two methods (r = 0.78, p < 0.01).

View larger version (16K): [in this window] [in a new window]   Figure 3.   Quantitation of the autoantibody against GM-CSF in sera from patients with I-PAP. The vertical axis is the titer determined by the latex agglutination test and the horizontal axis is the titer determined by the antigen capture assay. Significant correlation is found between the results of the two methods.

Serological Diagnosis of I-PAP by the Latex Agglutination Test

Using the latex beads, we examined sera from 110 donors including 24 I-PAP, four S-PAP, and two C-PAP patients, 40 patients with other lung diseases, and 40 normal subjects. Results of the latex agglutination test on 300-fold diluted sera are shown in Table 1 and Figure 4. Agglutination was observed in all sera from I-PAP patients, and in two of 40 sera from normal subjects. As described previously, blot assay with ¹²⁵I-rhGM-CSF and antigen capture assay using peroxidase-labeled anti-human IgG antibody showed that two sera from normal subjects with agglutination did not contain the autoantibody. Overall sensitivity of the latex agglutination test was 100% and specificity was 98% (chi-square value = 99.177, p < 0.0001). These results showed that the latex agglutination test is a reliable method for serological diagnosis of I-PAP.

View larger version (27K): [in this window] [in a new window]   Figure 4.   Latex agglutination test for serological diagnosis of I-PAP. A volume of 50 µl of 0.05% (wt/vol) rhGM-CSF-coupled latex beads and 10 µl of 300-fold diluted sera were transferred to a round bottom microplate and mixed well. The plate was kept at room temperature for 24 h. Latex beads agglutinated by the autoantibodies fall to the bottom of the well in a fine mat. In contrast, latex beads that are not agglutinated fall to the bottom in a button, therefore, a clear dot is observed as a negative result. Agglutination is observed in sera from eight I-PAP patients, but not in sera from eight patients with other lung diseases or eight normal subjects.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Previously, we reported the occurrence of the neutralizing autoantibody against GM-CSF in sera from five Japanese patients with I-PAP (12). In general, both specificity and consistency of serological markers are essential for definitive diagnosis. Examining the sera from 24 patients with I-PAP in different countries, we confirmed in this study that the neutralizing autoantibody against GM-CSF is consistently present in sera from I-PAP patients with high titer. The autoantibody was not detected in those from S-PAP or C-PAP patients, or from normal subjects. These findings support the rationality of detecting the autoantibody in the sera for diagnosis of I-PAP.

All four methods employed in this study to detect the autoantibody in the sera, namely, blot assay with ¹²⁵I-rhGM-CSF, antigen capture assay using anti-human IgG antibody, growth inhibition assay using a GM-CSF-dependent cell line, and a novel latex agglutination test using the rhGM-CSF coupled latex beads, can be used for serological diagnosis of I-PAP. Among these methods, however, the latex agglutination test is the simplest and the most convenient, and can be readily performed for diagnosis of I-PAP without special technique or equipment. The rhGM-CSF-coupled latex beads specifically bind the autoantibody in the sera from patients with I-PAP and hence we could isolate the autoantibody by simply eluting it from the latex beads by exposing them to low pH. By this isolation, we could identify the subclasses of the autoantibody to be a mixture of IgG₁ and IgG₂. The results indicate that the autoantibody that occurred in the patients is not monoclonal, though the reason why the autoantibody is generated in I-PAP patients is not clear at this stage.

In this study, we showed that the latex agglutination test is in fact useful for diagnosis of I-PAP. Because only the free form of the autoantibody but not the antibody bound to GM-CSF (immune complex form) can be adsorbed to the latex beads, we may not detect the total amount of the autoantibody in the patient's sera by this method. Similarly, all other assays except the blot assay also detect only the free form. To measure the total amount of the autoantibody, the blot assay must be performed, in which GM-CSF is dissociated from the autoantibody during SDS-PAGE. However, the values obtained with all the assays described can be taken as whole amounts in the sera, because the concentrations of GM-CSF in the patient's sera are 10⁸ order of magnitude lower than those of the neutralizing autoantibody (data not shown).

It was previously observed that 0.3 to 2% of sera from healthy volunteers contained autoantibodies against GM-CSF with a low titer (16, 17), and that GM-CSF administrated for biotherapy of malignant diseases induced neutralizing or non-neutralizing autoantibody against GM-CSF (17, 18). It is, therefore, possible that sera from such subjects may show positive results by the latex agglutination test. In this study, two sera from normal subjects were of positive results by the latex agglutination test. However, it was confirmed that these two sera did not contain the autoantibody by the blot assay and the antigen capture assay. It is unclear what component or components in the sera caused nonspecific bindings to the latex beads. Neither of these two normal subjects has developed any clinical evidence of any lung disorder.

Until now, a definitive diagnosis of PAP has been based on biochemical analysis of BALF, or pathological findings of specimens obtained by open lung biopsy or transbronchial lung biopsy, or a combination of these methods (7). These procedures are invasive, and are associated with moderate risks among those patients with severe respiratory failure. The availability of a reliable serological diagnosis of PAP will relieve patients from these invasive procedures. Serum levels of SP-A, SP-D, and mucin-like glycoprotein, KL-6 were reported to be elevated in patients with PAP, but these markers were also elevated in patients with other diseases, such as idiopathic pulmonary fibrosis and collagen vascular lung diseases (7, 11). Our results indicate that the autoantibody is a specific and useful marker for serological diagnosis of I-PAP. Although PAP is a rare disease, the latex agglutination test is useful to exclude I-PAP from other lung diseases described previously.

Finally, we propose a new concept for I-PAP based on our findings. I-PAP is autoimmune PAP, because it is quite likely that the neutralizing autoantibody against GM-CSF is the causative agent for I-PAP.