Patient-derived granulocyte/macrophage colony-stimulating factor autoantibodies reproduce pulmonary alveolar proteinosis in nonhuman primates

Abstract

Rationale: Granulocyte/macrophage colony-stimulating factor (GM-CSF) autoantibodies (GMAb) are strongly associated with idiopathic pulmonary alveolar proteinosis (PAP) and are believed to be important in its pathogenesis. However, levels of GMAb do not correlate with disease severity and GMAb are also present at low levels in healthy individuals.

Objectives: Our primary objective was to determine whether human GMAb would reproduce PAP in healthy primates. A secondary objective was to determine the concentration of GMAb resulting in loss of GM-CSF signaling in vivo (i.e., critical threshold).

Methods: Nonhuman primates (Macaca fascicularis) were injected with highly purified, PAP patient-derived GMAb in dose-ranging (2.2-50 mg) single and multiple administration studies, and after blocking antihuman immunoglobulin immune responses, in chronic administration studies maintaining serum levels greater than 40 microg/ml for up to 11 months.

Measurements and main results: GMAb blocked GM-CSF signaling causing (1) a milky-appearing bronchoalveolar lavage fluid containing increased surfactant lipids and proteins; (2) enlarged, foamy, surfactant-filled alveolar macrophages with reduced PU.1 and PPARgamma mRNA, and reduced tumor necrosis factor-alpha secretion; (3) pulmonary leukocytosis; (4) increased serum surfactant protein-D; and (5) impaired neutrophil functions. GM-CSF signaling varied inversely with GMAb concentration below a critical threshold of 5 microg/ml, which was similar in lungs and blood and to the value observed in patients with PAP.

Conclusions: GMAb reproduced the molecular, cellular, and histopathologic features of PAP in healthy primates, demonstrating that GMAb directly cause PAP. These results have implications for therapy of PAP and help define the therapeutic window for potential use of GMAb to treat other disorders.

Figure 1.

Effects of granulocyte/macrophage colony–stimulating factor (GM-CSF) affinity-purified, pulmonary alveolar proteinosis (PAP) patient-derived GM-CSF autoantibodies (GMAb) (23, 25) on GM-CSF signaling in nonhuman primates in vitro and in vivo. (A) GMAb blocked GM-CSF–dependent growth of TF-1 cells in vitro. Results of four separate experiments were fit by nonlinear regression to a four parameter logistic curve (y = min + [max – min]/ 1 + [x/EC50]^Hillslope, where min = 0 ± 2.26, max = 102 ± 1.77, EC50 = 0.208 ± 0.008, and Hillslope = 4.8159 ± 0.691; P = 0.0001). The correlation coefficient is indicated. (B) GMAb blocked GM-CSF–stimulated phosphorylation of STAT5 in blood leukocytes in heparinized whole blood (top) and in alveolar macrophages (without removing from bronchoalveolar lavage [BAL] fluid) obtained from passively immunized (+ GMAb) or control (no GMAb) primates. Clinical specimens were obtained in Study 6 on Day 125 and incubated without or with GM-CSF (10 ng/ml, 15 min) followed by Western blotting to antibodies against phosphorylated or total STAT5 (phosphoSTAT5 and STAT5, respectively). (C) Reversible inhibitory effects of GMAb on GM-CSF signaling in blood neutrophils in vivo. Heparinized blood samples from passively immunized (filled circle) and control (open circle) primates were incubated in the absence or presence of various concentrations of GM-CSF and the increase in leukocyte CD11b levels (i.e., CD11b stimulation index) was measured by flow cytometry (25). Specimens were obtained in Study 6 on Day 295, at which time the levels of GMAb in GMAb-injected and control primate were 60.9 μg/ml and zero, respectively. Similar results were obtained with samples obtained for five other study dates. The GM-CSF concentration used in the standard assay throughout this study is indicated by the dashed line. (D) Effect of GMAb dose on serum levels 30 minutes after intravenous administration. The regression line and 95% confidence interval are shown. (E) Serum half-life of GMAb in passively immunized primates without (open circle) or with (filled circle) B-cell depletion, clinical studies 1–5 or 6–7, respectively.

Figure 2.

Effects of granulocyte/macrophage colony–stimulating factor autoantibodies (GMAb) on serum and leukocyte biomarkers of pulmonary alveolar proteinosis (PAP) and homeostatic B-cell expansion following pharmacologic B-cell depletion. Healthy nonhuman primates received rituximab (open diamond) and cyclophosphamide (open triangle) to block antihuman immunoglobulin immune responses and GMAb (50 [filled diamonds] or 20 [filled triangles] mg/dose) as indicated continuously to maintain serum levels above 40 μg/ml for 327 days in Study 6 (left), or 110 days in Study 7 (right). Serum levels of GMAb, whole-blood CD11b stimulation index (CD11b SI), serum surfactant protein (SP)-D, and B-cell counts were measured in GMAb-exposed (filled circle) and control (open circle) primates as indicated. In Study 6, serum SP-D levels were fit by polynomial regression (R² = 0.52, GMAb-exposed primate). Serum SP-D concentration in untreated primates (n = 10) is indicated (hatched region). Bronchoalveolar lavage (BAL) (open square) and lung biopsies (open circle) were performed on GMAb-exposed and control primates on the days indicated in parentheses. Measurements obtained before first injection of GMAb are indicated (asterisk). Following the last dose of GMAb in Study 6, the serum concentrations of GMAb were 44, 39, and 24 μg/ml on Days 321, 328, and 348, respectively. Similarly, in Study 7 serum concentrations of GMAb were 47, 28, and 12 μg/ml on Days 105, 119, and 147, respectively.

Figure 3.

Effects of granulocyte/macrophage colony–stimulating factor autoantibodies (GMAb) on lung histopathology. Primates were those described in the legend to Figure 2; lung specimens were obtained at 112 days (A–C, E–L) or 321 days (D). (A–C) Typical histopathologic lesions in the lung of a primate exposed to GMAb for 112 days showing alveoli with well-preserved wall architecture (solid arrows) that are filled with foamy alveolar macrophages (asterisks) and intraalveolar eosinophilic debris (open arrows). The lung histology of the corresponding phosphate-buffered saline (PBS)–injected primate was normal (not shown). Hematoxylin and eosin stain. (D) Typical histopathologic lesion in the lung of a primate exposed to GMAb for 321 days. Note that alveoli have well-preserved walls (arrows) and are filled with foamy alveolar macrophages (asterisk). (E and F) Histochemical staining of neutral lipids in frozen lung sections from a GMAb-exposed (E) or PBS-exposed (F) primate. The dark red color indicates neutral lipid accumulation in alveolar macrophages. Oil red O staining with hematoxylin counterstaining. (G–L) Lung biopsy specimens from GMAb-injected (G, I, and K) or PBS-injected (H, J, and L) primates were immunostained to detect surfactant protein (SP)-A (G and H), SP-B (I and J), or SP-D (K and L) and counterstained with nuclear fast red. Original photomicrographs were obtained at magnifications of ×20 (A, C, and G–L), ×100 (B and C), or ×10 (E and F). The bar represents 20 μm (A–D, G–L) or 50 μm (E and F).

Figure 4.

Effects of granulocyte/macrophage colony–stimulating factor autoantibodies (GMAb) on bronchoalveolar lavage (BAL) cytology. Cytology was evaluated in BAL cell preparations from GMAb-exposed (A, D, G, J, and L) or control (B, E, H, K, and L) primates obtained in Study 6 on Day 225. (Similar results were obtained in Study 6 on Days 125, 202, and 348, and in Study 7 on Days 99 and 139.) Original magnification ×20. Scale bar = 50 μm. (A and B) Diff-Quick staining. (C) Alveolar macrophage size relative to the control mean, determined morphometrically in 600 cells from each BAL sample evaluated (n = 3 per primate). Alveolar macrophages were larger in GMAb-exposed primates (P < 0.001). (D and E) Periodic acid–Schiff (PAS) staining. (F) Visual grading scale used to quantify PAS staining of alveolar macrophages (left) and the results of staining of alveolar macrophages from passively immunized (+ GMAb) and control (− GMAb) primates (right). A total of 600 alveolar macrophages were evaluated for BAL specimen (n = 3 per primate). Each cell was assigned a number (0, 1, 2, or 3) reflecting the degree of PAS staining (0, +, ++, or +++) as indicated. The number of cells at each grade was multiplied by the numerical grade and products added and divided by 600 to obtain the PAS staining score. The mean (± SEM) score of three separate BAL specimens was increased in the GMAb-exposed compared with control primate (P < 0.001). (G and H) Oil red O staining. (I) Quantification of oil red O staining in alveolar macrophages was analogous to that used for PAS staining and was increased in the GMAb-exposed compared with control primate (P = 0.008). (J and K) surfactant protein-B immunostaining. (L) Numbers of alveolar macrophages (AM), neutrophils (PMN), and lymphocytes (Lym) in BAL fluid from GMAb (n = 12) and control (n = 11) primates. Cell differentials were determined by examination of 500 cells per sample using Diff-Quick–stained slides (20). Neutrophils were not detected (ND) in controls. Asterisks indicate a P < 0.001 for the comparison between GMAb-exposed and controls.

Figure 5.

Effects of granulocyte/macrophage colony–stimulating factor autoantibodies (GMAb) on alveolar macrophage ultrastructure. (A and B) Lung ultrastructure after exposure to phosphate-buffered saline (PBS) (A) or GMAb (B) for 321 days. Alveoli in GMAb-exposed primates have well-preserved walls (double arrows) and contain alveolar macrophages (arrows) that are enlarged and foamy. (Epon-embedded semithin lung sections stained with toluidine blue, original magnification ×100). Bar = 10 μm. Electron micrographs of alveolar macrophages obtained by bronchoalveolar lavage (BAL) from primates exposed to PBS (C) or GMAb (D–H) for 112 days. Alveolar macrophages from GMAb-exposed primates had a characteristic, foamy appearance caused by numerous lamellar body (arrows) and lipid droplet inclusions (asterisks). Some alveolar macrophages were normal in size but filled nearly completely with lamellar bodies (D), whereas others were enlarged and had varying numbers of lamellar bodies and lipid droplets (E and F) and some were very large and were filled completely with lipid droplets (G). Examination at higher magnification revealed that lamellar inclusions had a visible limiting membrane (arrows), whereas lipid droplets did not (asterisks). (H) (Epon-embedded thin sections stained with Uranyl acetate/lead citrate, original magnification ×7,000). Bar = 2 μm (C–G) or 0.5 μm (H). The numbers of lamellar bodies (I) and lipid droplets (J) were significantly increased in alveolar macrophages after only 112 days of GMAb exposure compared with alveolar macrophages from the control primate. Compared with controls, the mean diameter of lipid droplets in alveolar macrophages was greater in GMAb-exposed primates (K). For each comparison (I–K), 100 cells were evaluated per primate (P < 0.01). (L) Number of inclusions (lamellar plus lipid) per cell versus cell size alveolar macrophages from the GMAb-exposed primate. Alveolar macrophage size was the area of the cell in pixels (divided by 10⁶) determined using electronic digital photomicrograph files of 100 cells using Axiophot (Leica) software. Linear regression analysis was performed using SigmaPlot software.

Figure 6.

Effects of granulocyte/macrophage colony–stimulating factor autoantibodies (GMAb) on alveolar macrophage function. Compared with controls, alveolar macrophages from GMAb-exposed primates had significantly decreased PU.1 (A) and PPARγ (B) mRNA transcript levels, and secreted a reduced amount of tumor necrosis factor-α in response to lipopolysaccharide stimulation (C) (n = 3 determinations per condition; P < 0.001 for all comparisons with control). Results are for samples obtained in Study 6 on Day 225, and are similar to results on three or one other determinations in Study 6 and 7, respectively (A and B).

Figure 7.

Effects of granulocyte/macrophage colony–stimulating factor autoantibodies (GMAb) pulmonary surfactant homeostasis. Primates were those described in the legend to Figure 2 and bronchoalveolar lavage (BAL) specimens (n = 16, +GMAb; n = 15, −GMAb) were obtained as illustrated. (A) Gross appearance of BAL fluid. Specimens were from Study 6, Day 308, similar to results on five other occasions. (B) Turbidity of the BAL fluid from GMAb-exposed (n = 16) and control (n = 15) primates. (C) Total phospholipid concentration in BAL fluid (n = 10 each). (D) Saturated phosphatidylcholine concentration in BAL fluid (n = 10 each). (E) Total protein in BAL fluid (n = 6 samples per group; P = 0.088). (F) surfactant protein-A, -B, -C, and -D levels in BAL fluid relative to respective mean values in controls. Bars represent the mean of triplicate (GMAb-exposed) or duplicate (control) determinations. Double asterisks indicate P value < 0.001 for the comparison of results from GMAb-exposed (black bars) and phosphate-buffered saline–exposed (gray bars) primates.

Figure 8.

Measurement of the critical threshold of granulocyte/macrophage colony–stimulating factor autoantibodies (GMAb) in blood and lung. (A) CD11b stimulation index. Specimens (n = 57) were from Studies 6 and 7. (B) Phagocytic capacity (PC) of neutrophils in whole blood. Specimens (n = 7, evaluated in triplicate) were from Study 6 on Days 210, 238, 245, and 252, and from Study 7 on Days 43, 57, and 63. (C) Effect of serum GMAb concentration on the CD11b stimulation index in passively immunized primates. At concentrations of GMAb above the putative critical threshold (solid line), the CD11b stimulation index was reduced to zero (dashed line). Specimens are from Study 3 (single 14.7-mg GMAb dose; open circle; n = 8 determinations); Studies 4 and 5 (multiple 20-mg GMAb doses; triangle up, triangle down; n = 6 and 15 determinations, respectively); and Studies 6 and 7 (multiple 50-mg or 20-mg GMAb doses with B-cell depletion; open diamond, open square; n = 52 and 39 determinations, respectively). (D) GMAb in serum and lung (ELF) from GMAb-exposed primates. Specimens (n = 12) were from Study 6, Days 125, 225, and 308; Study 7, Day 99. (E) Serum surfactant protein-D concentration. Specimens (n = 65) were from Studies 6 and 7. (F) Number of days that the GMAb concentration in lung epithelial lining fluid (ELF) was maintained above specified levels in Studies 6 (open diamond) and 7 (open square).

Figure 9.

Proposed mechanism by which granulocyte/macrophage colony–stimulating factor (GM-CSF) autoantibodies cause pulmonary alveolar proteinosis. High levels of GM-CSF autoantibodies (GMAb) block GM-CSF signaling to alveolar macrophages and their precursors, thereby arresting terminal differentiation (dashed line) and impairing surfactant catabolism (open arrow) and host defense functions (dotted line) of alveolar macrophages. Surfactant homeostasis is disrupted by the slow, progressive accumulation of surfactant, resulting in eventual respiratory dysfunction of insidious onset with impaired oxygen diffusion, hypoxemia, and dyspnea. Loss of GM-CSF signaling does not alter basal neutrophil blood counts, cellular morphology, or expression of PU.1 and multiple differentiation markers. GMAb do reduce the capacity of some neutrophil functions in rheostatic fashion (dotted line) [25] contributing to an increased risk of infection but without affecting neutrophil differentiation. GM-CSF also blocks the homeostatic expansion of B lymphocytes after acute B lymphopenia (not shown).